Hi omer, I'm not sure the following is clear because I'm reeeally sleepy, but for what it's worth:
It's true that because this study only used an XMRV positive cohort of CFS patients it cannot draw definitive conclusions about the specificy of this cytokine profile to XMRV infection. [However, just because it cannot do so definitively does not mean that it does not provide evidence in support of that conclusion; the fact that the particular bias of the profile is consistent with retroviral infection is evidence that this profile is specific for XMRV positive CFS.]
The value of this study, aside from revealing an underlying cytokine pattern, is as a crucial first step to developing a diagnostic tool for XMRV+ CFS (and perhaps, in the future, other forms of X-associated disease), which requires very accurately defining the terms used. There are serious scientific problems with trying to generalize any conclusions to traditionally defined CFS, as it is not a rigorously defined term. So the WPI has here established a foothold for XMRV by characterizing "XMRV+ CFS" patients, in this case by immune parameters.
For that reason the scope of this study was narrower; it confined itself to identifying a cytokine profile that is predictive of at least XMRV+ CFS. This approach is understable given the difficulty of being certain that someone is X negative; given those circumstances, this was a rational first step in developing a potentially diagnostic profile, which requires initial characterization with a tightly defined cohort.
Does this study 'prove' that the cytokine profile it has found is caused by XMRV? No, but that was not its intention and at this stage it would not be expected to. Even if an XMRV negative Fukuda CFS cohort had been included as a control, and had different results, we wouldn't be able to conclusively say that XMRV is the determining factor because Fukuda CFS is such a heterogeneous label that the difference could be explained by other variables. If instead a CCC, acute onset CFS cohort was to be used as a control, there would be the problem of finding true XMRV negatives in that cohort. How would you find, say, 100 such patients who are XMRV negative if by a combination of WPI assays only 5% or so of such patients are XMRV negative.. and we don't know for certain if even those are true negatives? As a result, you couldn't be certain that there aren't XMRV infected individuals in such a cohort, and therefore the value of its overall immune profile as a control would be questionable. So to initially develop a baseline from which to assess the sensitivity and specificity of the profile as a diagnostic tool, this study had to exclude such cohorts.
What this study does do, like the Lombardi 2009 study, is document an association between XMRV and a particular characteristic (in this case a particular cytokine/chemokine profile in CFS patients). It makes no comment about the nature of this association, such as if XMRV causes the profile. It is a necessary first step to clarifying that issue, however. Such clarification will not be easy, though, as I mentioned above, because of the difficulty in isolating a cohort of well-defined CFS cases who are truly XMRV negative.
In this case, it is a cytokine/chemokine profile, and the association is so strong that the profile generated can be used to predict XMRV/CFS status with high accuracy as opposed to healthy controls. We do not know yet what happens when it is used on CFS populations defined by one criteria or another.
Had they been able to find an XMRV negative cohort of CFS patients and test them as well, they might have been able to make a more definitive association between XMRV and this cytokine profile, or make a strong case for the abandonment of vague Fukuda or Reeves cohorts, as well as build evidence for a causal role for XMRV in X+ patients. That was all outside the scope of this study...However, I guarantee you that such studies are in the works. This is just a first step, and you can see some of its value within the context of other WPI research; for example, one way to look at it is as an extension of the Science paper: a further investigation of the CFS patients found positive for XMRV as compared with healthy controls revealing a distinct cytokine profile. The WPI clearly has more data which they will publish soon enough that will investigate the potential causal role of XMRV in generating this cytokine profile.