Research integrity sleuths may have found a new red flag for identifying fraudulent papers, at least in cancer research: Findings about human cell lines that apparently do not exist. That’s the conclusion of a recent study investigating eight cell lines that are consistently misspelled across 420 papers published from 2004 to 2023, including in highly ranked journals in cancer research. Some of the misspellings may have been inadvertent errors, but a subset of 235 papers provided details about seven of the eight lines that indicate the reported experiments weren’t actually conducted, the sleuths say.
“Unfortunately, this just looks like a massive invention of data and experiments that probably never happened,” says study lead author Jennifer Byrne, a cancer researcher and data sleuth at the University of Sydney. Some of the nonexistent cell lines have already been cited in literature reviews and could confuse and mislead scientists conducting similar studies, she adds. “It’s a hell of a mess.”
Chao Shen, a cell biologist at Wuhan University and deputy director of the China Center for Type Culture Collection, a repository of human cell lines, hopes the findings gain attention. “These revelations underline the urgent need for concerted efforts to address the challenges posed by [these] cell lines to research integrity and reproducibility,” such as standardized reporting of cell lines, he says.
Problems with human cell lines had already drawn attention in recent years, as scientists discovered that many have been contaminated by other, more robust cell lines that corrupted results. But the new study reveals a different kind of flaw, stemming not from the misidentification of a known cell line, but from possible fabrication.
The study, published last week in the International Journal of Cancer, began as an examination of misspelled cell lines in papers about cancer research to determine whether the lines were contaminated or misidentified. Some of the misspellings may have begun as mistakes by inexperienced authors, Byrne says. But the team became suspicious about a subset of these papers that, in various ways, referred to the same seven cell lines as if they were not the same as similarly spelled, known cell lines. For example, some reported separate, different results from experiments that used the incorrectly and correctly spelled names of the same cell line.
These papers also had other red flags: They lacked a description of how the suspicious cell line was derived and didn’t provide its unique genetic fingerprint commonly used by researchers, which is based on short specific DNA sequences known as tandem repeats. What’s more, several papers identified three repositories from which researchers could purchase many of the seven cell lines, including the largest such resource, the American Type Culture Collection. But when the research team searched for the lines in their directories, none appeared.
The team eventually identified 235 such papers in 150 journals, including high-impact publications such as Cancer Letters and Oncogene. Most list authors in China who are affiliated with hospitals—a group previously identified as a source of customers for paper mills, businesses that sell authorship on papers that are often fake or shoddy, because they may lack research experience and have faced publish-or-perish pressure to gain professional promotion.
“It’s difficult to prove unequivocally that something doesn’t exist,” Byrne says, “but from our analyses, we’re pretty confident.”
She speculates that paper mill writers may have copied the misspelled names from otherwise legitimate papers, unaware of the correct spellings, instead of fabricating new names, because those might have attracted greater scrutiny from knowledgeable readers. The suspect characteristics are “just really unlikely to be made by a genuine researcher,” she says. “I worked with cell lines 30 years ago, and I can still recite their names.”
And these papers may be the tip of the iceberg, Byrne says. Since January 2023, more than 50,000 scholarly articles about human cancer cell lines have been published. Byrne’s team identified a total of 23 misspelled lines but limited its analysis to eight mentioned in 420 papers to keep the workload manageable. Her team plans further examinations, and she hopes others will as well. “We’re a small group, and going through these papers is very tedious.”