January 11, 2006
Has Schatten Revealed All?
University of Pittsburgh stem cell researcher Gerald Schatten played a larger role in authoring the bogus Korean stem cell paper than previously revealed, according to the today's Pittsburgh Tribune-Review. Schatten helped Hwang Woo-Suk and his Korean colleagues respond to peer reviewers' questions about the 2004 paper that announced the initial breakthrough in human cloning. Schatten's name didn't appear on that paper.
But he co-authored and served "as an adviser to help analyze, interpret and write up the results for the 2005 Science paper," which announced a huge breakthrough in the technology's efficiency to worldwide acclaim. Schatten had previously said he served only as a conduit to getting the paper published in the prestigious journal, which is published by the American Association for the Advancement of Science.
Today I took another look at the latter paper as well as the brief communication that appeared in Nature last August announcing the cloning of Snuppy the Afghan hound (Schatten and Hwang were co-authors on that paper as well). I discovered something interesting: neither Schatten nor Hwang declared any financial conflicts of interest even though both scientists had applied for patents on related technologies.
Schatten's failure to disclose a U.S. patent application he filed in April 2003 appears to be a clear violation of Nature's policy. The patent is titled "Methods for producing transgenic animals." The Nature article is headlined "Dogs cloned from adult somatic cells." The Nature conflict of interest disclosure policy states that authors must disclose "patents or patent applications whose value may be affected by publication."
It's less clear whether Schatten's failure to disclose his patent applications in Science represented a violation of that journal's policy, which states the author should disclose "a planned, pending, or awarded patent on this (emphasis added) work." Schatten's other pending patent application involves "methods for correcting mitotic spindle defects associated with somatic cell nuclear transfer (SCNT) in animals" including humans. This might prove useful to SCNT some day, but it isn't SCNT itself. Under the Nature policy, this would have to have been disclosed.
Based on statements he made at press conferences last year and his patent applications on file at the World Intellectual Property Organization, Hwang clearly violated both journal's conflict of interest disclosure policies. The latest press reports out of Korea suggest Hwang may be under criminal investigation for the scientific fraud that went on in his lab.
Earlier this week, Science editor Donald Kennedy pledged to overhaul the journal's scientific review policies in the wake of the stem cell scandal. "We are doing a systematic review of the editorial history of both papers and our procedures for evaluating them, to search for ways in which we might improve those," he said in a prepared statement. "I have pointed out in the past that even unusually rigorous peer review of the kind we undertook in this case may fail to detect cases of well-constructed fraud."
Just as the tightest peer review can't catch deliberate fraud, a conflict of interest disclosure policy without teeth is no policy at all. Science should tighten its conflict-of-interest disclosure policy to require authors divulge all relevant conflicts, not just the ones directly related to the paper at hand. It should also publish them all alongside the paper. And it should put authors on notice that if they fail to disclose those conflicts -- as appears to have happened in this case -- it will result in their being banned from publishing within the journal's pages for, say, at least three years.
As we've seen in this case, a policy without meaningful sanctions for violators is no policy at all.