Muscarinic Cholinergic Receptor Agonist and Peripheral Antagonist for Schizophrenia

Abstract

Background: The muscarinic receptor agonist xanomeline has antipsychotic properties and is devoid of dopamine receptor-blocking activity but causes cholinergic adverse events. Trospium is a peripherally restricted muscarinic receptor antagonist that reduces peripheral cholinergic effects of xanomeline. The efficacy and safety of combined xanomeline and trospium in patients with schizophrenia are unknown.

Methods: In this double-blind, phase 2 trial, we randomly assigned patients with schizophrenia in a 1:1 ratio to receive twice-daily xanomeline-trospium (increased to a maximum of 125 mg of xanomeline and 30 mg of trospium per dose) or placebo for 5 weeks. The primary end point was the change from baseline to week 5 in the total score on the Positive and Negative Syndrome Scale (PANSS; range, 30 to 210, with higher scores indicating more severe symptoms of schizophrenia). Secondary end points were the change in the PANSS positive symptom subscore, the score on the Clinical Global Impression-Severity (CGI-S) scale (range, 1 to 7, with higher scores indicating greater severity of illness), the change in the PANSS negative symptom subscore, the change in the PANSS Marder negative symptom subscore, and the percentage of patients with a response according to a CGI-S score of 1 or 2.

Results: A total of 182 patients were enrolled, with 90 assigned to receive xanomeline-trospium and 92 to receive placebo. The PANSS total score at baseline was 97.7 in the xanomeline-trospium group and 96.6 in the placebo group. The change from baseline to week 5 was -17.4 points with xanomeline-trospium and -5.9 points with placebo (least-squares mean difference, -11.6 points; 95% confidence interval, -16.1 to -7.1; P<0.001). The results for the secondary end points were significantly better in the xanomeline-trospium group than in the placebo group, with the exception of the percentage of patients with a CGI-S response. The most common adverse events in the xanomeline-trospium group were constipation, nausea, dry mouth, dyspepsia, and vomiting. The incidences of somnolence, weight gain, restlessness, and extrapyramidal symptoms were similar in the two groups.

Conclusions: In a 5-week trial, xanomeline-trospium resulted in a greater decrease in the PANSS total score than placebo but was associated with cholinergic and anticholinergic adverse events. Larger and longer trials are required to determine the efficacy and safety of xanomeline-trospium in patients with schizophrenia. (Funded by Karuna Therapeutics and the Wellcome Trust; ClinicalTrials.gov number, NCT03697252.).

Figure 1. Assessment, Randomization, and Analysis.

The safety population included all the patients who had undergone randomization and had received at least one dose of xanomeline–trospium or placebo. The modified intention-to-treat population included all the patients who had undergone randomization, had received at least one dose of xanomeline–trospium or placebo, and had received a Positive and Negative Syndrome Scale assessment at baseline and at least once after baseline. One patient who was assigned to receive placebo was incorrectly given xanomeline–trospium and was therefore included in the safety analyses for the xanomeline–trospium group.

Figure 2 (facing page). Efficacy End Points.

The primary end point was the change from baseline to week 5 in the total score on the Positive and Negative Syndrome Scale (PANSS; range, 30 to 210, with higher scores indicating more severe symptoms of schizophrenia) (Panel A). The prespecified secondary efficacy end points are presented in the hierarchical order in which they were tested in the statistical analysis plan: the change in the PANSS positive symptom subscore (Panel B), the score on the Clinical Global Impression–Severity (CGI-S) scale (Panel C), the change in the PANSS negative symptom subscore (Panel D), the change in the PANSS Marder negative symptom subscore (Panel E), and the percentage of patients with a response according to a CGI-S score of 1 or 2 (Panel F). CGI-S scores range from 1 to 7, with higher scores indicating greater severity of illness. The PANSS positive symptom subscore, negative symptom subscore, and Marder negative symptom subscore each range from 7 to 49, with higher scores indicating greater severity of symptoms. All efficacy analyses were conducted in the modified intention-to-treat population. Changes in efficacy in continuous variable outcome measures (Panels A, B, D, and E) were evaluated with a mixed model for repeated measures, and differences in categorical variables were evaluated with nonparametric statistical tests (Mann–Whitney–Wilcoxon test, Panel C; Cochran–Mantel–Haenszel test, Panel F).