Top Harvard Cancer Institute Will Retract Six Studies and Correct 31 More After Photoshop Claims

British biologist and blogger Sholto David alleged that executives at the Dana-Farber Cancer Institute published papers with manipulated data and images

the glass facade of a building marked Dana-Farber Cancer Institute
The Dana-Farber Cancer Institute Craig F. Walker / The Boston Globe via Getty Images

More than 50 research papers published by the Dana-Farber Cancer Institute (DFCI), a prestigious teaching center affiliated with Harvard Medical School, are under review this week following online claims that their images and data had been manipulated.

Sholto David, a British molecular biologist and blogger, wrote a post for the independent research integrity blog For Better Science on January 2, in which he accuses researchers at the institute of “data forgery.” The long and at times scathing blog post featured images such as protein bands, data plots and PCR results from the papers, which David alleges were manipulated via copy and paste or Photoshop. The studies in question had been published between 1999 and 2017.

In one paper, an image of mice on the first day of a study seems to reappear on the 16th day. In others, western blots, used by scientists to detect proteins, and images of stained cells appear to be flipped, stretched and resized, David writes.

“In conclusion: a swath of research coming out of DFCI authored by the most senior researchers and managers appears to be hopelessly corrupt with errors that are obvious from just a cursory reading,” he writes. David used both the naked eye and the artificial intelligence software ImageTwin to detect the allegedly doctored images.

nine different views of mice, some with colors measuring luminescence. One image of mice on day one appears identical to one on day 16, which David has highlighted with arrows
Sholto David's annotations (in red) from a study published in the journal Blood in 2014, showing a photograph of mice that has allegedly been duplicated. Sholto David via For Better Science

Three weeks later, as internal reviews of the institute’s research are ongoing, DFCI has announced that it will be retracting at least six studies and correcting 31, per the New York Times’ Benjamin Mueller. The studies in question had been published in more than a dozen major medical and biological peer-reviewed journals, including Molecular and Cellular BiologyClinical Cancer Research and Blood.

DFCI has not yet released the specifics of which studies will be corrected or retracted, though many of the works David scrutinized were published by the institute’s leading scientists—including president and CEO Laurie Glimcher, COO William Hahn and senior researchers Irene Ghobrial and Kenneth Anderson, STAT News’ Angus Chen and Jonathan Wosen report.

The institute also conceded that an undisclosed number of studies mentioned by David had already been flagged for review over data and image concerns, per the publication.

“We knew about many of these papers and their allegations before the blog post,” Barrett Rollins, the institute’s research-integrity officer, tells the Wall Street Journal’s Nidhi Subbaraman.

Rollins himself is a co-author of two studies mentioned by David and will be recused from investigations about them. But he tells the Harvard Crimson’s Veronica H. Paulus and Akshaya Ravi that some of these studies were based on data from outside laboratories.

The “presence of image discrepancies in a paper is not evidence of an author’s intent to deceive,” he says in a statement to the publication, which was the first to report about these allegations. “That conclusion can only be drawn after a careful, fact-based examination, which is an integral part of our response.”

several images of crystal violet stained cells, with a few sections that appear visually similar highlighted in boxes. David alleges these were copy and pasted
Sholto David's annotations (in red and blue) from a study published in 2012 in the journal Clinical Cancer Research, alleging copy-and-pasted images. Sholto David via For Better Science

Some researchers have responded to the accusations more quickly than others. The Harvard Crimson reports that both Hahn and Ghobrial have submitted corrected images to journals containing work that had been mentioned by David.

The allegations come as Harvard University’s research is held under a microscope. In early January, the former university president Claudine Gay, facing plagiarism accusations in her own research, stepped down.

For DFCI, the wide scope of corrections has some outside scientists questioning the center’s internal review process.

“If it’s one or two, that could still be an honest error,” Elisabeth M. Bik, a microbiologist and scientific integrity expert, tells the Harvard Crimson. “If it’s ten overlaps, then you have to ask yourself, that is pretty sloppy if you make that many errors.”

Steven Salzberg, a biologist at Johns Hopkins University, speaking about scientific misconduct in general, tells STAT News: “I know professors who found out [scientific misconduct] was happening in their labs and retracted the papers. … It’s a black mark on your record, but if you’re a scientist with integrity, then that’s what you do, and you recover from it.”

Get the latest stories in your inbox every weekday.