Breast Cancer Surveillance for High-Risk Women
April 1, 2001
Breast Cancer Surveillance for High-Risk Women
ABSTRACT & COMMENTARY
Synopsis: The effectiveness of breast cancer surveillance in high-risk women, including BRCA1/2 mutation carriers, was evaluated in a retrospective study of 1198 women — 449 moderate risk, 15-30% estimated lifetime breast cancer risk; 621 high risk, 30-50% estimated lifetime breast cancer risk; and 128 BRCA1/2 mutation carriers, and 60-85% estimated lifetime breast cancer risk. Screening consisted of instructions for monthly breast self-examination, clinical breast exam every 6-12 months, yearly mammography, optional use of breast MRI in more recently screened patients, and median follow-up was 3 years. Thirty-one invasive breast cancers and 4 ductal carcinoma-in-situ lesions were detected, and the ratio of observed vs. expected numbers of invasive breast cancer varied from 23.7 (95% CI, 1.2 to 483) in BRCA1/2 carriers to 7.0 (95% CI, 1.9 to 26.1) in the high-risk group and 2.7 (95% CI, 0.4 to 17.6) in the moderate-risk group. Screening parameters were comparable to population screening data. Screening detected a significantly greater number of cancers in these high-risk groups than would be expected in an average risk population.
Source: Brekelmans CT, et al. J Clin Oncol. 2001;19:924-930.
Breast cancer screening of average-risk women using mammography and clinical examination has been shown to decrease breast cancer mortality for women aged 50-69 years, and benefit is suggested, but controversial, for average-risk women aged 40-49 years and older than 70 years of age.1 Guidelines for screening of average-risk women include regular (monthly) breast self-exam as well as annual clinical breast exam and mammogram starting at age 50, with consideration of starting the screening process at 1-2 year intervals at the age of 40 years.1 While these recommendations are for average-risk women, the identification of risk factors for breast cancer allow identification of women who have an increased risk of developing this disease.2,3 It is possible that increased screening of women with a high risk of breast cancer would be beneficial, as a greater number of breast cancer events is expected in high-risk groups. However, biological differences may exist in breast cancers occurring in BRCA1/2 carriers or individuals at increased genetic risk of breast cancer.4,5 The potential clinical importance of some of these biological differences is under investigation.6 The current study was performed to evaluate the effectiveness of breast cancer screening in high-risk women, including BRCA1/2 mutation carriers.
This study by Brekelmans and colleagues evaluated 1198 women with an increased risk for breast cancer. The risk categories evaluated included individuals with a moderate increased breast cancer risk (15-30% estimated lifetime breast cancer risk, 449 women); a high increased breast cancer risk (30-50% estimated lifetime breast cancer risk, 621 women); and proven carriers of a BRCA1/2 mutation (60-85% lifetime breast cancer risk, 128 women). The screening intervention consisted of monthly breast self-examination, yearly mammogram, and clinical breast exam every 6 months except in some moderate-risk individuals who had a yearly screening interval. Data collection was primarily retrospective and included patients evaluated between 1978 and the late 1990s. Since 1995, breast MRI was optionally included with the screening for individuals with dense mammographic breast tissue and/or BRCA1/2 mutations. Overall, 35 breast tumors (31 invasive breast cancer and 4 ductal carcinoma in situ) were detected within the median follow-up period of 3 years (range, 0-22 years). Twenty-six of the 35 tumors were detected at the screening intervals, and 9 tumors were detected in the intervals between screens. The detection rates of invasive breast cancer were 33 per 1000 person-years in BRCA1/2 mutation carriers (observed vs expected breast cancer case ratio of 23.7 [95% CI, 1.2-483]); 8.4 per 1000 in the high-risk group (observed vs expected breast cancer case ratio of 7.0 [95% CI, 1.9-26.1]); and 3.3 per 1000 in the moderate-risk group (observed vs expected breast cancer case ratio of 2.7 [95% CI, 0.4-17.6]). A clear trend of increasing detection rates with age was demonstrated.
The characteristics of the invasive tumors were reported, and 20 of 31 (65%) were node negative. The percent of node-positive tumors was similar for the incident screen detected and the interval cancers. Interval cancers were defined as cancers detected between screens. Overall, the sensitivity of the screening test was 74% and had an expected increase with age. Brekelmans et al conclude that it is clearly possible to identify young women at high familial risk, and screening this population detected approximately seven times more breast cancers than expected in an average-risk population of comparable age. The problem of interval cancers was considered due to either missed cancers (ie, poor mammographic visibility) or due to tumors with a high growth rate between screening sessions. Consideration of either more intensive screening efforts or new technologies, such as MRI, was suggested for evaluation in these high-risk patients.
Comment by Mark R. Albertini, MD
Patients with an increased risk of breast cancer can be identified, and these patients are appropriate candidates for intensified breast cancer screening efforts. The opportunity to design breast cancer screening efforts that reflect overall breast cancer risk provides a means for increased surveillance in patients with the highest expected number of events. The important outcome of detection of a large number of cancers by screening was demonstrated in this study. However, another important outcome of this study was the identification of cancers developing during the intervals between screening events for BRCA1/2 mutation carriers and women with high familial risk. Additional evaluation is needed to determine the relationship of the surrogate end points of screen detected and interval cancers with the overall end point of breast cancer mortality.7
Additional genetic determinants, such as polygenes that may act in conjuction with environmental factors, are being recognized as potentially important determinants of overall breast cancer risk.8 Detailed risk assessment profiles will likely be possible to identify larger numbers of individuals at an increased risk for breast cancer. Many of these patients may be candidates for chemoprevention studies, and these high-risk patients may benefit from increased breast cancer surveillance. Prospective studies to evaluate more intensive screening strategies, as well as screening strategies using more sensitive imaging modalities such as breast MRI, are needed for women at high-genetic risk for breast cancer.
References
1. Abeloff MD, et al. Breast. In: Abeloff MD, Armitage JO, Lichter AS, Niederhuber JE, eds. Clinical Oncology. 2nd ed. London, England: Churchill Livingstone; 2000:2051-2159.
2. National Cancer Institute: Breast Cancer Risk Assessment Tool. Available at: http://cancernet.nci.nih.gov/bcra_tool.html.
3. Claus EB, et al. Cancer. 1994;73:643-651.
4. Lancet. 1997;349:1505-1510.
5. Lakhani SR, et al. J Natl Cancer Inst. 1998;90:1138-1145.
6. Verhoog LC, et al. J Clin Oncol. 1999;17:3396-3402.
7. Armstrong K, Weber BL. J Clin Oncol. 2001;19:919-920.
8. Chang-Claude J. IARC Sci Publ. 2001;154:177-190.