Altman, Lawrence K. "Probe Into Flawed Cancer Study Prompts Federal Reforms," New York


Altman, Lawrence K. "Probe Into Flawed Cancer Study Prompts Federal Reforms," New York
Times, 26 April 1994, p. C3.

There's nothing like a few Congressional investigations to get bureaucrats to clean up their act.
The fudging of Roger Poisson, a major contributor to the patient sample for one of the breast
cancer studies conducted by the National Surgical Adjuvant Breast and Bowel Project, (NSABP)
and coordinated at the University of Pittsburgh by Bernard Fisher, has elicited several
Congressional inquiries. Representative John Dingell has already held one meeting and
additional meetings are planned by him and by Senator Diane Feinstein and Senator Connie
Mack, both members of the Senate's ad-hoc breast cancer coalition.

The National Cancer Institute is reported to be tightening procedures, controls over the
physicians and hospitals participating in these longitudinal and hyperexpensive studies. Thus, on
18 March NCI suspended two institutions, Louisiana State and Tulane from recruiting patients
for the breast cancer study; audits at those institutions found "sloppy record keeping" and
incomplete documentation. In the past the findings at Louisiana and Tulane would not have been
made public because the "were considered poor data management rather than fraud..." but times
are changing.

"Since the scandal became public, the cancer institute has created a new branch to oversee the
integrity of clinical trials and officials said they were exercising administrative powers they had
never used. The drastic new procedures being put in force affect research in medical schools and
hospitals throughout the country.

"The institute is changing its contracts with researchers to include provisions that require them to
relinquish data tapes of tainted studies and to allow the institute to perform its own re- analysis in
cases of fraud, as it is doing in the breast cancer studies.

"The agency has pledged to take a more active role in informing the public of future cases of
fraud. It said it would not bury such reports in The Federal Register in language that even experts
cannot understand, as it did with the Canadian fraud.

"Participating hospitals will now be required to notify patients in the trials of fraud and to
arrange for appropriate follow-up evaluation."

"Stung by Dr. Poisson's explanation that he cheated in the studies to provide his patients the best
care, which the cancer institute officials characterized as "farcical," they are now requiring ethics
training for each investigator in their studies.

"The institute will also directly notify editors of medical journals where flawed studies were
published that retractions should be published, rather than leave it to the authors of the papers to
publish corrections at their own speed."

"The institute's new efforts are aimed at restoring public confidence in medical research that
many members of Congress and witnesses at the hearing said had been damaged by the breast
cancer scandal."

"One further data error involving Dr. Fisher has come to light. In the late 1980s, officials of
University of Pittsburgh solicited a gift from ICI Pharmaceuticals Group, the British
manufacturer of tamoxifen, a drug that Dr. Fisher's studies were beginning to find effective in
treating breast cancer, for an endowed chair in his honor. ICI gave $600,000, and other gifts
swelled the total to $1.2 million. But it was not enough to establish such a chair. In any case, said
Dr. Thomas Detre, the university's senior vice chancellor for health science, Dr. Fisher had
expressed a personal dislike for such recognition.

"Nevertheless, Mr. Dingell said at the time of his hearing, Dr. Fisher has listed himself as the ICI
Professor of Surgery in ‘American Men and Women of Science.' An even more specific listing in
‘Who's Who in America' says he is Bernard Fisher, ICI-Pharma professor of surgery. University
of Pittsburgh officials attribute the listing to a clerical error. It is just one of the many data errors
that could have been detected without a Congressional hearing."