Pt. IA: XMRV: the Big Test – the Blood Working Study (Taken From the Ottawa Conference/CFIDS Association Webinar)
The Blood Working Group was a large group (25) made of federal officials from various agencies (CDC, NIH, FDA), the Red Cross, several laboratories including the WP, several Universities and put together just a month after the XMRV Science paper published in Oct 2009.
The CAA brought together three members of the BWG – none of whom were employed by the federal government to explain the results in their latest webinar. They were Graham Simmons, PhD of Blood Systems Research Institute, Michael Busch, MD, PhD of Blood Systems Research Institute, Steven Kleinman, BSc, MD of University of British Columbia
The culmination of the BWG’s efforts was a blinded test involving 9 well known labs including the WPI, Abbot, two FDA labs (including Lo/Alter), the NCI Ruscetti lab and Gen-Probe that examined 15 CFS patients from the WPI and Lo/Alter and X controls. A grand total of over 1500 samples were tested.
The BIG TEST – Two years in the making the stakes in this study were high. XMRV had started off full of promise for the ME/CFS Community but after 40 studies had been unable to find XMRV present even in background levels doubts had settled in. The one constant, in the process however, was Dr. Mikovits emphatic assurance that the WPI’s test was accurate and the BWG study was going to indicate that .
The key to the studies rigor was its double-blinded nature, the presence of independent phlebotomists and the large and distinguished group of researchers overseeing it. Double-blinded studies are like doubled-edged swords; with their rigor – neither party knowing which samples belong to the patients or controls – they can and do make or break findings. Twenty years ago two double-blinded studies broke the back on the DeFreitas retrovirus finding in CFS.
With the BWG study, the buck finally stopped where it has started – at the WPI and Dr. Ruscetti’s National Cancer Institute (NCI) Lab. After the vigorous defenses, after all the talk of politics and researchers not wanting to find the virus or labs not knowing how to do PCR – the WPI had the opportunity to do what they said they alone knew how to do; to show, in a blinded trial, that they could find the virus in CFS patients. In a study they were intimately involved in creating, and in patients they themselves handpicked (and who undoubtedly had tested positive repeatedly), they had the opportunity to silence the naysayers and turn the XMRV field around. It was redemption time.
Not a good start – It didn’t start off well. Despite the high stakes the WPI, inexplicably, was able to come up with only 10 of the 25 ME/CFS patients it promised to provide and was late in delivering them to boot – delaying the study by several months. (According to the diagram, one set of samples was collected in Oct, 2010 but the WPI took until mid-April 2011 to deliver 10 samples. A similar lateness in the first BWG study held it up as well). With the Lo/Alter group delivering their 5 patients the study included 15 patients and 12 pedigreed negative samples from healthy control. All the labs, after extensive testing, agreed that the healthy controls were negative for XMRV. ….
Methods – Even with a limited sample size the study was a large one. With three tests (NAT, antibody, culture) being done multiple times and 9 labs participating a total of over 1500 samples were tested. Each lab picked the type of test they wished to carry out and carried them out in a manner of their choosing. For each test each lab first proved that they could detect XMRV in a spiked sample.
Independent phlebotomists went to the patients home, withdrew their blood and shipped it overnight to the Blood Safety Research Lab for processing. The samples were then coded and shipped to the individual labs for blinded testing.
Nucleic Acid Amplification (PCR) Test – The PCR results in the Science paper – showing that almost 70% of patient had tested positive for the virus – had captured the research communities attention.FINDINGS
WPI Finds Few Positives – Only the WPI lab found any positives but of the 30 tests done on CFS patients the WPI found only one positive sample. Of the 45 tests done on the agreed negative samples the WPI reported found 2 positives.
Thus the WPI failed to identify 9 of the ten CFS patients as having XMRV and identified two of the 15 healthy controls as having XMRV. Depending on whether they tested whole blood, plasma or PBMC’s they identified different CFS and different healthy controls as having XMRV. The findings suggested that the WPI’s PCR test was essentially worthless.
Lo/Alter Group Fails to find XMRV or pMLV’s – the Lo/Alter group did not find XMRV in their former study but they did find pMLV’s in 50% of CFS patients. Despite repeated testing, however, in this study, the Lo/Alter group did not find any positive samples dashing doubt on the veracity of their first study as well.
Contamination link – Contamination has been a constant theme in the research community over the past 9 months. Finding equal or more positive samples in the negative controls suggests contamination has occurred. In this study a genetic analysis of the positive samples indicated that it had occurred in the WPI. The analysis found that the XMRV the WPI picked up was almost identical to that produced by a lab cell line called 22RV1. This suggested the virus the WPI found had never entered a human body; ie was lab contaminant. (At the Ottawa conference Dr. Silverman, produced evidence of contamination in the original study, which resulted in several parts of the original Science paper being retracted, including the PCR results.)
Conclusion: The WPI test was more likely to identify healthy control samples as having XMRV than CFS patients and did not find XMRV in patients they identified as likely to have the virus. The XMRV they did find was most likely been a contaminant from a lab cell line. Lo and Alter’s inability to find pMLV’s in patients they provided suggested that their original finding was due to contamination as well.
Culture Wars – When the early PCR studies failing to validate the WPI’s findings Dr. Mikovits turned again and again to what she felt were more definitive tests; the culture and antibody tests. These tests, she stated repeatedly, would tell the tale.
Another Double Edged Sword – However good culture studies are at revealing rare viruses they also carry a hidden danger; the long culture times increase the risk of contamination substantially and contamination, again was a prominent theme in the results.
Unfortunately the WPI bowed out of the BWG culture study when its samples became contaminated with mycoplasma, leaving the Ruscetti lab at the NCI, the Lo/Alter group and the Hewlett lab at the FDA to do the culture studies. Both the Hewlett lab and the Ruscetti lab, however, had been trained by the WPI in how they did their culture and both did the lengthy culture tests (19-21 days) mandated by Dr. Mikovits.
The prospect of contamination reared its head again when the Ruscetti lab found XMRV in more healthy controls (6/15) than in CFS patients (3/10). The fact that the FDA/Hewlett lab, despite using the same protocols, found zero XMRV in both the CFS patient and healthy controls, suggested that a contaminant had gotten into Ruscetti’s samples.
Antibody Test – The antibody test was always the antidote to contamination for the WPI because contamination of a sample simply has no effect on whether or not antibodies to XMRV were present in humans.
In the past Dr. Mikovits has stated that other labs had not broadened their assays enough to look for the MLV’s she believes make up the human gamma retrovirus family the WPI uncovered in CFS. The idea that researchers weren’t looking for MLV’s or a broad ranges of gammaretroviruses caught hold in the ME/CFS community but the BWG reported the labs did exactly that after the Lo/Alter study came out as well as XMRV (as did many of the XMRV studies. )
The diagram of the antibody study told the tale; each lab used a broad rather than a specific assay for XMRVs in order to pick up the full ranges of MLV’s Dr. Mikovits believed were found in the CFS population.
Once again the NCI Ruscetti lab and the WPI were the only labs to find samples that were positive for XMRV and once again the positives showed the same scattershot pattern, with more healthy controls testing positive than CFS patients. Disturbingly, despite using the same test, the lab disagreed on which patients were positive. It was unsettling to find the WPI, for instance, reporting that patients 1, 5 and 8 say were positive for antibodies to XMRV while Ruscetti’s NCI lab reported that those patients were negative but patient 2,3 and 7 were.
Furthermore, despite the fact that each sample was tested multiple times by each lab no samples were found to be positive more than once.
In the CFID’s Association’s Webinar Dr. Simmons answered 10 questions patient on the web had raised concerning the validity of the BWG study. Had the healthy controls been validated? Yes, by five labs including the WPI and Ruscetti NCI labs. Were the labs given enough time for the culture experiments? Yes, each lab was allowed to take as much time as they wished. Were the samples located in an area where the 22RV1 cell line was present? No (and only the WPI found the contaminant anyway, suggesting it had come from them). For more click here.
The Blood Working Group’s Conclusion – The BWG was tasked to determine if XMRV was a threat to the nation’s blood supply. After two years this large group, which contained researchers from across the spectrum, concluded that it was not and Dr. Simmons stated the recommendation that the people with CFS not give blood would be revisited.
Dr. Simmons stated the BWG’s conclusions on XMRV starkest terms
“Certainly XMRV itself is a laboratory contaminant that was created in a lab by passaging through mice”
In response Dr. Ruscetti and Dr. Mikovits stated that the virus could have disappeared from the patients blood as it appeared to do in the macaque study and that all the BWG study indicated was that more work needed to be done to make the test more accurate.
Disappearing Virus Theory – The problem of a disappearing virus did not, however, arise in the original Science study and it would be strange indeed to find the virus disappear in all 10 patients who had presumably repeatedly tested positive for the virus.
The biggest problem with the results, though, was not that the ‘virus’ didn’t just seem to pop up in the CFS patients; the problem was that it popped up just as frequently in the healthy controls as it did in the CFS patients. That pattern suggests that a contaminant is intermittently popping up in all the samples.
The confusing part about this is that the contaminant is, in fact, XMRV and the key point is that XMRV is not coming from the patients but the evidence suggests that it is in the WPI’s lab and is making its way into the blood samples stored there.
MORE Time? Dr. Mikovits and Dr. Ruscetti suggested they just need more time to develop a better test. A discussion with Science mag journalist Jon Cohen, who has written a book on the HIV epidemic, suggested this was unlikely. In response to a question about the development of the antibody and PCR tests Cohen noted that accurate tests for HIV, which is much more variable and difficult to characterize than XMRV, were developed fairly quickly twenty-five years ago. Even in its primitive stage Cohen stated that the original HIV antibody test would have picked up 12 of 15 positives. Why two years after the XMRV paper publication, with the much improved technology of our time, the WPI or any other institution still have a wholly unreliable test for XMRV is unclear.
(On Facebook, WPI reported that the clinical lab test for XMRV; ie the one patients paid for, is validated – which throws another wrench into the equation. )
Immunomodulators in Patients – Not the Answer- In oral conversation at the Ottawa Conference Dr. Mikovits also asserted that some of the patients were on antivirals and immunomodulators and those probably obscured the results. The BWG spokesman in the webinar, however, was careful to point out that a) none of the patients in the study were taking these drugs, b) the WPI didn’t check for antiviral/immunomodulator use in the original study and c) there was no evidence that they would have affected the results anyway.
p-MLV focus Appears Dead – There were always the pMLV findings from the Lo/Alter and Hansen studies to fall back on if XMRV didn’t work out. The Hanson report at the IACFS/ME conference concluded, however, that her original MLV findings were almost certainly due to contamination, and in the BWG study the Lo/Alter group inability to pick up pMLV’s or XMRV in any of the samples they or the WPI provided, indicated that their original results may have been due to contamination as well.
This was not unexpected. Despite their embrace by the WPI the Lo/Alter findings were regarded by many retrovirologists with suspicion because of their close resemblance to harmless endogenous retroviruses and several retrovirologists were surprised that the WPI did not distance themselves from their results.
After the dozens of negative XMRV studies and the few positive ‘XMRV’ studies a pattern does seem to have emerged. Retrovirologist were never able to find the virus in CFS while a few non-retrovirologists detected what they believed to be XMRV’s close cousins. Neither Lo/Alter or Hansen, whatever their accomplishments elsewhere, are retrovirologists and that may have made the difference in this oh so complex field.
Human Gamma Retroviruses? BWG Member Says HGRV’s Do Not Exist- Dr. Mikovits coined the name “Human gamma retroviruses (HGRV’s) last year to refer to the family of gamma retroviruses she asserted were present in humans.
Dr. Mikovits decision to coin a new name for the gamma retroviruses she believed were present in ME/CFS has taken on a life of its own in the ME/CFS community but it fell with a thud in the research community and Dr. Simmons of the BSRI explained why.
Reporting that there is no such thing as a ‘human gamma retrovirus’ Dr. Simmons stated that scientific standards require that a virus be isolated from a human, cloned and sequenced and evidence of that be established in a peer reviewed journal and then validated before an attempt is made to name it. While XMRV fit those criteria none of the other members of the putative family did. Indeed, researchers had long looked for evidence that gamma retroviruses, with all their potential for harm, are present in humans and had, until XMRV came along, concluded that they were not.
It was not surprising, therefore, to find considerable frustration within the research community at Dr Mikovits decision to unilaterally name what she believed were a family of new retroviruses. In an interview Dr. Singh seemed flabbergasted by the attempt to do this. Not only was there no published evidence that these viruses existed – a necessary foundation for creating a new name in the scientific community – but Dr. Singh stated if XMRV was to be renamed doing so was up to the discoverer of the virus – not a later researcher.
Dr. Mikovits has stated she has been aware since the time of the original Science paper of an array of XMRV-like retroviruses in ME/CFS yet has been unable to provide evidence in the form of unique genetic sequences. Even if the labs were not making as broad a search as the WPI (something that is not clear) their broadened search should have picked up at least some evidence of what Dr. Mikovits believes is a fairly large family of human retroviruses – yet study after study has turned up empty.
You can find more of the charts and images at http://www.research1st.com/2011/10/14/xmrv-updates/
Check out the slides here: http://www.cfids.org/xmrv/srwg-webinar-oct2011.pdf