Leptomeningeal Disease in Breast Cancer: Pre-Treatment Prognostic Factors and Outcomes

Abstract

Leptomeningeal disease (LMD) is a devastating diagnosis with limited treatment options and poor prognosis. With recent advances in treatment, women with breast cancer (BC) are living longer and the incidence of LMD in this population is increasing. There remains little data regarding the prognosis and optimal treatment of these patients. As such, we conducted a multi-institutional retrospective review of women with BC LMD and analyzed outcomes based on prognostic factors and treatments.We analyzed the records of women diagnosed with BC LMD via MRIs of their brain and/or spine between 2000 and 2020 at two academic medical centers. Demographic and clinical characteristics were abstracted from electronic medical records and an in-depth radiographic review was performed by a neuro-radiologist for each case. Univariate and multivariate Cox proportional-hazards models were generated to identify factors associated with prolonged survival. Collection and analysis of the data was approved by institutional IRB.We identified 227 women with BC LMD. Prior to developing LMD, most women initially presented with Stage I-III BC (62.1%) with ductal histology (67.0%) and biological subtypes including ER-positive/HER2-negative (56.8%), HER2-positive (17.6%), and triple-negative (22.0%). LMD was first diagnosed a median of 4.6 years (range: 0 to 34.1 years) after the initial diagnosis of BC, and the median age at LMD diagnosis was 54.6 years (range: 29.4 to 84.5 years). At time of LMD diagnosis, 146 (64.3%) women presented with a Karnofsky Performance Status (KPS) ≥ 70, 90 (39.6%) had a preceding parenchymal brain metastasis, and 23 (10.1%) had LMD as their only site of metastatic disease. Radiographically, classical LMD (61.7%) was observed more frequently than nodular-only LMD (13.2%). Brain-only LMD was present in 55.1% of patients, spine-only in 6.1% of patients, and both brain and spine in 38.8% of patients. Median survival from the time of LMD diagnosis was 4.5 months (95% CI: 3.4-7.8). Pre-treatment clinical factors associated with improved survival on univariate analysis included ER-positive/HER2-negative subtype (P = 0.010), KPS ≥ 70 (P < 0.001), LMD as the only site of metastasis (P = 0.004), stable or controlled extracranial metastases (P = 0.042), absence of leukopenia (P = 0.040) and the presence of nodular LMD (P < 0.001). After adjusting for these clinical factors, the receipt of radiation therapy (adjusted hazard ratio [AHR] 0.50, 95% CI: 0.37 - 0.68), systemic therapy (AHR 0.27, 95% CI: 0.19 - 0.38) and intrathecal chemotherapy (AHR 0.49, 95% CI: 0.33 - 0.75) were all associated with improved survival.In one of the largest multi-institutional reviews to date, we found that several pre-treatment clinical factors were significantly associated with improved survival among women with breast cancer LMD. However, despite treatment overall survival remains poor, and there is a need for new, effective therapies.M.G. Milligan: None. A. Aitelli: None. W.A. Mehan: Consultant; Kura Oncology. D.D. Shi: None. D.N. Cagney: Research Grant; Viewray, NhTheraguix. K.S. Oh: None. N. Wang: None. P. Brastianos: Research Grant; Merck, Eli Lily, BMS. Honoraria; Merck, Genentech. Consultant; Angiochem, Dantari Pharmaceuticals, Elevatebio, Eli Lily, Genentech, Pfizer, SK Life Sciences, Tesaro. B. Moy: Research Grant; PUMA biotechnology. N.U. Lin: Research Grant; Genentech, Merck, Pfizer, Seattle Genetics. Consultant; PUMA Biotechnology, Seattle Genetics, Daiichi Sankyo, Astra Zeneca, Denali Therapeutics, California Institute for Regenerative Medicine, Prelude Therapeutics. H.A. Shih: Employee; Dartmouth Hitchcock. Research Grant; AbbVie, NIH. Honoraria; UpToDate. Consultant; Cleveland Clinic. Speaker's Bureau; prIME Oncology. advisory; The Radiosurgery Society. director of clinical operations; Massachusetts General Hospital. clinical operational leader; Massachusetts General Hospital.