All Stem Cell Data Fake, South Korea Panel Finds
South Korean cloning pioneer Hwang Woo Suk never made any of the 11 stem cell lines he claimed were derived from the DNA of sick and injured patients, an expert panel investigating the controversial scientist said today.
Roe Jung Hye, Seoul National University’s research chief, said the panel could find no evidence to support any of the claims in a blockbuster study Hwang published in May. The research was hailed at the time as a milestone in treating patients with spinal cord injuries and diseases such as Alzheimer’s, Parkinson’s and diabetes.
Last week, the panel announced that nine of the 11 stem cell lines Hwang claimed to have created did not exist, and it said today that the other two were actually human egg cells from a Seoul hospital.
As the panel continues to examine whether Hwang succeeded in producing the first human embryo clones or the first cloned dog, scientists, patients and others around the world are contemplating a more fundamental question:
Why did he do it?
“That’s one of the great mysteries,” said Marcel C. LaFollette, a science historian and author of “Stealing Into Print: Fraud, Plagiarism, and Misconduct in Scientific Publishing.” “It does seem to boggle the mind. This is a high-wire act of the most astonishing kind.”
Scientific fraud has been around almost as long as science itself. Perpetrators are driven by a variety of factors, including the need to secure research funding, win tenure and gain fame and recognition from their peers.
“Publishing is absolutely vital for your career, it’s vital for your prestige, it’s vital for your income, and it’s vital for your funding,” said Dr. Harvey Marcovitch, chairman of the Committee on Publication Ethics, a London-based group that advises scientific journal editors on ways to combat fraud in publications.
Sometimes, scientists do it simply because they’re convinced their theories are correct and they are too impatient to take the time to prove it.
Take the case of Dr. William Summerlin.
As a young researcher at Sloan-Kettering Institute for Cancer Research in New York in the 1970s, Summerlin claimed he could transplant skin from black mice onto genetically unrelated white mice, a feat with enormous implications for medicine. The trick, he said, was to grow the tissue in culture for four to six weeks to prevent it from being rejected by the recipient.
But try as they might, Summerlin’s colleagues could not repeat his experiments. Then they began challenging his results. So, to bolster his case, he used a black felt-tip pen to darken the grafts on two of his white mice.
The proof dissolved when lab assistants washed away the black patches with alcohol.
“Scientists are not disinterested truth seekers,” according to David Goodstein, vice provost of Caltech and an expert on research ethics. “They are more like players in an intense, winner-take-all competition for scientific prestige and the resources that follow from that prestige.”
Given the high stakes of big science, some experts say, they are surprised fraud isn’t uncovered more often.
Government agencies that fund scientific research track fraud, and they have found that the number of reported cases is “pretty consistent” from year to year, said Mark Frankel, director of the Scientific Freedom, Responsibility and Law Program at the American Assn. for the Advancement of Science in Washington, D.C.
Of course, it’s impossible to know just how much trickery is going on. Cases of fraud can be as subtle as omitting inconvenient data or fudging a statistical analysis. Even when the data are real, they can be manipulated to support conclusions that aren’t scientifically warranted.
A survey of 3,247 scientists published in June by the journal Nature found that 33% admitted to engaging in some sort of questionable research practice, such as failing to present contradictory evidence or changing the design of a study to satisfy a funding source. Only 3% said they had deliberately falsified research data.
Incidents as egregious as Hwang’s in South Korea are even more rare, experts said.
“This is maybe the worst one I’ve come across,” Marcovitch said.
He said he wasn’t surprised that a case like this came up in the high-profile field of stem cell research, which has captivated government health ministers, venture capitalists and patient-advocacy groups.
“Vast amounts of money will be poured into stem cell research in the coming decades because everyone thinks that’s the future of medicine,” Marcovitch said. “The temptation [to cheat] is great.”
No one may ever know why Hwang -- who became a national hero in South Korea after announcing a string of breakthroughs in some of the world’s most prestigious scientific journals -- committed such a blatant deception. The 53-year-old professor has admitted some errors in his work and offered to resign his post at Seoul National University, but he continued to insist his research results are real.
“I sincerely apologize to the people for creating a shock and disappointment,” Hwang told reporters last week.
Today Roe said, “The panel couldn’t find stem cells that match patients’ DNA regarding the 2005 paper, and it believes that Hwang’s team didn’t secure scientific data to prove” the cells were made.
Hwang’s first blockbuster paper in 2004 detailed the creation of the world’s first human embryo clones, one of which was used to derive stem cells. In May 2005, he claimed to have repeated that feat using DNA from sick and injured patients to create 11 customized stem cell lines -- a development that seemed to pave the way for creating tailor-made tissue repair kits for patients.
To complete his scientific trifecta, Hwang unveiled the world’s first cloned dog, a frisky Afghan hound named Snuppy in August.
All those accomplishments began to unravel last month, when Hwang’s collaborators and others in South Korea questioned some of the photographs and DNA fingerprints included in his studies.
Hwang’s deception may have slid past the usual gatekeepers of scientific accuracy -- the peer-review committees that vet studies for scientific journals -- because so many people think the achievement is possible.
Hundreds of researchers had been scrambling to create stem cells from cloned human embryos, a technique known as therapeutic cloning.
Hwang may have been counting on one of them to succeed where he could not, thereby seeming to corroborate his results, experts say.
So when Hwang’s stem cell papers were published in Science, his fellow researchers saw little reason to doubt them.
Neither did the public, which had put its faith in the power of embryonic stem cells to cure a variety of intractable diseases.
“We want to believe it,” Marcovitch said. “We all want to believe that somebody has worked out the basics of how to produce stem cells.”
A similar kind of optimism masked problems with a string of stunning papers from Jan Hendrik Schon, whose work at Lucent Technologies Inc.'s Bell Labs, starting in the late 1990s, had scientists speculating about a Nobel Prize for the German physicist.
In a flurry of publications, Schon described a series of breakthroughs in semiconductor physics.
His data matched up with theoretical predictions, and his results were quickly embraced. As many as 100 laboratories around the world attempted to duplicate his results.
But none were able to do so, prompting suspicion among other physicists.
On closer examination, they noticed the same graphs purporting to show results from different experiments. In other cases, the data seemed too perfect to have been produced experimentally.
A committee convened by Bell Labs concluded that Schon had simply extrapolated his results from physics theories without conducting any experiments.
After Bell Labs fired Schon in 2002, he continued to insist his experiments -- and his results -- were real.
Like Schon, Hwang may have been clouded by a fervent belief in the truth of his scientific conclusions, even if he couldn’t prove them, said LaFollette, the historian.
“What’s puzzling,” she said, “is if they really did intentionally deceive, how did they think they were going to get away with it?”
The answer is simple, said Goodstein of Caltech.
“Nobody ever cheats for the sake of cheating,” he said. “They cheat for the sake of getting into the record things they believe to be true. They think it will be proved right eventually.”