Carey Anders

Brain metastasis arising from breast cancer is associated with a poor prognosis. Various systemic chemotherapy and targeted therapies which are effective against breast cancer often fail to provide benefits against brain metastasis. This is mainly due to limited penetration of the therapies across the blood-brain barrier, and divergent evolution of brain metastasis compared to the primary tumor. Thus, brain metastasis is typically treated upfront with local therapies, such as surgery and radiation, followed by systemic therapies. Systemic therapies with CNS permeability are favored in patients with brain metastasis. This paper reviews various systemic therapy options for breast cancer brain metastasis.

PURPOSE: Brain metastases (BM) are a complication of advanced breast cancer (BC). Histology of melanoma BM offers prognostic value; however, understanding the microenvironment of breast cancer brain metastases (BCBM) is less characterized. This study reports on four histological biomarkers, gliosis, immune infiltrate, hemorrhage, necrosis, and their prognostic significance in BCBM. METHODS: A biobank of 203 human tissues from patients who underwent craniotomy for BCBM was created across four academic institutions. Degree of gliosis, immune infiltrate, hemorrhage, and necrosis were identified and scored via representative H&E stain (0-3+). Overall survival (OS) was estimated using the Kaplan-Meier method. Cox proportional hazards regression evaluated prognostic value of the biomarkers in the context of standard clinical characteristics. RESULTS: BCBM subtype (available for n = 158) was 36% Her2+, 26% hormone receptor (HR)+/Her2- 38% HR-/Her2- (triple negative, TN). Gliosis was observed in 82% (116/141) of BCBM, with immune infiltrate 44% (90/201), hemorrhage 82% (166/141), and necrosis 87% (176/201). Necrosis was significantly higher in TNBC (p < 0.01). Presence of gliosis, immune infiltrate, and hemorrhage correlated with improved OS (p = 0.03, p = 0.03, p = 0.1), while necrosis correlated with inferior OS (p = 0.01). Improved OS was associated with gliosis in TN (p = 0.02), and immune infiltrate (p = 0.001) and hemorrhage (p = 0.07) in HER2+. In a multivariable model for OS, incorporating these biomarkers with traditional clinical variables improved the model fit (p < 0.001). CONCLUSION: Gliosis confers superior prognosis in TNBC BM; immune infiltrate and hemorrhage correlate with superior prognosis in HER2+ BCBM. Understanding the metastatic microenvironment of BCBM refines prognostic considerations and may unveil novel therapeutic strategies.