A retrospective analysis of data from two trials of sunitinib in patients with advanced renal cell carcinoma (RCC): Pitfalls of efficacy subgroup analyses based on dose-reduction status.
363 Background: Patients requiring sunitinib dose reduction who received a novel dose/schedule modification scheme had longer progression-free survival (PFS) than patients who did not require dose reduction and remained on 50 mg Schedule 4/2 (4 weeks on, 2 weeks off) (Bjarnason 2011). Our analysis compared sunitinib efficacy in advanced RCC patients with/without dose reduction using the label-approved, dose-reduction scheme (ie, 50 mg/37.5 mg/25 mg) and schedule (Schedule 4/2) and explored potential causes of any differences. Methods: Data from a Phase 3 and a Phase 2 trial, sunitinib Schedule 4/2 arms only (N=375 and 146, respectively), were retrospectively analyzed, and pharmacokinetics and baseline characteristics of patients with/without dose reduction compared. Results: In the Phase 3 trial, median (95% CI) PFS was 14.0 (13.1–16.2) months (mos) and 8.1 (6.3–10.6) mos with (n=194) and without (n=181) dose reduction, respectively. In the Phase 2 trial, corresponding PFS values were 13.4 (9.8–19.8) mos and 5.8 (3.9–8.5) mos (n=51 and 95, respectively). In the Phase 2 trial, steady-state mean (SD) total drug trough concentrations were 96.0 (42.2) ng/mL and 85.8 (43.4) ng/mL on Day 29 of Cycle 1 in patients with/without dose reduction, respectively. In both studies, the percent of patients with baseline Memorial Sloan-Kettering Cancer Center risk factors of 0 (favorable), 1–2 (intermediate), and 3 (poor) were 37–47%, 53–57% and 0–6% with dose reduction vs. 25–28%, 65–72%, and 0–10% without dose reduction. The mean (range) time to dose reduction was 7.2 (0.03–40.4) mos in the Phase 3 trial and 4.5 (0.4–22.6) mos in the Phase 2 trial. Conclusions: Patients with dose reduction remaining on Schedule 4/2 appeared to have longer PFS than patients with no dose reduction. The differences were not caused by differences in plasma drug exposures; they appeared to be due, at least in part, to 1) differences in patients’ baseline prognostic factors and 2) patients’ PFS or longevity affecting their dose-reduction status. Thus, efficacy subgroup analysis based on patients’ dose-reduction status appears to be confounded, leading to biased results.