Zhou, Maggie

Abstract BACKGROUND Brain metastases (BMs) are a significant cause of morbidity and mortality in metastatic breast cancer (MBC), yet prognostic and predictive biomarkers remain poorly understood. STUDY DESIGN 90 consecutive patients with breast cancer BMs who underwent surgical resection from 2006-2022 were retrospectively identified. Review of the electronic medical record and institutional cancer registry were completed to identify baseline clinical parameters and outcomes. BM mutation analysis was performed using a CLIA-certified targeted DNA mutation sequencing assay as part of routine clinical care. RESULTS All patients were female, median age at first BM resection was 57.9 years (range 27.0-81.3 years), and receptor subtypes included HR+/HER2- (n=27, 30.0%), HER2+ (n=33, 36.7%), and triple negative BC (TNBC; n=30, 33.3%). At time of first BM resection, most patients had a solitary BM (n=55, 61.1%), with anatomic locations including cerebellar (n=29, 32.2%), frontal (n=26, 28.9%), and parietal (n=17, 18.9%) sites. BM DNA mutation analysis (n= 20) revealed pathogenic alterations in TP53 (n=17, 85.0%), ERBB2 amplification (n=7, 35.0%), PIK3CA (n=4, 20.0%), BRCA1 (n=3, 15.0%), and BRCA2 (n=2, 10.0%). Most patients received postoperative radiation (n=82, 90.8%) and/or systemic therapy (n=67, 82.7%). Cerebellar location was significantly associated with the risk of developing leptomeningeal disease compared to other BM resection sites (HR = 2.75, p=0.02). Among all patients, median overall survival (OS) from first BM resection was 27.8 months (range 0.1-196.6 months). HR+/HER2- (median 37 months) or HER2+ (median 36 months) MBC was associated with significantly increased OS compared to TNBC (14.4 months) (p<0.01, log-rank test). CONCLUSIONS Patients with breast cancer BMs exhibited a median OS of >2 years following BM resection, and patients with TNBC had a significantly worse prognosis. Targetable mutations were observed in 40% of resected BM specimens, supporting the utility of molecular analysis of resected BMs to help guide clinical management.

Abstract Tissue-derived extracellular ATP (eATP) is a “danger signal” sensed by immune cells via purinergic receptors (e.g. P2RX7). In addition, healthy cells can export eATP through transmembrane channels, such as the hemichannel Pannexin-1 (Panx1). CD8+ T cells express Panx1 constitutively throughout developmental and immune response stages. However, the biological role of Panx1 in CD8+ T cells is undefined. Here, we used genetic and pharmacological tools to address this question. T-cell specific Panx1 deletion did not affect CD8+ T cell thymic development. In contrast, CD8+ T cell responses to viral infection (LCMV) are impaired by Panx1 inhibition or deletion. This is true for effector expansion of CD8+ T cells and for the survival of memory CD8+ T cells. Using adoptive transfer of tamoxifen-conditional Panx1 knockout LCMV-specific (P14) cells into LCMV-infected mice, we showed that memory CD8+ T cell survival requires sustained Panx1 – except for small intestinal-resident cells. Mechanistically, Panx1 promotes memory CD8+ T cell survival via P2RX7, which also favors memory CD8+ T cells - as suggested by our data with Panx1/P2RX7-double knockout P14 cells. The role of Panx1 for effector CD8+ T cell expansion, however, is unlikely to be via P2RX7 since this receptor does not have a role in effector CD8+ T cells. Our transcriptional and metabolic data suggests that, while memory CD8+ T cells require Panx1 for optimal mitochondrial function and expression of pro-memory genes, effector CD8+ T cells lacking Panx1 had decreased activation of the glycolytic pathway. Collectively, these results suggest that, in response to acute virus, Panx1 induces CD8+ T cell effector expansion and memory survival through distinct metabolic pathways. Supported by grants from NIH (R00 AI139381, NIAID).