In a Prostate Tumor, Each Cancer Clone Has Unique Transcriptomic Topology of the Immune Response

Prostate tumor heterogeneity questions the value of the widely used biomarker paradigm for cancer diagnostic and therapy. We hypothesize that each histopathologically distinct tissue region has an unrepeatable dynamic transcriptome organization. The hypothesis was validated by quantifying the transcriptomic topologies and interplay of the toll-like receptors and chemokine signaling pathways in point biopsies of three cancer nodules and surrounding normal tissue from a surgically removed cancerous prostate. Raw expression data was collected from publicly accessible GEO datasets, but the theory and associated mathematical algorithms can be applied to any gene expression datasets of at least four biological replicas of each condition/tissue region to be compared. Our analysis revealed that the deep cancer-related topology remodeling is different even between equally graded cancer nodules within the same tumor, indicating distinct molecular mechanisms of the immune response. Moreover, cancer reorganized the transcriptomic networks interconnecting the two pathways via their shared genes, inducing novel expression coordination while removing others. We also found that genes associated with the immune response have low commanding heights, explaining the limited effectiveness of anti-prostate cancer immunotherapies in treating prostate cancer to date. Our results indicate the necessity to personalize the anti-cancer therapy beyond individual patient to his/her cancer nodules.