Spatially aware deep learning reveals tumor heterogeneity patterns that encode distinct kidney cancer states

• DOI: 10.1016/j.xcrm.2023.101189
• Publication Date: 2023-09-19T17:47:37Z
• Authors: Jackson Nyman; Thomas Denize; Ziad Bakouny; Chris Labaki; Breanna M. Titchen; Kevin Bi; Surya Narayanan Hari; Jacob Rosenthal; Nicita Mehta; Bowen Jiang; Bijaya Sharma; Kristen Felt; Renato Umeton; David A. Braun; Scott Rodig; Toni K. Choueiri; Sabina Signoretti; Eliezer M. Van Allen
• Keywords: 0303 health sciences; 03 medical and health sciences; Deep Learning; Phenotype; Carcinoma; Humans; Carcinoma, Renal Cell; Article; Kidney Neoplasms; 3. Good health

Clear cell renal cell carcinoma (ccRCC) is molecularly heterogeneous, immune infiltrated, and selectively sensitive to immune checkpoint inhibition (ICI). However, the joint tumor-immune states that mediate ICI response remain elusive. We develop spatially aware deep-learning models of tumor and immune features to learn representations of ccRCC tumors using diagnostic whole-slide images (WSIs) in untreated and treated contexts (n = 1,102 patients). We identify patterns of grade heterogeneity in WSIs not achievable through human pathologist analysis, and these graph-based "microheterogeneity" structures associate with PBRM1 loss of function and with patient outcomes. Joint analysis of tumor phenotypes and immune infiltration identifies a subpopulation of highly infiltrated, microheterogeneous tumors responsive to ICI. In paired multiplex immunofluorescence images of ccRCC, microheterogeneity associates with greater PD1 activation in CD8+ lymphocytes and increased tumor-immune interactions. Our work reveals spatially interacting tumor-immune structures underlying ccRCC biology that may also inform selective response to ICI.