Background: Single agent or combination regimens using different IO and TKI agents have dramatically improved survival of patients (pt) with advanced ccRCC. Nevertheless, selecting optimal combination and sequencing of treatments remains challenging given the current absence of predictive biomarkers. Considering substantial disease heterogeneity and tumor clonal evolution, CTCs could provide a more comprehensive and longitudinal assessment of disease and response to treatment when compared to single site biopsy. In this study we explore changes in CTC characteristics in correlation with pt’s response (Resp) to treatment and progression (PD). Methods: 152 blood sample time points from 44 ccRCC pts (mean=3.5 sample/pt, range= 1-9) were included. An automated microfluidic platform using exclusion-based sample technology was employed for high sensitivity enrichment of RCC CTCs using EpCAM and CAIX antibodies. High-specificity CTC identification was achieved with positive staining for pCK or CAXII, and negative staining for CD45, CD34 and CD66b. Single cell phenotyping of PDL1 and HLA1 expression on each CTC was quantified with automated image analysis algorithms followed by manual visual QA of the putative CTCs. Cellular-level expression of PDL1 and HLA1 were compared within treatment-response groupings, using Wilcoxon rank-sum tests (Table). Results: Pts were 70% male, 93% Caucasian, 51% de novo metastatic, 98% received systemic treatment. CTC samples were collected at baseline (12) or while on treatment with IO (71), TKI (55), or IO/TKI (14). Across treatments, the number of CTCs increased from Resp to PD but this trend was not statistically significant. PDL1 expression was higher at PD compared to Resp on IO and IO/TKI but not significantly different on TKI. HLA1 expression was higher at PD compared to Resp on IO and TKI but not significantly different on IO/TKI. Average CTC PD-L1 expression was higher at Resp to IO compared to baseline. Conclusions: Single cell phenotyping of PDL1 and HLA1 were done at Resp compared to PD time points on IO and/or TKI to identify potential mechanisms of resistance. Further evaluation of these preliminary results in prospective trials is ongoing to advance new predictive biomarkers for ccRCC.

* for significant tests and NS for non-significant.