Most breast cancer risk prediction models have been developed and validated primarily in data from white women. These models have underperformed in black women, researchers said in a study in the Journal of Clinical Oncology.
“Lack of a breast cancer risk prediction model tailored to Black women represents a critical gap,” wrote Julie Palmer, ScD, director of the Slone Epidemiology Center at Boston University School of Medicine, and colleagues. “US Black women have, on average, earlier ages at diagnosis than US white women and are more likely to be diagnosed with poor-prognosis breast cancers. Many young Black women are diagnosed with and die from breast cancer before they even reach the ages at which mammographic screening is typically recommended.”
Palmer’s team developed and externally validated two absolute risk prediction models for breast cancer in Black women — one that considered all breast cancers together and another that incorporated heterogeneity by estrogen receptor (ER) status.
The overall model, derived from 3,468 cases and 3,578 controls in three studies and validated with prospective follow-up data from the Black Women’s Health Study [BWHS], had excellent calibration overall, by age, and in both low- and high-risk women, the researchers said. “Discriminatory accuracy of the new model was similar to that of the most frequently used questionnaire-based breast cancer risk prediction models in white women, suggesting that effective risk stratification for Black women is now possible.”
“It is based on variables that can be easily obtained from women themselves and entered into an online risk calculator, available at BWHS Breast Cancer Risk Calculator,” the researchers added. “Validation data indicated better performance among women younger than 40 years, for whom personalized referral for breast cancer screening may be most important.”
In the following interview, Palmer elaborated on the research.
How have currently available risk prediction models tended to underperform in black women?
Palmers: Discriminatory accuracy of the currently available risk prediction models has been shown to be lower in Black women than white women.
What risk prediction models did you compare your new model to, and how did it compare in terms of accuracy?
Palmers: We directly evaluated the NCI BCRAT [Breast Cancer Risk Assessment Tool] model, also known as the Gail model, in the same data from the Black Women’s Health Study. Discriminatory accuracy was lower: 0.56 as compared with 0.58 in our model. Although it’s not a big difference, that magnitude of improvement is exactly what is being sought with advancements in breast cancer risk prediction models.
Are there any specific predictors or variables in your model that differ from other models, or differ in terms of their weight or importance?
Palmers: Our model included first-degree family history of pro-cancer state, a variable that has not been included in previous breast cancer risk prediction models. This may be a more important factor in predicting risk for Black women, because the incidence of prostate cancer is markedly higher in Black men and therefore the prevalence of having a family history of prostate cancer will be higher in Black women.
The model also included breastfeeding, a variable that has not been included in most previous models. Breastfeeding has been associated with a reduced risk of breast cancer, and the prevalence of breastfeeding is historically lower in Black women.
You hypothesized that a model considering specific ER subtypes would improve breast cancer risk prediction in Black women. Did your research bear this out?
Palmers: No. We found that risk prediction was just as good in the model that considered all breast cancers together.
What would be necessary for breast cancer risk prediction models to be further improved in the future?
Palmers: Breast cancer risk prediction models may be improved with the addition of a polygenic risk score — that is, genetic information beyond knowledge of family history of cancer. There are two barriers, however.
First of all, genetic data across the genome is not currently available for most patients, although that could change in the near future with the decreased costs of genotyping and the increased potential of the knowledge.
Second, at this point, the existing polygenic risk scores for breast cancer work well for most other populations but have little predictive ability for Black women. That is because of the substantially greater genetic diversity in individuals of African ancestry and the relative paucity of genetic research in African ancestry populations. We simply do not have the data yet to derive and validate useful polygenic risk scores for breast cancer in women of African ancestry.
Read the study here and expert commentary about it here.