Pat Schneider, 42, a nurse-midwife at the University of Chicago with three children, chose a lumpectomy in 1990. “Once you have cancer, you may have periods where you really forget about it for a while,” she reflects. “But something like this brings up all your old emotions. This is a big issue. This isn’t just deciding to have a more attractive [Caesarean] incision, a bikini cut as opposed to a longitudinal scar. This means that you would live to go to your children’s graduation, to see them grow up. You never want to be in a situation where you find out you were in the wrong group.”
The study that concerns Schneider is part of the National Surgical Adjuvant Breast and Bowel Project. Begun in 1957 and funded by the government, the NSABP is an ongoing series of trials that evaluate a variety of cancer treatments. So far, more than 44,000 patients from nearly 500 separate medical centers have taken part. In 1990, a routine data review by the University of Pittsburgh, which administers the project, discovered that one participating surgeon, Dr. Roger Poisson of Saint-Luc Hospital in Montreal, had falsified information about some of his patients. He included women, for example, whose tumors were bigger than the study’s strict criteria permitted. Further investigation revealed that he had tampered with data from 99 patients, but only six were enrolled in the much-vaunted study that found lumpectomy was as effective as mastectomy for early cancers. Dr. Bernard Fisher, a pioneering researcher in the biology and treatment of breast cancer who heads the NSABP, is satisfied that the study’s results are perfectly valid even when Poisson’s data are excluded. While there didn’t seem to be any attempt at a cover-up, it was not until last July that the Office of Research Integrity published the facts of the data fraud in the Federal Register-and even sent out a press release. But because the information was buried in a long litany of other fraud cases, it went unnoticed. No public statement was made by the government at the time, says Dr. Michael Friedman, who directs treatment evaluations for the National Cancer Institute, because health officials expected “that this material would be published in complete form soon.” Indeed, a re-analysis of the NSABP data by its directors will be submitted to The New England Journal of Medicine by next week.
Medical ethicists can’t understand how Poisson could have compromised such an important study, even though his lawyers claim he included ineligible patients (and then altered his data) only because he thought they’d benefit from the treatment and follow-up the study provided. “It’s not just some little goofy study,” notes Arthur Caplan, a medical ethicist at the University of Minnesota’s Center for Biomedical Ethics. “It’s the basis for a number of interventions that affect hundreds of thousands of women. . . . This is the backbone study of how medicine approaches breast cancer.”
Women often get mixed signals about lumpectomy. Many surgeons don’t think the procedure has proven itself yet. Now, concern over Poisson’s altered data will likely increase women’s uncertainty. “This might have been much less sensational,” says Cindy Pearson of the National Women’s Health Network, “if it had come from Fisher or NCI and they said ‘We’re horrified and we’re going to make sure it never happens again’.” But because the fraud was presented in a newspaper report, “Women feel they weren’t being told information that affects their lives and are upset because they think someone made a conscious decision to hide it.” When callers to the Network ask if lumpectomy is safe, they are now told “probably yes,” says Pearson, “which isn’t as reassuring as a solid yes.” And that’s a continuing worry for women who, like Pat Schneider, want to be sure they’ll be able to show up at graduation.
Candidates for lumpectomy are women with small, localized, early-stage cancers that are not too deep within the breast.
The course of radiation that usually follows lumpectomy weakens the skin of the breast. If the cancer spreads and a mastectomy is needed later, reconstruction may not be possible.
About 117,000 women had mastectomies in 1992 and 45,000 had lumpectomies. Among women whose cancer was diagnosed at an early stage, a third chose lumpectomies.
In New England, 53 percent of early breast-cancer surgeries in 1988 were lumpectomies, compared with only 15 percent in the south-central U.S.
