Abstract P4-01-09: Characterization of breast cancer brain metastases overall and progression-free survival and timing of cyclin-dependent kinase 4/6 inhibitor use: Retrospective single institution experience

Abstract

Abstract Background Approximately 15% of patients (pts) with hormone receptor (HR)+ human epidermal growth factor receptor 2 (HER2)- breast cancer will develop brain metastases (BM) (Kuksis et al, NeuroOnc 2021). Cyclin dependent kinase 4/6 inhibitors (CDK4/6i) with an endocrine therapy partner are recommended 1st line treatments in HR+HER2- metastatic breast cancer (MBC). Preclinical models show that CDK4/6i can cross the blood brain barrier (BBB). In vitro assays have shown that abemaciclib crosses the BBB more effectively than palbociclib or ribociclib (George et al, Front Oncol 2021). The efficacy of CDK4/6i in patients with breast cancer BM is not well described. Methods We examined prior treatment data for 368 pts with HR+HER2- BM who received a CDK4/6i between 2015 to 2021. The primary endpoint was overall survival (OS) from the time of starting CDK4/6i after BM development. CNS progression free survival (PFS) was assessed in pts who received CDK4/6i after BM development. We examined the relationship between OS, type and timing of CDK4/6i in multivariate analyses. Statistical analyses were conducted using R 4.1.2 software. Results Of the total cohort of 368 pts, 23% (n=86) had de novo MBC and 77% (n=282) had relapsed MBC. At initial presentation of MBC 79% (n=290) of pts had no BM, 19% (n=71) had BM and extracranial disease and 2% (n=7) had BM only. 56% (n=205) received a CDK4/6i before BM development, 37% (n=136) received a CDK4/6i after BM development and 7% (n=27) received a CDK 4/6i both before and after BM development. The most common CDK4/6i used first was palbociclib (85%, n=312) followed by abemaciclib (13%, n=47) and ribociclib (2%, n=9). At the time of data cutoff 277 pts were dead, 55 were alive and 36 were lost to follow up. The median follow-up of surviving pts from BM development was 32 months. 163 pts received a CDK4/6i post BM: palbociclib (66%, n=108), abemaciclib (31%, n=51) and ribociclib (3%, n=4). Of these 163 pts 83% (n=136) received a CDK4/6i as their 1st or 2nd systemic treatment post BM. In the cohort of 163 pts who received a CDK4/6i post BM, 9% did not receive local BM therapy, 40% had whole brain radiotherapy (WBRT), 34% had stereotactic surgery (SRS) alone and 17% had surgery +/- SRS. The median CNS PFS for pts who received a CDK4/6i after BM was 21 months with palbociclib and 14 months with abemaciclib (p value 0.11). Too few pts received ribociclib for analysis. CNS PFS was 21 months for pts receiving a CDK4/6i only after BM development and 10 months for those who received CDK4/6i both before and after BM diagnosis (p = 0.01). Pts who died prior to CNS progression were censored. Median OS from the time of starting CDK4/6i after BM development for pts receiving a CDK4/6i only after BM development was 25 months versus 12 months for those who received it both before and after BM development (p = 0.03). There was no statistically significant difference in OS when adjusting for the type of local BM treatment received, time from initial MBC diagnosis to BM, or the Breast Graded Prognostic Assessment (GPA) score (Sperduto et al, IJROBP 2020). Conclusions This observation suggests that there is a greater OS benefit from the time of starting CDK4/6i after BM development in pts who receive CDK4/6i solely after BM development compared to pts who received a CDK4/6i both before and after BM development. This is not unexpected given the known OS benefit associated with early use of these agents. Our unique observation of longer CNS PFS for patients who did not receive CDK4/6i prior to BM but who received it afterward suggests that CDK4/6i exposure prior to BM development may lead to development of resistance mechanisms that reduces CNS efficacy upon rechallenging with CDK4/6i after BM development. There was no statistically significant difference in post-BM CNS PFS by type of CDK4/6i received. This motivates investigation of biomarkers for patient selection and our ongoing work in collecting a matched comparison cohort of HR+HER2- pts with BM who never received CDK4/6i. Citation Format: Sonya Chew Minmin, Yuan Chen, Daniel Kelly, Mark E. Robson, Andrew D Seidman. Characterization of breast cancer brain metastases overall and progression-free survival and timing of cyclin-dependent kinase 4/6 inhibitor use: Retrospective single institution experience [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr P4-01-09.