Several factors may contribute to the varied detection of XMRV in different populations, including geographic distribution, patient selection, analyte choice (DNA, RNA, antigen), and detection methodology. Geography may play a pivotal role. Henrich et al [24] studied samples from immunodeficient patients who were well characterized at facilities in Boston and found no evidence of XMRV, whereas Lombardi et al [6] studied similarly affected patients from a distinctly different geographic region and found substantial rates of XMRV. Assay sensitivity and specificity may certainly affect detection. Reports of no XMRV detection, including the 2 in this issue of the Journal have superb sensitivity, generally in the range of 1–10 copies per cells or 10–100 copies/mL plasma. However, viremia may be chronically low (as it is in HTLV), transient, or episodic, complicating detection. XMRV detection is likely to be especially sensitive to potential false positive results by contamination with mouse‐derived material. Many of the hundreds of mouse endogenous retroviruses present in the mouse genome may amplify with XMRV primers; as such, a few copies of the mouse genome may represent a substantial source of contamination.
The new studies in the Journal highlight the following measures needed to resolve conflicting reports on detection of XMRV:
1.
Standardization of detection assays. This aspect is currently being addressed by a consortium of laboratories that will develop clear performance characteristics for testing.
2.
Prospective epidemiologic surveys. In response to the initial identification of XMRV in patients with prostate cancer, the search for XMRV infection was driven by a few clues and the availability of local sample sets. Such “molecular geocaching” is an entirely appropriate first response, but now larger, prospectively designed and appropriately powered epidemiologic studies with longitudinal sampling and standardized sample processing are in order. The data from existing studies, especially power analyses from Henrich et al [24] presented in this issue of the Journal, represent useful starting benchmarks.
3.
Sharing reagents and samples. Because XMRV has been reported in otherwise normal individuals, there are no a priori, gold standard XMRV‐positive or ‐negative populations. A working set of standards will require sharing samples among groups to independently confirm results. This painstaking process will yield a reference panel that would be an invaluable resource to the research community. Researchers have already contributed a number of XMRV molecular biology reagents for general use through the National Institutes of Health AIDS Research and Reference Reagent Program. The establishment of a reference panel will propel research in this field.
4.
Comprehensive and rigorous phylogenetic sequence analysis. Standard phylogenetic analyses established well described lineages for mouse viruses; adding all new sequences to a detailed phylogeny will yield useful observations [25].
5.
Development of tractable animal models, such as macaques [26], will be necessary to dissect XMRV pathogenesis.
Defining any XMRV disease association is a distinct and more challenging prospect. Over the past 40 years, a number of retroviruses have been discovered to be bona fide infections of humans (HTLV‐1–4, HIV‐1, HIV‐2, spumaviruses). Not all infections are associated with disease; no diseases have been unequivocally associated with infection with HTLV‐II or spumaviruses. The XMRV disease association will require “shoe leather” epidemiology, exhaustive molecular biology, and gimlet‐eyed statistical analysis to reach conclusions regarding etiology. As many virologists have long realized, disease association with retroviruses goes beyond traditional Koch’s postulates as benchmarks. As Rowe [27] summarized, Koch’s postulates assume an acute and persistent association of a readily detectable pathogen, criteria that poorly serve investigation of illnesses such as neoplasms and chronically debilitating diseases. The collective experience and utility of retroviruses for these studies [28] will provide invaluable support for XMRV investigations.
Reports of a viral agent in prostate cancer and chronic fatigue syndrome immediately established a new area of research. The XMRV field has strong magnetic properties, attracting basic researchers (particularly old time murine leukemia virus labs), clinicians, patients, and other interested parties to opposing poles regarding etiology, pathogenesis, and therapeutics. With accumulating publications, including those here, there are increasing discussions regarding possibilities for widespread testing and therapeutic intervention. XMRV may be detected using PCR or serologic techniques, and detection technologies are being made commercially available. None of the research or commercial assays has been exhaustively tested, and none is FDA approved. Further research is necessary before any of these tests can be used reliably; at this stage, discussions concerning blood donation deferral for patients with chronic fatigue syndrome are based on the general principle that there might be an infectious etiology, not on specific diagnostic testing. With respect to therapeutics, XMRV is sensitive to some antiretrovirals in vitro [11, 20, 29], and there are calls for use of antiretroviral agents for therapy, even though early observations suggest XMRV replication may be minimal in humans. At this time, such an approach is premature and medically indefensible outside the secure oversight of a well‐controlled clinical trial. “Real world” coping with severe diseases like chronic fatigue syndrome and prostate cancer creates understandable desperation on the part of patients, caregivers, and health care professionals. Such pressures are not justification for testing of therapies in an uncontrolled manner. Indeed, because they are of no help whatsoever to other patients, physicians, pharmaceutical companies, or regulatory agencies, such uncontrolled therapy works directly against the goal of providing effective therapy to the million or more individuals experiencing these serious conditions.
