COPENHAGEN—The erythropoiesis-stimulating agent (ESA) darbepoetin alfa can provide a clinical benefit in patients with lower-risk myelodysplastic syndromes (MDS), a phase 3 trial suggests.
In the ARCADE trial, darbepoetin alfa significantly reduced the incidence of red blood cell (RBC) transfusions in patients with low- and intermediate-1 risk myelodysplastic syndrome (MDS), when compared to placebo.
The ESA also significantly improved erythroid response.
In addition, researchers said adverse events (AEs) were generally balanced between the darbepoetin alfa and placebo arms.
Uwe Platzbecker, MD, of University Hospital Carl Gustav Carus Dresden in Germany, presented these results at the 21st Congress of the European Hematology Association (abstract S128). The ARCADE trial was sponsored by Amgen.
Dr Platzbecker noted that, although ESAs are recommended in clinical guidelines to treat anemia in patients with lower-risk MDS, the drugs are not widely approved for this indication.
So, in the ARCADE trial, he and his colleagues assessed darbepoetin alfa in patients with low- or intermediate-1 risk MDS who had not previously taken ESAs or biologic response modifiers.
The patients had hemoglobin levels ≤10 g/dL, endogenous erythropoietin levels ≤500 mU/mL, and low transfusion burden (<4 RBC units in each of 2 consecutive 8-week periods prior to randomization).
During a 24-week period, 147 patients received either darbepoetin alfa at 500 μg (n=97) or placebo (n=49) every 3 weeks. The ESA dose was withheld if patients’ hemoglobin was >12.0 g/dL and decreased if hemoglobin increased by >1.5 g/dL in 3 weeks without transfusion.
At week 25, when the primary and key secondary endpoints were assessed, patients underwent an end-of-treatment period visit. They could then enter a 48-week active treatment period and cross over to receive darbepoetin alfa, with dose escalation allowed beginning on week 31. Treatment continued until week 72 or 73, and patients continue to be assessed every 26 weeks, for a minimum of 3 years.
Patient characteristics
Dr Platzbecker said baseline demographic and disease characteristics were generally similar between the treatment arms. All patients were Caucasian, and about 55% were male. The median age was 74 (range, 67-79). About half of patients in each treatment arm belonged to the low-risk IPSS category.
In both arms, most patients had refractory cytopenia with multilineage dysplasia (38.8% in the placebo arm and 46.4% in the darbepoetin alfa arm). Patients also had refractory anemia with excess blasts-1 (20.4% and 13.4%, respectively), refractory anemia (26.5% and 9.3%), refractory anemia with ring sideroblasts (8.2% and 17.5%), 5q deletion (4.1% and 11.3%), unclassifiable MDS (2.0% and 1.0%), and MDS of an unknown type (0% and 1.0%).
In the 16 weeks before randomization, 58.2% of all patients—53.1% in the placebo arm and 60.8% in the darbepoetin alfa arm—did not have any RBC transfusions. About 25% (24.7%)—22.4% in the placebo arm and 25.8% in the darbepoetin alfa arm—received 1 to 3 RBC units. And 17.1%—24.5% in the placebo arm and 13.4% in the darbepoetin alfa arm—received 4 or more RBC units.
Dosing
During the 24-week double-blind period of the study, 77% (37/48) of patients in the placebo arm and 79% (77/98) in the darbepoetin alfa arm received all 8 doses of treatment.
Sixteen percent (n=16) of patients in the darbepoetin alfa arm had a single dose reduction, and 2% (n=2) had 2 dose reductions. None of the patients in the placebo arm had a dose reduction.
Eleven percent of patients in the darbepoetin alfa arm had doses withheld due to increased hemoglobin. The dose was withheld once for 6 patients, twice for 4 patients, and 3 times for 1 patient. None of the placebo-treated patients had a dose withheld for this reason.
Ten percent (n=5) of placebo-treated patients and 2% (n=2) of darbepoetin alfa-treated patients had a dose withheld due to an AE. Two percent (n=1) and 3% (n=3) of patients, respectively, had a dose withheld for “other” reasons (noncompliance, investigator decision, and no investigational product on site).
During the 48-week open-label period of the study, 81% (102/126) of patients who received darbepoetin alfa increased their dose frequency from every 3 weeks to every 2 weeks. Dr Platzbecker said this suggests the optimal dose of the drug was not achieved during the 24-week double-blind period of the study.
Efficacy
During the 24-week double-blind period, there was a significant difference between the treatment arms with regard to RBC transfusions. The transfusion incidence was 59.2% (29/49) in the placebo arm and 36.1% (35/97) in the darbepoetin alfa arm (P=0.008).
During the 48-week open-label period, the incidence of RBC transfusion was 50.8% (64/126) among patients receiving darbepoetin alfa.
During the 24-week double-blind period, 11 patients (14.7%) in the darbepoetin alfa arm had an erythroid hematologic improvement (HI-E), but none of the patients in the placebo arm had such an improvement.
All 11 patients with HI-E had a baseline serum erythropoietin level less than 100 mU/mL, 1 of the patients had 2 RBC units transfused in the 16 weeks prior to randomization, but none had transfusions in the 8 weeks prior to randomization. Four of the patients had a dose withheld due to having hemoglobin levels greater than 12 g/dL.
During the 48-week open-label period, the HI-E rate was 34.7% (34/98) among patients receiving darbepoetin alfa.
Dr Platzbecker said the nature of the HI-E criteria likely underestimated the clinical benefit of darbepoetin alfa in this trial, and this was further complicated by the trial design. Specifically, hemoglobin was measured every 3 weeks, some patients may have had their doses reduced even if they were still anemic, and the optimal dose of darbepoetin alfa was likely not given during the double-blind period (as evidenced by the increase in doses during the open-label period).
For these reasons, Dr Platzbecker and his colleagues are exploring alternative response analyses to determine if there were additional patients who received a clinical benefit from darbepoetin alfa but did not meet HI-E criteria.
Safety
During the 24-week double-blind period, 4.2% (n=2) of patients in the placebo arm and 3.1% (n=3) in the darbepoetin alfa arm had AEs that led to treatment discontinuation. In the placebo arm, these events were pulmonary hypertension and renal failure. In the darbepoetin alfa arm, the events were pulmonary thrombosis, thrombocytopenia, and increased blast cell count.
The incidence of grade 3 or higher AEs was 27.1% (n=13) in the placebo arm and 15.3% (n=15) in the darbepoetin alfa arm. The incidence of grade 4 or higher AEs was 12.5% (n=6) and 5.1% (n=5), respectively. And the incidence of serious AEs was 16.7% (n=8) and 11.2% (n=11), respectively.
The incidence of fatal AEs was 4.2% (n=2) and 1% (n=1), respectively, but none of these were treatment-related. The deaths in the placebo arm were due to cardiac failure and cerebral hemorrhage, while the death in the darbepoetin alfa arm was due to hemorrhagic proctitis.
One patient in the darbepoetin alfa arm experienced a treatment-related serious AE.
AEs occurring at least 5% more frequently in the darbepoetin alfa arm than the placebo arm were fatigue (17.3% and 8.3%), pyrexia (9.2% and 2.1%), headache (7.1% and 2.1%), and myalgia (5.1% and 0%).
During the 48-week double-blind period, 7.9% (n=3) of patients formerly in the placebo arm and 3.4% (n=3) of patients formerly in the darbepoetin alfa arm had AEs that led to treatment discontinuation.
The incidence of grade 3 or higher AEs was 23.7% (n=9) and 31.0% (n=27), respectively. The incidence of grade 4 or higher AEs was 10.5% (n=4) and 10.3% (n=9), respectively. And the incidence of serious AEs was 18.4% (n=7) and 25.3% (n=22), respectively.
The incidence of fatal AEs was 2.6% (n=1) and 1.1% (n=1), respectively, but none of these were treatment-related. Two patients experienced a treatment-related serious AE—1 from each of the former treatment arms.
