Outcomes in participants (pts) with bacillus Calmette-Guérin (BCG)–unresponsive high-risk non–muscle-invasive bladder cancer (NMIBC) who underwent radical cystectomy (RC) after pembrolizumab (pembro): KEYNOTE-057 cohort A post hoc analysis.

745 Background: Based on results from cohort A of the phase 2 KEYNOTE-057 trial (NCT02625961), pembro monotherapy is a bladder-sparing treatment option for pts with BCG-unresponsive high-risk NMIBC with carcinoma in situ (CIS) with or without papillary disease. A post hoc analysis of KEYNOTE-057 suggested that pembro may provide a clinically meaningful delay in RC. We present updated outcomes of pts in cohort A of KEYNOTE-057 who underwent RC after discontinuing pembro after a median follow-up of 6 years. Methods: Adults in cohort A with histologically confirmed BCG-unresponsive high-risk NMIBC with CIS who were ineligible for or declined RC received pembro 200 mg IV Q3W for ≤2 years. End points of this analysis were to evaluate cystectomy-free survival (CFS) for complete and non–complete responders, time to cystectomy (TTC), pathologic staging at RC, and subsequent therapies/procedures other than RC following pembro discontinuation. CFS was calculated from first dose of pembro until death or RC. TTC was calculated from last dose of pembro. Complete responders were defined as pts with best overall response of CR. Results: Ninety-six pts were enrolled in cohort A; 44 underwent RC after discontinuing pembro (1 additional pt since previous analysis). Median follow-up (May 30, 2023) was 72.6 mo (range, 62.5-83.1). Before undergoing RC, 13/44 pts had initial CR. Median CFS (95% CI) was 56.8 mo (30.9-NR) for pts with CR and 18.5 mo (7.8-36.5) for pts without CR; 12-mo CFS rates were 87.2% and 54.1%, respectively. Median TTC (range) was 15.0 mo (9.0-69.8) for pts with CR and 6.4 mo (4.0-42.2) for pts without CR. Fifteen pts (34.1%; 2 with CR, 13 without) underwent subsequent therapy/procedures following pembro discontinuation. Six pts (13.6%) had upstaging to MIBC at RC (1 additional pt since previous analysis). The subset of pts (n = 3) with ≥pT3 stage, which is associated with higher recurrence risk, underwent RC ≥1 year after last dose of pembro. Conclusions: Long-term results from this post hoc analysis of KEYNOTE-057 continue to suggest that intravenous pembro can offer durable bladder preservation in high-risk BCG-unresponsive NMIBC with CIS, with sustained delays of RC among initial responders and minimal additional upstaging. Pts with CR had > 3-fold longer median CFS than nonresponders. Results suggest pembro can provide durable responses and can delay time to subsequent RC in most pts. Clinical trial information: NCT02625961 . Pts(n=44) Maximumpathologic T stage Pathologic N stage, n Prior CR on pembro, n TTC, median (range), mo NMIBC (n=38) N0/NX/N1/N2 Yes/No 7 pT0 6/1/0/0 5/2 4.5 (2.0-9.3) 4 pTa 4/0/0/0 0/4 5.1 (2.6-27.0) 21 pTis 19/2/0/0 6/15 2.5 (1.4-38.7) 6 pT1 6/0/0/0 0/6 4.4 (1.7-19.0) MIBC (n=6) 3 pT2 2/0/1/0 1/2 24.2 (2.0-46.5) (N0)2.8 (N1) 2 pT3 1/0/1/0 0/2 31.0 (N0)15.0 (N1) 1 pT4 0/0/0/1 1/0 12.7