Estimated net benefit of avelumab (AVE) + best supportive care (BSC) vs BSC alone for patients (pts) with advanced urothelial carcinoma (aUC) using a quality-adjusted time without cancer symptoms or toxicity (Q-TWiST) analysis.

Abstract

4515 Background: The JAVELIN Bladder 100 (JB-100) trial (NCT02603432) showed that treatment with AVE first-line maintenance (1LM) + BSC significantly prolonged overall survival (OS) and progression-free survival (PFS) vs BSC alone in pts with aUC that had not progressed after 1L platinum-based chemotherapy. Safety and patient-reported outcomes (PROs) were previously reported. This post hoc analysis used JB-100 data to assess between-treatment differences using a Q-TWiST analysis, an integrated measure that incorporates efficacy, safety, and PROs into a single value. Methods: Mean OS was partitioned into 3 health states: time with all-cause grade ≥3 toxicity (TOX) prior to progression, time without all-cause grade ≥3 toxicity or symptoms of disease progression (TWiST), and time after progression or relapse (REL). To ensure equal observation time for each arm, analysis time was restricted to a period of 47.7 months from randomization, which was the shorter follow-up time between arms. A range of utility values were explored and weighted by the time in each health state then summed to estimate Q-TWiST to account for differences in quality of life. Mean between-treatment differences for each health state were also calculated. Bootstrap methods were used to estimate CIs for means and difference between means. Results: Time in TOX was slightly longer with AVE + BSC vs BSC alone; however, time spent in TWiST was much longer with AVE + BSC. Time in REL with AVE + BSC appears shorter vs BSC alone because pts experienced more time in the progression-free state, which led to less time in REL. Regardless of the range of utility values explored for each health state, the Q-TWIST estimate was always longer with AVE + BSC, with mean differences ranging from 3.7 to 7.1 months (95% CI not including 1). Conclusions: Pts receiving AVE + BSC demonstrated a consistently longer Q-TWiST (vs BSC alone), indicating a net benefit. Pts receiving AVE + BSC achieved greater quality-adjusted time by progressing more slowly (shorter REL) and living longer (greater OS). The Q-TWiST advantage reflects the safety profile of AVE in the context of a survival benefit. Restricted health state durations in months (mean, 95% CI). Clinical trial information: NCT02603432 . [Table: see text]