Studies Retracted at Harvard-Affiliated Dana-Farber

Scientists at Dana-Farber Cancer Institute, which is affiliated with Harvard, are planning to correct 31 studies and retract six more as part of a probe into discrepancies and duplications of images from some of its top researchers after a post on For Better Science, a scientific integrity blog, accused them of data manipulation.

The research at Boston's Dana-Farber, a top institute in cancer research and treatment, is largely supported by taxpayer money, including receiving more than $160 million through the National Institutes of Health in 2023, The Washington Post reported Tuesday.

Barrett Rollins, the research integrity officer at Dana-Farber, said that as a result of the blog's research, one paper containing a potential error remains under examination in addition to the pending corrections and retractions. 

Sixteen other papers relied on data that was gathered outside Dana-Farber, and those laboratories have been made aware of the issue, Rollins said.

However, he said the blog's allegations were not all correct, as no action was needed on three of the papers.

"The presence of image discrepancies in a paper is not evidence of an author's intent to deceive," Rollins said in a statement. "That conclusion can only be drawn after a careful, fact-based examination which is an integral part of our response. Our experience is that errors are often unintentional and do not rise to the level of misconduct."

Rollins has been recused from reviews on three papers that were flagged, as he was a co-author on those works, said Ellen Berlin, a spokeswoman for Dana-Farber.

Sholto David, an independent molecular biologist who wrote about the papers on the blog, said he started examining cancer research studies while he was between jobs, looking at image problems through the use of an AI-based tool, Imagetwin.

David noted problematic studies coming out of Dana-Farber and started researching works by its key leaders: President Laurie Glimcher and Chief Operating Officer William Hahn. 

Four of the papers David flagged were co-authored by Glimcher, and Hahn co-wrote 17 papers. Some of the issues that were flagged were caught more than 10 years ago on PubPeer, where researchers critique papers after they are punished, including on allegations of image manipulation.

• Dana-Farber Researchers Accused of Data Manipulation
• FDA to Require Warning on CAR-T Cancer Therapies