ORF11-01 - Understanding Unique Patterns of Dissemination and Bone Remodeling Heterogeneity in Metastatic Lobular Breast Cancer

Invasive lobular carcinoma (ILC) is the most common special histological subtype of breast cancer after No Special Type (NST, ductal carcinoma), accounting for ~15% of cases. ILC is characterized by E-cadherin loss, Estrogen Receptor (ER) positivity, and a distinct single file invasive growth pattern which likely promotes ILC’s unique metastatic progression. Metastatic ILC (mILC) disseminates to lung and bone like NST, but uniquely metastasizes to the ovary, GI, and leptomeninges. Moreover, while NST bone metastases are predominantly osteolytic, ILC bone metastases show heterogeneous osteolytic and osteosclerotic activity even within a single lesion. While the implications of this heterogeneity are unclear, 80% of patients with mILC will develop bone metastases and those with bone-only disease have a poorer prognosis than similar NST patients, suggesting that the distinct bone remodeling phenotype of ILC may confound current standard-of-care therapies for breast cancer bone metastases.

Models of ILC metastasis are under-developed, and models of spontaneous metastasis from ER+ breast cancer are critically limited. To address this, we leveraged mammary intraductal (MIND) xenografting of two ER+ ILC cell lines (MM134,44PE) and confirmed prior reports that these lines spontaneously metastasize to sites associated with mILC. Importantly, we find drinking water supplementation with low dose estradiol (E2) advances metastatic progression and development of macro-metastatic disease, including clinically relevant ILC-specific sites like uterus/ovary and leptomeninges. Bone metastases were prevalent with both cell lines, which we further assessed by microCT, observing multi-focal, heterogeneous osteosclerotic/lytic lesions, similar to clinical mILC. We then isolated cells from metastatic lesions and established these as putative organotropic models. RNAseq shows bone lesion-derived cells maintain estrogen responsiveness but are less estrogen-dependent; we also observed remodeling of gene expression related to cell adhesion and intercellular signaling. Using conditioned media (CM) from these cells to differentiate osteoblasts (OB) and osteoclasts (OC), we observed increased OB differentiation and mineral deposition with suppressed OC differentiation with ILC CM, versus the opposite with NST CM, showing that ILC cells have a distinct impact on the bone microenvironment consistent with their in vivo bone remodeling phenotype.

Collectively, our data supports that ER+ ILC cell line metastases recapitulate key features of clinical mILC. With our models, we are investigating the ILC metastatic cascade, cellular adaptation to unique metastatic niches, and attributes of ILC metastases that can be exploited for therapeutic advancement.