Abstracts

Rationale: Everolimus is approved for the treatment of renal angiomyolipoma and subependymal giant cell astrocytoma in tuberous sclerosis complex (TSC). Stomatitis and infections are the most common adverse events (AEs) reported with everolimus. Here, we report on these AEs and their time to occurrence in EXIST-3 (clinicaltrials.gov NCT01713946), a 3-arm, randomized, double-blind, phase 3 study that demonstrated a significant reduction in TSC-associated seizures in patients treated with everolimus. Methods: A total of 366 patients were randomized to everolimus low-exposure (LE [3-7 ng/mL]; n=117), high-exposure (HE [9-15 ng/mL]; n=130) and placebo (n=119). Everolimus dose was adjusted during titration (weeks 1-6) and, as needed, during the subsequent 12-week maintenance phase. After core phase, patients were offered to switch in an extension phase in which all patients received everolimus. Safety analyses included frequency of AEs by randomized treatment arms, a standard Cox regression analysis by randomized treatment arm and an extended Cox regression analysis stratified by age to explore the relationship between time to first occurrence of stomatitis or infections and exposure as a time-dependent covariate (expressed as time-normalized [TN] Cmin which denotes an estimated average of Cmin over four periods: titration, maintenance, and 2 periods in the extension phase). Results: The most common AEs observed in >15% of patients in at least one arm during the core phase are presented by age group in Table 1. Stomatitis occurred early in the study, with median time to first event was 8 weeks for LE (95% confidence interval [CI], 2.9-NE) and 4 weeks for HE (95% CI, 2.6-8.9). Compared to placebo, the risk of reporting these AEs was higher in LE (hazard ratio [HR] stomatitis: 8.61; 95% CI, 4.53-16.37; HR infections: 1.30; 95% CI, 0.91-1.87) and HE arms (HR stomatitis: 11.02; 95% CI, 5.86-20.74; HR infections: 1.63; 95% CI, 1.16-2.30). The extended Cox model showed that a 2-fold increase in everolimus exposure was not associated with a statistically significant increase in the risk of stomatitis (HR, 1.09; 95% CI, 0.87-1.38) or infections (HR, 1.06; 95% CI, 0.85-1.32). Conclusions: Overall frequency of AEs was similar among patients randomized to LE and HE across age groups with fewer infections in adults. On an intent-to-treat basis, there was a trend towards more and earlier occurrence of stomatitis and infections with HE compared to LE, during the first 18-weeks of treatment. Yet, a model-based approach indicated that a 2-fold increase in the TN Cmin of everolimus was not associated with an increase in the risk of stomatitis or infection, but was associated (as previously reported) with an approximate 2-fold increase in the odds of achieving a 50% reduction in seizure frequency. The increased benefit:risk associated with an increase in everolimus exposure should be considered in patients who have not yet demonstrated an adequate response in seizure frequency. Funding: This study was sponsored by Novartis Pharmaceuticals Corporation.