Figure 2.

Identification of genomic features relevant for treatment in tumors. A, Transcriptional signatures were used to identify tumor tissues with increased MAPK signaling, proliferation, mTOR signaling, immune infiltration, and BRCAness. Of 61 patients, 14 did not have a gene signature associated with a response to an existing therapy. Analysis of DEGs between these patients and the remaining members of the cohort found an upregulation of ADH1B gene (B) that corresponds to worse patient prognosis, and pathway analysis revealed that these patients had increased DEGs associated with immune response pathways (C). We calculated whether MAPK (P = 0.188), mTOR (P = 0.504), and/or WNT (P < 0.001) signaling were correlated to T-cell infiltration (D). E, A more detailed analysis of specific immune cell types showed WNT signaling activity was inversely correlated with infiltration of cytotoxic cells (P < 0.01), NK cells (P < 0.001), and M1 macrophages (P < 0.05).