Genomic and immune profiling of breast cancer brain metastases.

BACKGROUND: Brain metastases (BrM) arising from breast cancer (BC) are an increasing consequence of advanced disease, with up to half of patients with metastatic HER2 + or triple negative BC experiencing central nervous system (CNS) recurrence. The genomic alterations driving CNS recurrence, along with contributions of the immune microenvironment, particularly by intrinsic subtype, remain unclear. METHODS: We characterized the genomic and immune landscape of BCBrM from a cohort of 42 patients by sequencing whole-exome DNA (WES) and total RNA libraries from frozen and FFPE BrM and FFPE extracranial tumors (ECT). Analyses included PAM50 intrinsic subtypes, somatic mutations, copy number variations (CNV), pathway alterations, immune cell type deconvolution, and associations with clinical outcomes RESULTS: Intrinsic subtype calls were concordant for the majority of BrM-ECT pairs (60%). Across all BrM and ECT samples, the most common somatic gene mutation was TP53 (64%, 30/47). For patients with matched FFPE BrM-FFPE ECT, alterations tended to be conserved across tissue type, although differential somatic mutations and CNV in specific genes were observed. Several genomic pathways were differentially expressed between patient-matched BrM-ECT; MYC targets, DNA damage repair, cholesterol homeostasis, and oxidative phosphorylation were higher in BrM, while immune-related pathways were lower in BrM. Deconvolution of immune populations between BrM-ECT demonstrated activated dendritic cell populations were higher in BrM compared to ECT. Increased expression of several oncogenic preselected pathways in BrM were associated with inferior survival, including DNA damage repair, inflammatory response, and oxidative phosphorylation CONCLUSIONS: Collectively, this study illustrates that while some genomic alterations are shared between BrM and ECT, there are also unique aspects of BrM including somatic mutations, CNV, pathway alterations, and immune landscape. A deeper understanding of differences inherent to BrM will contribute to the development of BrM-tailored therapeutic strategies. Additional analyses are warranted in larger cohorts, particularly with additional matched BrM-ECT.

Duke Scholars

Author Carey Kernodle Anders Medicine, Medical Oncology

Author Amanda Van Swearingen  

Author Kouros Owzar Biostatistics & Bioinformatics, Division of Integrative Geno ...