I have some thoughts on the Shin et. al. study.
The first is that I concur this is a solid study. We know the reputation of several of the scientists in the study, and Singh especially is acknowledged as an MLV expert. Here are the main discrepances from the study that I can see (I am not doubting the study results, just comparing to Lombardi et. al. and Lo et. al.).
Shin et. al. have shown, again, that something is very wrong with the studies so far. They have shown that even good testing for XMRV can fail. However, even though they have been very careful, they cannot prove that contamination has occurred - nobody has yet found the contamination source, nor definitively shown how the controls have low levels of positives. They do make one very important point on this though: potential sources of contamination that not usually checked need to be looked at. One that I would like to put forward, again, as a potential contamination source is people - lab workers, researchers, cleaning staff etc.
Singh is not saying XMRV is not real, and does not infect people - quite the contrary in my view. She is saying that she cannot find an association with XMRV and ME/CFS. The association with cancers, including now breast cancer, is growing however.
Now further unpublished studies may take a different view on this, I really wish some of them could be made public to help fight the contamination claims, presuming they can do so of course. It has always been possible that XMRV is either not associated or not causal. We just need to know for sure - this is too important to give up on just because the science is getting difficult. The worst case scenario is NOT that XMRV may not be associated with, or causal for ME/CFS, but that it may be causal and we abandoned it too early, perhaps not returning to it for a decade or two.
I do wonder if XMRV/PMLV contamination is not from people handling the samples or working in the labs. If XMRV is present in even only a couple of percent of people, any lab that employs one of them might be a source of contamination, and so labs with infected employees "find" XMRV, and labs that do not have infected employees don't find XMRV. I am not saying this is the case, but it is a source of contamination that is not always tested, although I do know that at least one lab has done so.
I agree with those who have been saying that the BWG and Lipkin studies will end the debate on contamination. I am not sure of the timeline on the BWG, but Lipkin is potentially years from completion. Having to wait that long for definitive confirmation of XMRV association is not a pleasing prospect.
Things that would help enormously include:
1. Tissue studies - why not run a few simple biopsies? Singh is very good at finding XMRV in tissues. Several people have made this point recently, and many of us have been saying this for many many months. Its kind of obvious. If Singh cannot find any XMRV in a fair number of ME/CFS patient biopsies, that would be very negative for XMRV association, far more negative than a blood study.
2. Someone commented on going back to tissue positives (prostate cancer patients) and validating their XMRV blood tests on these patients. It is very likely that something like this will be the only way to properly validate tests for some time to come, but I do wonder if there might not be something different between ME/CFS and prostate cancer patients that might make such testing problematic.
3. Publication of some of the positive studies that are rumoured is important. Several people have commented in recent months, and I concur, that publishing a few of these could really help. Not all of them need to be published, and they do not have to be published in high impact journals - one or two select papers in a very minor journal, just to get the other side of the story out there, could make an enormous difference. One of the possible problems with getting positive papers published is they are trying to get these published in high impact journals, and are finding this difficult.
4. Nagalase is not the only thing we should be looking at closely. We need a study on finding reverse transcriptase in ME/CFS patients. While there might be viruses or other pathogens that make nagalase, not that I know of any, reverse transciptase is a defining feature of retroviruses. A high prevalence of reverse transcriptase would by definition prove an association between a retrovirus and ME/CFS, though not specific for XMRV or PMLVs. A statistical correlation with nagalase, reverse transcriptase and low NK cell function would be even better. Of course while they are looking at biopsies, why not store some for another study on enteroviruses etc. An ME/CFS gastrointestinal biopsy bank is a good idea.
I am not committed to XMRV being associated or causal for ME/CFS. I am committed to the view that we need to know for sure. I never ever want to have to read that we found the cause in XMRV and then let it languish for a couple of decades. That would be much harder to take than the news that XMRV is not associated with ME/CFS.
Bye
Alex
