Should women start breast cancer screening at age 40 or 50? Every year or every other year? By mammography or breast MRI? As patients and physicians ask these and related questions, confident answers require solid evidence. That’s why in 2011, the National Cancer Institute awarded a $20 million Program Project grant to the Breast Cancer Surveillance Consortium (BCSC), a national research network of mammography registries, and a five-year, $7.5 million contract to make previously collected BCSC data available to outside researchers.
Group Health Research Institute (GHRI) Senior Investigator Diana L. Miglioretti, PhD, is principal investigator of the contract and co-PI of the Program Project with Karla Kerlikowske, MD, University of California San Francisco. GHRI’s Casey Luce, MSPH, directs the Program Project Administrative Core, and Dr. Miglioretti directs the Biostatistics and Data Management Core, extending her previous role as PI of the BCSC Statistical Coordinating Center, which oversees collection, management, and analysis of BCSC data.
And these are amazing data. The BCSC is the nation’s largest longitudinal collection of breast imaging data, consisting of 9.5 million mammograms, 2.3 million women, and 114,000 breast cancer cases so far. Data include individual risk factors and information on previous screenings and breast procedures. Linkages to pathology, radiology, cancer, and mortality registries allow longitudinal analysis of diagnosis, treatment, and outcomes. The diversity of represented women and communities will increase this funding cycle with the addition of the Metro Chicago Breast Cancer Registry to the existing registries.
“The BCSC aims to take screening recommendations from one-size-fits-all to personalized,” says Dr. Miglioretti. “For that, we need better individual-level risk-prediction models. We need to be better at determining which women are likely to benefit from starting screening earlier or later, or being screened at different frequencies. We need to be better at predicting who will get aggressive vs. slow-growing tumors. And we need to determine which women will benefit from new technologies, and make sure those who benefit have access.”
The funding comes as the national debate about screening heats up. In 2009, the U.S. Preventive Services Task Force issued new guidelines recommending that women aged 40-49 discuss breast screening with their physicians, rather than automatically starting at age 40. While controversial, says Dr. Miglioretti, this means we are talking not only about the benefits of screening, but the harms such as overdiagnosis leading to invasive treatment for non-life-threatening conditions, and false-positive test results. “We tend to downplay the harms of false positives,” she says. “Many seem minor, but they affect millions of women every year.” GHRI Assistant Investigator Rebecca Hubbard, PhD, in a 2010 Annals of Internal Medicine study using BCSC data on more than 169,000 women, found that over 10 years, a majority of women receiving annual screening mammograms will have a false positive.
GHRI’s history of breast screening research makes it a natural home for the BCSC leadership. In 1985, led by Senior Investigator Robert S. Thompson, MD, now emeritus, Group Health initiated population-based breast cancer screening with mailed risk-factor surveys and personal schedules for screening mammograms. The program reduced breast cancer in the Group Health population and spurred groundbreaking GHRI studies on improving screening effectiveness. This record, and GHRI’s experience as the BCSC Statistical Coordinating Center, are the foundation for leading the integrated, multiproject BCSC program. Dr. Miglioretti will head an established collaborative team of expert biostatisticians and data managers and stellar administrative support, including Grants and Contracts Unit personnel skilled in handling large grants. Leading the Group Health BCSC registry is GHRI Senior Investigator Diana S.M. Buist, PhD, MPH.
The BCSC is taking a three-pronged approach to improve screening. One project develops better risk-prediction models that include individual factors such as race/ethnicity, hormone levels, and genetic profiles. Another examines the effectiveness of different imaging types: ultrasound, magnetic resonance imaging, film and digital mammography, and emerging technologies such as breast tomosynthesis (3D mammography). The third project addresses access to screening, including advanced imaging technologies, in different community groups. Project leads are Dr. Kerlikowske; GHRI Affiliate Investigator Constance Lehman, MD, University of Washington; and Tracy Onega, PhD, Dartmouth University.
A continuing nationwide conversation over cancer screening guidelines shows that talking about benefits and harms can be challenging. The BCSC hopes to give women and physicians a solid starting point for discussions with up-to-date, individually tailored evidence about effective breast screening strategies.
By Chris Tachibana
Oct. 17, 2011