Pan-cancer analysis distinguishes transcriptional changes of aneuploidy from proliferation

Patterns of gene expression differ strikingly between cancerous and normal tissues, and can arise as a consequence of or result in genetic instability characterized by the accumulation of somatic copy number alterations (SCNAs) and point mutations (PMs). SCNAs and PMs arise via different molecular processes and impose distinct stresses upon the cell, and hence accumulations of both variant classes have been presumed to associate with different gene expression patterns in cancer. Here we investigate this presumption by employing SCNA, exonic PM and transcriptome data from 2660 tumors representing 11 tumor types to investigate, in a tumor-type specific manner, gene expression changes that are associated with the accumulation of different classes of somatic genetic alteration. Notably, we find that both SCNAs and PMs are generally associated with similar expression changes, whereby expression changes reflecting cell growth and proliferation are associated positively, and immunoregulatory signaling negatively, indicating that general patterns of expression are not per se predictive for the acquisition of either PMs or SCNAs. Detailed characterization of expression changes identified genetic events affecting transcription factor signaling including TP53 and MYC specifically in highly aneuploid tumors. We additionally compared in vivo mouse breast tumorigenesis models for chromosome instability and demonstrated that cells undergoing CIN exhibit significant decrease in the expression of genes involved in cell growth and proliferation whereas an upregulation of immunity-related genes. Our results indicate that patterns of expression dysregulation of cell cycle related genes is a characteristic of tumor progression and extensive proliferation rather than genomic instability, and that independent trajectories for tumorigenesis converge on the same gene expression signatures to enforce increased rates of proliferation.