QuackeryWatch.com

Canadian Quackery Watch — documenting health fraud, since the 1990s
Case summary: A Montreal surgeon kept two sets of patient files — one marked "true," one marked "false" — to sneak ineligible women into the world's largest breast cancer research project, triggering a U.S. congressional investigation and shaking public confidence in the science behind the lumpectomy.

Dr. Roger Poisson: The Montreal Surgeon Behind the 1994 Breast Cancer Trial Fraud Scandal

Roger Poisson was a professor of surgery at the Université de Montréal and head of oncology at St. Luc Hospital, where he served as one of the leading patient-enrollers for the National Surgical Adjuvant Breast and Bowel Project (NSABP) — a joint Canadian-American research consortium, coordinated out of the University of Pittsburgh, that had been running large-scale breast cancer treatment trials since 1958. Poisson's site alone contributed roughly 16 percent of the patients in some of the project's most influential studies, including the landmark trial establishing that lumpectomy plus radiation was as effective as full mastectomy in preventing breast cancer recurrence — a finding, published in the New England Journal of Medicine in 1985, that reshaped breast cancer treatment across North America.

Two Sets of Books

From 1977 to 1990, Poisson falsified and fabricated patient records to enroll women who did not actually meet the trials' eligibility criteria — altering surgery dates, inventing lab results that were never performed, and in one case continuing to submit progress reports on a patient two years after she had died. Investigators later found he kept two separate sets of patient files in his office, one labeled "true" and the other "false," to conceal the discrepancies from NSABP's American coordinators. Poisson later characterized his conduct not as fraud but as "a nonobservance of strict criteria that had no oncological value," saying his goal had simply been to give as many women as possible access to advanced, grant-funded treatment.

The fraud was first identified in June 1990, when an NSABP statistician noticed a discrepancy between two supposedly identical operative reports for the same patient. Poisson was quietly removed from the NSABP in 1991, but the organization did not disclose the fraud publicly for roughly three years. The story broke in March 1994 after reporting by the Chicago Tribune, prompting the U.S. Office of Research Integrity to investigate. It concluded that Poisson had committed research misconduct over nearly a decade. Poisson lost his access to U.S. grant funding, his professorships at the Université de Montréal, and his position as chief of oncology at St. Luc Hospital as a result.

Fallout for the NSABP

The scandal escalated well beyond Poisson himself. Members of the U.S. Congress, including the Congressional Caucus on Women's Issues, publicly questioned the integrity of federally funded cancer research, and the National Cancer Institute seized the NSABP's data and suspended new patient recruitment across its trials — including an ongoing 16,000-woman tamoxifen prevention study — while it investigated. NSABP's own director, Dr. Bernard Fisher of the University of Pittsburgh, was forced to step down amid scrutiny over whether he and his chief biostatistician had knowingly published results built in part on Poisson's falsified data. A subsequent reanalysis of the NSABP's major trials with Poisson's contributed data entirely removed found it did not change any of the trials' major conclusions — meaning the underlying finding that lumpectomy was as effective as mastectomy for many patients held up. The case nonetheless remains one of the most consequential clinical-trial fraud scandals in medical history, and prompted lasting reforms in how multi-site clinical trials audit and verify patient eligibility data.

Timeline

Sources