SAN DIEGO – The glucagon-like peptide-1 (GLP-1) receptor agonist liraglutide significantly lessened glucose tolerance, glycemic control, and other cardiometabolic risk factors in overweight or obese prediabetic patients receiving clozapine or olanzapine for schizophrenia, according to the findings of a randomized, double-blind, placebo-controlled trial.
Liraglutide was generally well tolerated and conferred cardiometabolic benefits similar to those in past studies of patients who were not on antipsychotic therapy, said Louise Vedtofte, PhD, of the University of Copenhagen. After 16 weeks of treatment, a 75-gram oral glucose tolerance test showed that 2-hour plasma glucose levels were 23% lower with liraglutide compared with placebo (P less than .001), she said at the annual scientific sessions of the American Diabetes Association.
Liraglutide patients also lost an average of 5.2 kg more body weight, cut 4.1 cm more from their waist circumference, and were significantly more likely to normalize their fasting plasma glucose and hemoglobin A1c levels compared with the placebo group (P less than .001 for each comparison), she said. The report by Dr. Vedtofte and her associates was published in JAMA Psychiatry simultaneously with the presentation at the meeting (2017 June 10. doi: 10.1001/jamapsychiatry.2017.1220).
Antipsychotics are core therapies in schizophrenia spectrum disorders but also cause increased appetite, weight gain, and cardiometabolic disturbances, noted Dr. Vedtofte. About a third of patients with schizophrenia who receive antipsychotics develop metabolic syndrome, and some 15% go on to develop diabetes. Clozapine and olanzapine cause more weight gain than do other antipsychotics, but swapping either medication for a more weight-neutral alternative can threaten the well-being of patients with schizophrenia, she emphasized. “Olanzapine and clozapine are good for the patients’ mental health, but are detrimental for their somatic health.”
As a GLP-1 receptor agonist, liraglutide inhibits appetite and food intake, and the Food and Drug Administration has approved treatment at doses of 1.8 mg for type 2 diabetes and 3 mg once daily for obesity. To explore how liraglutide affects patients on antipsychotics, Dr. Vedtofte and her associates randomly assigned 103 adults with schizophrenia spectrum disorders from two clinical sites in Denmark to receive placebo or 0.6 mg liraglutide subcutaneously once daily, up-titrated by 0.6 mg weekly to a maximum dose of 1.8 mg. At baseline, all patients were receiving stable treatment with clozapine, olanzapine, or both; had a body mass index of at least 27 kg/m2; and were prediabetic, with fasting plasma glucose levels of 110 to 125 mg/dL, HbA1c levels of 6.1%-6.4%, or 2-hour plasma glucose levels of 140 mg/dL or higher during the 75-gram oral glucose tolerance test.
Dr. Louise Vedtofte
Fully 93% of patients completed the study. At week 16, glucose tolerance improved significantly in the liraglutide group (P less than .001) but not in the placebo group (P less than .001 for difference between groups) after the researchers controlled for age, sex, illness duration, BMI, Clinical Global Impressions Scale severity score, and antipsychotic treatment. Glucose tolerance normalized in 30 patients who received liraglutide (64%) compared with 8 in the placebo group (16%; P less than .001), Dr. Vedtofte said.
Besides its benefits to waist circumference and weight loss, liraglutide was associated with reductions in systolic blood pressure (average, 5 mm Hg; visceral fat, 0.25 kg; and low-density lipoprotein levels, 15 mg/dL). These changes occurred even though liraglutide did not significantly alter C-peptide or glucagon secretion in the multivariate analysis of glucose tolerance test results, Dr. Vedtofte said. “The rate of nausea was 62% in liraglutide patients and 32% with placebo, but nausea was transient and did not explain the weight loss in subgroup analyses,” she added.
Liraglutide did not alter clinical liver function, but treated patients experienced a small (1.3 U/L), statistically significant rise in amylase levels. Rates of other adverse events were similar between groups except that patients were more likely to experience orthostatic hypotension on liraglutide (8.2%) than on placebo (0%; P = .04). Treatment with liraglutide did not significantly increase the likelihood of serious somatic or psychiatric adverse events, but one patient died while on therapy. “This patient was a man in his 60s with longstanding schizophrenia who was admitted to the hospital with vomiting and diarrhea, and died 3 days later,” Dr. Vedtofte said. Autopsy did not reveal the cause of death, but the patients showed no change in mental status or other signs of adverse therapeutic effects, she added.
Novo Nordisk funded the study and provided liraglutide and placebo injections. Additional support came from Capital Region Psychiatry Research Group, the foundation of King Christian X of Denmark, and the Lundbeck Foundation. Dr. Vedtofte had no disclosures.