Chronic fatigue syndrome has long been controversial and mysterious. Is the mystery clearing up, or getting deeper? There have been diagnoses of something like CFS for a long time, under a lot of different names. The common sign is persistent fatigue with no obvious physical cause, often accompanied by joint pain, disrupted sleep, and other symptoms. It's more common in women than in men - but then, so are a lot of autoimmune disorders, which has made some sort of immune syndrome a popular explanation. All sorts of contradictory data have been generated around that idea, but nothing convincing has emerged.
There's a preprint in Science from teams at the National Cancer Institute, the Cleveland Clinic, and Whittemore Peterson Institute that's attracting a lot of interest. It presents evidence for a viral infection which is far more common in patients diagnosed with CFS. What's even more intriguing is that the virus (XMRV, a mouse retrovirus) is already one that's suspected of involvement in some cases of prostate cancer, as shown by analysis of biopsy samples. (Commentary on that work here). About two-thirds of the CFS patients were found to be positive for the virus, as opposed to about three per cent of the control group. The WPI people are now saying that since the manuscript went in that further work has shown 98% of a 300-CFS-patient sample as positive for XMRV. More on that below.
In the case of the prostate patients, there seems to be a link with a deficiency in the RNAse L pathway, which is part of the interferon-induced antiviral response. It may be that patients with this immune system vulnerability are more susceptible to infection by XMRV, which then goes on to cause (or exacerbate) prostate cancer. There may be a link between RNAse L function and a diagnosis of CFS as well. It makes a neat story, and I hope that it's true.
But we're not quite there yet. No one's seen the data yet on that 300-patient cohort mentioned above, and it's not clear if a different diagnostic method was used on them compared to the group in the Science paper. And that paper itself doesn't have enough details on the patients to satisfy some readers - a specialist at the CDC complained about this to the New York Times, and said that his team would try to reproduce the results, but that he wasn't hopeful. (Working on chronic fatigue has not been the sort of thing that breeds a hopeful outlook, to be sure). Other researchers in the field have voiced their doubts to Science (who, to be sure, did accept the original paper).
One of the problems in this area has been defining who's a patient and who isn't. It's a bit of a catch-all diagnosis, or can be, so there's always the suspicion that even if there's a solid underlying cause that the data are hard to dig out of a heterogeneous patient sample. And there's the whole psychological-or-physical question, too, which is a sure route to raised voices and waving fists. My thinking is that there are very likely a number of people with other issues (which I will leave undefined) piled into this area, and that the necessary attempts to draw boundaries will be sure to leave someone upset.
As for this retrovirus angle, there are a number of other steps that need to be taken. Looking over historical blood and tissue samples will be very interesting - could you find that a person showed no sign of the virus when younger, then went positive before showing signs of the disease? Or does it stay latent for a longer period before finally breaking through? Are there animals that are susceptible to infection, and do they show similar symptoms to humans?
Can we at least demonstrate infection of cultured cells in vitro? (Update: I see that they've shown that, which is a very good step). Do any of the existing antiretroviral drugs have any effect on either of those processes, and if so, what happens when you give them to patients with CFS? What about the 3% or so of the population that seems to be positive for XMRV but shows no sign of either prostate cancer or CFS - what's different about them, if anything? And so on.
The Whittemore Peterson Institute people are way out in front on these questions, for better or worse. You may have said, as I did, "Who they?", but it turns out that they've only been around since 2004. The institute was set up by the parents of a CFS patient to do research in the field, and they've apparently been quite busy. Their web site gives the impression that the question of CFS as a retroviral infection is basically settled, but I'm not there yet. I have a lot of sympathy for the unidentified-infectious-agent line of thinking, and I believe that there are probably several things out there that will eventually fit into this category, but it can be a hard thing to prove. Let's hope this one is solid, so we can get to work.