After enduring more than 2 years of criticism that included evidence of contamination and misrepresentation of data, a Science paper that linked a mouse retrovirus to chronic fatigue syndrome (CFS) today received its last rites: Editor-in-Chief Bruce Alberts issued a full retraction. The study's 13 authors in September signed a partial retraction after one of the three collaborating labs found that a contamination had marred its contribution, but they could not agree on the wording of the full retraction, so Alberts issued it without their approval. "Science has lost confidence in the Report and the validity of its conclusions," wrote Alberts in a rare "editorial" retraction, which appears in the 23 December issue of Science. "It is Science's opinion that a retraction signed by all the authors is unlikely to be forthcoming."
Researchers from the Whittemore Peterson Institute for Neuro-Immune Disease (WPI) in Reno, Nevada, led the controversial study, teaming up with investigators from the National Cancer Institute (NCI) and the Cleveland Clinic in Ohio. As they reported online in the 8 October 2009 issue of Science, they found evidence of a mouse virus known as xenotropic murine leukemia virus-related virus (XMRV) in the blood of 67% of the 101 CFS patients they analyzed. Alarmingly, 3.7% of controls also tested positive, leading to fears that XMRV could be widely contaminating the blood supply in many countries.
Soon after publication, researchers around the world began reporting that they could not find the virus in CFS patients. One group discovered that XMRV was likely created in laboratory experiments with mice that made an immortalized cell line to study prostate cancer, and another showed that variants of this line had evolved more than isolates of XMRV, exactly the opposite of what would be expected if the mouse virus truly infected humans and was subject to immune pressure. The U.S. Department of Health and Human Services organized a nine-lab study to assess whether the blood supply was at risk from XMRV or related mouse retroviruses. This so-called Blood Working Group—which included the researchers from WPI and NCI who co-authored the original Science study using whatever assays they chose—reported online 22 September in Science that no one could reliably detect the virus in previously positive samples from patients.
Alberts says the Blood Working Group finding was the final straw that led Science to request the full retraction. "The blood group study to me was dramatic evidence of poor science," says Alberts. "It gave us absolutely no confidence in the ability of the major labs involved to do the assays. I find that enormously disturbing." NCI's Francis Ruscetti, a prominent retrovirologist and one of the co-authors, attempted to coordinate a retraction with his colleagues but a dispute arose over wording that suggested some of the findings in the original paper were still valid. "We tried to get all of the authors to agree, but it got endless," says Alberts. "The responsibility that Science magazine has to the scientific community is to make a strong statement that we don't think anything in that paper can be relied on."
WPI's Judy Mikovits, who led the study with Ruscetti, says she and two of her contributing lab assistants refused to sign the retraction. The day after the publication of the Blood Working Group study, Mikovits presented new data at a CFS meeting in Ottawa, Canada, that purported to show evidence of human gamma retroviruses—the family XMRV belongs to—in patients. She essentially argued that the original paper focused too narrowly on one variant of XMRV. (She also showed a slide at the meeting that led to Science to discover that the original paper had a mislabeled image, which factored into the full retraction.) "We were confident of our data," Mikovits told ScienceInsider, explaining why they wanted to include a line in the retraction that said they still trusted their data and conclusions. Ruscetti refused to comment about the full retraction.
Mikovits was fired by WPI a week after the Ottawa meeting for insubordination and then accused in a civil suit by her former employer of misappropriating laboratory notebooks and computer data about her studies. Police at the University of Nevada, Reno, then filed a warrant for her arrest in relation to the allegedly stolen material, and she was briefly jailed. Both the civil and criminal cases are now being adjudicated.
Mikovits and Ruscetti are currently taking part in a multilab study coordinated by pathogen sleuth Ian Lipkin at Columbia University in New York City that will look for XMRV and related viruses in many more CFS patients than were analyzed in the Blood Working Group study. Mikovits says this $2.3 million study, funded by the U.S. National Institute of Allergy and Infectious Diseases, also factored in to the decision not to sign the full retraction. "We think it's premature to do anything before it's complete," says Mikovits, who estimates they will have results within 2 months.
Alberts strongly disagrees. "I think they should cancel that study," says Alberts. "It's over. They can't do the assays, so what's the point? Why should that give any different result than the blood group study? Maybe us retracting will help them scale back how much money they've spent on that. It seems like an incredible waste."
Researchers who have closely followed this saga and invested their own efforts into finding XMRV in CFS patients commend Science for issuing a full retraction. "It's very sad, but the writing has been on the wall for some time now—and the font size has gotten bigger over the course of the year," says retrovirologist Jonathan Stoye of the Medical Research Council in London, who co-authored an editorial in Science supporting the original paper. John Coffin, a retrovirologist at Tufts University in Boston who wrote the editorial with Stoye, says the full retraction could have happened much earlier. "It's kind of a surprise that it took so long," says Coffin.
Science Executive Editor Monica Bradford says the journal always prefers authors to sign retractions. "It's the authors' work, and it's a very clear signal to scientific community that there can't be accusations of other agendas," says Bradford. Alberts says they simply had been "spun" by the authors too many times for too long. "If our editorial retraction helps to end the resources to go into this fruitless endeavor, I think we've made a contribution to the scientific community," he says.
Updated, 3:35 p.m.: Paper co-author Robert Silverman of the Cleveland Clinic Lerner College of Medicine of Case Western Reserve University sent this statement to ScienceInsider:
I requested a full retraction of our findings this summer after discovering that the blood samples were contaminated. I was in favor of a retraction of the entire paper at that time. I am pleased to see that the Journal has now granted a retraction of the entire paper, and I agree with that decision.