I agree with Esther12. There was no point in doing a blinded study with only 4 samples.
There was no point in publishing negative papers using no controls at all, or samples screened to remove people who had any sign of viral infection.
There was no point in not trying to replicate Lombardi et al.
There is no point in so long delaying an independently-blinded study with an appropriate number of samples, where WPI gets a chance to try to replicate its own work.
At this point, journals should go ahead and publish any positive papers which have been submitted, even if they think they are junk, because they have already published negative papers that are junk.
The way science is done is to look at the data, positive and negative, and figure out why it differs. Not to hold up studies because it bothers you that they differ. And not to come to instant conclusions,
whether it's "Oh! CFS is definitely not caused by a retrovirus! we know it to be associated to _____" (doesn't matter what, even if it's biomedical--good science always allows for new data),
or "Yeah! We found the cause!" (maybe not; it's a good model, but what in the body differs between people who get this MLV-associated disease and that MLV-associated disease? )
We still have many more questions than answers, and yet many (probably most) deem the case closed. Contamination only partially explains results in 1 labs (Shin/Singh; Singh says prostate cancer results are unaffected), and does not explain positive results in 2-4 labs (Lombardi, Lo, Cleveland Clinic, NCI?), and does not explain negative results in 1 lab (I forgot which negative study found contamination), and does not explain how all studies who were able to find XMRV (or MLVs) at all, find similar low prevalance in the healthy population (despite having completely different contamination routes, where contamination was found to be present). So contamination does not close the case.