A phase II study of nivolumab in patients with high-risk biochemically recurrent (BCR) prostate cancer (PCa).

Abstract

TPS341 Background: Although androgen deprivation therapy (ADT) may be used for high-risk BCR PCa prior to radiographic metastases, evidence is lacking and toxicities are clear. Recent data show higher prevalence of programmed death-ligand 1 (PD-L1) expression in primary PCa than previously seen. Tumor PD-L1 expression also significantly correlates with risk of BCR, suggesting that inhibition of innate immune function could be associated with recurrence. Finally, immune-suppressive immune cells may increase after castration resistance, so the early-stage tumor microenvironment may differ from the castration-resistant setting in which prior studies were conducted. These data justify a novel trial investigating PD-1 inhibition in the setting of BCR. We hypothesize that PD-L1 expression in primary PCa is a marker of immunogenicity and therefore may predict for response to PD-1 inhibition. Methods: We designed a single-arm phase 2 clinical trial of nivolumab 480 mg every 28 days for men experiencing high-risk BCR. The primary endpoint is the proportion of patients experiencing PSA decline or stabilization (without symptomatic/radiographic progression) after 12 weeks of nivolumab; secondary endpoints are best PSA response, change in PSA doubling time, time to ADT initiation and to metastatic disease, and safety. The study will include one cohort with PD-L1 expression on tumor cells >5% (“PD-L1-positive”), the other, PD-L1-negative, thus enriched for but not limited to PD-L1-expressing primary PCa. We are conducting correlative studies using tumor tissue and blood samples collected during treatment. We will perform next-generation sequencing on tissue to identify genomic biomarkers correlated with PD-L1 expression and with response to nivolumab. Using neoantigen prediction models, we will examine whether there are peripheral T cells present in nivolumab responders that recognize tumor-associated neoantigens and are expanded by PD-1 inhibition. We will evaluate subsets of tumor-infiltrating lymphocytes, including “exhausted” phenotypes, and other infiltrating immune cells. Finally, we will assess changes in peripheral immune cells and plasma cytokines as potential biomarkers. Clinical trial information: NCT03637543.