Q&A: Why I delayed XMRV paper

After a weeks-long delay, a linkurl:paper;http://www.pnas.org/content/early/2010/08/16/1006901107.abstract reporting a strong association between the retrovirus xenotropic murine leukemia virus-related virus (XMRV) and chronic fatigue syndrome was published this week in the __Proceedings of the National Academy of Sciences__ (PNAS). The study, carried out by researchers from the National Institutes of Health and the Food and Drug Administration, found gene sequences pertaining to a closely related class of viruses, known as murine leukemia virus (MLV)-related viruses, in 86.5 percent of patients diagnosed with the syndrome, in contrast to less than 10 percent of healthy people.

Randy Schekman, editor in chief of PNAS

The path to publication took a few unexpected turns. A few days after the paper had been accepted in late May, last author Harvey Alter contacted the staff at __PNAS__ asking to delay its publication after having found that a linkurl:paper;http://www.retrovirology.com/content/7/1/57/abstract/ reporting no such association between the syndrome and the retrovirus was in the process of being published in the journal __Retrovirology__. The rival study, carried out by a team from the Centers for Disease Control and Prevention, had analyzed blood samples from 51 people with chronic fatigue and 56 healthy controls and found no evidence of an XMRV infection. The CDC team also opted for a delay to review their methodology. Their study was published online July 1st, three weeks after the original publication date, with no additional changes made. Chronic fatigue syndrome affects around one million people in the US. If a virus indeed has close ties to the disease, then hope may come in the form of antiretroviral treatments. __PNAS__ editor in chief linkurl:Randy Schekman;http://mcb.berkeley.edu/labs/schekman/index.html discusses a journal's role in making scientific knowledge available as soon as possible while still responsibly framing a public health debate. TS: Is there an established protocol that journals -- or at least __PNAS__ -- turn to in such situations? Randy Schekman: This is nothing routine. I wasn't aware of [Alter's] paper until I was called by someone in the retrovirus field who expressed concerns. So I alerted my staff in Washington and they told me they had just had a note from Alter requesting that the paper be held, pending discussions with various government agencies. I got concerned with that. And as I became aware of [the __Retrovirology__] paper, I decided on my own authority to solicit an independent review of the work from an established person in the field -- someone who has very high standards and who has not been involved in this controversy. He wrote a critique, which I then conveyed to Alter, which called for the paper not to be published until they could demonstrate that the virus' genes were integrated into the human genome. Alter then responded several weeks later saying he of course understood this. This was indeed the highest standard that would prove the case, but he felt that his data and the care that he had used in the selection and maintenance of the samples made his data very strong. He felt that by delaying this further, that would be a disservice to the community to hold the data that much longer. So then he sent us a revised version of the paper and I sent it to an independent member of my editorial board, another retrovirus expert. He said on balance he felt the paper should be published now. He agreed that this additional work was crucial but that the data were strong enough to be published now. And so we accepted it. TS: So would this be a case where delaying, as opposed to releasing peer-reviewed data immediately, is the responsible thing to do? RS: I think so. We're publishing a science journal here. My strongest feeling is that the data that we publish, we want to be of the highest caliber. We're not publishing a blog. Timeliness is important, but it is not the most important thing. Accuracy and the strongest evidence possible is what I feel we must uphold. TS: How could the process have gone smoother? RS: For me, and as far as __PNAS__ is concerned, where I feel we might have done a better job was in identifying how this particular paper might have caused [a controversy]. I was not aware of this paper until I got a call from this outside retrovirologist. If we're going to be faulted at all at __PNAS__ is that we weren't aware of the implications of this work until after it was already accepted. And we try. The staff at Washington, they're not trained scientists, but we try to examine the papers that we're reviewing to see if they have anything unusual about them that would cause additional publicity. We have cases of papers that come to controversial conclusions that are in a variety of disciplines. We have papers in our sustainability section in __PNAS__ which often touch on climate change issues. In this case, the paper was reviewed by two well-known retrovirus experts and the staff in Washington had no reason to question the work or bring it to my attention. TS: Anything else you'd like to add? RS: Alter went through considerable pains to try to eliminate [contamination] by using samples that were sealed for many years from patients who had chronic fatigue syndrome and where the signal that he detected was consistent over many years. It looks like he's done as good a job as he could without actually show[ing] that the sequences were contiguous with human DNA. That I trust will be what he does next. I would hope that he would do that next or someone would do that next before the NIH calls for clinical trials and certainly before patients start being treated with antiretroviral drugs off-label. That could be a very bad consequence of this. S.C. Lo, et al., "Detection of MLV-related virus gene sequences in blood of patients with chronic fatigue syndrome and healthy blood donors," PNAS, doi: 10.1073/pnas.1006901107, 2010. R. Schekman, "Patients, patience, and the publication process," PNAS, doi: 10.1073/pnas.1012027107, 2010. W.M. Switzer, et al., "Absence of evidence of Xenotropic Murine Leukemia Virus-related virus infection in persons with Chronic Fatigue Syndrome and healthy controls in the United States," Retrovirology, doi:10.1186/1742-4690-7-57, 2010. Correction (August 24): When originally posted, the article read that the PNAS study found XMRV genes in chronic fatigue syndrome patients. The genes analyzed are not particular to XMRV, but are shared by a closely related family of viruses known as murine leukemia virus (MLV)-related viruses. The Scientist regrets the error.
**__Related stories:__***linkurl:CDC paper downplays XMRV, chronic fatigue link;http://blog.the-scientist.com/2010/07/06/news-in-a-nutshell-5/
[6th July 2010]*linkurl:Viral cause for chronic fatigue?;http://www.the-scientist.com/blog/display/56048/
[8th October 2009]*linkurl:More talk of XMRV, chronic fatigue;http://blog.the-scientist.com/2010/06/28/news-in-a-nutshell-4/
[28th June 2010]