Phase II study of oral APL-1202 plus tislelizumab or tislelizumab alone as neoadjuvant therapy in patients with muscle-invasive bladder cancer (MIBC).

Abstract

793 Background: Interim results of the randomized phase II trial of neoadjuvant tislelizumab +/- oral APL-1202 (nitroxoline) revealed promising pathological complete response (pCR) rates and met prespecified thresholds for study expansion (ASCO GU 2024). Here, we report the final primary endpoint analysis of this trial. Methods: Patients with cT2-T4aN0M0 urothelial cancer of the bladder based on local assessment, planned for radical cystectomy (RC), and ineligible for or refusing cisplatin-based chemotherapy were eligible. Patients were randomly assigned to APL-1202 plus tislelizumab (A+T) or tislelizumab (T), stratified by PD-L1 expression. Neoadjuvant tislelizumab was administered q3weeks for 3 cycles and APL1202 was administered orally tid. The primary endpoint was centrally assessed (pCR, pT0N0) rate and secondary endpoints included central pathologic response (PaR, < ypT2N0) rate and safety. Results: A total of 103 patients were enrolled; 28 patients declined RC after neoadjuvant treatment (A+T=16, T=12). Consequently, 75 patients remained in the efficacy analysis set (EAS); RC was completed after A+T in 42/43 patients and after T in 31/32 patients. The pCR and PaR results in the EAS are presented in the table. On retrospective central pathology review of baseline TURBT specimens, a large subset of patients were determined to be ineligible due to <cT2 disease (33/75). There was a numerically higher pCR rate in an exploratory analysis of this retrospectively defined ‘protocol eligible’ subset (A+T: 9/22, T: 4/20). Treatment related adverse events (TRAEs) occurred in 35 (59%) on the A+T arm and 19 (44%) on the T arm. TRAEs (94%) were predominantly ≤ CTCAE grade 2. Conclusions: Neoadjuvant A+T in cisplatin-ineligible patients prior to RC was safe. Evaluation of the efficacy in MIBC is complicated by the large subset of patients enrolled retrospectively determined to have <cT2 disease but pCR rates were similar between groups in the EAS. A signal of higher activity of A+T may be present in PD-L1 “low” tumors supporting further exploration of the immunomodulatory effects of A+T. Evaluation of efficacy is further complicated by a large subset of patients declining RC after neoadjuvant treatment underlining the need for novel bladder-sparing approaches. Clinical trial information: NCT04813107 . pCR rate, pT0N0 PaR rate, < ypT2N0 EAS (n=75) n(%) A+T (n=43) T (n=32) A+T (n=43) T (n=32) All evaluable patients 12 (33)* 7 (26)* 19 (44) 13 (41) PD-L1 h igh 5/18 (28) 4/16 (25) 7/18 (39) 8/16 (50) PD-L1 l ow 7/25 (28) 3/16 (19) 12/25 (48) 5/16 (31) *pCR rate is calculated using UMVUE method when the final sample size of Simon's two-stage is changed.