The Politics of Women's Health
Mammography Screening Controversy
Read this Before You Have a Mammogram
An interview with Cornelia J. Baines, MD by Maryann Napoli, Associate Director of the Center for Medical Consumers, October 2002.
Last month, the Canadian National Breast Screening Study published follow-up results showing, once again, that mammography screening did not reduce the breast cancer death rate for women in their 40s (Annals of Internal Medicine, 9/3/02). The Study's findings have challenged the prevailing belief that early breast cancer detection saves lives. Worse, they show that mammography screening leads many women to be treated unnecessarily with mastectomy or radiation therapy. Though 40 more cases of non-palpable invasive breast cancer were detected in the mammography-screened women, their breast cancer death rate was no different from that of the women who did not get mammograms. Similarly, there were 42 more cases of ductal carcinoma in situ, a non-invasive cancer, detected in the mammography-screened women. This shows that mammography screening causes a significant number of younger women to suffer treatment-related harm without reducing their odds of dying of breast cancer.
Mammography proponents have criticized the Canadian Study ever since it first published results more than a decade ago. The Study now has 11 to 16 years worth of follow-up for women in their 40s. Its deputy director, Cornelia J. Baines, MD, is interviewed about the fact that-in the early years of this trial-there were more breast cancer deaths among women given mammograms. This was dismissed as a statistical fluke when it first showed up. Now some researchers are having second thoughts.
MN: When you published your seven-year results, there were more breast cancer deaths (38) in the mammography-screened women, compared with those in the control group who had no mammograms (28). Were there any surprises now that you have 11-16 year results?
Dr. Baines: No, I knew by 1983 that more breast cancer deaths were occurring in the mammography-screened group rather than the control group. Of course, that's not what we expected. When we started out, we were sure that we were going to show a major benefit. After all, the HIP Study [the first mammography trial conducted in the 1960s] had shown a benefit to women ages 50-69, and we assumed that the only reason a benefit wasn't shown for younger women was that the mammography was archaic by todayís standards.
MN: When I interviewed you at the time you published seven-year results, you said that the excess of ten breast cancer deaths in the mammography group was not statistically significant because of the large number of study participants (50,000). I thought that meant it could be ignored.
Dr. Baines: You are quite right it's not statistically significant, but what is disturbing is that this excess has happened in all screening trials in three different countries. 1985 was a landmark year for mammography screening trials. A Swedish study headed by Laszlo Tabar was published in The Lancet (4/13/85). When you read the abstract [summary] of that study, it says that women ages 40-74 showed a 31% reduction in breast cancer deaths. But if you look in the text of the article, you see that the number of deaths in the [small subset of] women in their 40s given mammograms was higher than in the control group. Similar results were observed in the Stockholm and HIP trials. The consistency of this trend demands further evaluation.
MN: Is anyone looking into it?
Dr. Baines: When we published our first results in 1992, it never entered my head that the people who have been promoting mammography would try to completely destroy the credibility of our study and ignore this excess mortality phenomenon which had been clearly shown in Tabar's study and which had also been shown in the HIP study. I started out saying that this needs investigating at the basic science level and believing that screening researchers would pay attention to these trends. Well, was I ever out to lunch. People, when they strongly believe in something, don't waste time looking at evidence that challenges their beliefs. That's just not human nature.
MN: Dr. Tabar is a recipient of an American Cancer Society award for his promotion of mammography screening and a teacher of Continuing Medical Education courses for American radiologists. He and the other mammography researchers might not want to look at the "why" behind the increase in breast cancer deaths, but haven't some researchers begun to investigate a possible underlying biological mechanism for the deaths?
Dr. Baines: Yes, Michael Retsky, at Harvard Medical School, and Romano Demicheli and William Hrushesky. They studied the recurrence patterns of 251 premenopausal women with node-positive breast cancer treated only with surgery and followed for 16-20 years. Retsky and colleagues found that the breast cancer mortality rates show two peaks: one occurs three years after diagnosis, the other at nine years, and after that, women seem to survive quite well. This, of course, corresponds with what we have been observing in mammography screening trials. Increasingly, researchers like Michael Retsky and Michael Baum speculate that something associated with the biopsy or surgery stimulates growth factors. In some women with micrometastases [undetectable spread of cancer outside the breast], these growth factors may stimulate the micrometastases, and the woman goes on to die. This is consistent with the suggestion made a long time ago by Bernard Fisher [America's leading breast cancer researcher]-that micrometastases has already occurred in 90% of all breast cancers before clinical or radiological detection.
MN: Are you talking only about women in their 40s?
Dr. Baines: The finding was more prominent in younger women, but Tabarís study showed a breast cancer mortality increase in older women as well.
This interview was originally published in the October 2002 issue of the Center for Medical Consumers monthly newsletter, HealthFacts.
Next Page >
< Return to The Politics of Women's Health Overview