Abstract 5559: Exploratory analysis of hormone receptor-positive (HR+), human epidermal growth factor receptor 2-negative (HER2-) PIK3CA mutated (mut) advanced breast cancer (aBC) treated with taselisib (TAS) and/or fulvestrant (FUL) reveals increased CCNE2 expression with poor outcome to placebo (PBO) + FUL but not TAS + FUL in SANDPIPER

Abstract

Abstract PIK3CA mutations occur in ~40% of patients (pts) with HR+/HER2- aBC. TAS is a potent and selective β-sparing PI3K inhibitor that has shown modest clinical benefit when combined with FUL endocrine therapy in patients (pts) with PIK3CAmut aBC. To better understand the molecular features associated with response to TAS and/or FUL, we examined the gene expression profiles of archival tissue from pts enrolled in SANDPIPER (NCT02340221). Most patients did not have prior treatment with a CDK4/6i. RNA sequencing was performed on baseline tissue samples from 456 pts with PIK3CAmut tumors. Exploratory biomarker analyses revealed 28 genes associated with worse outcome. Of these, CCNE2 [HR=1.11 (1.04-1.19), p=0.002], RRM2 [HR=1.14 (1.07-1.22), p<0.0001] and TUB4A [HR=1.15 (1.05-1.26), p=0.004] are involved in the regulation of mitotic cell cycle. CCNE2 encodes CYCLIN E2, a G1 cyclin that binds and activates CDK2. Elevated expression of CCNE2 has been associated with poor outcome in ER-positive breast cancer, and has been identified as part of gene signatures that predict disease progression in tamoxifen-resistant breast cancer or advanced breast cancer. Since CCNE2 expression may be a marker of more proliferative tumors, we assessed whether the inclusion of CCNE2 gene expression would add value to a prognostic model that already includes cell cycle as a factor. Analysis of deviance after sequential addition of variables into Coxph models with stratification factors showed that elevated CCNE2 expression is significant, in addition to cell cycle signature scores, for association with worse PFS (likelihood ratio test p = 0.001). When we grouped patients based on CCNE2 expression split by quartile with the highest values (Q4) grouped as CCNE2 high vs lower values (Q1-Q3), we found that the CCNE2 high group does worse in the PBO + FUL treatment arm. Median PFS for CCNE2-high group in the placebo arm was 3.75 months compared to the other groups (log-rank p < 0.0001; PBO, CCNE2-low 7.3 months, TAS, CCNE2-high 7.6 months; TAS, CCNE2-low 7.3 months). A significant statistical interaction was observed between high CCNE2 gene expression and PBO + FUL treatment arm HR=1.51 (1.03-2.20), p=0.04], suggesting that CCNE2 expression may be associated with poor PFS to in 2L ER+/HER2- PIK3CAmut aBC patients that receive PBO + FUL. Our analysis revealed an association between high CCNE2 expression and worse PFS in patients that received PBO + FUL. Further, this association was not observed in the patients that received the combination of TAS+FUL, which may inform future rational combination strategies for the clinical development of PI3K inhibitors in ER+/HER2- PIK3CAmut aBC. Citation Format: Radia M. Johnson, Bonnie Liu, Javier Cortés, Susan Dent, Nadia Harbeck, Ian E. Krop, William Jacot, Robert L. Yauch, Thomas J. Stout, Frauke Schimmoller, Steven Gendreau, Timothy R. Wilson. Exploratory analysis of hormone receptor-positive (HR+), human epidermal growth factor receptor 2-negative (HER2-) PIK3CA mutated (mut) advanced breast cancer (aBC) treated with taselisib (TAS) and/or fulvestrant (FUL) reveals increased CCNE2 expression with poor outcome to placebo (PBO) + FUL but not TAS + FUL in SANDPIPER. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 5559.