Hosts: Vincent Racaniello, Rich Condit, and Dickson Despommier
Vincent, Dickson, and Rich meant to do an all-email episode, but first they review results of the Blood XMRV Scientific Research Working Group, and partial retraction of the paper associating XMRV with chronic fatigue syndrome.
With this episode TWiV is three years old.
[powerpress url=”http://traffic.libsyn.com/twiv/TWiV150.mp3″]
Click the arrow above to play, or right-click to download TWiV 150 (56 MB .mp3, 93 minutes).
Subscribe to TWiV (free) in iTunes , at the Zune Marketplace, by the RSS feed, by email, or listen on your mobile device with the Microbeworld app.
Links for this episode:
- Partial retraction (Science)
- Failure to confirm (Science)
- False positive (Science)
- Why a partial retraction? (Retraction Watch)
- TWiV on Facebook
- Letters read on TWiV 150
Weekly Science Picks
Dickson – The Tree of Life
Vincent – When do you fact-check article content with sources? (take as directed)
Rich – io9
Listener Pick of the Week
Luis – NIH videocasting and podcasting
Send your virology questions and comments (email or mp3 file) to twiv@microbe.tv, or call them in to 908-312-0760. You can also post articles that you would like us to discuss at microbeworld.org and tag them with twiv.
What an interesting story XMRV has been. Thanks for covering it all as it came along.
re blinding: I think that the samples for the Science paper were blinded prior to testing, but they were not drawn/collected in the same way. The blinding only came in half way through, and contamination could have occurred prior to that point. (I could be wrong about this, as it’s never been clearly explained in one place, but I think that’s right).
re the Guardian CFS piece: It was a shame that they didn’t take any time to consider if there were any legitimate complaints to be had against these researchers – which is surely an important part of understanding any hostility they may attract.
While sending threatening e-mails to researchers is clearly unacceptable, there are a number who could do with a good telling off, including both Simon Wessely and Judy Mikovits imo. When researchers behave badly, it can have real and damaging consequences for patients.
The T-shirt is obscene. Regardless of how the XMRV story plays out it treats this serious disease as a joke. I am rather stunned that you’ve stooped to that level. If you don’t grasp the seriousness of ME/CFS in destroying lives of millions then I’m not sure anything you have to say is worth listening to.
To the poster who compared Mikovits to Wessley, what planet do you live on? Putting aside validity of either person’s research, Wessley’s theory of CFS is that it is caused by “false illness beliefs”. That nonsense alone discredits him from any discussion. I say this a clinical psychologist who’s been disabled by ME/CFS. The worst one can say about Mikovits is that she may have made some errors in looking for a retrovirus. And the jury is still out on that. But she clearly understands the terrible seriousness of the disease.
michael allen, ph.d.
Your overreaction is what is obscene. The t-shirt doesn’t say anything about the disease. False positive results of this virus are not just an idea, it is a fact. Moreover, this virus virus was first found in a prostate cancer, and it looks like that was also false positive. People who don’t have CFS also got tested and got false positive results (to test for the ridiculous autism connection). If you actually listened to the podcast you would’ve seen that Drs are advocating for more research in the field just not in the XMRV direction.
.the vp-62 clone is a composite made from different patient,s tissues
the primers originally used by silverman were based on vp-35
dr mikovits and dr Ruscetti lowered the annealing temperatures and increased the magnesium concentration in their PCR to increase sensitivity and this is why they picked up the entire community of gammaretroviruses present in the cohort
They could not detect env sequences because the env primers could only pick up vp-62 and they were too different even for the lower annealing temperatures to detect
the VP-62 clone does not exist in nature so all the 00 studies which set their PCR to detect the synthetic VP-62 clone would have no hope of detecting a real virus
The interesting point is how does john coffin now explain how he demonstrated that two gammaretroviruses recombined to form an artificial clone which does not exist in nature when one of those gammaretroviruses has not been demonstrated to exist in nature either
The virus (XMRV) has been shown to exist, it just hasn’t been shown to be present in patients, that’s all.
It would be hard to say no-one can find the virus if it’s not there. A bit like trying to see my dog.
(I don’t have a dog so matter how amazing your dog detecting modalities are, you won’t find it.)
Thanks for another nice TWIV (I see this is the place where the comments are coming in, so I am “spamming” my response from TWIV.tv here….)The authors of the original Science study did address the blinding issue. They did so in the Science response to some comments by other scientists that were sent in: http://www.sciencemag.org/content/328/5980/825.4.fullI think the ‘discrimination’ of the contaminant is better explained by the fact that controls and patients were not collected and handled in exactly the same way before they were blinded/tested. It seems likely that contamination crept into (some of) the CFS patients’ samples before they were “mixed” with the control samples.What I do find very strange however is that Silverman’s co-authors from the Lombardi et al. study refused to share their data with Science regarding their own experiments following Silverman’s revelation (which they assert show that they didn’t contaminate their samples with VP62). From the Retraction Watch article you’ve addressed in the talk, Science exc. editor Bruce Alberts is quoted as saying:”While we were aware that other co-authors had tested samples and claimed to not find evidence of plasmid contamination, those co-authors were unwilling to provide their data for examination so we were unable to comment on the validity of the other experiments.I don’t get why they didn’t do that and I’d like to hear your thoughts on this.Anyway, from the Science conditions on data availability:”All data necessary to understand, assess, and extend the conclusions of the manuscript must be available to any reader of Science. All computer codes involved in the creation or analysis of data must also be available to any reader of Science. After publication, all reasonable requests for data and materials must be fulfilled.”Given this, I think they should have demanded relevant data from these co-authors. Even better, they should ask the co-authors to let their positive Lombardi et al. samples tested by other labs (e.g. the Silverman lab), of course using proper blinding and controls. This would seem (at least, to me) a pretty easy way to check if the contamination reported by Silverman originated in the WPI lab.
Too bad the copy/pasting from the TWIV site messed up the paragraphs… 🙁
Why exactly would the autism connection be ridiculous, drosha?Â
Â
It is well known that HIV infection in children often VERY closely matches both various pathologies found in idiopathic autism, and more importantly the neurological symptoms of HIV-encephalopathic children are IDENTICAL to symptoms and presentations of idiopathic autism. (with ‘autism’ itself being nothing more than a collection of symptoms and presentations…).
Â
Paediatric HIVÂ provides us with a ready off-the-shelf, living, breathing human model of autism.Â
Â
Various retroviruses (including murine retroviruses!) have been shown to cause: widespread central nervous system dysfunction, microglial activation and inflammation, vascular endothelial inflammation, vasoconstriction and permeability, oxidative stress, systemic glutathione depletion, mitochondrial dysfunction, gluten and casein reactivity, autoimmune reactivity, glutamate toxicity, intestinal permeability, microbial translocation, hyperplasia of intestinal epithelial cells, pancreatic enzyme deficiency, disaccharide intolerance and malabsorption, autonomic/vagal stem dysfunction, abnormal cytokine profiles and abnormal gene methylation.
Â
ALL of those pathologies have to date been observed in idiopathic AUTISM.
HIV encephalopathy is very different from autism.
There is no connection between XMRV and any disease at the present time.
“… Pamela Walters
and colleagues report that children with HIV- caused encephalopathy (brain
inflammation) frequently develop
symptoms of autism…”
Neurodevelopmental
findings in children infected with HIV retrovirus include impairments in language, especially expressive
language, behavioural symptoms: irritability, lack of social skills, repetitive
actions (rocking etc). Severity of autistic symptoms in HIV positive children
is correlated to levels of retroviral load/replication, as well as CD4+ levels.
Symptoms of autism – deficits in language, behaviour and social skills – in HIV
infected children often recover upon administration of single or combination
antiretroviral treatments, at least to some degree. Sometimes recovery is
complete, with total remission of autistic symptoms.
From
Moss at al ‘The Development of a Q-Sort Behavioral Rating Procedure
for Pediatric HTV Patients’. Journal of Pediatric Psychology, Vol. 19, No. I,
1994, pp. 27-46.
Paediatric HIV
encephalopathy symptoms:
Rarely
has eye to eye contact with others
Difficult
to engage in social interaction
Emotional
responses not appropriate
Has
frequent temper tantnuns/irritable
Rarely
initiates activities or interaction
Rarely
smiles
Responds
minimally to social interaction
Does not
seem to enjoy play activities
Is easily
frustrated
Appears
underactive/passive
Seems
withdrawn
Does not
respond to comforting when distressed
Is mainly
interested in objects rather than people
Rocks
head, body in a repetitive way
Has
difficulty remaining focused on a task
Has a
short attention span
Shows a
lack of social responsiveness
Is fussy
and cranky for no apparent reason
Â
Additional references:
Impairment of Expressive Behavior in Pediatric HIV-Infected
Patients with Evidence of CNS Disease, Moss AH et al , Journal of Pedialric
Psychology. Vol 21. No. 3. 1996. pp. 379-4O0
Early language development in children
exposed to or infected with human immunodeficiency virus. Coplan J et al
Pediatrics. 1998 Jul;102(1):e8.
Neurodevelopmental/neuroradiologic
recovery of a child infected with HIV after treatment with combination
antiretroviral therapy using the HIV-specific protease inhibitor ritonavir.
Tepper VJ et al Pediatrics. 1998 Mar;101(3):E7.
Neurologic, neurocognitive, and brain
growth outcomes in human immunodeficiency virus-infected children receiving
different nucleoside antiretroviral regimens. Pediatric AIDS Clinical Trials
Group 152 Study Team. Raskino C et al Pediatrics. 1999 Sep;104(3):e32.
Neuropsychological functioning and
viral load in stable antiretroviral therapy-experienced HIV-infected children.
Jeremy RJ et al Pediatrics. 2005 Feb;115(2):380-7.
Neurocognitive functioning in pediatric
human immunodeficiency virus infection: effects of combined therapy. Shanbhag
MC et al Arch Pediatr Adolesc Med. 2005 Jul;159(7):651-6
Â
“Why exactly would the autism connection be ridiculous”
Couple of reasons:
1. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2964727/?tool=pubmed
2. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3043069/?tool=pubmed
“Various retroviruses (including murine retroviruses!) have been shown to cause:…”
You forgot to add the fact that those finding were in animal models since there are only 2 human retroviruses.
By that logic, we also cured cancer and various genetic diseases in mice, should we just close the chapter on those too?
Incorrect. Most of those findings are actually from human retroviruses, and a good number are from findings in children (who accidentally also develop autism, see references below)
all of your references are from HIV patients, how does that qualify as “Various retroviruses (including murine retroviruses!)”
all of your references are from HIV patients, how does that qualify as “Various retroviruses (including murine retroviruses!)”
I said ‘most’, not ‘all’.
Lance and Drosha,
Â
The key is age of onset of the HIV encephalopathy. When the brain of a very young child is affected, the outcome is quite different than for an adult with a fully developped brain. Adults don’t develop autism, only very young children do. HIV encephalopathy does not present as autism in an adult, but it does in very young children, as stated by the Guest poster (and supported by the refreences Guest posted).
I also don’t think that proposing that many cases of autism could be caused by a (yet unknown) retrovirus is ridiculous at all. Quite the opposite, actually, I think it is a hypothesis that makes a lot of sense (again, read the studies posted by Guest), and so far there aren’t many other hypothesis on the table to explain why the rates of autism have been exploding in the past decades. We still don’t have the beginning of an answer there, and it is a tragic shame (I am saying that as the mother of an autistic child myself).
Could a yet unknown retrovirus be causing the autism epidemic? Nobody seem to be interested in looking into this. What kind of studies could be done to start answering this question, or at least try to rule out this hypothesis?
I have been thinking of comparing the rates of autism in children who were never breastfed versus those who were breastfed at least six months. If a retrovirus is the root cause of the autism epidemic, then it has to be infecting babies at birth or shortly thereafter (since autistic regression usually starts in the second year), and an obvious route of contamination would be from the mother, through breast milk.
I am aware of on small study that looked into this, I wish there were larger scale studies to confirm it:
“The percent of ASD patients who were breastfed at 6 months was 37% as compared to 13% of controls (p of 0.003)….In contrast to previous studies, our interview study yielded significantly higher rates of breastfeeding in subjects with autism as compared to controls with other developmental disabilities.”http://imfar.confex.com/imfar/2010/webprogram/Paper6362.htmlBreastfeeding and AutismSaturday, May 22, 2010Franklin Hall B Level 4 (Philadelphia Marriott Downtown)10:00 AMP. G. Williams , Pediatrics, University of Louisville, Louisville, KYL. L. Sears , Pediatrics, University of Louisville, Louisville, KY
Sorry the link isn’t showing up right. here I try again:
http://imfar.confex.com/imfar/2010/webprogram/Paper6362.html
“again, read the studies posted by Guest”
1. How about you read the studies that I posted! 2 recent papers showing no correlation with autism and MLVs.
2. For your breastfeeding retrovirus transmission theory to work, every sibling of an autistic child has to be also autistic but that’s not the case. “Retrovirus is hiding” idea doesn’t work there either because maternal transmission of HIV is well established and breastfeeding is a frequent event.
3. There are a lot of theories about increase in autism rates and one important one being the change in the diagnosis criteria. Not every autistic child has the same symptoms, it’s a spectrum of diseases. It’s extremely unlikely for such a wide spectrum of diseases to be caused by a simple retrovirus.
Actually you said “various retroviruses” even though you only came up with one (HIV) and you said most of the findings are from human retroviruses even though you only came up with HIV references.
1. I did read them, thanks. I wasn’t talking about MLVs here, but yet unknown retroviruses. I am glad the connection with MLVs was looked into though, because it was a plausible hypothesis, in my opinion.
2. Why would all siblings be affected? Not all carriers might develop symptoms. It might depend on other environmental factors, illnesses, timing of infection, etc…
3. It might not be a simple virus. Take HIV for example, not all patients develop the same symptoms: not all of them develop encephalopathy, patients (used to) die of various infections, the latency period is very variable etc… Same for HTLV, the symptom presentation and timing is very variable from patient to patient.
What’s in a name?   XMRV has come to mean the V62 clone that was created in Silverman’s lab, as expressed by 22rv1 cell line, which has never been in the WPI labs.Lombardi et al found gag and env sequences 99% similar to the prostate tumor-associated strains of XMRV (VP62, VP35 and VP42) yet Silverman’s retraction was only for VP62, which incidentally is the clone used as a positive sample in just about all papers that failed to find any wild HGRV’s at all.  I wonder if all the conflicting results have something to do with the semantics surrounding the way the labels are applied?Why are patients invested in the HGRV hypothesis?  Because, if you had this disease, with the inside knowledge that that brings, it makes perfect sense.  When you bring in the years of research on what MLV’s do in animal models, when you bring in the several earlier discoveries of retroviruses in patients with ME (called CFS since 1988), when you know people who have at the least improved their illness experience using anti-retrovirals (early days yet), you too would not be so quick to dismiss the hypothesis.To get to the BWG 3 results, the following criticisms of the methodology has been made by scientists, patients and advocates:*The samples used were from patients who were continuing with their medication regimes, to include Ampligen and anti-virals.*No preservative was used to preserve PBMC’s in samples*Negative controls were not proven to be negative*Spiked positives were spiked with VP62*Sample size was too small to draw meaningful results.I would love to hear Vince’s response to these bullet points.
What happened to my formatting? Â
Reposting to improve formatting. Â You may not know that one of the cognitive problems people with ME have is difficulty in reading blocks of text. Â Please delete the previous version of this comment. Â Thank you.
What’s in a name?   XMRV has come to mean the V62 clone that was created in Silverman’s lab, as expressed by 22rv1 cell line, which has never been in the WPI labs.Lombardi et al found gag and env sequences 99% similar to the prostate tumor-associated strains of XMRV (VP62, VP35 and VP42) yet Silverman’s retraction was only for VP62, which incidentally is the clone used as a positive sample in just about all papers that failed to find any wild HGRV’s at all.  I wonder if all the conflicting results have something to do with the semantics surrounding the way the labels are applied?Why are patients invested in the HGRV hypothesis?  Because, if you had this disease, with the inside knowledge that that brings, it makes perfect sense.  When you bring in the years of research on what MLV’s do in animal models, when you bring in the several earlier discoveries of retroviruses in patients with ME (called CFS since 1988), when you know people who have at the least improved their illness experience using anti-retrovirals (early days yet), you too would not be so quick to dismiss the hypothesis.To get to the BWG 3 results, the following criticisms of the methodology has been made by scientists, patients and advocates:*The samples used were from patients who were continuing with their medication regimes, to include Ampligen and anti-virals.*No preservative was used to preserve PBMC’s in samples*Negative controls were not proven to be negative*Spiked positives were spiked with VP62*Sample size was too small to draw meaningful results.I would love to hear Vince’s response to these bullet points.
Something seriously wrong with the formatting on this comments section. Â Please help! Â Delete reposts! Add para breaks! Â Thank you.
Patient samples tested positive for gag sequences at the NCI and WPI using nested reverse transcriptase PCR. Â Of these samples a portion was retained from each and the remainder exported to Dr Silvermans lab for sequencing.The retained sample aliqots have since been tested by an independent lab and found negative for any trace of mouse mitochondrial DNA, IAP sequences or the VP62 plasmid. Â Additionally, gag sequences were detected which were nothing to do with VP-62. Thus when the samples left the WPI and NCI they were not contaminated.Following Dr Silvermans single round PCR the VP62 plasmid was detected in several of the samples and the virus he sequenced was VP62. Â In short he had sequenced the contaminant that he had introduced into some of those samples.VP62 is not correct sequence for the viruses in the samples exported to Dr Silverman by the NCI and the WPI. Â The WPI and NCI were not able to detect gag or env sequences in retained portion of the samples using silvermans methods and primers.This explains the negative studies and the Blood Working group. Â In particular it should be noted that J Levy replicated Silvermans methods and not those of the NCI and the WPI. Â Unsurprisingly he could not find a HGRV in patients testing positive using the methods of the WPI and NCI.
I’ve just listened to the first 25 minutes of your podcast.
I would like to draw your attention to comments made by Harvey Alter on 8th September in Glasgow at the annual BBTS conference. He said that it had been a mistake that this study with Lo 2010 had not been a true replication of Lombardi 2009, and that there had indeed still not been an attempt at a true replication of Lombardi. Therefore I find your comments re lack of “reproducibility”
ill-informed. He also stated in response to a question re antiretroviral studies that he would not expect to see the success so far observed were there not a retrovirus at play in ME/CFS.
I am disappointed that none of the panel on the podcast saw fit to draw attention to the quality of the so called “negative” XMRV studies – eg assays used (unvalidated/ unproven), cohort selection, failure to use techniques/ methodology as per WPI specification. Deeply reminiscent of de Freitas’ efforts to get the scientific community to closely follow/ understand the intricacies of research protocol (eg, don’t freeze the samples) way back in 1991.
You seem confused by the low rates of “contamination” in controls as compared to those with a neuroimmune disease. I suggest that if you rid yourselves of this “contamination theory contagion”, some clarity might be gained.
I also think that if you viewed ME/CFS in the context of the wider history, as detailed in Osler’s Web by H Johnson, some of the pieces of this puzzle might make more sense. Possibly even more sense to yourselves than to those who have lived with the illness and the politics for decades.
We need you to talk of the science, not the scientists. Please do so – lives are at stake.
Someone with a background in the pawn shop business has been flogging meaningless diagnostic tests to seriously ill, vulnerable people. Is anyone ever going to get upset about that? I hope a serious investigative effort to follow the money will be conducted, as opposed to just believing reassurances that it didn’t end up in someone’s pockets. The Science reporting has been excellent, but I think there could have been more clarity regarding where the responsibility lies for the people who paid for bogus tests – is anyone responsible for communicating the new information to them? What about refunds? Is the fact that the WPI/NCI assays give different results on **replicates of the same sample** being sufficiently emphasized?Â
Another nice bit of science to add to the list of references below 🙂
“Staff members and parents rated 29 pediatric AIDS patients (approximate
mean age of 5 years, 6 months) at baseline and again after 6 months of
AZT treatment on 49 behavioral items using a Q sort format. Item
analysis and factor analysis resulted in 5 robust scales. 1) autism, 2)
depression, 3) hyperactivity and attentional problems, 4) irritability,
5) adaptive behavior. The mean inter-rater reliability for these scales
was satisfactory. Patients who were classified clinically as
encephalopathic were rated before treatment as SIGNIFICANTLY MORE AUTISTIC *, depressed and less adaptive than a group of
non-encephalopathic patients. These differences generally disappeared
after 6 months of treatment, leading to a statistically significant
DECREASE IN AUTISTIC and depressed behavior and an increase in adaptive
behavior after 6 months of AZT treatment for the encephalopathic
patients.”
http://gateway.nlm.nih.gov/MeetingAbstracts/ma?f=102177501.html
* my caps
I’m disappointed that you decided to change the figure on this episode. There was no need to give in to a group of people who deny science and harass scientists. I think this is the first group of people since the dawn of molecular biology that somehow find the reason to threaten and harass basic scientists. There are so many who suffer in this world. Unfortunately only those who have access to internet and keyboard have the ability to scream.
If anyone is wondering where the discussion and the red T-shirt image has gone: https://virology.ws/2011/09/29/admit-when-you-are-wrong/#
Could you please respond to the comments below re your podcast Professor Racaniello?
The words of Prof Racaniello:
https://virology.ws/2010/12/22/xmrv-and-cfs-its-not-the-end/
In reference to the papers released on 20th and 21st December 2010: “As a
consequence of this additional research I decided that my initial
impression of the papers was incorrect…”, “I make mistakes…” and “I
had difficulties interpreting these
papers…”.
Perhaps you would wish to reveiw your recent comments after a wee reminder of those made by yourself 9 months ago? I hope so. We need good commentary.
give me a break!
It has been shown so far to exist in patients, produce an immune response, be infectious and you can photograph it. Â No hypothesis exists to explain the immune response other than the human gammaretroviruses. Â
Gammaretroviruses can be hypothesised to produce the same response seen in children with autism. Â
The WPI and others are finding multiple HGRVs in autism as they are in ME/CFS. Â HTLV is a model and virus you should try to learn about. Â Xenotropic is merely the host range of a virus. Â This type is also the largest host range found in MLV viruses. Â
The VP62 plasmid that was infecting Silvermans samples and caused him to incorrectly call the viruses found by the WPI and NCI, XMRV, was never in the WPI or NCI labs and has been conclusively proven by an independent lab to have never infected the samples in their labs. Â The negative papers and the Blood study can be explained by their use of VP62, which has never been found to exist in nature.Â
The WPI and NCI are now known to have had their findings backed by the Lo findings, as they are of the same type. Â Hanson is also finding those viruses. Â Â Â
There is no evidence that would prove any patient has ever sent any threatening letters and individuals would never be consider to represent a sick population. Â The entire notion is silly. Â
Mikovits and Frank Ruscetti have done nothing to warrant your critisim. Â Their findings still stand. Â They discovered HGRVs.
Immune responses alone are a very poor method of identifying novel infectious agents. The combination of an immune response plus an infectious agent provides very good evidence for real infection, but you have to have both.
Infectious agent only = questionable relevance & unknown relation to disease
Immune response only = very prone to false positivity and hard to know what you are looking at
In many immunological assays getting the negatives to be negative is much more difficult than getting the positives to be positive. Positive assays such as Western blots and ELISAs only tell you that one protein binds another – nothing about whether an infectious agent is present. The immune system can be fantastically cross reactive – remember it has evolved to keep us alive not to be a taxonomist – the greater the number of agents it can recognise with the least number of receptors the better.
An immune response to an MLV virus, using an assay that cannot detect human or mouse endogenous retroviruses has narrowed the identification to HGRVs. Â It is not possible to have an immune response to a contaminant. Â
No, but it is certainly possible to have a band on a gel. An immune assay alone can’t identify a novel pathogen, it’s far too non-specific. The question here is whether there is an immune response at all, which given all the other data that show virus is not infecting humans is very unlikely. A positive assay does not equate to an immune response.