News blog

Science raises questions about XMRV study – updated


XMRV is running out of legs to stand on. The journal Science, which published a controversial 2009 paper linking the retrovirus to chronic fatigue syndrome (CFS), has published two follow-up papers undermining the association, as well as an “editorial expression of concern” indicating that the association was probably the result of contamination.

According to the Wall Street Journal, Science has also asked the authors of the 2009 paper to voluntarily retract it.

“The request was almost inevitable at some point,” says Jonathan Stoye, a retrovirologist at the MRC National Institute for Medical Research in London who emerged as one of the staunchest critics of the link between XMRV and CFS. “It’s a bust.”

In 2009, a team led by Judy Mikovits at the Whittemore-Peterson Institute for Neuro-Immune Disease (WPI) in Reno Nevada, reported evidence of XMRV infection in up to two-thirds of people with CFS, which is also known as mylagic encephalomyelitis, and only a small percentage of healthy people. For a full rundown on the two-year saga, check out my recent profile of Mikovits and XMRV.

In that story, we reported on the two studies released today in Science. One paper, led by Vinay Pathak at the National Cancer Institute in Frederick, Maryland and John Coffin at Tufts University in Boston, suggests that XMRV was created in a laboratory in the 1990s and was unlikely to have infected people with CFS.

Pathak and Coffin’s team discovered viral sequences perfectly matching XMRV’s in two different mouse strains in which a prostate tumour biopsy was implanted to create a cell line called 22Rv1. “They just sort of snapped together like two puzzle pieces,” Coffin told Nature in March.

The second study, led by retrovirologist Jay Levy at the University of California, San Francisco, found no evidence for XMRV infection in 61 CFS patients, including some who tested positive for the virus at WPI or a company it owns that markets an XMRV test.

Levy’s team failed to detect the virus or an immune response to it using several different methods. Levy’s team also found sequences of mouse viruses related to XMRV in laboratory reagents, echoing similar findings by other groups. Notably, the patients in Levy’s study were referred by Daniel Peterson, the physician who co-founded the WPI and left the institute last year.

Based on these results, as well as a growing heap of studies failing to find any sign of XMRV, Science released an “expression of concern” from its Editor-in-Chief Bruce Alberts, a potential step towards retracting the paper.

A letter sent from Alberts to Mikovits’ team and obtained by WSJ has asked them to do just that. However Mikovits told WSJ in a statement that “it is premature to retract our paper.”

In a 30 May letter addressed to Science and forwarded to Nature by Mikovits, she stands by her group’s work and questions Science’s expression of concern and their request to retract her paper. “We feel this is an extremely premature action which is not in the best interest of the scientific community or human health and again we respectfully request that you allow the scientific process to run its course unhindered by bias,” she writes.

The latest Science papers are not the final word on XMRV. The US National Institutes of Health is sponsoring two studies in which several different labs are testing blinded samples from CFS patients and healthy controls.

One, known as the Blood Working Group, should have the results of tests for XMRV in about 30 people with CFS by this summer. While a second, larger study, coordinated by Columbia University virologist Ian Lipkin, should be complete by year’s end.

Stoye, who co-authored an enthusiastic preview of the original paper linking XMRV to CFS questions whether both studies are needed to come to an increasingly obvious conclusion. “Perhaps none of us were sufficiently critical early on, but who could tell? I think we have to move on,” he says.

Micrograph of XMRV via Wikimedia Commons


  1. Report this comment

    Kelly Latta said:

    The XMRV research may be a bust, however like Dr. Ila Singh and colleagues a few weeks ago, Dr. Levy (Knox et al 2011)recommends that virological studies into the etiology of ME/CFS continue. He cites the infectious nature of the disease.

    Perhaps more telling at this point than his XMRV study will be Dr. Lipkin’s metagenomic study on ME/CFS patients.

    As an aside, according to Professor Vincent Racaniello, it is rare for a journal to retract a study unless there is fraud, which he is careful to point out is not the case here.

  2. Report this comment

    gerwyn said:

    how can science publish a study where the author (levi)

    did not even use the same primer combinations as Lombardi et al in the reverse transcription PCR assay and used different annealing temperatures.Both are crucial variables in determining the success or failure of a PCR reaction.

    How can they publish a study regarding the presence of pre XMRV in strains of mice when the authors provide no evidence that these strains were used in the creation of the 22rv1 cell line

    I would invite the authors to deny that these mice were not in fact used in the creation of the cell line

  3. Report this comment

    Alex said:

    “(…) we respectfully request that you allow the scientific process to run its course unhindered by bias.”

    … I agree? 😛

  4. Report this comment

    pimsainnum said:

    Based on this genetic analysis, the scientists concluded that XMRV was not present in the original prostate tumor samples but arose only after they had been put into mice. The probability that an identical recombination event occurred independently is about 1 in 1 trillion, making it extremely unlikely that XMRV arose from another source. The researchers concluded that the association of XMRV with human disease is due to contamination of samples with virus originating from this recombination event.

    “After the reports of XMRV in human prostate cancer, and later of XMRV in people with CFS, retrovirologists all over the world were excited to explore its role in human infection and disease. The results published today are not what we would have expected, but due to the time and resources dedicated to the understanding of this virus by researchers at NCI and NIH as well as others, scientists can now concentrate on identifying the real causes of these diseases,” said Pathak.

  5. Report this comment

    Jane Clout said:

    At the top of this page second para, you will find a link to Annette Whittemore’s letter to Dr. Bruce Alberts and Ms. Monica Bradford, Editors of Science, in response to this event. The letter gives a table showing the differences between those high-profile studies that have failed to find XMRV and the Lombardi 2009 study.

    Clearly, we are still waiting for a replication study. Until then, as Dr Mikovits says, calling for retraction is premature and feels to me more like politics than science.

  6. Report this comment

    Michael Lerman said:

    The saddest part of this story is that NIH/NCI are going to spend more time and more money to replicate the original “discovery” to convince the NCI team to back off and retract the Science paper. Science itself made a mistake publishing this paper and has now the right to “disown” it. The reviewers of the manuscript should explain their recommendation to publish.

    In ancient Iran the physician who treated the deceased patient was required to walk upfront the funeral procession.

  7. Report this comment

    Richard said:

    It’s time consuming for scoence to own a unhindered future,just like the high-technology,but we indeed are sure that it will get there in the near future.

  8. Report this comment

    Ed Rybicki said:

    I can’t help thinking that folk are missing the point here: step back a bit, forget how “Science shouldn’t have published the original paper” or what a tragedy it is that more money will be sent after money “wasted” on finding XMRV originally – and see science WORKING, as it should!

    That is, something is published which is contentious, which is then vigorously disputed, and is then either disproved, or not – out in the open.

    Not publishing the original study would be a perfect example of paradigms not shifting because people don’t think they should – rather than because an evidence-based study said they should, WHETHER IT WAS WRONG OR NOT.

  9. Report this comment

    Brian Hildebrandt said:

    Good point Ed. Most too often researchers tend to get married to their theories rather than looking at the problem with a more objective mind.

    We all have our selected biases but some can be more easily swayed by financial or political interests.

Comments are closed.