The mysterious link between the mysterious disease known as chronic fatigue syndrome (CFS) and the mysterious new virus called XMRV just got more mysterious (BBC).
XMRV was discovered by an American group in 2006 during a search for viruses linked to prostate cancer. Then, this October, another American group published in Science a strong correlation between CFS and infection with XMRV — 67% of CFS patients were infected with the virus, compared to 3.7% of controls. The paper made a big splash not only because it’s exceptional to find a correlation of 67% between a virus — or any infectious agent — and a chronic disease, but also because apparently about 4% of the healthy American population is walking around with this curious virus lurking around in their bodies. Despite these links, there is no evidence that XMRV causes either of these diseases.
But the link is completely missing on the other side of the pond. Researchers in Germany and in Ireland had a tough time finding a correlation between XMRV and prostate cancer, and now a group of UK scientists are reporting in PLoS-ONE that none of the CFS patients they tested carry the virus. The team boldly asserts they are “’one thousand per cent’ confident in their result” (Daily Mail).
There are a few possibilities to explain this discrepancy.
First, the tests in Europe may not have been sensitive enough to detect the virus. Ars Tecnica notes it has “never seen a DNA gel with that many empty lanes in a publication before,” but this should be a cause for unease — people don’t often publish wholly negative results (and one third of the lanes contained nothing but water). The PLoS ONE paper’s only positive “control” was the XMRV plasmid — not a sample from a human known to be infected with XMRV, or even different concentrations of plasmid to demonstrate sensitivity. As a result, it’s impossible to know whether their assay would have picked up the virus had it been there, especially since the Science paper showed that the virus was present at very low levels — only a small percentage of the cells tested had even one copy of the virus.
Another possibility is the positive tests could have been contaminated by a virus called MLV, which is highly related to XMRV and is common in any lab that handles mice. This is certainly a danger, but is tougher to swallow because the authors of the Science paper did a pretty thorough job showing contamination was unlikely, given the difference in the XMRV and MLV sequences, the results of negative controls and testing by multiple independent labs.
A third possibility is the virus is more prevalent in the US than in Europe. It would seem relatively simple to resolve the first two issues — all it would take is sharing a few reagents and a few vials of human samples. If only scientific collaboration was so simple.