Date of Award

12-2023

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Program

Biomedical Sciences

Track

Cancer and Developmental Biology

Research Advisor

Ramesh Narayanan, PhD

Committee

Taosheng Chen, PhD; Evan Glazer, MD,PhD; Ronald N. Laribee, PhD; Lawrence Pfeffer, PhD

Keywords

Androgen receptor;Breast cancer;JAK/STAT;Oncogene;SARDs;Tumor suppressor

Abstract

Breast cancer is classified based on the expression of three markers estrogen receptor (ER), progesterone receptor, and human epidermal growth factor receptor (HER2). Breast cancers that express these targets are classified as hormone receptor (HR) positive breast cancer and the tumors that lack the expression of these markers are classified as triple negative breast cancer (TNBC). Though HR-positive breast cancer is treated with hormonal and cyclin-dependent kinase 4/6 targeted therapies, the cancer relapses and becomes resistant to the treatment options. Androgen receptor (AR) that is expressed in over 90\% of ER-positive breast cancer and in 20\% of TNBC is a pharmacologically targetable tumor suppressor in ER-positive breast cancer and a targetable oncogene in TNBC (luminal AR TNBC (LAR TNBC)). Here we show that the efficacy of AR agonists in ER-positive breast cancer depends on the cellular elasticity wherein AR agonists promote a more luminal state as a consequence of inhibiting oncogenic pathways. Sustained treatment of ER-positive breast cancer in vitro and in vivo with AR agonists results in rearrangement of AR, ER, and FOXA1 cistromes, transformation of tumor suppressor AR into oncogenic AR, and activation of the JAK/STAT pathway, all culminating in lineage plasticity and resistance. This unique shift of AR into an oncogene (previously unreported for the tumor suppressor class) and increase in JAK/STAT signaling switches the tumor cells from luminal ER-positive breast cancer to an intermediate ER/AR-positive basal subtype, whose growth was inhibited by AR antagonists or JAK/STAT inhibitor. Treatment with a JAK inhibitor also re-sensitizes the resistant cells to AR agonists.

While we show that AR can shift from tumor suppressor to an oncogene in ER-positive breast cancer, in TNBC we identified that the tumors not only express full-length AR (AR-FL) at a much higher rate (~80\%) than previously reported, but also express AR splice variants (AR-SVs) (~20\%), subclassifying the LAR-TNBC subtype into LAR-FL, LAR-AR-SV, or dual positive. Higher AR and AR-SV expression and corresponding aggressive phenotypes were observed predominantly in specimens obtained from African American women. RNA sequencing of LAR TNBC specimens indicates an enrichment of hallmark interferon, JAK/STAT, and androgen signaling pathways, which were exclusive to AR-expressing epithelial cancer cells as demonstrated by spatial genomics. LAR and LAR-AR-SV TNBC cell proliferation, orthotopic cell line and patient-derived xenograft, and patient-derived tumor explants growth were inhibited by AR N-terminal domain (NTD)-binding selective AR degrader (SARD) that irreversibly inhibits AR or by the JAK inhibitor ruxolitinib. Subsequent biochemical analysis suggests that STAT1 is an AR coactivator, and SARD or ruxolitinib blocks this coactivation.

Collectively, we identified that AR has multifaceted role in breast cancer and co-expression of JAK/STAT pathway defines the function of AR as a tumor suppressor or oncogene, providing a mechanism of action for AR in breast cancer and also a roadmap for therapeutic development targeting the two important pathways.

Declaration of Authorship

Declaration of Authorship is included in the supplemental files.

ORCID

0000-0003-0720-2222

DOI

10.21007/etd.cghs.2023.0634

Additional Files
2023-022-Asemota-DOA.pdf (116 kB)
Download

Available for download on Monday, August 11, 2025

Share

COinS