Prognostic Risk Model for African American Women With Breast Cancer
A prognostic model developed using a machine learning approach may be able to identify African American patients with breast cancer who have an increased risk of death, according to results of a study presented by Bhattarai et al at the 11th AACR Conference on The Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved (Abstract C101).
“Using gene-expression data, we have developed a machine learning pattern to accurately stratify African American [patients with] breast cancer with high and low risks of death, which could help inform clinical decision-making,” said Shristi Bhattarai, PhD candidate in the lab of Ritu Aneja, PhD, in the Department of Biology, Georgia State University. “As African American women tend to have worse breast cancer outcomes, this study will help us to identify race-based differences in this cohort, which could potentially lead to specific therapeutic regimens for African American women with breast cancer.”
Although the incidence of breast cancer is similar between European American and African American women in the United States, the age-adjusted mortality rates are 40% higher in African American women with breast cancer, said Ms. Bhattarai. “The etiology of this … outcome disparity is multifactorial, arising from the combination of socioeconomic inequality with inherently more aggressive tumor biology in women of African ancestry,” she noted. “We wanted to identify a fingerprint that could stratify African American [patients with] breast cancer with different prognostic risks.”
Utilizing data from The Cancer Proteome Atlas, Ms. Bhattarai and colleagues analyzed protein-expression levels of 224 proteins in 754 patients. Of these patients, 620 were of European descent and 134 were African American. The algorithm they developed enabled the researchers to identify significant protein combinations that were associated with breast cancer survival.
Study Findings
The deep-learning algorithm identified a combination of four proteins for optimal prognostic prediction: Bcl2-like protein (BAX); inositol polyphosphate-4-phosphatase, type II (INPP4B); x-ray repair cross-complementing protein 1 (XRCC1); and cleaved poly (ADP-ribose) polymerase (c-PARP). This combination of proteins could stratify high-risk African American patients with breast cancer with 86% accuracy (hazard ratio [HR] = 5.0, P < .001). The model retained its significant prognostic ability (HR = 10.741, P = .0006) when controlling for clinicopathologic variables like stage, age, and positive lymph nodes.
“Interestingly, these proteins did not have a significant prognostic value individually,” said coauthor Sergey Klimov, PhD candidate in the Aneja lab at Georgia State University. “However, their combined effect within the machine-learning model could identify an African American cohort that had five times increased risk of death.”
The researchers were not able to stratify European American patients with breast cancer into low- and high-risk populations using this specific model, suggesting that this model is only prognostic for African American patients.
“We are moving toward the phase of clinical research where we can identify very specific patterns for understudied demographic groups to find high-risk patients so that they can be recruited for additional therapies,” said Dr. Aneja. “We are excited that our model has the potential to inform clinicians to prioritize African American [patients with] breast cancer for appropriate clinical trials and also help patients make decisions about enrolling in specific clinical trials.”
Limitations of this study include a lack of validation in other cohorts. “We want to make sure that this model is generalizable to different methodologies,” concluded Dr. Aneja.
This study was funded by grants from the National Cancer Institute and the National Institutes of Health. Dr. Aneja, Ms. Bhattarai, and Mr. Klimov declared no conflict of interest.
The content in this post has not been reviewed by the American Society of Clinical Oncology, Inc. (ASCO®) and does not necessarily reflect the ideas and opinions of ASCO®.
