Assessment of the Effectiveness of Risk Evaluation and Mitigation Strategy for the Prevention of Fetal Exposure to Phentermine-Topiramate Using Two Large Nationwide Cohorts

Background: Upon approval of phentermine-topiramate in 2012, the US Food and Drug Administration required a Risk Evaluation and Mitigation Strategy (REMS) to prevent fetal exposure due to concerns about its teratogenicity. However, the effectiveness of the REMS, which includes regular pregnancy testing and contraception use, in preventing pregnancy occurrence is unknown.

Objectives: To evaluate the effectiveness of phentermine-topiramate REMS for prevention of pregnancy occurrence.

Methods: Using the nationwide Medicaid 2012-2018 and MarketScan 2012-2020 databases, we assembled two cohorts of females who newly initiated either phentermine-topiramate (exposed group) or phentermine (reference, which does not have a REMS). Patients were considered on treatment for 30 days after the end of the period covered by the days’ supply. If the gap between the end of one prescription and the start of the next was > 30 days, the next prescription was counted as a new treatment episode. We excluded patients with all causes of infertility. To determine pregnancy occurrence, we implemented a hierarchical algorithm which identifies pregnancy start date. Patients were followed from the index date (i.e., treatment initiation date), until treatment discontinuation, estimated pregnancy start date, enrollment end, or end of available data, whichever occurred first. For both groups, we estimated the rate of pregnancy, live and non-live births per 10,000 person-days (pd), and hazard ratios (HR) with 95% confidence intervals (CIs) using Cox proportional hazards models. We further adjusted for age, year, and race (Medicaid only).

Results: We identified 821 phentermine-topiramate and 14,846 phentermine treatment episodes in Medicaid, and 10,643 and 138,023 episodes respectively in MarketScan. The pregnancy rate was 2.1 per 10,000 pd in both groups in Medicaid (HR: 1.00; 95% CI, 0.63 - 1.58), while the rate in MarketScan was 0.51 per 10,000 pd among phentermine-topiramate initiators and 0.98 per 10,000 pd among phentermine initiators (0.54; 0.43 - 0.68). Similarly, the incidence rates of live births (0.98; 0.54 - 1.75) and non-live births (1.04; 0.50 - 2.15) were similar between the two groups in Medicaid, while they were lower in the phentermine-topiramate exposed group in MarketScan (0.45, 0.34 - 0.60; 0.78, 0.54 - 1.14). The adjusted HR remained consistent with the crude for all pregnancy outcomes.

Conclusions: We observed an effect of the REMS program for phentermine-topiramate in reducing the incidence of pregnancy relative to phentermine exposed females only in the commercially insured patients. Further research is required to examine the apparent differential pregnancy rates among commercially vs publicly insured populations.