Transcriptomically-inferred PI3K activity and stemness show a counterintuitive correlation with PIK3CA genotype in breast cancer

Abstract

A PI3K-selective inhibitor has recently been approved for use in breast tumours harbouring mutations in PIK3CA, the gene encoding PI3K. Preclinical studies have suggested that the PI3K/AKT/mTORC1 signalling pathway influences stemness, a dedifferentiation-related cellular phenotype associated with aggressive cancer. No direct evidence for such a correlation has been demonstrated to date in human tumours. In two independent human breast cancer cohorts, encompassing nearly 3,000 tumour samples, transcriptional footprint-based analysis uncovered a positive linear association between transcriptionally-inferred PI3K signalling scores and stemness scores. Unexpectedly, stratification of tumours according to PIK3CA genotype revealed a "biphasic" relationship of mutant PIK3CA allele dosage with these scores. Relative to tumour samples without PIK3CA mutations, the presence of a single copy of a hotspot PIK3CA variant was associated with lower PI3K signalling and stemness scores, whereas tumours with multiple copies of PIK3CA hotspot mutations showed higher PI3K signalling and stemness scores. This observation was recapitulated in a human cell model of heterozygous and homozygous PIK3CAH1047R expression. Collectively, our analysis provides evidence for a signalling strength-dependent PI3K-stemness relationship in human breast cancer, which may aid future patient stratification for PI3K-targeted therapies.