Correlation of histological immunophenotype in papillary renal cell carcinoma with gene signatures related to the therapeutic effect of systemic therapy.

Abstract

To predict the therapeutic response of systemic therapy, comprehensive analyses of the tumor microenvironment in papillary renal cell carcinoma (pRCC) have been conducted previously using immunohistochemistry and RNA sequencing. This study aimed to evaluate the correlation between hematoxylin and eosin-based histological immunophenotypes and gene signatures employed in several clinical trials predicting responsiveness to immune checkpoint inhibitors and tyrosine kinase inhibitors, using data from the Cancer Genome Atlas (TCGA)-KIRP cohort (n = 254). Herein, we evaluated tumor-associated immune cells (TAICs) using three methodologies previously reported in clear cell RCC: a 3-tier immunophenotype (desert, excluded, and inflamed) based on the spatial distribution of TAICs; a 4-tier immunophenotype (cold, immune-low, excluded, and hot) considering both the location and degree of TAICs; and an inflammation score (score 0, 1, and 2) focusing only on the degree of TAICs. Furthermore, we compared the predictive ability of the three immunophenotypes. The histological immunophenotype in pRCC exhibited a correlation with adverse clinicopathological factors (including higher stage, WHO/ISUP grade, and the presence of sarcomatoid/rhabdoid changes), gene signatures related to angiogenesis, Teff, myeloid cells, JAVELIN Renal 101 Immuno, and immune checkpoints, as well as a poorer prognosis. Among the three methodologies, the 4-tier immunophenotype demonstrated the strongest correlation with gene signatures. In conclusion, the 4-tier immunophenotype may yield potential predictive biomarkers for pRCC and guide treatment decisions.