Background: Patients (pts) with muscle-invasive bladder cancer (MIBC) ineligible for cisplatin-based chemotherapy have no standard systemic therapy options and are prioritized for radical cystectomy (RC) alone. This prospective clinical trial investigated the safety and efficacy of escalating doses of neoadjuvant atezolizumab (N-ATZ) prior to RC in MIBC pts. Methods: This is a single-institution, phase II trial of escalating doses of N-ATZ (1200 mg IV every 3 weeks) in pts with MIBC. Key inclusion criteria were resectable urothelial carcinoma of the bladder (T2-T4a,N0-1,M0) and inability to receive cisplatin-based treatment (eGFR < 60 mL/min, G≥2 neuropathy/hearing loss, pt decision). Pts with high-risk disease at RC were eligible to receive adjuvant ATZ for up to 16 total doses. Pts were followed for up to 2 years following RC. Primary efficacy endpoint was pathologic complete response (pCR; pT0N0). Secondary endpoints were safety of treatment, rate of pathologic downstaging (≤pT1N0), response based on PD-L1 status, and overall survival (OS) and recurrence-free survival (RFS) at 1 and 2 years from RC. Results: A total of 20 pts were enrolled and sequentially treated with one (n=6), two (n=5), and three (n=9) cycles of N-ATZ prior to RC. Median age was 69 (range 61-81), 75% were male and 85% Caucasian. Pts were cisplatin-ineligible due to low GFR (35%), hearing loss (25%) or neuropathy (10%); the rest refused cisplatin (30%). Most pts had pT2 disease (80%); the remainder, pT3/pT4 (15%/5%), and 10% had cN1. Among 17 pts with available tumor PD-L1 status, 76% had PD-L1 positive (CPS≥10) tumors. pCR was observed in 2 pts (10%) with 1 and 2 ATZ doses, whereas pathologic downstaging was observed in 5 pts (25%) across all 3 doses (Table). All pts completed intended treatment and RC within the trial-defined timeframe. Perioperative TRAEs of any grade occurred in 75%, but only 10% had G3 TRAEs (diarrhea, fecal incontinence). There were no G4/G5 events. Median follow-up from RC was 23.6 months and 75% were still followed at the time of data cutoff in 2/2021. Among evaluable pts, 1-year RFS and OS were 72% and 94% while 2-year RFS and OS were 64% and 69%. PD-L1 positive pts had superior OS (logrank p=0.06) and RFS (p=0.10) relative to PD-L1 negative pts. Conclusions: N-ATZ was well tolerated at all three dose levels and did not delay or prevent surgery. As few as 1 to 2 ATZ doses resulted in pathologic downstaging, including pCR. Although pCR rate in this trial was lower than expected, most pts had a durable recurrence-free period and all evaluable pts with tumor downstaging were alive and recurrence-free at 2 years following RC. Increased tumor PD-L1 expression was suggestive of improved outcomes and further biomarker analyses are ongoing. Clinical trial information: NCT02451423