Virus previously tied to syndrome was lab contaminant, studies suggest

By Laura Sanders

Web edition: September 22, 2011
Print edition: October 22, 2011; Vol.180 #9 (p. 5)

A meticulous study has failed to confirm a connection between chronic fatigue syndrome and a family of viruses that includes XMRV. Nine laboratories — including the two that originally identified a link — could not reliably detect the viruses in blood cells from patients with the mysterious and controversial condition, researchers report online September 22 in Science.

In another blow, one of the labs that contributed to the original study retracted its key result because some of the samples used in the research were found to be contaminated with genetically engineered DNA from XMRV. The partial retraction also appears online September 22 in Science.

Together, the new data largely exonerate the virus as a cause of chronic fatigue syndrome — a disorder with a constellation of symptoms, no known cause and no effective treatment.

“The data certainly contradict the original findings,” says infectious disease specialist Michael Busch of the University of California, San Francisco and the Blood Systems Research Institute in San Francisco, who coordinated the new nine-lab study. “To me, it demonstrates that these labs have serious problems with false positive results and that those earlier data are not credible.”

For some researchers, the end to the XMRV saga can’t come soon enough. “The whole scientific world pretty much knows this whole thing is hocus-pocus,” says Robert Gallo, director of the Institute of Human Virology at the University of Maryland School of Medicine in Baltimore. “Basically, it’s not real.”

In 2009, researchers reported in Science that two-thirds of blood samples from people with chronic fatigue syndrome tested positive for the mouse leukemia virus XMRV, compared with only 3.7 percent of samples from healthy controls (SN: 11/07/09, p. 13).

In addition to hinting at the cause of chronic fatigue syndrome, the finding also raised serious concerns about the safety of the blood supply. If the virus infected a substantial fraction of the population, it could turn out to be widespread in donated blood.

Busch set up the nine-lab study to assess the potential threat to the blood supply. In the study, scientists studied blood samples from 14 people with chronic fatigue syndrome that had previously been found to harbor the virus, and 15 healthy people. The samples were blinded and mailed out to the nine laboratories, where the blood was put through a battery of techniques to find the virus.

Seven of the labs found not a whiff of the virus in any of the specimens. The groups that had previously published the virus-CFS link did turn up some positive samples. But both these groups reported inconsistent test results on the same samples. More importantly, these two groups found that the virus didn’t show up any more often in samples from people with chronic fatigue syndrome than in samples from healthy participants.

“Their positives are scattered randomly between cases and controls, leaving you with the inescapable conclusion that this is a fairly high level of false positives,” says John Coffin of Tufts University Medical School, who is a coauthor on the new paper. He says that the idea that the virus is tied to chronic fatigue syndrome is “fading fast.”

The partial retraction resulted from a reanalysis of DNA sequences from blood samples of people with chronic fatigue syndrome and people without by two researchers involved in the original study. Robert Silverman and Jaydip Das Gupta, both of the Cleveland Clinic Foundation, reanalyzed seven samples that had tested positive for the virus in the original report. Just as before, tests were positive for the viral DNA. But the viral DNA wasn’t ensconced in the cells’ DNA, as the researchers had initially assumed. Instead, the virus came from a piece of artificially introduced DNA called a plasmid that was unmistakably lab engineered. In the samples that had originally tested positive, the XMRV came from the artificially introduced DNA — not the patients’ cells.

It’s not clear how or when the contamination occurred. “The contaminated samples originated outside of the Cleveland Clinic,” a statement from the Cleveland Clinic reads.

Still unclear is how the original work found such a high percentage of samples with the virus in the chronic fatigue samples and such a low percentage in the healthy samples. A high rate of random false positives caused by a contaminant should have been the same for both groups. Instead, it appears that the chronic fatigue samples were preferentially affected.

“There is no way of knowing how the plasmid got into those particular samples,” says Judy Mikovits of the Whittemore Peterson Institute and University of Nevada in Reno, a coauthor on the original and the new papers. She says that the two-thirds finding may be called into question. But even considering the contamination problem, Mikovits says that the team still has evidence of viral infection in chronic fatigue patients.

She cites several reasons that the recent multi-lab project may have not detected the XMRV virus: Virus loads may fluctuate over time in CSF patients, the virus could hide out in tissues other than blood and the amount of virus present may be too low to detect. What’s more, other viruses similar to XMRV may be to blame. She and her team are aggressively pursuing the question, she says.

Virologist Ian Lipkin of Columbia University, who was not involved in the studies, says that the case is not closed yet. He is leading another large collaborative effort to test for the presence of retroviruses, including XMRV, among different populations of people with and without chronic fatigue syndrome. “We don’t want to have to revisit this, and there’s a risk of that,” he says. He and his colleagues anticipate results by the end of 2011. “I don’t know what we’re going to find,” he says. “I have no horse in this race.”

---

Back Story – A theory’s birth and death

October 23, 2009
XMRV turns up in two-thirds of blood samples from people with chronic fatigue syndrome, researchers report in Science.

August 23, 2010
Another group reports online in the Proceedings of the National Academy of Sciences that it has found high rates of viruses related to XMRV in people with chronic fatigue. Other scientists try to replicate the findings, without success.

May 31, 2011
At least 10 independent studies have failed to find XMRV in patients with chronic fatigue syndrome. Two online publications in Science cast still more doubt: One traces XMRV’s origins to a lab-engineered mashup between two mouse viruses in the 1990s. The other finds no evidence of XMRV in 61 people with chronic fatigue syndrome. Science publishes an “Expression of Concern” about the 2009 study.

September 22, 2011
Seven of nine labs fail to find virus in samples from chronic fatigue syndrome patients, researchers report in Science. The two others are the same labs that had previously found a link, but their hits are not clustered in patients with chronic fatigue. The authors of the 2009 Science paper retract a key result after it is shown to have been due to contamination.

End of 2011
Another large-scale study of retroviruses in chronic fatigue syndrome patients, using a new group of subjects, is slated to conclude.