Abstract

4509 Background: In the CheckMate 901 trial, combination nivolumab (NIVO) + gemcitabine-cisplatin (GC) demonstrated significant improvements in overall survival (OS) and progression-free survival (PFS) with compelling objective response rates (ORR; 57.6% with NIVO+GC vs 43.1% with GC alone) and deep, durable complete responses (CR; 21.7% with NIVO+GC vs 11.8% with GC alone) in patients (pts) with previously untreated unresectable or metastatic urothelial carcinoma (mUC). Lymph node (LN)-only metastatic disease is a favorable prognostic factor in pts with mUC and a subset of pts achieve durable disease-free, treatment-free survival with GC +/- surgical consolidation. We conducted a post hoc analysis to characterize the subset of pts with CR, with further analysis of pts with LN-only mUC. Methods: In the global, open-label, randomized, phase 3 CheckMate 901 (NCT03036098) trial, cisplatin-eligible pts received NIVO 360 mg + GC every 3 wk for ≤6 cycles followed by NIVO 480 mg every 4 wk until disease progression/unacceptable toxicity or up to a maximum of 2 yrs, or GC every 3 wk for ≤6 cycles. Primary endpoints were OS and PFS by blinded independent central review (BICR). ORR per BICR and safety were exploratory endpoints. These post hoc analyses evaluated treatment outcomes in complete responders and in pts with LN-only disease. Results: Of the 608 pts randomized, 102 (16.8%) achieved a CR. Baseline disease characteristics of these pts are shown in the Table. As pts with LN-only mUC were enriched in the CR group, additional analysis of this subgroup was performed. Of all randomized pts, 54 treated with NIVO+GC and 56 treated with GC had LN-only mUC. In these pts, the ORR and CR rate was 81.5% (95% CI 68.6-90.7) and 63.0% versus 64.3% (50.4-76.6%) and 33.9% for NIVO+GC and GC, respectively. Median OS (95% CI) in LN-only pts was 46.3 (24.0-NE) mos with NIVO+GC vs 24.9 (21.4-29.9) with GC (HR, 0.58, 95% CI 0.34-1.00), and median PFS (95% CI) was 30.5 (9.6-NE) mos with NIVO+GC vs 8.8 (7.5-10.9) mos with GC (HR 0.38, 95% CI 0.22-0.66). Conclusions: NIVO+GC generated deep responses in CheckMate 901 with a fixed duration of chemotherapy and up to 2 years NIVO. Exploratory characterization of pts with CR identified a group of pts enriched with LN-only disease.In pts with LN-only mUC, NIVO+GC induced durable disease control and clinically meaningful improvements in OS and PFS vs GC alone. These results provide additional support for NIVO plus cisplatin-based chemo as a first-line treatment option for pts with mUC. Clinical trial information: NCT03036098 . [Table: see text]