Toronto researchers alleged to have falsified data; ask court to review claims

TORONTO -- The retraction of two scientific papers and a concern raised about a third have led to allegations of data falsification in a laboratory run by two Toronto researchers.

The scientists -- Dr. Sylvia Asa, a pathologist, and her husband, Dr. Shereen Ezzat, an endocrinologist -- have denied the findings of an internal review at the conglomeration of hospitals where they work, University Health Network (UHN).

That report stated the two retracted papers contained data that had been manipulated and-or falsified. It said the third contained an image that was identical to one used in a different paper published in another journal, a scientific no-no.

Asa was head of pathology at UHN from 2000 until this spring, when she stepped aside. It has not been revealed if that move was a response to the internal investigation, the findings of which were submitted in April to the editors of the journal that published the three articles, the American Journal of Pathology.

Both Asa and Ezzat remain on staff at UHN. But it has been reported that work in their lab has been suspended.

The scientists, who study the genetics of breast cancer, are refusing to comment about the situation.

But their lawyer, Brian Moher, confirmed he has filed an application for a judicial review of the investigation's findings with Ontario's divisional court. Moher said a preliminary court date has been set for Aug. 27.

"The doctors disagree with the allegation that they engaged in the falsification of data," Moher said in an interview, adding the application is for "an impartial review."

Questions about their work arose in September 2012 when a reader wrote to the journal to raise concerns about two papers published in 2010. In the notice of retraction, published in the journal's August issue, editors said they approached Ezzat, the corresponding author, but the exchange "failed to resolve the matter."

The journal then asked UHN to investigate. Dr. Christopher Paige, vice-president of research, and Dr. Charles Chan, vice-president of medical affairs and quality, conducted the investigation. In April, they informed the journal of their findings.

University Health Network will not release the report. But the retraction notice says Paige and Chan informed the journal that multiple figures in both studies contained manipulated and-or fabricated data.

The researchers agreed that the papers should be retracted, though they insisted the findings were valid and could be reproduced, which is of key importance in science.

"On behalf of all of the other authors, we wish to state that we have collectively confirmed the validity and reproducibility of the findings reported in these articles. Nevertheless, we request that these papers be retracted," Ezzat wrote.

Asa and Ezzat had one co-author on one of the papers and two on the other.

The journal's August issue also contains a "Note of Concern" about a third article written by Asa, Ezzat and another author.

Published in the journal in September 2003, that article contained an image that was identical to one used in an article the pair -- along with three other authors -- published the previous year in the Journal of Clinical Investigation. After the investigation confirmed the problem, Asa supplied the journal with a replacement image and the article has been corrected.

The journal's editors declined to be interviewed Monday. But in an email, managing editor Audra Cox said the journal is not currently looking into other papers Asa and Ezzat published in the journal.

Gillian Howard, UHN's vice-president for public affairs and communications, also declined to be interviewed. But she released a statement CEO Dr. Peter Pisters wrote to staff about the situation.

Pisters said researchers at the University Health Network publish more than 3,000 scientific articles a year and from time to time problems come to light.

"In the context of our significant research efforts, issues arise that span a spectrum from sloppy record keeping to plagiarism to inappropriate alteration of data," Pisters said, adding that on average, UHN investigates three or four such issues a year.

"We take each allegation seriously."