PIK3CA is the most frequently mutated oncogene (13%) across all human cancers. PIK3CA codes for p110?, the catalytic subunit of phosphoinositide-3-kinase (PI3K), and ?hotspots? mutations in this gene are known to hyperactivate PI3K and drive oncogenicity. Specific p110? PI3K inhibitors are currently in late clinical development and show higher activity in PIK3CA-mutant tumors. However, while some patients show marked and durable responses to PI3K inhibitors, others are intrinsically resistant to therapy. Interestingly, we observed that among the best responders there is an enrichment of patients with tumors bearing two PIK3CA mutations. Using next-generation sequencing (NGS), we sequenced over one thousand breast cancer patients and identified 337 samples with PIK3CA mutations, 16% of which have two or more PIK3CA mutations. Dual mutations in any gene can be either compound mutations (on the same allele) or ?in trans? mutations (on different alleles). We have identified that patients with breast cancer and other cancers with two PIK3CA mutations in close proximity to each other are compound mutations. We will use exome sequencing, cDNA sequencing and single molecule real time (SMRT) sequencing of tumor DNA from fresh frozen tumor samples from breast cancer patients enrolled in clinical trials testing the activity of PI3K inhibitors. We will investigate whether also the more frequent dual PIK3CA-mutations that are far from each other in the gene are in cis. We hypothesize that dual compound PIK3CA mutations cause an increased oncogenic phenotype, increased biochemical PI3K activity, and, consequently, increased sensitivity to PI3K inhibitors. We will use isogenic stable cell lines and patient-derived models to elucidate the transforming capacity of the most frequent dual PIK3CA mutants and their response to PI3K inhibition. The same mutants will be expressed in insect cells and the resulting recombinant mutant PI3K protein complexes will be used to measure their lipid binding capacity and kinase activity in unperturbed conditions or in the presence of specific PI3K inhibitors. Almost all additional mutations rising in mutated driver oncogenes confer acquired resistance to inhibitors targeting these kinases. Dual PIK3CA mutations, however, may confer primary sensitivity to PI3K inhibitors as these genomic aberrations are early events detected in the tumors before exposure to these agents. These findings may define a new paradigm in cancer biology and provide the rationale to design ad hoc clinical trials testing the activity of PI3K p110?-specific inhibitors in breast cancer patients bearing tumors with dual compound PIK3CA mutations. Importantly, these findings may also be expanded to other PIK3CA-mutated cancers (e.g. endometrial, cervical, anal, bladder) and broaden the number of patients that can potentiually benefit from pharmacological PI3K inhibition.
Mutations in the PIK3CA gene lead to hyperactivation of the PI3K pathway and occur frequently in breast cancer and other tumors. We found that in a significant proportion of cases, these tumors bear two mutations in PIK3CA, and this occurrence is more frequent in patients that respond clinically to PI3K inhibitors. In this proposal, we aim to elucidate whether the presence of two distinct mutations in PIK3CA results in increase activity of PI3K, which would translate in increase dependency on this pathway of the tumor cells and exquisite sensitivity to pharmacological PI3K inhibition.