Association of C-MYC, MYC target gene, and unfolded protein response (UPR) expression with clinical benefit from the oral aurora kinase A (AURKA) inhibitor, alisertib (A), in combination with paclitaxel (P) compared with P alone in patients (Pts) with HER2-negative metastatic breast cancer (MBC).

Abstract

1037 Background: AURKA is a key regulator of the mitotic spindle, G2/M transition and epithelial-mesenchymal transition. AURKA is amplified and/or overexpressed in breast cancer and is associated with therapy resistance and worse survival. A randomized phase II trial in hormone receptor (HR)-positive, HER2-negative and triple negative (TN) MBC pts showed that addition of A to weekly P significantly improved progression-free survival (PFS) compared with P alone (O’Shaughnessy J et al. JAMA N etwork Open, 2021). Pts’ primary or metastatic disease tissues were analyzed for biomarkers associated with clinical benefit from A. Methods: Retrospective analysis of tumor whole exome and whole transcriptome sequencing of formalin-fixed paraffin embedded pre-treatment tissues from 96 pts (77 HR+, 19 TN) was performed. Clinical benefit from A+P or P was defined as having PFS of at least 6 mos and lack of benefit as PFS less than 6 mos. Enrichment for cancer gene mutations was assessed using Fisher's exact test. Transcriptome data were evaluated for differential expression with DeSeq2 and gene set enrichment analysis (GSEA) and compared to cancer hallmark gene sets. Molecular features were compared between A+P responders and non-responders, P responders and non-responders, and A+P and P responders. P-values <0.05 were considered significant. Results: PIK3CA, TP53, and CDH1 were altered in 41%, 39%, and 13% of tumors, respectively, and these and other known cancer genes were not associated with A+P benefit. Alterations in AKT1, PIK3CA, ERBB2, CDH1, TP53, and KMT2C/ KMT2D, as well as amplifications of CCND1, MYB, and MDM2, did not preclude benefit from A+P. RNA expression of AURKA, or its upstream regulator, FOXM1, was not significantly different based on treatment group or benefit from A+P or P. Increased C-MYC RNA expression (p=0.008), and enrichment for MYC targets and UPR by GSEA were enriched in pts with lack of benefit from P compared to those with benefit from P (p<0.05), but not in pts with lack of benefit from A+P. Wnt/Beta-catenin signaling was enriched in pts with lack of benefit from P compared to those with benefit from P (p=0.03), and with lack of benefit from A+P compared to those with benefit from A+P (p=0.086). Benefit from A+P was associated with enrichment for MYC targets and UPR compared with benefit from P (p=0.024), suggesting activity of A+P, but not P, with high MYC activation and UPR. Conclusions: A added to P improved PFS in HR+ HER2- and TN MBC pts compared with P alone. Pts whose breast cancers had increased C-MYC expression, high MYC activation and increased UPR on RNA expression derived greater clinical benefit from A+P than from P alone. Clinical trial information: NCT02187991 .