Abstract PS14-19: Characteristics of breast cancer brain metastases presentation by subtype and validation of the modified breast graded prognostic assessment

Abstract

Abstract Purpose: Breast cancer brain metastases (BCBM) diagnosis is increasing in frequency due to improved systemic control and imaging techniques. Differences have been noted in rates of central nervous system (CNS) relapse and biologic subtype. The modified breast graded prognostic assessment (breast-GPA) was initially validated in patients treated between 1996-2013 and considers biologic subtype. In this study, we characterize patients diagnosed with BCBM by subtype and validate the breast-GPA in a modern cohort of patients. Methods: All patients with BCBM treated at our institution with radiotherapy between 2016 and 2019 were identified. Characteristics of patients’ initial brain metastasis diagnosis were retrieved from the clinical chart and radiologic examinations. To test differences between cohorts, the Kruskal-Wallis and Pearson’s chi-square tests were used when appropriate. Overall survival (OS) was calculated from the date of brain metastasis diagnosis to the date of death using the Kaplan-Meier (KM) method, with the log-rank test used to examine differences between groups. Results: A total of 122 BCBM patients were identified. Breast cancer subtypes included hormone receptor (HR)+/HER2- (45%), triple negative (TN) (25%), HR-/HER2+ (16%), and HR+/HER2+ (14%). The first treatment for BCBM patients following diagnosis was whole brain radiation (51%), surgery followed by stereotactic radiation (28%), and stereotactic radiation (21%). The interval between breast cancer diagnosis and diagnosis of BCBM was longest for HR+/HER2- 4.5 years, followed by TN 2.8 years, HR+/HER2+ 2.3 years, HR-/HER2+ 1.9 years, p=0.003. The interval from systemic metastases to BCBM diagnosis trended towards the shortest for TN patients 6.6 months, p=0.15. A total of 34 patients (28%) were diagnosed with leptomeningeal disease (LMD) at initial brain metastases presentation. LMD was diagnosed most commonly at presentation in HR+/HER2- (36%) followed by , TN (26%), HR-/HER2+ (26%), and HR+/HER2+ (6%), p=0.06. No differences were noted based on receptor typessubtype and age, symptomatic intracranial disease, number of brain metastases, type of first intracranial treatment or concurrent systemic metastases at initial BCBM presentation, all p > 0.05. Twenty-four month KM OS rates following diagnosis of brain metastasis for breast-GPA 0-1, 1.5-2, 2.5-3, and 3.5-4 groups were 14%, 27%, 33%, and 86% (p=0.0005), respectively. Conclusions: In our institutional analysis, similarities were noted in the initial presentation of BCBM based on receptor typesubtype. Significant differences were noted in OS based on the modified breast-GPA. Further investigation is needed to determine which subtypes of asymptomatic breast cancer patients are at sufficient risk to warrant brain MRI screening. Citation Format: Yuki Kawahara, Matthew Fahey, Thrisha K Potluri, Matthew N Mills, Nicholas B Figura, Iman R Washington, Roberto Diaz, Timothy J Robinson, Hsiang-Hsuan M Yu, Arnold B Etame, James Liu, Michael A Vogelbaum, Brian J Czerniecki, Peter A Forsyth, Hatem H Soliman, Hyo S Han, Kamran A Ahmed. Characteristics of breast cancer brain metastases presentation by subtype and validation of the modified breast graded prognostic assessment [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PS14-19.