Lancet: Screening mammo prevents breast cancer deaths in mid-risk women under 50
Screening mammography with breast tomosynthesis in a 57-year-old woman. Image source: RadioGraphics
Annual screening mammography of women with a medium familial risk of breast cancer is likely to prevent breast cancer deaths, according to results of the FH01 study published online in Lancet/Oncology on Nov. 18.

Existing data indicate mammographic screening of the general population of women aged 40 to 49 years yields reduced mortality. However, the impact of screening on disease stage and projected mortality of women under age 50 with a family history of breast cancer is unknown. The authors conjectured that earlier or more frequent mammographic surveillance of the medium-risk group might offer an appropriate screening strategy. (The current U.K. National Health Service model offers three-yearly, two-view mammograms to women aged 50 to 70 years.)

The research team, led by Stephen W. Duffy, MD, of Cancer Research U.K. Centre for Epidemiology, Mathematics, Statistics in London, devised a quantitative assessment. The single arm FH01 study offered women with substantial family breast cancer risk annual screening mammography for five years.

The authors recruited 6,710 women from 76 centers between Jan. 16, 2003 and Feb. 28, 2007 and planned to assess mammographic surveillance on disease stage at presentation and estimate the probable future effect on breast cancer mortality, using external datasets from the U.K. Age Trial (which was  designed to assess whether annual mammography from age 40 years reduces breast cancer mortality and by what percentage) and a Dutch study of women with a family history of breast cancer.

Duffy and colleagues reported data acquired by the ongoing surveillance program as of Nov. 30, 2009. Ninety one percent of enrolled women were 40 to 44 years of age at enrollment, and mean age at first screen was 42 years, stated the researchers.

“In total, in 26,244 person-years, 136 women were diagnosed with breast cancer, which is 5.2 diagnoses per 1,000 person years,” according to Duffy. Seventy-seven percent of cancers were detected at screening, yielding a screening sensitivity of 79 percent.

When researchers compared the pathological characteristic of tumors detected in the FH01 group with controls in the U.K. Age Trial and the Dutch study of women with non-BRCA cancer, they found invasive cancers in the FH01 study were significantly smaller, less likely to be node positive and of a more favorable histological grade.

Duffy and colleagues calculated the Nottingham prognostic index (NPI) to estimate 10-year survival and predicted average 10-year survival of 81 percent in the FH01 study, 73 percent in the U.K. Age Trial control group and 71 percent in the Dutch study. After calculating risk of death, they added that the FH01 surveillance program likely prevented five deaths in the 96 women with invasive cancer, which is equivalent to two deaths prevented per 10,000 screens. The researchers pointed out that the estimate is higher than the U.K. Age Trial estimate of 0.56 deaths prevented per 10,000 screens over seven to nine years.

The authors identified several limits to the study: the absence of a control group and use of predicted mortality from the NPI score, a combination of tumor size, node status and grade.

Radiological and pathological analyses of FH01 data are ongoing; the authors suggested “our results provide some confidence in the interim that the surveillance strategy is effective.”