Atezolizumab plus personalized neoantigen vaccination (PGV001) in patients with urothelial cancer.

597 Background: Most patients (pts) with urothelial cancer (UC) treated with immune checkpoint inhibitors (ICIs) do not respond. Given that features associated with restrained antitumor immunity and high neoantigen burden have correlated with response to ICI, stimulating antitumor immunity with neoantigen vaccination is an attractive approach to overcome ICI resistance. This study assessed atezolizumab in combination with PGV001, a personalized genomic neoantigen vaccine, in the adjuvant and metastatic settings. Methods: This single-arm pilot study (NCT03359239) enrolled pts with UC for treatment in the adjuvant (≥pT3/pN+ who declined or were ineligible for adjuvant chemotherapy; or ≥ypT2/ypN+ after neoadjuvant chemotherapy) or metastatic (disease progression after chemotherapy, or “switch maintenance” for stable disease after chemotherapy) setting. Neoantigen prediction was performed using the OpenVax computational pipeline. After an initial priming vaccination course (cycle 1), pts received atezolizumab 1200 mg intravenously every 3 weeks for up to 12 months (adjuvant) or 24 months (metastatic) and PGV001 with poly-ICLC on subsequent days every 3 weeks during cycles 2-5 and 9. The primary endpoints were feasibility and safety, including the number of neoantigens identified and peptides synthesized per pt, vaccine production time, proportion of consented pts with prepared vaccine, and proportion of treatment-eligible pts completing the priming course. Secondary endpoints included immune monitoring of vaccine-induced neoantigen specific T cells. Results: Between 12/2018 and 8/2020 12 pts were enrolled. The median neoantigens identified per pt was 103 (range 60-388). Vaccines containing 9-10 synthetic long peptides (24-25mer) were prepared for 92% (11/12) of consented pts over a median of 153 days (95% CI 151-181 days). 100% (10/10: 4 adjuvant, 6 metastatic) of pts eligible for the treatment phase completed the priming course. All evaluated pts demonstrated the emergence of ex vivo T cell responses as reflected by IFNγ production, including circulating multi-functional CD4 + and CD8 + neoantigen-specific T cell responses upon expansion. At a median follow up of 32 months, 4/4 patients treated in the adjuvant setting were free of recurrence and 2/5 pts with metastatic UC achieved an objective response. The single pt treated in the switch maintenance setting had no evidence of disease, off all treatment, at 33 months. The most common treatment-related adverse events were grade 1 injection site reactions, fatigue, and fever. One pt developed grade 3 immune-related hepatitis. Conclusions: Atezolizumab plus PGV001 was feasible, safe, and induced neoantigen-specific T cell immunity in all evaluated pts. To our knowledge, this is the first report of combination neoantigen vaccination plus ICI in the adjuvant setting in UC. Neoantigen vaccination plus ICI warrants further evaluation. Clinical trial information: NCT03359239 .