ANTIPSYCHOTIC SWITCHING AMONG PATIENTS WITH BIPOLAR I DEPRESSION AFTER INITIATING CARIPRAZINE VS LUMATEPERONE: A REAL-WORLD CLAIMS-BASED STUDY

Introduction

Switching between atypical antipsychotics (AAs) is common in patients with bipolar I disorder (BP-I), and is associated with an increased risk of relapse and hospitalization. Understanding real-world patterns of AA switching could inform prescribing practices to potentially improve clinical and economic outcomes. This study compared realworld AA switch rates in patients with BP-I depression initiating cariprazine versus those initiating lumateperone.

Methods

Commercially insured adults with BP-I depression who initiated cariprazine or lumateperone (first dispensing=index date) were identified using claims data from the Komodo Research Database (12/20/2020–11/30/2024). Patients were required to have ≥12 months of continuous health insurance coverage preindex (baseline period), ≥2 pharmacy claims for the index AA, and ≥3 months of follow-up postindex. Exclusion criteria included administration of a long-acting injectable antipsychotic within 3 months preindex, dispensing of an antidepressant or mood stabilizer on the index date, dispensing of a nonindex AA from 14 days preindex to 30 days postindex, or a primary diagnosis of schizophrenia during the baseline period or on the index date. Baseline demographic and clinical characteristics of AA cohorts were balanced using inverse probability of treatment weighting based on propensity scores. Rates of switching within 3, 6, 9, or 12 months postindex were calculated from the weighted cohorts, where a switch was defined as ≥2 postindex dispensings of a nonindex AA (or ≥1 such dispensing in a separate sensitivity analysis). Logistic regression models compared switch rates between cohorts using odds ratios (ORs), 95% CIs, and P values.

Results

The analysis included patients who initiated cariprazine (n=2330) or lumateperone (n=571). At each time point considered, the proportion of patients who switched to a different AA was lower among those who initiated cariprazine relative to those who initiated lumateperone; switch rates among cariprazine and lumateperone cohorts were 12.7% and 19.2% at 3 months, 20.0% and 28.1% at 6 months, 26.2% and 31.6% at 9 months, and 28.7% and 38.6% at 12 months, respectively. Compared to patients who initiated lumateperone, those who initiated cariprazine were 39% less likely to switch by 3 months (OR [95% CI]=0.61 [0.48, 0.78], P < .001), 36% less likely by 6 months (0.64 [0.51, 0.80], P < .001), 23% less likely by 9 months (0.77 [0.60, 0.98], P < .05), and 36% less likely by 12 months (0.64 [0.49, 0.83], P < .001). ORs were also significantly in favor of cariprazine at 3, 6, and 12 months in a sensitivity analysis which relaxed the definition of a switch to ≥1 dispensing of a nonindex AA postindex. Patients switched to a variety of AAs, most commonly (≥5% for either cohort) aripiprazole (cariprazine cohort: 4.5%; lumateperone cohort: 7.0%), lurasidone (4.2%; 5.4%), and olanzapine (3.6%; 5.8%). Numerically more patients switched from lumateperone to cariprazine (4.7%) than from cariprazine to lumateperone (2.5%).

Conclusions

Among patients with BP-I depression, those who began cariprazine were significantly less likely to switch to a different AA at 3, 6, 9, and 12 months from treatment initiation compared to those who began lumateperone.