C3NIRA: Randomized phase II study of carboplatin-cabazitaxel-cetrelimab (anti-PD-1) induction followed by niraparib +/- cetrelimab maintenance in men with aggressive variant prostate cancers (AVPC).

5008 Background: AVPC have limited therapeutic options and a dismal prognosis. Although early data indicate that adding carboplatin (Carb) to cabazitaxel (Cab) benefits men with AVPC, responses are short-lived. In a previous study (NCT03263650) a subset of AVPC patients had long-term responses to CabCarb induction followed by PARP inhibitor maintenance. Tumor tissue analyses revealed upregulation of inflammatory and immune pathways in post-CabCarb treated tumors from prolonged responders. We therefore hypothesized that adding anti-PD-1 (cetrelimab, Cet) would increase the efficacy of CabCarb induction plus niraparib (Nira) maintenance in AVPC patients, and that correlates would expose tumor cell intrinsic and microenvironment pathways implicated in therapy response and resistance. Methods: In a single-institution study, men with CRPC meeting ≥ 1 of 9 AVPC criteria (listed in https://clinicaltrials.gov/study/NCT04592237 ) and ECOG performance status 0-2 received 6 cycles of Cab 20-25mg/m 2 and Carbo (AUC 3-4). Cet (360mg) was added to Cycles 2-6 of CabCarb. Men completing 6 cycles without progressive disease (PD) were randomized 1:1 to Nira 300mg BID ± Cet. The primary endpoint was progression free survival (PFS) in randomized patients. Subgroup comparisons between patients with PD in the first 6 cycles (Early PD) vs. patients who achieved randomization (Randomized) were made with chi-square and Kruskal-Wallis tests. Paired tumor biopsies were obtained after 1 cycle of CabCarb and 2 cycles of CabCarbCet. Results: 120 patients started CabCarb. Their median age was 68 years (30-83). Most were White/non-Hispanic (78%) and had not received prior docetaxel (58%). Of the 120, 20 (16.7%) went off study for reasons other than PD and 40 (33.3%) had Early PD and were not randomized. The Early PD group did not differ from the Randomized group in terms of age, race ethnicity, prior docetaxel, or baseline PSA. Post randomization median follow up was 20.5 months; median PFS was 3.4 months (95% confidence interval: 2.0, 4.4) with Nira (n=30) and 5.6 (3.7, 16.8) months with Nira+Cet (n=30, p=0.01); median overall survival was 10.2 (6.4, 19.6) with Nira and 24.3 (9.5, not reached) months with Nira+Cet (p=0.01). Single-cell RNA sequencing of paired biopsies revealed increased progenitor-like exhausted CD8+ T cells and decreased FoxP3+ Treg cells in Randomized patients (n=5) while Early PD patients (n=5) showed the opposite following the addition of Cet to CabCarb. Conclusions: A subset of men with AVPC derive meaningful benefit from the addition of anti-PD-1 to PARP inhibitor maintenance following platinum-taxane-anti-PD-1 induction. Ongoing correlates aim to identify biomarkers to select patients for this treatment strategy and reveal candidate mechanisms of resistance to guide future therapeutic combinations. Clinical trial information: NCT04592237 .