Background: Avelumab 1L maintenance is approved in the United States for patients with advanced UC that has not progressed with 1L platinum-containing chemotherapy based on significantly prolonged overall survival (OS) vs BSC seen in the phase III JAVELIN Bladder 100 trial (NCT02603432). However, optimal duration of 1L chemotherapy is unknown and some patients are unable to receive 6 cycles. We report a post hoc analysis of efficacy by duration or number of cycles of 1L chemotherapy. Methods: Eligible patients with unresectable locally advanced or metastatic UC that had not progressed with 4-6 cycles of 1L gemcitabine + cisplatin or carboplatin were randomized to receive maintenance avelumab + BSC or BSC alone within 4-10 weeks. Subgroups were defined by quartiles (Qs) for duration (<Q1 [<15.0 weeks], Q1-Q2 [15.0 to <18.0 weeks], Q2-Q3 [18.0 to <20.1 weeks], and >Q3 [>20.1 weeks]) or estimated number of cycles (4, 5, or 6) of 1L chemotherapy. Duration of chemotherapy included dosing delays/interruptions, and the decision to stop 1L chemotherapy was at the investigator’s discretion. Treatment arms were compared using an unstratified Cox proportional hazards model for OS. The potential impact of baseline characteristics on treatment patterns or dose intensity was not explored. Results: Numbers of patients in 1L chemotherapy subgroups were generally well balanced between arms (Table). An OS benefit was observed for avelumab + BSC vs BSC alone across subgroups with differing durations or cycles of 1L chemotherapy (Table). A progression-free survival benefit was also observed for avelumab + BSC vs BSC alone across subgroups. No significant treatment-by-cycle interaction (at 0.05 level) was observed. Conclusions: Improved OS was observed with avelumab 1L maintenance vs BSC alone irrespective of duration or cycles of 1L chemotherapy received prior to entering the trial. Among patients who stopped 1L chemotherapy prior to 6 cycles, avelumab 1L maintenance still provided an OS benefit. Clinical trial information: NCT02603432. Research Sponsor: Pfizer Inc, Pharmaceutical/Biotech Company

HR, hazard ratio; m, median months from randomization (after chemotherapy); NE, not estimable * Derived using exposure records of 1L chemotherapy based on a 21-day cycle assumption.