Genomic Classifier for Luminal Androgen Receptor Triple-Negative Breast Cancer
TRIPLE-NEGATIVE breast cancer is a heterogeneous disease that comprises several subtypes, which may respond differently to therapy. Breast cancer researchers at the Mayo Clinic are developing a novel genomic signature that may improve the identification of the luminal androgen receptor vs basal (non–luminal androgen receptor) triple-negative breast cancer subtypes, which may better inform treatment selection. This research was presented as two posters at the 2018 San Antonio Breast Cancer Symposium.1.2
Roberto A. Leon-Ferre, MD
“Genomically distinct subtypes have been identified in triple-negative breast cancer, yet clinically triple-negative breast cancer is approached as a single entity. The most accurate way to identify the subtypes of triple-negative breast cancer is not clear,” said lead author Roberto A. Leon-Ferre, MD, Senior Associate Consultant in the Division of Medical Oncology at the Mayo Clinic, Rochester, Minnesota.
The novel genomic classifier, called CABAL (Clustering Among BAsal and Luminal androgen), comprises 426 differently expressed genes in luminal androgen receptor and non–luminal androgen receptor triple-negative breast cancers and was developed by Drs. Kevin Thompson and Krishna Rani Kalari. The senior author of the CABAL method, Krishna Rani Kalari, PhD, Associate Professor of Biomedical Informatics, and Dr. Leon-Ferre evaluated the CABAL signature in a large Mayo Clinic cohort of centrally confirmed triple-negative breast cancers, and found that luminal androgen receptor tumors behave differently from non–luminal androgen receptor triple-negative breast tumors. Further, the genomic signature appears to be more accurate than immunohistochemistry (IHC) staining to differentiate luminal androgen receptor from non–luminal androgen receptor triple-negative breast cancer.
Matthew Goetz, MD
“There is a lot of interest in the luminal androgen receptor subtype, which accounts for 20% of all triple-negative breast cancer,” said senior author Matthew Goetz, MD, also of the Mayo Clinic. “Our data suggest that this molecularly distinct subtype is also clinically distinct, with the median age nearly 15 years older comparing the luminal androgen receptor vs the non-luminal androgen receptor subtype.”
THE RESEARCHERS looked retrospectively at tumor samples from 9,982 women who had surgery for clinically HER2-negative or unknown breast cancer. They identified 605 cases of centrally confirmed triple-negative breast cancer. Among them, RNA-sequencing data and androgen receptor IHC staining was available in 213. The CABAL signature was used to classify the tumors as luminal androgen receptor or non–luminal androgen receptor.
Luminal androgen receptor tumors occurred in older, postmenopausal women compared with non–luminal androgen receptor tumors (median age 67 years vs 53; P < .001). The median androgen receptor IHC score was significantly higher with luminal androgen receptor than non–luminal androgen receptor tumors (70% vs 0%, respectively; P < .001). Luminal androgen receptor tumors were more often grade 1 or 2 (32% vs 6%, respectively; P < .001) and had lower Ki67 expression than non–luminal androgen receptor triple-negative breast cancer (P = .009). They also had higher levels of androgen receptor expression than non–luminal androgen receptor triple-negative breast cancer. Furthermore, patients with luminal androgen receptor triple-negative breast cancer had a lower recurrence rate during the first 3 years.
“These properties suggest that luminal androgen receptor tumors are not as immunogenic as non–luminal androgen receptor triple-negative breast cancers,” Dr. Leon-Ferre explained. “The interesting pattern of recurrence we saw suggests that luminal androgen receptor triple-negative breast cancer behaves more like estrogen receptor– positive breast cancer, particularly those with low Ki67 expression.”
“This research has therapeutic implications. Three clinical trials evaluating androgen receptor blockers in luminal androgen receptor tumors showed only modest results.… We think our genomic model may be better able to identify luminal androgen receptor tumors,” said Dr. Leon-Ferre.
“Ki67 may to be a promising marker within luminal androgen receptor triple-negative breast cancer to help identify patients at risk for earlier vs later disease recurrence,” Dr. Leon-Ferre stated. ■
DISCLOSURE: Dr. Leon-Ferre received travel funds from Immunomedics. Dr. Goetz reported financial relationships with Biotheranostics, Biovica, Eisai, Eli Lilly, Pfizer, Novartis, Contex Therapeutics, Sermonix Pharmaceuticals, Genomic Health, and Myriad Genetics.
1. Leon-Ferre RA, et al: Characteristics outcomes and prognostic factors of luminal androgen receptor triple-negative breast cancer. 2018 San Antonio Breast Cancer Symposium. Abstract P3-08-01. Presented December 6, 2018.
2. Thompson KJ, et al: Molecular subtyping of androgen receptor-positive patients using gene expression profiles. 2018 San Antonio Breast Cancer Symposium Abstract P1-03-04. Presented December 5, 2018.