Comparing progression-free survival (PFS) in second-line (2L) urothelial carcinoma (UC) treatments from single-arm trials: Importance of using appropriate methods.

Abstract

573 Background: Population-adjusted comparisons of PFS from single-arm trials of cancer treatments can be derived using matching-adjusted indirect comparison (MAIC); however, results are still susceptible to bias, particularly if the trials had different tumor assessment schedules (favoring longer intervals). This is shown with an MAIC of avelumab vs atezolizumab, nivolumab, durvalumab, pembrolizumab or chemotherapy in 2L UC. Methods: The MAIC used patient-level data for avelumab from the JAVELIN Solid Tumor trial (NCT01772004) and compared PFS with published curves for other treatments to obtain population-adjusted hazard ratios (HRs). The first assessment was scheduled at 6 weeks for avelumab and durvalumab and at 8 or 9 weeks for the others. The MAIC was repeated after aligning the first assessment time for avelumab to each comparator’s schedule with the Assessment Schedule Matching (ASM) method. Results: Naïve analyses (Table) suggest similar PFS for avelumab vs comparators. MAIC based on population adjustment (PA) altered the HR of avelumab vs atezolizumab and nivolumab (23.0% and 9.2% reductions in HR, respectively) but had little effect in other comparisons. ASM adjustment led to further changes in HR (13.2-32.7% reductions from naïve) favoring avelumab in MAICs where schedules differed. Of note, assessment schedule differences fully accounted for differences in PFS between avelumab and pembrolizumab. Conclusions: Indirect comparisons of PFS from single-arm trials must include adjustment for differences in both population characteristics and tumor assessment schedules. In this 2L UC MAIC, adjustment reversed some naïve comparisons, revealing more favorable PFS with avelumab up to 30%. Clinical trial information: NCT01772004. [Table: see text]