XMRV and chronic fatigue syndrome: Scientific Blue Balls

Posted by ERV on August 23, 2010
(95)
More »

So, a while back, there was HUUUUUGE DRAMA. BIG HUBUB. Because a study that CONFIRMED the initial findings regarding XMRV and CFS was being HELD UP by TEH GOVERNMENT! Conspiracy, conspiracy, yada yada yada.

Apparently, this was that paper:

Detection of MLV-related virus gene sequences in blood of patients with chronic fatigue syndrome and healthy blood donors

I can honestly say I was really excited for this paper to come out. I wanted to see 1) what this group did to replicate those findings that no other lab could do, and 2) see the same results come from samples that had never seen Judy Mikovits laboratory.

Yeah.

Here is a graphic representation of this paper:

I have, like, the worst case of scientific blue balls ever. Like, ever.


*rubs temples*

Okay.

In the original paper, they found XMRV proviruses in the PBMC of 68 of 101 from patients with Chronic Fatigue Syndrome, 8 of 218 healthy controls.

In this paper, they found sequences in 32 of 37 (!!!) Chronic Fatigue patients, and 3 of 44 healthy controls.

Sequences.

Gag sequences.

From endogenous mouse retroviruses.

Not XMRV.

*blink*

*blink*

Noooooot XMRV…

Im granting them, that conclusion, which I honestly dont necessarily believe. PNAS paper, using standard PCR to ID ‘positives’ and ‘negatives’, with insane numbers of bands as a result of non-specific primer binding. Insane numbers. As if this is 1990 and we dont know what ‘Real Time PCR’ and ‘Taqman Probes’ are. Im going to ignore that and grant the premise they really have IDed viral sequences.

And they arent XMRV. They dont cluster anywhere near XMRV on a phylogenetic tree.

Theyre something else entirely.

Okay, well, logically it could be contamination from mouse DNA. So they checked for that… by looking for mouse mitochondrial DNA.

*blink*

I understand why they did it– mitochondrial DNA is easier to find than genomic DNA, because there are more mitochondria in a cell than genome. But you kinda need to look for mouse genomic DNA contamination when one of your phyologenetic trees has your ‘viral sequences’ so closely related to mouse ERVs on four different chromosomes that it doesnt even form a proper branch. Cause ERVs are in genomes. Not mitochondrial DNA.

*blink*

But wait, it gets better.

32/37 (86.5%) of the CFS patients were ‘something weird’ positive when they looked at PBMC. This group actually looked for ‘something weird’ as viral RNA. Cell-free virus in patient plasma. 42% were positive. Now, Im not bitching about the differences in percentage– when you freeze virus, even high-titer HIV-1, you lose some of it. What Im amazed by, is this is the first time anyone has isolated cell-free virus in the bloodstream of CFS patients… and this finding is just talked over in one awkwardly worded paragraph.

In 42% of samples, we also detected and sequence-confirmed the presence of MLV-related viral RNA in the frozen plasma samples of these CFS patients, using an RT-PCR assay (Fig. 1C). With one exception, all of the patients who tested positive for viral RNA gag gene sequences in the plasma samples also tested positive in the DNA prepared from PBMCs and/or whole blood. On the other hand, only about half of the patients with MLV-related virus gag gene sequences detected in PBMC DNA also had viral gag RNA sequences detected in the plasma.

1C is the only figure that doesnt look like shit. Its the first time anyone has identified cell-free virus in CFS patients. And thats what it gets.

*blink*

Im ignoring the fact that in figure 1C, all the CFS patients are lined up in a row, and the word ‘blinded’ is never used in this paper or the supplemental. That figure still looks nice.

And thats it.

Thats the whole paper.

And yet we get this:

“‘I think it settles the issue of whether the initial report was real or not,’ said K. Kimberly McCleary, president of the CFIDS Association of America, the leading organization for people with chronic fatigue syndrome.”

Nooooo… not exactly. Not at all, actually. This paper very superficially ‘supports’ the WPI, in that they found something weird in CFS patients blood (and didnt investigate it beyond standard PCR). But they didnt find XMRV. They found mouse ERVs. Which makes almost as much sense as Darth Vader using a Brita water filter to fill a plastic jug with ocean water.

Almost.

Keywords:
(95)
More »

Comments

  1. #1 Brian
    August 23, 2010

    SENSE.

    This controversy makes none.

  2. #2 gf1
    August 23, 2010

    I stayed up late for this!

    It’s CFS. It was always going to be odd. Much odder than a thirsty sci-fi fan.

    re the Mcleary quote: I take it to mean “It now seems very unlikely that the WPI’s work was straight up bull-shit.” It does now seem like they found something worth investigating (and I wonder if it could have been passed over if this Alter study happened to have been another negative?). I get these things mixed up, but haven’t the CFIDS previously been rather restrained about a possible XMRV link? Politically, they’ve got to be worried about totally missing the boat on it and looking like divs.

  3. #3 gf1
    August 23, 2010

    Actually – that last post sounds kind of arsey now that i see it again. How dare I imply this post wasn’t worth staying up for? It would be nice to get a bit more clear cut progress, one way or the other, with all this stuff though.

  4. #4 Chris
    August 23, 2010

    Reason for the condescending and hostile tone?

  5. #5 K
    August 23, 2010

    Sory but…. you don’t even have balls…

  6. #6 pochoams
    August 23, 2010

    I guess you need to let go and admit that you owe many apologies to many people… This NIH & FDA paper is not the last positive XMRV study coming out, in September 8th you will see at least 2 additional studies coming out in the first international XMRV workshop. But today´s news coming out of NIH is already big enough to invalidate the negative studies published to date on XMRV and CFS patients, were they basically found ZERO XMRV in patients and controls…
    But I guess you big ego will just not let you let go, and you will keep on trying to convince us that you were right all the time…

  7. #7 ERV
    August 23, 2010

    gf1– Oh, no arseyness noted! I agree that the WPIs stuff might not be straight up bullshit. I see this as one more lab saying “There is something fucked up with CFSers.” Whether its XMRV or endogenous mouse viruses or EBV or HHV6 or whatever. All this stuff is ‘associated’ with CFS, but I just dont see any of it being causative.

    Chris– Hi! You must be new to ERV! If you are really interested in XMRV, and why I might be a bit annoyed with this paper and the publicity surrounding it, you can click on the ‘Archives’ tab above, and then click on the ‘XMRV’ category. Go back to the very beginning, and read all my posts on it, if you are interested in my journey from ‘Huh! This is neat!’ to ‘I hate this entire field of research!’

    If you just want me to be nice and cuddly and ‘civil’, you have come to the wrong blog. :-D

  8. #8 Brian H
    August 23, 2010

    Chris #4: You must be new to this corner of the intertruck, or else perhaps you haven’t been paying attention.

  9. #9 Richard Jefferys
    August 23, 2010

    It’s confusing. The new paper says it used the 419F/1154R gag primer set and it looks like that’s the same one used in the Lombardi paper – is it possible that the Lombardi paper jumped the gun on designating those gag results as XMRV based on relatively limited sequencing? The reagents used to identify “XMRV proteins” were all MLV-based, could that mean that they would also react with proteins from the viruses described in the new paper? But that would still leave a yawning gap between these results and the negative papers.

    It sounds like the CFS samples date back to when Shyh-Ching Lo was interested in whether the disease might be linked to mycoplasmas (“Initially, we tested 41 whole-blood samples that had been obtained for culture isolation of mycoplasmal agents in the mid-1990s.”).

  10. #10 ERV
    August 23, 2010

    Richard– The age of samples doesnt really bother me. Someone found HIV-1 in tissue soaked in formaldehyde and sitting at room temp for 50 years…

    Hmm. Ill have to go back and look at the Lombardi sequences. Didnt they make trees too?

  11. #11 Andrew
    August 23, 2010

    BTW, The mt-DNA assay was from the CDC (Bill Switzer), that’s one of the reasons why they used it.

  12. #12 Ernie
    August 23, 2010

    Maybe you should just take it up with Dr. Alter.

  13. #13 Levi
    August 23, 2010

    ERV,

    I don’t think you owe anyone an apology, and your easily understood breakdowns on this stuff is really appeciated by many. The PNAS paper was not quite what a lot of CFS folks were expecting because, well, they did not find any XMRV. The other zero/zero studies did not find any XMRV either, and that is where they stopped looking. Apparently.

    But the PNAS guys did find something, and that will be likely be enough to take this game into extra innings. My questions for you at this point;

    1)Do you feel the question of contamination was handled adequately with this paper? Or is what they found actually entirely contamination, i.e. mouse ervs?

    2)Is it fair to say that what they found was a retrovirus group that infects both mice and humans? While XMRV would only infect humans, but not mice?

    3)If you take requests, sometime I would like to see a short primer on retroviral clades on your site.

  15. #15 ERV
    August 23, 2010

    Thats a terrible read. Nowhere in this paper do they say they found XMRV. They are well aware of the fact that they did not find XMRV, nothing I am saying on that is in dispute.

    But that article says:
    “In the new study, conducted by scientists at the National Institutes of Health (NIH), the U.S. Food and Drug Administration (FDA), and Harvard University, researchers scanned for traces of a virus known as XMRV in samples taken from 37 CFS patients, collected by Harvard Medical School CFS specialist Anthony Komaroff in the mid-1990s. They found evidence for the virus in 32 (87%) of the patients, but in only three out of 44 healthy controls (6.8%).”

    Nowhere in the actual paper do they say they found XMRV.

    They call what they found ‘MLV-like’ throughout the entire paper. I dont think this paper is good, but Alter isnt a moron.

  16. #16 Andrew
    August 23, 2010

    Readers of this blog might be interested in this also:
    “Q&A: Why I delayed XMRV paper”
    http://www.the-scientist.com/blog/display/57628/

  17. #17 ERV
    August 23, 2010

    Levi–

    1– No. An exogenous mouse ERV in humans makes no sense. But thats what their tree says. Mouse ERV is even more incredible than XMRV. Might be able to figure this out more if they upload their sequences to genbank. I realize they tried very hard not to contaminate their samples with mouse cells. That doesnt mean mouse DNA isnt in any of their store-bought reagents. There are H2O lanes in the mitochondral gels, but not the MLV gels (Fig 1, Fig 2). Why? Positive and negative controls go on every gel, end of story. First lesson every rotating student in our lab learns.

    2– No. They have done zero virology. There is nothing in this paper but PCR, so its hard to say much beyond that data.

    3– Sure!

  18. #18 qetzal
    August 24, 2010

    But today´s news coming out of NIH is already big enough to invalidate the negative studies published to date on XMRV and CFS patients, were they basically found ZERO XMRV in patients and controls…

    It does nothing of the sort. If anything, it’s yet another case of a lab that could not detect XMRV in CFS patients. In that respect, it’s quite consistent with the 3 or 4 other labs that were also unable to detect XMRV in CFS.

    The fact that Alter’s group is seeing something like mouse ERVs in CFS suggests something really weird may be going on, and I agree it lends some credence to WPI’s initial study. But it doesn’t confirm WPI’s conclusions about XMRV, and it doesn’t invalidate the negative findings either. It suggests that maybe everyone’s right, for reasons we don’t currently understand.

  19. #19 John
    August 24, 2010

    “That doesnt mean mouse DNA isnt in any of their store-bought reagents”.

    I’ve heard this mentioned before in regards to a study done by Brigette Huber which was recently presented at the Invest in ME Conference in the UK, and is scheduled to be presented again at the upcoming NIH XMRV workshop in Sept. My question is why don’t they test all of their supplies before doing anything to see if their stuff is pre-contaminated?

    And please keep rocking the boat, ERV. Some of us are getting desperate for a truly objective take on the subject since there are so few people without any axe to grind, even if they think they are totally neutral. Thanks for the review.

  20. #20 Richard Jefferys
    August 24, 2010

    Seems like the main “68 of 101″ result in the Lombardi paper was based on the same gag primer used in the PNAS paper, but there were an additional 7 samples where gag and env was sequenced and two complete and one partial genome – all XMRV. The sequences are in the supplemental fig #1, completely beyond my ken though.

    I’m trying to understand the claim in the PNAS paper that “”our results clearly support the central argument by Lombardi et al. that MLV-related viruses are associated with CFS and are present in some blood donors.” Doesn’t seem clear at all to me unless they’re suggesting that some of the Lombardi gag results weren’t XMRV but these MLVs. It’s also rather generous of them to helpfully define the “central argument” of the Lombardi paper in a way that makes their findings seem confirmatory.

    The discussion of the negative findings seems disingenuous, as it suggests that “some studies” used a primer that would’ve missed the MLV sequences but only cites one. And several of the negative studies did use the same primer. The big discrepancy in the findings in healthy controls also goes undiscussed, the paper closes by citing only their own and the Lombardi data and doesn’t mention the negative studies at all.

  21. #21 Andrew
    August 24, 2010

    The part that annoys me is that all of the published results are in line with the pre-conceived biases of the researchers, with the possible exception of the Groom study.

    Dr Alter and Dr Lo wanted to find something, the CDC guys did not…

  22. #22 Andrew
    August 24, 2010

    By researchers I mean the research leaders.

  23. #23 Dr Matthew
    August 24, 2010

    I still stand by this – the CFS “industry” and its discouragement of persons suffering these symptoms from pursuing help from mental health professionals – the same folk that help people with the concomitant depressions that persons with cancer, or HIV, or nerve damage might have…. all because it underminds their quest for a pure, neuron-free, non-psychologically mediated disease? Bunkum.

    I just finished writing up an abstract examining an indirect effect between internalized homophobia and viral load based through the pathway of…. wait for it….. positive affect. Google Scholar search Moskowitz and positive affect. I’ll take the whole CFS crowd more seriously once they drop the bizarre notion that theirs is the first disease involving fatigue and pain completely independent from psychological and neurological interaction.

  24. #24 RRM
    August 24, 2010

    I don’t quite agree, Andrew. Groom was considered to be ‘friendly’ to Lombardi (until) they found nothing) , but I don’t think it’s fair to say that Van Kuppenveld or McClure had a pre-conceived bias toward not finding the (or ‘a’) virus. I’m no scientist, but it seems to me that the ‘psych-lobby’ researchers that were involved in the negative studies (Van der Meer, Wesseley, Reeves) had no influence on the actual testing, but were merely responsible for the researched cohort and/or for providing blood samples to the people actually doing the testing.

    The blog you’ve posted (http://www.the-scientist.com/blog/display/57628/) is an interesting read. Shows the crazy conspiracy theories were way off, as always.

  25. #25 Andrew
    August 24, 2010

    This is becoming associative ad hominem, but have you read the CFS papers published by the guys at Nijmegen (including Bleijenberg? By the way, a paper published by the Nijmegen guys in August has a big “what the” in their conclusion. But that is a topic for another place.

    Also by the way the other way to bias the outcome besides biasing the cohort is influencing the amount of time spent optimizing the method. (if you get negative results, you may ask – am I looking for the right thing? If you don’t have the time and funds then you are simply going to publish a negative result.)

  26. #26 Nero
    August 24, 2010

    Van Der Meer is pretty known in my country (he’s dutch – like me) and has from the beginning on always tried to make CFS sound like something purely psychological. Even when the first WPI science article came out, he totally ditched it in the papers. In all those years before he also has always tried to make it look like something totally psychological. Not ever has he really questioned it.

    I’d love to give him some ME for breakfast

  27. #27 sadpanda
    August 24, 2010

    The first time I remember hearings about XMRV was about a possible connection to prostate cancer, wonder what’s happened since then.

    see: http://www.plospathogens.org/article/info:doi/10.1371/journal.ppat.0020025

  28. #28 RRM
    August 24, 2010

    @Andrew

    Like ERV pointed out, there really is no argument for ‘biasing’ the cohort when positive studies find 3.9%-7% of this virus in healthy controls. The Lo/Alter study also ‘just’ used the CDC Fukuda criteria instead of the CCC (which weren’t available at the time).

    Regarding assay methodology, if anything, some of the other researchers (at least Van Kuppenveld) went beyond the sensitivity of the original paper. ‘The amount of time spent optimizing the method’ seems more like an issue for the actual virologists that put their name on the line instead of the psych-lobbyists in the research group. To conclude, I believe the CDC assay proved to be the most sensitive in a ‘spiked sample’ test by the BPAC, so I guess there is not much factual basis to support the assertion that not enough was done to detect this bug.

    In a few months this will probably be figured out, but I’m not betting the house right now…

  29. #29 Peter W
    August 24, 2010

    ERV (or anyone else),
    Sorry if I’m about to repeat an old subject but I just don’t have enough steam to read all them archives.

    Your comment “I can honestly say I was really excited for this paper to come out. I wanted to see ……. 2) see the same results come from samples that had never seen Judy Mikovits laboratory.

    In the 2009 study, didn’t the samples tested by the National Cancer Institute travel straight from patient to the institute. Is this relevant and how does it fit in with your assessment? http://www.landesbioscience.com/journals/virulence/article/MikovitisVIRU1-5.pdf

  30. #30 qetzal
    August 24, 2010

    John (#19) asked:

    My question is why don’t they test all of their supplies before doing anything to see if their stuff is pre-contaminated?

    They did. According to the paper, during the course of the study, they ran over 300 negative control samples, and none of them came up positive. It’s always hard to rule out every possible source of contamination, but these results don’t seem consistent with simple sporadic contamination of their supplies.

  31. #31 RedDirtDevil
    August 24, 2010

    At this point we are just trading talking points. But..MLV and XMRV are completely different taxa, now and forever. But Dr. Alter isolated Hep C and B even though they are completely unrelated organisms. Life isn’t fair. Some of us make huge leaps of scientific discovery. Some write hyperbolic blogs about their abs, moutain dew, etc.

  33. #33 Eric Johnson
    August 24, 2010

    So ERV, you truly want already-refereed papers to be fucked with more often by vanity-addled politicians who are numb from the left anterior cingulate down to the tip of the cock? Sure, sure you do.

    > one of your phyologenetic trees has your ‘viral sequences’ so closely related to mouse ERVs on four different chromosomes that it doesnt even form a proper branch.

    My half Yiddische half English girlfriend was all paraphyletic like that too, or rather created paraphylies unless you cut her little oval on the family tree in half. Very disturbing. I suspect her parents may have had sex – even more disturbing (she and I only held hands). Let him that hath ears hear.

  34. #34 Eric Johnson
    August 24, 2010

    I agree that in general, stuff is problematic.

  35. #35 ERV
    August 24, 2010

    sadpanda– Though the XMRV–>prostate cancer association was found first, the field is still plagued with ‘I can find it’ ‘I cant’ ‘Its there’ ‘No, its not.’ You are very right in pointing out, again, that this isnt only about CFS.

    Peter– I will look it up!

    qetzal– 300 negative controls. And they couldnt publish one? And they did not operationally define ‘negative’. What is ‘negative’/'not positive’? Were the lanes totally clean? Was there a ton of non-specific binding, but no products the right size? This is not something you tell readers when contamination is a very big concern. You show it and let people see and decide for themselves. Not showing even one ‘negative’ really bothers me.

    Eric– This was a PNAS paper submitted by a NAS member. It was not subject to normal peer review (as I stated before, there is nothing ‘wrong’ with this, nor is it ‘cheating’, but its not normal). This paper is emaciated. I cant imagine how sad it must have looked before the editors independent reviewer asked for more. And, this paper was also not held up by any politicians, as you can see in the article in comment #16.

  36. #36 Eric Johnson
    August 24, 2010

    Dr Matthew, since you sound like you have some sincere inquisitiveness in you. What your teacher said about lupus was interesting, but there is more to learn. CFS is from the /physis/ and indeed, very few diseases are likely to be psychogenic. When I got it, though I am only one tiny example, I couldn’t have been happier. I lived with 10 people and smoked pot three times a day. College was not very troublesome considering my 1550 SATs and GREs.

    What are your postulates. Surely you agree that the rather common psychosis of lupus is most likely physio-genic? The depression, OCD, etc found in lupus are probably physio-genic, but in any case secondary to lupus. Elevated prevalences of depression and altered personality variables are found in many other systemic diseases.

    I previously typed this for some friends, from a late edition of the undisputed number one reference book on lupus, namely Dubois’.

    =====================
    Psychiatric disorders have been reported to a variable extent in SLE patients (refs) and are likely multifactorial in etiology (ref). Eg in a review of 21 studies, Wekking (ref) found that the overall prevalence of psychiatric disorders ranged from 17% to 71% with depression being the most common syndrome. THey found no consist. relation. with other manifest. of SLE and emphas. the import. of psychosocial stress as an assoc. factor (refs). Indeed, those studies that have incl. other chronic disease control groups, such as patients with RA, have reported comparable freq.s and types of psychiat. disorders in control and SLE groups (ref). But zees, mein Herrn, does not negate the import. of recog. and managing psych. disorders in SLE, and one must remember that a partic. tragic outcome of such disorders in SLE patients is suicide (refs). Alors! – allonz-y!
    =======================

    CFS patients also have elevated prevalences of generalized anxiety, depression, OCD, bipolarism, etc. I have experienced all of these as purely physio-genic. I have had all of them severely, except for the mania which was marked but quite mild (my disease is far, far milder now and was very severe for ~18 months). I’ve also experienced very mild paranoia, but rarely without being alert to the fact that I was experiencing it. When I had the very severe anxiety, I was not ruminating on something over and over. I simply had it and it came and went as did heart palps (thousands per day), muscle fasiculations (once /30 seconds), flu-like stuff, etc.

    This is why CFSers seem aberrant to you on average. Of course one who thinks so is quite correct. You will also see nasty depression, ‘brain fog’, and fatigue, in patients given any sort of immunergic cytokine such as IFNg – this finding has been ratified so many times, without dissent, that it is extremely solid.

    What can I say, I don’t dislike you, but you write shit.

    “The reality we know from basic science is that our brain gauges our energy level to our routine activities: if any person, regardless of health, reduces their physical activity for an extended length of time to 2 hours a day, then a 4-5 hour day of driving to a doctor’s appointment, sitting in a waiting room, being seen, etc., will result in feeling exhausted.”

    This is utter nonsense of course. Go ahead an lie down 22 hours a day for as long as you like. You will not lose the ability to drive, sit, stand for 5 hours by practicing this way, until you reach the age of 75. Also you are starting from a false premise. There are plenty of CFS patients – a good fraction – who walk a mile every day. I guaran-fucking-tee you that fact. A hypothesis of deconditioning simply doesn’t survive outside the hothouse of certain intellects.

    “or as some patient advocates now call it, myalgic encephalomyelitis (ME; probably in part because the latter sounds so much more technical and appears to imply a clear biomedical etiology)”

    That’s accurate, about it sounding technical-biomedical. But it’s a very old term, not a novel one as you say. How can you be a scholar if you just spout stuff? You need to pound, pound, pound, right on the head of the nail, perfectly like a punctual machine, and go on so, so long at a stretch before once erring about fact. Instead I find an error as soon as I begin reading.

    What can I say. You seem like a nice guy and fairly intelligent. You are simply not capable of scholarship or of probing hypotheses with any rigor. You are dispensing horseshit, it is just as brilliant and scintillating as invading Iraq. This is not said out of love or hate, but indifferent philosophizing. It is unwise to despair about this matter; this incapability for scholarship and science is totally normal. This is the order of the universe. Rather it is simply a waste of energy to despair, a mild shame to indulge in. Just quit and do something real as your work. If you are rather in want of pride at the point of recognizing how wrong all this stuff is, simply take pride in overcoming your own BS. It works for me vis-a-vis my long youth, filled with total garbage.

    Vanity of vanities, saith the preacher – so nearly everything is vanity. Also of the making of books there is no end. Let us hear the conclusion of the entire matter: fear God and keep his commandments. Name for yourself the God or ideal that pertains to you, perhaps.

  37. #37 Eric Johnson
    August 24, 2010

    Fair enough, it was initially held up by politicians, and later (for longer) by Prof Schekman, on his own wishes, to protect PNAS (which is fair for him to do, perhaps even a duty).

    I agree it would be nice to have serology after all this time.

  38. #38 Mu
    August 24, 2010

    The one interesting thing I’m getting from all of this is, while everyone thinks that isolating and sequencing genomes is the gold standard of research, there’s still a lot of rust hiding beneath all that glitter in form of contamination, non-specific primers and other results subject to experimental error and interpretation. For some reason they never mention that on CSI or google-U. Please keep up the reporting on this kind of research.

  39. #39 qetzal
    August 24, 2010

    ERV,

    I admit, I was also bothered by their failure to include any of the negative controls in any of the figures. Especially in a situation like this one, I’d be running multiple controls in every PCR study and on every gel. Why don’t we see any of them? So yeah, that’s a bit bothersome.

    300 negative controls with no false positives rules out ubiquitous pre-contamination of their reagents, which was John’s question. But without more detail on exactly what the controls were and how they looked, it’s hard to fully evaluate other possible contamination scenarios.

    I’d especially like to know about any controls during DNA or RNA isolation. Ideally, every time they were processsing patient blood or plasma, they would have also processed some negative control blood or plasma in parallel, preferably interspersed with the patient samples.

    Of course, they only saw 7% positives in healthy subjects, versus 87% positives in CFS patients. Also, the methods state:

    All patient and control samples were coded and tested in parallel.

    If that really means they were processed from start to finish as part of a single group, that would make contamination during DNA/RNA isolation pretty unlikely.

    As always, more study is needeed. :-)

  40. #40 daedalus2u
    August 24, 2010

    Ruling out contamination in the reagents the lab used does not rule out contamination in the reagents used for blood collection and blood storage.

    Contamination in the reagents used for blood drawing and/or blood storage might explain the finding of sequences in plasma.

    For samples taken years ago, there probably are not blanks remaining to test for reagent contamination.

  41. #41 CAT
    August 24, 2010

    Dr Mathew

    “I’ll take the whole CFS crowd more seriously once they drop the bizarre notion that theirs is the first disease involving fatigue and pain completely independent from psychological and neurological interaction.”

    Not everyone in the “CFS” crowd thinks this.

    RPM

    “The Lo/Alter study also ‘just’ used the CDC Fukuda criteria instead of the CCC (which weren’t available at the time).”

    No, they did not. Fukuda was created in 94, these patients were from 93. That would be Holmes.

  42. #42 Dr Matthew
    August 24, 2010

    Eric – You don’t seem to offer anything new to the discussion. My complaint isn’t that the psychological disorders associated with CFS are unique, but rather that advocates of the diagnostic label are overly dismissive of them and frequently discourage treating them. The idea that, as you say, you can lie down 22 hours a day and suffer no pain, discomfort, or loss of energy when only occasionally being mobile until age 75 is, quite frankly, too illogical a fantasy to bother refuting. I call bullshit. I get why you skip right to ad hominem attacks (though seriously…. pontificating on your SAT scores and own intelligence as a form of argument from authority is one of the weakest routes I’ve seen), but you really need to understand more than the history of these labels to contribute effectively in the debate.

    Thanks for playing.

  43. #43 Prometheus
    August 24, 2010

    Dr Matthew,

    “…ponificating on your own SAT scores….”

    Yea.

    What the hell is that? Is every CFS sufferer on the way to present their I. A. S. dissertation in theoretical math with their devoted fiance/jeans model at their side when “disaster strikes”.

    No stories of CFS ending a career on the semipro bowling circuit or how it threw a wrench in twiddling your thumbs in a cubicle till the pension vests?

    Is part of the Fukada criteria “formerly promising asshole”?

    I could garner a lot more sympathy and attend more of the arguments if either pointless anecdote were given a mince or just once, the wall of text was introduced with “I was looking forward to taking over the old man’s Cheese-on-a-Stick concession. I had the hat, apron and everything when….”.

  44. #44 CAT
    August 24, 2010

    Dr Matthew.

    Advocates for the organic cause of ME/CFS do not dismiss associated psychological disorders. They will treat them, if those disorders are present. This is a myth.

    Not everyone with the disease is flat on their back all day, another myth.

    Again, the patients were diagnosed in 93.

  45. #45 Zootfloggin
    August 24, 2010

    Oh, Lord………save me.

    Nested PCR has the following steps:

    1) Amplification of a piece of DNA in a PCR tube.
    2) Taking said DNA PCR amplification reaction, opening the tube and placing these amplification products into a second PCR reaction.
    3) A second “nested” amplification is then performed using internal primers (semi-nested as described for the mouse MT DNA reuses one of the original primers).

    The reason for doing nested PCR is to really, really, REALLY!!!! increase the sensitivity of the PCR reaction. It is SO sensitive that it often detects less than one molecule of DNA in a PCR reaction mixture (insert laughs here).

    By definition, nested PCR invalidates precautions listed in this paper to prevent PCR contamination. This is because the scientists are working with the first amplification products in their second amplification. Harvey Alter has done good work on hepatitis viruses but there is so much dumb in this paper that it makes my head hurt.

    Your description of the conundrum of the genetic distance between XMRV and the MLV-like sequences described in the PNAS paper is really insightful. Its about like saying folks with HTLV I and HIV are more less the same thing. I certainly don’t know if XMRV is real and associated with CFS but Irving Langmuir, the famous physical chemist, had some wise words of caution 50 years ago (http://en.wikipedia.org/wiki/Pathological_science).

  46. #46 CAT
    August 24, 2010

    Why compare HIV & HTLV, why not HTLV-1 & HTLV-1V

  47. #47 ERV
    August 24, 2010

    CAT– I think you mean HTLV-IV, as in, 4 ;)

    Without having more of these sequences, its hard to say. And it doesnt help that viral nomenclature is really hard for normal people to understand.

    The endogenous MPMV that, apparently, ‘look like’ what this group found look nothing like XMRV. Their LTRs are different. Their gags are different. Their envs are different. Endogenous MLVs (MPMV, PMV, XMV) and XMRV are different creatures. John Coffin published the hell out of this when I was a toddler.

    And endogenous MLVs? They are tens of thousands (millions?) of years old.

    Heres an analogy that might work. Say a police officer has a tip. She KNOWS that Suspect X killed Victim A, and if she is right, the body will be found in Suspect Xs backyard. So they send in an excavation team and they find… a fossilized mammoth. “AHA!” the police woman says “Suspect X was the murderer! Heres the body in his backyard!!”

    Mammoths and humans are mammals, and probably have lots of genes in common. But their genomes are fundamentally different, and separated by a very long stretch of time. So for this group to say “AHA! We found endogenous MPMV sequences! XMRV is the culprit!” is equally baffling.

    NOTE: The above analogy might be shit. And it might be even worse shit after they upload their sequences to genbank and we can mathematically figure out a more apt analogy. Or I might be right, which might signal the coming apocalypse. Requesting /b/ackup on this.

  48. #48 RRM
    August 25, 2010

    To make matters even stranger, the FDA are reporting on their MLV/CFS FAQ page that they’ve checked 34 samples from the CDC study and that none have come up positive (31 negatie, 3 undetermined). See question 10 at: http://www.fda.gov/BiologicsBloodVaccines/SafetyAvailability/ucm223232.htm

    @ CAT (“No, they did not. Fukuda was created in 94, these patients were from 93. That would be Holmes.”)

    From the paper: “Each [of the 25 patients] met the 1988 ["Holmes"] CDC criteria for CFS, and 21 also met the 1994 ["Fukuda"] CDC criteria.”

    The methodology used to select the other 12 patients, is “unknown” to the researchers.

  49. #49 CAT
    August 25, 2010

    1 I, ha

    A few thoughts.

    All those attending the XMRV international conference agree that this is an exogenous human retrovirus. McClure does too, as she is now finding it.

    Also, the Lomardi et al and Lo et al. have never said that their research proves causation.

    The Lo team looked for XMRV and other MLV-like viruses suspecting that there may be more diversity than had been previously found.

    This all seems reasonable

  50. #50 flower21
    August 25, 2010

    Some good points made.

    As we all know there are many isolates of HIV 1, The Sequences aren’t all exactly the same but there still all HIV 1.
    Like you said ERV once the virus from people are sequenced – we will have a more clear understanding!

  51. #51 CAT
    August 25, 2010

    These are a couple of quotes from the authors:

    Dr Alter: “Does raise questions because there isn’t consensus in different CFS populations. Many things still have to be looked at. It does at least confirm the findings of Whittemore Peterson. I think one wants to go back to their studies because they have had more time and more they have done extensive work… so its not just finding isolated viral sequences. But they have found gag, env sequences. They have been able to transmit this to an animal model (the macaque). They have seen antibody in the macaque and they have seen ab in pts and they have seen the viral particles. So this study is more advanced than ours. But with them having done the groundwork, I think our study is highly confirmatory of their work and we are in the process of trying to culture virus from our patients and to find antibodies. We have already shown that some of our pts also have envelope sequences, and preliminary work appears that there is antibody present as well but this has not been confirmed. But there are many things that need to be done, many more CFS populations that need to be tested. (More) numbers of blood donors that need to be tested. Other diseases that need to be tested. More culture work. More robust assays.”

    Dr Lo. “No, actually this group of virus that we call polytropic MLV – this is actually more characteristic for the retrovirus infection instead of a single viral gene. They very quickly can evolve and have multiple different kind of a sequence found. In fact what we found is that actually gave us good confidence that this cannot be a laboratory artifact or PCR contamination product because the sequences they are so varied from one patient to another.”

    Dr Alter. “My thought came back. Since their original publication the WPI and NCI groups have now found that they too are finding greater variability among their patients, so it is not just XMRV even in the original cohort of patients.”

    Dr McCluskey: “One thing that hasn’t been mentioned so far is that this virus at least in our hands is present in a variable titre which increases the difficulty of detection so that may be also complicating or adding to the variability among the labs.”

    Dr Lo: “I would also add on this. I think most of the irregularity or the different findings. Many probably due to a very low copy or viral gene copy in the blood and that is a main concern for a consistent finding and using different kind of assay systems.”

    Dr Lo: In a way it was a little bit surprise when the WPI’s first publication in Science and they show it as a single kind of kind of a sequence. That is unusual for retrovirus, but now they are also stating they do see the variation of the sequence and also appear to be more closely related to polytropic related MLV’s.

  52. #52 DrDuke
    August 25, 2010

    @ERV comment 46.

    Viral nomenclature and classification is more messed up than almost anybody realizes. But the nomenclature and classification of bacteria and some groups of single-celled eukaryotes is a mess too. How do we define a “species” or “genus” of these organisms, anyway? Classes like “serotypes” make some sense in some cases, but the genes for the proteins that are being recognized by the serology can often be horizontally transferred.

    The Murine Leukemia Viruses, whether they are polytropic, xenotropic, amphotropic, endogenous or exogenous are all quite highly related to each other and could arguably be defined as a single “species”. In fact the terms used to name these viruses are actually a property of the virus-host pair under study. For example amphotropic means it does not cause disease, so a virus can be amphotropic for one strain of mice and not amphotropic for another strain of mice. Xenotropic means the virus can grow in the cells of other species beside the host species, but none of these viruses has been tested on hundreds of mammalian species, they only get tested on a few very diverse species (ie think lab mouse vs human, rather than lab mouse vs deer mouse).

    XMRV is not very “unique”. Almost every section of it’s genome is between 98% and 100% identical to some previously sequenced MuLV. Some regions seem to come from endogenous MuLVs and some from exogenous MuLVs. This is typical of all exogenous MuLV’s, they recombine with endogenous MuLVs to form new variant exogenous MuLVs. They also sometimes recombine with host non-viral nucleic acids (more likely by packaging an mRNA than by picking up genomic DNA) and if that gene is an oncogene the virus can become a replication incompetent (because the oncogene usually replaces a viral gene needed for repication) oncovirus which needs a “helper virus” to replicate.

    At any rate, all of these MuLV’s and XMRV are highly related, but still easily recognizable as being distinct different lineages of MuLV. The analogy might be that all humans belong to homo sapiens sapiens but it is still quite possible to tell that Asians, Africans, Native Americans and other lineages have distinct recent histories. It is also possible to easily recognize individuals such as Tiger Woods as being of mixed ancestry as far as these recent lineages go.

    To take the analogy a bit further, we can also classify modern humans by occupation, religious belief, hair color or hundreds of other traits, each of which may or may not have any close relationship to relatively recent genetic ancestry.
    If we start mixing up the features used for classification we end up with a mess, not knowing where to put a Buddhist computer programmer Caucasian.

  53. #53 ERV
    August 25, 2010

    CAT– I cannot emphasize enough how little I care about rumors and comments. Remember a few months ago, when “OMFG THE NIH CONFIRMED WPI!!”? And then all we got was this emaciated publication that did nothing of the sort? I want to read papers. Thats it.

    Thanks, Duke! The endogenous MPMVs and PMVs dont clade out from one another in the little trees in this paper, but the XMRVs clearly do… So when are they gonna upload their awesome sequences (comment #50), eh? hehehe!

  54. #54 DrDuke
    August 25, 2010

    It is not clear how long MuLVs have been infecting mice, we have only studied them in various lineages of house mouse really and not in the hundreds of other species of mice such as deer mice and kangaroo mice out there.

    What is clear, is that some of the specific locations or movements of endogenous MuLVs or solo-LTRs left behind by those movements, have been rather recent in the house mouse lineage:

    Adaptive evolution of Mus Apobec3 includes retroviral insertion and positive selection at two clusters of residues flanking the substrate groove.
    Sanville B, Dolan MA, Wollenberg K, Yan Y, Martin C, Yeung ML, Strebel K, Buckler-White A, Kozak CA.
    PLoS Pathog. 2010 Jul 1;6:e1000974.
    PMID: 20617165

  55. #55 CAT
    August 25, 2010

    How is it a rumour if they these quotes are from the press teleconference?

  56. #56 CAT
    August 25, 2010

    “Although we find evidence of a broader group of MLV-related viruses, rather than just XMRV, in patients with CFS and healthy blood donors, our results clearly support the central argument by Lombardi et al. (3) that MLV-related viruses are associated with CFS and are present in some blood donors. ”

    This is from the PNAS paper.

  57. #57 DrDuke
    August 25, 2010

    @CAT 54 and 55.

    Scientists are as bad at spreading rumors as anyone else. What Abbie is talking about is what these scientists say their data means vs what it actually may or may not mean. The work of Dr. Silverman, the Reno group including Dr. Mikovits, and a couple others (a German group with their lung lavage samples) all stressed that “this cannot be contamination because it is unique, not similar to MuLVs”. That assertion is not quite correct because every segment of “XMRV” is essentially identical to previously sequence endogenous or exogenous MuLVs, it is only the particular combination of the fragments that is unique.

    So now the Alter group finds a different MuLV. They did not find some XMRV and some MuLV, they found only one particular sequence of MuLV. Likewise all the other groups find only “XMRV” and never other MuLVs. All of this looks very much like what is expected from PCR contamination.

    Lacking in all of these studies, and discussion of them in the press, is any thought of the epidemiology. How did so many humans get infected with a mouse virus or two mouse viruses? Cats have been eating mice for a few thousand years at least and they have an FeLV that is somewhat related to MuLVs but it is not anywhere in the MuLV family we are discussing here. Humans have been living with mice for thousands of years, and working in the labs with MuLVs for dozens of years and no infections have been detected. It is not impossible that humans are infected with MuLVs of one or more lineages, but if I were interested in this I would START with the epidemiology. Retroviruses don’t just float around in the air and infect hosts at random, they always require close physical contact.

    I can’t think of a virus that was discovered without starting with the epidemiology. If you know of one, please post it here.

  58. #58 CAT
    August 25, 2010

    They also find antibodies.

    The CDC has also tested the WPI samples and been unable to find a contaminant.

    The epidemiology is certainly a puzzle, but not likely to be resolved any time soon.

  59. #59 DrDuke
    August 25, 2010

    @CAT 57

    What titer of antibodies against MuLVs have been found in any of these patients? Which proteins are they binding to? I can imagine people who are not infected with MuLVs having low-level titers of anti-gag or anti-pol antibodies. If they have high levels of antibodies to gag, pol and env it is a different story. So far, I have not seen any publications which specify exactly how strongly reactive, how much anti-MuLV antibody, is being detected in these blood samples.
    Can you tell us where it has been published?

  60. #60 CAT
    August 25, 2010

    In the ‘Science’ paper.

  61. #61 DrDuke
    August 25, 2010

    Yes, in “the Science paper” by Lombardi et al, the dis show that plasma from one patient (WPI-1104) had antibodies that would bind to BaF3ER-SFFV cells and not to BaF3ER cells that were not expressing SFFV envelope. It is not clear which SFFV virus this envelope was from. Some SFFV viruses such as the Mink Cell Focus Forming virus with env sequence K02533 in GenBank, are quite similar to the sequences reported for XMRV. Others, such as the Friend and Rauscher spleen focus forming viruses (GenBank L27432 and K02375) have almost no similarity to XMRV.

    At any rate, there is no titer of antibody with the type of assay done, as far as I can tell. But you are correct that at least it was some apparent anti-envelope antibody found in at least one patient.

  62. #62 DrDuke
    August 25, 2010

    I may have just been unlucky in my search for “spleen focus forming virus” in GenBank. There may be several SFFVs that have high envelope similarity to XMRV. It is also possible that the Friend and Rauscher SFFV sequences I found at first were badly annotated or strange in some other way.

    But at any rate, I’d like to know the titer of virus-specific antibodies in a few patients. It is one think to have a few antibodies to something and another to have some some 5,000 fold more than the background level.

  63. #63 summer21
    August 25, 2010

    http://blogs.wsj.com/health/2010/08/25/does-x-the-virus-that-is-mark-the-spot-in-chronic-fatigue-syndrome/

    “Mikovits says. One of the big unknowns: Are patients who test positive for more than one MLV-related virus at the same time sicker than people who test positive for only one of the viruses? She says that issue is the focus of intense research efforts not only at her institution but in labs around the country.”

  64. #64 CAT
    August 25, 2010

    They found antibodies in plasma in more than 30 patients tested. Around another 10 tested were not positive to antibodies.

  65. #65 CAT
    August 25, 2010

    ” Plasma from 9 out of 18 CFS patients in- fected with XMRV reacted with a mouse B cell line expressing recombinant SFFV Env (BaF3ER- SFFV-Env) but not to SFFV Env negative con- trol cells (BaF3ER), analogous to the binding of the SFFV Env mAb to these cells (Fig. 4D and S6A). In contrast, plasma from seven healthy donors did not react. Furthermore, all nine positive plasma samples from CFS patients but none of the plasma sam- ples from healthy donors blocked the binding of the SFFV Env mAb to SFFV Env on the cell surface “

  66. #66 Levi
    August 25, 2010

    Dr. Duke,

    Your antibody titer levels reasoning is interesting. I looked though the WPI Serology test patent, but got lost.
    http://www.faqs.org/patents/app/20100167268

    I found no mention of titer levels that relate to a positive test finding, other than a repetitive “at least on copy” antigen reference.

  67. #67 Richard Jefferys
    August 25, 2010

    Harvey Alter is quoted as saying: “We have already shown that some of our pts also have envelope sequences.” According to the paper, they found a 240bp env segment in one healthy control and a 206bp segment in one patient. That was it. It looks like they used the same primers WPI used, but WPI detected env in 7 out of 11 patients tested and no controls.

    It’s interesting to me that Randy Schekman is very specific about the need to prove integration in his interview, he mentions it a couple of times (“That I trust will be what he does next”). Is that difficult to do?

  68. #68 ghholm
    August 25, 2010

    @DrDuke #56

    Plenty of viruses have been discovered without epidemiology. Many people know of SV40 (Simian Virus 40), but what about SV’s #1 -39, etc.? These were all discovered through EM studies of various simian cell lines/tissues or similar techniuqes. Mammalian reovirus (respiratory and enteric orphan) was discovered the same way. But, these are all non-pathogenic; if you are talking about the etiologic agent for viral diseases being discovered without epidemiology, I don’t know of any offhand.

  69. #69 ERV
    August 25, 2010

    People talking about antibodies– Dont forget, another lab looked for antibodies, investigated them further, and found exactly what Duke said: they bind to ‘MLV’, but they are NOT MLV specific.

    Richard– I think they only looked in one. Again, this paper is emaciated, and if it were any other person submitting it to any other journal, I frankly cant imagine it would have been accepted.

    On what planet can someone tell a reviewer ‘Yeah… thats a good idea. But I actually want to publish now, so Im gonna ignore your review. PEACE OUT HOMES!’

    Whatever. I said I wasnt going to bitch about PNAS, Im not gonna bitch (anymore).

    Honestly, not much in virology is ‘difficult to do’. We have a lot of shit that is a pain in the ass to do (expensive, time consuming, labor intensive), but not ‘difficult’. I mean, there isnt really anything that one person can do that other people/labs simply cannot do for difficulty reasons (financial reasons, facility reasons, but not difficulty).

    What was asked of Alter is a pain in the ass, but an excellent suggestion. If I submitted this paper, either I would have dug in and done it, or it wouldnt have been accepted.

    But Alter chose to ignore that suggestion.

    A luxury non-NAS members submitting to non-PNAS journals dont have. Aww Im bitching about PNAS again, arent I? Damn. Okay, no more. *zips-lips*

  70. #70 CAT
    August 26, 2010

    The paper was accepted before that though, so getting in a third reviewer was only really paranoia.

  71. #71 ERV
    August 26, 2010

    *heavy sigh*

    *rubs temples*

    Arrogant ignorance… I never get tired of it, really…

  72. #72 CAT
    August 26, 2010

    Randy Schekman is no wet behind the ears push over, so why would he let it through then?

  73. #73 BlimeyWhale
    August 26, 2010

    The results of both the WPI and PNAS groups are perfectly consistent with contamination. In retrovirus terms, the WPI sequences are just about identical with each other and with the plasmid control from Silverman. Notice also that the PNAS study used healthy controls isolated and stored at a different time and geographical location from the other samples.

    Contamination or not, why would you use only healthy controls for such studies? Take a look at the fibromyalgia literature, where the best studies compare FM with healthy AND with MS or other conditions similar in some ways to FM. If you don’t, you can get spurious results. Like those brain imaging studies on FM. Compare FM to healthy controls, and you see different activity patterns. But you can get similar results with even a healthy control group asked to IMAGINE they’re experiencing pain.

    CFS studies should get control samples from people who are bedridden with other disorders, or from people who are stressed or suffer with severe depression. Also, match patients with controls by age, geography AND post-collection handling.

  74. #74 ERV
    August 26, 2010

    CAT– Randy Schekman is no wet behind the ears push over, so why would he let it through then?
    You know everything, so why dont you tell me?

    [/patience]

    Blimey– *nod* I didnt think of that, but youre very right. Im not an immunologist, but in immunology journal club I remember reading paper after paper where they used ‘similar but should be different’ diseases as control groups. Though, I cant think of any XMRV paper that has used these kinds of controls (eg looking at aggressive prostate cancer vs aggressive testicular cancer or aggressive bladder cancer whatever).

  75. #75 John
    August 27, 2010

    Goddamn it, there’s so many layers to this whole thing and virtually nobody tells it like it is- they all leave shit out or spin shit or whatever (or is it just me?).

    Dr. Duke’s question about antibodies reminded me about this CDC XMRV/prostate cancer study abstract which was presented at CROI 2010 and yep, guess what- CDC found 2 positives and neither had detectable antibodies, so CDC itself is saying that they can’t find antibodies in patients they consider positive, yet CDC trumpets the fact that they didn’t find antibodies in WPI’s ‘XMRV-positive’ samples that they tested.

    From the CDC abstract- “Of 165 prostate tissues, 2 (1.2%) were positive by pol and env polymerase chain reaction and had undetectable mouse mtDNA…Plasma from both persons (5956 and 6203) were negative by reverse transcriptase-polymerase chain reaction indicating absence of viremia. Both patients were Western blot-negative…The finding of undetectable antibodies and viremia in 2 patients is noteworthy and may reflect sequestered or cleared infections.”

    I wonder what they mean by ‘and had undetectable mouse mtDNA’. Did they find other ‘positives’ that did have what they considered to be mouse DNA? The entire presentation is online for those who know more about this stuff and are interested. CDC’s ‘positives’ also were only related to XMRV while not being exactly the same XMRV as reported thus far by the WPI/Cleveland Clinic.

    Abstract- http://www.retroconference.org/2010/Abstracts/37160.htm

    Presentation- http://www.retroconference.org/2010/data/files/webcast_2010.htm (XMRV abstract presentations took place on Friday; scroll down about 1/3 of the way to find the oral abstract presentations. There are several talks before getting to the XMRV presentations, scrolling down and clicking on slide 137 will take you to the start of the XMRV presentations)

    There also was a paper published recently by Abbott on antibodies in infected monkeys, where they also tested a thousand or so human blood donors-
    ‘Characterization of antibodies elicited by XMRV infection and development of immunoassays useful for epidemiologic studies’
    http://www.retrovirology.com/content/7/1/68

  76. #76 DrDuke
    August 27, 2010

    John in #74 wrote:
    “From the CDC abstract- “Of 165 prostate tissues, 2 (1.2%) were positive by pol and env polymerase chain reaction and had undetectable mouse mtDNA…Plasma from both persons (5956 and 6203) were negative by reverse transcriptase-polymerase chain reaction indicating absence of viremia. Both patients were Western blot-negative…The finding of undetectable antibodies and viremia in 2 patients is noteworthy and may reflect sequestered or cleared infections.”"

    It also may reflect FALSE POSITIVE PCR results for those 2 patients. No test is 100% perfect, the ones with the greatest sensitivity tend to have a higher false positive rate. Tests designed to eliminate false positives tend to have a higher false negative rate. PCR in general is prone to problems with PCR product from earlier reactions (such as testing the primers on a positive control sample) winding up in the input to later reactions. Thus, proving that there is no mouse mitochondrial DNA in the samples does not eliminate the possibility of this type of contamination.

    PCR contamination does not indicate sloppy or careless work. It happens in every lab that does PCR. The best labs can nearly eliminate it, but the key is to have methods of detecting it when it happens, as well as methods to prevent it in the first place.

    The real problem with all this work is not who can or cannot find PCR products. The problem is that nobody is even thinking about the epidemiology, and whether or not that could possibly line up with one or more murine retroviruses infecting thousands of humans. Cats have been hunting mice for tens of thousands of years, and they indeed do have a feline leukemia virus that is somewhat related to MuLVs. But that looks like one transfer from mice to cats in the past thousands of years. Millions of cats eating tens of millions of mice and one transfer. Human lab workers have been handling not only mice but also high-titer preparations of dozens of MuLVs for more than 30 years, and no infections have been noticed among them.

    None of that proves that MuLV could not possibly infect humans, but it sure does throw up a large red flag saying that the epidemiology needs to be looked into. And this is true not only for XMRV but also for “Human Mammary Tumor Virus” and a few other alleged viruses that can only be found by PCR.

  77. #77 Impish
    August 27, 2010

    “Human lab workers have been handling not only mice but also high-titer preparations of dozens of MuLVs for more than 30 years, and no infections have been noticed among them.”

    Did anyone look? Did any of them come down with CFS? Just because something wasn’t found previously or no one made the CFS –> MuLV link doesn’t mean it wasn’t there. That is some poor logic. HIV was kicking around in humans for a long time with no one noticing and it seems to be relatively easier to find.

    I apologize for my limited biology but from my reading of papers it seems that in mice genetics comes into play with infections with MuLV. Some mice can resist infection and some cannot. There also was a recent paper that showed a co-infection can reduce the genetic resistance and allow infection. These are all possible paths to explain what is happening in humans, correct?

    Also doesn’t this http://www.nature.com/ng/journal/v21/n2/abs/ng0299_216.html suggest that the same receptor that allows XMRV to attack cells also allows MLV in? If humans have XPR1(SYG1) and that allows in XMRV as seen in http://www.pnas.org/content/104/5/1655.abstract wouldn’t that also allow MulV in humans?

  78. #78 BlimeyWhale
    August 27, 2010

    HIV is relatively easier to find because in almost all cases it provokes a rapid and sustained antibody response. Unfortunately, that doesn’t help the host as much as it does with other infections because HIV evolves rapidly. So do all retroviruses except the so-called XMRV, which looks to have been in genetic stasis since the 1980s. More likely, the constant is the VP62 plasmid that many of us suspect is contaminating these studies.

    Re: XPR1 and the Villinger/Abbott results, monkeys and humans have approximately the same XPR1 protein, but only two of three monkeys had detectable viremia despite inoculation with 3.6 million (!!) tissue culture infectious doses (50) of “XMRV,” and in those two monkeys, the viremia was low and could be detected only until about a month after exposure. Ultra-sensitive antibody tests developed with this exposure model to detect mouse viruses came back positive for only 3 of 2382 human samples, and none of the 3 was genuinely positive for more than one of the three antibody tests. Cross-reactivity.

    The paper tells us there’s no immunologic evidence of mouse viruses in 2382 of 2382 healthy or HIV-1 or HTLV-1 infected humans. It also suggests that mouse viruses replicate very poorly and only briefly in primates even when massive doses are introduced directly into the bloodstream. And that nucleic acid and antibody tests might work only during a short window early after exposure.

  79. #79 John
    August 28, 2010

    “The paper tells us there’s no immunologic evidence of mouse viruses in 2382 of 2382 healthy or HIV-1 or HTLV-1 infected humans. It also suggests that mouse viruses replicate very poorly and only briefly in primates even when massive doses are introduced directly into the bloodstream. And that nucleic acid and antibody tests might work only during a short window early after exposure.”

    Is that statement entirely accurate? Would it possibly be more accurate to say that ‘It also suggests that mouse viruses replicate very poorly and only briefly in the bloodstream in primates?

    I ask this because the same group that presented the antibody paper also presented another paper at that conference, with the second one having this to say- “We experimentally infected 5 healthy rhesus macaques with XMRV intravenously…Extensive tissue collections were done from various organs at necropsy to evaluate both the tissue and cell tropism at various times post infection using FISH to the entire genome and IHC via detection of XMRV gag using a cross reactive monoclonal antibodies to murine spleen focus-forming virus (SFFV)…Both methods were concordant for the detection of XMRV in the various organs tested and showed a wide dissemination of replicating virus even when the plasma viral load was undetectable.”
    http://www.retroconference.org/2010/Abstracts/39855.htm

    So by combining the two papers, would it possibly be accurate to say that both PCR and antibody tests done on blood might not work even though a potentially infected individual could have a ‘wide dissemination of replicating virus’?

  80. #80 daedalus2u
    August 28, 2010

    John, but they found that the antibody titres went down over time. That implies that there wasn’t an ongoing infection that was replenishing the population of virus that was causing the antibody formation. The antibody titers went back up with a reinoculation, so it wasn’t some kind of immune system dyfunction but low antigen load.

  81. #81 aidan walsh
    August 28, 2010

    45% of the hiv virus envelope inserted into mycoplasma is the cause and not any strains of a mouse model!! garth nicolson is the real genius out of all this…aidan walsh southampton,u.k. maybe this will stop those phyco-babble theories once and for all! i tested positive to garth’s method and would not ever consider testing for any mouse type strains and stay very far from any aids drugs!!

  82. #82 Levi
    August 29, 2010

    Garth,

    I agree that although a medical treatment would be nice to offer ME/CFS patients, I believe its way too premature to even think of messing around with anti-retrovirals for CFS patients; the drugs can have very serious side effects.

    I doubt folks here will be interested in a PDL discussion. Also, since most of the CFS samples in this study come from a former mycoplasma study by Dr. Lo with across/board negative findings, your own status with regard to the microbe is not likely to be germane to this particular study. Best wishes.

  83. #83 CAT
    August 29, 2010

    CAT– Randy Schekman is no wet behind the ears push over, so why would he let it through then?
    ERV –You know everything, so why dont you tell me?

    I don’t know, you tell me?

    As for ARV’s, would they be worse than the disease? I think not in a lot of cases. Even the CDC says it’s as serious as AIDS, end stage Cancer, MS.

  84. #84 Levi
    August 29, 2010

    CAT,

    ME/CFS patients run along a wide continuum of symptomology. There are likely to be different causes for unexplained illness in different ME/CFS patient subsets. I submit that it would be very bad practice to subject patients from the Wichita/Georgia CDC dial-up cohort to ARVs. Lo/Alter tested their blood, and could not find XMRV/MuLVs in those patients either.

    When it comes to the practice of medicine, first do no harm. Right? I think the issue of what a “well-pedigreed” CFS patient is will be a sensitive and sticky one in any future research. Researchers will likely need to use biomarker profiles to arrive at a meaningful cohort, rather than the vague and non-uniform definitions used in the zero/zero studies.

    Assuming for the sake of argument that ME/CFS is as serious as AIDS, end stage Cancer, and MS; in two of three cases with your example, you would just make the patient much worse using ARV’s on them. Anecdotal initial accounts of ME/CFS patients who have actually forged ahead with ARV treatments so far have not been promising. Some have become much worse. Everyone should proceed with caution and intelligence with ARV’s for ME/CFS patients. There will be no substitute for good science if what is desired is a good result.

  85. #85 CAT
    August 29, 2010

    So don’t use the CDC revised Fukuda definition. Select patients from expert doctors.

    As for making patients worse on ARV’s it’s never been tried before, so no one has any idea.

    So far only a handful of patients have tried ARV’s, and reports have been fair. But it’s still anecdotal.

  86. #86 RRM
    August 30, 2010

    Criticizing the cohort does not make much sense to me at this moment. Even if the cohorts were as bad as some would like us to believe, that circumstance could never explain all the negative results. Only if you would like to believe that the criteria of the ‘zero studies’ not only managed to exclude ALL ‘true’ CFS patients, but ALL healty but infected conrols too, it would start to make some sense.

    Add to this that the Van Kuppenveld group was able to detect XMRV in samples provided by the WPI and that the WPI was able to detect XMRV in some negative (according to the Van Kuppendveld group) samples, and you must conclude the real solution must lie within the testing methodology (false positives/false negatives/contamination/teh tubezz).

    And CAT: the third reviewer was really only the first *independent* reviewer, as the paper was initially accepted by two reviewers of Alter’s own choosing. This had nothing to do with “paranoia” or some high level conspiracy to suppress the truth. As for you second remark (‘why would he let it through?’) it’s also pretty simple: as a NAS member, Alter had the RIGHT to publish without independent peer review. Although the editor could still reject a paper if it was apparent tripe, it doesn’t make sense to apply the exact same standards to Alter that normally would apply to non-members. Why would you otherwise grant these special (lower) standards to NAS members in the first place? All Schekman could reasonably do in the end, was to publish the paper, but with some explanation to the community.

    (English is not my native language BTW, so sorry for my far from perfect writings)

  87. #87 BlimeyWhale
    August 30, 2010

    John, I respect the scientists whose abstract you quoted, but I hope their submitted paper has more specific detection techniques. “FISH to the entire genome” is sometimes necessary to find retroviruses, but it’s not ideal because compared with standard FISH with one 100-200 nucleotide probe sequence, the entire genome probe collection is randomly fragmented sequences covering 8000 nucleotides. That much more opportunity for nonspecific hybridization.

    And you can’t be sure of “detection of XMRV gag” when “using using a cross reactive monoclonal antibodies to murine spleen focus-forming virus (SFFV).” In the published PNAS paper they mention PCR, but did they do Taqman qPCR for DNA and RNA in these tissues? Did they sequence clones from these tissues? Should be interesting to read about.

  88. #88 KAL
    August 30, 2010

    Good range of comments here. Several metagenomics studies are already underway which include CFS patients and results have the potential to be interesting.

    As for viruses discussed here – this is a relatively new area and it is not surprising that there is conflict. There are more studies that should be published in the coming year. Whether they will clarify or add to the confusion is unknown at this time. As Abbie correctly notes, hard to say until you read the actual paper.

    As for anti-virals in CFS patients, there are already pilot studies in the literature showing anti-virals can be successful against other viruses associated with CFS unless the patient has multiple viruses in the mix.

    Or if there is not an anti-viral developed for the specific virus, immune modulators have proven successful as well.

    Pilot studies of anti-virals that target retroviruses would certainly tell you something is up if they work. I wouldn’t spend a great deal of time worrying that hordes of patients are going on anti-virals outside of clinical trials or even in cost recovery trials. Have you priced them lately?

    Dr. Duke – not sure what you mean by the CFS crowd. I’m afraid you need to distinguish between researchers and patients although there are some patients who are qualified to do professional research – viruses and diseases don’t give a rat’s ass what your education or profession is.

    If you are unsure about the research, look it up. Navigating databases isn’t all that hard. Don’t have time? Contact someone at the International Association of Chronic Fatigue Syndrome/Myalgic Encephalomyelitis. They are the largest group of professional researchers, clinicians and educators specializing in both biomedical and behavioral areas. IACFS.org

    And there are appropriate neuropsychiatric studies if you look. Results however, are dependent on the appropriateness of the instruments partly due to confusion because without assessing severity symptoms can be attributed to both organic disease and depression. Other variables are how the instruments are used and how much depends on assumptions by administrators of the instruments – the DIS for example. The issue in research isn’t whether there are psychosocial overlays – those are present in any organic disease.

    The main issue is defining patients so that you aren’t mixing in people who do not have a disease. One can hardly attribute signs or symptoms or even viruses to a disease if the people studied don’t actually have the disease in question.

    Sorry ERV – wandered off topic.

    Good thread

  89. #89 Alex
    September 1, 2010

    Who manufactures most PCR reagents? I understand Roche has a patent on part of the process. It might be an interesting experiment to buy reagents from all the manufacturers and set up a lot of parallel negative controls – as long as you yourself could be confident of your own.

    The whole story reminds me of the famous computer security paper “On Trusting Trust”.

  90. #90 DrDuke
    September 1, 2010

    @Alex 88 If all labs followed standard lab practice to blind all samples and only break the blinding code at the right time at the end of a study, we could eliminate many “trust” issues.

    The best researchers are not those who are unbiased, but those who realize that all humans can be expected to be biased and to screw up in dozens of other ways, and therefor design truly fool-proof systems that can’t be broken by any type of conscious or unconscious human error.

    Experiments have to be designed with a lot more attention to disproving the hypothesis than supporting it. Not all people who call themselves scientists are equally well trained and rigorous about following the scientific method at every step and turn in their research.

  91. #91 DrDuke
    September 1, 2010

    @self #89

    I am not implying that I know or suspect that the labs studying XMRV are not following every standard protocol including blinding of all samples. It is possible for example that in the paper published by Lo and Alter, the researchers ran all samples and controls randomly, and then only re-created the 2 gels shown in the paper, one with mostly negative and the other mostly positive, after unblinding the coded samples. They might have done that for more simple and clear illustration in the publication.

    However, if they did know which samples were CFS and which were the healthy donors, during the experiments, the possibility of some type of conscious or unconscious bias cannot be completely eliminated.

  92. #92 RRM
    September 2, 2010

    @Duke, regarding blinding:

    Q: Were the samples tested under blinded conditions?
    A: We did not specifically blind and mix the two sample groups (CFS and blood donors). However, they were studied in parallel by polymerase chain reaction (PCR). As shown in the figures of the paper, those samples with positive amplicons of the predicted sizes were all sequenced; no sample was considered positive unless sequencing confirmed PCR reactivity

  93. #93 thomasbernhard
    September 2, 2010

    Pseudo-Doc Matthew:
    “My complaint isn’t that the psychological disorders associated with CFS are unique, but rather that advocates of the diagnostic label are overly dismissive of them and frequently discourage treating them.”

    This is just bullshit. It’s cheap, it’s banal, and it’s a massive generalization, based on a tired cliche. Not even deserving of a response, but I feel like sticking up for Eric.

    There are a lot of wackos on the internet. Some of these wackos have, or claim to have CFS. They don’t represent all ailing CFS-ers – how could they? The majority suffer in silence. And the majority are sensible, but sick. The fact is that the vast majority of CFS suffers would prefer an illumination of their pain as corporal manifestation of psychic turbulence – because that’s better than a multi-system, unexplainable, chronic and incurable debilitating illness.

    The whole cliche of the war between the psychologizers and the neurasthenics is so fucking tediously boring.

    Re: your other ridiculous assertion that laziness begets fatigue: perhaps, but this has no relevance with CFS. It’s just the opposite. Many CFS-ers are active gym-rats and runners and cyclists and hikers,etc, when they become sick. In fact, recovering from CFS sometimes necessitates lassitude. Lying in bed all day is sometimes the only way they can get better.

  94. #94 synapse13
    April 9, 2011

    Let’s keep in mind that there may be a very good reason why the WPI found XMRV and the NIH/FDA found other MLVs. The cell line used by WPI may preferentially produce XMRV over other MLVs and is thus showing evidence of just one of the retroviruses that may be involved. I believe this is why Alter/Lo are thinking towards the whole family of MLVs being involved.

  95. #95 Charl
    April 9, 2011

    “The cell line used by WPI may preferentially produce XMRV over other MLVs and is thus showing evidence of just one of the retroviruses that may be involved.” Maybe it’s because it’s late here in the UK, but… that just makes no sense. What biological mechanisms are you imagining for the co-culture method to result in one kind of murine leukaemia virus being produced but not another? Why would only one kind be detected?

    There’s some thought that in double-positive (KSHV, EBV infected) cell lines, KSHV lytic genes may inhibit EBv lytic reactivation. But you can still detect both of them, and not by bloody co-culturing them with another highly permissive cell line! If the WPI have to co-culture the cell they take to find them “positive”, how is the virus supposed to be replicating and causing disease in vivo?!?