Jonathan M. Meyer, MD

Among patients with bipolar I or II disorder (BD I or II), major depressive episodes represent the predominant mood state when not euthymic, and are disproportionately associated with the functional disability of BD and its suicide risk.¹ Long-term naturalistic studies of weekly mood states in patients with BD I or II found that the proportion of time spent depressed greatly exceeded that spent in a mixed, hypomanic, or manic state during >12 years of follow-up (Figure 1²and Figure 2³). In the 20th century, traditional antidepressants represented the sole option for management of bipolar depression despite concerns of manic switching or lack of efficacy.^4,5 Efficacy concerns were subsequently confirmed by placebo-controlled studies, such as the Systematic Treatment Enhancement Program for Bipolar Disorder (STEP-BD) trial, which found limited effectiveness of adjunctive antidepressants for bipolar depression.⁶ Comprehensive reviews of randomized controlled trials and observational studies documented the risk of mood cycling and manic switching, especially in patients with BD I, even if antidepressants were used in the presence of mood-stabilizing medications.^7,8

Several newer antipsychotics have been FDA-approved for treating depressive episodes associated with BD (Table 1). Approval of olanzapine/fluoxetine combination (OFC) in December 2003 for depressive episodes associated with BD I established that mechanisms exist which can effectively treat acute depressive episodes in patients with BD without an inordinate risk of mood instability. Subsequent approval of quetia­pine in October 2006 for depression associated with BD I or II, lurasidone in June 2013, and cariprazine in May 2019 for major depression in BD I greatly expanded the options for management of acute bipolar depression. However, despite the array of molecules available, for certain patients these agents presented tolerability issues such as sedation, weight gain, akathisia, or parkinsonism that could hamper effective treatment.⁹ Safety and efficacy data in bipolar depression for adjunctive use with lithium or divalproex/valproate (VPA) also are lacking for quetiapine, OFC, and cariprazine.^10,11 Moreover, despite the fact that BD II is as prevalent as BD I, and that patients with BD II have comparable rates of comorbidity, chronicity, disability, and suicidality,¹² only quetiapine was approved for acute treatment of depression in patients with BD II. This omission is particularly problematic because the depressive episodes of BD II predominate over the time spent in hypomanic and cycling/mixed states (50.3% for depression vs 3.6% for hypomania/cycling/mixed combined), much more than is seen with BD I (31.9% for depression vs 14.8% for hypomania/cycling/mixed combined).^2,3 The paucity of data for the use of newer antipsychotics in BD II depression presents a problem when patients cannot tolerate or refuse to consider quetiapine. This prevents clinicians from engaging in evidence-based efficacy discussions of other options, even if it is assumed that the tolerability profile for BD II depression patients may be similar to that seen when these agents are used for BD I depression.

Continue to: Table 1...

Lumateperone (Caplyta) is a novel oral antipsychotic initially approved in 2019 for the treatment of adult patients with schizophrenia. It was approved in December 2021 for the management of depression associated with BD I or II in adults as monotherapy or when used adjunctively with the mood stabilizers lithium or VPA (Table 2).¹³ Lumateperone possesses certain binding affinities not unlike those in other newer antipsychotics, including high affinity for serotonin 5HT2A receptors (Ki 0.54 nM), low affinity for dopamine D2 receptors (Ki 32 nM), and low affinity for alpha 1-adrenergic receptors (Ki 73 nM), muscarinic and histaminergic receptors (Ki >100 nM for both).^13,14 However, there are some distinguishing features: the ratio of 5HT2A receptor affinity to D2 affinity is 60, greater than that for other second-generation antipsychotics (SGAs) such as risperidone (12), olanzapine (12.4) or aripiprazole (0.18).¹⁵ At steady state, D2 receptor occupancy remains <40%, and the corresponding rates of extrapyramidal side effects (EPS)/akathisia differed by only 0.4% for lumateperone vs placebo in short-term adult clinical schizophrenia trials,^13,16 by 0.2% for lumateperone vs placebo in the monotherapy BD depression study, and by 1.7% in the adjunctive BD depression study.^13,17,18 Lumateperone also exhibited no clinically significant impact on metabolic measures or serum prolactin during the 4-week schizophrenia trials, with mean weight gain ≤1 kg for the 42 mg dose across all studies.¹⁹ This favorable tolerability profile for endocrine and metabolic adverse effects was also seen in the BD depression studies. Across the 2 BD depression monotherapy trials and the single adjunctive study, the only adverse reactions occurring in ≥5% of lumateperone-treated patients and more than twice the rate of placebo were somnolence/sedation, dizziness, nausea, and dry mouth.¹³ There was also no single adverse reaction leading to discontinuation in the BD depression studies that occurred at a rate >2% in patients treated with lumateperone.¹³

Continue to: Clinical implications...

Clinical implications

The approval of lumateperone for both BD I and BD II depression, and for its use as monotherapy and for adjunctive use with lithium or VPA, satisfies several unmet needs for the management of acute major depressive episodes in patients with BD. Clinicians now have both safety and tolerability data to present to their bipolar spectrum patients regardless of subtype, and regardless of whether the patient requires mood stabilizer therapy. The tolerability advantages for lumateperone seen in schizophrenia trials were replicated in a diagnostic group that is very sensitive to D2-related adverse effects, and for whom any signal of clinically significant weight gain or sedation often represents an insuperable barrier to patient acceptance.²³

Efficacy in adults with BD I or II depression.

The efficacy of lumateperone for major depressive episodes has been established in 2 pivotal, double-blind, placebo-controlled trials in BD I or II patients: 1 monotherapy study,¹⁷ and 1 study when used adjunctively to lithium or VPA.¹⁸ The first study was a 6-week, double-blind, placebo-controlled monotherapy trial (study 404) in which 377 patients age 18 to 75 with BD I or BD II experiencing a major depressive episode were randomized in a 1:1 manner to lumateperone 42 mg/d or placebo given once daily in the evening. Symptom entry criteria included a Montgomery-Åsberg Depression Rating Scale (MADRS) total score ≥20, and scores ≥4 on the depression and overall BD illness subscales of the Clinical Global Impressions Scale–Bipolar Version Severity scale (CGI-BP-S) at screening and at baseline.¹⁷ Study entry also required a score ≤12 on the Young Mania Rating Scale (YMRS) at screening and at baseline. The duration of the major depressive episode must have been ≥2 weeks but <6 months before screening, with symptoms causing clinically significant distress or functional impairment. The primary outcome measure was change from baseline in MADRS. Several secondary efficacy measures were examined, including the proportion of patients meeting criteria for treatment response (≥50% decrease in MADRS), or remission (MADRS score ≤12), and differential changes in MADRS scores from baseline for BD I and BD II subgroups.¹⁷

The patient population was 58% female and 91% White, with 79.9% diagnosed as BD I and 20.1% as BD II. The least squares mean changes on the MADRS total score from baseline to Day 43 were lumateperone 42 mg/d: -16.7 points; placebo: -12.1 points (P < .0001), and the effect size for this difference was moderate: 0.56. Secondary analyses indicated that 51.1% of those taking lumateperone 42 mg/d and 36.7% taking placebo met response criteria (P < .001), while 39.9% of those taking lumateperone 42 mg/d and 33.5% taking placebo met remission criteria (P = .018). Importantly, depression improvement was observed both in patients with BD I (effect size 0.49, P < .0001) and in those with BD II (effect size 0.81, P < .001).

The second pivotal trial (study 402) was a 6-week, double-blind, placebo-controlled adjunctive trial in which 528 patients age 18 to 75 with BD I or BD II experiencing a major depressive episode despite treatment with lithium or VPA were randomized in a 1:1:1 manner to lumateperone 28 mg/d, lumateperone 42 mg/d, or placebo given once daily in the evening.¹⁸ Like the monotherapy trial, symptom entry criteria included a MADRS total score ≥20, and scores ≥4 on the depression and overall illness CGI-BP-S subscales at screening and baseline.¹⁸ Study entry also required a score ≤12 on the YMRS at screening and baseline. The duration of the major depressive episode must have been ≥2 weeks but <6 months before screening, with symptoms causing clinically significant distress or functional impairment. The primary outcome measure was change from baseline in MADRS for lumateperone 42 mg/d compared to placebo. Secondary efficacy measures included MADRS changes for lumateperone 28 mg/d and the proportion of patients meeting criteria for treatment response (≥50% decrease in MADRS) or remission (MADRS score ≤12).

The patient population was 58% female and 88% White, with 83.3% diagnosed as BD I, 16.7% diagnosed as BD II, and 28.6% treated with lithium vs 71.4% on VPA. The effect size for the difference in MADRS total score from baseline to Day 43 for lumateperone 42 mg/d was 0.27 (P < .05), while that for the lumateperone 28 mg/d dose did not reach statistical significance. Secondary analyses indicated that response rates for lumateperone 28 mg/d and lumateperone 42 mg/d were significantly higher than for placebo (both P < .05). Response rates were placebo: 39%; lumateperone 28 mg/d: 50%; and lumateperone 42 mg/d: 45%. Remission rates were similar at Day 43 in both lumateperone groups compared with placebo: placebo: 31%, lumateperone 28 mg/d: 31%, and lumateperone 42 mg/d: 28%.¹⁸ As of this writing, a secondary analysis by BD subtype has not yet been presented.

A third study examining lumateperone monotherapy failed to establish superiority of lumateperone over placebo (NCT02600494). The data regarding tolerability from that study were incorporated in product labeling describing adverse reactions.

Continue on to: Adverse reactions...

Adverse reactions

In the positive monotherapy trial, there were 376 patients in the modified intent-to-treat efficacy population to receive lumateperone (N = 188) or placebo (N = 188) with nearly identical completion rates in the active treatment and placebo cohorts: lumateperone, 88.8%; placebo, 88.3%.¹⁷ The proportion experiencing mania was low in both cohorts (lumateperone, 1.1%; placebo, 2.1%), and there was 1 case of hypomania in each group. One participant in the lumateperone group and 1 in the placebo group discontinued the study due to a serious adverse event of mania. There was no worsening of mania in either group as measured by mean change in the YMRS score. There was also no suicidal behavior in either cohort during the study. Pooling the 2 monotherapy trials, the adverse events that occurred at ≥5% in lumateperone-treated patients and at more than twice the rate of the placebo group were somnolence/sedation (lumateperone 42 mg/d: 13%, placebo: 3%), dizziness (lumateperone 42 mg/d: 8%, placebo: 4%), and nausea (lumateperone 42 mg/d: 8%, placebo: 3%).¹³ Rates of EPS were low for both groups: lumateperone 42 mg/d: 1.3%, placebo: 1.1%.¹³ Mean weight change at Day 43 was +0.11 kg for lumateperone and +0.03 kg for placebo in the positive monotherapy trial.¹⁷ Moreover, compared to placebo, lumateperone exhibited comparable effects on serum prolactin and all metabolic parameters, including fasting insulin, fasting glucose, and fasting lipids, none of which were clinically significant. No patient exhibited a corrected QT interval >500 ms at any time, and increases ≥60 ms from baseline were similar between the lumateperone (n = 1, 0.6%) and placebo (n = 3, 1.8%) cohorts.

Complete safety and tolerability data for the adjunctive trial has not yet been published, but discontinuation rates due to treatment-emergent adverse effects for the 3 arms were: lumateperone 42 mg/d: 5.6%; lumateperone 28 mg/d: 1.7%; and placebo: 1.7%. Overall, 81.4% of patients completed the trial, with only 1 serious adverse event (lithium toxicity) occurring in a patient taking lumateperone 42 mg/d. While this led to study discontinuation, it was not considered related to lumateperone exposure by the investigator. There was no worsening of mania in either lumateperone dosage group or the placebo cohort as measured by mean change in YMRS score: -1.2 for placebo, -1.4 for lumateperone 28 mg/d, and -1.6 for lumateperone 42 mg/d. Suicidal behavior was not observed in any group during treatment. The adverse events that occurred at rates ≥5% in lumateperone-treated patients and at more than twice the rate of the placebo group were somnolence/sedation (lumateperone, 13%; placebo, 3%), dizziness (lumateperone, 11%; placebo, 2%), and nausea (lumateperone, 9%; placebo, 4%).¹³ Rates of EPS were low for both groups: lumateperone, 4.0%, placebo, 2.3%.¹³ Mean weight changes at Day 43 were +0.23 kg for placebo, +0.02 kg for lumateperone 28 mg/d, and 0.00 kg for lumateperone 42 mg/d.¹⁸ Compared to placebo, both doses of lumateperone exhibited comparable effects on serum prolactin and all metabolic parameters, including fasting insulin, fasting glucose, and fasting lipids, none of which were clinically significant.¹⁸

Lastly, the package insert notes that in an uncontrolled, open-label trial of lumateperone for up to 6 months in patients with BD depression, the mean weight change was -0.01 ± 3.1 kg at Day 175.¹³

Continue on to: Pharmacologic profile...

Pharmacologic profile

Lumateperone’s preclinical discovery program found an impact on markers associated with increased glutamatergic neurotransmission, properties that were predicted to yield antidepressant benefit.^14,15,24 This is hypothesized to be based on the complex pharmacology of lumateperone, including dopamine D1 agonism, modest SERT occupancy, and near saturation of the 5HT2A receptor.^15,22 Dopamine D2 affinity is modest (32 nM), and the D2 receptor occupancy at the 42 mg dose is low. These properties translate to rates of EPS in clinical studies of schizophrenia and BD that are close to that of placebo. Lumateperone has very high affinity for serotonin 5HT2A receptors (Ki 0.54 nM), which also helps mitigate D2-related adverse effects and may be part of the therapeutic antidepressant mechanism. Underlying the tolerability profile is the low affinity for alpha 1-adrenergic receptors (Ki 73 nM), muscarinic and histaminergic receptors (Ki >100 nM for both).

Clinical considerations

Data from the lumateperone BD depression trials led to it being only the second agent approved for acute major depression in BD II patients, and the only agent which has approvals as monotherapy and adjunctive therapy for both BD subtypes. The monotherapy trial results substantiate that lumateperone was robustly effective regardless of BD subtype, with significant improvement in depressive symptoms experienced by patients with BD I (effect size 0.49, P < .0001) and those with BD II (effect size 0.81, P < .001). Effect sizes in acute BD depression studies are much larger in monotherapy trials than in adjunctive trials, as the latter group represents patients who have already failed pretreatment with a mood stabilizer.^25,26 In the lurasidone BD I depression trials, the effect size based on mean change in MADRS score over the course of 6 weeks was 0.51 in the monotherapy study compared to 0.34 when used adjunctively with lithium or VPA.^25,26 In the lumateperone adjunctive study, the effect size for the difference in mean MADRS total score from baseline for lumateperone 42 mg/d, was 0.27 (P < .05). Subgroup analyses by BD subtype are not yet available for adjunctive use, but the data presented to FDA were sufficient to permit an indication for adjunctive use across both diagnostic groups.

The absence of clinically significant EPS, the minimal impact on metabolic or endocrine parameters, and the lack of a need for titration are all appealing properties. At the present there is only 1 marketed dose (42 mg capsules), so the package insert includes cautionary language regarding situations when a patient might encounter less drug exposure (concurrent use of cytochrome P450 [CYP] 3A4 inducers), or greater drug exposure due to concurrent use of moderate or strong CYP3A4 inhibitors, as well as in patients with moderate or severe hepatic impairment as defined by Child-Pugh Criteria (Child-Pugh B or C). These are not contraindications.

Unique properties of lumateperone include efficacy established as monotherapy for BD I and BD II patients, and efficacy for adjunctive use with lithium or VPA. Additionally, the extremely low rates of significant EPS and lack of clinically significant metabolic or endocrine adverse effects are unique properties of lumateperone.¹³

Why Rx? Reasons to prescribe lumateperone for adult BD depression patients include:

• data support efficacy for BD I and BD II patients, and for monotherapy or adjunctive use with lithium/VPA
• favorable tolerability profile, including no significant signal for EPS, endocrine or metabolic adverse effects, or QT prolongation
• no need for titration.

Dosing. There is only 1 dose available for lumateperone: 42 mg capsules (Table 3). As the dose cannot be modified, the package insert contains cautionary language regarding situations with less drug exposure (use of CYP3A4 inducers), or greater drug exposure (use with moderate or strong CYP3A4 inhibitors or in patients with moderate or severe hepatic impairment as defined by Child-Pugh Criteria [Child-Pugh B or C]). These are not contraindications. Based on newer pharmacokinetic studies, lumateperone does not need to be dosed with food, and there is no clinically significant interaction with UGT1A4 inhibitors such as VPA.

Bottom Line

Data support the efficacy of lumateperone for treating depressive episodes in adults with bipolar I or bipolar II disorder, either as monotherapy or adjunctive to lithium or divalproex/valproate. Potential advantages of lumateperone for this indication include a favorable tolerability profile and no need for titration.

Among patients with bipolar I or II disorder (BD I or II), major depressive episodes represent the predominant mood state when not euthymic, and are disproportionately associated with the functional disability of BD and its suicide risk.¹ Long-term naturalistic studies of weekly mood states in patients with BD I or II found that the proportion of time spent depressed greatly exceeded that spent in a mixed, hypomanic, or manic state during >12 years of follow-up (Figure 1²and Figure 2³). In the 20th century, traditional antidepressants represented the sole option for management of bipolar depression despite concerns of manic switching or lack of efficacy.^4,5 Efficacy concerns were subsequently confirmed by placebo-controlled studies, such as the Systematic Treatment Enhancement Program for Bipolar Disorder (STEP-BD) trial, which found limited effectiveness of adjunctive antidepressants for bipolar depression.⁶ Comprehensive reviews of randomized controlled trials and observational studies documented the risk of mood cycling and manic switching, especially in patients with BD I, even if antidepressants were used in the presence of mood-stabilizing medications.^7,8

Several newer antipsychotics have been FDA-approved for treating depressive episodes associated with BD (Table 1). Approval of olanzapine/fluoxetine combination (OFC) in December 2003 for depressive episodes associated with BD I established that mechanisms exist which can effectively treat acute depressive episodes in patients with BD without an inordinate risk of mood instability. Subsequent approval of quetia­pine in October 2006 for depression associated with BD I or II, lurasidone in June 2013, and cariprazine in May 2019 for major depression in BD I greatly expanded the options for management of acute bipolar depression. However, despite the array of molecules available, for certain patients these agents presented tolerability issues such as sedation, weight gain, akathisia, or parkinsonism that could hamper effective treatment.⁹ Safety and efficacy data in bipolar depression for adjunctive use with lithium or divalproex/valproate (VPA) also are lacking for quetiapine, OFC, and cariprazine.^10,11 Moreover, despite the fact that BD II is as prevalent as BD I, and that patients with BD II have comparable rates of comorbidity, chronicity, disability, and suicidality,¹² only quetiapine was approved for acute treatment of depression in patients with BD II. This omission is particularly problematic because the depressive episodes of BD II predominate over the time spent in hypomanic and cycling/mixed states (50.3% for depression vs 3.6% for hypomania/cycling/mixed combined), much more than is seen with BD I (31.9% for depression vs 14.8% for hypomania/cycling/mixed combined).^2,3 The paucity of data for the use of newer antipsychotics in BD II depression presents a problem when patients cannot tolerate or refuse to consider quetiapine. This prevents clinicians from engaging in evidence-based efficacy discussions of other options, even if it is assumed that the tolerability profile for BD II depression patients may be similar to that seen when these agents are used for BD I depression.

Continue to: Table 1...

Lumateperone (Caplyta) is a novel oral antipsychotic initially approved in 2019 for the treatment of adult patients with schizophrenia. It was approved in December 2021 for the management of depression associated with BD I or II in adults as monotherapy or when used adjunctively with the mood stabilizers lithium or VPA (Table 2).¹³ Lumateperone possesses certain binding affinities not unlike those in other newer antipsychotics, including high affinity for serotonin 5HT2A receptors (Ki 0.54 nM), low affinity for dopamine D2 receptors (Ki 32 nM), and low affinity for alpha 1-adrenergic receptors (Ki 73 nM), muscarinic and histaminergic receptors (Ki >100 nM for both).^13,14 However, there are some distinguishing features: the ratio of 5HT2A receptor affinity to D2 affinity is 60, greater than that for other second-generation antipsychotics (SGAs) such as risperidone (12), olanzapine (12.4) or aripiprazole (0.18).¹⁵ At steady state, D2 receptor occupancy remains <40%, and the corresponding rates of extrapyramidal side effects (EPS)/akathisia differed by only 0.4% for lumateperone vs placebo in short-term adult clinical schizophrenia trials,^13,16 by 0.2% for lumateperone vs placebo in the monotherapy BD depression study, and by 1.7% in the adjunctive BD depression study.^13,17,18 Lumateperone also exhibited no clinically significant impact on metabolic measures or serum prolactin during the 4-week schizophrenia trials, with mean weight gain ≤1 kg for the 42 mg dose across all studies.¹⁹ This favorable tolerability profile for endocrine and metabolic adverse effects was also seen in the BD depression studies. Across the 2 BD depression monotherapy trials and the single adjunctive study, the only adverse reactions occurring in ≥5% of lumateperone-treated patients and more than twice the rate of placebo were somnolence/sedation, dizziness, nausea, and dry mouth.¹³ There was also no single adverse reaction leading to discontinuation in the BD depression studies that occurred at a rate >2% in patients treated with lumateperone.¹³

Continue to: Clinical implications...

Clinical implications

The approval of lumateperone for both BD I and BD II depression, and for its use as monotherapy and for adjunctive use with lithium or VPA, satisfies several unmet needs for the management of acute major depressive episodes in patients with BD. Clinicians now have both safety and tolerability data to present to their bipolar spectrum patients regardless of subtype, and regardless of whether the patient requires mood stabilizer therapy. The tolerability advantages for lumateperone seen in schizophrenia trials were replicated in a diagnostic group that is very sensitive to D2-related adverse effects, and for whom any signal of clinically significant weight gain or sedation often represents an insuperable barrier to patient acceptance.²³

Efficacy in adults with BD I or II depression.

The efficacy of lumateperone for major depressive episodes has been established in 2 pivotal, double-blind, placebo-controlled trials in BD I or II patients: 1 monotherapy study,¹⁷ and 1 study when used adjunctively to lithium or VPA.¹⁸ The first study was a 6-week, double-blind, placebo-controlled monotherapy trial (study 404) in which 377 patients age 18 to 75 with BD I or BD II experiencing a major depressive episode were randomized in a 1:1 manner to lumateperone 42 mg/d or placebo given once daily in the evening. Symptom entry criteria included a Montgomery-Åsberg Depression Rating Scale (MADRS) total score ≥20, and scores ≥4 on the depression and overall BD illness subscales of the Clinical Global Impressions Scale–Bipolar Version Severity scale (CGI-BP-S) at screening and at baseline.¹⁷ Study entry also required a score ≤12 on the Young Mania Rating Scale (YMRS) at screening and at baseline. The duration of the major depressive episode must have been ≥2 weeks but <6 months before screening, with symptoms causing clinically significant distress or functional impairment. The primary outcome measure was change from baseline in MADRS. Several secondary efficacy measures were examined, including the proportion of patients meeting criteria for treatment response (≥50% decrease in MADRS), or remission (MADRS score ≤12), and differential changes in MADRS scores from baseline for BD I and BD II subgroups.¹⁷

The patient population was 58% female and 91% White, with 79.9% diagnosed as BD I and 20.1% as BD II. The least squares mean changes on the MADRS total score from baseline to Day 43 were lumateperone 42 mg/d: -16.7 points; placebo: -12.1 points (P < .0001), and the effect size for this difference was moderate: 0.56. Secondary analyses indicated that 51.1% of those taking lumateperone 42 mg/d and 36.7% taking placebo met response criteria (P < .001), while 39.9% of those taking lumateperone 42 mg/d and 33.5% taking placebo met remission criteria (P = .018). Importantly, depression improvement was observed both in patients with BD I (effect size 0.49, P < .0001) and in those with BD II (effect size 0.81, P < .001).

The second pivotal trial (study 402) was a 6-week, double-blind, placebo-controlled adjunctive trial in which 528 patients age 18 to 75 with BD I or BD II experiencing a major depressive episode despite treatment with lithium or VPA were randomized in a 1:1:1 manner to lumateperone 28 mg/d, lumateperone 42 mg/d, or placebo given once daily in the evening.¹⁸ Like the monotherapy trial, symptom entry criteria included a MADRS total score ≥20, and scores ≥4 on the depression and overall illness CGI-BP-S subscales at screening and baseline.¹⁸ Study entry also required a score ≤12 on the YMRS at screening and baseline. The duration of the major depressive episode must have been ≥2 weeks but <6 months before screening, with symptoms causing clinically significant distress or functional impairment. The primary outcome measure was change from baseline in MADRS for lumateperone 42 mg/d compared to placebo. Secondary efficacy measures included MADRS changes for lumateperone 28 mg/d and the proportion of patients meeting criteria for treatment response (≥50% decrease in MADRS) or remission (MADRS score ≤12).

The patient population was 58% female and 88% White, with 83.3% diagnosed as BD I, 16.7% diagnosed as BD II, and 28.6% treated with lithium vs 71.4% on VPA. The effect size for the difference in MADRS total score from baseline to Day 43 for lumateperone 42 mg/d was 0.27 (P < .05), while that for the lumateperone 28 mg/d dose did not reach statistical significance. Secondary analyses indicated that response rates for lumateperone 28 mg/d and lumateperone 42 mg/d were significantly higher than for placebo (both P < .05). Response rates were placebo: 39%; lumateperone 28 mg/d: 50%; and lumateperone 42 mg/d: 45%. Remission rates were similar at Day 43 in both lumateperone groups compared with placebo: placebo: 31%, lumateperone 28 mg/d: 31%, and lumateperone 42 mg/d: 28%.¹⁸ As of this writing, a secondary analysis by BD subtype has not yet been presented.

A third study examining lumateperone monotherapy failed to establish superiority of lumateperone over placebo (NCT02600494). The data regarding tolerability from that study were incorporated in product labeling describing adverse reactions.

Continue on to: Adverse reactions...

Adverse reactions

In the positive monotherapy trial, there were 376 patients in the modified intent-to-treat efficacy population to receive lumateperone (N = 188) or placebo (N = 188) with nearly identical completion rates in the active treatment and placebo cohorts: lumateperone, 88.8%; placebo, 88.3%.¹⁷ The proportion experiencing mania was low in both cohorts (lumateperone, 1.1%; placebo, 2.1%), and there was 1 case of hypomania in each group. One participant in the lumateperone group and 1 in the placebo group discontinued the study due to a serious adverse event of mania. There was no worsening of mania in either group as measured by mean change in the YMRS score. There was also no suicidal behavior in either cohort during the study. Pooling the 2 monotherapy trials, the adverse events that occurred at ≥5% in lumateperone-treated patients and at more than twice the rate of the placebo group were somnolence/sedation (lumateperone 42 mg/d: 13%, placebo: 3%), dizziness (lumateperone 42 mg/d: 8%, placebo: 4%), and nausea (lumateperone 42 mg/d: 8%, placebo: 3%).¹³ Rates of EPS were low for both groups: lumateperone 42 mg/d: 1.3%, placebo: 1.1%.¹³ Mean weight change at Day 43 was +0.11 kg for lumateperone and +0.03 kg for placebo in the positive monotherapy trial.¹⁷ Moreover, compared to placebo, lumateperone exhibited comparable effects on serum prolactin and all metabolic parameters, including fasting insulin, fasting glucose, and fasting lipids, none of which were clinically significant. No patient exhibited a corrected QT interval >500 ms at any time, and increases ≥60 ms from baseline were similar between the lumateperone (n = 1, 0.6%) and placebo (n = 3, 1.8%) cohorts.

Complete safety and tolerability data for the adjunctive trial has not yet been published, but discontinuation rates due to treatment-emergent adverse effects for the 3 arms were: lumateperone 42 mg/d: 5.6%; lumateperone 28 mg/d: 1.7%; and placebo: 1.7%. Overall, 81.4% of patients completed the trial, with only 1 serious adverse event (lithium toxicity) occurring in a patient taking lumateperone 42 mg/d. While this led to study discontinuation, it was not considered related to lumateperone exposure by the investigator. There was no worsening of mania in either lumateperone dosage group or the placebo cohort as measured by mean change in YMRS score: -1.2 for placebo, -1.4 for lumateperone 28 mg/d, and -1.6 for lumateperone 42 mg/d. Suicidal behavior was not observed in any group during treatment. The adverse events that occurred at rates ≥5% in lumateperone-treated patients and at more than twice the rate of the placebo group were somnolence/sedation (lumateperone, 13%; placebo, 3%), dizziness (lumateperone, 11%; placebo, 2%), and nausea (lumateperone, 9%; placebo, 4%).¹³ Rates of EPS were low for both groups: lumateperone, 4.0%, placebo, 2.3%.¹³ Mean weight changes at Day 43 were +0.23 kg for placebo, +0.02 kg for lumateperone 28 mg/d, and 0.00 kg for lumateperone 42 mg/d.¹⁸ Compared to placebo, both doses of lumateperone exhibited comparable effects on serum prolactin and all metabolic parameters, including fasting insulin, fasting glucose, and fasting lipids, none of which were clinically significant.¹⁸

Lastly, the package insert notes that in an uncontrolled, open-label trial of lumateperone for up to 6 months in patients with BD depression, the mean weight change was -0.01 ± 3.1 kg at Day 175.¹³

Continue on to: Pharmacologic profile...

Pharmacologic profile

Lumateperone’s preclinical discovery program found an impact on markers associated with increased glutamatergic neurotransmission, properties that were predicted to yield antidepressant benefit.^14,15,24 This is hypothesized to be based on the complex pharmacology of lumateperone, including dopamine D1 agonism, modest SERT occupancy, and near saturation of the 5HT2A receptor.^15,22 Dopamine D2 affinity is modest (32 nM), and the D2 receptor occupancy at the 42 mg dose is low. These properties translate to rates of EPS in clinical studies of schizophrenia and BD that are close to that of placebo. Lumateperone has very high affinity for serotonin 5HT2A receptors (Ki 0.54 nM), which also helps mitigate D2-related adverse effects and may be part of the therapeutic antidepressant mechanism. Underlying the tolerability profile is the low affinity for alpha 1-adrenergic receptors (Ki 73 nM), muscarinic and histaminergic receptors (Ki >100 nM for both).

Clinical considerations

Data from the lumateperone BD depression trials led to it being only the second agent approved for acute major depression in BD II patients, and the only agent which has approvals as monotherapy and adjunctive therapy for both BD subtypes. The monotherapy trial results substantiate that lumateperone was robustly effective regardless of BD subtype, with significant improvement in depressive symptoms experienced by patients with BD I (effect size 0.49, P < .0001) and those with BD II (effect size 0.81, P < .001). Effect sizes in acute BD depression studies are much larger in monotherapy trials than in adjunctive trials, as the latter group represents patients who have already failed pretreatment with a mood stabilizer.^25,26 In the lurasidone BD I depression trials, the effect size based on mean change in MADRS score over the course of 6 weeks was 0.51 in the monotherapy study compared to 0.34 when used adjunctively with lithium or VPA.^25,26 In the lumateperone adjunctive study, the effect size for the difference in mean MADRS total score from baseline for lumateperone 42 mg/d, was 0.27 (P < .05). Subgroup analyses by BD subtype are not yet available for adjunctive use, but the data presented to FDA were sufficient to permit an indication for adjunctive use across both diagnostic groups.

The absence of clinically significant EPS, the minimal impact on metabolic or endocrine parameters, and the lack of a need for titration are all appealing properties. At the present there is only 1 marketed dose (42 mg capsules), so the package insert includes cautionary language regarding situations when a patient might encounter less drug exposure (concurrent use of cytochrome P450 [CYP] 3A4 inducers), or greater drug exposure due to concurrent use of moderate or strong CYP3A4 inhibitors, as well as in patients with moderate or severe hepatic impairment as defined by Child-Pugh Criteria (Child-Pugh B or C). These are not contraindications.

Unique properties of lumateperone include efficacy established as monotherapy for BD I and BD II patients, and efficacy for adjunctive use with lithium or VPA. Additionally, the extremely low rates of significant EPS and lack of clinically significant metabolic or endocrine adverse effects are unique properties of lumateperone.¹³

Why Rx? Reasons to prescribe lumateperone for adult BD depression patients include:

• data support efficacy for BD I and BD II patients, and for monotherapy or adjunctive use with lithium/VPA
• favorable tolerability profile, including no significant signal for EPS, endocrine or metabolic adverse effects, or QT prolongation
• no need for titration.

Dosing. There is only 1 dose available for lumateperone: 42 mg capsules (Table 3). As the dose cannot be modified, the package insert contains cautionary language regarding situations with less drug exposure (use of CYP3A4 inducers), or greater drug exposure (use with moderate or strong CYP3A4 inhibitors or in patients with moderate or severe hepatic impairment as defined by Child-Pugh Criteria [Child-Pugh B or C]). These are not contraindications. Based on newer pharmacokinetic studies, lumateperone does not need to be dosed with food, and there is no clinically significant interaction with UGT1A4 inhibitors such as VPA.

Bottom Line

Data support the efficacy of lumateperone for treating depressive episodes in adults with bipolar I or bipolar II disorder, either as monotherapy or adjunctive to lithium or divalproex/valproate. Potential advantages of lumateperone for this indication include a favorable tolerability profile and no need for titration.

Among patients with bipolar I or II disorder (BD I or II), major depressive episodes represent the predominant mood state when not euthymic, and are disproportionately associated with the functional disability of BD and its suicide risk.¹ Long-term naturalistic studies of weekly mood states in patients with BD I or II found that the proportion of time spent depressed greatly exceeded that spent in a mixed, hypomanic, or manic state during >12 years of follow-up (Figure 1²and Figure 2³). In the 20th century, traditional antidepressants represented the sole option for management of bipolar depression despite concerns of manic switching or lack of efficacy.^4,5 Efficacy concerns were subsequently confirmed by placebo-controlled studies, such as the Systematic Treatment Enhancement Program for Bipolar Disorder (STEP-BD) trial, which found limited effectiveness of adjunctive antidepressants for bipolar depression.⁶ Comprehensive reviews of randomized controlled trials and observational studies documented the risk of mood cycling and manic switching, especially in patients with BD I, even if antidepressants were used in the presence of mood-stabilizing medications.^7,8

Several newer antipsychotics have been FDA-approved for treating depressive episodes associated with BD (Table 1). Approval of olanzapine/fluoxetine combination (OFC) in December 2003 for depressive episodes associated with BD I established that mechanisms exist which can effectively treat acute depressive episodes in patients with BD without an inordinate risk of mood instability. Subsequent approval of quetia­pine in October 2006 for depression associated with BD I or II, lurasidone in June 2013, and cariprazine in May 2019 for major depression in BD I greatly expanded the options for management of acute bipolar depression. However, despite the array of molecules available, for certain patients these agents presented tolerability issues such as sedation, weight gain, akathisia, or parkinsonism that could hamper effective treatment.⁹ Safety and efficacy data in bipolar depression for adjunctive use with lithium or divalproex/valproate (VPA) also are lacking for quetiapine, OFC, and cariprazine.^10,11 Moreover, despite the fact that BD II is as prevalent as BD I, and that patients with BD II have comparable rates of comorbidity, chronicity, disability, and suicidality,¹² only quetiapine was approved for acute treatment of depression in patients with BD II. This omission is particularly problematic because the depressive episodes of BD II predominate over the time spent in hypomanic and cycling/mixed states (50.3% for depression vs 3.6% for hypomania/cycling/mixed combined), much more than is seen with BD I (31.9% for depression vs 14.8% for hypomania/cycling/mixed combined).^2,3 The paucity of data for the use of newer antipsychotics in BD II depression presents a problem when patients cannot tolerate or refuse to consider quetiapine. This prevents clinicians from engaging in evidence-based efficacy discussions of other options, even if it is assumed that the tolerability profile for BD II depression patients may be similar to that seen when these agents are used for BD I depression.

Continue to: Table 1...

Lumateperone (Caplyta) is a novel oral antipsychotic initially approved in 2019 for the treatment of adult patients with schizophrenia. It was approved in December 2021 for the management of depression associated with BD I or II in adults as monotherapy or when used adjunctively with the mood stabilizers lithium or VPA (Table 2).¹³ Lumateperone possesses certain binding affinities not unlike those in other newer antipsychotics, including high affinity for serotonin 5HT2A receptors (Ki 0.54 nM), low affinity for dopamine D2 receptors (Ki 32 nM), and low affinity for alpha 1-adrenergic receptors (Ki 73 nM), muscarinic and histaminergic receptors (Ki >100 nM for both).^13,14 However, there are some distinguishing features: the ratio of 5HT2A receptor affinity to D2 affinity is 60, greater than that for other second-generation antipsychotics (SGAs) such as risperidone (12), olanzapine (12.4) or aripiprazole (0.18).¹⁵ At steady state, D2 receptor occupancy remains <40%, and the corresponding rates of extrapyramidal side effects (EPS)/akathisia differed by only 0.4% for lumateperone vs placebo in short-term adult clinical schizophrenia trials,^13,16 by 0.2% for lumateperone vs placebo in the monotherapy BD depression study, and by 1.7% in the adjunctive BD depression study.^13,17,18 Lumateperone also exhibited no clinically significant impact on metabolic measures or serum prolactin during the 4-week schizophrenia trials, with mean weight gain ≤1 kg for the 42 mg dose across all studies.¹⁹ This favorable tolerability profile for endocrine and metabolic adverse effects was also seen in the BD depression studies. Across the 2 BD depression monotherapy trials and the single adjunctive study, the only adverse reactions occurring in ≥5% of lumateperone-treated patients and more than twice the rate of placebo were somnolence/sedation, dizziness, nausea, and dry mouth.¹³ There was also no single adverse reaction leading to discontinuation in the BD depression studies that occurred at a rate >2% in patients treated with lumateperone.¹³

Continue to: Clinical implications...

Clinical implications

The approval of lumateperone for both BD I and BD II depression, and for its use as monotherapy and for adjunctive use with lithium or VPA, satisfies several unmet needs for the management of acute major depressive episodes in patients with BD. Clinicians now have both safety and tolerability data to present to their bipolar spectrum patients regardless of subtype, and regardless of whether the patient requires mood stabilizer therapy. The tolerability advantages for lumateperone seen in schizophrenia trials were replicated in a diagnostic group that is very sensitive to D2-related adverse effects, and for whom any signal of clinically significant weight gain or sedation often represents an insuperable barrier to patient acceptance.²³

Efficacy in adults with BD I or II depression.

The efficacy of lumateperone for major depressive episodes has been established in 2 pivotal, double-blind, placebo-controlled trials in BD I or II patients: 1 monotherapy study,¹⁷ and 1 study when used adjunctively to lithium or VPA.¹⁸ The first study was a 6-week, double-blind, placebo-controlled monotherapy trial (study 404) in which 377 patients age 18 to 75 with BD I or BD II experiencing a major depressive episode were randomized in a 1:1 manner to lumateperone 42 mg/d or placebo given once daily in the evening. Symptom entry criteria included a Montgomery-Åsberg Depression Rating Scale (MADRS) total score ≥20, and scores ≥4 on the depression and overall BD illness subscales of the Clinical Global Impressions Scale–Bipolar Version Severity scale (CGI-BP-S) at screening and at baseline.¹⁷ Study entry also required a score ≤12 on the Young Mania Rating Scale (YMRS) at screening and at baseline. The duration of the major depressive episode must have been ≥2 weeks but <6 months before screening, with symptoms causing clinically significant distress or functional impairment. The primary outcome measure was change from baseline in MADRS. Several secondary efficacy measures were examined, including the proportion of patients meeting criteria for treatment response (≥50% decrease in MADRS), or remission (MADRS score ≤12), and differential changes in MADRS scores from baseline for BD I and BD II subgroups.¹⁷

The patient population was 58% female and 91% White, with 79.9% diagnosed as BD I and 20.1% as BD II. The least squares mean changes on the MADRS total score from baseline to Day 43 were lumateperone 42 mg/d: -16.7 points; placebo: -12.1 points (P < .0001), and the effect size for this difference was moderate: 0.56. Secondary analyses indicated that 51.1% of those taking lumateperone 42 mg/d and 36.7% taking placebo met response criteria (P < .001), while 39.9% of those taking lumateperone 42 mg/d and 33.5% taking placebo met remission criteria (P = .018). Importantly, depression improvement was observed both in patients with BD I (effect size 0.49, P < .0001) and in those with BD II (effect size 0.81, P < .001).

The second pivotal trial (study 402) was a 6-week, double-blind, placebo-controlled adjunctive trial in which 528 patients age 18 to 75 with BD I or BD II experiencing a major depressive episode despite treatment with lithium or VPA were randomized in a 1:1:1 manner to lumateperone 28 mg/d, lumateperone 42 mg/d, or placebo given once daily in the evening.¹⁸ Like the monotherapy trial, symptom entry criteria included a MADRS total score ≥20, and scores ≥4 on the depression and overall illness CGI-BP-S subscales at screening and baseline.¹⁸ Study entry also required a score ≤12 on the YMRS at screening and baseline. The duration of the major depressive episode must have been ≥2 weeks but <6 months before screening, with symptoms causing clinically significant distress or functional impairment. The primary outcome measure was change from baseline in MADRS for lumateperone 42 mg/d compared to placebo. Secondary efficacy measures included MADRS changes for lumateperone 28 mg/d and the proportion of patients meeting criteria for treatment response (≥50% decrease in MADRS) or remission (MADRS score ≤12).

The patient population was 58% female and 88% White, with 83.3% diagnosed as BD I, 16.7% diagnosed as BD II, and 28.6% treated with lithium vs 71.4% on VPA. The effect size for the difference in MADRS total score from baseline to Day 43 for lumateperone 42 mg/d was 0.27 (P < .05), while that for the lumateperone 28 mg/d dose did not reach statistical significance. Secondary analyses indicated that response rates for lumateperone 28 mg/d and lumateperone 42 mg/d were significantly higher than for placebo (both P < .05). Response rates were placebo: 39%; lumateperone 28 mg/d: 50%; and lumateperone 42 mg/d: 45%. Remission rates were similar at Day 43 in both lumateperone groups compared with placebo: placebo: 31%, lumateperone 28 mg/d: 31%, and lumateperone 42 mg/d: 28%.¹⁸ As of this writing, a secondary analysis by BD subtype has not yet been presented.

A third study examining lumateperone monotherapy failed to establish superiority of lumateperone over placebo (NCT02600494). The data regarding tolerability from that study were incorporated in product labeling describing adverse reactions.

Continue on to: Adverse reactions...

Adverse reactions

In the positive monotherapy trial, there were 376 patients in the modified intent-to-treat efficacy population to receive lumateperone (N = 188) or placebo (N = 188) with nearly identical completion rates in the active treatment and placebo cohorts: lumateperone, 88.8%; placebo, 88.3%.¹⁷ The proportion experiencing mania was low in both cohorts (lumateperone, 1.1%; placebo, 2.1%), and there was 1 case of hypomania in each group. One participant in the lumateperone group and 1 in the placebo group discontinued the study due to a serious adverse event of mania. There was no worsening of mania in either group as measured by mean change in the YMRS score. There was also no suicidal behavior in either cohort during the study. Pooling the 2 monotherapy trials, the adverse events that occurred at ≥5% in lumateperone-treated patients and at more than twice the rate of the placebo group were somnolence/sedation (lumateperone 42 mg/d: 13%, placebo: 3%), dizziness (lumateperone 42 mg/d: 8%, placebo: 4%), and nausea (lumateperone 42 mg/d: 8%, placebo: 3%).¹³ Rates of EPS were low for both groups: lumateperone 42 mg/d: 1.3%, placebo: 1.1%.¹³ Mean weight change at Day 43 was +0.11 kg for lumateperone and +0.03 kg for placebo in the positive monotherapy trial.¹⁷ Moreover, compared to placebo, lumateperone exhibited comparable effects on serum prolactin and all metabolic parameters, including fasting insulin, fasting glucose, and fasting lipids, none of which were clinically significant. No patient exhibited a corrected QT interval >500 ms at any time, and increases ≥60 ms from baseline were similar between the lumateperone (n = 1, 0.6%) and placebo (n = 3, 1.8%) cohorts.

Complete safety and tolerability data for the adjunctive trial has not yet been published, but discontinuation rates due to treatment-emergent adverse effects for the 3 arms were: lumateperone 42 mg/d: 5.6%; lumateperone 28 mg/d: 1.7%; and placebo: 1.7%. Overall, 81.4% of patients completed the trial, with only 1 serious adverse event (lithium toxicity) occurring in a patient taking lumateperone 42 mg/d. While this led to study discontinuation, it was not considered related to lumateperone exposure by the investigator. There was no worsening of mania in either lumateperone dosage group or the placebo cohort as measured by mean change in YMRS score: -1.2 for placebo, -1.4 for lumateperone 28 mg/d, and -1.6 for lumateperone 42 mg/d. Suicidal behavior was not observed in any group during treatment. The adverse events that occurred at rates ≥5% in lumateperone-treated patients and at more than twice the rate of the placebo group were somnolence/sedation (lumateperone, 13%; placebo, 3%), dizziness (lumateperone, 11%; placebo, 2%), and nausea (lumateperone, 9%; placebo, 4%).¹³ Rates of EPS were low for both groups: lumateperone, 4.0%, placebo, 2.3%.¹³ Mean weight changes at Day 43 were +0.23 kg for placebo, +0.02 kg for lumateperone 28 mg/d, and 0.00 kg for lumateperone 42 mg/d.¹⁸ Compared to placebo, both doses of lumateperone exhibited comparable effects on serum prolactin and all metabolic parameters, including fasting insulin, fasting glucose, and fasting lipids, none of which were clinically significant.¹⁸

Lastly, the package insert notes that in an uncontrolled, open-label trial of lumateperone for up to 6 months in patients with BD depression, the mean weight change was -0.01 ± 3.1 kg at Day 175.¹³

Continue on to: Pharmacologic profile...

Pharmacologic profile

Lumateperone’s preclinical discovery program found an impact on markers associated with increased glutamatergic neurotransmission, properties that were predicted to yield antidepressant benefit.^14,15,24 This is hypothesized to be based on the complex pharmacology of lumateperone, including dopamine D1 agonism, modest SERT occupancy, and near saturation of the 5HT2A receptor.^15,22 Dopamine D2 affinity is modest (32 nM), and the D2 receptor occupancy at the 42 mg dose is low. These properties translate to rates of EPS in clinical studies of schizophrenia and BD that are close to that of placebo. Lumateperone has very high affinity for serotonin 5HT2A receptors (Ki 0.54 nM), which also helps mitigate D2-related adverse effects and may be part of the therapeutic antidepressant mechanism. Underlying the tolerability profile is the low affinity for alpha 1-adrenergic receptors (Ki 73 nM), muscarinic and histaminergic receptors (Ki >100 nM for both).

Clinical considerations

Data from the lumateperone BD depression trials led to it being only the second agent approved for acute major depression in BD II patients, and the only agent which has approvals as monotherapy and adjunctive therapy for both BD subtypes. The monotherapy trial results substantiate that lumateperone was robustly effective regardless of BD subtype, with significant improvement in depressive symptoms experienced by patients with BD I (effect size 0.49, P < .0001) and those with BD II (effect size 0.81, P < .001). Effect sizes in acute BD depression studies are much larger in monotherapy trials than in adjunctive trials, as the latter group represents patients who have already failed pretreatment with a mood stabilizer.^25,26 In the lurasidone BD I depression trials, the effect size based on mean change in MADRS score over the course of 6 weeks was 0.51 in the monotherapy study compared to 0.34 when used adjunctively with lithium or VPA.^25,26 In the lumateperone adjunctive study, the effect size for the difference in mean MADRS total score from baseline for lumateperone 42 mg/d, was 0.27 (P < .05). Subgroup analyses by BD subtype are not yet available for adjunctive use, but the data presented to FDA were sufficient to permit an indication for adjunctive use across both diagnostic groups.

The absence of clinically significant EPS, the minimal impact on metabolic or endocrine parameters, and the lack of a need for titration are all appealing properties. At the present there is only 1 marketed dose (42 mg capsules), so the package insert includes cautionary language regarding situations when a patient might encounter less drug exposure (concurrent use of cytochrome P450 [CYP] 3A4 inducers), or greater drug exposure due to concurrent use of moderate or strong CYP3A4 inhibitors, as well as in patients with moderate or severe hepatic impairment as defined by Child-Pugh Criteria (Child-Pugh B or C). These are not contraindications.

Unique properties of lumateperone include efficacy established as monotherapy for BD I and BD II patients, and efficacy for adjunctive use with lithium or VPA. Additionally, the extremely low rates of significant EPS and lack of clinically significant metabolic or endocrine adverse effects are unique properties of lumateperone.¹³

Why Rx? Reasons to prescribe lumateperone for adult BD depression patients include:

• data support efficacy for BD I and BD II patients, and for monotherapy or adjunctive use with lithium/VPA
• favorable tolerability profile, including no significant signal for EPS, endocrine or metabolic adverse effects, or QT prolongation
• no need for titration.

Dosing. There is only 1 dose available for lumateperone: 42 mg capsules (Table 3). As the dose cannot be modified, the package insert contains cautionary language regarding situations with less drug exposure (use of CYP3A4 inducers), or greater drug exposure (use with moderate or strong CYP3A4 inhibitors or in patients with moderate or severe hepatic impairment as defined by Child-Pugh Criteria [Child-Pugh B or C]). These are not contraindications. Based on newer pharmacokinetic studies, lumateperone does not need to be dosed with food, and there is no clinically significant interaction with UGT1A4 inhibitors such as VPA.

Bottom Line

Data support the efficacy of lumateperone for treating depressive episodes in adults with bipolar I or bipolar II disorder, either as monotherapy or adjunctive to lithium or divalproex/valproate. Potential advantages of lumateperone for this indication include a favorable tolerability profile and no need for titration.

Antipsychotic nonadherence is a known contributor to relapse risk among patients with schizophrenia.¹ Because relapse episodes may be associated with antipsychotic treatment resistance, this must be avoided as much as possible by appropriate medication selection.² Adverse effect burden is an important factor leading to oral antipsychotic nonadherence, with patient-derived data indicating that extrapyramidal symptoms (EPS) (odds ratio [OR] 0.57, P = .0007), sedation/cognitive adverse effects (OR 0.70, P = .033), prolactin/endocrine effects (OR 0.69, P = .0342), and metabolic adverse effects (OR 0.64, P = .0079) are all significantly related to lower rates of adherence.³ With this in mind, successive generations of antipsychotics have been released, with fewer tolerability issues present than seen with earlier compounds.^1,4 Although these newer second-generation antipsychotics (SGAs) have not proven more effective for schizophrenia than those first marketed in the 1990s, they generally possess lower rates of EPS, hyperprolactinemia, anticholinergic and antihistaminic properties, metabolic adverse effects, and orthostasis.⁵ This improved adverse effect profile will hopefully increase the chances of antipsychotic acceptance in patients with schizophrenia, and thereby promote improved adherence.

Lumateperone (Caplyta) is a novel oral antipsychotic approved for the treatment of adult patients with schizophrenia (Table 1). It possesses some properties seen with other SGAs, including high affinity for serotonin 5HT2A receptors (Ki 0.54 nM) and lower affinity for dopamine D2 receptors (Ki 32 nM), along with low affinity for alpha1-adrenergic receptors (Ki 73 nM), and muscarinic and histaminergic receptors (Ki > 100 nM).^6,7 However, there are some distinguishing features: the ratio of 5HT2A receptor affinity to D2 affinity is 60, greater than that of other SGAs such as risperidone (12), olanzapine (12.4) or aripiprazole (0.18)⁸; at steady state, the D2 occupancy remains <40% (Figure) and the corresponding rates of EPS/akathisia were only 6.7% for lumateperone vs 6.3% for placebo in short-term clinical trials.^7,9

How it works

A unique aspect of lumateperone’s pharmacology may relate to differential actions at presynaptic and postsynaptic dopamine D2 receptors. Other antipsychotics possess comparable antagonist (or partial agonist) properties at postsynaptic D2 receptors (the D2 long isoform) and the presynaptic autoreceptor (the D2 short isoform). By blocking the presynaptic autoreceptor, feedback inhibition on dopamine release is removed; therefore, the required higher levels of postsynaptic D2 receptor occupancy needed for effective antipsychotic action (eg, 60% to 80% for antagonists, and 83% to 100% for partial agonists) may be a product of the need to oppose this increased presynaptic release of dopamine. In vitro assays show that lumateperone does not increase presynaptic dopamine release, indicating that it possesses agonist properties at the presynaptic D2 short receptor.¹⁰ That property may explain how lumateperone functions as an antipsychotic despite low levels of D2 receptor occupancy.¹⁰

Another hypothesis is based on our understanding of pimavanserin’s pharmacology. Pimavanserin is a selective 5HT2A antagonist FDA-approved for the treatment of Parkinson’s disease psychosis (PDP), with extremely high receptor affinity (Ki 0.087 nM) and no appreciable binding at dopamine receptors.⁵ Pimavanserin not only treats PDP, but is being evaluated in clinical trials for dementia-related psychosis, and has positive data for augmenting antipsychotics when there is a low level of D2 blockade.^11,12 In a controlled trial, pimavanserin added to risperidone, 2 mg/d, was as effective as risperidone, 6 mg/d, illustrating the point that near-saturation of the 5HT2A receptor can increase antipsychotic efficacy when dopamine blockade is relatively low. For risperidone, 2 mg/d, the expected D2 occupancy is only 60%.¹³

Lumateperone also has moderate binding affinity for serotonin transporters (SERT) (Ki 33 nM). Serotonin transporter occupancy at the dose approved for schizophrenia (42 mg/d) is approximately 30%,¹⁴ below the ≥80% SERT occupancy seen with selective serotonin reuptake inhibitor (SSRI) antidepressants; nevertheless, there is evidence for antidepressant effects seen in preclinical assays, schizophrenia studies, and phase III trials for bipolar depression.^8,15,16 It is hypothesized that near-saturation of the 5HT2A receptor might act synergistically with the modest extent of 5HT reuptake inhibition to promote downstream effects associated with effective antidepressant treatments.⁸ In vivo data also showed phosphorylation of N-methyl-d-aspartate receptor subunits and glycogen synthase kinase 3 beta (GSK3B), properties that could impact cognition or negative symptoms in patients with schizophrenia.⁸

Clinical implications

Nonadherence with oral antipsychotics among patients with schizophrenia is often related to adverse effects.¹⁷ The SGAs marketed since 2000 generally have lower rates of sedation and metabolic and/or endocrine adverse events than earlier compounds, yet each has limitations:

• asenapine: sedation and weight gain
• the partial agonists (aripiprazole, brexpiprazole, cariprazine): akathisia
• lurasidone: dose-dependent EPS and akathisia
• iloperidone: orthostasis.¹⁸

Ziprasidone is an exception, because it had low rates of most adverse effects in schizophrenia trials, but the need to take it twice daily with a 500 kcal meal hampers its use. A meta-analysis of 32 oral antipsychotics, including first-generation agents, noted that the efficacy differences between medications are slight for patients without treatment-resistant schizophrenia, but “differences in side-effects are more marked.”¹⁸

Continue to: Until novel mechanisms are discovered...

Efficacy in adults with schizophrenia. The efficacy of lumateperone has been established in 2 pivotal, double-blind, placebo-controlled trials. The first was a 4-week, phase II trial (Study 005) in which 335 adults age 18 to 55 with an acute exacerbation of schizophrenia were randomized in a 1:1:1:1 manner to lumateperone, 42 mg/d (60 mg of lumateperone tosylate), luma­teperone, 84 mg/d (120 mg of lumateperone tosylate), risperidone, 4 mg/d, or placebo, all taken once daily.²⁰ For the 4 treatment arms, the least squares mean changes from baseline to the Day 28 endpoint on the primary outcome measure, Positive and Negative Syndrome Scale (PANSS) total score, were: lumateperone, 42 mg/d: −13.2 points; lumateperone, 84 mg/d: −8.3 points; risperidone, 4 mg/d: −13.4 points; and placebo: −7.4 points. Both lumateperone, 42 mg/d, and risperidone, 4 mg/d, were significantly different than placebo, and with identical moderate effect sizes of 0.4.²⁰ Lumateperone, 84 mg/d, did not separate from placebo on the primary outcome. The responder analysis also indicated that a similar proportion of patients (40%) randomized to lumateperone, 42 mg/d, or risperidone, 4 mg/d, improved by ≥30% on PANSS total score.

The second pivotal trial (Study 301) was a phase III, double-blind, placebo-controlled trial of 450 adults, age 18 to 60, with an acute exacerbation of schizophrenia who were randomized in 1:1:1 manner to receive lumateperone, 42 mg/d (lumateperone tosylate 60 mg), lumateperone, 28 mg/d (lumateperone tosylate 40 mg), or placebo once daily for 4 weeks.²¹ For the 3 treatment arms, the least squares mean changes on PANSS total score from baseline to the Day 28 endpoint were: lumateperone, 42 mg/d: −14.5 points; lumateperone, 28 mg/d: −12.9 points; and placebo: −10.3 points. Lumateperone, 28 mg/d, did not separate from placebo on the primary outcome. The responder analysis also indicated that 36.5% of those receiving lumateperone, 42 mg/d, and 36.3% of those receiving lumateperone, 28 mg/d, improved by ≥30% on PANSS total score, compared with 25.5% of patients treated with placebo.

Unlike the 2 positive trials in which placebo change in total PANSS scores were −7.4 and −10.3 points, respectively, in a phase III trial (Study 302) with 696 participants, placebo showed a 15.1-point decrease from baseline PANSS total score.¹⁹ Among the 3 treatment arms of this study (lumateperone, 14 mg/d, lumateperone, 42 mg/d, and risperidone, 4 mg/d), only risperidone was superior to placebo.

Adverse events

In the phase II pivotal study, completion rates among the 4 arms were comparable: lumateperone, 42 mg/d: 71%; lumateperone, 84 mg/d: 76%; risperidone, 4 mg/d: 77%; and placebo: 72%.²⁰ There were no serious adverse events (SAEs) associated with lumateperone; the 2 SAEs that occurred involved worsening of schizophrenia/psychotic disorder for risperidone (n = 1) and for placebo (n = 1). Five participants discontinued the study due to an adverse event: 2 who were receiving lumateperone (1 due to dry mouth, and 1 due to worsening of schizophrenia) and 3 who were receiving risperidone (2 due to akathisia, and 1 due to blood creatine phosphokinase increase).²⁰ The most frequent adverse event was somnolence/sedation (placebo: 13%; lumateperone, 42 mg/d: 17%; risperidone, 4 mg/d: 21%; and lumateperone, 84 mg/d: 32.5%). Neither dose of lumateperone was associated with increased rates of EPS. Median weight gain to Day 28 was 1 kg for placebo and for each dose of lumateperone, and 2.5 kg for risperidone. Compared with risperidone, lumateperone showed statistically significantly lower prolactin levels (lumateperone, 42 mg/d and 84 mg/d: P < .001), and metabolic parameters, including fasting glucose (lumateperone 42 mg/d: P = .007; lumateperone, 84 mg/d: P = .023), total cholesterol (lumateperone, 42 mg/d: P = .012; lumateperone, 84 mg/d: P = .004), and triglycerides (lumateperone, 42 mg/d: P = .074; lumateperone, 84 mg/d: P = .002).²⁰ There was no significant impact on the corrected QT interval.

Continue to: In the phase III trial...

In the phase III trial, completion rates among the 3 arms were lumateperone, 42 mg/d: 85%; lumateperone, 28 mg/d: 80%; and placebo: 74%. There was 1 SAE in a patient receiving lumateperone, 28 mg/d. This individual had preexisting risk factors and a history of seizures, and experienced a seizure during the study. Adverse events that occurred in either lumateperone group at a rate ≥5% and more than twice the rate of placebo were somnolence (lumateperone, 42 mg/d: 17.3%; lumateperone, 28 mg/d: 11.3%; and placebo: 4.0%); sedation (lumateperone, 42 mg/d: 12.7%; lumateperone, 28 mg/d: 9.3%; and placebo: 5.4%); fatigue (lumateperone, 42 mg/d: 5.3%; lumateperone, 28 mg/d: 4.7%; and placebo: 1.3%); and constipation (lumateperone, 42 mg/d: 6.7%; lumateperone, 28 mg/d: 4.0%; and placebo: 2.7%).²¹ No EPS-related adverse events occurred in ≥5% patients in any treatment arm. Median change in weight from baseline to Day 28 was 0.9 kg for lumateperone, 42 mg/d, 0.6 kg for lumateperone, 28 mg/d, and 0.7 kg for placebo. There were no significant mean changes in metabolic parameters for any treatment arm, and none of the patients had a corrected QT interval (QTc) >500 ms or a change in QTc >60 ms from baseline.²¹

Pharmacologic profile

Lumateperone’s in vitro binding profile includes high affinity for serotonin 5HT2A receptors (Ki 0.54 nM), lower affinity for dopamine D2 receptors (Ki 32 nM), moderate binding affinity for SERT (Ki 33 nM), and lower affinity for alpha 1-adrenergic receptors (Ki 73 nM) and muscarinic and histaminergic receptors (Ki >100 nM).^6,7 As noted above, this 60-fold ratio of 5HT2A to D2 affinity is extremely high; moreover, imaging data reveal low D2 receptor occupancy, consistent with the lack of clinically significant EPS seen in the trials. In vitro assays also reveal impact on glutamate pathways, and pathways associated with antidepressant response.⁸ The clinical benefits of the glutamatergic properties remain theoretical, but the antidepressant benefit has been seen in a positive phase III trial for bipolar depression.¹⁹

Clinical considerations

Effect sizes in the 2 positive pivotal trials were 0.3 and 0.4, comparable with those for other antipsychotics approved within the last decade: brexpiprazole, 0.26; cariprazine, 0.34; and lurasidone, 0.36.²¹ The absence of clinically significant EPS, lack of impact on metabolic or endocrine parameters, and lack of titration are all appealing properties. That only 42 mg/d proved effective may reflect the fact that the other doses studied to date in randomized, fixed-dose studies were 14 mg/d (Study 302) and 84 mg/d (Study 005), evaluated in one study each. While those 2 doses might indeed be outside the therapeutic window, given the heterogeneity of schizophrenia, future studies might help further refine the therapeutic range for schizophrenia, especially for doses closer to 42 mg/d (eg, 28 mg/d, 63 mg/d). Should 42 mg/d not prove effective, there is no data for now to suggest whether a dose increase may be helpful. As there is only 1 marketed dose of lumateperone (42-mg capsules), and no easy way to modify this dose, lumateperone’s package insert includes cautionary language regarding situations where there will be less-than-expected drug exposure (use of cytochrome P450 [CYP] 3A4 inducers), greater-than-expected drug exposure (moderate or strong CYP 3A4 inhibitors), or use in patients with moderate or severe hepatic impairment as defined by Child-Pugh Criteria (Child-Pugh B or C). These are not contraindications.

Unique properties of lumateperone include the lack of presynaptic dopamine D2 antagonism, low D2 receptor occupancy, and the absence of significant EPS and metabolic or endocrine adverse effects. In vitro data indicate glutamatergic effects, and human data indicate antidepressant effects in bipolar patients. Despite the absence of significant histamine H1 or muscarinic affinity, the rate of somnolence/sedation was twice that of placebo (lumateperone 24%, placebo 10%).⁷

Why Rx? Reasons to prescribe lumateperone for adult patients with schizophrenia include:

• Favorable tolerability profile, including no significant signal for EPS or endocrine or metabolic adverse effects, and no QT prolongation
• No need for titration.

Dosing. There is only 1 dose available for lumateperone, 42-mg capsules (Table 2). As the dose cannot be modified, the package insert contains cautionary language regarding situations with less-than-expected drug exposure (use of CYP 3A4 inducers), greater-than-expected drug exposure (moderate or strong CYP 3A4 inhibitors), or use in patients with moderate or severe hepatic impairment as defined by Child-Pugh criteria (Child-Pugh B or C). These are not contraindications.

Contraindications. The only contraindication is known hypersensitivity to lumateperone.

Continue to: Bottom Line

Lumateperone is a novel oral antipsychotic indicated for treating adults with schizophrenia. Its unique properties include the lack of presynaptic dopamine D2 antagonism, low D2 receptor occupancy, and the absence of significant extrapyramidal symptoms and metabolic or endocrine adverse effects. In clinical trials, the most frequent adverse event was somnolence/sedation.

Related Resource

• Fulton D. FDA approves Caplyta to treat schizophrenenia in adults. https://www.mdedge.com/psychiatry/article/214733/schizophrenia-other-psychotic-disorders/fda-approves-caplyta-treat.

Drug Brand Names

Aripiprazole • Abilify
Asenapine • Saphris
Brexpiprazole • Rexulti
Cariprazine • Vraylar
Iloperidone • Fanapt
Lumateperone • Caplyta
Lurasidone • Latuda
Olanzapine • Zyprexa
Pimavanserin • Nuplazid
Risperidone • Risperdal
Ziprasidone • Geodon

Antipsychotic nonadherence is a known contributor to relapse risk among patients with schizophrenia.¹ Because relapse episodes may be associated with antipsychotic treatment resistance, this must be avoided as much as possible by appropriate medication selection.² Adverse effect burden is an important factor leading to oral antipsychotic nonadherence, with patient-derived data indicating that extrapyramidal symptoms (EPS) (odds ratio [OR] 0.57, P = .0007), sedation/cognitive adverse effects (OR 0.70, P = .033), prolactin/endocrine effects (OR 0.69, P = .0342), and metabolic adverse effects (OR 0.64, P = .0079) are all significantly related to lower rates of adherence.³ With this in mind, successive generations of antipsychotics have been released, with fewer tolerability issues present than seen with earlier compounds.^1,4 Although these newer second-generation antipsychotics (SGAs) have not proven more effective for schizophrenia than those first marketed in the 1990s, they generally possess lower rates of EPS, hyperprolactinemia, anticholinergic and antihistaminic properties, metabolic adverse effects, and orthostasis.⁵ This improved adverse effect profile will hopefully increase the chances of antipsychotic acceptance in patients with schizophrenia, and thereby promote improved adherence.

Lumateperone (Caplyta) is a novel oral antipsychotic approved for the treatment of adult patients with schizophrenia (Table 1). It possesses some properties seen with other SGAs, including high affinity for serotonin 5HT2A receptors (Ki 0.54 nM) and lower affinity for dopamine D2 receptors (Ki 32 nM), along with low affinity for alpha1-adrenergic receptors (Ki 73 nM), and muscarinic and histaminergic receptors (Ki > 100 nM).^6,7 However, there are some distinguishing features: the ratio of 5HT2A receptor affinity to D2 affinity is 60, greater than that of other SGAs such as risperidone (12), olanzapine (12.4) or aripiprazole (0.18)⁸; at steady state, the D2 occupancy remains <40% (Figure) and the corresponding rates of EPS/akathisia were only 6.7% for lumateperone vs 6.3% for placebo in short-term clinical trials.^7,9

How it works

A unique aspect of lumateperone’s pharmacology may relate to differential actions at presynaptic and postsynaptic dopamine D2 receptors. Other antipsychotics possess comparable antagonist (or partial agonist) properties at postsynaptic D2 receptors (the D2 long isoform) and the presynaptic autoreceptor (the D2 short isoform). By blocking the presynaptic autoreceptor, feedback inhibition on dopamine release is removed; therefore, the required higher levels of postsynaptic D2 receptor occupancy needed for effective antipsychotic action (eg, 60% to 80% for antagonists, and 83% to 100% for partial agonists) may be a product of the need to oppose this increased presynaptic release of dopamine. In vitro assays show that lumateperone does not increase presynaptic dopamine release, indicating that it possesses agonist properties at the presynaptic D2 short receptor.¹⁰ That property may explain how lumateperone functions as an antipsychotic despite low levels of D2 receptor occupancy.¹⁰

Another hypothesis is based on our understanding of pimavanserin’s pharmacology. Pimavanserin is a selective 5HT2A antagonist FDA-approved for the treatment of Parkinson’s disease psychosis (PDP), with extremely high receptor affinity (Ki 0.087 nM) and no appreciable binding at dopamine receptors.⁵ Pimavanserin not only treats PDP, but is being evaluated in clinical trials for dementia-related psychosis, and has positive data for augmenting antipsychotics when there is a low level of D2 blockade.^11,12 In a controlled trial, pimavanserin added to risperidone, 2 mg/d, was as effective as risperidone, 6 mg/d, illustrating the point that near-saturation of the 5HT2A receptor can increase antipsychotic efficacy when dopamine blockade is relatively low. For risperidone, 2 mg/d, the expected D2 occupancy is only 60%.¹³

Lumateperone also has moderate binding affinity for serotonin transporters (SERT) (Ki 33 nM). Serotonin transporter occupancy at the dose approved for schizophrenia (42 mg/d) is approximately 30%,¹⁴ below the ≥80% SERT occupancy seen with selective serotonin reuptake inhibitor (SSRI) antidepressants; nevertheless, there is evidence for antidepressant effects seen in preclinical assays, schizophrenia studies, and phase III trials for bipolar depression.^8,15,16 It is hypothesized that near-saturation of the 5HT2A receptor might act synergistically with the modest extent of 5HT reuptake inhibition to promote downstream effects associated with effective antidepressant treatments.⁸ In vivo data also showed phosphorylation of N-methyl-d-aspartate receptor subunits and glycogen synthase kinase 3 beta (GSK3B), properties that could impact cognition or negative symptoms in patients with schizophrenia.⁸

Clinical implications

Nonadherence with oral antipsychotics among patients with schizophrenia is often related to adverse effects.¹⁷ The SGAs marketed since 2000 generally have lower rates of sedation and metabolic and/or endocrine adverse events than earlier compounds, yet each has limitations:

• asenapine: sedation and weight gain
• the partial agonists (aripiprazole, brexpiprazole, cariprazine): akathisia
• lurasidone: dose-dependent EPS and akathisia
• iloperidone: orthostasis.¹⁸

Ziprasidone is an exception, because it had low rates of most adverse effects in schizophrenia trials, but the need to take it twice daily with a 500 kcal meal hampers its use. A meta-analysis of 32 oral antipsychotics, including first-generation agents, noted that the efficacy differences between medications are slight for patients without treatment-resistant schizophrenia, but “differences in side-effects are more marked.”¹⁸

Continue to: Until novel mechanisms are discovered...

Efficacy in adults with schizophrenia. The efficacy of lumateperone has been established in 2 pivotal, double-blind, placebo-controlled trials. The first was a 4-week, phase II trial (Study 005) in which 335 adults age 18 to 55 with an acute exacerbation of schizophrenia were randomized in a 1:1:1:1 manner to lumateperone, 42 mg/d (60 mg of lumateperone tosylate), luma­teperone, 84 mg/d (120 mg of lumateperone tosylate), risperidone, 4 mg/d, or placebo, all taken once daily.²⁰ For the 4 treatment arms, the least squares mean changes from baseline to the Day 28 endpoint on the primary outcome measure, Positive and Negative Syndrome Scale (PANSS) total score, were: lumateperone, 42 mg/d: −13.2 points; lumateperone, 84 mg/d: −8.3 points; risperidone, 4 mg/d: −13.4 points; and placebo: −7.4 points. Both lumateperone, 42 mg/d, and risperidone, 4 mg/d, were significantly different than placebo, and with identical moderate effect sizes of 0.4.²⁰ Lumateperone, 84 mg/d, did not separate from placebo on the primary outcome. The responder analysis also indicated that a similar proportion of patients (40%) randomized to lumateperone, 42 mg/d, or risperidone, 4 mg/d, improved by ≥30% on PANSS total score.

The second pivotal trial (Study 301) was a phase III, double-blind, placebo-controlled trial of 450 adults, age 18 to 60, with an acute exacerbation of schizophrenia who were randomized in 1:1:1 manner to receive lumateperone, 42 mg/d (lumateperone tosylate 60 mg), lumateperone, 28 mg/d (lumateperone tosylate 40 mg), or placebo once daily for 4 weeks.²¹ For the 3 treatment arms, the least squares mean changes on PANSS total score from baseline to the Day 28 endpoint were: lumateperone, 42 mg/d: −14.5 points; lumateperone, 28 mg/d: −12.9 points; and placebo: −10.3 points. Lumateperone, 28 mg/d, did not separate from placebo on the primary outcome. The responder analysis also indicated that 36.5% of those receiving lumateperone, 42 mg/d, and 36.3% of those receiving lumateperone, 28 mg/d, improved by ≥30% on PANSS total score, compared with 25.5% of patients treated with placebo.

Unlike the 2 positive trials in which placebo change in total PANSS scores were −7.4 and −10.3 points, respectively, in a phase III trial (Study 302) with 696 participants, placebo showed a 15.1-point decrease from baseline PANSS total score.¹⁹ Among the 3 treatment arms of this study (lumateperone, 14 mg/d, lumateperone, 42 mg/d, and risperidone, 4 mg/d), only risperidone was superior to placebo.

Adverse events

In the phase II pivotal study, completion rates among the 4 arms were comparable: lumateperone, 42 mg/d: 71%; lumateperone, 84 mg/d: 76%; risperidone, 4 mg/d: 77%; and placebo: 72%.²⁰ There were no serious adverse events (SAEs) associated with lumateperone; the 2 SAEs that occurred involved worsening of schizophrenia/psychotic disorder for risperidone (n = 1) and for placebo (n = 1). Five participants discontinued the study due to an adverse event: 2 who were receiving lumateperone (1 due to dry mouth, and 1 due to worsening of schizophrenia) and 3 who were receiving risperidone (2 due to akathisia, and 1 due to blood creatine phosphokinase increase).²⁰ The most frequent adverse event was somnolence/sedation (placebo: 13%; lumateperone, 42 mg/d: 17%; risperidone, 4 mg/d: 21%; and lumateperone, 84 mg/d: 32.5%). Neither dose of lumateperone was associated with increased rates of EPS. Median weight gain to Day 28 was 1 kg for placebo and for each dose of lumateperone, and 2.5 kg for risperidone. Compared with risperidone, lumateperone showed statistically significantly lower prolactin levels (lumateperone, 42 mg/d and 84 mg/d: P < .001), and metabolic parameters, including fasting glucose (lumateperone 42 mg/d: P = .007; lumateperone, 84 mg/d: P = .023), total cholesterol (lumateperone, 42 mg/d: P = .012; lumateperone, 84 mg/d: P = .004), and triglycerides (lumateperone, 42 mg/d: P = .074; lumateperone, 84 mg/d: P = .002).²⁰ There was no significant impact on the corrected QT interval.

Continue to: In the phase III trial...

In the phase III trial, completion rates among the 3 arms were lumateperone, 42 mg/d: 85%; lumateperone, 28 mg/d: 80%; and placebo: 74%. There was 1 SAE in a patient receiving lumateperone, 28 mg/d. This individual had preexisting risk factors and a history of seizures, and experienced a seizure during the study. Adverse events that occurred in either lumateperone group at a rate ≥5% and more than twice the rate of placebo were somnolence (lumateperone, 42 mg/d: 17.3%; lumateperone, 28 mg/d: 11.3%; and placebo: 4.0%); sedation (lumateperone, 42 mg/d: 12.7%; lumateperone, 28 mg/d: 9.3%; and placebo: 5.4%); fatigue (lumateperone, 42 mg/d: 5.3%; lumateperone, 28 mg/d: 4.7%; and placebo: 1.3%); and constipation (lumateperone, 42 mg/d: 6.7%; lumateperone, 28 mg/d: 4.0%; and placebo: 2.7%).²¹ No EPS-related adverse events occurred in ≥5% patients in any treatment arm. Median change in weight from baseline to Day 28 was 0.9 kg for lumateperone, 42 mg/d, 0.6 kg for lumateperone, 28 mg/d, and 0.7 kg for placebo. There were no significant mean changes in metabolic parameters for any treatment arm, and none of the patients had a corrected QT interval (QTc) >500 ms or a change in QTc >60 ms from baseline.²¹

Pharmacologic profile

Lumateperone’s in vitro binding profile includes high affinity for serotonin 5HT2A receptors (Ki 0.54 nM), lower affinity for dopamine D2 receptors (Ki 32 nM), moderate binding affinity for SERT (Ki 33 nM), and lower affinity for alpha 1-adrenergic receptors (Ki 73 nM) and muscarinic and histaminergic receptors (Ki >100 nM).^6,7 As noted above, this 60-fold ratio of 5HT2A to D2 affinity is extremely high; moreover, imaging data reveal low D2 receptor occupancy, consistent with the lack of clinically significant EPS seen in the trials. In vitro assays also reveal impact on glutamate pathways, and pathways associated with antidepressant response.⁸ The clinical benefits of the glutamatergic properties remain theoretical, but the antidepressant benefit has been seen in a positive phase III trial for bipolar depression.¹⁹

Clinical considerations

Effect sizes in the 2 positive pivotal trials were 0.3 and 0.4, comparable with those for other antipsychotics approved within the last decade: brexpiprazole, 0.26; cariprazine, 0.34; and lurasidone, 0.36.²¹ The absence of clinically significant EPS, lack of impact on metabolic or endocrine parameters, and lack of titration are all appealing properties. That only 42 mg/d proved effective may reflect the fact that the other doses studied to date in randomized, fixed-dose studies were 14 mg/d (Study 302) and 84 mg/d (Study 005), evaluated in one study each. While those 2 doses might indeed be outside the therapeutic window, given the heterogeneity of schizophrenia, future studies might help further refine the therapeutic range for schizophrenia, especially for doses closer to 42 mg/d (eg, 28 mg/d, 63 mg/d). Should 42 mg/d not prove effective, there is no data for now to suggest whether a dose increase may be helpful. As there is only 1 marketed dose of lumateperone (42-mg capsules), and no easy way to modify this dose, lumateperone’s package insert includes cautionary language regarding situations where there will be less-than-expected drug exposure (use of cytochrome P450 [CYP] 3A4 inducers), greater-than-expected drug exposure (moderate or strong CYP 3A4 inhibitors), or use in patients with moderate or severe hepatic impairment as defined by Child-Pugh Criteria (Child-Pugh B or C). These are not contraindications.

Unique properties of lumateperone include the lack of presynaptic dopamine D2 antagonism, low D2 receptor occupancy, and the absence of significant EPS and metabolic or endocrine adverse effects. In vitro data indicate glutamatergic effects, and human data indicate antidepressant effects in bipolar patients. Despite the absence of significant histamine H1 or muscarinic affinity, the rate of somnolence/sedation was twice that of placebo (lumateperone 24%, placebo 10%).⁷

Why Rx? Reasons to prescribe lumateperone for adult patients with schizophrenia include:

• Favorable tolerability profile, including no significant signal for EPS or endocrine or metabolic adverse effects, and no QT prolongation
• No need for titration.

Dosing. There is only 1 dose available for lumateperone, 42-mg capsules (Table 2). As the dose cannot be modified, the package insert contains cautionary language regarding situations with less-than-expected drug exposure (use of CYP 3A4 inducers), greater-than-expected drug exposure (moderate or strong CYP 3A4 inhibitors), or use in patients with moderate or severe hepatic impairment as defined by Child-Pugh criteria (Child-Pugh B or C). These are not contraindications.

Contraindications. The only contraindication is known hypersensitivity to lumateperone.

Continue to: Bottom Line

Lumateperone is a novel oral antipsychotic indicated for treating adults with schizophrenia. Its unique properties include the lack of presynaptic dopamine D2 antagonism, low D2 receptor occupancy, and the absence of significant extrapyramidal symptoms and metabolic or endocrine adverse effects. In clinical trials, the most frequent adverse event was somnolence/sedation.

Related Resource

• Fulton D. FDA approves Caplyta to treat schizophrenenia in adults. https://www.mdedge.com/psychiatry/article/214733/schizophrenia-other-psychotic-disorders/fda-approves-caplyta-treat.

Drug Brand Names

Aripiprazole • Abilify
Asenapine • Saphris
Brexpiprazole • Rexulti
Cariprazine • Vraylar
Iloperidone • Fanapt
Lumateperone • Caplyta
Lurasidone • Latuda
Olanzapine • Zyprexa
Pimavanserin • Nuplazid
Risperidone • Risperdal
Ziprasidone • Geodon

Antipsychotic nonadherence is a known contributor to relapse risk among patients with schizophrenia.¹ Because relapse episodes may be associated with antipsychotic treatment resistance, this must be avoided as much as possible by appropriate medication selection.² Adverse effect burden is an important factor leading to oral antipsychotic nonadherence, with patient-derived data indicating that extrapyramidal symptoms (EPS) (odds ratio [OR] 0.57, P = .0007), sedation/cognitive adverse effects (OR 0.70, P = .033), prolactin/endocrine effects (OR 0.69, P = .0342), and metabolic adverse effects (OR 0.64, P = .0079) are all significantly related to lower rates of adherence.³ With this in mind, successive generations of antipsychotics have been released, with fewer tolerability issues present than seen with earlier compounds.^1,4 Although these newer second-generation antipsychotics (SGAs) have not proven more effective for schizophrenia than those first marketed in the 1990s, they generally possess lower rates of EPS, hyperprolactinemia, anticholinergic and antihistaminic properties, metabolic adverse effects, and orthostasis.⁵ This improved adverse effect profile will hopefully increase the chances of antipsychotic acceptance in patients with schizophrenia, and thereby promote improved adherence.

Lumateperone (Caplyta) is a novel oral antipsychotic approved for the treatment of adult patients with schizophrenia (Table 1). It possesses some properties seen with other SGAs, including high affinity for serotonin 5HT2A receptors (Ki 0.54 nM) and lower affinity for dopamine D2 receptors (Ki 32 nM), along with low affinity for alpha1-adrenergic receptors (Ki 73 nM), and muscarinic and histaminergic receptors (Ki > 100 nM).^6,7 However, there are some distinguishing features: the ratio of 5HT2A receptor affinity to D2 affinity is 60, greater than that of other SGAs such as risperidone (12), olanzapine (12.4) or aripiprazole (0.18)⁸; at steady state, the D2 occupancy remains <40% (Figure) and the corresponding rates of EPS/akathisia were only 6.7% for lumateperone vs 6.3% for placebo in short-term clinical trials.^7,9

How it works

A unique aspect of lumateperone’s pharmacology may relate to differential actions at presynaptic and postsynaptic dopamine D2 receptors. Other antipsychotics possess comparable antagonist (or partial agonist) properties at postsynaptic D2 receptors (the D2 long isoform) and the presynaptic autoreceptor (the D2 short isoform). By blocking the presynaptic autoreceptor, feedback inhibition on dopamine release is removed; therefore, the required higher levels of postsynaptic D2 receptor occupancy needed for effective antipsychotic action (eg, 60% to 80% for antagonists, and 83% to 100% for partial agonists) may be a product of the need to oppose this increased presynaptic release of dopamine. In vitro assays show that lumateperone does not increase presynaptic dopamine release, indicating that it possesses agonist properties at the presynaptic D2 short receptor.¹⁰ That property may explain how lumateperone functions as an antipsychotic despite low levels of D2 receptor occupancy.¹⁰

Another hypothesis is based on our understanding of pimavanserin’s pharmacology. Pimavanserin is a selective 5HT2A antagonist FDA-approved for the treatment of Parkinson’s disease psychosis (PDP), with extremely high receptor affinity (Ki 0.087 nM) and no appreciable binding at dopamine receptors.⁵ Pimavanserin not only treats PDP, but is being evaluated in clinical trials for dementia-related psychosis, and has positive data for augmenting antipsychotics when there is a low level of D2 blockade.^11,12 In a controlled trial, pimavanserin added to risperidone, 2 mg/d, was as effective as risperidone, 6 mg/d, illustrating the point that near-saturation of the 5HT2A receptor can increase antipsychotic efficacy when dopamine blockade is relatively low. For risperidone, 2 mg/d, the expected D2 occupancy is only 60%.¹³

Lumateperone also has moderate binding affinity for serotonin transporters (SERT) (Ki 33 nM). Serotonin transporter occupancy at the dose approved for schizophrenia (42 mg/d) is approximately 30%,¹⁴ below the ≥80% SERT occupancy seen with selective serotonin reuptake inhibitor (SSRI) antidepressants; nevertheless, there is evidence for antidepressant effects seen in preclinical assays, schizophrenia studies, and phase III trials for bipolar depression.^8,15,16 It is hypothesized that near-saturation of the 5HT2A receptor might act synergistically with the modest extent of 5HT reuptake inhibition to promote downstream effects associated with effective antidepressant treatments.⁸ In vivo data also showed phosphorylation of N-methyl-d-aspartate receptor subunits and glycogen synthase kinase 3 beta (GSK3B), properties that could impact cognition or negative symptoms in patients with schizophrenia.⁸

Clinical implications

Nonadherence with oral antipsychotics among patients with schizophrenia is often related to adverse effects.¹⁷ The SGAs marketed since 2000 generally have lower rates of sedation and metabolic and/or endocrine adverse events than earlier compounds, yet each has limitations:

• asenapine: sedation and weight gain
• the partial agonists (aripiprazole, brexpiprazole, cariprazine): akathisia
• lurasidone: dose-dependent EPS and akathisia
• iloperidone: orthostasis.¹⁸

Ziprasidone is an exception, because it had low rates of most adverse effects in schizophrenia trials, but the need to take it twice daily with a 500 kcal meal hampers its use. A meta-analysis of 32 oral antipsychotics, including first-generation agents, noted that the efficacy differences between medications are slight for patients without treatment-resistant schizophrenia, but “differences in side-effects are more marked.”¹⁸

Continue to: Until novel mechanisms are discovered...

Efficacy in adults with schizophrenia. The efficacy of lumateperone has been established in 2 pivotal, double-blind, placebo-controlled trials. The first was a 4-week, phase II trial (Study 005) in which 335 adults age 18 to 55 with an acute exacerbation of schizophrenia were randomized in a 1:1:1:1 manner to lumateperone, 42 mg/d (60 mg of lumateperone tosylate), luma­teperone, 84 mg/d (120 mg of lumateperone tosylate), risperidone, 4 mg/d, or placebo, all taken once daily.²⁰ For the 4 treatment arms, the least squares mean changes from baseline to the Day 28 endpoint on the primary outcome measure, Positive and Negative Syndrome Scale (PANSS) total score, were: lumateperone, 42 mg/d: −13.2 points; lumateperone, 84 mg/d: −8.3 points; risperidone, 4 mg/d: −13.4 points; and placebo: −7.4 points. Both lumateperone, 42 mg/d, and risperidone, 4 mg/d, were significantly different than placebo, and with identical moderate effect sizes of 0.4.²⁰ Lumateperone, 84 mg/d, did not separate from placebo on the primary outcome. The responder analysis also indicated that a similar proportion of patients (40%) randomized to lumateperone, 42 mg/d, or risperidone, 4 mg/d, improved by ≥30% on PANSS total score.

The second pivotal trial (Study 301) was a phase III, double-blind, placebo-controlled trial of 450 adults, age 18 to 60, with an acute exacerbation of schizophrenia who were randomized in 1:1:1 manner to receive lumateperone, 42 mg/d (lumateperone tosylate 60 mg), lumateperone, 28 mg/d (lumateperone tosylate 40 mg), or placebo once daily for 4 weeks.²¹ For the 3 treatment arms, the least squares mean changes on PANSS total score from baseline to the Day 28 endpoint were: lumateperone, 42 mg/d: −14.5 points; lumateperone, 28 mg/d: −12.9 points; and placebo: −10.3 points. Lumateperone, 28 mg/d, did not separate from placebo on the primary outcome. The responder analysis also indicated that 36.5% of those receiving lumateperone, 42 mg/d, and 36.3% of those receiving lumateperone, 28 mg/d, improved by ≥30% on PANSS total score, compared with 25.5% of patients treated with placebo.

Unlike the 2 positive trials in which placebo change in total PANSS scores were −7.4 and −10.3 points, respectively, in a phase III trial (Study 302) with 696 participants, placebo showed a 15.1-point decrease from baseline PANSS total score.¹⁹ Among the 3 treatment arms of this study (lumateperone, 14 mg/d, lumateperone, 42 mg/d, and risperidone, 4 mg/d), only risperidone was superior to placebo.

Adverse events

In the phase II pivotal study, completion rates among the 4 arms were comparable: lumateperone, 42 mg/d: 71%; lumateperone, 84 mg/d: 76%; risperidone, 4 mg/d: 77%; and placebo: 72%.²⁰ There were no serious adverse events (SAEs) associated with lumateperone; the 2 SAEs that occurred involved worsening of schizophrenia/psychotic disorder for risperidone (n = 1) and for placebo (n = 1). Five participants discontinued the study due to an adverse event: 2 who were receiving lumateperone (1 due to dry mouth, and 1 due to worsening of schizophrenia) and 3 who were receiving risperidone (2 due to akathisia, and 1 due to blood creatine phosphokinase increase).²⁰ The most frequent adverse event was somnolence/sedation (placebo: 13%; lumateperone, 42 mg/d: 17%; risperidone, 4 mg/d: 21%; and lumateperone, 84 mg/d: 32.5%). Neither dose of lumateperone was associated with increased rates of EPS. Median weight gain to Day 28 was 1 kg for placebo and for each dose of lumateperone, and 2.5 kg for risperidone. Compared with risperidone, lumateperone showed statistically significantly lower prolactin levels (lumateperone, 42 mg/d and 84 mg/d: P < .001), and metabolic parameters, including fasting glucose (lumateperone 42 mg/d: P = .007; lumateperone, 84 mg/d: P = .023), total cholesterol (lumateperone, 42 mg/d: P = .012; lumateperone, 84 mg/d: P = .004), and triglycerides (lumateperone, 42 mg/d: P = .074; lumateperone, 84 mg/d: P = .002).²⁰ There was no significant impact on the corrected QT interval.

Continue to: In the phase III trial...

In the phase III trial, completion rates among the 3 arms were lumateperone, 42 mg/d: 85%; lumateperone, 28 mg/d: 80%; and placebo: 74%. There was 1 SAE in a patient receiving lumateperone, 28 mg/d. This individual had preexisting risk factors and a history of seizures, and experienced a seizure during the study. Adverse events that occurred in either lumateperone group at a rate ≥5% and more than twice the rate of placebo were somnolence (lumateperone, 42 mg/d: 17.3%; lumateperone, 28 mg/d: 11.3%; and placebo: 4.0%); sedation (lumateperone, 42 mg/d: 12.7%; lumateperone, 28 mg/d: 9.3%; and placebo: 5.4%); fatigue (lumateperone, 42 mg/d: 5.3%; lumateperone, 28 mg/d: 4.7%; and placebo: 1.3%); and constipation (lumateperone, 42 mg/d: 6.7%; lumateperone, 28 mg/d: 4.0%; and placebo: 2.7%).²¹ No EPS-related adverse events occurred in ≥5% patients in any treatment arm. Median change in weight from baseline to Day 28 was 0.9 kg for lumateperone, 42 mg/d, 0.6 kg for lumateperone, 28 mg/d, and 0.7 kg for placebo. There were no significant mean changes in metabolic parameters for any treatment arm, and none of the patients had a corrected QT interval (QTc) >500 ms or a change in QTc >60 ms from baseline.²¹

Pharmacologic profile

Lumateperone’s in vitro binding profile includes high affinity for serotonin 5HT2A receptors (Ki 0.54 nM), lower affinity for dopamine D2 receptors (Ki 32 nM), moderate binding affinity for SERT (Ki 33 nM), and lower affinity for alpha 1-adrenergic receptors (Ki 73 nM) and muscarinic and histaminergic receptors (Ki >100 nM).^6,7 As noted above, this 60-fold ratio of 5HT2A to D2 affinity is extremely high; moreover, imaging data reveal low D2 receptor occupancy, consistent with the lack of clinically significant EPS seen in the trials. In vitro assays also reveal impact on glutamate pathways, and pathways associated with antidepressant response.⁸ The clinical benefits of the glutamatergic properties remain theoretical, but the antidepressant benefit has been seen in a positive phase III trial for bipolar depression.¹⁹

Clinical considerations

Effect sizes in the 2 positive pivotal trials were 0.3 and 0.4, comparable with those for other antipsychotics approved within the last decade: brexpiprazole, 0.26; cariprazine, 0.34; and lurasidone, 0.36.²¹ The absence of clinically significant EPS, lack of impact on metabolic or endocrine parameters, and lack of titration are all appealing properties. That only 42 mg/d proved effective may reflect the fact that the other doses studied to date in randomized, fixed-dose studies were 14 mg/d (Study 302) and 84 mg/d (Study 005), evaluated in one study each. While those 2 doses might indeed be outside the therapeutic window, given the heterogeneity of schizophrenia, future studies might help further refine the therapeutic range for schizophrenia, especially for doses closer to 42 mg/d (eg, 28 mg/d, 63 mg/d). Should 42 mg/d not prove effective, there is no data for now to suggest whether a dose increase may be helpful. As there is only 1 marketed dose of lumateperone (42-mg capsules), and no easy way to modify this dose, lumateperone’s package insert includes cautionary language regarding situations where there will be less-than-expected drug exposure (use of cytochrome P450 [CYP] 3A4 inducers), greater-than-expected drug exposure (moderate or strong CYP 3A4 inhibitors), or use in patients with moderate or severe hepatic impairment as defined by Child-Pugh Criteria (Child-Pugh B or C). These are not contraindications.

Unique properties of lumateperone include the lack of presynaptic dopamine D2 antagonism, low D2 receptor occupancy, and the absence of significant EPS and metabolic or endocrine adverse effects. In vitro data indicate glutamatergic effects, and human data indicate antidepressant effects in bipolar patients. Despite the absence of significant histamine H1 or muscarinic affinity, the rate of somnolence/sedation was twice that of placebo (lumateperone 24%, placebo 10%).⁷

Why Rx? Reasons to prescribe lumateperone for adult patients with schizophrenia include:

• Favorable tolerability profile, including no significant signal for EPS or endocrine or metabolic adverse effects, and no QT prolongation
• No need for titration.

Dosing. There is only 1 dose available for lumateperone, 42-mg capsules (Table 2). As the dose cannot be modified, the package insert contains cautionary language regarding situations with less-than-expected drug exposure (use of CYP 3A4 inducers), greater-than-expected drug exposure (moderate or strong CYP 3A4 inhibitors), or use in patients with moderate or severe hepatic impairment as defined by Child-Pugh criteria (Child-Pugh B or C). These are not contraindications.

Contraindications. The only contraindication is known hypersensitivity to lumateperone.

Continue to: Bottom Line

Lumateperone is a novel oral antipsychotic indicated for treating adults with schizophrenia. Its unique properties include the lack of presynaptic dopamine D2 antagonism, low D2 receptor occupancy, and the absence of significant extrapyramidal symptoms and metabolic or endocrine adverse effects. In clinical trials, the most frequent adverse event was somnolence/sedation.

Related Resource

• Fulton D. FDA approves Caplyta to treat schizophrenenia in adults. https://www.mdedge.com/psychiatry/article/214733/schizophrenia-other-psychotic-disorders/fda-approves-caplyta-treat.

Drug Brand Names

Aripiprazole • Abilify
Asenapine • Saphris
Brexpiprazole • Rexulti
Cariprazine • Vraylar
Iloperidone • Fanapt
Lumateperone • Caplyta
Lurasidone • Latuda
Olanzapine • Zyprexa
Pimavanserin • Nuplazid
Risperidone • Risperdal
Ziprasidone • Geodon

Oral antipsychotic nonadherence is a significant contributor to relapse in patients with schizophrenia spectrum disorders. Long-acting injectable (LAI) antipsychotics have been developed to provide sustained antipsychotic exposure, with evidence that use of LAIs significantly reduces hospitalization rates.¹ One limiting factor in transitioning patients to certain LAIs is the need for prolonged oral coverage at the onset of treatment for agents that cannot be loaded. Nonadherence with this bridging oral therapy places the patient at risk for symptom exacerbation until effective antipsychotic plasma levels are achieved from the LAI.² Although risperidone is one of the more widely used antipsychotics for treating schizophrenia, until recently the only available LAI preparation, risperidone microspheres (Risperdal Consta), required 3 weeks of oral coverage upon initiation.³

Clinical implications

Oral medication nonadherence remains a significant public health issue for patients with schizophrenia, with an estimated 50% of patients failing to achieve 80% adherence even when enrolled in clinical trials specifically designed to track adherence.⁵ Although LAI atypical antipsychotics have been available since the approval of Risperdal Consta, the LAI form of risperidone, and both LAI forms of aripiprazole, were not designed to be loaded. A 1-day initiation regimen for aripiprazole lauroxil has been developed to avoid the need for 3 weeks of oral medication coverage,^6,7 but aripiprazole monohydrate and risperidone microspheres mandate oral bridging of 2 and 3 weeks, respectively.² Because one of the primary indications for LAI antipsychotic therapy is oral medication nonadherence, this prolonged period of oral coverage creates a risk for symptom exacerbation when the bridging period occurs outside of a controlled setting, as is common when patients are discharged from inpatient hospitalization.

One solution to this problem has its antecedents in the development of the Atrigel biodegradable injectable polymer, which was designed to deliver prolonged medication exposure after subcutaneous injection.⁸ This biodegradable polymer drug delivery system suspends and dissolves the medication of interest (in this case, risperidone) in a poly DL-lactide-coglycolide gel and its biocompatible carrier.⁹ The viscous liquid undergoes a phase transition upon contact with tissue fluids after subcutaneous injection, resulting in an implant that releases risperidone in a controlled manner as it is resorbed. Importantly, the kinetic parameters of RBP-7000 are such that effective drug levels are seen within the first week without the need for oral coverage.¹⁰

Use in adults with schizophrenia. After establishing tolerability with oral risperidone, the recommended doses are 90 mg or 120 mg monthly, which correspond to oral daily risperidone doses of 3 mg or 4 mg. RBP-7000 must be administered as a subcutaneous abdominal injection by a health care professional. It is recommended that the patient be in the supine position for the injection and that the injection sites be rotated monthly among 4 quadrants in the abdominal region. The injection volumes for the 90 mg and 120 mg doses are 0.6 mL and 0.8 mL, respectively.¹⁰ As the gel implant becomes firmer, the patient will notice a lump for several weeks that will decrease in size over time. Patients should be advised not to rub or massage the injection site, and to be aware of the placement of any belts or clothing with waistbands.¹⁰

Pharmacologic profile, adverse reactions

Risperidone is an atypical antipsychotic that has been commercially available in the U.S. since December 29, 1993, and its adverse effect profile is well characterized. The most common adverse effects associated with risperidone include those related to dopamine D2 antagonism, metabolic adverse effects, and an increase in serum prolactin. In the 12-month long-term safety study of RBP-7000, 1-minute post-dose injection site pain scores (on a 100-point scale) were highest on Day 1 (mean of 25) and decreased over time with subsequent injections (14 to 16 following the last injection).¹⁰

Continue to: How the Atrigel system works

The kinetics for RBP-7000 are markedly different than those for oral risperidone (Figure 2¹⁵). After a single subcutaneous injection, RBP-7000 shows 2 absorption peaks for risperidone. The first lower peak occurs with a T_max of 4 to 6 hours due to initial release of risperidone during the implant formation process; a second risperidone peak occurs after 10 to 14 days and is associated with slow release from the subcutaneous depot.^9,16,17 For both 9-OH risperidone levels and the total active moiety (risperidone plus 9-OH risperidone levels), the median T_max of the first peak ranges from 4 to 48 hours and the second peak ranges from 7 to 11 days. Following a single subcutaneous injection of RBP-7000, the apparent terminal half-life of risperidone ranges from 9 to 11 days, on average. The mean apparent terminal half-life of the active moiety ranges from 8 to 9 days.^9,16,17 Based on population pharmacokinetic modeling, the 90 mg and 120 mg doses of RBP-7000 are estimated to provide drug exposure equivalent to 3 mg/d and 4 mg/d of oral risperidone, respectively.^9,16,17

Continue to: Efficacy of RBP-7000

Efficacy of RBP-7000 was established in an 8-week, double-blind, placebo-controlled trial of adult patients experiencing an acute exacerbation of schizo­phrenia (age 18 to 55).⁴ Eligible participants had:

• An acute exacerbation of schizophrenia that occurred ≤8 weeks before the screening visit and would have benefited from psychiatric hospitalization or continued hospitalization
• Positive and Negative Syndrome Scale (PANSS) total score between 80 and 120 at visit 1 and a score of >4 on at least 2 of the following 4 items: hallucinatory behavior, delusions, conceptual disorganization, or suspiciousness/persecution
• The diagnosis of acute exacerbation of schizophrenia and PANSS total score were confirmed through an independent video-conference interview conducted by an experienced rater.

Participants were excluded if they:

• Experienced a ≥20% improvement in PANSS total score between the initial screening visit and the first injection
• had been treated at any time with clozapine for treatment-resistant schizophrenia
• had met DSM-IV-TR criteria for substance dependence (with the exception of nicotine or caffeine) before screening.

During the initial screening visit, participants received a 0.25-mg tablet of oral risperidone on 2 consecutive days to assess the tolerability of risperidone.

Outcome. Participants were randomized in a 1:1:1 manner to placebo (n = 112) or 1 of 2 monthly doses of RBP-7000: 90 mg (n = 111) or 120 mg (n = 114). Using the least squares means of repeated-measures changes from baseline in PANSS total scores, there was a significant improvement in the difference in PANSS total scores from baseline to the end of the study compared with placebo: 90-mg RBP-7000, -6.148 points (95% confidence interval [CI], -9.982 to -2.314, P = .0004); 120-mg RBP-7000, -7.237 points (95% CI, -11.045 to -3.429, P < .0001). The absolute change from baseline in PANSS total score was -15.367 points for the 90-mg dose and -16.456 points for the 120-mg dose.⁴ Completion rates across all 3 arms were comparable: placebo 70.6%, RBP-7000 90 mg 77.6%, and RBP-7000 120 mg 71.4%.

Tolerability. In the 8-week phase III efficacy trial of RBP-7000, adverse effects occurring with an incidence ≥5% and at least twice the rate of placebo were weight gain (placebo 3.4%, 90 mg 13.0%, 120 mg 12.8%) and sedation (placebo 0%, 90 mg 7.0%, 120 mg 7.7%).¹⁰ Compared with baseline, participants had a mean weight gain at the end of the study of 2.83 kg in the placebo group, 5.15 kg in the 90-mg RBP-7000 group, and 4.69 kg in the 120-mg RBP-7000 group. There were no clinically significant differences at study endpoint in glucose and lipid parameters. Consistent with the known effects of risperidone, there were increases in mean prolactin levels during the 8-week study, the effects of which were greater for women. For men, mean prolactin levels from baseline to study end were: placebo: 9.8 ± 7.9 vs 9.9 ± 8.0 ng/mL; 90 mg: 8.9 ± 6.9 vs 22.4 ± 11.2 ng/mL; and 120 mg: 8.2 ± 5.2 vs 31.3 ± 14.8 ng/mL. For women, mean prolactin levels from baseline to study end were: placebo: 12.8 ± 11.7 vs 10.4 ± 8.0 ng/mL; 90 mg: 7.7 ± 5.3 vs 60.3 ± 46.9 ng/mL; and 120 mg: 10.9 ± 8.6 vs 85.5 ± 55.1 ng/mL. In the pivotal study, discontinuations due to adverse events were low across all treatment groups: 2.5% for placebo vs 0% for 90 mg and 1.7% for 120 mg.⁴ There was no single adverse reaction leading to discontinuation that occurred at a rate of ≥2% and greater than placebo in patients treated with RBP-7000.¹⁰ There were no clinically relevant differences in mean changes from baseline in corrected QT, QRS, and PR intervals, and in heart rate. Similarly, in the 12-month, long-term safety study, there were no clinically relevant changes in mean electrocardiography interval values from baseline to post-dose assessments.¹⁰

Using a 100-point visual analog scale (VAS), injection site pain scores 1 minute after the first dose decreased from a mean of 27 to the range of 3 to 7 for scores obtained 30 to 60 minutes post-dose. In the 12-month long-term safety study, 1-minute post-dose injection site pain VAS scores were highest on Day 1 (mean of 25) and decreased over time with subsequent injections (14 to 16 following last injection).¹⁰

Clinical considerations

Unique properties. RBP-7000 uses the established Atrigel system to provide effective antipsychotic levels in the first week of treatment, without the need for bridging oral coverage or a second loading injection. The abdominal subcutaneous injection volume is relatively small (0.6 mL or 0.8 mL).

Why Rx? The reasons to prescribe RBP-7000 for adult patients with schizophrenia include:

• no oral coverage required at the initiation of treatment
• effective plasma active moiety levels are seen within the first week without the need for a second loading injection
• monthly injection schedule.

Dosing. The recommended dosage of RBP-7000 is 90 mg or 120 mg once monthly, equivalent to 3 mg/d or 4 mg/d of oral risperidone, respectively. Oral risperidone tolerability should be established before the first injection. No oral risperidone coverage is required. RBP-7000 has not been studied in patients with renal or hepatic impairment and should be used with caution in these patients. Prior to initiating treatment in these patients, it is advised to carefully titrate up to at least 3 mg/d of oral risperidone. If a patient can tolerate 3 mg/d of oral risperidone and is psychiatrically stable, then the 90-mg dose of RBP-7000 can be considered.¹⁰ 

Contraindications. The only contraindications for RBP-7000 are known hypersensitivity to risperidone, paliperidone (9-OH risperidone), or other components of the injection.

Bottom Line

RBP-7000 (Perseris) is the second long-acting injectable (LAI) form of risperidone approved in the U.S. Unlike risperidone microspheres (Consta), RBP-7000 does not require any oral risperidone coverage at the beginning of therapy, provides effective drug levels within the first week of treatment with a single injection, and uses a monthly dosing interval. RBP-7000 does not require loading upon initiation. The monthly injection is <1 mL, is administered in abdominal subcutaneous tissue, and uses the Atrigel system.

Related Resource

• Carpenter J, Wong KK. Long-acting injectable antipsychotics: What to do about missed doses. Current Psychiatry. 2018;17(7):10-12,14-19,56.
   

Drug Brand Names
Aripiprazole • Abilify
Carbamazepine • Carbatrol, Tegretol
Doxycycline • Atridox
Leuprolide acetate injectable suspension • Eligard
Paliperidone palmitate • Invega Sustenna
Risperidone • Risperdal
Risperidone extended-release injectable suspension • Perseris
Risperidone long-acting injection • Risperdal Consta

Oral antipsychotic nonadherence is a significant contributor to relapse in patients with schizophrenia spectrum disorders. Long-acting injectable (LAI) antipsychotics have been developed to provide sustained antipsychotic exposure, with evidence that use of LAIs significantly reduces hospitalization rates.¹ One limiting factor in transitioning patients to certain LAIs is the need for prolonged oral coverage at the onset of treatment for agents that cannot be loaded. Nonadherence with this bridging oral therapy places the patient at risk for symptom exacerbation until effective antipsychotic plasma levels are achieved from the LAI.² Although risperidone is one of the more widely used antipsychotics for treating schizophrenia, until recently the only available LAI preparation, risperidone microspheres (Risperdal Consta), required 3 weeks of oral coverage upon initiation.³

Clinical implications

Oral medication nonadherence remains a significant public health issue for patients with schizophrenia, with an estimated 50% of patients failing to achieve 80% adherence even when enrolled in clinical trials specifically designed to track adherence.⁵ Although LAI atypical antipsychotics have been available since the approval of Risperdal Consta, the LAI form of risperidone, and both LAI forms of aripiprazole, were not designed to be loaded. A 1-day initiation regimen for aripiprazole lauroxil has been developed to avoid the need for 3 weeks of oral medication coverage,^6,7 but aripiprazole monohydrate and risperidone microspheres mandate oral bridging of 2 and 3 weeks, respectively.² Because one of the primary indications for LAI antipsychotic therapy is oral medication nonadherence, this prolonged period of oral coverage creates a risk for symptom exacerbation when the bridging period occurs outside of a controlled setting, as is common when patients are discharged from inpatient hospitalization.

One solution to this problem has its antecedents in the development of the Atrigel biodegradable injectable polymer, which was designed to deliver prolonged medication exposure after subcutaneous injection.⁸ This biodegradable polymer drug delivery system suspends and dissolves the medication of interest (in this case, risperidone) in a poly DL-lactide-coglycolide gel and its biocompatible carrier.⁹ The viscous liquid undergoes a phase transition upon contact with tissue fluids after subcutaneous injection, resulting in an implant that releases risperidone in a controlled manner as it is resorbed. Importantly, the kinetic parameters of RBP-7000 are such that effective drug levels are seen within the first week without the need for oral coverage.¹⁰

Use in adults with schizophrenia. After establishing tolerability with oral risperidone, the recommended doses are 90 mg or 120 mg monthly, which correspond to oral daily risperidone doses of 3 mg or 4 mg. RBP-7000 must be administered as a subcutaneous abdominal injection by a health care professional. It is recommended that the patient be in the supine position for the injection and that the injection sites be rotated monthly among 4 quadrants in the abdominal region. The injection volumes for the 90 mg and 120 mg doses are 0.6 mL and 0.8 mL, respectively.¹⁰ As the gel implant becomes firmer, the patient will notice a lump for several weeks that will decrease in size over time. Patients should be advised not to rub or massage the injection site, and to be aware of the placement of any belts or clothing with waistbands.¹⁰

Pharmacologic profile, adverse reactions

Risperidone is an atypical antipsychotic that has been commercially available in the U.S. since December 29, 1993, and its adverse effect profile is well characterized. The most common adverse effects associated with risperidone include those related to dopamine D2 antagonism, metabolic adverse effects, and an increase in serum prolactin. In the 12-month long-term safety study of RBP-7000, 1-minute post-dose injection site pain scores (on a 100-point scale) were highest on Day 1 (mean of 25) and decreased over time with subsequent injections (14 to 16 following the last injection).¹⁰

Continue to: How the Atrigel system works

The kinetics for RBP-7000 are markedly different than those for oral risperidone (Figure 2¹⁵). After a single subcutaneous injection, RBP-7000 shows 2 absorption peaks for risperidone. The first lower peak occurs with a T_max of 4 to 6 hours due to initial release of risperidone during the implant formation process; a second risperidone peak occurs after 10 to 14 days and is associated with slow release from the subcutaneous depot.^9,16,17 For both 9-OH risperidone levels and the total active moiety (risperidone plus 9-OH risperidone levels), the median T_max of the first peak ranges from 4 to 48 hours and the second peak ranges from 7 to 11 days. Following a single subcutaneous injection of RBP-7000, the apparent terminal half-life of risperidone ranges from 9 to 11 days, on average. The mean apparent terminal half-life of the active moiety ranges from 8 to 9 days.^9,16,17 Based on population pharmacokinetic modeling, the 90 mg and 120 mg doses of RBP-7000 are estimated to provide drug exposure equivalent to 3 mg/d and 4 mg/d of oral risperidone, respectively.^9,16,17

Continue to: Efficacy of RBP-7000

Efficacy of RBP-7000 was established in an 8-week, double-blind, placebo-controlled trial of adult patients experiencing an acute exacerbation of schizo­phrenia (age 18 to 55).⁴ Eligible participants had:

• An acute exacerbation of schizophrenia that occurred ≤8 weeks before the screening visit and would have benefited from psychiatric hospitalization or continued hospitalization
• Positive and Negative Syndrome Scale (PANSS) total score between 80 and 120 at visit 1 and a score of >4 on at least 2 of the following 4 items: hallucinatory behavior, delusions, conceptual disorganization, or suspiciousness/persecution
• The diagnosis of acute exacerbation of schizophrenia and PANSS total score were confirmed through an independent video-conference interview conducted by an experienced rater.

Participants were excluded if they:

• Experienced a ≥20% improvement in PANSS total score between the initial screening visit and the first injection
• had been treated at any time with clozapine for treatment-resistant schizophrenia
• had met DSM-IV-TR criteria for substance dependence (with the exception of nicotine or caffeine) before screening.

During the initial screening visit, participants received a 0.25-mg tablet of oral risperidone on 2 consecutive days to assess the tolerability of risperidone.

Outcome. Participants were randomized in a 1:1:1 manner to placebo (n = 112) or 1 of 2 monthly doses of RBP-7000: 90 mg (n = 111) or 120 mg (n = 114). Using the least squares means of repeated-measures changes from baseline in PANSS total scores, there was a significant improvement in the difference in PANSS total scores from baseline to the end of the study compared with placebo: 90-mg RBP-7000, -6.148 points (95% confidence interval [CI], -9.982 to -2.314, P = .0004); 120-mg RBP-7000, -7.237 points (95% CI, -11.045 to -3.429, P < .0001). The absolute change from baseline in PANSS total score was -15.367 points for the 90-mg dose and -16.456 points for the 120-mg dose.⁴ Completion rates across all 3 arms were comparable: placebo 70.6%, RBP-7000 90 mg 77.6%, and RBP-7000 120 mg 71.4%.

Tolerability. In the 8-week phase III efficacy trial of RBP-7000, adverse effects occurring with an incidence ≥5% and at least twice the rate of placebo were weight gain (placebo 3.4%, 90 mg 13.0%, 120 mg 12.8%) and sedation (placebo 0%, 90 mg 7.0%, 120 mg 7.7%).¹⁰ Compared with baseline, participants had a mean weight gain at the end of the study of 2.83 kg in the placebo group, 5.15 kg in the 90-mg RBP-7000 group, and 4.69 kg in the 120-mg RBP-7000 group. There were no clinically significant differences at study endpoint in glucose and lipid parameters. Consistent with the known effects of risperidone, there were increases in mean prolactin levels during the 8-week study, the effects of which were greater for women. For men, mean prolactin levels from baseline to study end were: placebo: 9.8 ± 7.9 vs 9.9 ± 8.0 ng/mL; 90 mg: 8.9 ± 6.9 vs 22.4 ± 11.2 ng/mL; and 120 mg: 8.2 ± 5.2 vs 31.3 ± 14.8 ng/mL. For women, mean prolactin levels from baseline to study end were: placebo: 12.8 ± 11.7 vs 10.4 ± 8.0 ng/mL; 90 mg: 7.7 ± 5.3 vs 60.3 ± 46.9 ng/mL; and 120 mg: 10.9 ± 8.6 vs 85.5 ± 55.1 ng/mL. In the pivotal study, discontinuations due to adverse events were low across all treatment groups: 2.5% for placebo vs 0% for 90 mg and 1.7% for 120 mg.⁴ There was no single adverse reaction leading to discontinuation that occurred at a rate of ≥2% and greater than placebo in patients treated with RBP-7000.¹⁰ There were no clinically relevant differences in mean changes from baseline in corrected QT, QRS, and PR intervals, and in heart rate. Similarly, in the 12-month, long-term safety study, there were no clinically relevant changes in mean electrocardiography interval values from baseline to post-dose assessments.¹⁰

Using a 100-point visual analog scale (VAS), injection site pain scores 1 minute after the first dose decreased from a mean of 27 to the range of 3 to 7 for scores obtained 30 to 60 minutes post-dose. In the 12-month long-term safety study, 1-minute post-dose injection site pain VAS scores were highest on Day 1 (mean of 25) and decreased over time with subsequent injections (14 to 16 following last injection).¹⁰

Clinical considerations

Unique properties. RBP-7000 uses the established Atrigel system to provide effective antipsychotic levels in the first week of treatment, without the need for bridging oral coverage or a second loading injection. The abdominal subcutaneous injection volume is relatively small (0.6 mL or 0.8 mL).

Why Rx? The reasons to prescribe RBP-7000 for adult patients with schizophrenia include:

• no oral coverage required at the initiation of treatment
• effective plasma active moiety levels are seen within the first week without the need for a second loading injection
• monthly injection schedule.

Dosing. The recommended dosage of RBP-7000 is 90 mg or 120 mg once monthly, equivalent to 3 mg/d or 4 mg/d of oral risperidone, respectively. Oral risperidone tolerability should be established before the first injection. No oral risperidone coverage is required. RBP-7000 has not been studied in patients with renal or hepatic impairment and should be used with caution in these patients. Prior to initiating treatment in these patients, it is advised to carefully titrate up to at least 3 mg/d of oral risperidone. If a patient can tolerate 3 mg/d of oral risperidone and is psychiatrically stable, then the 90-mg dose of RBP-7000 can be considered.¹⁰ 

Contraindications. The only contraindications for RBP-7000 are known hypersensitivity to risperidone, paliperidone (9-OH risperidone), or other components of the injection.

Bottom Line

RBP-7000 (Perseris) is the second long-acting injectable (LAI) form of risperidone approved in the U.S. Unlike risperidone microspheres (Consta), RBP-7000 does not require any oral risperidone coverage at the beginning of therapy, provides effective drug levels within the first week of treatment with a single injection, and uses a monthly dosing interval. RBP-7000 does not require loading upon initiation. The monthly injection is <1 mL, is administered in abdominal subcutaneous tissue, and uses the Atrigel system.

Related Resource

• Carpenter J, Wong KK. Long-acting injectable antipsychotics: What to do about missed doses. Current Psychiatry. 2018;17(7):10-12,14-19,56.
   

Drug Brand Names
Aripiprazole • Abilify
Carbamazepine • Carbatrol, Tegretol
Doxycycline • Atridox
Leuprolide acetate injectable suspension • Eligard
Paliperidone palmitate • Invega Sustenna
Risperidone • Risperdal
Risperidone extended-release injectable suspension • Perseris
Risperidone long-acting injection • Risperdal Consta

Oral antipsychotic nonadherence is a significant contributor to relapse in patients with schizophrenia spectrum disorders. Long-acting injectable (LAI) antipsychotics have been developed to provide sustained antipsychotic exposure, with evidence that use of LAIs significantly reduces hospitalization rates.¹ One limiting factor in transitioning patients to certain LAIs is the need for prolonged oral coverage at the onset of treatment for agents that cannot be loaded. Nonadherence with this bridging oral therapy places the patient at risk for symptom exacerbation until effective antipsychotic plasma levels are achieved from the LAI.² Although risperidone is one of the more widely used antipsychotics for treating schizophrenia, until recently the only available LAI preparation, risperidone microspheres (Risperdal Consta), required 3 weeks of oral coverage upon initiation.³

Clinical implications

Oral medication nonadherence remains a significant public health issue for patients with schizophrenia, with an estimated 50% of patients failing to achieve 80% adherence even when enrolled in clinical trials specifically designed to track adherence.⁵ Although LAI atypical antipsychotics have been available since the approval of Risperdal Consta, the LAI form of risperidone, and both LAI forms of aripiprazole, were not designed to be loaded. A 1-day initiation regimen for aripiprazole lauroxil has been developed to avoid the need for 3 weeks of oral medication coverage,^6,7 but aripiprazole monohydrate and risperidone microspheres mandate oral bridging of 2 and 3 weeks, respectively.² Because one of the primary indications for LAI antipsychotic therapy is oral medication nonadherence, this prolonged period of oral coverage creates a risk for symptom exacerbation when the bridging period occurs outside of a controlled setting, as is common when patients are discharged from inpatient hospitalization.

One solution to this problem has its antecedents in the development of the Atrigel biodegradable injectable polymer, which was designed to deliver prolonged medication exposure after subcutaneous injection.⁸ This biodegradable polymer drug delivery system suspends and dissolves the medication of interest (in this case, risperidone) in a poly DL-lactide-coglycolide gel and its biocompatible carrier.⁹ The viscous liquid undergoes a phase transition upon contact with tissue fluids after subcutaneous injection, resulting in an implant that releases risperidone in a controlled manner as it is resorbed. Importantly, the kinetic parameters of RBP-7000 are such that effective drug levels are seen within the first week without the need for oral coverage.¹⁰

Use in adults with schizophrenia. After establishing tolerability with oral risperidone, the recommended doses are 90 mg or 120 mg monthly, which correspond to oral daily risperidone doses of 3 mg or 4 mg. RBP-7000 must be administered as a subcutaneous abdominal injection by a health care professional. It is recommended that the patient be in the supine position for the injection and that the injection sites be rotated monthly among 4 quadrants in the abdominal region. The injection volumes for the 90 mg and 120 mg doses are 0.6 mL and 0.8 mL, respectively.¹⁰ As the gel implant becomes firmer, the patient will notice a lump for several weeks that will decrease in size over time. Patients should be advised not to rub or massage the injection site, and to be aware of the placement of any belts or clothing with waistbands.¹⁰

Pharmacologic profile, adverse reactions

Risperidone is an atypical antipsychotic that has been commercially available in the U.S. since December 29, 1993, and its adverse effect profile is well characterized. The most common adverse effects associated with risperidone include those related to dopamine D2 antagonism, metabolic adverse effects, and an increase in serum prolactin. In the 12-month long-term safety study of RBP-7000, 1-minute post-dose injection site pain scores (on a 100-point scale) were highest on Day 1 (mean of 25) and decreased over time with subsequent injections (14 to 16 following the last injection).¹⁰

Continue to: How the Atrigel system works

The kinetics for RBP-7000 are markedly different than those for oral risperidone (Figure 2¹⁵). After a single subcutaneous injection, RBP-7000 shows 2 absorption peaks for risperidone. The first lower peak occurs with a T_max of 4 to 6 hours due to initial release of risperidone during the implant formation process; a second risperidone peak occurs after 10 to 14 days and is associated with slow release from the subcutaneous depot.^9,16,17 For both 9-OH risperidone levels and the total active moiety (risperidone plus 9-OH risperidone levels), the median T_max of the first peak ranges from 4 to 48 hours and the second peak ranges from 7 to 11 days. Following a single subcutaneous injection of RBP-7000, the apparent terminal half-life of risperidone ranges from 9 to 11 days, on average. The mean apparent terminal half-life of the active moiety ranges from 8 to 9 days.^9,16,17 Based on population pharmacokinetic modeling, the 90 mg and 120 mg doses of RBP-7000 are estimated to provide drug exposure equivalent to 3 mg/d and 4 mg/d of oral risperidone, respectively.^9,16,17

Continue to: Efficacy of RBP-7000

Efficacy of RBP-7000 was established in an 8-week, double-blind, placebo-controlled trial of adult patients experiencing an acute exacerbation of schizo­phrenia (age 18 to 55).⁴ Eligible participants had:

• An acute exacerbation of schizophrenia that occurred ≤8 weeks before the screening visit and would have benefited from psychiatric hospitalization or continued hospitalization
• Positive and Negative Syndrome Scale (PANSS) total score between 80 and 120 at visit 1 and a score of >4 on at least 2 of the following 4 items: hallucinatory behavior, delusions, conceptual disorganization, or suspiciousness/persecution
• The diagnosis of acute exacerbation of schizophrenia and PANSS total score were confirmed through an independent video-conference interview conducted by an experienced rater.

Participants were excluded if they:

• Experienced a ≥20% improvement in PANSS total score between the initial screening visit and the first injection
• had been treated at any time with clozapine for treatment-resistant schizophrenia
• had met DSM-IV-TR criteria for substance dependence (with the exception of nicotine or caffeine) before screening.

During the initial screening visit, participants received a 0.25-mg tablet of oral risperidone on 2 consecutive days to assess the tolerability of risperidone.

Outcome. Participants were randomized in a 1:1:1 manner to placebo (n = 112) or 1 of 2 monthly doses of RBP-7000: 90 mg (n = 111) or 120 mg (n = 114). Using the least squares means of repeated-measures changes from baseline in PANSS total scores, there was a significant improvement in the difference in PANSS total scores from baseline to the end of the study compared with placebo: 90-mg RBP-7000, -6.148 points (95% confidence interval [CI], -9.982 to -2.314, P = .0004); 120-mg RBP-7000, -7.237 points (95% CI, -11.045 to -3.429, P < .0001). The absolute change from baseline in PANSS total score was -15.367 points for the 90-mg dose and -16.456 points for the 120-mg dose.⁴ Completion rates across all 3 arms were comparable: placebo 70.6%, RBP-7000 90 mg 77.6%, and RBP-7000 120 mg 71.4%.

Tolerability. In the 8-week phase III efficacy trial of RBP-7000, adverse effects occurring with an incidence ≥5% and at least twice the rate of placebo were weight gain (placebo 3.4%, 90 mg 13.0%, 120 mg 12.8%) and sedation (placebo 0%, 90 mg 7.0%, 120 mg 7.7%).¹⁰ Compared with baseline, participants had a mean weight gain at the end of the study of 2.83 kg in the placebo group, 5.15 kg in the 90-mg RBP-7000 group, and 4.69 kg in the 120-mg RBP-7000 group. There were no clinically significant differences at study endpoint in glucose and lipid parameters. Consistent with the known effects of risperidone, there were increases in mean prolactin levels during the 8-week study, the effects of which were greater for women. For men, mean prolactin levels from baseline to study end were: placebo: 9.8 ± 7.9 vs 9.9 ± 8.0 ng/mL; 90 mg: 8.9 ± 6.9 vs 22.4 ± 11.2 ng/mL; and 120 mg: 8.2 ± 5.2 vs 31.3 ± 14.8 ng/mL. For women, mean prolactin levels from baseline to study end were: placebo: 12.8 ± 11.7 vs 10.4 ± 8.0 ng/mL; 90 mg: 7.7 ± 5.3 vs 60.3 ± 46.9 ng/mL; and 120 mg: 10.9 ± 8.6 vs 85.5 ± 55.1 ng/mL. In the pivotal study, discontinuations due to adverse events were low across all treatment groups: 2.5% for placebo vs 0% for 90 mg and 1.7% for 120 mg.⁴ There was no single adverse reaction leading to discontinuation that occurred at a rate of ≥2% and greater than placebo in patients treated with RBP-7000.¹⁰ There were no clinically relevant differences in mean changes from baseline in corrected QT, QRS, and PR intervals, and in heart rate. Similarly, in the 12-month, long-term safety study, there were no clinically relevant changes in mean electrocardiography interval values from baseline to post-dose assessments.¹⁰

Using a 100-point visual analog scale (VAS), injection site pain scores 1 minute after the first dose decreased from a mean of 27 to the range of 3 to 7 for scores obtained 30 to 60 minutes post-dose. In the 12-month long-term safety study, 1-minute post-dose injection site pain VAS scores were highest on Day 1 (mean of 25) and decreased over time with subsequent injections (14 to 16 following last injection).¹⁰

Clinical considerations

Unique properties. RBP-7000 uses the established Atrigel system to provide effective antipsychotic levels in the first week of treatment, without the need for bridging oral coverage or a second loading injection. The abdominal subcutaneous injection volume is relatively small (0.6 mL or 0.8 mL).

Why Rx? The reasons to prescribe RBP-7000 for adult patients with schizophrenia include:

• no oral coverage required at the initiation of treatment
• effective plasma active moiety levels are seen within the first week without the need for a second loading injection
• monthly injection schedule.

Dosing. The recommended dosage of RBP-7000 is 90 mg or 120 mg once monthly, equivalent to 3 mg/d or 4 mg/d of oral risperidone, respectively. Oral risperidone tolerability should be established before the first injection. No oral risperidone coverage is required. RBP-7000 has not been studied in patients with renal or hepatic impairment and should be used with caution in these patients. Prior to initiating treatment in these patients, it is advised to carefully titrate up to at least 3 mg/d of oral risperidone. If a patient can tolerate 3 mg/d of oral risperidone and is psychiatrically stable, then the 90-mg dose of RBP-7000 can be considered.¹⁰ 

Contraindications. The only contraindications for RBP-7000 are known hypersensitivity to risperidone, paliperidone (9-OH risperidone), or other components of the injection.

Bottom Line

RBP-7000 (Perseris) is the second long-acting injectable (LAI) form of risperidone approved in the U.S. Unlike risperidone microspheres (Consta), RBP-7000 does not require any oral risperidone coverage at the beginning of therapy, provides effective drug levels within the first week of treatment with a single injection, and uses a monthly dosing interval. RBP-7000 does not require loading upon initiation. The monthly injection is <1 mL, is administered in abdominal subcutaneous tissue, and uses the Atrigel system.

Related Resource

• Carpenter J, Wong KK. Long-acting injectable antipsychotics: What to do about missed doses. Current Psychiatry. 2018;17(7):10-12,14-19,56.
   

Drug Brand Names
Aripiprazole • Abilify
Carbamazepine • Carbatrol, Tegretol
Doxycycline • Atridox
Leuprolide acetate injectable suspension • Eligard
Paliperidone palmitate • Invega Sustenna
Risperidone • Risperdal
Risperidone extended-release injectable suspension • Perseris
Risperidone long-acting injection • Risperdal Consta

Clinical implications

Nonadherence with oral antipsychotics is a common problem for patients with schizophrenia, one that is often underappreciated by clinicians.⁵ Whether one uses 70% or 80% as the measure of oral medication adherence, at least 50% of schizophrenia patients are nonadherent, with resultant increased risks for symptom exacerbation and hospitalization.^5,6 Although 2 LAI forms of aripiprazole have been introduced over the past few years, neither was designed to be loaded, resulting in the need for 2 or 3 weeks of oral antipsychotic coverage following the first injectable dose.¹ The primary reason for LAI antipsychotic therapy is oral medication nonadherence, and thus the need for 14 to 21 days of oral coverage at the outset of treatment creates a risk for symptom exacerbation if the patient is nonadherent with this oral bridging therapy which is needed to achieve the necessary serum concentrations until the long-acting formulation takes over.

One approach was to create a new form of AL using smaller nanomolecular particles rather than the micron-sized particles used for maintenance AL injections.^3,4 This nanocrystal suspension is called Aristada Initio (ALncd) and has a median T_max that ranges from 16 to 35 days, compared with 41 days for single-dose injections of AL. ALncd also has a much shorter median half-life of 15 to 18 days, compared with 53.9 to 57.2 days for AL (Table 2^7,8). Utilizing these kinetic differences, a 1-day initiation regimen was developed to dispense with the need for 3 weeks of oral medication coverage when commencing AL treatment.^3,4 In lieu of 3 weeks of oral coverage starting at the time of the first AL injection, patients instead will receive an injection of ALncd 675 mg, and a single oral 30 mg aripiprazole dose. The combination of ALncd and the single 30 mg oral dose when added to the initial AL injection provides aripiprazole levels in the first weeks of therapy that are comparable to those seen in the previous paradigm, when patients took 21 days of oral aripiprazole after the first AL injection.³

Use in adults with schizophrenia. After establishing tolerability with oral aripiprazole, ALncd675 mg is administered as an IM injection by a health care professional, and the patient is concomitantly given a single 30 mg oral dose of aripiprazole. Only one dosage form of ALncd is available: 675 mg. The maintenance AL dose chosen by the clinician (441, 662, 882, or 1,064 mg) is also administered at the same time, but must be injected in the other deltoid or gluteal muscle. The injection volume for ALncd is 2.4 mL and can be administered in the deltoid or gluteus muscle.⁹ If the patient prefers not to have 2 injections on the same day, the AL dose can be administered up to 10 days thereafter.⁹ This 10-day window for administering AL relates to the long time to maximum plasma levels from single AL injections. The relevant drug levels during the first weeks are provided predominantly from the initiation regimen of ALncd injection plus the single 30 mg oral dose.³ In instances when a patient agrees to receive both ALncd and AL injections but refuses the 30 mg oral dose, effective plasma levels will be seen in the middle of the second week of therapy.

Continue to: Pharmacologic profile, adverse reactions

Aripiprazole is a dopamine partial agonist atypical antipsychotic that has been commercially available in the United States since November 15, 2002, and its adverse effect profile is well characterized. The LAI formulation AL was approved on October 5, 2015. In the pivotal, 12-week, fixed-dose, placebo-controlled clinical trial of AL 441 mg or 882 mg monthly for adults with an acute exacerbation of schizophrenia, the only adverse effect that occurred in ≥5% of AL-treated patients and a rate at least twice that of placebo was akathisia (441 mg: 11%; 882 mg: 11%; placebo: 4%).¹⁰ Only 2 of 415 AL-treated patients discontinued the study due to akathisia. Injection-site reactions were reported by 4%, 5%, and 2% of patients treated with AL 441 mg, AL 882 mg, and placebo, respectively. Most of these were injection-site pain associated with the first injection, and decreased with each subsequent injection. Other injection-site reactions (induration, swelling, and redness) occurred at rates <1%.¹¹

Having established that the range of plasma aripiprazole levels consistent with effective treatment is bounded by levels seen with AL 441 mg or 882 mg monthly, the FDA did not require additional efficacy studies for new AL doses provided that pharmacokinetic (PK) studies could demonstrate levels within the effective range. This is consistent with how new doses of other LAI antipsychotic preparations have been addressed in the past. For example, the 37.5 mg dose of risperidone microspheres was approved based on PK data, although the pivotal efficacy trials included doses of 25 mg, 50 mg, and 75 mg.¹² Based on PK studies, AL doses of 662 mg monthly, 882 mg every 6 weeks, and 1,064 mg every 8 weeks were previously approved.¹³ The approval process for ALncd followed a similar pathway, and is based on PK results combined with tolerability data amassed during the PK studies. The package insert thus notes that in PK studies the safety profile of ALncd was generally consistent with that observed for AL (see Tolerability).

Pharmacokinetic outcomes. A comparative phase 1 PK study was performed to evaluate initiation regimens: either 21 days of oral aripiprazole (15 mg/d) and one AL dose (n = 81) or one injection of ALncd plus a single dose of 30 mg oral aripiprazole and one AL dose (n = 80). Patients were randomized 1:1:1:1 to receive an AL dose of either 441 mg or 882 mg combined with the oral or the new ALncdinitiation regimen. As shown in Figure 1 and Figure 2, the mean plasma levels seen with 675 mg IM ALncd plus a single dose of 30 mg oral aripiprazole were comparable with levels achieved using 21 days of oral aripiprazole coverage, regardless of whether the regimen was paired with a maintenance AL dose of 441 mg or 882 mg.⁴

Tolerability. In PK studies, the safety profile and incidences of injection site reactions of ALncd were generally consistent with those observed for aripiprazole lauroxil.⁹ In the phase I PK study comparing oral initiation with ALncd plus a single 30 mg oral aripiprazole dose, there were 2 mild cases of akathisia in the 21-day oral aripiprazole groups (n = 81) and 4 cases in the ALncd groups (n = 80) (3 mild cases, 1 moderate case). None of the adverse events related to akathisia were deemed serious, and no patients discontinued participation in the trial due to akathisia.⁹

Continue to: Clinical considerations

ALncd is not a substitute for AL due to the very different kinetic properties of the 2 preparations. ALncd is approved only to be used for initiating treatment with AL, or in those instances where the revised missed dose guidelines for AL permit use of ALncd to obviate the need for oral coverage.⁹ Table 3⁹ presents these revised AL missed dose guidelines focusing on those time periods when some form of supplementation is required in addition to the established maintenance AL dose. Clinicians should be reminded that ALncd must be paired with a dose of AL, although the latter can be given up to 10 days later when commencing therapy.

• it obviates the need for 21 days of oral coverage previously required at the initiation of AL treatment
• clinically relevant plasma levels are seen within the first week when ALncd is combined with a single 30 mg oral aripiprazole dose
• per the revised missed dose guidelines for AL, it can be used in those situations that previously demanded 7 days of oral coverage, and, when combined with a single 30 mg oral dose, can be used for resumption of therapy after prolonged absences that required 21 days of oral coverage. In all instances, the patient will also receive their usual maintenance dose of AL.

Dosing. There is only one dose available for ALncd, 675 mg IM. As the dose cannot be modified, the package insert contains cautionary language regarding situations with less-than-expected drug exposure (use of cytochrome P450 [CYP] 3A4 inducers), greater-than-expected drug exposure (strong CYP3A4 or 2D6 inhibitors or known 2D6 poor metabolizers), or increased pharmacodynamic effects (concurrent use of antihypertensives or benzodiazepines).

Contraindications. The only contraindication is a known hypersensitivity to aripiprazole.

Bottom Line

Aripiprazole lauroxil nanocrystal suspension (Aristada Initio) was specifically developed to obviate the need for 21 days of oral aripiprazole coverage when commencing treatment with aripiprazole lauroxil (Aristada). The plasma levels achieved when an injection of aripiprazole lauroxil nanocrystal suspension is combined with a single 30 mg oral dose are comparable to those achieved with 21 days of oral coverage. This initiation regimen, including a aripiprazole lauroxil nanocrystal injection and a 30 mg oral dose, should be administered on the same day as the maintenance aripiprazole lauroxil injection, although the latter can be administered on any of the next 10 days.

Related Resource

• Khan AY, Ovais DM. Long-acting injectable aripiprazole lauroxil for schizophrenia. Current Psychiatry. 2016;15(7):50-52,58.

Drug Brand Names
Aripiprazole lauroxil • Aristada
Aripiprazole lauroxil nanocrystal • Aristada Initio
Risperidone microspheres • Risperdal Consta

Clinical implications

Nonadherence with oral antipsychotics is a common problem for patients with schizophrenia, one that is often underappreciated by clinicians.⁵ Whether one uses 70% or 80% as the measure of oral medication adherence, at least 50% of schizophrenia patients are nonadherent, with resultant increased risks for symptom exacerbation and hospitalization.^5,6 Although 2 LAI forms of aripiprazole have been introduced over the past few years, neither was designed to be loaded, resulting in the need for 2 or 3 weeks of oral antipsychotic coverage following the first injectable dose.¹ The primary reason for LAI antipsychotic therapy is oral medication nonadherence, and thus the need for 14 to 21 days of oral coverage at the outset of treatment creates a risk for symptom exacerbation if the patient is nonadherent with this oral bridging therapy which is needed to achieve the necessary serum concentrations until the long-acting formulation takes over.

One approach was to create a new form of AL using smaller nanomolecular particles rather than the micron-sized particles used for maintenance AL injections.^3,4 This nanocrystal suspension is called Aristada Initio (ALncd) and has a median T_max that ranges from 16 to 35 days, compared with 41 days for single-dose injections of AL. ALncd also has a much shorter median half-life of 15 to 18 days, compared with 53.9 to 57.2 days for AL (Table 2^7,8). Utilizing these kinetic differences, a 1-day initiation regimen was developed to dispense with the need for 3 weeks of oral medication coverage when commencing AL treatment.^3,4 In lieu of 3 weeks of oral coverage starting at the time of the first AL injection, patients instead will receive an injection of ALncd 675 mg, and a single oral 30 mg aripiprazole dose. The combination of ALncd and the single 30 mg oral dose when added to the initial AL injection provides aripiprazole levels in the first weeks of therapy that are comparable to those seen in the previous paradigm, when patients took 21 days of oral aripiprazole after the first AL injection.³

Use in adults with schizophrenia. After establishing tolerability with oral aripiprazole, ALncd675 mg is administered as an IM injection by a health care professional, and the patient is concomitantly given a single 30 mg oral dose of aripiprazole. Only one dosage form of ALncd is available: 675 mg. The maintenance AL dose chosen by the clinician (441, 662, 882, or 1,064 mg) is also administered at the same time, but must be injected in the other deltoid or gluteal muscle. The injection volume for ALncd is 2.4 mL and can be administered in the deltoid or gluteus muscle.⁹ If the patient prefers not to have 2 injections on the same day, the AL dose can be administered up to 10 days thereafter.⁹ This 10-day window for administering AL relates to the long time to maximum plasma levels from single AL injections. The relevant drug levels during the first weeks are provided predominantly from the initiation regimen of ALncd injection plus the single 30 mg oral dose.³ In instances when a patient agrees to receive both ALncd and AL injections but refuses the 30 mg oral dose, effective plasma levels will be seen in the middle of the second week of therapy.

Continue to: Pharmacologic profile, adverse reactions

Aripiprazole is a dopamine partial agonist atypical antipsychotic that has been commercially available in the United States since November 15, 2002, and its adverse effect profile is well characterized. The LAI formulation AL was approved on October 5, 2015. In the pivotal, 12-week, fixed-dose, placebo-controlled clinical trial of AL 441 mg or 882 mg monthly for adults with an acute exacerbation of schizophrenia, the only adverse effect that occurred in ≥5% of AL-treated patients and a rate at least twice that of placebo was akathisia (441 mg: 11%; 882 mg: 11%; placebo: 4%).¹⁰ Only 2 of 415 AL-treated patients discontinued the study due to akathisia. Injection-site reactions were reported by 4%, 5%, and 2% of patients treated with AL 441 mg, AL 882 mg, and placebo, respectively. Most of these were injection-site pain associated with the first injection, and decreased with each subsequent injection. Other injection-site reactions (induration, swelling, and redness) occurred at rates <1%.¹¹

Having established that the range of plasma aripiprazole levels consistent with effective treatment is bounded by levels seen with AL 441 mg or 882 mg monthly, the FDA did not require additional efficacy studies for new AL doses provided that pharmacokinetic (PK) studies could demonstrate levels within the effective range. This is consistent with how new doses of other LAI antipsychotic preparations have been addressed in the past. For example, the 37.5 mg dose of risperidone microspheres was approved based on PK data, although the pivotal efficacy trials included doses of 25 mg, 50 mg, and 75 mg.¹² Based on PK studies, AL doses of 662 mg monthly, 882 mg every 6 weeks, and 1,064 mg every 8 weeks were previously approved.¹³ The approval process for ALncd followed a similar pathway, and is based on PK results combined with tolerability data amassed during the PK studies. The package insert thus notes that in PK studies the safety profile of ALncd was generally consistent with that observed for AL (see Tolerability).

Pharmacokinetic outcomes. A comparative phase 1 PK study was performed to evaluate initiation regimens: either 21 days of oral aripiprazole (15 mg/d) and one AL dose (n = 81) or one injection of ALncd plus a single dose of 30 mg oral aripiprazole and one AL dose (n = 80). Patients were randomized 1:1:1:1 to receive an AL dose of either 441 mg or 882 mg combined with the oral or the new ALncdinitiation regimen. As shown in Figure 1 and Figure 2, the mean plasma levels seen with 675 mg IM ALncd plus a single dose of 30 mg oral aripiprazole were comparable with levels achieved using 21 days of oral aripiprazole coverage, regardless of whether the regimen was paired with a maintenance AL dose of 441 mg or 882 mg.⁴

Tolerability. In PK studies, the safety profile and incidences of injection site reactions of ALncd were generally consistent with those observed for aripiprazole lauroxil.⁹ In the phase I PK study comparing oral initiation with ALncd plus a single 30 mg oral aripiprazole dose, there were 2 mild cases of akathisia in the 21-day oral aripiprazole groups (n = 81) and 4 cases in the ALncd groups (n = 80) (3 mild cases, 1 moderate case). None of the adverse events related to akathisia were deemed serious, and no patients discontinued participation in the trial due to akathisia.⁹

Continue to: Clinical considerations

ALncd is not a substitute for AL due to the very different kinetic properties of the 2 preparations. ALncd is approved only to be used for initiating treatment with AL, or in those instances where the revised missed dose guidelines for AL permit use of ALncd to obviate the need for oral coverage.⁹ Table 3⁹ presents these revised AL missed dose guidelines focusing on those time periods when some form of supplementation is required in addition to the established maintenance AL dose. Clinicians should be reminded that ALncd must be paired with a dose of AL, although the latter can be given up to 10 days later when commencing therapy.

• it obviates the need for 21 days of oral coverage previously required at the initiation of AL treatment
• clinically relevant plasma levels are seen within the first week when ALncd is combined with a single 30 mg oral aripiprazole dose
• per the revised missed dose guidelines for AL, it can be used in those situations that previously demanded 7 days of oral coverage, and, when combined with a single 30 mg oral dose, can be used for resumption of therapy after prolonged absences that required 21 days of oral coverage. In all instances, the patient will also receive their usual maintenance dose of AL.

Dosing. There is only one dose available for ALncd, 675 mg IM. As the dose cannot be modified, the package insert contains cautionary language regarding situations with less-than-expected drug exposure (use of cytochrome P450 [CYP] 3A4 inducers), greater-than-expected drug exposure (strong CYP3A4 or 2D6 inhibitors or known 2D6 poor metabolizers), or increased pharmacodynamic effects (concurrent use of antihypertensives or benzodiazepines).

Contraindications. The only contraindication is a known hypersensitivity to aripiprazole.

Bottom Line

Aripiprazole lauroxil nanocrystal suspension (Aristada Initio) was specifically developed to obviate the need for 21 days of oral aripiprazole coverage when commencing treatment with aripiprazole lauroxil (Aristada). The plasma levels achieved when an injection of aripiprazole lauroxil nanocrystal suspension is combined with a single 30 mg oral dose are comparable to those achieved with 21 days of oral coverage. This initiation regimen, including a aripiprazole lauroxil nanocrystal injection and a 30 mg oral dose, should be administered on the same day as the maintenance aripiprazole lauroxil injection, although the latter can be administered on any of the next 10 days.

Related Resource

• Khan AY, Ovais DM. Long-acting injectable aripiprazole lauroxil for schizophrenia. Current Psychiatry. 2016;15(7):50-52,58.

Drug Brand Names
Aripiprazole lauroxil • Aristada
Aripiprazole lauroxil nanocrystal • Aristada Initio
Risperidone microspheres • Risperdal Consta

Clinical implications

Nonadherence with oral antipsychotics is a common problem for patients with schizophrenia, one that is often underappreciated by clinicians.⁵ Whether one uses 70% or 80% as the measure of oral medication adherence, at least 50% of schizophrenia patients are nonadherent, with resultant increased risks for symptom exacerbation and hospitalization.^5,6 Although 2 LAI forms of aripiprazole have been introduced over the past few years, neither was designed to be loaded, resulting in the need for 2 or 3 weeks of oral antipsychotic coverage following the first injectable dose.¹ The primary reason for LAI antipsychotic therapy is oral medication nonadherence, and thus the need for 14 to 21 days of oral coverage at the outset of treatment creates a risk for symptom exacerbation if the patient is nonadherent with this oral bridging therapy which is needed to achieve the necessary serum concentrations until the long-acting formulation takes over.

One approach was to create a new form of AL using smaller nanomolecular particles rather than the micron-sized particles used for maintenance AL injections.^3,4 This nanocrystal suspension is called Aristada Initio (ALncd) and has a median T_max that ranges from 16 to 35 days, compared with 41 days for single-dose injections of AL. ALncd also has a much shorter median half-life of 15 to 18 days, compared with 53.9 to 57.2 days for AL (Table 2^7,8). Utilizing these kinetic differences, a 1-day initiation regimen was developed to dispense with the need for 3 weeks of oral medication coverage when commencing AL treatment.^3,4 In lieu of 3 weeks of oral coverage starting at the time of the first AL injection, patients instead will receive an injection of ALncd 675 mg, and a single oral 30 mg aripiprazole dose. The combination of ALncd and the single 30 mg oral dose when added to the initial AL injection provides aripiprazole levels in the first weeks of therapy that are comparable to those seen in the previous paradigm, when patients took 21 days of oral aripiprazole after the first AL injection.³

Use in adults with schizophrenia. After establishing tolerability with oral aripiprazole, ALncd675 mg is administered as an IM injection by a health care professional, and the patient is concomitantly given a single 30 mg oral dose of aripiprazole. Only one dosage form of ALncd is available: 675 mg. The maintenance AL dose chosen by the clinician (441, 662, 882, or 1,064 mg) is also administered at the same time, but must be injected in the other deltoid or gluteal muscle. The injection volume for ALncd is 2.4 mL and can be administered in the deltoid or gluteus muscle.⁹ If the patient prefers not to have 2 injections on the same day, the AL dose can be administered up to 10 days thereafter.⁹ This 10-day window for administering AL relates to the long time to maximum plasma levels from single AL injections. The relevant drug levels during the first weeks are provided predominantly from the initiation regimen of ALncd injection plus the single 30 mg oral dose.³ In instances when a patient agrees to receive both ALncd and AL injections but refuses the 30 mg oral dose, effective plasma levels will be seen in the middle of the second week of therapy.

Continue to: Pharmacologic profile, adverse reactions

Aripiprazole is a dopamine partial agonist atypical antipsychotic that has been commercially available in the United States since November 15, 2002, and its adverse effect profile is well characterized. The LAI formulation AL was approved on October 5, 2015. In the pivotal, 12-week, fixed-dose, placebo-controlled clinical trial of AL 441 mg or 882 mg monthly for adults with an acute exacerbation of schizophrenia, the only adverse effect that occurred in ≥5% of AL-treated patients and a rate at least twice that of placebo was akathisia (441 mg: 11%; 882 mg: 11%; placebo: 4%).¹⁰ Only 2 of 415 AL-treated patients discontinued the study due to akathisia. Injection-site reactions were reported by 4%, 5%, and 2% of patients treated with AL 441 mg, AL 882 mg, and placebo, respectively. Most of these were injection-site pain associated with the first injection, and decreased with each subsequent injection. Other injection-site reactions (induration, swelling, and redness) occurred at rates <1%.¹¹

Having established that the range of plasma aripiprazole levels consistent with effective treatment is bounded by levels seen with AL 441 mg or 882 mg monthly, the FDA did not require additional efficacy studies for new AL doses provided that pharmacokinetic (PK) studies could demonstrate levels within the effective range. This is consistent with how new doses of other LAI antipsychotic preparations have been addressed in the past. For example, the 37.5 mg dose of risperidone microspheres was approved based on PK data, although the pivotal efficacy trials included doses of 25 mg, 50 mg, and 75 mg.¹² Based on PK studies, AL doses of 662 mg monthly, 882 mg every 6 weeks, and 1,064 mg every 8 weeks were previously approved.¹³ The approval process for ALncd followed a similar pathway, and is based on PK results combined with tolerability data amassed during the PK studies. The package insert thus notes that in PK studies the safety profile of ALncd was generally consistent with that observed for AL (see Tolerability).

Pharmacokinetic outcomes. A comparative phase 1 PK study was performed to evaluate initiation regimens: either 21 days of oral aripiprazole (15 mg/d) and one AL dose (n = 81) or one injection of ALncd plus a single dose of 30 mg oral aripiprazole and one AL dose (n = 80). Patients were randomized 1:1:1:1 to receive an AL dose of either 441 mg or 882 mg combined with the oral or the new ALncdinitiation regimen. As shown in Figure 1 and Figure 2, the mean plasma levels seen with 675 mg IM ALncd plus a single dose of 30 mg oral aripiprazole were comparable with levels achieved using 21 days of oral aripiprazole coverage, regardless of whether the regimen was paired with a maintenance AL dose of 441 mg or 882 mg.⁴

Tolerability. In PK studies, the safety profile and incidences of injection site reactions of ALncd were generally consistent with those observed for aripiprazole lauroxil.⁹ In the phase I PK study comparing oral initiation with ALncd plus a single 30 mg oral aripiprazole dose, there were 2 mild cases of akathisia in the 21-day oral aripiprazole groups (n = 81) and 4 cases in the ALncd groups (n = 80) (3 mild cases, 1 moderate case). None of the adverse events related to akathisia were deemed serious, and no patients discontinued participation in the trial due to akathisia.⁹

Continue to: Clinical considerations

ALncd is not a substitute for AL due to the very different kinetic properties of the 2 preparations. ALncd is approved only to be used for initiating treatment with AL, or in those instances where the revised missed dose guidelines for AL permit use of ALncd to obviate the need for oral coverage.⁹ Table 3⁹ presents these revised AL missed dose guidelines focusing on those time periods when some form of supplementation is required in addition to the established maintenance AL dose. Clinicians should be reminded that ALncd must be paired with a dose of AL, although the latter can be given up to 10 days later when commencing therapy.

• it obviates the need for 21 days of oral coverage previously required at the initiation of AL treatment
• clinically relevant plasma levels are seen within the first week when ALncd is combined with a single 30 mg oral aripiprazole dose
• per the revised missed dose guidelines for AL, it can be used in those situations that previously demanded 7 days of oral coverage, and, when combined with a single 30 mg oral dose, can be used for resumption of therapy after prolonged absences that required 21 days of oral coverage. In all instances, the patient will also receive their usual maintenance dose of AL.

Dosing. There is only one dose available for ALncd, 675 mg IM. As the dose cannot be modified, the package insert contains cautionary language regarding situations with less-than-expected drug exposure (use of cytochrome P450 [CYP] 3A4 inducers), greater-than-expected drug exposure (strong CYP3A4 or 2D6 inhibitors or known 2D6 poor metabolizers), or increased pharmacodynamic effects (concurrent use of antihypertensives or benzodiazepines).

Contraindications. The only contraindication is a known hypersensitivity to aripiprazole.

Bottom Line

Aripiprazole lauroxil nanocrystal suspension (Aristada Initio) was specifically developed to obviate the need for 21 days of oral aripiprazole coverage when commencing treatment with aripiprazole lauroxil (Aristada). The plasma levels achieved when an injection of aripiprazole lauroxil nanocrystal suspension is combined with a single 30 mg oral dose are comparable to those achieved with 21 days of oral coverage. This initiation regimen, including a aripiprazole lauroxil nanocrystal injection and a 30 mg oral dose, should be administered on the same day as the maintenance aripiprazole lauroxil injection, although the latter can be administered on any of the next 10 days.

Related Resource

• Khan AY, Ovais DM. Long-acting injectable aripiprazole lauroxil for schizophrenia. Current Psychiatry. 2016;15(7):50-52,58.

Drug Brand Names
Aripiprazole lauroxil • Aristada
Aripiprazole lauroxil nanocrystal • Aristada Initio
Risperidone microspheres • Risperdal Consta

Monoamine oxidase inhibitors (MAOIs) have well-established efficacy for treating depression, panic disorder, and social phobia. However, a lack of familiarity with these agents and misconceptions about the risks associated with their use have led to MAOIs being substantially underutilized. The goal of this 2-part guide to MAOIs is to educate clinicians about this often-overlooked class of medications. Part 1 (“A concise guide to monoamine inhibitors,” Current Psychiatry. December 2017, p. 14-23,47,A) described the pharmacology of MAOIs, the mechanism by which tyramine induces hypertension in patients taking MAOIs, and what to tell patients about dietary restrictions associated with these medications. Part 2 covers how to avoid potential drug interactions, including serotonin syndrome (SS) and pressor effects, that could affect patients receiving an MAOI; other factors to consider when starting a patient on an MAOI; and augmentation strategies for depressed patients who do not achieve remission from MAOI monotherapy.

MAOIs and potential drug interactions

One source of concern in patients receiving irreversible nonselective MAOIs is the development of excessive serotonergic neurotransmission resulting in SS. In the 1960s, researchers noted that administering large doses of tryptophan to MAOI-treated patients resulted in clonus and hyperactive reflexes without hypertensive events.¹ In 1991, Sternbach² provided an extensive case series and described the first set of criteria for SS. Features of SS include:

• mild symptoms: tremor, akathisia, inducible clonus
• moderate symptoms: spontaneous or sustained clonus, muscular hypertonicity
• severe symptoms: hyperthermia, diaphoresis.²

Although SS can be induced by significant exposure to individual agents that promote excess synaptic serotonin (eg, overdose of selective serotonin reuptake inhibitors [SSRIs]), the majority of fatal cases have occurred among patients taking MAOIs who were coadministered an agent that inhibited serotonin reuptake (Table 1³). Animal studies have determined that excessive stimulation of the 5HT2A receptor is primarily responsible for SS,⁴ and that 5HT2A antagonists, such as mirtazapine, can block the development of SS in a mouse coadministered fluoxetine and tranylcypromine.⁵ In addition to a patient’s medication history, the clinical hallmark of SS that helps distinguish it from neuroleptic malignant syndrome, delirium, and other acute syndromes is clonus, which becomes spontaneous and sustained as the severity increases.

Risk for SS. Most medications that promote serotonergic activity are well known for their use as antidepressants, but other agents that have 5HT reuptake properties (eg, the antihistamine chlorpheniramine) must be avoided. Although older literature suggests that the use of lower doses of certain tricyclic antidepressants concurrently with MAOIs may not be as dangerous as once believed,⁶ there are sufficient reports of serious outcomes that tricyclics should be avoided in patients taking MAOIs because of the risk of SS, and also because, in general, tricyclics are poorly tolerated.⁷

Desipramine, a potent norepinephrine transporter (NET) inhibitor, blocks the entry of tyramine into cells by NET, thereby preventing hypertensive events in animal models of tyramine overexposure. However, in some assays, the affinity for the serotonin transporter is not insignificant, so at higher doses desipramine may pose the same theoretical risk for SS as seen with other tricyclics.³

Lastly, rasagiline is an MAO-B selective inhibitor that has been available in the United States since 2008 for the treatment of Parkinson’s disease (PD). Although this drug lacks MAO-A antagonism, its package insert carries SS warnings; however, analysis of outcomes from a large multicenter rasagiline trial (N = 1,504) found no SS events in the 471 patients receiving rasagiline plus antidepressants (74.5% on SSRIs).⁸ Because depression is a common comorbidity in PD, clinicians who encounter rasagiline-treated patients who need antidepressant therapy should consult with the patient’s neurologist regarding their experience and comfort level with combining rasagiline and SSRIs or serotonin and norepinephrine reuptake inhibitors (SNRIs).

Astute clinicians will recognize that antidepressants that lack 5HT reuptake (eg, bupropion, mirtazapine) are not on this list of agents that may increase SS risk when taken with an MAOI. Older papers often list mirtazapine, but as a 5HT2A antagonist, it does not possess a plausible mechanism by which it can induce 5HT toxicity.^9,10 Most atypical antipsychotics have significant 5HT2A antagonism and can be combined with MAOIs, but ziprasidone is an exception: as a moderate SNRI, it has been associated with SS when administered with an MAOI.¹¹

Pressor reactions. The only theoretical sources of concern for pressor effects are medications that act as norepinephrine releasers through interactions at the trace amine-associated receptor 1 (TAAR1) (for more information on TAAR1, see “A concise guide to monoamine inhibitors,” Current Psychiatry. December 2017, p. 14-23,47,A). Amphetamines are one such class, but a 2004 review did not find cases of hypertensive crises when amphetamines were combined with MAOIs,¹² nor did a recent article that described the combined use of lisdexamfetamine and transdermal selegiline.¹³ Presumably, the low level of intracellular exposure mitigates the risk of excessive TAAR1 agonism, but amphetamine derivatives should be approached cautiously and with careful blood pressure monitoring. On the other hand, methylphenidate is an inhibitor of dopamine reuptake with no affinity for TAAR1¹⁴ or the serotonin transporter,¹⁵ and does not induce a pressor response nor increase risk for SS when combined with MAOIs.³ Concerns about the use of α1-adrenergic agonists in patients taking MAOIs are not universal, as the deleterious effects on blood pressure are seen only in certain vulnerable patients, typically those with preexisting hypertension. Nonetheless, all patients should be cautioned about the use of phenylephrine and pseudoephedrine.¹⁶
 

Starting a patient on an MAOI

Contraindicated medications need to be tapered before beginning MAOI treatment. The duration of the washout period depends on the half-life of the medication and any active metabolites. Antidepressants with half-lives of approximately ≤24 hours should be tapered over 7 to 14 days (depending on the dose) to minimize the risk of withdrawal syndromes, while those with long half-lives (eg, fluoxetine, vortioxetine) can be stopped abruptly. After stopping a medication for 5 half-lives, 96.875% of the medication is removed, so adequate time must elapse after the last dose before starting an MAOI. Table 2^17-22 lists the half-lives of commonly used newer antidepressants and any active metabolites or isomers. Clinicians should always err on the side of caution before starting an MAOI, and give their patients a brief overview of SS symptoms; however, be mindful of not extending time the patient is without effective antidepressant levels.

Initiation of an MAOI is always based on whether the patient can reliably follow the basic dietary advice (see “A concise guide to monoamine inhibitors,” Current Psychiatry. December 2017, p. 14-23,47,A), and they agree to check with their clinician before starting new medications. Titration of MAOIs should be based on tolerability; orthostasis is the primary dose-limiting adverse effect associated with rapid titration or higher dosages. This may be especially true in older patients with poor vasomotor tone, or those on α1-adrenergic antagonists or other agents that may induce orthostatic effects. The rapid titration schedules present in certain package inserts (eg, phenelzine²³) should not be followed.

The orthostasis management strategy is similar to that employed for clozapine: minimize the use of concurrent α1-adrenergic antagonists, lower the doses of antihypertensives as much as possible, and encourage adequate fluid intake. For patients with ongoing orthostasis and without a history of congestive heart failure, consider using the potent mineralocorticoid fludrocortisone starting at 0.1 mg/d, and titrating every 10 to 14 days if needed to a maximum of 0.3 mg/d.²⁴ Older literature noted weight gain, peripheral edema, and sexual dysfunction as common adverse effects. Research on the most recently studied MAOI, selegiline transdermal, reported rates of these adverse effects as follows: weight gain: 2.1% for selegiline transdermal vs 2.4% for placebo; all forms of sexual dysfunction: 0 to 1% for selegiline transdermal vs 0 to 0.4% for placebo.²⁵

Augmentation options for patients taking MAOIs

For depressed patients who do not achieve remission of symptoms from MAOI therapy, augmentation options should be sought, as patients who respond but fail to remit are at increased risk of relapse.²⁶ Lithium augmentation is one of the more common strategies, with abundant data supporting its use.^27,28 Case reports dating back >12 years describe the concurrent use of bupropion and MAOIs.^12,29 A recent review of augmentation of MAOIs with second-generation antipsychotics found multiple positive reports for most agents, including aripiprazole, with the sole exception of ziprasidone, a moderate SNRI for which cases of SS have been reported.¹¹ As of November 2017, there are no case reports for asenapine, lurasidone, brexpiprazole, or cariprazine. Triiodothyronine is often a neglected strategy, but older case reports of combined treatment with MAOIs found no obvious concerns beyond those related to the use of thyroid hormone.^30,31

Monoamine oxidase inhibitors (MAOIs) have well-established efficacy for treating depression, panic disorder, and social phobia. However, a lack of familiarity with these agents and misconceptions about the risks associated with their use have led to MAOIs being substantially underutilized. The goal of this 2-part guide to MAOIs is to educate clinicians about this often-overlooked class of medications. Part 1 (“A concise guide to monoamine inhibitors,” Current Psychiatry. December 2017, p. 14-23,47,A) described the pharmacology of MAOIs, the mechanism by which tyramine induces hypertension in patients taking MAOIs, and what to tell patients about dietary restrictions associated with these medications. Part 2 covers how to avoid potential drug interactions, including serotonin syndrome (SS) and pressor effects, that could affect patients receiving an MAOI; other factors to consider when starting a patient on an MAOI; and augmentation strategies for depressed patients who do not achieve remission from MAOI monotherapy.

MAOIs and potential drug interactions

One source of concern in patients receiving irreversible nonselective MAOIs is the development of excessive serotonergic neurotransmission resulting in SS. In the 1960s, researchers noted that administering large doses of tryptophan to MAOI-treated patients resulted in clonus and hyperactive reflexes without hypertensive events.¹ In 1991, Sternbach² provided an extensive case series and described the first set of criteria for SS. Features of SS include:

• mild symptoms: tremor, akathisia, inducible clonus
• moderate symptoms: spontaneous or sustained clonus, muscular hypertonicity
• severe symptoms: hyperthermia, diaphoresis.²

Although SS can be induced by significant exposure to individual agents that promote excess synaptic serotonin (eg, overdose of selective serotonin reuptake inhibitors [SSRIs]), the majority of fatal cases have occurred among patients taking MAOIs who were coadministered an agent that inhibited serotonin reuptake (Table 1³). Animal studies have determined that excessive stimulation of the 5HT2A receptor is primarily responsible for SS,⁴ and that 5HT2A antagonists, such as mirtazapine, can block the development of SS in a mouse coadministered fluoxetine and tranylcypromine.⁵ In addition to a patient’s medication history, the clinical hallmark of SS that helps distinguish it from neuroleptic malignant syndrome, delirium, and other acute syndromes is clonus, which becomes spontaneous and sustained as the severity increases.

Risk for SS. Most medications that promote serotonergic activity are well known for their use as antidepressants, but other agents that have 5HT reuptake properties (eg, the antihistamine chlorpheniramine) must be avoided. Although older literature suggests that the use of lower doses of certain tricyclic antidepressants concurrently with MAOIs may not be as dangerous as once believed,⁶ there are sufficient reports of serious outcomes that tricyclics should be avoided in patients taking MAOIs because of the risk of SS, and also because, in general, tricyclics are poorly tolerated.⁷

Desipramine, a potent norepinephrine transporter (NET) inhibitor, blocks the entry of tyramine into cells by NET, thereby preventing hypertensive events in animal models of tyramine overexposure. However, in some assays, the affinity for the serotonin transporter is not insignificant, so at higher doses desipramine may pose the same theoretical risk for SS as seen with other tricyclics.³

Lastly, rasagiline is an MAO-B selective inhibitor that has been available in the United States since 2008 for the treatment of Parkinson’s disease (PD). Although this drug lacks MAO-A antagonism, its package insert carries SS warnings; however, analysis of outcomes from a large multicenter rasagiline trial (N = 1,504) found no SS events in the 471 patients receiving rasagiline plus antidepressants (74.5% on SSRIs).⁸ Because depression is a common comorbidity in PD, clinicians who encounter rasagiline-treated patients who need antidepressant therapy should consult with the patient’s neurologist regarding their experience and comfort level with combining rasagiline and SSRIs or serotonin and norepinephrine reuptake inhibitors (SNRIs).

Astute clinicians will recognize that antidepressants that lack 5HT reuptake (eg, bupropion, mirtazapine) are not on this list of agents that may increase SS risk when taken with an MAOI. Older papers often list mirtazapine, but as a 5HT2A antagonist, it does not possess a plausible mechanism by which it can induce 5HT toxicity.^9,10 Most atypical antipsychotics have significant 5HT2A antagonism and can be combined with MAOIs, but ziprasidone is an exception: as a moderate SNRI, it has been associated with SS when administered with an MAOI.¹¹

Pressor reactions. The only theoretical sources of concern for pressor effects are medications that act as norepinephrine releasers through interactions at the trace amine-associated receptor 1 (TAAR1) (for more information on TAAR1, see “A concise guide to monoamine inhibitors,” Current Psychiatry. December 2017, p. 14-23,47,A). Amphetamines are one such class, but a 2004 review did not find cases of hypertensive crises when amphetamines were combined with MAOIs,¹² nor did a recent article that described the combined use of lisdexamfetamine and transdermal selegiline.¹³ Presumably, the low level of intracellular exposure mitigates the risk of excessive TAAR1 agonism, but amphetamine derivatives should be approached cautiously and with careful blood pressure monitoring. On the other hand, methylphenidate is an inhibitor of dopamine reuptake with no affinity for TAAR1¹⁴ or the serotonin transporter,¹⁵ and does not induce a pressor response nor increase risk for SS when combined with MAOIs.³ Concerns about the use of α1-adrenergic agonists in patients taking MAOIs are not universal, as the deleterious effects on blood pressure are seen only in certain vulnerable patients, typically those with preexisting hypertension. Nonetheless, all patients should be cautioned about the use of phenylephrine and pseudoephedrine.¹⁶
 

Starting a patient on an MAOI

Contraindicated medications need to be tapered before beginning MAOI treatment. The duration of the washout period depends on the half-life of the medication and any active metabolites. Antidepressants with half-lives of approximately ≤24 hours should be tapered over 7 to 14 days (depending on the dose) to minimize the risk of withdrawal syndromes, while those with long half-lives (eg, fluoxetine, vortioxetine) can be stopped abruptly. After stopping a medication for 5 half-lives, 96.875% of the medication is removed, so adequate time must elapse after the last dose before starting an MAOI. Table 2^17-22 lists the half-lives of commonly used newer antidepressants and any active metabolites or isomers. Clinicians should always err on the side of caution before starting an MAOI, and give their patients a brief overview of SS symptoms; however, be mindful of not extending time the patient is without effective antidepressant levels.

Initiation of an MAOI is always based on whether the patient can reliably follow the basic dietary advice (see “A concise guide to monoamine inhibitors,” Current Psychiatry. December 2017, p. 14-23,47,A), and they agree to check with their clinician before starting new medications. Titration of MAOIs should be based on tolerability; orthostasis is the primary dose-limiting adverse effect associated with rapid titration or higher dosages. This may be especially true in older patients with poor vasomotor tone, or those on α1-adrenergic antagonists or other agents that may induce orthostatic effects. The rapid titration schedules present in certain package inserts (eg, phenelzine²³) should not be followed.

The orthostasis management strategy is similar to that employed for clozapine: minimize the use of concurrent α1-adrenergic antagonists, lower the doses of antihypertensives as much as possible, and encourage adequate fluid intake. For patients with ongoing orthostasis and without a history of congestive heart failure, consider using the potent mineralocorticoid fludrocortisone starting at 0.1 mg/d, and titrating every 10 to 14 days if needed to a maximum of 0.3 mg/d.²⁴ Older literature noted weight gain, peripheral edema, and sexual dysfunction as common adverse effects. Research on the most recently studied MAOI, selegiline transdermal, reported rates of these adverse effects as follows: weight gain: 2.1% for selegiline transdermal vs 2.4% for placebo; all forms of sexual dysfunction: 0 to 1% for selegiline transdermal vs 0 to 0.4% for placebo.²⁵

Augmentation options for patients taking MAOIs

For depressed patients who do not achieve remission of symptoms from MAOI therapy, augmentation options should be sought, as patients who respond but fail to remit are at increased risk of relapse.²⁶ Lithium augmentation is one of the more common strategies, with abundant data supporting its use.^27,28 Case reports dating back >12 years describe the concurrent use of bupropion and MAOIs.^12,29 A recent review of augmentation of MAOIs with second-generation antipsychotics found multiple positive reports for most agents, including aripiprazole, with the sole exception of ziprasidone, a moderate SNRI for which cases of SS have been reported.¹¹ As of November 2017, there are no case reports for asenapine, lurasidone, brexpiprazole, or cariprazine. Triiodothyronine is often a neglected strategy, but older case reports of combined treatment with MAOIs found no obvious concerns beyond those related to the use of thyroid hormone.^30,31

Monoamine oxidase inhibitors (MAOIs) have well-established efficacy for treating depression, panic disorder, and social phobia. However, a lack of familiarity with these agents and misconceptions about the risks associated with their use have led to MAOIs being substantially underutilized. The goal of this 2-part guide to MAOIs is to educate clinicians about this often-overlooked class of medications. Part 1 (“A concise guide to monoamine inhibitors,” Current Psychiatry. December 2017, p. 14-23,47,A) described the pharmacology of MAOIs, the mechanism by which tyramine induces hypertension in patients taking MAOIs, and what to tell patients about dietary restrictions associated with these medications. Part 2 covers how to avoid potential drug interactions, including serotonin syndrome (SS) and pressor effects, that could affect patients receiving an MAOI; other factors to consider when starting a patient on an MAOI; and augmentation strategies for depressed patients who do not achieve remission from MAOI monotherapy.

MAOIs and potential drug interactions

One source of concern in patients receiving irreversible nonselective MAOIs is the development of excessive serotonergic neurotransmission resulting in SS. In the 1960s, researchers noted that administering large doses of tryptophan to MAOI-treated patients resulted in clonus and hyperactive reflexes without hypertensive events.¹ In 1991, Sternbach² provided an extensive case series and described the first set of criteria for SS. Features of SS include:

• mild symptoms: tremor, akathisia, inducible clonus
• moderate symptoms: spontaneous or sustained clonus, muscular hypertonicity
• severe symptoms: hyperthermia, diaphoresis.²

Although SS can be induced by significant exposure to individual agents that promote excess synaptic serotonin (eg, overdose of selective serotonin reuptake inhibitors [SSRIs]), the majority of fatal cases have occurred among patients taking MAOIs who were coadministered an agent that inhibited serotonin reuptake (Table 1³). Animal studies have determined that excessive stimulation of the 5HT2A receptor is primarily responsible for SS,⁴ and that 5HT2A antagonists, such as mirtazapine, can block the development of SS in a mouse coadministered fluoxetine and tranylcypromine.⁵ In addition to a patient’s medication history, the clinical hallmark of SS that helps distinguish it from neuroleptic malignant syndrome, delirium, and other acute syndromes is clonus, which becomes spontaneous and sustained as the severity increases.

Risk for SS. Most medications that promote serotonergic activity are well known for their use as antidepressants, but other agents that have 5HT reuptake properties (eg, the antihistamine chlorpheniramine) must be avoided. Although older literature suggests that the use of lower doses of certain tricyclic antidepressants concurrently with MAOIs may not be as dangerous as once believed,⁶ there are sufficient reports of serious outcomes that tricyclics should be avoided in patients taking MAOIs because of the risk of SS, and also because, in general, tricyclics are poorly tolerated.⁷

Desipramine, a potent norepinephrine transporter (NET) inhibitor, blocks the entry of tyramine into cells by NET, thereby preventing hypertensive events in animal models of tyramine overexposure. However, in some assays, the affinity for the serotonin transporter is not insignificant, so at higher doses desipramine may pose the same theoretical risk for SS as seen with other tricyclics.³

Lastly, rasagiline is an MAO-B selective inhibitor that has been available in the United States since 2008 for the treatment of Parkinson’s disease (PD). Although this drug lacks MAO-A antagonism, its package insert carries SS warnings; however, analysis of outcomes from a large multicenter rasagiline trial (N = 1,504) found no SS events in the 471 patients receiving rasagiline plus antidepressants (74.5% on SSRIs).⁸ Because depression is a common comorbidity in PD, clinicians who encounter rasagiline-treated patients who need antidepressant therapy should consult with the patient’s neurologist regarding their experience and comfort level with combining rasagiline and SSRIs or serotonin and norepinephrine reuptake inhibitors (SNRIs).

Astute clinicians will recognize that antidepressants that lack 5HT reuptake (eg, bupropion, mirtazapine) are not on this list of agents that may increase SS risk when taken with an MAOI. Older papers often list mirtazapine, but as a 5HT2A antagonist, it does not possess a plausible mechanism by which it can induce 5HT toxicity.^9,10 Most atypical antipsychotics have significant 5HT2A antagonism and can be combined with MAOIs, but ziprasidone is an exception: as a moderate SNRI, it has been associated with SS when administered with an MAOI.¹¹

Pressor reactions. The only theoretical sources of concern for pressor effects are medications that act as norepinephrine releasers through interactions at the trace amine-associated receptor 1 (TAAR1) (for more information on TAAR1, see “A concise guide to monoamine inhibitors,” Current Psychiatry. December 2017, p. 14-23,47,A). Amphetamines are one such class, but a 2004 review did not find cases of hypertensive crises when amphetamines were combined with MAOIs,¹² nor did a recent article that described the combined use of lisdexamfetamine and transdermal selegiline.¹³ Presumably, the low level of intracellular exposure mitigates the risk of excessive TAAR1 agonism, but amphetamine derivatives should be approached cautiously and with careful blood pressure monitoring. On the other hand, methylphenidate is an inhibitor of dopamine reuptake with no affinity for TAAR1¹⁴ or the serotonin transporter,¹⁵ and does not induce a pressor response nor increase risk for SS when combined with MAOIs.³ Concerns about the use of α1-adrenergic agonists in patients taking MAOIs are not universal, as the deleterious effects on blood pressure are seen only in certain vulnerable patients, typically those with preexisting hypertension. Nonetheless, all patients should be cautioned about the use of phenylephrine and pseudoephedrine.¹⁶
 

Starting a patient on an MAOI

Contraindicated medications need to be tapered before beginning MAOI treatment. The duration of the washout period depends on the half-life of the medication and any active metabolites. Antidepressants with half-lives of approximately ≤24 hours should be tapered over 7 to 14 days (depending on the dose) to minimize the risk of withdrawal syndromes, while those with long half-lives (eg, fluoxetine, vortioxetine) can be stopped abruptly. After stopping a medication for 5 half-lives, 96.875% of the medication is removed, so adequate time must elapse after the last dose before starting an MAOI. Table 2^17-22 lists the half-lives of commonly used newer antidepressants and any active metabolites or isomers. Clinicians should always err on the side of caution before starting an MAOI, and give their patients a brief overview of SS symptoms; however, be mindful of not extending time the patient is without effective antidepressant levels.

Initiation of an MAOI is always based on whether the patient can reliably follow the basic dietary advice (see “A concise guide to monoamine inhibitors,” Current Psychiatry. December 2017, p. 14-23,47,A), and they agree to check with their clinician before starting new medications. Titration of MAOIs should be based on tolerability; orthostasis is the primary dose-limiting adverse effect associated with rapid titration or higher dosages. This may be especially true in older patients with poor vasomotor tone, or those on α1-adrenergic antagonists or other agents that may induce orthostatic effects. The rapid titration schedules present in certain package inserts (eg, phenelzine²³) should not be followed.

The orthostasis management strategy is similar to that employed for clozapine: minimize the use of concurrent α1-adrenergic antagonists, lower the doses of antihypertensives as much as possible, and encourage adequate fluid intake. For patients with ongoing orthostasis and without a history of congestive heart failure, consider using the potent mineralocorticoid fludrocortisone starting at 0.1 mg/d, and titrating every 10 to 14 days if needed to a maximum of 0.3 mg/d.²⁴ Older literature noted weight gain, peripheral edema, and sexual dysfunction as common adverse effects. Research on the most recently studied MAOI, selegiline transdermal, reported rates of these adverse effects as follows: weight gain: 2.1% for selegiline transdermal vs 2.4% for placebo; all forms of sexual dysfunction: 0 to 1% for selegiline transdermal vs 0 to 0.4% for placebo.²⁵

Augmentation options for patients taking MAOIs

For depressed patients who do not achieve remission of symptoms from MAOI therapy, augmentation options should be sought, as patients who respond but fail to remit are at increased risk of relapse.²⁶ Lithium augmentation is one of the more common strategies, with abundant data supporting its use.^27,28 Case reports dating back >12 years describe the concurrent use of bupropion and MAOIs.^12,29 A recent review of augmentation of MAOIs with second-generation antipsychotics found multiple positive reports for most agents, including aripiprazole, with the sole exception of ziprasidone, a moderate SNRI for which cases of SS have been reported.¹¹ As of November 2017, there are no case reports for asenapine, lurasidone, brexpiprazole, or cariprazine. Triiodothyronine is often a neglected strategy, but older case reports of combined treatment with MAOIs found no obvious concerns beyond those related to the use of thyroid hormone.^30,31

Despite an abundance of evidenced-based literature supporting monoamine oxidase inhibitors (MAOIs) as an effective treatment for depression, use of these agents has decreased drastically in the past 3 decades. A lack of industry support and the ease of use of other agents are contributing factors, but the biggest impediments to routine use of MAOIs are unfamiliarity with their efficacy advantages and concerns about adverse effects, particularly the risk of hypertensive crises and serotonin syndrome. Many misconceptions regarding these medications are based on outdated data and studies that are no longer reliable.

The goal of this 2-part review is to provide clinicians with updated information regarding MAOIs. Part 1 provides a brief description of:

• the pharmacology of nonselective irreversible MAOIs
• the mechanism by which tyramine induces hypertension
• sources of clinically significant tyramine exposure
• what to tell patients about dietary restrictions and MAOIs.

Part 2 of this guide will cover the risk of serotonin syndrome when MAOIs are combined with inhibitors of serotonin reuptake, how to initiate MAOI therapy, and augmenting MAOIs with other agents.

The pharmacology of MAOIs

First used clinically in the 1950s to treat tuberculosis, MAOIs have a long and interesting history (see the Box “A brief history of monoamine oxidase inhibitors”). Table 1¹ lists MAOIs currently available in the United States, including the MAO-B–specific agent rasagiline, which is used for Parkinson’s disease.

Manipulation of the monoamines serotonin, norepinephrine, and dopamine is fundamental to managing major depressive disorder (MDD), yet only nonselective MAOIs directly promote neurotransmission of all 3 by inhibiting MAO-A and MAO-B.² The Sequenced Treatment Alternatives to Relieve Depression (STAR*D) study demonstrated that <50% of MDD patients achieve remission in monotherapy trials of selective serotonin reuptake inhibitors, serotonin norepinephrine reuptake inhibitors, mirtazapine, or bupropion, necessitating consideration of antidepressant combinations, augmentation options, and eventually irreversible, nonselective MAOIs such as phenelzine, tranylcypromine, or isocarboxazid.^3,4 Nonselective MAOIs thus offer a therapeutic opportunity for patients who do not respond to single or dual-mechanism strategies; moreover, nonselective MAOIs have compelling effectiveness data for other conditions, including panic disorder and social phobia.⁵ Although MAOIs are among the most effective pharmacologic agents for MDD,⁶ they are underutilized because of an inadequate understanding of risk mechanisms and resultant fear of catastrophic outcomes. Because of the difficulties encountered in achieving clinical remission for MDD, the nonselective MAOIs deserve a second look.
 

Differentiation of MAO-A from MAO-B. It is essential to understand the mechanism of action of MAOIs, specifically the impact of MAO-A inhibition. Although the enzyme MAO was known in the 1950s, it wasn’t until 1968 that Johnston⁷ postulated the existence of >1 form. In 1971, Goridis and Neff⁸ used clorgyline to examine the deamination rate by MAO of tyramine and norepinephrine. They found that tyramine appeared to be a substrate of both MAO isoforms, but only 1 of the MAO types was sensitive to the inhibitory effects of clorgyline. They also discerned that norepinephrine was only a substrate for MAO-A, and that this form of MAO was sensitive to clorgyline inhibition. Thus, the forms of MAO were characterized by their preferred substrates (Table 2^9,10), and then later by their tissue distribution. Phenylethylamine is a naturally occurring compound found in foods, such as chocolate, and has an in vitro pharmacology similar to amphetamine but with 1 important difference: it has a short half-life of 5 to 10 minutes after oral ingestion, and therefore no appreciable CNS impact.

Within the CNS, norepinephrine and dopamine neurons possess both MAO forms, with the MAO-A content greater than MAO-B. Serotonergic neurons only contain MAO-B.¹¹ Outside of the CNS, MAO-A predominates, with only platelets and lymphocytes possessing MAO-B activity.¹¹ The overall relative tissue proportions of MAO-A to MAO-B activity are: brain, 25% MAO-A, 75% MAO-B; liver, 50% MAO-A, 50% MAO-B; intestine, 80% MAO-A, 20% MAO-B; and peripheral adrenergic neurons, 90% MAO-A, 10% MAO-B.

Because of its specificity for serotonin and norepinephrine, CNS MAO-A inhibition is necessary for antidepressant effects. MAO-B inhibition by itself does not appear to raise CNS dopamine levels unless exogenous dopamine is supplied.¹¹ All MAOIs used in the United States to treat depression are irreversible, nonselective inhibitors of MAO-A and MAO-B.

Selegiline in oral form generates low plasma levels and primarily inhibits MAO-B. The transdermal form of selegiline achieves significantly greater systemic exposure, and at these higher plasma levels selegiline is a nonselective, irreversible MAOI effective for MDD (Figure 1¹²). Administering selegiline systemically via a transdermal patch avoids clinically significant MAOI effects in the gut, so no dietary warnings exist for the lowest dose (6 mg/24 hours), although there are warnings for the higher dosages (9 mg/24 hours and 12 mg/24 hours).

MAOIs and tyramine: Dietary requirements

Clinicians who are familiar with MAOIs recognize that there are dietary restrictions to minimize patients’ exposure to tyramine. As most clinicians know, significant tyramine ingestion may cause an increase in blood pressure (BP) in patients taking an MAOI, but many overestimate the prevalence of foods high in tyramine content since the original reports emerged in the early 1960s.¹⁸ In a recent monograph, one of the leading experts on MAOIs, Professor Ken Gillman, stated:

Very few foods now contain problematically high tyramine levels, that is a result of great changes in international food production methods and hygiene regulations. Cheese is the only food that, in the past, has been associated with documented fatalities resulting from hypertension. Nowadays most cheeses are quite safe, and even ‘matured’ cheeses are usually safe in healthy-sized portions. The variability of sensitivity to tyramine between individuals, and the sometimes unpredictable amount of tyramine content in foods, means a little knowledge and care are still required.¹⁹

What is tyramine? Tyramine is a biogenic amine that is virtually absent in fresh animal protein sources but is enriched after decay or fermentation.²⁰ Modern food processing and handling methods have significantly limited the tyramine content in processed foods, with the exception of certain cheeses and sauces, as discussed below. Moreover, modern assaying techniques using high-performance liquid chromatography have generated extremely accurate assessments of the tyramine content of specific foods.²¹ Data published prior to 2000 are not reliable, because many of these publications employed outdated methods.¹⁷

When ingested, tyramine is metabolized by gut MAO-A, with doses up to 400 mg causing no known effects, although most people rarely ingest >25 mg during a meal.²² In addition to being a substrate for MAO-A, tyramine is also a substrate for the dopamine transporter, norepinephrine transporter (NET), the vesicular monoamine transporter 2, and TAAR1.²³ Tyramine enters the cell via NET, where it interacts with TAAR1, a G protein-coupled receptor that is responsive to trace amines, such as tyramine, as well as amphetamines.²⁰ The agonist properties at TAAR1 are the presumed site of action for the BP effects of tyramine, because binding results in potent release of norepinephrine.^20,24 When tyramine is supplied to an animal in which MAO-A is inhibited, the decreased peripheral catabolism of tyramine results in markedly increased norepinephrine release by peripheral adrenergic neurons. Moreover, the absence of MAO-A activity in those neurons prevents any norepinephrine breakdown, resulting in robust synaptic norepinephrine delivery and peripheral effects.

All orally administered irreversible MAOIs potently inhibit gut and systemic MAO-A, and are susceptible to the impact of significant tyramine ingestion. The exception is selegiline transdermal (Figure 1¹²), as appreciable gut MAO-A inhibition does not occur until doses >6 mg/24 hours are reached.²² No significant pressor response was seen in participants taking selegiline transdermal, 6 mg/24 hours for 13 days, who consumed a meal that provided 400 mg of tyramine.²² Conversely, for oral agents that produce gut MAO-A inhibition, tyramine doses as low as 8 to 10 mg (when administered as tyramine capsules) may increase systolic pressure by 30 mm Hg.²⁵ The dietary warnings do not apply to rasagiline, which is a selective MAO-B inhibitor, although rasagiline may have an impact on resting BP; the prescribing information for rasagiline includes warnings about hypotension and hypertension.²⁶

What to tell patients about tyramine. Although administering pure tyramine capsules can induce a measurable change in systolic BP, when ingested as food, tyramine doses <50 mg are unlikely to cause an increase in BP sufficient to warrant clinical intervention, although some individuals can be sensitive to 10 to 25 mg.¹⁹ When discussing with patients safety issues related to diet, there are a few important concepts to remember¹⁹:

• In an era when the tyramine content of foods was much higher (1960 to 1964) and MAOI users received no dietary guidance, only 14 deaths were reported among an estimated 1.5 million patients who took MAOIs.
• MAOIs do not raise BP, and their use is associated with orthostasis in some patients.
• Routine exercise or other vigorous activities (eg, weightlifting) can raise systolic pressure well above 200 mm Hg, and routine baseline systolic pressures, ranging from 180 to 220 mm Hg, do not increase the risk of subarachnoid hemorrhage.
• Hospital evaluation is needed only if a substantial amount of tyramine is ingested (eg, estimated ≥100 mg), and self-monitoring shows a systolic BP ≥220 mm Hg over a prolonged period (eg, 2 hours). Ingestion of 100 mg of tyramine would almost certainly have to be intentional, as it would require one to consume 3.5 oz of the most highly tyramine-laden cheeses.
   

Emphasize to patients that only a small number of highly aged cheeses, foods, and sauces contain high quantities of tyramine, and that even these foods can be enjoyed in small amounts. All patients who are prescribed an MAOI also should purchase a portable BP cuff for those rare instances when a dietary indiscretion may have occurred and the person experiences a headache within 1 to 2 hours after tyramine ingestion. Most reactions are self-limited and resolve over 2 to 4 hours.

Patients who ingest ≥100 mg of tyramine should be evaluated by a physician. Under no circumstances should a patient be given a prescription for nifedipine or other medications that can abruptly lower BP, because this may result in complications, including myocardial infarction.^27,28 Counsel patients to remain calm. Some clinicians endorse the use of low doses of benzodiazepines (the equivalent of alprazolam 0.5 mg) to facilitate this, because anxiety elevates BP. A recent emergency room study of patients with an initial systolic BP ≥160 mm Hg or diastolic BP ≥100 mm Hg without end organ damage demonstrated that alprazolam, 0.5 mg, was as effective as captopril, 25 mg, in lowering BP.²⁹

Also, tell patients that if a food is unfamiliar and highly aged or fermented, they should avoid it until they can further inquire about it. In a review, Gillman¹⁹ provides the tyramine content of an exhaustive list of cheeses, aged meats, and sauces (see Related Resources). For other products, patients often can obtain information directly from the manufacturer. In many parts of the world, assays for tyramine content are required as a demonstration of adequate product safety procedures. Even the most highly aged cheeses with a tyramine content of 1,000 g/kg can be enjoyed in small amounts (<1 oz), and most products would require heroic intake to achieve clinically significant tyramine ingestion (Table 3¹⁹).

Improved education can clarify the risks

Medications such as lithium, clozapine, and MAOIs have a proven record of efficacy, yet often are underused due to fears engendered by lack of systematic training. A recent initiative in New York thus aimed to increase rates of clozapine prescribing by providing clinicians with an education consultation center.³⁰ Similarly, enhanced education regarding MAOIs could increase the use of these highly effective medications. With a better understanding of MAOIs, clinicians can become adept at using these medications, and therefore expand the armamentarium of agents available to patients with MDD, as well as to those with panic disorder and social phobia.
 

Despite an abundance of evidenced-based literature supporting monoamine oxidase inhibitors (MAOIs) as an effective treatment for depression, use of these agents has decreased drastically in the past 3 decades. A lack of industry support and the ease of use of other agents are contributing factors, but the biggest impediments to routine use of MAOIs are unfamiliarity with their efficacy advantages and concerns about adverse effects, particularly the risk of hypertensive crises and serotonin syndrome. Many misconceptions regarding these medications are based on outdated data and studies that are no longer reliable.

The goal of this 2-part review is to provide clinicians with updated information regarding MAOIs. Part 1 provides a brief description of:

• the pharmacology of nonselective irreversible MAOIs
• the mechanism by which tyramine induces hypertension
• sources of clinically significant tyramine exposure
• what to tell patients about dietary restrictions and MAOIs.

Part 2 of this guide will cover the risk of serotonin syndrome when MAOIs are combined with inhibitors of serotonin reuptake, how to initiate MAOI therapy, and augmenting MAOIs with other agents.

The pharmacology of MAOIs

First used clinically in the 1950s to treat tuberculosis, MAOIs have a long and interesting history (see the Box “A brief history of monoamine oxidase inhibitors”). Table 1¹ lists MAOIs currently available in the United States, including the MAO-B–specific agent rasagiline, which is used for Parkinson’s disease.

Manipulation of the monoamines serotonin, norepinephrine, and dopamine is fundamental to managing major depressive disorder (MDD), yet only nonselective MAOIs directly promote neurotransmission of all 3 by inhibiting MAO-A and MAO-B.² The Sequenced Treatment Alternatives to Relieve Depression (STAR*D) study demonstrated that <50% of MDD patients achieve remission in monotherapy trials of selective serotonin reuptake inhibitors, serotonin norepinephrine reuptake inhibitors, mirtazapine, or bupropion, necessitating consideration of antidepressant combinations, augmentation options, and eventually irreversible, nonselective MAOIs such as phenelzine, tranylcypromine, or isocarboxazid.^3,4 Nonselective MAOIs thus offer a therapeutic opportunity for patients who do not respond to single or dual-mechanism strategies; moreover, nonselective MAOIs have compelling effectiveness data for other conditions, including panic disorder and social phobia.⁵ Although MAOIs are among the most effective pharmacologic agents for MDD,⁶ they are underutilized because of an inadequate understanding of risk mechanisms and resultant fear of catastrophic outcomes. Because of the difficulties encountered in achieving clinical remission for MDD, the nonselective MAOIs deserve a second look.
 

Differentiation of MAO-A from MAO-B. It is essential to understand the mechanism of action of MAOIs, specifically the impact of MAO-A inhibition. Although the enzyme MAO was known in the 1950s, it wasn’t until 1968 that Johnston⁷ postulated the existence of >1 form. In 1971, Goridis and Neff⁸ used clorgyline to examine the deamination rate by MAO of tyramine and norepinephrine. They found that tyramine appeared to be a substrate of both MAO isoforms, but only 1 of the MAO types was sensitive to the inhibitory effects of clorgyline. They also discerned that norepinephrine was only a substrate for MAO-A, and that this form of MAO was sensitive to clorgyline inhibition. Thus, the forms of MAO were characterized by their preferred substrates (Table 2^9,10), and then later by their tissue distribution. Phenylethylamine is a naturally occurring compound found in foods, such as chocolate, and has an in vitro pharmacology similar to amphetamine but with 1 important difference: it has a short half-life of 5 to 10 minutes after oral ingestion, and therefore no appreciable CNS impact.

Within the CNS, norepinephrine and dopamine neurons possess both MAO forms, with the MAO-A content greater than MAO-B. Serotonergic neurons only contain MAO-B.¹¹ Outside of the CNS, MAO-A predominates, with only platelets and lymphocytes possessing MAO-B activity.¹¹ The overall relative tissue proportions of MAO-A to MAO-B activity are: brain, 25% MAO-A, 75% MAO-B; liver, 50% MAO-A, 50% MAO-B; intestine, 80% MAO-A, 20% MAO-B; and peripheral adrenergic neurons, 90% MAO-A, 10% MAO-B.

Because of its specificity for serotonin and norepinephrine, CNS MAO-A inhibition is necessary for antidepressant effects. MAO-B inhibition by itself does not appear to raise CNS dopamine levels unless exogenous dopamine is supplied.¹¹ All MAOIs used in the United States to treat depression are irreversible, nonselective inhibitors of MAO-A and MAO-B.

Selegiline in oral form generates low plasma levels and primarily inhibits MAO-B. The transdermal form of selegiline achieves significantly greater systemic exposure, and at these higher plasma levels selegiline is a nonselective, irreversible MAOI effective for MDD (Figure 1¹²). Administering selegiline systemically via a transdermal patch avoids clinically significant MAOI effects in the gut, so no dietary warnings exist for the lowest dose (6 mg/24 hours), although there are warnings for the higher dosages (9 mg/24 hours and 12 mg/24 hours).

MAOIs and tyramine: Dietary requirements

Clinicians who are familiar with MAOIs recognize that there are dietary restrictions to minimize patients’ exposure to tyramine. As most clinicians know, significant tyramine ingestion may cause an increase in blood pressure (BP) in patients taking an MAOI, but many overestimate the prevalence of foods high in tyramine content since the original reports emerged in the early 1960s.¹⁸ In a recent monograph, one of the leading experts on MAOIs, Professor Ken Gillman, stated:

Very few foods now contain problematically high tyramine levels, that is a result of great changes in international food production methods and hygiene regulations. Cheese is the only food that, in the past, has been associated with documented fatalities resulting from hypertension. Nowadays most cheeses are quite safe, and even ‘matured’ cheeses are usually safe in healthy-sized portions. The variability of sensitivity to tyramine between individuals, and the sometimes unpredictable amount of tyramine content in foods, means a little knowledge and care are still required.¹⁹

What is tyramine? Tyramine is a biogenic amine that is virtually absent in fresh animal protein sources but is enriched after decay or fermentation.²⁰ Modern food processing and handling methods have significantly limited the tyramine content in processed foods, with the exception of certain cheeses and sauces, as discussed below. Moreover, modern assaying techniques using high-performance liquid chromatography have generated extremely accurate assessments of the tyramine content of specific foods.²¹ Data published prior to 2000 are not reliable, because many of these publications employed outdated methods.¹⁷

When ingested, tyramine is metabolized by gut MAO-A, with doses up to 400 mg causing no known effects, although most people rarely ingest >25 mg during a meal.²² In addition to being a substrate for MAO-A, tyramine is also a substrate for the dopamine transporter, norepinephrine transporter (NET), the vesicular monoamine transporter 2, and TAAR1.²³ Tyramine enters the cell via NET, where it interacts with TAAR1, a G protein-coupled receptor that is responsive to trace amines, such as tyramine, as well as amphetamines.²⁰ The agonist properties at TAAR1 are the presumed site of action for the BP effects of tyramine, because binding results in potent release of norepinephrine.^20,24 When tyramine is supplied to an animal in which MAO-A is inhibited, the decreased peripheral catabolism of tyramine results in markedly increased norepinephrine release by peripheral adrenergic neurons. Moreover, the absence of MAO-A activity in those neurons prevents any norepinephrine breakdown, resulting in robust synaptic norepinephrine delivery and peripheral effects.

All orally administered irreversible MAOIs potently inhibit gut and systemic MAO-A, and are susceptible to the impact of significant tyramine ingestion. The exception is selegiline transdermal (Figure 1¹²), as appreciable gut MAO-A inhibition does not occur until doses >6 mg/24 hours are reached.²² No significant pressor response was seen in participants taking selegiline transdermal, 6 mg/24 hours for 13 days, who consumed a meal that provided 400 mg of tyramine.²² Conversely, for oral agents that produce gut MAO-A inhibition, tyramine doses as low as 8 to 10 mg (when administered as tyramine capsules) may increase systolic pressure by 30 mm Hg.²⁵ The dietary warnings do not apply to rasagiline, which is a selective MAO-B inhibitor, although rasagiline may have an impact on resting BP; the prescribing information for rasagiline includes warnings about hypotension and hypertension.²⁶

What to tell patients about tyramine. Although administering pure tyramine capsules can induce a measurable change in systolic BP, when ingested as food, tyramine doses <50 mg are unlikely to cause an increase in BP sufficient to warrant clinical intervention, although some individuals can be sensitive to 10 to 25 mg.¹⁹ When discussing with patients safety issues related to diet, there are a few important concepts to remember¹⁹:

• In an era when the tyramine content of foods was much higher (1960 to 1964) and MAOI users received no dietary guidance, only 14 deaths were reported among an estimated 1.5 million patients who took MAOIs.
• MAOIs do not raise BP, and their use is associated with orthostasis in some patients.
• Routine exercise or other vigorous activities (eg, weightlifting) can raise systolic pressure well above 200 mm Hg, and routine baseline systolic pressures, ranging from 180 to 220 mm Hg, do not increase the risk of subarachnoid hemorrhage.
• Hospital evaluation is needed only if a substantial amount of tyramine is ingested (eg, estimated ≥100 mg), and self-monitoring shows a systolic BP ≥220 mm Hg over a prolonged period (eg, 2 hours). Ingestion of 100 mg of tyramine would almost certainly have to be intentional, as it would require one to consume 3.5 oz of the most highly tyramine-laden cheeses.
   

Emphasize to patients that only a small number of highly aged cheeses, foods, and sauces contain high quantities of tyramine, and that even these foods can be enjoyed in small amounts. All patients who are prescribed an MAOI also should purchase a portable BP cuff for those rare instances when a dietary indiscretion may have occurred and the person experiences a headache within 1 to 2 hours after tyramine ingestion. Most reactions are self-limited and resolve over 2 to 4 hours.

Patients who ingest ≥100 mg of tyramine should be evaluated by a physician. Under no circumstances should a patient be given a prescription for nifedipine or other medications that can abruptly lower BP, because this may result in complications, including myocardial infarction.^27,28 Counsel patients to remain calm. Some clinicians endorse the use of low doses of benzodiazepines (the equivalent of alprazolam 0.5 mg) to facilitate this, because anxiety elevates BP. A recent emergency room study of patients with an initial systolic BP ≥160 mm Hg or diastolic BP ≥100 mm Hg without end organ damage demonstrated that alprazolam, 0.5 mg, was as effective as captopril, 25 mg, in lowering BP.²⁹

Also, tell patients that if a food is unfamiliar and highly aged or fermented, they should avoid it until they can further inquire about it. In a review, Gillman¹⁹ provides the tyramine content of an exhaustive list of cheeses, aged meats, and sauces (see Related Resources). For other products, patients often can obtain information directly from the manufacturer. In many parts of the world, assays for tyramine content are required as a demonstration of adequate product safety procedures. Even the most highly aged cheeses with a tyramine content of 1,000 g/kg can be enjoyed in small amounts (<1 oz), and most products would require heroic intake to achieve clinically significant tyramine ingestion (Table 3¹⁹).

Improved education can clarify the risks

Medications such as lithium, clozapine, and MAOIs have a proven record of efficacy, yet often are underused due to fears engendered by lack of systematic training. A recent initiative in New York thus aimed to increase rates of clozapine prescribing by providing clinicians with an education consultation center.³⁰ Similarly, enhanced education regarding MAOIs could increase the use of these highly effective medications. With a better understanding of MAOIs, clinicians can become adept at using these medications, and therefore expand the armamentarium of agents available to patients with MDD, as well as to those with panic disorder and social phobia.
 

Despite an abundance of evidenced-based literature supporting monoamine oxidase inhibitors (MAOIs) as an effective treatment for depression, use of these agents has decreased drastically in the past 3 decades. A lack of industry support and the ease of use of other agents are contributing factors, but the biggest impediments to routine use of MAOIs are unfamiliarity with their efficacy advantages and concerns about adverse effects, particularly the risk of hypertensive crises and serotonin syndrome. Many misconceptions regarding these medications are based on outdated data and studies that are no longer reliable.

The goal of this 2-part review is to provide clinicians with updated information regarding MAOIs. Part 1 provides a brief description of:

• the pharmacology of nonselective irreversible MAOIs
• the mechanism by which tyramine induces hypertension
• sources of clinically significant tyramine exposure
• what to tell patients about dietary restrictions and MAOIs.

Part 2 of this guide will cover the risk of serotonin syndrome when MAOIs are combined with inhibitors of serotonin reuptake, how to initiate MAOI therapy, and augmenting MAOIs with other agents.

The pharmacology of MAOIs

First used clinically in the 1950s to treat tuberculosis, MAOIs have a long and interesting history (see the Box “A brief history of monoamine oxidase inhibitors”). Table 1¹ lists MAOIs currently available in the United States, including the MAO-B–specific agent rasagiline, which is used for Parkinson’s disease.

Manipulation of the monoamines serotonin, norepinephrine, and dopamine is fundamental to managing major depressive disorder (MDD), yet only nonselective MAOIs directly promote neurotransmission of all 3 by inhibiting MAO-A and MAO-B.² The Sequenced Treatment Alternatives to Relieve Depression (STAR*D) study demonstrated that <50% of MDD patients achieve remission in monotherapy trials of selective serotonin reuptake inhibitors, serotonin norepinephrine reuptake inhibitors, mirtazapine, or bupropion, necessitating consideration of antidepressant combinations, augmentation options, and eventually irreversible, nonselective MAOIs such as phenelzine, tranylcypromine, or isocarboxazid.^3,4 Nonselective MAOIs thus offer a therapeutic opportunity for patients who do not respond to single or dual-mechanism strategies; moreover, nonselective MAOIs have compelling effectiveness data for other conditions, including panic disorder and social phobia.⁵ Although MAOIs are among the most effective pharmacologic agents for MDD,⁶ they are underutilized because of an inadequate understanding of risk mechanisms and resultant fear of catastrophic outcomes. Because of the difficulties encountered in achieving clinical remission for MDD, the nonselective MAOIs deserve a second look.
 

Differentiation of MAO-A from MAO-B. It is essential to understand the mechanism of action of MAOIs, specifically the impact of MAO-A inhibition. Although the enzyme MAO was known in the 1950s, it wasn’t until 1968 that Johnston⁷ postulated the existence of >1 form. In 1971, Goridis and Neff⁸ used clorgyline to examine the deamination rate by MAO of tyramine and norepinephrine. They found that tyramine appeared to be a substrate of both MAO isoforms, but only 1 of the MAO types was sensitive to the inhibitory effects of clorgyline. They also discerned that norepinephrine was only a substrate for MAO-A, and that this form of MAO was sensitive to clorgyline inhibition. Thus, the forms of MAO were characterized by their preferred substrates (Table 2^9,10), and then later by their tissue distribution. Phenylethylamine is a naturally occurring compound found in foods, such as chocolate, and has an in vitro pharmacology similar to amphetamine but with 1 important difference: it has a short half-life of 5 to 10 minutes after oral ingestion, and therefore no appreciable CNS impact.

Within the CNS, norepinephrine and dopamine neurons possess both MAO forms, with the MAO-A content greater than MAO-B. Serotonergic neurons only contain MAO-B.¹¹ Outside of the CNS, MAO-A predominates, with only platelets and lymphocytes possessing MAO-B activity.¹¹ The overall relative tissue proportions of MAO-A to MAO-B activity are: brain, 25% MAO-A, 75% MAO-B; liver, 50% MAO-A, 50% MAO-B; intestine, 80% MAO-A, 20% MAO-B; and peripheral adrenergic neurons, 90% MAO-A, 10% MAO-B.

Because of its specificity for serotonin and norepinephrine, CNS MAO-A inhibition is necessary for antidepressant effects. MAO-B inhibition by itself does not appear to raise CNS dopamine levels unless exogenous dopamine is supplied.¹¹ All MAOIs used in the United States to treat depression are irreversible, nonselective inhibitors of MAO-A and MAO-B.

Selegiline in oral form generates low plasma levels and primarily inhibits MAO-B. The transdermal form of selegiline achieves significantly greater systemic exposure, and at these higher plasma levels selegiline is a nonselective, irreversible MAOI effective for MDD (Figure 1¹²). Administering selegiline systemically via a transdermal patch avoids clinically significant MAOI effects in the gut, so no dietary warnings exist for the lowest dose (6 mg/24 hours), although there are warnings for the higher dosages (9 mg/24 hours and 12 mg/24 hours).

MAOIs and tyramine: Dietary requirements

Clinicians who are familiar with MAOIs recognize that there are dietary restrictions to minimize patients’ exposure to tyramine. As most clinicians know, significant tyramine ingestion may cause an increase in blood pressure (BP) in patients taking an MAOI, but many overestimate the prevalence of foods high in tyramine content since the original reports emerged in the early 1960s.¹⁸ In a recent monograph, one of the leading experts on MAOIs, Professor Ken Gillman, stated:

Very few foods now contain problematically high tyramine levels, that is a result of great changes in international food production methods and hygiene regulations. Cheese is the only food that, in the past, has been associated with documented fatalities resulting from hypertension. Nowadays most cheeses are quite safe, and even ‘matured’ cheeses are usually safe in healthy-sized portions. The variability of sensitivity to tyramine between individuals, and the sometimes unpredictable amount of tyramine content in foods, means a little knowledge and care are still required.¹⁹

What is tyramine? Tyramine is a biogenic amine that is virtually absent in fresh animal protein sources but is enriched after decay or fermentation.²⁰ Modern food processing and handling methods have significantly limited the tyramine content in processed foods, with the exception of certain cheeses and sauces, as discussed below. Moreover, modern assaying techniques using high-performance liquid chromatography have generated extremely accurate assessments of the tyramine content of specific foods.²¹ Data published prior to 2000 are not reliable, because many of these publications employed outdated methods.¹⁷

When ingested, tyramine is metabolized by gut MAO-A, with doses up to 400 mg causing no known effects, although most people rarely ingest >25 mg during a meal.²² In addition to being a substrate for MAO-A, tyramine is also a substrate for the dopamine transporter, norepinephrine transporter (NET), the vesicular monoamine transporter 2, and TAAR1.²³ Tyramine enters the cell via NET, where it interacts with TAAR1, a G protein-coupled receptor that is responsive to trace amines, such as tyramine, as well as amphetamines.²⁰ The agonist properties at TAAR1 are the presumed site of action for the BP effects of tyramine, because binding results in potent release of norepinephrine.^20,24 When tyramine is supplied to an animal in which MAO-A is inhibited, the decreased peripheral catabolism of tyramine results in markedly increased norepinephrine release by peripheral adrenergic neurons. Moreover, the absence of MAO-A activity in those neurons prevents any norepinephrine breakdown, resulting in robust synaptic norepinephrine delivery and peripheral effects.

All orally administered irreversible MAOIs potently inhibit gut and systemic MAO-A, and are susceptible to the impact of significant tyramine ingestion. The exception is selegiline transdermal (Figure 1¹²), as appreciable gut MAO-A inhibition does not occur until doses >6 mg/24 hours are reached.²² No significant pressor response was seen in participants taking selegiline transdermal, 6 mg/24 hours for 13 days, who consumed a meal that provided 400 mg of tyramine.²² Conversely, for oral agents that produce gut MAO-A inhibition, tyramine doses as low as 8 to 10 mg (when administered as tyramine capsules) may increase systolic pressure by 30 mm Hg.²⁵ The dietary warnings do not apply to rasagiline, which is a selective MAO-B inhibitor, although rasagiline may have an impact on resting BP; the prescribing information for rasagiline includes warnings about hypotension and hypertension.²⁶

What to tell patients about tyramine. Although administering pure tyramine capsules can induce a measurable change in systolic BP, when ingested as food, tyramine doses <50 mg are unlikely to cause an increase in BP sufficient to warrant clinical intervention, although some individuals can be sensitive to 10 to 25 mg.¹⁹ When discussing with patients safety issues related to diet, there are a few important concepts to remember¹⁹:

• In an era when the tyramine content of foods was much higher (1960 to 1964) and MAOI users received no dietary guidance, only 14 deaths were reported among an estimated 1.5 million patients who took MAOIs.
• MAOIs do not raise BP, and their use is associated with orthostasis in some patients.
• Routine exercise or other vigorous activities (eg, weightlifting) can raise systolic pressure well above 200 mm Hg, and routine baseline systolic pressures, ranging from 180 to 220 mm Hg, do not increase the risk of subarachnoid hemorrhage.
• Hospital evaluation is needed only if a substantial amount of tyramine is ingested (eg, estimated ≥100 mg), and self-monitoring shows a systolic BP ≥220 mm Hg over a prolonged period (eg, 2 hours). Ingestion of 100 mg of tyramine would almost certainly have to be intentional, as it would require one to consume 3.5 oz of the most highly tyramine-laden cheeses.
   

Emphasize to patients that only a small number of highly aged cheeses, foods, and sauces contain high quantities of tyramine, and that even these foods can be enjoyed in small amounts. All patients who are prescribed an MAOI also should purchase a portable BP cuff for those rare instances when a dietary indiscretion may have occurred and the person experiences a headache within 1 to 2 hours after tyramine ingestion. Most reactions are self-limited and resolve over 2 to 4 hours.

Patients who ingest ≥100 mg of tyramine should be evaluated by a physician. Under no circumstances should a patient be given a prescription for nifedipine or other medications that can abruptly lower BP, because this may result in complications, including myocardial infarction.^27,28 Counsel patients to remain calm. Some clinicians endorse the use of low doses of benzodiazepines (the equivalent of alprazolam 0.5 mg) to facilitate this, because anxiety elevates BP. A recent emergency room study of patients with an initial systolic BP ≥160 mm Hg or diastolic BP ≥100 mm Hg without end organ damage demonstrated that alprazolam, 0.5 mg, was as effective as captopril, 25 mg, in lowering BP.²⁹

Also, tell patients that if a food is unfamiliar and highly aged or fermented, they should avoid it until they can further inquire about it. In a review, Gillman¹⁹ provides the tyramine content of an exhaustive list of cheeses, aged meats, and sauces (see Related Resources). For other products, patients often can obtain information directly from the manufacturer. In many parts of the world, assays for tyramine content are required as a demonstration of adequate product safety procedures. Even the most highly aged cheeses with a tyramine content of 1,000 g/kg can be enjoyed in small amounts (<1 oz), and most products would require heroic intake to achieve clinically significant tyramine ingestion (Table 3¹⁹).

Improved education can clarify the risks

Medications such as lithium, clozapine, and MAOIs have a proven record of efficacy, yet often are underused due to fears engendered by lack of systematic training. A recent initiative in New York thus aimed to increase rates of clozapine prescribing by providing clinicians with an education consultation center.³⁰ Similarly, enhanced education regarding MAOIs could increase the use of these highly effective medications. With a better understanding of MAOIs, clinicians can become adept at using these medications, and therefore expand the armamentarium of agents available to patients with MDD, as well as to those with panic disorder and social phobia.
 

Compared with first-generation antipsychotics, second-generation antipsychotics (SGAs) have a lower risk for extrapyramidal symptoms. Yet tardive dyskinesia (TD) remains a concern because of the widespread use of SGAs for multiple indications.¹ Prior to April 2017, clinicians had no FDA-approved TD treatment options. The most widely used agent worldwide, tetrabenazine, had positive efficacy data in TD trials over the past 45 years but was not available in the United States until 2008, and its sole indication was for chorea associated with Huntington’s disease.² Moreover, the use of tetrabenazine involved slow titration, multiple daily dosing, cytochrome P450 (CYP) 2D6 genotyping for doses >50 mg/d, and tolerability issues.

Tetrabenazine is an inhibitor of vesicular monoamine transport type 2 (VMAT2), a transport protein located almost exclusively in the CNS whose role is to place monoamine neurotransmitters (dopamine, serotonin, norepinephrine) into presynaptic vesicles. By decreasing dopamine transport into these presynaptic vesicles, synaptic dopamine release is lessened, thus reducing postsynaptic dopamine D2 receptor activity and the severity of dyskinetic movements.¹

In 2 pivotal 12-week clinical trials, deutetrabenazine significantly reduced TD severity as measured by Abnormal Involuntary Movement Scale (AIMS) scores (see Efficacy).^6,7

Clinical implications

TD remains a substantial public health concern due to the increasing use of antipsychotics for mood and other disorders beyond the initial indications for schizophrenia.¹ Although exposure to dopamine D2antagonism results in postsynaptic receptor upregulation and supersensitivity that underlies the development of dyskinesia, this process is often rapidly reversible in animal models.¹ The persistence of TD symptoms in up to 80% of patients after dopamine receptor blocking agents (DRBAs) are stopped has led to hypotheses that the underlying pathophysiology of TD is also a problem with neuroplasticity. Aside from DRBA exposure, environmental factors (eg, oxidative stress) and genetic predisposition might contribute to TD risk.¹

Before 2017, only 1 medication (branched-chain amino acids) had been FDA-approved for treating TD in the United States, and only a few existing medications (clonazepam, amantadine, and ginkgo biloba extract [EGb-761]) had positive results from controlled trials, most with small effect sizes.⁸ Moreover, there was only 1 controlled trial each for clonazepam and EGb-761.¹ A branched-chain amino acid preparation received FDA approval for managing TD in male patients, but is no longer commercially available, except from compounding pharmacies.⁹

Tetrabenazine was developed in the mid-1950s to avoid orthostasis and sedation associated with reserpine.¹⁰ Both reserpine and tetrabenazine proved effective for TD,¹¹ but tetrabenazine lacked reserpine’s peripheral adverse effects. However, the kinetics of tetrabenazine necessitated multiple daily doses, and CYP2D6 genotyping was required for doses >50 mg/d.²

Receptor blocking. The mechanism that distinguishes the clinical profiles of reserpine and tetrabenazine relates to their differential properties at VMAT.¹² VMAT exists in 2 forms (VMAT1 and VMAT2) that vary in distribution, with VMAT1 expressed mainly in the peripheral nervous system and VMAT2 expressed mainly in monoaminergic cells of the CNS.¹³ Tetrabenazine is a specific and reversible VMAT2 inhibitor, whereas reserpine is an irreversible and nonselective antagonist of VMAT1 and VMAT2. It is reserpine’s VMAT1 inhibition that results in peripheral adverse effects such as orthostasis. Tetrabenazine is rapidly and extensively converted into 2 isomers, alpha-dihydrotetrabenazine (α-DHTBZ) and beta-dihydrotetrabenazine (β-DHTBZ), both of which are metabolized by CYP2D6, with a role for CYP3A4 in α-DHTBZ metabolism.¹ These DHTBZ metabolites have a short half-life when generated from oral tetrabenazine, a feature that necessitates multiple daily dosing; moreover, the existence of 2D6 polymorphisms led to FDA-mandated CYP2D6 genotyping for tetrabenazine doses >50 mg/d when it was approved for Huntington’s chorea. The concern is that 2D6 poor metabolizers will have excessive exposure to the VMAT2 effects of DHTBZ, resulting in sedation, akathisia, parkinsonism, and mood symptoms.²

How deuterium impacts medication kinetics. Deuterium is a naturally occurring, stable, nontoxic isotope of hydrogen. Humans have 5 g of deuterium in their body at any time, mostly in the form of heavy water (D₂O).¹⁴ When deuterium is used to replace selected hydrogen atoms, the resulting molecule will have similar configuration and receptor-binding properties but markedly different kinetics. Because the carbon–deuterium covalent bond requires 8 times more energy to break than a carbon–hydrogen bond, the half-life is prolonged.¹⁵ Utilizing this knowledge, a deuterated form of tetrabenazine, deutetrabenazine, was synthesized with such a purpose in mind. While the active metabolites of deutetrabenazine retain the VMAT2 affinity of non-deuterated tetrabenazine, the substitution of deuterium for hydrogen at specific positions slows the breakdown of metabolites, resulting in sustained duration of action, greater active drug exposure, and less impact of 2D6 genotype on drug exposure, thus eliminating the need for genotyping, unless one wants to exceed 36 mg/d.

Deutetrabenazine was first studied in Huntington’s chorea in a 13-week, double-blind, placebo-controlled, parallel-group study (N = 90).⁴ The maximum daily deutetrabenazine dose was 48 mg, but reduced to 36 mg in those taking strong CYP2D6 inhibitors (bupropion, fluoxetine, or paroxetine). Blinded 2D6 genotyping was performed, but there was no dose modification required based on 2D6 genotype. There was a 36.4% reduction in total maximal chorea score for deutetrabenazine compared with 14.4% for placebo (P < .001).⁴ Importantly, adverse effects were comparable between both groups, with 1 drop-out in the deutetrabenazine arm vs 2 in the placebo arm. The only adverse event occurring in ≥5% of deutetrabenazine participants and at a rate ≥2 times that of placebo was somnolence: 11.1% for deutetrabenazine vs 4.4% for placebo.⁴ The mean deutetrabenazine daily dose at the end of the treatment period was 39.7 ± 9.3 mg, and for those with impaired CYP2D6 function (poor metabolizers or those taking strong CYP2D6 inhibiting medications), the mean daily dose was 34.8 mg ± 3.8 mg.⁴

Use in tardive dyskinesia. The recommended starting dosage for TD treatment is 6 mg, twice daily with food. The dose may be increased at weekly intervals in increments of 6 mg/d to a maximum recommended daily dosage of 48 mg.⁵ The maximum daily dose is 36 mg (18 mg, twice daily) in patients receiving strong CYP2D6 inhibitors or who are 2D6 poor metabolizers.⁵

Deutetrabenazine has not been studied in those with moderate or severe hepatic impairment, and its use is contraindicated in these patients.⁵ No clinical studies have been conducted to assess the effect of renal impairment on the pharmacokinetics of deutetrabenazine.⁵

Pharmacologic profile, adverse reactions

When the data from the two 12-week, phase 3 placebo-controlled studies were pooled, the most common adverse reactions occurring in >3% of deutetrabenazine patients and greater than placebo were nasopharyngeal symptoms (4% vs 2% placebo) and insomnia (4% vs 1% placebo).⁵ Importantly, in neither TD study were there clinically significant changes in rating scales for depression, suicidal ideation and behavior, or parkinsonism. There also were no clinically significant changes in measures of schizophrenia symptoms. The mean QT prolongation for a single 24 mg dose of deutetrabenazine in healthy volunteers was 4.5 milliseconds, with the upper bound of the double-sided 90% confidence interval reaching 6.5 milliseconds.⁵ For tetrabenazine, single 50 mg doses administered to volunteers resulted in mean QT prolongation of 8 milliseconds.⁵ In patients requiring deutetrabenazine doses >24 mg/d who are taking other medications known to prolong QTc, assess the QTc interval before and after increasing the dose of deutetrabenazine or other medications that are known to prolong QTc.⁵

How it works

Tetrabenazine is the only agent that has demonstrated significant efficacy for TD management, but its use involves slow titration, multiple daily dosing, CYP2D6 genotyping for doses >50 mg/d, and tolerability issues. For example, the most common adverse effects in the pivotal tetrabenazine Huntington’s disease trial were sedation/somnolence (tetrabenazine 31% vs 3% for placebo), insomnia (tetrabenazine 22% vs 0% for placebo), depression (tetrabenazine 19% vs 0% for placebo), fatigue (tetrabenazine 22% vs 13% for placebo), and akathisia (tetrabenazine 19% vs 0% for placebo).² For comparison, the only adverse event occurring in ≥5% of deutetrabenazine participants and at a rate ≥2 times that of placebo in the pivotal Huntington’s disease trial was somnolence (11.1% for deutetrabenazine vs 4.4% for placebo).⁴

Pharmacokinetics

Deutetrabenazine has 80% oral bioavailability, and is rapidly converted to its active metabolites after oral dosing (Table 2).⁵ Linear dose dependence of Cmax and area under the curve (AUC) was observed for the active metabolites following single or multiple doses of deutetrabenazine (6 to 24 mg and 7.5 to 22.5 mg, twice daily).¹⁵ Cmax of deuterated α-DHTBZ and β-DHTBZ is reached within 3 to 4 hours after dosing, with a steady state ratio of 3:1 for the α-DHTBZ vs the β-DHTBZ form. Food had no effect on AUC, but did increase Cmax by 50%.⁵

Deutetrabenazine is metabolized through carbonyl reductase enzymes to its active metabolites, and these are further metabolized through multiple CYP pathways, predominantly 2D6 and to a lesser extent 3A4. The effect of CYP2D6 inhibition on the pharma­cokinetics of deutetrabenazine and its α-DHTBZ and β-DHTBZ metabolites was studied in 24 healthy participants following a single 22.5 mg dose of deutetrabenazine given after 8 days of administration of the strong CYP2D6 inhibitor paroxetine, 20 mg/d. In the presence of paroxetine, systemic exposure (AUC) of α-DHTBZ was 1.9-fold higher and β-DHTBZ was 6.5-fold higher, resulting in an approximately 3-fold increase in AUC for total (α+β)-DHTBZ, with corresponding increases in mean half-life of approximately 1.5-fold and 2.7-fold, respectively.⁵ Neither deutetrabenazine or its metabolites are inhibitors or inducers of major CYP enzymes. Aside from VMAT2, the results of in vitro studies suggest that deutetrabenazine and its active metabolites are unlikely to inhibit most major drug transporters at clinically relevant concentrations.

Efficacy

Efficacy was established in two 12-week, double-blind, placebo-controlled trials of adult patients with TD (ages 18 to 80).^6,7 Eligible participants had:

• TD diagnosis for ≥3 months before screening and a history of DRBA treatment for ≥3 months (≥1 month if age ≥60)
• Total AIMS motor score ≥6 (items 1 to 7) at both screening and baseline, verified by a blinded central rater at screening via central video rating
• Patients with an underlying psychiatric illness had to be stable. Psychoactive medication use, including antipsychotics, was allowed if stable for ≥30 days before screening (antidepressants, ≥45 days).

Exclusion criteria included treatment with tetrabenazine, reserpine, α-methyl-p-tyrosine, strong anticholinergic medications, dopamine antagonizing antiemetics (eg, metoclopramide, prochlorperazine, promethazine), dopamine agonists, levodopa, stimulants, or a monoamine oxidase inhibitor (MAOI) within 30 days of screening or baseline, or treatment with botulinum toxin within 3 months of screening; and presence of a neurologic condition that could confound TD assessments, serious untreated or undertreated psychiatric illness, or unstable medical illness. Patients with a history of or active suicidal ideation or behavior within 6 months of screening or score ≥11 on the depression subscale of the Hospital Anxiety and Depression Scale were excluded. Those participants with Fridericia-corrected QT interval values >450 milliseconds in men, >460 milliseconds in women, or >480 milli­seconds in patients with a right bundle branch block on electrocardiography at screening also were excluded.

The flexible-dose TD study was performed in 117 participants randomized in a 1:1 manner to deutetrabenazine or placebo, both administered twice daily, titrated to optimal dosage (12 to 48 mg/d) over 6 weeks, and then administered at that dose for another 6 weeks.⁷ The population demographics were: mean age, 54.6 ± 10.3 years, 52.1% female, 69.2% white, and 80.3% receiving ongoing dopamine antagonists, with a mean TD duration of 74.7 ± 81.5 months. Sixty-eight percent had schizophrenia spectrum disorders, and 30% had mood disorders. The primary outcome was change in total AIMS score (items 1 to 7) assessed by central, independent raters. The mean baseline AIMS score for items 1 to 7 was 9.6 ± 3.9, with 82.9% of participants with baseline AIMS scores ≥6. Study treatment retention was high: placebo 88.1%, deutetrabenazine 89.7%.⁷ There was a mean 3 point decrease in AIMS score for deutetrabenazine compared with 1.4 for placebo (P = .019). Among those with baseline AIMS scores ≥6, there was a 3.4 point decrease in AIMS scores for deutetrabenazine compared with a 1.9 point decrease for placebo (P = .027). The only adverse effects that occurred in ≥5% of deutetrabenazine participants and at a rate ≥2 times the rate in placebo were insomnia (deutetra­benazine 6.9% vs placebo 1.7%) and akathisia (deutetrabenazine 5.2% vs placebo 0%).

The fixed-dose TD study was performed in 293 participants randomized in 1:1:1:1 manner to 1 of 3 fixed doses of deutetrabenazine (12 mg/d, 24 mg/d, or 36 mg/d) or placebo, both administered twice daily.⁶ The starting dose of deutetrabenazine was 6 mg twice daily. During the dose escalation period (through Week 4), the dose of study drug was increased weekly in increments of 6 mg/d until the randomized dose was achieved. Patients continued to receive the dose they were assigned to over a maintenance period of 8 weeks.⁶ The population demographics were: mean age, 56.4 ± 11.3 years, 55% female, 79% white, 76% receiving ongoing dopamine antagonists, with a mean TD duration of 67.2 ± 66 months. Sixty percent had schizophrenia spectrum disorders, and 36% had mood disorders. The primary outcome was change in AIMS total score (items 1 to 7) assessed by central, independent raters. The mean AIMS score at baseline was 9.5 ± 2.7 in the placebo group, and for deutetrabenazine: 9.6 ± 2.4 in the 12 mg/d group, 9.4 ± 2.9 in the 24 mg/d group, and 10.1 ± 3.2 in the 36 mg/d group. The 24 mg/d and 36 mg/d doses significantly reduced AIMS scores from baseline vs placebo: 36 mg: −3.3 (0.42) vs −1.4 (0.41) (P = .001); 24 mg: −3.2 (0.45) vs −1.4 (0.41) (P = .003). Study treatment retention rates were high: placebo 90.5%, deutetrabenazine 88%. Across all doses, only 1 adverse effect occurred in ≥5% of deutetrabenazine participants: headache (5% deutetrabenazine vs 6% placebo). At the highest dose, 36 mg/d, the only adverse effects that occurred in ≥5% of participants were diarrhea (7% deutetrabenazine vs 3% placebo) and headache (7% deutetrabenazine vs 6% placebo).

Outcome. In the flexible-dose study (mean dose 38.8 ± 7.92 mg/d), the deutetrabenazine arm experienced a mean 30% reduction in AIMS scores from baseline at the Week 12 endpoint. Compared with placebo, the mean reduction in AIMS scores (standard error) was: −3.0 (0.45) deutetrabenazine vs −1.6 (0.46) placebo (P = .019).⁷ For the fixed-dose study, the 24 mg/d and 36 mg/d doses significantly reduced AIMS scores from baseline vs placebo: 36 mg: −3.3 (0.42) vs −1.4 (0.41) (P = .001); 24 mg: −3.2 (0.45) vs −1.4 (0.41) (P = .003). In addition to these mean changes from baseline, 35% of the 24 mg/d group and 33% of the 36 mg/d group demonstrated ≥50% reduction in AIMS scores.⁶

Tolerability

In the 2 phase 3 trials, there were no adverse effects occurring with an incidence ≥5% and at least twice the rate of placebo.⁵ Discontinuations because of adverse events were low in both pivotal studies across all treatment groups: 3.4% for placebo vs 1.7% for deutetrabenazine in the flexible-dose trial,⁷ and 3% for placebo vs 4% for deutetrabenazine in the fixed-dose study.⁶ In neither trial were there clinically significant changes in ratings of depression, suicidality, parkinsonism, or schizophrenia symptoms. The mean QT prolongation in healthy volunteers is described above.

Clinical considerations

Unique properties. Deutetrabenazine utilizes the greater bond strength of the carbon–deuterium bond to slow CYP metabolism, resulting in prolonged duration of action that is well tolerated, and provides significant efficacy.

Why Rx? The reasons to prescribe deutetrabenazine for TD patients include:

• only 1 of 2 agents with FDA approval for TD
• fewer tolerability issues than with tetrabenazine
• lower sedation rates in TD trials than with valbenazine
• no signal for effects on mood para­meters or rates of parkinsonism when used for TD.

Dosing

The recommended starting dosage of deutetrabenazine is 6 mg twice daily taken with food, increasing by 6 mg/d weekly as needed, with a maximum dose of 48 mg/d or 36 mg/d in those taking strong CYP2D6 inhibitors or who are 2D6 poor metabolizers. Deutetrabenazine is contraindicated in patients with hepatic impairment (as determined by Child-Pugh criteria¹⁶). There are no data in patients with renal impairment. The combined efficacy and tolerability of dosages >48 mg/d has not been evaluated. Overdoses of tetrabenazine ranging from 100 to 1,000 mg have been reported in the literature and were associated with acute dystonia, oculogyric crisis, nausea and vomiting, sweating, sedation, hypotension, confusion, diarrhea, hallucinations, rubor, and tremor.⁵

Contraindications

When used for TD, deutetrabenazine is contraindicated for patients taking reserpine, tetrabenazine, valbenazine, or MAOIs, and for patients with hepatic impairment. As with most medications, there are no data on deutetrabenazine use in pregnant women; however, oral administration of deutetrabenazine (5, 10, or 30 mg/kg/d) or tetrabenazine (30 mg/kg/d) to pregnant rats during organogenesis had no clear effect on embryofetal development. The highest dose tested was 6 times the maximum recommended human dose of 48 mg/d on a body surface area (mg/m²) basis. There are no data on the presence of deutetrabenazine or its metabolites in human milk, the effects on the breastfed infant, or the effects of the drug on milk production.

Compared with first-generation antipsychotics, second-generation antipsychotics (SGAs) have a lower risk for extrapyramidal symptoms. Yet tardive dyskinesia (TD) remains a concern because of the widespread use of SGAs for multiple indications.¹ Prior to April 2017, clinicians had no FDA-approved TD treatment options. The most widely used agent worldwide, tetrabenazine, had positive efficacy data in TD trials over the past 45 years but was not available in the United States until 2008, and its sole indication was for chorea associated with Huntington’s disease.² Moreover, the use of tetrabenazine involved slow titration, multiple daily dosing, cytochrome P450 (CYP) 2D6 genotyping for doses >50 mg/d, and tolerability issues.

Tetrabenazine is an inhibitor of vesicular monoamine transport type 2 (VMAT2), a transport protein located almost exclusively in the CNS whose role is to place monoamine neurotransmitters (dopamine, serotonin, norepinephrine) into presynaptic vesicles. By decreasing dopamine transport into these presynaptic vesicles, synaptic dopamine release is lessened, thus reducing postsynaptic dopamine D2 receptor activity and the severity of dyskinetic movements.¹

In 2 pivotal 12-week clinical trials, deutetrabenazine significantly reduced TD severity as measured by Abnormal Involuntary Movement Scale (AIMS) scores (see Efficacy).^6,7

Clinical implications

TD remains a substantial public health concern due to the increasing use of antipsychotics for mood and other disorders beyond the initial indications for schizophrenia.¹ Although exposure to dopamine D2antagonism results in postsynaptic receptor upregulation and supersensitivity that underlies the development of dyskinesia, this process is often rapidly reversible in animal models.¹ The persistence of TD symptoms in up to 80% of patients after dopamine receptor blocking agents (DRBAs) are stopped has led to hypotheses that the underlying pathophysiology of TD is also a problem with neuroplasticity. Aside from DRBA exposure, environmental factors (eg, oxidative stress) and genetic predisposition might contribute to TD risk.¹

Before 2017, only 1 medication (branched-chain amino acids) had been FDA-approved for treating TD in the United States, and only a few existing medications (clonazepam, amantadine, and ginkgo biloba extract [EGb-761]) had positive results from controlled trials, most with small effect sizes.⁸ Moreover, there was only 1 controlled trial each for clonazepam and EGb-761.¹ A branched-chain amino acid preparation received FDA approval for managing TD in male patients, but is no longer commercially available, except from compounding pharmacies.⁹

Tetrabenazine was developed in the mid-1950s to avoid orthostasis and sedation associated with reserpine.¹⁰ Both reserpine and tetrabenazine proved effective for TD,¹¹ but tetrabenazine lacked reserpine’s peripheral adverse effects. However, the kinetics of tetrabenazine necessitated multiple daily doses, and CYP2D6 genotyping was required for doses >50 mg/d.²

Receptor blocking. The mechanism that distinguishes the clinical profiles of reserpine and tetrabenazine relates to their differential properties at VMAT.¹² VMAT exists in 2 forms (VMAT1 and VMAT2) that vary in distribution, with VMAT1 expressed mainly in the peripheral nervous system and VMAT2 expressed mainly in monoaminergic cells of the CNS.¹³ Tetrabenazine is a specific and reversible VMAT2 inhibitor, whereas reserpine is an irreversible and nonselective antagonist of VMAT1 and VMAT2. It is reserpine’s VMAT1 inhibition that results in peripheral adverse effects such as orthostasis. Tetrabenazine is rapidly and extensively converted into 2 isomers, alpha-dihydrotetrabenazine (α-DHTBZ) and beta-dihydrotetrabenazine (β-DHTBZ), both of which are metabolized by CYP2D6, with a role for CYP3A4 in α-DHTBZ metabolism.¹ These DHTBZ metabolites have a short half-life when generated from oral tetrabenazine, a feature that necessitates multiple daily dosing; moreover, the existence of 2D6 polymorphisms led to FDA-mandated CYP2D6 genotyping for tetrabenazine doses >50 mg/d when it was approved for Huntington’s chorea. The concern is that 2D6 poor metabolizers will have excessive exposure to the VMAT2 effects of DHTBZ, resulting in sedation, akathisia, parkinsonism, and mood symptoms.²

How deuterium impacts medication kinetics. Deuterium is a naturally occurring, stable, nontoxic isotope of hydrogen. Humans have 5 g of deuterium in their body at any time, mostly in the form of heavy water (D₂O).¹⁴ When deuterium is used to replace selected hydrogen atoms, the resulting molecule will have similar configuration and receptor-binding properties but markedly different kinetics. Because the carbon–deuterium covalent bond requires 8 times more energy to break than a carbon–hydrogen bond, the half-life is prolonged.¹⁵ Utilizing this knowledge, a deuterated form of tetrabenazine, deutetrabenazine, was synthesized with such a purpose in mind. While the active metabolites of deutetrabenazine retain the VMAT2 affinity of non-deuterated tetrabenazine, the substitution of deuterium for hydrogen at specific positions slows the breakdown of metabolites, resulting in sustained duration of action, greater active drug exposure, and less impact of 2D6 genotype on drug exposure, thus eliminating the need for genotyping, unless one wants to exceed 36 mg/d.

Deutetrabenazine was first studied in Huntington’s chorea in a 13-week, double-blind, placebo-controlled, parallel-group study (N = 90).⁴ The maximum daily deutetrabenazine dose was 48 mg, but reduced to 36 mg in those taking strong CYP2D6 inhibitors (bupropion, fluoxetine, or paroxetine). Blinded 2D6 genotyping was performed, but there was no dose modification required based on 2D6 genotype. There was a 36.4% reduction in total maximal chorea score for deutetrabenazine compared with 14.4% for placebo (P < .001).⁴ Importantly, adverse effects were comparable between both groups, with 1 drop-out in the deutetrabenazine arm vs 2 in the placebo arm. The only adverse event occurring in ≥5% of deutetrabenazine participants and at a rate ≥2 times that of placebo was somnolence: 11.1% for deutetrabenazine vs 4.4% for placebo.⁴ The mean deutetrabenazine daily dose at the end of the treatment period was 39.7 ± 9.3 mg, and for those with impaired CYP2D6 function (poor metabolizers or those taking strong CYP2D6 inhibiting medications), the mean daily dose was 34.8 mg ± 3.8 mg.⁴

Use in tardive dyskinesia. The recommended starting dosage for TD treatment is 6 mg, twice daily with food. The dose may be increased at weekly intervals in increments of 6 mg/d to a maximum recommended daily dosage of 48 mg.⁵ The maximum daily dose is 36 mg (18 mg, twice daily) in patients receiving strong CYP2D6 inhibitors or who are 2D6 poor metabolizers.⁵

Deutetrabenazine has not been studied in those with moderate or severe hepatic impairment, and its use is contraindicated in these patients.⁵ No clinical studies have been conducted to assess the effect of renal impairment on the pharmacokinetics of deutetrabenazine.⁵

Pharmacologic profile, adverse reactions

When the data from the two 12-week, phase 3 placebo-controlled studies were pooled, the most common adverse reactions occurring in >3% of deutetrabenazine patients and greater than placebo were nasopharyngeal symptoms (4% vs 2% placebo) and insomnia (4% vs 1% placebo).⁵ Importantly, in neither TD study were there clinically significant changes in rating scales for depression, suicidal ideation and behavior, or parkinsonism. There also were no clinically significant changes in measures of schizophrenia symptoms. The mean QT prolongation for a single 24 mg dose of deutetrabenazine in healthy volunteers was 4.5 milliseconds, with the upper bound of the double-sided 90% confidence interval reaching 6.5 milliseconds.⁵ For tetrabenazine, single 50 mg doses administered to volunteers resulted in mean QT prolongation of 8 milliseconds.⁵ In patients requiring deutetrabenazine doses >24 mg/d who are taking other medications known to prolong QTc, assess the QTc interval before and after increasing the dose of deutetrabenazine or other medications that are known to prolong QTc.⁵

How it works

Tetrabenazine is the only agent that has demonstrated significant efficacy for TD management, but its use involves slow titration, multiple daily dosing, CYP2D6 genotyping for doses >50 mg/d, and tolerability issues. For example, the most common adverse effects in the pivotal tetrabenazine Huntington’s disease trial were sedation/somnolence (tetrabenazine 31% vs 3% for placebo), insomnia (tetrabenazine 22% vs 0% for placebo), depression (tetrabenazine 19% vs 0% for placebo), fatigue (tetrabenazine 22% vs 13% for placebo), and akathisia (tetrabenazine 19% vs 0% for placebo).² For comparison, the only adverse event occurring in ≥5% of deutetrabenazine participants and at a rate ≥2 times that of placebo in the pivotal Huntington’s disease trial was somnolence (11.1% for deutetrabenazine vs 4.4% for placebo).⁴

Pharmacokinetics

Deutetrabenazine has 80% oral bioavailability, and is rapidly converted to its active metabolites after oral dosing (Table 2).⁵ Linear dose dependence of Cmax and area under the curve (AUC) was observed for the active metabolites following single or multiple doses of deutetrabenazine (6 to 24 mg and 7.5 to 22.5 mg, twice daily).¹⁵ Cmax of deuterated α-DHTBZ and β-DHTBZ is reached within 3 to 4 hours after dosing, with a steady state ratio of 3:1 for the α-DHTBZ vs the β-DHTBZ form. Food had no effect on AUC, but did increase Cmax by 50%.⁵

Deutetrabenazine is metabolized through carbonyl reductase enzymes to its active metabolites, and these are further metabolized through multiple CYP pathways, predominantly 2D6 and to a lesser extent 3A4. The effect of CYP2D6 inhibition on the pharma­cokinetics of deutetrabenazine and its α-DHTBZ and β-DHTBZ metabolites was studied in 24 healthy participants following a single 22.5 mg dose of deutetrabenazine given after 8 days of administration of the strong CYP2D6 inhibitor paroxetine, 20 mg/d. In the presence of paroxetine, systemic exposure (AUC) of α-DHTBZ was 1.9-fold higher and β-DHTBZ was 6.5-fold higher, resulting in an approximately 3-fold increase in AUC for total (α+β)-DHTBZ, with corresponding increases in mean half-life of approximately 1.5-fold and 2.7-fold, respectively.⁵ Neither deutetrabenazine or its metabolites are inhibitors or inducers of major CYP enzymes. Aside from VMAT2, the results of in vitro studies suggest that deutetrabenazine and its active metabolites are unlikely to inhibit most major drug transporters at clinically relevant concentrations.

Efficacy

Efficacy was established in two 12-week, double-blind, placebo-controlled trials of adult patients with TD (ages 18 to 80).^6,7 Eligible participants had:

• TD diagnosis for ≥3 months before screening and a history of DRBA treatment for ≥3 months (≥1 month if age ≥60)
• Total AIMS motor score ≥6 (items 1 to 7) at both screening and baseline, verified by a blinded central rater at screening via central video rating
• Patients with an underlying psychiatric illness had to be stable. Psychoactive medication use, including antipsychotics, was allowed if stable for ≥30 days before screening (antidepressants, ≥45 days).

Exclusion criteria included treatment with tetrabenazine, reserpine, α-methyl-p-tyrosine, strong anticholinergic medications, dopamine antagonizing antiemetics (eg, metoclopramide, prochlorperazine, promethazine), dopamine agonists, levodopa, stimulants, or a monoamine oxidase inhibitor (MAOI) within 30 days of screening or baseline, or treatment with botulinum toxin within 3 months of screening; and presence of a neurologic condition that could confound TD assessments, serious untreated or undertreated psychiatric illness, or unstable medical illness. Patients with a history of or active suicidal ideation or behavior within 6 months of screening or score ≥11 on the depression subscale of the Hospital Anxiety and Depression Scale were excluded. Those participants with Fridericia-corrected QT interval values >450 milliseconds in men, >460 milliseconds in women, or >480 milli­seconds in patients with a right bundle branch block on electrocardiography at screening also were excluded.

The flexible-dose TD study was performed in 117 participants randomized in a 1:1 manner to deutetrabenazine or placebo, both administered twice daily, titrated to optimal dosage (12 to 48 mg/d) over 6 weeks, and then administered at that dose for another 6 weeks.⁷ The population demographics were: mean age, 54.6 ± 10.3 years, 52.1% female, 69.2% white, and 80.3% receiving ongoing dopamine antagonists, with a mean TD duration of 74.7 ± 81.5 months. Sixty-eight percent had schizophrenia spectrum disorders, and 30% had mood disorders. The primary outcome was change in total AIMS score (items 1 to 7) assessed by central, independent raters. The mean baseline AIMS score for items 1 to 7 was 9.6 ± 3.9, with 82.9% of participants with baseline AIMS scores ≥6. Study treatment retention was high: placebo 88.1%, deutetrabenazine 89.7%.⁷ There was a mean 3 point decrease in AIMS score for deutetrabenazine compared with 1.4 for placebo (P = .019). Among those with baseline AIMS scores ≥6, there was a 3.4 point decrease in AIMS scores for deutetrabenazine compared with a 1.9 point decrease for placebo (P = .027). The only adverse effects that occurred in ≥5% of deutetrabenazine participants and at a rate ≥2 times the rate in placebo were insomnia (deutetra­benazine 6.9% vs placebo 1.7%) and akathisia (deutetrabenazine 5.2% vs placebo 0%).

The fixed-dose TD study was performed in 293 participants randomized in 1:1:1:1 manner to 1 of 3 fixed doses of deutetrabenazine (12 mg/d, 24 mg/d, or 36 mg/d) or placebo, both administered twice daily.⁶ The starting dose of deutetrabenazine was 6 mg twice daily. During the dose escalation period (through Week 4), the dose of study drug was increased weekly in increments of 6 mg/d until the randomized dose was achieved. Patients continued to receive the dose they were assigned to over a maintenance period of 8 weeks.⁶ The population demographics were: mean age, 56.4 ± 11.3 years, 55% female, 79% white, 76% receiving ongoing dopamine antagonists, with a mean TD duration of 67.2 ± 66 months. Sixty percent had schizophrenia spectrum disorders, and 36% had mood disorders. The primary outcome was change in AIMS total score (items 1 to 7) assessed by central, independent raters. The mean AIMS score at baseline was 9.5 ± 2.7 in the placebo group, and for deutetrabenazine: 9.6 ± 2.4 in the 12 mg/d group, 9.4 ± 2.9 in the 24 mg/d group, and 10.1 ± 3.2 in the 36 mg/d group. The 24 mg/d and 36 mg/d doses significantly reduced AIMS scores from baseline vs placebo: 36 mg: −3.3 (0.42) vs −1.4 (0.41) (P = .001); 24 mg: −3.2 (0.45) vs −1.4 (0.41) (P = .003). Study treatment retention rates were high: placebo 90.5%, deutetrabenazine 88%. Across all doses, only 1 adverse effect occurred in ≥5% of deutetrabenazine participants: headache (5% deutetrabenazine vs 6% placebo). At the highest dose, 36 mg/d, the only adverse effects that occurred in ≥5% of participants were diarrhea (7% deutetrabenazine vs 3% placebo) and headache (7% deutetrabenazine vs 6% placebo).

Outcome. In the flexible-dose study (mean dose 38.8 ± 7.92 mg/d), the deutetrabenazine arm experienced a mean 30% reduction in AIMS scores from baseline at the Week 12 endpoint. Compared with placebo, the mean reduction in AIMS scores (standard error) was: −3.0 (0.45) deutetrabenazine vs −1.6 (0.46) placebo (P = .019).⁷ For the fixed-dose study, the 24 mg/d and 36 mg/d doses significantly reduced AIMS scores from baseline vs placebo: 36 mg: −3.3 (0.42) vs −1.4 (0.41) (P = .001); 24 mg: −3.2 (0.45) vs −1.4 (0.41) (P = .003). In addition to these mean changes from baseline, 35% of the 24 mg/d group and 33% of the 36 mg/d group demonstrated ≥50% reduction in AIMS scores.⁶

Tolerability

In the 2 phase 3 trials, there were no adverse effects occurring with an incidence ≥5% and at least twice the rate of placebo.⁵ Discontinuations because of adverse events were low in both pivotal studies across all treatment groups: 3.4% for placebo vs 1.7% for deutetrabenazine in the flexible-dose trial,⁷ and 3% for placebo vs 4% for deutetrabenazine in the fixed-dose study.⁶ In neither trial were there clinically significant changes in ratings of depression, suicidality, parkinsonism, or schizophrenia symptoms. The mean QT prolongation in healthy volunteers is described above.

Clinical considerations

Unique properties. Deutetrabenazine utilizes the greater bond strength of the carbon–deuterium bond to slow CYP metabolism, resulting in prolonged duration of action that is well tolerated, and provides significant efficacy.

Why Rx? The reasons to prescribe deutetrabenazine for TD patients include:

• only 1 of 2 agents with FDA approval for TD
• fewer tolerability issues than with tetrabenazine
• lower sedation rates in TD trials than with valbenazine
• no signal for effects on mood para­meters or rates of parkinsonism when used for TD.

Dosing

The recommended starting dosage of deutetrabenazine is 6 mg twice daily taken with food, increasing by 6 mg/d weekly as needed, with a maximum dose of 48 mg/d or 36 mg/d in those taking strong CYP2D6 inhibitors or who are 2D6 poor metabolizers. Deutetrabenazine is contraindicated in patients with hepatic impairment (as determined by Child-Pugh criteria¹⁶). There are no data in patients with renal impairment. The combined efficacy and tolerability of dosages >48 mg/d has not been evaluated. Overdoses of tetrabenazine ranging from 100 to 1,000 mg have been reported in the literature and were associated with acute dystonia, oculogyric crisis, nausea and vomiting, sweating, sedation, hypotension, confusion, diarrhea, hallucinations, rubor, and tremor.⁵

Contraindications

When used for TD, deutetrabenazine is contraindicated for patients taking reserpine, tetrabenazine, valbenazine, or MAOIs, and for patients with hepatic impairment. As with most medications, there are no data on deutetrabenazine use in pregnant women; however, oral administration of deutetrabenazine (5, 10, or 30 mg/kg/d) or tetrabenazine (30 mg/kg/d) to pregnant rats during organogenesis had no clear effect on embryofetal development. The highest dose tested was 6 times the maximum recommended human dose of 48 mg/d on a body surface area (mg/m²) basis. There are no data on the presence of deutetrabenazine or its metabolites in human milk, the effects on the breastfed infant, or the effects of the drug on milk production.

Compared with first-generation antipsychotics, second-generation antipsychotics (SGAs) have a lower risk for extrapyramidal symptoms. Yet tardive dyskinesia (TD) remains a concern because of the widespread use of SGAs for multiple indications.¹ Prior to April 2017, clinicians had no FDA-approved TD treatment options. The most widely used agent worldwide, tetrabenazine, had positive efficacy data in TD trials over the past 45 years but was not available in the United States until 2008, and its sole indication was for chorea associated with Huntington’s disease.² Moreover, the use of tetrabenazine involved slow titration, multiple daily dosing, cytochrome P450 (CYP) 2D6 genotyping for doses >50 mg/d, and tolerability issues.

Tetrabenazine is an inhibitor of vesicular monoamine transport type 2 (VMAT2), a transport protein located almost exclusively in the CNS whose role is to place monoamine neurotransmitters (dopamine, serotonin, norepinephrine) into presynaptic vesicles. By decreasing dopamine transport into these presynaptic vesicles, synaptic dopamine release is lessened, thus reducing postsynaptic dopamine D2 receptor activity and the severity of dyskinetic movements.¹

In 2 pivotal 12-week clinical trials, deutetrabenazine significantly reduced TD severity as measured by Abnormal Involuntary Movement Scale (AIMS) scores (see Efficacy).^6,7

Clinical implications

TD remains a substantial public health concern due to the increasing use of antipsychotics for mood and other disorders beyond the initial indications for schizophrenia.¹ Although exposure to dopamine D2antagonism results in postsynaptic receptor upregulation and supersensitivity that underlies the development of dyskinesia, this process is often rapidly reversible in animal models.¹ The persistence of TD symptoms in up to 80% of patients after dopamine receptor blocking agents (DRBAs) are stopped has led to hypotheses that the underlying pathophysiology of TD is also a problem with neuroplasticity. Aside from DRBA exposure, environmental factors (eg, oxidative stress) and genetic predisposition might contribute to TD risk.¹

Before 2017, only 1 medication (branched-chain amino acids) had been FDA-approved for treating TD in the United States, and only a few existing medications (clonazepam, amantadine, and ginkgo biloba extract [EGb-761]) had positive results from controlled trials, most with small effect sizes.⁸ Moreover, there was only 1 controlled trial each for clonazepam and EGb-761.¹ A branched-chain amino acid preparation received FDA approval for managing TD in male patients, but is no longer commercially available, except from compounding pharmacies.⁹

Tetrabenazine was developed in the mid-1950s to avoid orthostasis and sedation associated with reserpine.¹⁰ Both reserpine and tetrabenazine proved effective for TD,¹¹ but tetrabenazine lacked reserpine’s peripheral adverse effects. However, the kinetics of tetrabenazine necessitated multiple daily doses, and CYP2D6 genotyping was required for doses >50 mg/d.²

Receptor blocking. The mechanism that distinguishes the clinical profiles of reserpine and tetrabenazine relates to their differential properties at VMAT.¹² VMAT exists in 2 forms (VMAT1 and VMAT2) that vary in distribution, with VMAT1 expressed mainly in the peripheral nervous system and VMAT2 expressed mainly in monoaminergic cells of the CNS.¹³ Tetrabenazine is a specific and reversible VMAT2 inhibitor, whereas reserpine is an irreversible and nonselective antagonist of VMAT1 and VMAT2. It is reserpine’s VMAT1 inhibition that results in peripheral adverse effects such as orthostasis. Tetrabenazine is rapidly and extensively converted into 2 isomers, alpha-dihydrotetrabenazine (α-DHTBZ) and beta-dihydrotetrabenazine (β-DHTBZ), both of which are metabolized by CYP2D6, with a role for CYP3A4 in α-DHTBZ metabolism.¹ These DHTBZ metabolites have a short half-life when generated from oral tetrabenazine, a feature that necessitates multiple daily dosing; moreover, the existence of 2D6 polymorphisms led to FDA-mandated CYP2D6 genotyping for tetrabenazine doses >50 mg/d when it was approved for Huntington’s chorea. The concern is that 2D6 poor metabolizers will have excessive exposure to the VMAT2 effects of DHTBZ, resulting in sedation, akathisia, parkinsonism, and mood symptoms.²

How deuterium impacts medication kinetics. Deuterium is a naturally occurring, stable, nontoxic isotope of hydrogen. Humans have 5 g of deuterium in their body at any time, mostly in the form of heavy water (D₂O).¹⁴ When deuterium is used to replace selected hydrogen atoms, the resulting molecule will have similar configuration and receptor-binding properties but markedly different kinetics. Because the carbon–deuterium covalent bond requires 8 times more energy to break than a carbon–hydrogen bond, the half-life is prolonged.¹⁵ Utilizing this knowledge, a deuterated form of tetrabenazine, deutetrabenazine, was synthesized with such a purpose in mind. While the active metabolites of deutetrabenazine retain the VMAT2 affinity of non-deuterated tetrabenazine, the substitution of deuterium for hydrogen at specific positions slows the breakdown of metabolites, resulting in sustained duration of action, greater active drug exposure, and less impact of 2D6 genotype on drug exposure, thus eliminating the need for genotyping, unless one wants to exceed 36 mg/d.

Deutetrabenazine was first studied in Huntington’s chorea in a 13-week, double-blind, placebo-controlled, parallel-group study (N = 90).⁴ The maximum daily deutetrabenazine dose was 48 mg, but reduced to 36 mg in those taking strong CYP2D6 inhibitors (bupropion, fluoxetine, or paroxetine). Blinded 2D6 genotyping was performed, but there was no dose modification required based on 2D6 genotype. There was a 36.4% reduction in total maximal chorea score for deutetrabenazine compared with 14.4% for placebo (P < .001).⁴ Importantly, adverse effects were comparable between both groups, with 1 drop-out in the deutetrabenazine arm vs 2 in the placebo arm. The only adverse event occurring in ≥5% of deutetrabenazine participants and at a rate ≥2 times that of placebo was somnolence: 11.1% for deutetrabenazine vs 4.4% for placebo.⁴ The mean deutetrabenazine daily dose at the end of the treatment period was 39.7 ± 9.3 mg, and for those with impaired CYP2D6 function (poor metabolizers or those taking strong CYP2D6 inhibiting medications), the mean daily dose was 34.8 mg ± 3.8 mg.⁴

Use in tardive dyskinesia. The recommended starting dosage for TD treatment is 6 mg, twice daily with food. The dose may be increased at weekly intervals in increments of 6 mg/d to a maximum recommended daily dosage of 48 mg.⁵ The maximum daily dose is 36 mg (18 mg, twice daily) in patients receiving strong CYP2D6 inhibitors or who are 2D6 poor metabolizers.⁵

Deutetrabenazine has not been studied in those with moderate or severe hepatic impairment, and its use is contraindicated in these patients.⁵ No clinical studies have been conducted to assess the effect of renal impairment on the pharmacokinetics of deutetrabenazine.⁵

Pharmacologic profile, adverse reactions

When the data from the two 12-week, phase 3 placebo-controlled studies were pooled, the most common adverse reactions occurring in >3% of deutetrabenazine patients and greater than placebo were nasopharyngeal symptoms (4% vs 2% placebo) and insomnia (4% vs 1% placebo).⁵ Importantly, in neither TD study were there clinically significant changes in rating scales for depression, suicidal ideation and behavior, or parkinsonism. There also were no clinically significant changes in measures of schizophrenia symptoms. The mean QT prolongation for a single 24 mg dose of deutetrabenazine in healthy volunteers was 4.5 milliseconds, with the upper bound of the double-sided 90% confidence interval reaching 6.5 milliseconds.⁵ For tetrabenazine, single 50 mg doses administered to volunteers resulted in mean QT prolongation of 8 milliseconds.⁵ In patients requiring deutetrabenazine doses >24 mg/d who are taking other medications known to prolong QTc, assess the QTc interval before and after increasing the dose of deutetrabenazine or other medications that are known to prolong QTc.⁵

How it works

Tetrabenazine is the only agent that has demonstrated significant efficacy for TD management, but its use involves slow titration, multiple daily dosing, CYP2D6 genotyping for doses >50 mg/d, and tolerability issues. For example, the most common adverse effects in the pivotal tetrabenazine Huntington’s disease trial were sedation/somnolence (tetrabenazine 31% vs 3% for placebo), insomnia (tetrabenazine 22% vs 0% for placebo), depression (tetrabenazine 19% vs 0% for placebo), fatigue (tetrabenazine 22% vs 13% for placebo), and akathisia (tetrabenazine 19% vs 0% for placebo).² For comparison, the only adverse event occurring in ≥5% of deutetrabenazine participants and at a rate ≥2 times that of placebo in the pivotal Huntington’s disease trial was somnolence (11.1% for deutetrabenazine vs 4.4% for placebo).⁴

Pharmacokinetics

Deutetrabenazine has 80% oral bioavailability, and is rapidly converted to its active metabolites after oral dosing (Table 2).⁵ Linear dose dependence of Cmax and area under the curve (AUC) was observed for the active metabolites following single or multiple doses of deutetrabenazine (6 to 24 mg and 7.5 to 22.5 mg, twice daily).¹⁵ Cmax of deuterated α-DHTBZ and β-DHTBZ is reached within 3 to 4 hours after dosing, with a steady state ratio of 3:1 for the α-DHTBZ vs the β-DHTBZ form. Food had no effect on AUC, but did increase Cmax by 50%.⁵

Deutetrabenazine is metabolized through carbonyl reductase enzymes to its active metabolites, and these are further metabolized through multiple CYP pathways, predominantly 2D6 and to a lesser extent 3A4. The effect of CYP2D6 inhibition on the pharma­cokinetics of deutetrabenazine and its α-DHTBZ and β-DHTBZ metabolites was studied in 24 healthy participants following a single 22.5 mg dose of deutetrabenazine given after 8 days of administration of the strong CYP2D6 inhibitor paroxetine, 20 mg/d. In the presence of paroxetine, systemic exposure (AUC) of α-DHTBZ was 1.9-fold higher and β-DHTBZ was 6.5-fold higher, resulting in an approximately 3-fold increase in AUC for total (α+β)-DHTBZ, with corresponding increases in mean half-life of approximately 1.5-fold and 2.7-fold, respectively.⁵ Neither deutetrabenazine or its metabolites are inhibitors or inducers of major CYP enzymes. Aside from VMAT2, the results of in vitro studies suggest that deutetrabenazine and its active metabolites are unlikely to inhibit most major drug transporters at clinically relevant concentrations.

Efficacy

Efficacy was established in two 12-week, double-blind, placebo-controlled trials of adult patients with TD (ages 18 to 80).^6,7 Eligible participants had:

• TD diagnosis for ≥3 months before screening and a history of DRBA treatment for ≥3 months (≥1 month if age ≥60)
• Total AIMS motor score ≥6 (items 1 to 7) at both screening and baseline, verified by a blinded central rater at screening via central video rating
• Patients with an underlying psychiatric illness had to be stable. Psychoactive medication use, including antipsychotics, was allowed if stable for ≥30 days before screening (antidepressants, ≥45 days).

Exclusion criteria included treatment with tetrabenazine, reserpine, α-methyl-p-tyrosine, strong anticholinergic medications, dopamine antagonizing antiemetics (eg, metoclopramide, prochlorperazine, promethazine), dopamine agonists, levodopa, stimulants, or a monoamine oxidase inhibitor (MAOI) within 30 days of screening or baseline, or treatment with botulinum toxin within 3 months of screening; and presence of a neurologic condition that could confound TD assessments, serious untreated or undertreated psychiatric illness, or unstable medical illness. Patients with a history of or active suicidal ideation or behavior within 6 months of screening or score ≥11 on the depression subscale of the Hospital Anxiety and Depression Scale were excluded. Those participants with Fridericia-corrected QT interval values >450 milliseconds in men, >460 milliseconds in women, or >480 milli­seconds in patients with a right bundle branch block on electrocardiography at screening also were excluded.

The flexible-dose TD study was performed in 117 participants randomized in a 1:1 manner to deutetrabenazine or placebo, both administered twice daily, titrated to optimal dosage (12 to 48 mg/d) over 6 weeks, and then administered at that dose for another 6 weeks.⁷ The population demographics were: mean age, 54.6 ± 10.3 years, 52.1% female, 69.2% white, and 80.3% receiving ongoing dopamine antagonists, with a mean TD duration of 74.7 ± 81.5 months. Sixty-eight percent had schizophrenia spectrum disorders, and 30% had mood disorders. The primary outcome was change in total AIMS score (items 1 to 7) assessed by central, independent raters. The mean baseline AIMS score for items 1 to 7 was 9.6 ± 3.9, with 82.9% of participants with baseline AIMS scores ≥6. Study treatment retention was high: placebo 88.1%, deutetrabenazine 89.7%.⁷ There was a mean 3 point decrease in AIMS score for deutetrabenazine compared with 1.4 for placebo (P = .019). Among those with baseline AIMS scores ≥6, there was a 3.4 point decrease in AIMS scores for deutetrabenazine compared with a 1.9 point decrease for placebo (P = .027). The only adverse effects that occurred in ≥5% of deutetrabenazine participants and at a rate ≥2 times the rate in placebo were insomnia (deutetra­benazine 6.9% vs placebo 1.7%) and akathisia (deutetrabenazine 5.2% vs placebo 0%).

The fixed-dose TD study was performed in 293 participants randomized in 1:1:1:1 manner to 1 of 3 fixed doses of deutetrabenazine (12 mg/d, 24 mg/d, or 36 mg/d) or placebo, both administered twice daily.⁶ The starting dose of deutetrabenazine was 6 mg twice daily. During the dose escalation period (through Week 4), the dose of study drug was increased weekly in increments of 6 mg/d until the randomized dose was achieved. Patients continued to receive the dose they were assigned to over a maintenance period of 8 weeks.⁶ The population demographics were: mean age, 56.4 ± 11.3 years, 55% female, 79% white, 76% receiving ongoing dopamine antagonists, with a mean TD duration of 67.2 ± 66 months. Sixty percent had schizophrenia spectrum disorders, and 36% had mood disorders. The primary outcome was change in AIMS total score (items 1 to 7) assessed by central, independent raters. The mean AIMS score at baseline was 9.5 ± 2.7 in the placebo group, and for deutetrabenazine: 9.6 ± 2.4 in the 12 mg/d group, 9.4 ± 2.9 in the 24 mg/d group, and 10.1 ± 3.2 in the 36 mg/d group. The 24 mg/d and 36 mg/d doses significantly reduced AIMS scores from baseline vs placebo: 36 mg: −3.3 (0.42) vs −1.4 (0.41) (P = .001); 24 mg: −3.2 (0.45) vs −1.4 (0.41) (P = .003). Study treatment retention rates were high: placebo 90.5%, deutetrabenazine 88%. Across all doses, only 1 adverse effect occurred in ≥5% of deutetrabenazine participants: headache (5% deutetrabenazine vs 6% placebo). At the highest dose, 36 mg/d, the only adverse effects that occurred in ≥5% of participants were diarrhea (7% deutetrabenazine vs 3% placebo) and headache (7% deutetrabenazine vs 6% placebo).

Outcome. In the flexible-dose study (mean dose 38.8 ± 7.92 mg/d), the deutetrabenazine arm experienced a mean 30% reduction in AIMS scores from baseline at the Week 12 endpoint. Compared with placebo, the mean reduction in AIMS scores (standard error) was: −3.0 (0.45) deutetrabenazine vs −1.6 (0.46) placebo (P = .019).⁷ For the fixed-dose study, the 24 mg/d and 36 mg/d doses significantly reduced AIMS scores from baseline vs placebo: 36 mg: −3.3 (0.42) vs −1.4 (0.41) (P = .001); 24 mg: −3.2 (0.45) vs −1.4 (0.41) (P = .003). In addition to these mean changes from baseline, 35% of the 24 mg/d group and 33% of the 36 mg/d group demonstrated ≥50% reduction in AIMS scores.⁶

Tolerability

In the 2 phase 3 trials, there were no adverse effects occurring with an incidence ≥5% and at least twice the rate of placebo.⁵ Discontinuations because of adverse events were low in both pivotal studies across all treatment groups: 3.4% for placebo vs 1.7% for deutetrabenazine in the flexible-dose trial,⁷ and 3% for placebo vs 4% for deutetrabenazine in the fixed-dose study.⁶ In neither trial were there clinically significant changes in ratings of depression, suicidality, parkinsonism, or schizophrenia symptoms. The mean QT prolongation in healthy volunteers is described above.

Clinical considerations

Unique properties. Deutetrabenazine utilizes the greater bond strength of the carbon–deuterium bond to slow CYP metabolism, resulting in prolonged duration of action that is well tolerated, and provides significant efficacy.

Why Rx? The reasons to prescribe deutetrabenazine for TD patients include:

• only 1 of 2 agents with FDA approval for TD
• fewer tolerability issues than with tetrabenazine
• lower sedation rates in TD trials than with valbenazine
• no signal for effects on mood para­meters or rates of parkinsonism when used for TD.

Dosing

The recommended starting dosage of deutetrabenazine is 6 mg twice daily taken with food, increasing by 6 mg/d weekly as needed, with a maximum dose of 48 mg/d or 36 mg/d in those taking strong CYP2D6 inhibitors or who are 2D6 poor metabolizers. Deutetrabenazine is contraindicated in patients with hepatic impairment (as determined by Child-Pugh criteria¹⁶). There are no data in patients with renal impairment. The combined efficacy and tolerability of dosages >48 mg/d has not been evaluated. Overdoses of tetrabenazine ranging from 100 to 1,000 mg have been reported in the literature and were associated with acute dystonia, oculogyric crisis, nausea and vomiting, sweating, sedation, hypotension, confusion, diarrhea, hallucinations, rubor, and tremor.⁵

Contraindications

When used for TD, deutetrabenazine is contraindicated for patients taking reserpine, tetrabenazine, valbenazine, or MAOIs, and for patients with hepatic impairment. As with most medications, there are no data on deutetrabenazine use in pregnant women; however, oral administration of deutetrabenazine (5, 10, or 30 mg/kg/d) or tetrabenazine (30 mg/kg/d) to pregnant rats during organogenesis had no clear effect on embryofetal development. The highest dose tested was 6 times the maximum recommended human dose of 48 mg/d on a body surface area (mg/m²) basis. There are no data on the presence of deutetrabenazine or its metabolites in human milk, the effects on the breastfed infant, or the effects of the drug on milk production.

Despite improvements in the tolerability of antipsychotic medications, the development of tardive dyskinesia (TD) still is a significant area of concern; however, clinicians have had few treatment options. Valbenazine, a vesicular monoamine transport type 2 (VMAT2) inhibitor, is the only FDA-approved medication for TD (Table 1).¹ By modulating dopamine transport into presynaptic vesicles, synaptic dopamine release is decreased, thereby reducing the postsynaptic stimulation of D2 receptors and the severity of dyskinetic movements.

Clinical implications

Tardive dyskinesia remains a significant public health concern because of the increasing use of antipsychotics for disorders beyond the core indication for schizophrenia. Although exposure to dopamine D2 antagonism could result in postsynaptic receptor upregulation and supersensitivity, this process best explains what underlies withdrawal dyskinesia.³ The persistence of TD symptoms in 66% to 80% of patients after discontinuing offending agents has led to hypotheses that the underlying pathophysiology of TD might best be conceptualized as a problem with neuroplasticity. As with many disorders, environmental contributions (eg, oxidative stress) and genetic predisposition might play a role beyond that related to exposure to D2 antagonism.³

There have been trials of numerous agents, but no medication has been FDA-approved for treating TD, and limited data support the efficacy of a few existing medications (clonazepam, amantadine, and ginkgo biloba extract [EGb-761]),⁴ albeit with small effect sizes. A medical food, consisting of branched-chain amino acids, received FDA approval for the dietary management of TD in males, but is no longer commercially available except from compounding pharmacies.⁵

Tetrabenazine, a molecule developed in the mid-1950s to improve on the tolerability of reserpine, was associated with significant adverse effects such as orthostasis.⁶ Like reserpine, tetrabenazine subsequently was found to be effective for TD⁷ but without the peripheral adverse effects of reserpine. However, the kinetics of tetrabenazine necessitated multiple daily doses, and required CYP2D6 genotyping for doses >50 mg/d.⁸

Receptor blocking. The mechanism that differentiated reserpine’s and tetrabenazine’s clinical properties became clearer in the 1980s when researchers discovered that transporters were necessary to package neurotransmitters into the synaptic vesicles of presynaptic neurons.⁹ The vesicular monoamine transporter (VMAT) exists in 2 isoforms (VMAT1 and VMAT2) that vary in distribution, with VMAT1 expressed mainly in the peripheral nervous system and VMAT2 expressed mainly in monoaminergic cells of the central nervous system.¹⁰

Tetrabenazine’s improved tolerability profile was related to the fact that it is a specific and reversible VMAT2 inhibitor, while reserpine is an irreversible and nonselective antagonist of both VMAT isoforms. Investigation of tetrabenazine’s metabolism revealed that it is rapidly and extensively converted into 2 isomers, α-dihydrotetrabenazine (DH-TBZ) and β-DH-TBZ. The isomeric forms of DH-TBZ have multiple chiral centers, and therefore numerous forms of which only 2 are significantly active at VMAT2.³ The α–DH-TBZ isomer is metabolized via CYP2D6 and 3A4 into inactive metabolites, while β-DH-TBZ is metabolized solely via 2D6.³ Because of the short half-life of DH-TBZ when generated from oral tetrabenazine, the existence of 2D6 polymorphisms, and the predominant activity deriving from only 2 isomers, a molecule was synthesized (valbenazine), that when metabolized would slowly be converted into the most active isomer of α–DH-TBZ designated as NBI-98782 (Table 2). This slower conversion to NBI-98782 from valbenazine (compared with its formation from oral tetrabenazine) yielded improved kinetics and permitted once-daily dosing; moreover, because the metabolism of NBI-98782 is not solely dependent on CYP2D6, the need for genotyping was removed. Neither of the 2 metabolites from valbenazine NBI-98782 and NB-136110 have significant affinity for targets other than VMAT2.¹¹

Use in tardive dyskinesia. Recommended starting dosage is 40 mg once daily with or without food, increased to 80 mg after 1 week, based on the design and results from the phase-III clinical trial.¹² The FDA granted breakthrough therapy designation for this compound, and only 1 phase-III trial was performed. Valbenazine produced significant improvement on the AIMS, with a mean 30% reduction in AIMS scores at the Week 6 endpoint from baseline of 10.4 ± 3.6.² The effect size was large (Cohen’s d = 0.90) for the 80-mg dosage. Continuation of 40 mg/d may be considered for some patients based on tolerability, including those who are known CYP2D6 poor metabolizers, and those taking strong CYP2D6 inhibitors. Patients taking strong 3A4 inhibitors should not exceed 40 mg/d. The maximum daily dose is 40 mg for those who have moderate or severe hepatic impairment (Child-Pugh score, 7 to 15). Dosage adjustment is not required for mild to moderate renal impairment (creatinine clearance, 30 to 90 mL/min).

Pharmacologic profile, adverse reactions

Valbenazine and its 2 metabolites lack affinity for receptors other than VMAT2, leading to an absence of orthostasis in clinical trials.^1,2 In the phase-II trial, 76% of participants receiving valbenazine (n = 51) were titrated to the maximum dosage of 75 mg/d. Common adverse reactions (incidence ≥5% and at least twice the rate of placebo) were headache (9.8% vs 4.1% placebo), fatigue (9.8% vs 4.1% placebo), and somnolence (5.9% vs 2% placebo).¹ In the phase-III trial, participants were randomized 1:1:1 to valbenazine, 40 mg (n = 72), valbenazine, 80 mg (n = 79), or placebo (n = 76). In the clinical studies the most common diagnosis was schizophrenia or schizoaffective disorder, and 40% and 85% of participants in the phase-II and phase-III studies, respectively, remained on antipsychotics.^1,2 There were no adverse effects with an incidence ≥5% and at least twice the rate of placebo in the phase-III trial.²

When data from all placebo-controlled studies were pooled, only 1 adverse effect occurred with an incidence ≥5% and twice that of placebo, somnolence with a rate of 10.9% for valbenazine vs 4.2% for placebo. The incidence of akathisia in the pooled analysis was 2.7% for valbenazine vs 0.5% for placebo. Importantly, in neither study was there a safety signal related to depression, suicidal ideation and behavior, or parkinsonism. There also were no clinically significant changes in measures of schizophrenia symptoms.

The mean QT prolongation for valbenazine in healthy participants was 6.7 milliseconds, with the upper bound of the double-sided 90% confidence interval reaching 8.4 milliseconds. For those taking strong 2D6 or 3A4 inhibitors, or known 2D6 poor metabolizers, the mean QT prolongation was 11.7 milliseconds (14.7 milliseconds upper bound of double-sided 90% CI). In the controlled trials, there was a dose-related increase in prolactin, alkaline phosphatase, and bilirubin. Overall, 3% of valbenazine-treated patients and 2% of placebo-treated patients discontinued because of adverse reactions.

As noted above, there were no adverse effects with an incidence ≥5% and at least twice the rate of placebo in the phase-III valbenazine trial. Aggregate data across all placebo-controlled studies found that somnolence was the only adverse effect that occurred with an incidence ≥5% and twice that of placebo (10.9% for valbenazine vs 4.2% for placebo).² As a comparsion, rates of sedation and akathisia for tetrabenazine were higher in the pivotal Huntington’s disease trial: sedation/somnolence 31% vs 3% for placebo, and akathisia 19% vs 0% for placebo.⁸

How it works

Tetrabenazine, a selective VMAT2 inhibitor, is the only agent that has demonstrated significant efficacy and tolerability for TD management; however, its complex metabolism generates numerous isomers of the metabolites α-DH-TBZ and β-DH-TBZ, of which only 2 are significantly active (Table 3). By choosing an active isomer (NBI-98782) as the metabolite of interest because of its selective and potent activity at VMAT2 and having a metabolism not solely dependent on CYP2D6, a compound was generated (valbenazine) that when metabolized slowly converts into NBI-98782.

Pharmacokinetics

Valbenazine demonstrates dose-proportional pharmacokinetics after single oral dosages from 40 to 300 mg with no impact of food or fasting status on levels of the active metabolite. Valbenazine has a Tmax of 0.5 to 1.0 hours, with 49% oral bioavailability. The plasma half-life for valbenazine and for NBI-98782 ranges from 15 to 22 hours. The Tmax for NBI-98782 when formed from valbenazine occurs between 4 and 8 hours, with a Cmax of approximately 30 ng/mL. It should be noted that when NBI-98782 is generated from oral tetrabenazine, the mean half-life and Tmax are considerably shorter (6 hours and 1.5 hours, respectively), while the Cmax is much higher (approximately 77 ng/mL) (Table 4).

Valbenazine is metabolized through endogenous esterases to NBI-98782 and NBI-136110. NBI-98782, the active metabolite, is further metabolized through multiple CYP pathways, predominantly 3A4 and 2D6. Neither valbenazine nor its metabolites are inhibitors or inducers of major CYP enzymes. Aside from VMAT2, the results of in vitro studies suggest that valbenazine and its active metabolite are unlikely to inhibit most major drug transporters at clinically relevant concentrations. However, valbenazine increased digoxin levels because of inhibition of intestinal P-glycoprotein; therefore plasma digoxin level monitoring is recommended when these 2 are co-administered.

Efficacy

Efficacy was established in a 6-week, fixed-dosage, double-blind, placebo-controlled trial of adult patients with TD. Eligible participants had:

• DSM-IV diagnosis of antipsychotic-induced TD for ≥3 months before screening and moderate or severe TD, as indicated by AIMS item 8 (severity of abnormal movement), which was rated by a blinded, external reviewer using a video of the participant’s AIMS assessment at screening
• a DSM-IV diagnosis of schizophrenia or schizoaffective disorder or mood disorder (and stable per investigator)
• Brief Psychiatric Rating Scale score <50 at screening.

Exclusion criteria included clinically significant and unstable medical conditions within 1 month before screening; comorbid movement disorder (eg, parkinsonism, akathisia, truncal dystonia) that was more prominent than TD; and significant risk for active suicidal ideation, suicidal behavior, or violent behavior.² Participants had a mean age of 56, 52% were male, and 65.7% of participants in the valbenazine 40-mg group had a schizophrenia spectrum disorder diagnosis, as did 65.8% in both the placebo and valbenazine 80-mg arms.

Antipsychotic treatments were permitted during the trial and >85% of participants continued taking these medications during the study. Participants (N = 234) were randomly allocated in a 1:1:1 manner to valbenazine 40 mg, 80 mg, or matched placebo. The primary outcome was change in AIMS total score (items 1 to 7) assessed by central, independent raters. Baseline AIMS scores were 9.9 ± 4.3 in the placebo group, and 9.8 ± 4.1 and 10.4 ± 3.6 in the valbenazine 40-mg and 80-mg arms, respectively.²

Outcome. A fixed-sequence testing procedure to control for family-wise error rate and multiplicity was employed, and the primary endpoint was change from baseline to Week 6 in AIMS total score (items 1 to 7) for valbenazine 80 mg vs placebo. Valbenazine, 40 mg, was associated with a 1.9 point decrease in AIMS score, while valbenazine, 80 mg, was associated with a 3.2 point decrease in AIMS score, compared with 0.1 point decrease for placebo (P < .05 for valbenazine, 40 mg, P < .001 for valbenazine, 80 mg). This difference for the 40-mg dosage did not meet the prespecified analysis endpoints; however, for the 80-mg valbenazine dosage, the effect size for this difference (Cohen’s d) was large 0.90. There also were statistically significant differences between 40 mg and 80 mg at weeks 2, 4, and 6 in the intent-to-treat population. Of the 79 participants, 43 taking the 80-mg dosage completed a 48-week extension. Efficacy was sustained in this group; however, when valbenazine was discontinued at Week 48, AIMS scores returned to baseline after 4 weeks.

Tolerability

Of the 234 randomized patients, 205 (87.6%) completed the 6-week trial. Discontinuations due to adverse events were low across all treatment groups: 2.6% and 2.8% in the placebo and valbenazine 40-mg arms, respectively, and 3.8% in valbenazine 80-mg cohort. There was no safety signal based on changes in depression, suicidality, parkinsonism rating, or changes in schizophrenia symptoms. Because valbenazine can cause somnolence, patients should not perform activities requiring mental alertness (eg, operating a vehicle or hazardous machinery) until they know how they will be affected by valbenazine.

Valbenazine should be avoided in patients with congenital long QT syndrome or with arrhythmias associated with a prolonged QT interval. For patients at increased risk of a prolonged QT interval, assess the QT interval before increasing the dosage.

Clinical considerations

Unique properties. Valbenazine is metabolized slowly to a potent, selective VMAT2 antagonist (NBI-98782) in a manner that permits once daily dosing, removes the need for CYP2D6 genotyping, and provides significant efficacy.


Why Rx? The reasons to prescribe valbenazine for TD patients include:

• currently the only agent with FDA approval for TD
• fewer tolerability issues seen with the only other effective agent, tetrabenazine
• no signal for effects on mood parameters or rates of parkinsonism
• lack of multiple daily dosing and possible need for 2D6 genotyping involved with TBZ prescribing.

Dosing

The recommended dosage of valbenazine is 80 mg/d administered as a single dose with or without food, starting at 40 mg once daily for 1 week. There is no dosage adjustment required in those with mild to moderate renal impairment; however, valbenazine is not recommended in those with severe renal impairment. The maximum dose is 40 mg/d for those who with moderate or severe hepatic impairment (Child-Pugh score, 7 to 15) however, valbenazine is not recommended for patients with severe renal impairment (creatinine clearance <30 mL/min) because the exposure to the active metabolite is reduced by approximately 75%. The combined efficacy and tolerability of dosages >80 mg/d has not been evaluated. Adverse effects seen with tetrabenazine at higher dosages include akathisia, anxiety, insomnia, parkinsonism, fatigue, and depression.

A daily dose of 40 mg may be considered for some patients based on tolerability, including those who are known CYP 2D6 poor metabolizers, and those taking strong CYP2D6 inhibitors.² For those taking strong 3A4 inhibitors, the maximum daily dose is 40 mg. Concomitant use of valbenazine with strong 3A4 inducers is not recommended as the exposure to the active metabolite is reduced by approximately 75%.² Lastly, because VMAT2 inhibition may alter synaptic levels of other monoamines, it is recommended that valbenazine not be administered with monoamine oxidase inhibitors, such as isocarboxazid, phenelzine, or selegiline.

Contraindications

There are no reported contraindications for valbenazine. As with most medications, there is limited available data on valbenazine use in pregnant women; however, administration of valbenazine to pregnant rats during organogenesis through lactation produced an increase in the number of stillborn pups and postnatal pup mortalities at doses under the maximum recommended human dose (MRHD) using body surface area based dosing (mg/m²). Pregnant women should be advised of the potential risk to a fetus. Valbenazine and its metabolites have been detected in rat milk at concentrations higher than in plasma after oral administration of valbenazine at doses 0.1 to 1.2 times the MRHD (based on mg/m²). Based on animal findings of increased perinatal mortality in exposed fetuses and pups, woman are advised not to breastfeed during valbenazine treatment and for 5 days after the final dose. No dosage adjustment is required for geriatric patients.

Despite improvements in the tolerability of antipsychotic medications, the development of tardive dyskinesia (TD) still is a significant area of concern; however, clinicians have had few treatment options. Valbenazine, a vesicular monoamine transport type 2 (VMAT2) inhibitor, is the only FDA-approved medication for TD (Table 1).¹ By modulating dopamine transport into presynaptic vesicles, synaptic dopamine release is decreased, thereby reducing the postsynaptic stimulation of D2 receptors and the severity of dyskinetic movements.

Clinical implications

Tardive dyskinesia remains a significant public health concern because of the increasing use of antipsychotics for disorders beyond the core indication for schizophrenia. Although exposure to dopamine D2 antagonism could result in postsynaptic receptor upregulation and supersensitivity, this process best explains what underlies withdrawal dyskinesia.³ The persistence of TD symptoms in 66% to 80% of patients after discontinuing offending agents has led to hypotheses that the underlying pathophysiology of TD might best be conceptualized as a problem with neuroplasticity. As with many disorders, environmental contributions (eg, oxidative stress) and genetic predisposition might play a role beyond that related to exposure to D2 antagonism.³

There have been trials of numerous agents, but no medication has been FDA-approved for treating TD, and limited data support the efficacy of a few existing medications (clonazepam, amantadine, and ginkgo biloba extract [EGb-761]),⁴ albeit with small effect sizes. A medical food, consisting of branched-chain amino acids, received FDA approval for the dietary management of TD in males, but is no longer commercially available except from compounding pharmacies.⁵

Tetrabenazine, a molecule developed in the mid-1950s to improve on the tolerability of reserpine, was associated with significant adverse effects such as orthostasis.⁶ Like reserpine, tetrabenazine subsequently was found to be effective for TD⁷ but without the peripheral adverse effects of reserpine. However, the kinetics of tetrabenazine necessitated multiple daily doses, and required CYP2D6 genotyping for doses >50 mg/d.⁸

Receptor blocking. The mechanism that differentiated reserpine’s and tetrabenazine’s clinical properties became clearer in the 1980s when researchers discovered that transporters were necessary to package neurotransmitters into the synaptic vesicles of presynaptic neurons.⁹ The vesicular monoamine transporter (VMAT) exists in 2 isoforms (VMAT1 and VMAT2) that vary in distribution, with VMAT1 expressed mainly in the peripheral nervous system and VMAT2 expressed mainly in monoaminergic cells of the central nervous system.¹⁰

Tetrabenazine’s improved tolerability profile was related to the fact that it is a specific and reversible VMAT2 inhibitor, while reserpine is an irreversible and nonselective antagonist of both VMAT isoforms. Investigation of tetrabenazine’s metabolism revealed that it is rapidly and extensively converted into 2 isomers, α-dihydrotetrabenazine (DH-TBZ) and β-DH-TBZ. The isomeric forms of DH-TBZ have multiple chiral centers, and therefore numerous forms of which only 2 are significantly active at VMAT2.³ The α–DH-TBZ isomer is metabolized via CYP2D6 and 3A4 into inactive metabolites, while β-DH-TBZ is metabolized solely via 2D6.³ Because of the short half-life of DH-TBZ when generated from oral tetrabenazine, the existence of 2D6 polymorphisms, and the predominant activity deriving from only 2 isomers, a molecule was synthesized (valbenazine), that when metabolized would slowly be converted into the most active isomer of α–DH-TBZ designated as NBI-98782 (Table 2). This slower conversion to NBI-98782 from valbenazine (compared with its formation from oral tetrabenazine) yielded improved kinetics and permitted once-daily dosing; moreover, because the metabolism of NBI-98782 is not solely dependent on CYP2D6, the need for genotyping was removed. Neither of the 2 metabolites from valbenazine NBI-98782 and NB-136110 have significant affinity for targets other than VMAT2.¹¹

Use in tardive dyskinesia. Recommended starting dosage is 40 mg once daily with or without food, increased to 80 mg after 1 week, based on the design and results from the phase-III clinical trial.¹² The FDA granted breakthrough therapy designation for this compound, and only 1 phase-III trial was performed. Valbenazine produced significant improvement on the AIMS, with a mean 30% reduction in AIMS scores at the Week 6 endpoint from baseline of 10.4 ± 3.6.² The effect size was large (Cohen’s d = 0.90) for the 80-mg dosage. Continuation of 40 mg/d may be considered for some patients based on tolerability, including those who are known CYP2D6 poor metabolizers, and those taking strong CYP2D6 inhibitors. Patients taking strong 3A4 inhibitors should not exceed 40 mg/d. The maximum daily dose is 40 mg for those who have moderate or severe hepatic impairment (Child-Pugh score, 7 to 15). Dosage adjustment is not required for mild to moderate renal impairment (creatinine clearance, 30 to 90 mL/min).

Pharmacologic profile, adverse reactions

Valbenazine and its 2 metabolites lack affinity for receptors other than VMAT2, leading to an absence of orthostasis in clinical trials.^1,2 In the phase-II trial, 76% of participants receiving valbenazine (n = 51) were titrated to the maximum dosage of 75 mg/d. Common adverse reactions (incidence ≥5% and at least twice the rate of placebo) were headache (9.8% vs 4.1% placebo), fatigue (9.8% vs 4.1% placebo), and somnolence (5.9% vs 2% placebo).¹ In the phase-III trial, participants were randomized 1:1:1 to valbenazine, 40 mg (n = 72), valbenazine, 80 mg (n = 79), or placebo (n = 76). In the clinical studies the most common diagnosis was schizophrenia or schizoaffective disorder, and 40% and 85% of participants in the phase-II and phase-III studies, respectively, remained on antipsychotics.^1,2 There were no adverse effects with an incidence ≥5% and at least twice the rate of placebo in the phase-III trial.²

When data from all placebo-controlled studies were pooled, only 1 adverse effect occurred with an incidence ≥5% and twice that of placebo, somnolence with a rate of 10.9% for valbenazine vs 4.2% for placebo. The incidence of akathisia in the pooled analysis was 2.7% for valbenazine vs 0.5% for placebo. Importantly, in neither study was there a safety signal related to depression, suicidal ideation and behavior, or parkinsonism. There also were no clinically significant changes in measures of schizophrenia symptoms.

The mean QT prolongation for valbenazine in healthy participants was 6.7 milliseconds, with the upper bound of the double-sided 90% confidence interval reaching 8.4 milliseconds. For those taking strong 2D6 or 3A4 inhibitors, or known 2D6 poor metabolizers, the mean QT prolongation was 11.7 milliseconds (14.7 milliseconds upper bound of double-sided 90% CI). In the controlled trials, there was a dose-related increase in prolactin, alkaline phosphatase, and bilirubin. Overall, 3% of valbenazine-treated patients and 2% of placebo-treated patients discontinued because of adverse reactions.

As noted above, there were no adverse effects with an incidence ≥5% and at least twice the rate of placebo in the phase-III valbenazine trial. Aggregate data across all placebo-controlled studies found that somnolence was the only adverse effect that occurred with an incidence ≥5% and twice that of placebo (10.9% for valbenazine vs 4.2% for placebo).² As a comparsion, rates of sedation and akathisia for tetrabenazine were higher in the pivotal Huntington’s disease trial: sedation/somnolence 31% vs 3% for placebo, and akathisia 19% vs 0% for placebo.⁸

How it works

Tetrabenazine, a selective VMAT2 inhibitor, is the only agent that has demonstrated significant efficacy and tolerability for TD management; however, its complex metabolism generates numerous isomers of the metabolites α-DH-TBZ and β-DH-TBZ, of which only 2 are significantly active (Table 3). By choosing an active isomer (NBI-98782) as the metabolite of interest because of its selective and potent activity at VMAT2 and having a metabolism not solely dependent on CYP2D6, a compound was generated (valbenazine) that when metabolized slowly converts into NBI-98782.

Pharmacokinetics

Valbenazine demonstrates dose-proportional pharmacokinetics after single oral dosages from 40 to 300 mg with no impact of food or fasting status on levels of the active metabolite. Valbenazine has a Tmax of 0.5 to 1.0 hours, with 49% oral bioavailability. The plasma half-life for valbenazine and for NBI-98782 ranges from 15 to 22 hours. The Tmax for NBI-98782 when formed from valbenazine occurs between 4 and 8 hours, with a Cmax of approximately 30 ng/mL. It should be noted that when NBI-98782 is generated from oral tetrabenazine, the mean half-life and Tmax are considerably shorter (6 hours and 1.5 hours, respectively), while the Cmax is much higher (approximately 77 ng/mL) (Table 4).

Valbenazine is metabolized through endogenous esterases to NBI-98782 and NBI-136110. NBI-98782, the active metabolite, is further metabolized through multiple CYP pathways, predominantly 3A4 and 2D6. Neither valbenazine nor its metabolites are inhibitors or inducers of major CYP enzymes. Aside from VMAT2, the results of in vitro studies suggest that valbenazine and its active metabolite are unlikely to inhibit most major drug transporters at clinically relevant concentrations. However, valbenazine increased digoxin levels because of inhibition of intestinal P-glycoprotein; therefore plasma digoxin level monitoring is recommended when these 2 are co-administered.

Efficacy

Efficacy was established in a 6-week, fixed-dosage, double-blind, placebo-controlled trial of adult patients with TD. Eligible participants had:

• DSM-IV diagnosis of antipsychotic-induced TD for ≥3 months before screening and moderate or severe TD, as indicated by AIMS item 8 (severity of abnormal movement), which was rated by a blinded, external reviewer using a video of the participant’s AIMS assessment at screening
• a DSM-IV diagnosis of schizophrenia or schizoaffective disorder or mood disorder (and stable per investigator)
• Brief Psychiatric Rating Scale score <50 at screening.

Exclusion criteria included clinically significant and unstable medical conditions within 1 month before screening; comorbid movement disorder (eg, parkinsonism, akathisia, truncal dystonia) that was more prominent than TD; and significant risk for active suicidal ideation, suicidal behavior, or violent behavior.² Participants had a mean age of 56, 52% were male, and 65.7% of participants in the valbenazine 40-mg group had a schizophrenia spectrum disorder diagnosis, as did 65.8% in both the placebo and valbenazine 80-mg arms.

Antipsychotic treatments were permitted during the trial and >85% of participants continued taking these medications during the study. Participants (N = 234) were randomly allocated in a 1:1:1 manner to valbenazine 40 mg, 80 mg, or matched placebo. The primary outcome was change in AIMS total score (items 1 to 7) assessed by central, independent raters. Baseline AIMS scores were 9.9 ± 4.3 in the placebo group, and 9.8 ± 4.1 and 10.4 ± 3.6 in the valbenazine 40-mg and 80-mg arms, respectively.²

Outcome. A fixed-sequence testing procedure to control for family-wise error rate and multiplicity was employed, and the primary endpoint was change from baseline to Week 6 in AIMS total score (items 1 to 7) for valbenazine 80 mg vs placebo. Valbenazine, 40 mg, was associated with a 1.9 point decrease in AIMS score, while valbenazine, 80 mg, was associated with a 3.2 point decrease in AIMS score, compared with 0.1 point decrease for placebo (P < .05 for valbenazine, 40 mg, P < .001 for valbenazine, 80 mg). This difference for the 40-mg dosage did not meet the prespecified analysis endpoints; however, for the 80-mg valbenazine dosage, the effect size for this difference (Cohen’s d) was large 0.90. There also were statistically significant differences between 40 mg and 80 mg at weeks 2, 4, and 6 in the intent-to-treat population. Of the 79 participants, 43 taking the 80-mg dosage completed a 48-week extension. Efficacy was sustained in this group; however, when valbenazine was discontinued at Week 48, AIMS scores returned to baseline after 4 weeks.

Tolerability

Of the 234 randomized patients, 205 (87.6%) completed the 6-week trial. Discontinuations due to adverse events were low across all treatment groups: 2.6% and 2.8% in the placebo and valbenazine 40-mg arms, respectively, and 3.8% in valbenazine 80-mg cohort. There was no safety signal based on changes in depression, suicidality, parkinsonism rating, or changes in schizophrenia symptoms. Because valbenazine can cause somnolence, patients should not perform activities requiring mental alertness (eg, operating a vehicle or hazardous machinery) until they know how they will be affected by valbenazine.

Valbenazine should be avoided in patients with congenital long QT syndrome or with arrhythmias associated with a prolonged QT interval. For patients at increased risk of a prolonged QT interval, assess the QT interval before increasing the dosage.

Clinical considerations

Unique properties. Valbenazine is metabolized slowly to a potent, selective VMAT2 antagonist (NBI-98782) in a manner that permits once daily dosing, removes the need for CYP2D6 genotyping, and provides significant efficacy.


Why Rx? The reasons to prescribe valbenazine for TD patients include:

• currently the only agent with FDA approval for TD
• fewer tolerability issues seen with the only other effective agent, tetrabenazine
• no signal for effects on mood parameters or rates of parkinsonism
• lack of multiple daily dosing and possible need for 2D6 genotyping involved with TBZ prescribing.

Dosing

The recommended dosage of valbenazine is 80 mg/d administered as a single dose with or without food, starting at 40 mg once daily for 1 week. There is no dosage adjustment required in those with mild to moderate renal impairment; however, valbenazine is not recommended in those with severe renal impairment. The maximum dose is 40 mg/d for those who with moderate or severe hepatic impairment (Child-Pugh score, 7 to 15) however, valbenazine is not recommended for patients with severe renal impairment (creatinine clearance <30 mL/min) because the exposure to the active metabolite is reduced by approximately 75%. The combined efficacy and tolerability of dosages >80 mg/d has not been evaluated. Adverse effects seen with tetrabenazine at higher dosages include akathisia, anxiety, insomnia, parkinsonism, fatigue, and depression.

A daily dose of 40 mg may be considered for some patients based on tolerability, including those who are known CYP 2D6 poor metabolizers, and those taking strong CYP2D6 inhibitors.² For those taking strong 3A4 inhibitors, the maximum daily dose is 40 mg. Concomitant use of valbenazine with strong 3A4 inducers is not recommended as the exposure to the active metabolite is reduced by approximately 75%.² Lastly, because VMAT2 inhibition may alter synaptic levels of other monoamines, it is recommended that valbenazine not be administered with monoamine oxidase inhibitors, such as isocarboxazid, phenelzine, or selegiline.

Contraindications

There are no reported contraindications for valbenazine. As with most medications, there is limited available data on valbenazine use in pregnant women; however, administration of valbenazine to pregnant rats during organogenesis through lactation produced an increase in the number of stillborn pups and postnatal pup mortalities at doses under the maximum recommended human dose (MRHD) using body surface area based dosing (mg/m²). Pregnant women should be advised of the potential risk to a fetus. Valbenazine and its metabolites have been detected in rat milk at concentrations higher than in plasma after oral administration of valbenazine at doses 0.1 to 1.2 times the MRHD (based on mg/m²). Based on animal findings of increased perinatal mortality in exposed fetuses and pups, woman are advised not to breastfeed during valbenazine treatment and for 5 days after the final dose. No dosage adjustment is required for geriatric patients.

Despite improvements in the tolerability of antipsychotic medications, the development of tardive dyskinesia (TD) still is a significant area of concern; however, clinicians have had few treatment options. Valbenazine, a vesicular monoamine transport type 2 (VMAT2) inhibitor, is the only FDA-approved medication for TD (Table 1).¹ By modulating dopamine transport into presynaptic vesicles, synaptic dopamine release is decreased, thereby reducing the postsynaptic stimulation of D2 receptors and the severity of dyskinetic movements.

Clinical implications

Tardive dyskinesia remains a significant public health concern because of the increasing use of antipsychotics for disorders beyond the core indication for schizophrenia. Although exposure to dopamine D2 antagonism could result in postsynaptic receptor upregulation and supersensitivity, this process best explains what underlies withdrawal dyskinesia.³ The persistence of TD symptoms in 66% to 80% of patients after discontinuing offending agents has led to hypotheses that the underlying pathophysiology of TD might best be conceptualized as a problem with neuroplasticity. As with many disorders, environmental contributions (eg, oxidative stress) and genetic predisposition might play a role beyond that related to exposure to D2 antagonism.³

There have been trials of numerous agents, but no medication has been FDA-approved for treating TD, and limited data support the efficacy of a few existing medications (clonazepam, amantadine, and ginkgo biloba extract [EGb-761]),⁴ albeit with small effect sizes. A medical food, consisting of branched-chain amino acids, received FDA approval for the dietary management of TD in males, but is no longer commercially available except from compounding pharmacies.⁵

Tetrabenazine, a molecule developed in the mid-1950s to improve on the tolerability of reserpine, was associated with significant adverse effects such as orthostasis.⁶ Like reserpine, tetrabenazine subsequently was found to be effective for TD⁷ but without the peripheral adverse effects of reserpine. However, the kinetics of tetrabenazine necessitated multiple daily doses, and required CYP2D6 genotyping for doses >50 mg/d.⁸

Receptor blocking. The mechanism that differentiated reserpine’s and tetrabenazine’s clinical properties became clearer in the 1980s when researchers discovered that transporters were necessary to package neurotransmitters into the synaptic vesicles of presynaptic neurons.⁹ The vesicular monoamine transporter (VMAT) exists in 2 isoforms (VMAT1 and VMAT2) that vary in distribution, with VMAT1 expressed mainly in the peripheral nervous system and VMAT2 expressed mainly in monoaminergic cells of the central nervous system.¹⁰

Tetrabenazine’s improved tolerability profile was related to the fact that it is a specific and reversible VMAT2 inhibitor, while reserpine is an irreversible and nonselective antagonist of both VMAT isoforms. Investigation of tetrabenazine’s metabolism revealed that it is rapidly and extensively converted into 2 isomers, α-dihydrotetrabenazine (DH-TBZ) and β-DH-TBZ. The isomeric forms of DH-TBZ have multiple chiral centers, and therefore numerous forms of which only 2 are significantly active at VMAT2.³ The α–DH-TBZ isomer is metabolized via CYP2D6 and 3A4 into inactive metabolites, while β-DH-TBZ is metabolized solely via 2D6.³ Because of the short half-life of DH-TBZ when generated from oral tetrabenazine, the existence of 2D6 polymorphisms, and the predominant activity deriving from only 2 isomers, a molecule was synthesized (valbenazine), that when metabolized would slowly be converted into the most active isomer of α–DH-TBZ designated as NBI-98782 (Table 2). This slower conversion to NBI-98782 from valbenazine (compared with its formation from oral tetrabenazine) yielded improved kinetics and permitted once-daily dosing; moreover, because the metabolism of NBI-98782 is not solely dependent on CYP2D6, the need for genotyping was removed. Neither of the 2 metabolites from valbenazine NBI-98782 and NB-136110 have significant affinity for targets other than VMAT2.¹¹

Use in tardive dyskinesia. Recommended starting dosage is 40 mg once daily with or without food, increased to 80 mg after 1 week, based on the design and results from the phase-III clinical trial.¹² The FDA granted breakthrough therapy designation for this compound, and only 1 phase-III trial was performed. Valbenazine produced significant improvement on the AIMS, with a mean 30% reduction in AIMS scores at the Week 6 endpoint from baseline of 10.4 ± 3.6.² The effect size was large (Cohen’s d = 0.90) for the 80-mg dosage. Continuation of 40 mg/d may be considered for some patients based on tolerability, including those who are known CYP2D6 poor metabolizers, and those taking strong CYP2D6 inhibitors. Patients taking strong 3A4 inhibitors should not exceed 40 mg/d. The maximum daily dose is 40 mg for those who have moderate or severe hepatic impairment (Child-Pugh score, 7 to 15). Dosage adjustment is not required for mild to moderate renal impairment (creatinine clearance, 30 to 90 mL/min).

Pharmacologic profile, adverse reactions

Valbenazine and its 2 metabolites lack affinity for receptors other than VMAT2, leading to an absence of orthostasis in clinical trials.^1,2 In the phase-II trial, 76% of participants receiving valbenazine (n = 51) were titrated to the maximum dosage of 75 mg/d. Common adverse reactions (incidence ≥5% and at least twice the rate of placebo) were headache (9.8% vs 4.1% placebo), fatigue (9.8% vs 4.1% placebo), and somnolence (5.9% vs 2% placebo).¹ In the phase-III trial, participants were randomized 1:1:1 to valbenazine, 40 mg (n = 72), valbenazine, 80 mg (n = 79), or placebo (n = 76). In the clinical studies the most common diagnosis was schizophrenia or schizoaffective disorder, and 40% and 85% of participants in the phase-II and phase-III studies, respectively, remained on antipsychotics.^1,2 There were no adverse effects with an incidence ≥5% and at least twice the rate of placebo in the phase-III trial.²

When data from all placebo-controlled studies were pooled, only 1 adverse effect occurred with an incidence ≥5% and twice that of placebo, somnolence with a rate of 10.9% for valbenazine vs 4.2% for placebo. The incidence of akathisia in the pooled analysis was 2.7% for valbenazine vs 0.5% for placebo. Importantly, in neither study was there a safety signal related to depression, suicidal ideation and behavior, or parkinsonism. There also were no clinically significant changes in measures of schizophrenia symptoms.

The mean QT prolongation for valbenazine in healthy participants was 6.7 milliseconds, with the upper bound of the double-sided 90% confidence interval reaching 8.4 milliseconds. For those taking strong 2D6 or 3A4 inhibitors, or known 2D6 poor metabolizers, the mean QT prolongation was 11.7 milliseconds (14.7 milliseconds upper bound of double-sided 90% CI). In the controlled trials, there was a dose-related increase in prolactin, alkaline phosphatase, and bilirubin. Overall, 3% of valbenazine-treated patients and 2% of placebo-treated patients discontinued because of adverse reactions.

As noted above, there were no adverse effects with an incidence ≥5% and at least twice the rate of placebo in the phase-III valbenazine trial. Aggregate data across all placebo-controlled studies found that somnolence was the only adverse effect that occurred with an incidence ≥5% and twice that of placebo (10.9% for valbenazine vs 4.2% for placebo).² As a comparsion, rates of sedation and akathisia for tetrabenazine were higher in the pivotal Huntington’s disease trial: sedation/somnolence 31% vs 3% for placebo, and akathisia 19% vs 0% for placebo.⁸

How it works

Tetrabenazine, a selective VMAT2 inhibitor, is the only agent that has demonstrated significant efficacy and tolerability for TD management; however, its complex metabolism generates numerous isomers of the metabolites α-DH-TBZ and β-DH-TBZ, of which only 2 are significantly active (Table 3). By choosing an active isomer (NBI-98782) as the metabolite of interest because of its selective and potent activity at VMAT2 and having a metabolism not solely dependent on CYP2D6, a compound was generated (valbenazine) that when metabolized slowly converts into NBI-98782.

Pharmacokinetics

Valbenazine demonstrates dose-proportional pharmacokinetics after single oral dosages from 40 to 300 mg with no impact of food or fasting status on levels of the active metabolite. Valbenazine has a Tmax of 0.5 to 1.0 hours, with 49% oral bioavailability. The plasma half-life for valbenazine and for NBI-98782 ranges from 15 to 22 hours. The Tmax for NBI-98782 when formed from valbenazine occurs between 4 and 8 hours, with a Cmax of approximately 30 ng/mL. It should be noted that when NBI-98782 is generated from oral tetrabenazine, the mean half-life and Tmax are considerably shorter (6 hours and 1.5 hours, respectively), while the Cmax is much higher (approximately 77 ng/mL) (Table 4).

Valbenazine is metabolized through endogenous esterases to NBI-98782 and NBI-136110. NBI-98782, the active metabolite, is further metabolized through multiple CYP pathways, predominantly 3A4 and 2D6. Neither valbenazine nor its metabolites are inhibitors or inducers of major CYP enzymes. Aside from VMAT2, the results of in vitro studies suggest that valbenazine and its active metabolite are unlikely to inhibit most major drug transporters at clinically relevant concentrations. However, valbenazine increased digoxin levels because of inhibition of intestinal P-glycoprotein; therefore plasma digoxin level monitoring is recommended when these 2 are co-administered.

Efficacy

Efficacy was established in a 6-week, fixed-dosage, double-blind, placebo-controlled trial of adult patients with TD. Eligible participants had:

• DSM-IV diagnosis of antipsychotic-induced TD for ≥3 months before screening and moderate or severe TD, as indicated by AIMS item 8 (severity of abnormal movement), which was rated by a blinded, external reviewer using a video of the participant’s AIMS assessment at screening
• a DSM-IV diagnosis of schizophrenia or schizoaffective disorder or mood disorder (and stable per investigator)
• Brief Psychiatric Rating Scale score <50 at screening.

Exclusion criteria included clinically significant and unstable medical conditions within 1 month before screening; comorbid movement disorder (eg, parkinsonism, akathisia, truncal dystonia) that was more prominent than TD; and significant risk for active suicidal ideation, suicidal behavior, or violent behavior.² Participants had a mean age of 56, 52% were male, and 65.7% of participants in the valbenazine 40-mg group had a schizophrenia spectrum disorder diagnosis, as did 65.8% in both the placebo and valbenazine 80-mg arms.

Antipsychotic treatments were permitted during the trial and >85% of participants continued taking these medications during the study. Participants (N = 234) were randomly allocated in a 1:1:1 manner to valbenazine 40 mg, 80 mg, or matched placebo. The primary outcome was change in AIMS total score (items 1 to 7) assessed by central, independent raters. Baseline AIMS scores were 9.9 ± 4.3 in the placebo group, and 9.8 ± 4.1 and 10.4 ± 3.6 in the valbenazine 40-mg and 80-mg arms, respectively.²

Outcome. A fixed-sequence testing procedure to control for family-wise error rate and multiplicity was employed, and the primary endpoint was change from baseline to Week 6 in AIMS total score (items 1 to 7) for valbenazine 80 mg vs placebo. Valbenazine, 40 mg, was associated with a 1.9 point decrease in AIMS score, while valbenazine, 80 mg, was associated with a 3.2 point decrease in AIMS score, compared with 0.1 point decrease for placebo (P < .05 for valbenazine, 40 mg, P < .001 for valbenazine, 80 mg). This difference for the 40-mg dosage did not meet the prespecified analysis endpoints; however, for the 80-mg valbenazine dosage, the effect size for this difference (Cohen’s d) was large 0.90. There also were statistically significant differences between 40 mg and 80 mg at weeks 2, 4, and 6 in the intent-to-treat population. Of the 79 participants, 43 taking the 80-mg dosage completed a 48-week extension. Efficacy was sustained in this group; however, when valbenazine was discontinued at Week 48, AIMS scores returned to baseline after 4 weeks.

Tolerability

Of the 234 randomized patients, 205 (87.6%) completed the 6-week trial. Discontinuations due to adverse events were low across all treatment groups: 2.6% and 2.8% in the placebo and valbenazine 40-mg arms, respectively, and 3.8% in valbenazine 80-mg cohort. There was no safety signal based on changes in depression, suicidality, parkinsonism rating, or changes in schizophrenia symptoms. Because valbenazine can cause somnolence, patients should not perform activities requiring mental alertness (eg, operating a vehicle or hazardous machinery) until they know how they will be affected by valbenazine.

Valbenazine should be avoided in patients with congenital long QT syndrome or with arrhythmias associated with a prolonged QT interval. For patients at increased risk of a prolonged QT interval, assess the QT interval before increasing the dosage.

Clinical considerations

Unique properties. Valbenazine is metabolized slowly to a potent, selective VMAT2 antagonist (NBI-98782) in a manner that permits once daily dosing, removes the need for CYP2D6 genotyping, and provides significant efficacy.


Why Rx? The reasons to prescribe valbenazine for TD patients include:

• currently the only agent with FDA approval for TD
• fewer tolerability issues seen with the only other effective agent, tetrabenazine
• no signal for effects on mood parameters or rates of parkinsonism
• lack of multiple daily dosing and possible need for 2D6 genotyping involved with TBZ prescribing.

Dosing

The recommended dosage of valbenazine is 80 mg/d administered as a single dose with or without food, starting at 40 mg once daily for 1 week. There is no dosage adjustment required in those with mild to moderate renal impairment; however, valbenazine is not recommended in those with severe renal impairment. The maximum dose is 40 mg/d for those who with moderate or severe hepatic impairment (Child-Pugh score, 7 to 15) however, valbenazine is not recommended for patients with severe renal impairment (creatinine clearance <30 mL/min) because the exposure to the active metabolite is reduced by approximately 75%. The combined efficacy and tolerability of dosages >80 mg/d has not been evaluated. Adverse effects seen with tetrabenazine at higher dosages include akathisia, anxiety, insomnia, parkinsonism, fatigue, and depression.

A daily dose of 40 mg may be considered for some patients based on tolerability, including those who are known CYP 2D6 poor metabolizers, and those taking strong CYP2D6 inhibitors.² For those taking strong 3A4 inhibitors, the maximum daily dose is 40 mg. Concomitant use of valbenazine with strong 3A4 inducers is not recommended as the exposure to the active metabolite is reduced by approximately 75%.² Lastly, because VMAT2 inhibition may alter synaptic levels of other monoamines, it is recommended that valbenazine not be administered with monoamine oxidase inhibitors, such as isocarboxazid, phenelzine, or selegiline.

Contraindications

There are no reported contraindications for valbenazine. As with most medications, there is limited available data on valbenazine use in pregnant women; however, administration of valbenazine to pregnant rats during organogenesis through lactation produced an increase in the number of stillborn pups and postnatal pup mortalities at doses under the maximum recommended human dose (MRHD) using body surface area based dosing (mg/m²). Pregnant women should be advised of the potential risk to a fetus. Valbenazine and its metabolites have been detected in rat milk at concentrations higher than in plasma after oral administration of valbenazine at doses 0.1 to 1.2 times the MRHD (based on mg/m²). Based on animal findings of increased perinatal mortality in exposed fetuses and pups, woman are advised not to breastfeed during valbenazine treatment and for 5 days after the final dose. No dosage adjustment is required for geriatric patients.