BREAST DENSITY AND REDUCED MAMOGRAM SENSITIVITY

July 29, 2010 — Poorer outcomes from conventional mammography screening of women in their 40s are primarily a result of sensitivity limits of the technology, rather than tumor biology, suggest authors of a study published online July 27 in the Journal of the National Cancer Institute.

Nearly 80% of the difference between outcomes in younger vs older women is accounted for by the inability of mammography to detect small tumors in the breast tissue of younger women, and about 20% is accounted for by faster tumor-volume doubling times, report Sylvia K. Plevritis, PhD, and colleagues from the Department of Radiology at Stanford University School of Medicine in California.

"Our results underscore the importance of continued efforts to improve technologies for early detection of breast cancer in younger women, particularly in women with dense breast tissue," they write.

Both digital mammography and ultrasound have been shown in recent trials to be more sensitive for detecting breast cancer in women with dense tissues and should be considered for improving the sensitivity of screening in younger women, the authors note.

A mammography expert who was not involved in the study told Medscape Medical News that the results are in keeping with current understanding of the limits of conventional mammography in women younger than a certain age, and point to the need for better screening methods in this population.

"If you're talking about tumor detectability at a point in time, it really doesn't matter whether the thing is doubling fast or slow if you can't see it, so more important, I think, is the density and the difficulty of detecting a tumor," said Avice O'Connell, MD, head of women's imaging at the University of Rochester Medical Center and James P. Wilmot Cancer Center in Rochester, New York.

Dr. O'Connell notes that Connecticut now requires that all women of any age who undergo mammography in that state be informed of breast density findings. Women who are deemed to have breast density of grade 3 (heterogeneous, with dense tissue making up more than 50% of the breast) or grade 4 (extremely dense breast with very little fat) receive a letter stating that they might benefit from supplementary screening tests, such as breast ultrasound, magnetic resonance imaging, or both, depending on individual risk factors.

"We can't make their breasts less dense, so we just have to work out what these women should have," Dr. O'Connell said. "If money were no object, every woman with breast density above 50% should have another test — whole breast ultrasound if you can find somebody to do it, [magnetic resonance imaging] if you can find somebody to pay for it, or one of the newer technologies coming up, molecular imaging, which is even more expensive."

Computer Simulation Model

In the new study, the Stanford investigators arrived at their conclusions by creating a computer simulation model of breast cancer screening by age. The model estimated the median tumor size detectable on a mammogram and the mean tumor volume doubling time. They derived their estimates by calibrating predicted vs actual breast cancer incidence rates from the Surveillance, Epidemiology, and End Results database, and the predicted vs actual distributions of screen-detected tumor sizes using data from the Breast Cancer Surveillance Consortium.

The model showed that younger women had higher estimated mammography thresholds — the smallest tumor diameter that could be detected on mammography — at a median of 1.63 cm for women younger than 40 years, 1.44 cm for women from 40 to 39 years of age, and 1.25 cm for women aged 50 to 59 years. Estimated mean doubling times were 179, 206, and 233 days, respectively.

The researchers then created various hypothetical scenarios to see what would happen if women in their 40s had the doubling times and/or mammography threshold values of women 10 to 20 years older. They found that when the model considered mean tumor size at detection, lifetime gained, and breast cancer mortality, the relative contributions of tumor detectability and doubling time were similar, and that there was a near-linear relationship between screen sensitivity and breast cancer death reduction with annual or biennial screening vs no screening.

However, when they compared screening once yearly to screening once every other year, the researchers found that screening sensitivity was more strongly affected by tumor detectability on mammography than on tumor doubling time.

The journal's editors point out the study's limitations, including the use of estimates that are subject to study biases and that it did not consider the possibility that reduced ability to detect a tumor on mammography could be an independent risk factor associated with a specific breast cancer subtype.

The study was supported by grants from the National Institutes of Health. Author conflicts of interest were not disclosed. Dr. O'Connell has disclosed no relevant financial relationships.

J Natl Cancer Inst. Published online July 27, 2010.