A retrospective cohort study of 318 patients with breast cancer with leptomeningeal metastases was conducted to describe patient characteristics and outcomes. Independent factors associated with survival after the diagnosis of leptomeningeal metastases were identified. Such information is useful in management and in clinical trial development of patients with breast cancer with leptomeningeal metastases.Background:
Disease presentation, prognostic factors, and treatment patterns for patients with breast cancer with leptomeningeal metastasis are not well characterized. In this study, we examined patient characteristics and prognostic factors for survival after a diagnosis of leptomeningeal metastasis.Patients and Methods:
Three hundred eighteen consecutive patients with breast cancer diagnosed with leptomeningeal metastasis from January 1998 to December 2013 at Memorial Sloan Kettering Cancer Center were identified. Clinicopathologic and treatment information were obtained in a retrospective review. Associations with time from leptomeningeal diagnosis to death were evaluated according to Kaplan–Meier curves, log rank tests, and Cox proportional hazard models.Results:
Of the 318 patients, 44% were hormone receptor-positive (HR+) HER2−, 18% were HR+HER2+, 8.5% were HR−HER2+, 25.5% were triple-negative; and 4% had missing information. The median survival was 3.5 months (95% confidence interval, 3.0-4.0) with 63 patients (20%) surviving >1 year. Recent diagnosis (after 2006), HER2+ subtype, higher performance status, cranial-only involvement, and no evidence of noncentral nervous system disease were independently associated with improved survival in multivariate analysis.Conclusion:
Despite the improvement noted with the more recent years of diagnosis, survival after a diagnosis of leptomeningeal metastasis remains poor. Similar to patients with parenchymal brain metastasis only, the survival differs among different receptor subtypes. A closer examination to identify factors, such as introduction of new systemic therapies that might contribute to longer-term survival might provide insight to improve management of these patients. In addition, factors we identified that are associated with survival might be considered as stratification variables in the design of future randomized clinical trials in this population.