To Örebro University

INTRODUCTION: In light of the clinically meaningful results of the PI3K inhibitors in PIK3CA-mutated metastatic breast cancer (BC) patients, the reliable identification of PIK3CA mutations is of outmost importance. However, lack of evidence on the optimal site and timing of assessment, presence of temporal heterogeneity and analytical factors pose several challenges in clinical routine. We aimed to study the discordance rates of PIK3CA mutational status between primary and matched metastatic tumors.

METHODS: A systematic literature search was performed in three different databases (Embase, Pubmed, Web of Science) and-upon screening-a total of 25 studies reporting PIK3CA mutational status both on primary breast tumors and their matched metastases were included in this meta-analysis. The random-effects model was used for pooled analyses of discordance of PIK3CA mutational status.

RESULTS: The overall discordance rate of PIK3CA mutational status was 9.8% (95% CI, 7.0-13.0; n = 1425) and did not significantly differ within BC subtypes or metastatic sites. The change was bi-directional, more commonly observed from PIK3CA mutated to wild-type status (14.9%, 95% CI 11.8-18.2; n tumor pairs = 453) rather than the opposite direction (8.9%, 95% CI 6.1-12.1; n tumor pairs = 943).

CONCLUSIONS: Our results indicate the need of obtaining metastatic biopsies for PIK3CA-mutation analysis and the possibility of testing of the primary tumor, in case a re-biopsy deemed non-feasible.