Joyce A. O’Shaughnessy, MD
A wide-ranging analysis of more than 5500 breast cancer tumors that combined genomic and protein expression testing has identified promising targets to explore for treating patients with poor prognoses, with particularly notable findings involving androgen receptor (AR) expression,1
according to Joyce A. O’Shaughnessy, MD. Nearly 36% of the samples were from patients with triple-negative breast cancer (TNBC), making the study the largest genomic analysis in that disease setting to date.
The clues yielded through the analysis confirm the utility of multiplatform molecular profiling in identifying targets that could help oncologists direct patients who have metastatic disease, or who are not responding to standard therapies, into clinical trials, said O’Shaughnessy. In some instances, drugs aimed at these targets already have been approved.
O’Shaughnessy, who is chair of Breast Cancer Research at Baylor Charles A. Sammons Cancer Center, Texas Oncology and US Oncology, presented findings from the analysis at the 2013 San Antonio Breast Cancer Symposium (SABCS).1
She discussed her research in an interview in advance of the 31st Annual Miami Breast Cancer Conference (MBCC) in March, where she served as a faculty member. O’Shaughnessy’s expertise includes using emerging technologies to explore novel therapies, particularly for TNBC. Her presentation was entitled “Human Subject Implications of Genomic Profiling of Tumors.”
The clinical utility of next-generation sequencing for managing patients with high-risk or metastatic breast cancer continues to progress with the development of gene expression assays, and genomic profiling of newly diagnosed patients in these settings “will increasingly allow for early intervention with rational targeted therapies that could very significantly improve patients’ outcomes,” O’Shaughnessy said in her MBCC presentation.
Study Generates Trove of Data
In the study that she led, O’Shaughnessy and colleagues used a multifaceted tool developed by Caris Life Sciences, a Texas-based company. The Caris Molecular Intelligence platform examines tumors for gene mutations, protein expression, and/or gene amplification using next-generation sequencing technologies, immunohistochemistry (IHC), and in situ hybridization testing. Some 15 protein markers and 17 gene mutations were explored in the study, including the potential correlations and trends among the values.
The 5521 samples evaluated in the study were collected from 2009 to 2013. “About half of them were primary breast cancer tissues and half of them were metastatic breast cancer tissues, and they were obtained from patients who had a poor outcome,” said O’Shaughnessy.
Overall, 52.8% of the samples were either estrogen receptor (ER)- or progesterone receptor (PR)-positive and HER2-negative. One of the study’s primary goals was to distinguish molecular differences between TNBC and other breast cancers, and a total of 35.8% of the samples turned out to be TNBC.
AR Findings Compelling
Several of the study’s most striking findings came from the information yielded through the analyses of AR expression. “The androgen receptor story is emerging as a potentially very important one in ER-negative breast cancer in general, be it triple-negative or ER-negative/HER2-positive,” said O’Shaughnessy.
She said the patient populations in the study were large enough to generate data that could guide drug development efforts. Key AR findings include:
AR expression was measured in 50% of the ER-negative/HER2-positive samples and 18% of TNBC cases, suggesting it is prevalent enough to serve as a therapeutic target (IHC positivity threshold of 0+, or <10%, or ≥1+ and ≥10%).
Nearly all of the AR-positive samples displayed evidence of aberrations in the PI3K pathway through PIK3CA mutations, or PTEN mutations, or loss of expression.
Decreased proliferation as measured by Ki67 is associated with AR expression in TNBC but not ER-positive or HER2-positive disease.
O’Shaughnessy said the proliferation analysis has clinical implications. “About half of the triple- negative patients who have AR positivity have a low Ki67, a low cell division rate under 30%, and that is low for triple-negative,” she said. “There are a number of clinical trials going on right now in triple-negative breast cancer patients looking for those who have androgen receptor positivity, and in terms of finding them in the clinic we want to be alert that if they have a slower-growing cancer or Ki67 under 30% they are probably enriched for the androgen receptor.”
However, more highly proliferative TNBCs can also be AR-positive, noted O’Shaughnessy.