Clozapine Induces Metformin-Resistant Prediabetes/Diabetes that Is Associated with Poor Clinical Efficacy in Patients with Early Treatment-Resistant Schizophrenia

Background: Two distinct subtypes of treatment-resistant schizophrenia (TRS) have been recently reported, including early-treatment resistance (E-TR) as the dominant subtype and late-treatment resistance (L-TR). However, the response rate to clozapine, incidence of clozapine-induced prediabetes/diabetes, and effects of metformin at preventing clozapine-induced prediabetes/diabetes in patients with schizophrenia E-TR subtype need investigating. To assess clozapine-induced metformin-resistant prediabetes/diabetes and its correlation with clinical efficacy in schizophrenia E-TR subtype. Methods: This prospective cohort study enrolled 230 patients with schizophrenia E-TR subtype from Tianjin Kangtai Hospital between January 2015 and December 2020.The patients were treated with adequate doses of clozapine for 16 weeks, during which patients with prediabetes/diabetes were assigned to receive add-on metformin. Incidence of clozapine-induced prediabetes/diabetes and metformin-resistant prediabetes/diabetes and the efficacy of clozapine as assessed by the Positive and Negative Syndrome Scale (PANSS) score. Findings: Clozapine-induced prediabetes/diabetes occurred in 176 patients (170 prediabetes and 6 diabetes), with 76.52% incidence. The blood sugar of 43 (24.43%) patients was controlled with metformin. Despite add-on metformin, 47.06% (74/170) of patients with clozapine-induced prediabetes progressed to diabetes. In total, the incidence of clozapine-induced metformin-resistant prediabetes/diabetes was 75.57% (133/176). On completion of 16-week clozapine treatment, 16.52% (38/230) patients showed clinical improvement with PANSS scores of ≥50% declining. Furthermore, clozapine-induced prediabetes/diabetes was significantly correlated with the poor clinical efficacy of clozapine for schizophrenia E-TR subtype. In comparison with patients that did not have clozapine-induced prediabetes/diabetes, clinical efficacy was significantly lower in patients with clozapine-induced prediabetes/diabetes. Of note, the lowest clinical efficacy detected in patients with clozapine-induced diabetes [odds ratio (OR) of 0.18; 95% confidence interval (CI), 0.07–0.44]. Interpretation: The incidence of clozapine-induced metformin-resistant prediabetes/diabetes was considerably high in the schizophrenia E-TR subtype. Clozapine-induced metformin-resistant prediabetes/diabetes represents an independent risk factor that adversely affects the clinical efficacy of clozapine for the schizophrenia E-TR subtype. This study provided new evidence for re-evaluating the use of clozapine for TRS, especially E-TR subtype, and the use of metformin for the glycemic control of clozapine-induced prediabetes/diabetes.  Funding: This study was supported by the National Natural Science Foundation of China (81871052 and 81571319). Declaration of Interests: All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and there is no conflict of interest to disclose. Ethics Approval Statement: The study protocol was reviewed and approved by the Ethic Committee of Tianjin Kangtai Hospital (Tianjin, China).