PD-L1 expression in primary clear cell renal cell carcinomas (ccRCCs) and their metastases.

467 Background: Clinical trials evaluating anti-PD-1 and anti-PD-L1 antibodies (Abs) in ccRCC have shown promising efficacy in a subset of patients. Preliminary studies have demonstrated that tumor PD-L1 expression increases the likelihood of benefit with anti-PD-1 Ab, but fails to identify all responders. One potential explanation for these results is that predictive biomarkers are usually evaluated in the primary tumors, which are more readily available; however, biomarker expression in nephrectomy samples may not accurately reflect expression in the metastases that are being targeted by therapy. In this study, we compared PD-L1 expression in a series of ccRCCs and their metastases. Methods: Formalin-fixed paraffin-embedded (FFPE) tissue blocks from 34 primary ccRCCs and corresponding metastases were retrieved. Multiple areas of the primary tumors, including areas of predominant and highest Fuhrman nuclear grade (FNG), were selected for analysis. Slides were immunostained with a mouse monoclonal anti-PD-L1 antibody (405.9A11). The assay was validated using FFPE cell line controls known to be positive or negative for PD-L1 expression by flow cytometry. The presence of tumor cells with membranous staining was assessed. A case was considered positive when any tumor cell positivity was detected. Results: Positive membranous PD-L1 expression in tumor cells was observed in 10/34 (29%) primary ccRCCs. In 3 of these 10 cases (30%), the metastases were negative. In 2 cases the primary tumor was negative but the metastases were positive. In twenty-two cases, both the primary tumor and the corresponding metastasis were negative. The pattern of PD-L1 staining was highly heterogeneous in the primary tumors and was restricted to areas of highest FNG. The staining was more homogeneous in the metastases. PD-L1 expression by the tumor infiltrating immune cells is currently being evaluated. Conclusions: Discordant expression of PD-L1 between the primary tumor and the corresponding metastases was detected in 5/34 (15%) cases, suggesting that accurate assessment of predictive biomarkers for PD-1 blockade in ccRCC might require analysis of metastatic lesions. Analysis of a larger patient cohort is ongoing to confirm these findings.