Development of a clinically relevant mouse model of ER+ breast cancer

The purpose of this proposed R21 project is to develop and validate a novel, clinically relevant mouse model for estrogen receptor-positive (ER+) breast cancer. Most human breast cancers are ER+ and are treated by endocrine therapy.

Current clinical challenges for treating ER+ breast cancer include: 1) some patients do not respond to or develop resistance to endocrine therapy; 2) some patients eventually progress to ER+ metastatic breast cancer.

Animal models that can faithfully recapitulate the biology of ER+ breast cancer are essential for addressing these challenges in a preclinical setting. However, current in vivo models are inadequate. It remains an unmet clinical need to more faithfully model ER+ breast cancer under the physiological setting.

The failure to model ER+ breast cancer faithfully in animals (e.g., mice) may be due to: 1) oncogenic events are not targeted to the correct cellular origin; 2) oncogenic mutations specifically associated with human ER+ breast tumors are rarely modeled in mice.

Taking these into consideration, we recently developed a new mouse model for ER+ breast cancer based on a recurrent genetic mutation uniquely found in human ER+ breast tumors, i.e., loss-of-function mutation or deletion of an estrogen/ER signaling- related transcription factor gene, RUNX1.

The model is based on intraductal injection of a Cre-expressing adenovirus (Ad-Cre) under the control of the Keratin 8 promoter (Ad-K8-Cre) to female mice carrying conditional knockout alleles of Runx1 and p53.

Ad-K8-Cre-induced loss of RUNX1 and p53 in K8+ luminal mammary epithelial cells (MECs) led to development of ER+ mammary tumors with hyperactive ribosome biogenesis and immune responses, which are also observed in human RUNX1-deficient ER+ breast cancers.

However, since Ad-K8-Cre targets both ER+ and ER- luminal MECs and the resulting tumors exhibited features of both ER+ and Claudin-low (i.e., tumor cells with mesenchymal-like appearance) cancers, it is unclear whether the mixed tumor phenotype is due to their different cellular origins (e.g., ER+ vs.

ER- luminal MECs) and/or due to p53-loss (which promotes cell fate plasticity).

We hypothesize that in order to make this model more clinically relevant to human ER+ breast cancer, it can be further improved by inducing RUNX1- loss and its cooperating oncogenic events (i.e., other than p53-loss) specifically in ER+ luminal MECs (i.e., the potential cellular origin of ER+ breast cancer).

To address this, two Specific Aims are proposed.

Aim 1 will focus on developing a new Ad-Cre tool to specifically target Cre expression to the ER+ luminal sublineage, so that mammary tumorigenesis is initiated only from ER+ luminal cells.

In Aim 2, we will test if oncogenic events affecting the RB-E2F pathway (as hyperactive ribosome biogenesis may lead to ribosomal stress, which may cause cell cycle arrest via this pathway) would cooperate with RUNX1-loss, leading to ER+ mammary tumors from luminal MECs.

The successful development of such model would provide an invaluable tool for better understanding the biology and therapy of ER+ breast cancer.