The fact that a woman has had a mammogram "within the past few years" is not sufficient to protect her from breast cancer, says a group of researchers from the Massachusetts General Hospital (MGH). In a study published in the August issue of Radiology, the MGH investigators show that the interval between mammograms may be the most critical factor for reducing the breast cancer death rate. The conclusions are based on a new computer model developed by James Michaelson, PhD, of the MGH Cancer Center, in collaboration with Elkan Halpern, PhD, and Daniel Kopans, MD, MGH director of Breast Imaging. The model suggests that, if women had mammograms more frequently than most of them currently do, more tumors would be discovered at a stage when they are curable by local surgery and radiation.
"The findings described here suggest that screening, if used frequently enough, has the potential to reduce dramatically breast cancer deaths beyond what has already been shown in trials of screening mammography," says Michaelson. He and his colleagues note that, while MGH physicians consistently have urged prompt annual screening, those women who have come to the hospital for their mammograms actually wait an average of a year and a half between examinations, with many going two, three or more years between screenings.
"This model indicates that procrastination is not a good idea," adds Kopans. "I suspect that many women find it easy to delay having their mammograms - healthy people are rarely anxious to go looking for cancer. However, this model suggests that the time between screenings is actually extremely important. If we are to save the most lives possible, women need to comply with the general recommendation of a prompt yearly mammogram."
Scientists have long known that rates of cancer growth and spread are fundamentally important. The computer model developed by Michaelson provides a mathematical prediction of the growth and spread of breast cancers that appears sufficiently accurate to allow assessment of new and possibly better ways to screen patients for tumors. For example, almost 20 years of clinical trials have shown that women who have mammograms every one to three years reduce their risk of dying from breast cancer by 30 percent. The new computer model came up with the same 30 percent reduction for a one-to-three-year interval as was seen in actual clinical experience.
In addition, the model indicated that prompt annual screening would produce a 51 percent reduction in breast cancer deaths. It also raised the possibility that even more frequent screening could cut death rates even further - with twice yearly screenings resulting in a death rate reduction as great as 80 percent. While the current data supplied to the model suggest that twice-yearly screening may be optimal, the researchers note that such a hypothesis would need to be tested in a large clinical trail before any such recommendation could be made.
While the general conclusion that more frequent screenings would reduce breast cancer death rates appears clear, exactly how much reduction would be achieved for each screening interval remains to be determined. Michaelson notes, however, that improving the data supplied to the model - a goal he and his colleagues are now addressing - should result in more reliable and useful screening recommendations.
Kopans adds that, while the current study does not endorse having mammograms more than once a year, it does confirm the importance of prompt, annual screenings. Annual screening is widely recognized as important - with the American Medical Association in June joining the American Cancer Society and the American College of Radiology and other organizations in recommending annual mammograms for women over 40.
"The recommendation should be firm and clear: women should be advised to be screened every year," says Kopans. "Since radiation risk in the 40-and-over age group is minimal, the only reason to recommend a longer screening interval would be to save money."
Journal
Radiology