XMRV – the fight continues

XMRV In October 2009 it was reported that 68 of 101 patients with chronic fatigue syndrome (CFS) in the USA were infected with a novel gamma retrovirus, xenotropic murine leukaemia virus-related virus (XMRV), a virus previously linked to prostate cancer. This finding, if confirmed, would have a profound effect on the understanding and treatment of an incapacitating disease affecting millions worldwide.

But does XMRV cause disease? It’s far from clear that it does.

No Evidence of XMRV or Related Retroviruses in a London HIV-1-Positive Patient Cohort. 2011 PLoS ONE 6(3): e18096. doi:10.1371/journal.pone.0018096
Several studies have implicated a recently discovered gammaretrovirus, XMRV (Xenotropic murine leukaemia virus-related virus), in chronic fatigue syndrome and prostate cancer, though whether as causative agent or opportunistic infection is unclear. It has also been suggested that the virus can be found circulating amongst the general population. The discovery has been controversial, with conflicting results from attempts to reproduce the original studies.
Methodology/Principal Findings
We extracted peripheral blood DNA from a cohort of 540 HIV-1-positive patients (approximately 20% of whom have never been on anti-retroviral treatment) and determined the presence of XMRV and related viruses using TaqMan PCR. While we were able to amplify as few as 5 copies of positive control DNA, we did not find any positive samples in the patient cohort.
In view of these negative findings in this highly susceptible group, we conclude that it is unlikely that XMRV or related viruses are circulating at a significant level, if at all, in HIV-1-positive patients in London or in the general population.

This entry was posted in Uncategorized and tagged , , , , , , , , , . Bookmark the permalink.

21 Responses to XMRV – the fight continues

  1. dave says:

    They have only just started studies to look at whether the virus causes disease, so why would they know by now.

    The obvious problems with the negative studies is that they are using new assays and assuming that the virus is represented by the VP62 clone, which was never used by those who found the virus. By refusing to calibrate their assays to a known clinical positive they are hindering scientific progress into the association of this retrovirus to ME and prostate cancer, and the general population.

  2. RRM says:

    Dave, you are making a critical logical error: the assertion that there are are ‘known clinical positives’ out there is exactly what is being investigated. Scientific validation should not rely on the results of the original study. Remember that ‘those who found the virus’ also didn’t have to rely on ‘clinical positives’ to report their findings.

    Also, if ‘those who found the virus’ feel that the VP62 clone is not sufficient for validation, they should report that (in a peer reviewed article and not ‘off the record’ BTW) and report a proper, reliable way to find it (of course, without actually relying on their own findings). However, since the original investigators (e.g. Bob Silverman) are sending VP62 clones to scientists trying to validate their findings, it seems people like Silverman also think the VP62 clone is a correct representation of XMRV.

  3. dave says:

    RRM you are assuming there are not clinical positives. The reviewers of Science magazine disagree, as do the other labs scattered around the world who are also detecting the virus and using the clinical positives to calibrate their assays to. Validation has been provided by Lo et al. XMRV is a poly and xeno.

    Lombardi et al did not use the VP62 clone to in their Nested RT-PCR.

    I’m guessing you are new to the field, as any scientist would expect more diversity in XMRV than that represented by one clone. What a ridiculous idea to push. Silverman and co
    offer any assistance in helping those without the experience in detecting XMRV. If others want to limit what they can detect to a clone, they cannot call themselves scientists.

  4. Justin Reilly says:

    Please go back and edit this title. You turned off the comments there so I am posting here. This title is parroting of an anti-science press release. Obviously, we don’t know for sure that XMRV doesn’t cause ME in the UK.

    “XMRV does not cause CFS in the UK”
    Wednesday, January 6th, 2010

  5. RRM says:


    What then did Lombardi use to calibrate?

    Why does the scientific community not accept these reports ‘from labs scattered around the world’ that are finding XMRV in blood? Because RedLabs and VipDx are exactly doing what I am arguing against: using ‘clinical positives’ as calibration. Thus, their results are not validation of the original finding, as their results may be just as faulty (or as good) as the original study. Their results add little if anything to the table, because their experiments are not independent.

    Also, your remark about the reviewers of Science indicates a misunderstanding of the peer review process. Per review is a procedural quality check by peers. It is in no way a guarantee that the findings are actually correct. That is why everybody in the scientific community agreed that this finding (just as any other finding) needed independent validation (thus, without relying on the original results).

  6. dave says:

    Lombardi et al, like Urisman et al, used a microarray technique. Indeed the variability found in Lombardi et al was greater than VP62, so why are these labs who claim to be unable to detect XMRV using assays that are limited to VP62. Why would you assume, when the published research has shown different, both in Lombardi et al and in other retrovirus papers, that VP62 is XMRV. Illogical and unscientific.

    What is the “scientific community”? What are you claiming has rejected these results? If they cannot reproduce the methodology of Lombardi et al and cannot prove contamination, and Lombardi et al showed it was not contamination through their serology and culture assays, then they cannot claim the results are false. You are also ignoring the validation study of Lo et al. It should also be remembered, that the Blood XMRV working group is now onto Phase 3 and has no intention of stopping, and Ian Lipkin is beginning his multi lab study. Further research is constantly being published on the detection of the retrovirus, which still includes positive studies from Silverman and Klein.

    Your logic, which is not how the scientific method works, is that you cannot use a previous positive to confirm it is positive. If you did that how would any positive ever be declared a positive? The way science works, and the way all people are taught from when they are little children, is that you retest under the same conditions. Only then can you start altering the parameters. No one has replicated Lombardi et al, but Lo et al validated it. You are also forgetting, if you even know about this, that labs in Spain and Norway are also detecting the virus. As is Fischer in Germany, and labs in Japan. It is a virus that is found all over the world. Some of whom presented at CROI 18.

    The review of Lombardi et al set further requirements to test the PCR results, and were replicated in the other two labs taking part in the study. It is therefore you who has misunderstood what a six month review is about. That data supports the finding that XMRV is associated with ME/CFS. Furthermore, and once again, Lo et al validated the results.

    When you make claims about Lombardi et al or other work perhaps you should provide evidence as I have done above, otherwise statements such as ” their results may be just as faulty (or as good) as the original study” are nothing more than hot air.

  7. dave says:

    There was no contamination in Lombardi et al.

  8. RRM says:

    “Your logic, which is not how the scientific method works, is that you cannot use a previous positive to confirm it is positive”

    You misrepresent my logic. It is perfectly normal to use confirmed positives for future studies. It is however unscientific for a VALIDATION study to use a sample that is positive ACCORDING to one (or a few) lab(s). If you don’t agree, can you please provide me with one example in the history of science where a scientists succecfully used the perceieved results (not the methodology) of the study he was trying to validate?

    In fact, finding a truly confirmed positive was a primary purpose in phase II of the BWG group. Although many patients are complaining about the low number of ‘Science’ patients ued (4), the setup was that Mikovits would select the ‘most positive’ patients, that were more or less positive without question. During phase II, she and others could check if the would agree on the status of these samples. However, in the blinded round of testing, neither Mikovits, Ruscetti or Switzer could reliable pick one positive patient from the control, let alone that two or more labs would agree on the status of a single sample. Thus, there are still no confirmed positive patient samples.

    Until there are confirmed positive patients, the question remains if XMRV is really in the blood of human beings, and thus labs must not rely on the results of the study that is under debate.

    The rest of your post makes no real sense. For instance, a six month review process indicates that the reviewers had problems with the original manuscript, my statement about WPI’s results is a matter of simple logic and Lo et al. did not validate Lombardi et al. Lo et al did not find XMRV but closely related (but, and this is important, disctinct) viruses.

    One more thing which you probaly won’t address: Lo et al. validated the Lombardi et al. paper in your view. I think we can agree that Lo et al. did not use ‘confirmed positives’ from WPI for this ‘validation’. I suppose you feel that Lo and Alter were not adhering to the scientific method? Were they lazy scientists just like the rest of them, only more lucky? Or do you regard them as great scientists, just because you happen to like their results?

  9. dave says:

    You are clearly not a scientist so perhaps your comments on the subject can be excused for looking foolish. It is perfectly reasonable to use known clinical positives to calibrate an assay to, how else can it then be claimed that an unproven assay is capable of detecting the virus. VP62 cannot represent the diversity of the virus, and it is illogical to expect it to. Gallo and Montagnier exchanged samples to prove HIV was there. You are clearly confusing the ability to check an assay to using ” the perceieved results (not the methodology) of the study”. Perception has nothing to do with the scientific method, that is subjective opinion. Thus there are known positives, that is what the six month review agreed.

    Your comments about the blood working group are a little off topic. What is the 4? You provide no references. The results from Phase IIb were scrapped as a procedural error had been spotted before unblinding. The serology results matched for the NCI and WPI. Switzer detected XMRV in phase IIa and then changed all his assays and was unable to detect in IIb. Not really a surprise. However, there are multiple variables at play, and the new phlebotomist could well have made a mistake, but as was reported serology did detect the virus.

    It again must be down to you not being a scientist that you would claim a six month review and publication would mean the study had problems. It was clearly of the highest calibre, unlike the negative papers that could not get published in a top ranking journal.

    XMRV is a xeno and poly hybrid. Lo et al validated Lombardi et al. They found different strains of the same virus. Lo et al used the same primers and cycling conditions, so there was no need to calibrate as they were using a proven method, unlike all the negative papers with novel assays. Chance of course plays a part in any discovery, but here they applied the scientific method, and also assumed that there would be greater diversity than the VB62 clone. A scientist would do this and should be expected to.

  10. AJ Cann says:

    “You are clearly not a scientist so perhaps your comments on the subject can be excused for looking foolish.”

    Please keep it polite Dave.

  11. dave says:

    Does anyone think that people who keep repeating the same mistakes, using unproven and unvalidated assays, when 3 to 7% of the general population could be infected, are being polite? Yes, lets all put the science last, those infected last, and be polite about CDC switching assays. Or PR firms dropping rude comments about Science magazine and Lombardi et al. Which received no comment about politeness.

    AJ Cann – do you not think something stinks?

    People have been dying of cancers and neuro-immune diseases such as ME. Why is MicrobiologyBytes even entertaining those who put out a press release to try and manipulate the world into believing the association to ME had been disproven? It has not and the virus is there, validated, integrated, and infectious.

    Good people need to take a stand on this and work to get the WPI, and other labs with successful assays, Government funding. Before you try to say what a journalist does, they do not put illogical incoherent press forward, they seek the truth. Science journalism is almost non existent.

  12. AJ Cann says:

    The paper that this post refers to is published in a peer-reviewed scientific journal. The function of MicrobiologyBytes is to bring academic research such as this to a wider audience.

  13. Paul Watton says:

    Having watched Judy Mikovits’ presentation to the New York Academy of Science on Tuesday, her comments with regards to VP62 and the G to A hypermutation effects of Apobec3G on the virus invivo, was to the effect that PCR calibration using VP62 is too restrictive to be able to pick up XMRV & PMLVrVs.

  14. dave says:

    AJ Cann, that would sound reasonable, however you post it without comment. Others are then forced to have to comment. Did you ever post Lo et al, the validation paper? What about Fischer et al?

    @Paul Watton. VP62 has always been too restrictive. Why would anyone think that a retrovirus would be the same as one clone? Mikovits highlighted this well known problem earlier last year.

  15. dave says:

    The following are the only articles on XMRV that I can find on microbiologybytes. Please point me to any more if I have missed them.

    Lombardi had no specific mention in Oct 09, neither did Lo et al. But McClure’s first negative paper, which had a one day review did, as did Robin A Weiss article about how he contaminated his lab, and finally this little HIV study from the Wellcome Trust.

    Viruses and Cancer: XMRV and Prostate Cancer
    Monday, March 19th, 2007

    Prostate Cancer Caused By a Virus?
    Wednesday, September 16th, 2009

    XMRV does not cause CFS in the UK
    Wednesday, January 6th, 2010

    Human rumour viruses – a cautionary tale of virus and disease
    Tuesday, October 5th, 2010

    XMRV – the fight continues
    Tuesday, March 29th, 2011

  16. RRM says:


    Can you provide me with any negative or positive validation studies that have been published by Science? For every one you provide me with, I will provide you with 100 original findings. That argument, that is used on “the forums” ad nauseum, is laughable to real scientists in the field. Science is a good journal, but they are famous for presenting the most spectacular research, and not necessarily the most reliable. In fact, they have been criticized by scientists because of going with “spectacular” over “reliable” research multiple times. Also, I have never read research that showed that Science studies are retracted less than studies in other respected journals, but I am sure you can provide me with that evidence, given your belief in the superior quality of Science studies?

    And yes, the Science study did have problems. The reviewers did not accept the original manuscript and asked for several revisions. With any paper, how long it takes to get these revisions depends on several factors, e.g. the complexity of the demanded revisions and the available time of the researchers to do some extra experiments. These “six months in peer review” say absolutely nothing about the quality of peer review or the quality of the results. In science, you let the data speak for itself. I will add that the final Lombardi paper did look pretty good, but there was also valid criticism from well respected virologists in the field with absolutely no stake in CFS (check for instance the remarks by the great Patrick S. Moore).

    And great example! Montagnier and Gallo exchanged samples but that did NOT mean that Gallo used Montagnier’s samples to calibrate his test with (Gallo contaminated his samples with Montagnier’s BTW, which also shows you to beware with exchanging samples in validation studies). I guess that means you cannot provide me with any validation study that used the results of the study it was trying to validate to calibrate its own assays. Please report back when you do find such a study (I promise you that you won’t find any such respectable study).

    A few quick bullets:

    – The results of phase IIb were not scrapped. After the results came out as bad as they were, WPI provided an explanation for it (a grad student had re-used some needles). That does not mean that the results are scrapped. At all. Sorry.
    – The 4 is the number of ‘most positive’ XMRV samples that WPI collected (and couldn’t reliably identify in a blinded setting later on)
    – The blood was collected by the same phlebomotist on both occasions. Only, the first time, he (or she) sent the samples to WPI for processing and further distribution, while the second time, the samples were distributed by another lab in blinded fashion.

  17. dave says:

    The negative papers haven’t even made journals like PNAS. You are desperate for journals to not have a ranking system. They do, and the negative papers are getting nowhere as they are flawed. VP62 being the most glaringly obvious example. I suggest you stop hanging out in forums and start reading, as you have a lot of science to catch up on.

    Science published Lombardi et al. You can’t ague against it. PNAS published Lo et al. You can’t argue against that. The length of review depends on how much proof a journal wants to approve of the results. Lombardi and Lo et al passed with flying colors. A 24 hour review tells you everything a scientist needs to know about the quality and reputation of a journal, and the paper that is published. But of course some of the negative papers didn’t even have a review and were passed by the editor.

    Claims of authority are rather pointless with the names that are backing the research, and also have no stake in ME. Gallo used Montagnier’s sample. Amazingly the reason why Gallo had contamination was because the samples were identical, that hasn’t happened with XMRV, and didn’t affect HIV being proven as a human retrovirus. Again, Lo et al validated Lombardi et al because they used the same primers and cycling conditions. That’s the only other way to do it. It is inexcusable to make claims that an assay is proven when it has never detected XMRV in another lab or has not been proven against a positive clinical sample. Most studies would just replicate and have, and this is why you are confused. Please report back when you have read something.

    Here are a few bullet points.
    -the WPI results of Phase IIb were scrapped, as a procedural error was discovered. Those running the working group said this was of no concern as the WPI isolate and sequence all their samples. So they would always know.
    -The serology results matched of the NCI in IIb matched with the WPI.
    -Again, the phlebotomist can still make mistakes, and the variables are numerous. The results in IIa matched with the CDC, and with the NCI serology in IIb.
    -Lombardi et al was blinded, and the NCI and CC results also matched.

    Here is a little summary for you.

    It’s over. The virus is there, it is a human exogenous retrovirus, it is infectious, it is probably spread through saliva, and it IS integrated.

    Don’t let the results of the Phase III of the blood working group upset you. The truth is better than nothing.

  18. Peep says:

    12 months ago Stephen Goff was saying that everyone needs to exchange samples, and they are not doing it. Obviously the WPI were. Now someone called RRM is questioning if that is a safe thing to do and when clones have been exchanged! Very funny.

  19. RRM says:


    The only XMRV study that wasn’t properly peer reviewed was the PNAS paper. This is because Harvey Alter is a NAS member and has “special” rights to publish.

    Science published the original study but that doesn’t mean it is “right” or “true” or that it has some special, magical higher chance of being “right” or “true”. It was a nice finding that was properly peer reviewed, but has not been validated since that time and several lines of evidence suggest that he original finding is simpy incorrect. The irony is that you are claiming I am making appeals to authority when you are clinging to the authority of ScienceMag like it is some infallible magazine.

    I would certainly be surprised by positive results in Phase III of the BWG, but I would not be upset. You see, I would have no problem with reassessing my opinion based on new evidence when it comes available. That is what every scientist does or should do, by the way. However, the results of Phase III would not be conclusive even when WPI identifies the correct samples, as there are some problems with the methodology in the BWG. Please note that Ï am not saying that the methodology is faulty, but it is designed with a different objective in mind than validating Lombardi et al. That is precisely why it’s so good that the Lipkin study, which doesn’t have the same methodological limitations, is underway.

    The serology results of the NCI did NOT match with the WPI’s results. Please check slide 36 of the following presentation: http://www.cfids.org/webinar/slides-121710.pdf. You are clearly misinterpreting a statement of Judy Mikovits in which she meant that she had compared samples with Ruscetti outside the setting of the BWG. The results of Pahse IIb have been presented and there were no matching positive results.

    How can you even seriously argue that the results of Phase IIb are matching when you argue that the results are scrapped at the same time?

    Finally, while Lombardi et al. might have been blinded, the matching results of NCI and CC did have some problems with them, as they used the same samples and did not validate using freshly collected samples. Thus, if there was anything wrong with these samples, those other labs would “just” have gotten the same erronous results as the WPI.

    @Peep: The only thing that is “very funny” is your incorrectly interpreting my very simple argument. I am all for exchanging samples, but that doesn’t mean that one shouldn’t beware for contamination (as history has shown).

  20. Dave says:

    Lo et al had already passed peer review when Government officials put it on hold. It was then that they, i.e. a government employee, wanted more. So it had passed muster, not because someone has special rights. Ask Alter, Lo or even Coffin. They will confirm this.

    The fact that Science published the paper is testiment to it’s calibur. This may be where you are confused and not for the first time. It had to undergo a six month review and undergo further testing to support the original findings. Everything that the reviewers asked them to do to back up their data and prove it was a genuine human retrovirus in those patients was done and proven. It is rather pathetic to try and suggest the finding was a nice one, it was a major breakthrough on the back of the Silverman and Klein work.

    Lo did indeed validate Lombardi. XMRV is not Xentropic, it is a hybrid of a xeno and poly. It is clear you are not grasping this.

    I’m am surprised you are attaching emotions to the next phase of the BWG. This is obvoiusly causing you trouble in general and yes I agree you are not a scientist. Science is not a consensus process, it is the data that is important. It does strongly support that this is a human replicating retrovirus and it is present in two disease.

    It is highly amusing that you believe the results in the blood working group are not evidence in themselves. Again, it is all supporting data. Nothing else at this time suggests anything but the results being correct. The Lipkin study is no different. It is one paper. Looking at the work that is due to be published, it is likely that this will be moving onto clinical trials long before the Lipkin study is out.

    The serology results of the NCI Phase IIb did match the results of the WPI. Pay attention and try to keep up won’t you. That is still data. You are confusing the PCR results from which no conclusions could be drawn. The CDC was also detecting the virus in IIa, then changed all of their assays and found negatives in IIb. Why they were allowed to do such an thing is baffling.

    I’m am certain from your claims that you have not understood what took place in Lombardi et al. Certain experiments were not performed for the reasons you have stated.

    Lombardi et al:

    “Detection of XMRV was
    confirmed in 7 of 11 WPI CFS samples at the
    Cleveland Clinic”

    The following is from the supporting material.

    “Banked samples were selected for this study”

    “PCR analysis performed on 20 of the identical patient PBMC DNA
    specimens stored at the NCI (Frederick, MD) since 2007 confirmed nearly
    identical gag sequences, thereby diminishing the possibility of laboratory
    contamination as a source of XMRV.”

    Thus the results would not have had the same results and indeed they did not.

    Your comment to Peep is very telling. No data to argue differently and thus contamination as a “posibiliity” is only something to design into ones study.

  21. RRM says:

    Again, you post several errors. They are easily verifiable by checking the appropriate sources.

    First, Lo et al had not passed *proper* peer review before the hold up (or after it). You see, the Lo/Alter paper is a PNAS track 2 publication instead of a PNAS track 3 publication. The difference is, that in track 2, a PNAS member that has (co-)written the study will choose his own reviewers. He could have theoretically sent the paper to 100 reviewers and would only need the thums up of two of that 100. Track 3 is a normal publication where the PNAS editor will choose independent reviewers. You can see this is a track 2 publication by the “contributed by Harvey Alter” header to the paper by the way. A track 1 publication (which is no longer possible since late 2010) could be identified by the “communicated by [PNAS member]” header.

    Second, not “everything” that the Science reviewers’ requested was done by the authors. By Judy Mikovits own admission, they could not perform one very important “gold standard” experiment (integration of thevirus into the human genome). Please read: http://www.nature.com/news/2011/110314/full/471282a.html – the part from “Reviewers wanted more evidence will do.

    “Looking at the work that is due to be published, it is likely that this will be moving onto clinical trials long before the Lipkin study is out.”

    In the light of the Singh study, this statement again shows why I posted it is not smart to base one’s opinion on those “future study that will prove us all wrong”. I even think the Singh paper was the actual study most people were referring to, because of this single patient who was telling people he had been tested positive for XMRV by Singh. But of course, you have already decided that Singh’s study does not mean anything, and you are waiting for the next study that will prove us all wrong…

    I must admit that I didn’t know about those NCI samples (please notice that I am capable of admitting that I am wrong, I am just not used to it ;-) ). Still, it seems the parsimonious explanation is that both sets of samples were handled in the same way (that would lead to false positivity), while the sets of controls were collected differently (for instance, heparin tubes vs other tubes).

    The rest of your post (concerning the BWG results) is incoherent and does not address my arguments at all.

Comments are closed.