Integration of Digital Pathologic and Transcriptomic Analyses Connects Tumor-Infiltrating Lymphocyte Spatial Density With Clinical Response to BRAF Inhibitors

Abstract

Metastatic melanoma is one of the most immunogenic malignancies due to its high rate of mutations and neoantigen formation. Response to BRAF inhibitors (BRAFi) may be determined by intratumoral immune activation within melanoma metastases. To evaluate whether CD8+ T cell infiltration and distribution within melanoma metastases can predict clinical response to BRAFi, we developed a methodology to integrate immunohistochemistry with automated image analysis of CD8+ T cell position. CD8+ distribution patterns were correlated with gene expression data to identify and quantify “hot” areas within a tumor. Furthermore, the relative activation of CD8+cells, based on transcriptomic analysis, and their relationship to other CD8+ T cells and non-CD8+ cells within the tumor suggested a less crowded distribution of cells around activated CD8+ T cells. Furthermore, the relative activation of these CD8+ T cells was associated with improved clinical outcomes and decreased tumor cell proliferation. This study demonstrates the potential of digital pathomics to incorporate immune cell spatial distribution within metastases and RNAseq analysis to predict clinical response to BRAF inhibition in metastatic melanoma.