Critics claim to find flaws in dozens of Alzheimer’s studies by Temple scientist

When Temple University scientist Domenico Praticò helped secure a $3.8 million state grant to study Alzheimer’s disease, school officials hailed the news as a sign that his research center was “poised to become a national leader in dementia research.”

Not mentioned in the announcement: School officials had been alerted to allegations that Praticò improperly reused or altered images of mouse brains that accompanied his studies, thereby making his findings appear stronger than they really were.

The next month, a group of Praticò’s Temple colleagues wrote the university’s board of trustees to urge an investigation into the allegations, which had been posted on a public peer-review site called PubPeer. In their October 2020 letter, a copy of which was obtained by The Inquirer, the faculty members wrote that Praticò’s research contained “numerous instances of data incongruity.”

Since then, scientific journals have retracted four of Praticò’s studies. Critics on PubPeer, some of them anonymous, have raised questions about more of his studies, bringing a total of 36 under scrutiny.

Research misconduct has drawn increased concern in recent years, and in 2023, an analysis by the journal Nature found that more than 10,000 research papers had been retracted that year — meaning they were formally withdrawn due to deliberate fabrication, major errors, or other serious flaws — a record high.

Although that represents a small fraction of studies published, the consequences can be severe. In July, the president of Stanford University stepped down after a school inquiry found that images in his studies had been manipulated, though it determined that he did not personally engage in the misconduct. Also that month, federal officials found that former University of Pennsylvania scientist William Armstead had falsified or fabricated 51 figures in his studies, and he agreed to a seven-year ban on conducting federally funded research.

Asked about the allegations against Praticò, Temple officials said the university does not comment on internal investigations or personnel issues.

“Allegations of research misconduct are reviewed and investigated centrally through Temple’s Office of the Vice President for Research in accordance with university policy and applicable federal regulations,” school officials said in a statement.

In a lawsuit he filed Jan. 9, Praticò acknowledged flaws with two of the 36 studies and blamed them on a former graduate student, whom he accused in the suit of defamation and fraud. Citing the litigation, Praticò‘s lawyer said the scientist would not answer questions about specific criticisms.

“Dr. Praticò has no comment, other than he categorically denies engaging in scientific misconduct,” lawyer Christopher Ezold said.

A physician trained in his native Italy, Praticò, 63, joined the faculty of the North Philadelphia university in 2007 and drew prominent supporters for his work. He is the founding director of the Alzheimer’s Center at Temple, established in 2018 with a gift from university trustee Phillip C. Richards and a philanthropic group that he chairs.

In a ceremony at the time, Praticò thanked the trustee with these words:

“Extraordinary advancements in science happen because there is a core belief and commitment that we can overcome the odds.”

Among the first to raise questions about Praticò’s research was Mu Yang, the director of a center for testing the behavior of laboratory mice at Columbia University Medical Center.

Yang’s concerns began when a graduate student asked for help replicating one of Praticò’s experiments, she said in a phone interview. The Temple scientist had found that mice swam faster through a water-filled maze after they were given an experimental drug, an indication of improvement in the animals’ cognitive ability.

But upon reviewing Praticò’s published results, Yang thought they looked too good to be true. The animals’ progress over a four-day period yielded a graph with an unnaturally straight line. There was little to none of the variability that would be expected in how quickly various animals navigated the maze.

“This is not real data,” she recalled saying. “It was impossible.”

Yang had similar misgivings upon looking at the graphs in three additional water maze studies from Praticò’s lab. She consulted two other behavioral neuroscientists and found that they shared her concerns. In February 2020, the trio wrote letters to the journals that had published the studies and to the Office of Research Integrity at the U.S. Department of Health and Human Services.

In June 2020, agency officials told Yang and her coauthors that the matter had been referred to Temple for review.

Since then, critics on the PubPeer site have flagged concerns with dozens of other studies from Praticò’s lab. Run by a California-based nonprofit, PubPeer allows critiques of studies from scientists and nonscientists alike, provided they follow strict guidelines, use polite, neutral language, and submit supporting documentation where possible.

Some targets of these critiques have pushed back against the PubPeer model, saying they had been wrongly maligned by anonymous critics.

Yet one of Praticò's chief critics commented under her own name: Elisabeth Bik, a former Stanford University microbiologist who now works as a science integrity consultant for publishers, universities, and lawyers.

In dozens of studies, Praticò or his colleagues appeared to have digitally cut and pasted images and graphs from different experiments he had published in the past, Bik said in an interview.

“It suggests an intention to mislead, or a sign of sloppiness, at best,” she said.

In one study, Praticò and his coauthors included a pair of images of mouse brains with varying levels of inflammatory proteins.

But in looking through Praticò’s previous studies, Bik found one with what looked like a flip-flopped version of the same pair of images. In the text accompanying the earlier version, Praticò wrote that he had been conducting a completely different experiment: treating mice with a drug.

Using software to reverse an image, as Bik thinks was done in this case, would make it look like a new image from a new experiment, she said.

“Typically, something like that doesn’t happen by accident,” she said.

After some of the concerns on PubPeer were highlighted by a blog called For Better Science, a group of Praticò's Temple colleagues wrote the October 2020 letter urging the school to investigate.

One of the letter’s authors was John W. Elrod, who has collaborated with Praticò on two studies, neither of which was among those flagged on PubPeer. Asked about the letter and whether he still had concerns about his colleague’s work, Elrod declined to comment.

Though the status of Temple’s review of Praticò’s research is unclear, two academic journals have concluded their reviews, retracting a total of four of Praticò's studies between October 2022 and February 2023.

One of them, in a journal called Molecular Neurobiology, was the study in which Bik had spotted the brain images that looked like mirror images of those in the earlier study.

The journal formally withdrew that study in February 2023, though it remains visible online with the words “RETRACTED ARTICLE” at the front of its title.

In an accompanying notice, the publisher and a “handling editor” wrote that the image of the mouse brain looked “very similar” to the one that had been published earlier in a different journal. They also listed a dozen concerns with other images, writing that the figures appeared to have been reused or copied from elsewhere in the same study or from earlier studies.

“The Handling Editor and Publisher therefore no longer have confidence in the data reported,” they wrote, adding that “Domenico Praticò does not agree to this retraction.”

Yet in his lawsuit, Praticò acknowledges flaws in that study and one other, for which he blamed the former graduate student. That student “duplicated, altered, and improperly used data” and also copied images from external sources and passed them off as his own, Praticò alleged in the Jan. 9 complaint, filed in Common Pleas Court. The Inquirer is not identifying the student, who could not be reached for comment. He collaborated with Praticò on several of the other studies that drew criticism on PubPeer, though they were not cited in the eight-page lawsuit.

The three other retracted Praticò studies were in a different journal, Molecular Psychiatry — two of them in October 2022 and a third in January 2023.

Three of the four retractions cited concerns with images. The fourth was relatively minor, saying the authors had copied language from one of their own previous studies.

Thirty-two additional Praticò studies look potentially problematic to Bik, Yang, and other commenters on PubPeer, who have notified the publishers and are waiting to hear the results of any review.

Bik said she has no direct knowledge of actions by Praticò and his lab members. But generally, she says, the cutthroat pressures of academic publishing can tempt a person to cut corners.

“When the results look better, you’ll have a better chance of getting the next grant or the next scientific award,” she said. “I think that’s a slippery slope that as a scientist, you need to be careful of falling into.”

Whether flaws in research are intentional or the result of mistakes, many go undetected for years, said physician Ivan Oransky, cofounder of the science media site Retraction Watch. But the rate at which flawed studies are retracted is on the rise, he said.

“It’s pretty clear this is because of better detection of misconduct and error thanks to technology and a growing army of sleuths,” he said. “Whether misconduct is actually on the rise is less clear.”

Yet the pace at which they are investigated by journals and universities is too slow, said Yang, the behavioral neuroscientist at Columbia — and this concerns her more than the actions of any one researcher. While four of Praticò studies have been retracted, she said some remain under review at other journals, years after concerns were raised.

“Working through this case, I think I’m more concerned about the system that’s protecting him,” she said. “That’s taxpayers’ money. But nobody’s willing to do anything. That is a very painful realization.”