INTRODUCTION AND OBJECTIVE: Immuno-oncology (IO) treatment has become the standard of care for locally-advanced and/or metastatic renal cell carcinoma (RCC). Yet, we do not fully understand whole pictures of IO treatment appearing one after another in this high-throughput era. The aim of this study is to propose the novel immune-based clustering (IC) algorithm using future immune-checkpoint targets TIGIT, LAG3, and TIM3 for RCC prognosis. METHODS: A total of 152 clear cell RCC (ccRCC) specimens, in which 105 were from primary lesions and 47 from metastases, were enrolled. The center of tumor was histologically selected from each case by one genitourinary pathologist. The expression levels of TIGIT, LAG3, and TIM3 were evaluated by the virtual slide scanner and automated single-cell count system for immunohistochemistry, and then categorized based on our clustering algorithm. RESULTS: In 105 cases of primary ccRCC, median follow-up period was 99.2 months and 27 (26%) death events were observed. By the IC algorithm based on TIGIT, LAG3, and TIM3expression levels, all of primary ccRCC were clearly categorized into TIM3- (53%), LAG3- (35%) and TIGIT-dominant (11%) clusters. (Figure1A). Remarkably, prognostic analysis showed that recurrence-free survival rates were significantly different among three clusters (p=0.009, Figure1B). Overall survival rates tended to be different but not statistically significant (p=0.144, Figure1C). Metastatic lesions were also categorized into TIM3- (36%), LAG3- (19%) and TIGIT-dominant (44%) clusters (Figure1D). However, the categorized ratio into three clusters were completely different from those of primary lesion (P=0.001). Notably, all of brain metastasis (n=4) were categorized into the TIM3-dominant cluster. CONCLUSIONS: We successfully divided all of 152 ccRCC specimens into three clusters based on new immune-checkpoint target TIGIT, LAG3, and TIM3expression levels. Our IC model may help to predict patient survivals as well clinically depict future candidates for targeting TIGIT, LAG3, and TIM3 in ccRCC.Source of Funding: None