Abstract PS17-41: In depth single cell profiling of a case of bone metastases with associated organoid models reveals a precision medicine approach to treatment

Abstract

Abstract Occurring in 65-80% of metastatic breast cancer (BC), bone metastasis (BoM) is the major cause of BC related mortality. Under current standard of care, 5 year overall survival for patients with BoM is 20-25%, warranting the need of improved treatments. Like other cancers, BC BoM has high inter- and intra-patient heterogeneity. Thus, understanding disease evolution and heterogeneity of BoM at the individual patient level will be key to guide precise application of targeted therapies. In this study, we describe in depth histologic and single cell molecular characterization of a case of invasive lobular breast cancer (ILC) metastasis to bone. We collected fixed primary tumor (ER+/PR-/HER2- ILC) and fresh pelvis and tibia BoMs from a patient. H&E/IHC staining and whole exome sequencing (WES) of DNA and RNA of the samples were performed. Organoids were derived from the two BoMs and single cell RNA sequencing was applied to these cultures and the BoMs. Based on H&E/IHC staining, this case evolved from a ER+ primary ILC to ER- BoM with mixed lobular and ductal carcinoma features. WES revealed 208 missense somatic mutations in the BoMs. 15 were predicted as BC drivers, two of which are druggable including PIK3CA (E545K) and BRCA1 (D1834H, H399P). Notably, BRCA1 (D1834H) was not found in the primary tumor. Based on RNA sequencing, estrogen signaling was downregulated, while TGF-β, Wnt beta catenin and PI3K signaling, epithelial to mesenchymal transition (EMT) and angiogenesis were upregulated, all of which have been reported to be potential targets in BoMs. scRNAseq of BoMs revealed 5 major cell populations and pronounced heterogeneity. The two BoMs showed similar cellular compositions, including epithelial (30-50%), fibroblasts (50%), immune (5-10%), osteoclasts (1-2%) and endothelial cells (1-2%). Within epithelial cells, there were 5 major clusters exhibiting unique transcriptomic features, including a TNF-α signaling high cluster, two clusters showing high partial EMT (pEMT) signatures majorly regulated by PRRX1/2, TWIST1/2, and FOXS1, a cluster in active cell proliferation, and a cluster showing high signatures related to endocrine resistance. In fibroblasts, 3 clusters were identified representing ECM remodeling, angiogenesis (VEGFA high) and myofibroblasts. Immune cells majorly composed of macrophages, CD4+, CD8+ and Treg T cells. 2.2% tumor and 3.2% T cells express PDL1, exceeding the 1% threshold to select patients eligible for anti-PDL1 therapy. Based on CellPhoneDB, epithelial cells with high pEMT signature show the most interactions and uniquely interact with Treg cells through TNFSF4-TNFRSF4 which has been reported to promote Treg cell proliferation, and is investigated in clinic for immune therapy. Organoids were developed to evaluate therapeutic potential of targetable mutations, upregulated genes and susceptible cell populations. WES showed organoids preserved the mutational signatures of the matched tumor. scRNAseq showed all epithelial populations in the tumor were preserved in the matched organoids at comparable abundance. Consistent with the BRCA1 and PIK3CA mutations, organids were responsive to a PARP (Talazoparib: IC50 1.3uM/1.3uM) and PI3K (Alpelisib: IC50 9uM/4uM) inhibitor. We are currently examining the therapeutic potential of inhibiting pEMT, TNF-α, and genes upregulated in BoM. In summary, we have demonstrated a precision medicine approach to understand the evolution and heterogeneity of BC BoM. We have identified potential therapeutic targets and evaluated those in patient-specific organoids, thereby providing insights for the design of a precision medicine based clinical treatment strategy. Citation Format: Kai Ding, Fangyuan Chen, Nolan Priedigkeit, Daniel D. Brown, Tanya Heim, Rebecca Watters, Kurt Weiss, Jennifer Atkinson, Steffi Oesterreich, Adrian Lee. In depth single cell profiling of a case of bone metastases with associated organoid models reveals a precision medicine approach to treatment [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PS17-41.