PIK3CA mutations in advanced or recurrent endometrial cancer: an exploratory molecular profiling from the phase II MITO END-3 study. A pre-planned translational molecular analysis on PIK3CA mutational status.

Objective: Immunotherapy with chemotherapy significantly improves progression-free survival in advanced endometrial cancer (EC) especially in microsatellite-instability high (MSI-H) cases. Despite this, for recurrent EC in the post-immunotherapy setting, new treatment options are urgently awaited. The PI3K pathway is frequently altered in gynecological cancers in EC representing a potential treatment target. Patients and methods: In a pre-planned translational analysis of the MITO END-3, we retrospectively analyzed the genomic abnormalities of 107 pts with samples eligible for Next Generation Sequencing (NGS) analysis (FoundationOne CDx®) focusing on PIK3CA mutational status. Results: In our analysis, the 4 most frequently mutated genes were TP53, PIK3CA, ARID1A, PTEN. PIK3CA mutations were detected in 51.4% (55/107 pts) of tissue samples. A total of 68 different PIK3CA mutations were observe, of which 64 were considered pathogenic and 4 variants of uncertain significance, with the co-expression of 2 different pathogenic mutations in 2 pts. The most frequent PIK3CA mutations were found in the kinase domain most commonly H1047R, (11/68), H1047Y (3/68), and in the helical domain such as E545K with a frequency of 4/68, and E542K with a frequency of 3/68. Notable, 24 different PIK3CA mutations out of 68 reported in MITO END-3 are considered as predictive of response to PI3K inhibition with alpelisib by the current literature in breast cancer. Interestingly, PIK3CA mutations were observed in all the TCGA categories. Conclusion: In this post-hoc analysis from the MITO END 3 trial, we confirmed the high frequency of PIK3CA mutations in EC establishing this tumor as an ideal candidate for testing of PI3K inhibitors in clinical trials.