Roanoke Times
Copyright (c) 1995, Landmark Communications, Inc.
DATE: SUNDAY, March 13, 1994 TAG: 9403130078
SECTION: NATIONAL/INTERNATIONAL PAGE: A-2 EDITION: METRO
SOURCE: Associated Press
DATELINE: CHICAGO LENGTH: Medium
The organizer of the study, the National Surgical Adjuvant Breast and Bowel Project, has known about the fraud for at least two years, but has not published a re-evaluation of the findings, the Tribune said.
Dr. Bernard Fisher, a Pittsburgh surgeon who heads the research project, did not know when or where a re-evaluation would be published. A statistician for the project, however, said the findings won't change.
The study was reported in 1985 in the New England Journal of Medicine. One of its major findings was that many women with early breast cancers have the same chance of survival following a partial mastectomy, or lumpectomy, as with a more disfiguring total mastectomy. The study is credited with increasing the willingness of doctors to perform lumpectomies, the Tribune said.
About 16 percent of the patients in the study came from Dr. Roger Poisson, a professor of surgery at the University of Montreal.
Investigators found that beginning in 1977, Poisson enrolled at least 100 of his cancer patients in Fisher's studies even though they were ineligible.
The Tribune said Fisher suspended Poisson from providing new data for his studies in February 1991, eight months after he learned of the discrepancies.
The falsifications ranged from changing the date of a surgical operation by a few weeks to supplying follow-up data on a patient for two years after her death.
Jerome Kassirer, editor of the New England Journal of Medicine, said he was never told about the fraud.
What effect the tainted data will have on re-evaluated test findings is unclear. For more than a year, officials at the federal Office of Research Integrity and the National Cancer Institute have urged Fisher to publish a corrected analysis so that other researchers can make an evaluation.
Dr. Dorothy MacFarlane, the National Cancer Institute's chief investigator in the case, said women with breast cancer could be making treatment decisions based on invalid information. "We were concerned that we might have a public health crisis," she said.
But Carol Redmond, NSABP's chief statistician, said a reanalysis of the data with Poisson's contributions entirely removed "did not affect any major conclusions of any of our trials." Redmond said the results of the reanalysis will not be made public before June.
by CNB