New Model Provides Improved Prediction of Breast Cancer Risk in African American Women
Breast cancer risk prediction models have underestimated the risk for African American women, a factor that has contributed to lower rates of recruitment to prevention trials in this population. In a study reported in the Journal of Clinical Oncology, Boggs et al developed a risk model, derived from the Black Women’s Health Study (BWHS), which showed good agreement with observed cancer rates and indicated that a greater proportion of women are at elevated risk compared with previous models.
Study Details
In the study, a risk model including family history, previous biopsy, body mass index at age 18 years, age at menarche, age at first birth, oral contraceptive use, bilateral oophorectomy, estrogen-plus-progestin use, and height was developed using relative risks derived from 10 years of follow-up of BWHS participants aged 30 to 69 years at baseline. With use of the subsequent 5 years of follow-up data, calibration of the model was assessed as the ratio of expected to observed numbers of breast cancers, and discriminatory ability was assessed using the concordance statistic.
Model Performance
The expected-to-observed ratio for breast cancers overall was 0.96 (95% confidence interval [CI] = 0.88–1.05), with similar good agreement observed in most risk factor categories. Discriminatory accuracy was higher for women aged < 50 years (area under the curve [AUC] = 0.62, 95% CI = 0.58–0.65) than for women aged ≥ 50 years (AUC = 0.56, 95% CI = 0.53–0.59; P = .03). The AUC was nonsignificantly greater for estrogen receptor–positive disease than for estrogen receptor–negative disease (0.62 vs 0.58, P = .10).
With a 5-year predicted risk of ≥ 1.66% as a cutpoint, 2.8% of women aged < 50 years and 32.2% of women aged ≥ 50 years were classified as at elevated risk of invasive breast cancer. By comparison, another reported risk prediction model (CARE model) indicated a 5-year risk ≥ 1.66% in 0.1% of younger women and 7.3% of older women.
The investigators concluded: “The BWHS model was well calibrated overall, and the predictive ability was best for younger women. The proportion of women predicted to meet the 1.66% cutpoint commonly used to determine eligibility for breast cancer prevention trials was greatly increased relative to previous models.”
Julie R. Palmer, ScD, of Slone Epidemiology Center at Boston University, is the corresponding author of the Journal of Clinical Oncology article.
The study was supported by Susan G. Komen for the Cure and the National Cancer Institute.
Deborah A. Boggs, ScD, is employed by Quintiles.
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®.
