Amy Dockser Marcus on Ila Singh's paper and Lipkin study

[Otis]

ooohhh Xombi Apocalypse, I like it, it has a nice ring, and after invalidating my mornings I often feel like I've become one of the Xombies. (girns)

Hey Otis, I can't help it I'm short and gutters are kind of at my eye level. (hhhheeeeehhhehe, grins) Besides I like your er, em, thingies. They remind me of mom, need several more though.

:tear::tear::tear:

Aren't they inside out? Is this like an inside view?

 

[xrayspex]

if thats the inside view than those babies are really some low hanging fruit!

 

[Cort]

Here's my take on the Singh paper - obviously a very well thought out paper

Dr. Singh on Detecting XMRV - the title of Dr. Singh's commentary in the Viruses journal went right to the heart of the matter. The problem right now is a simple one; "Detecting Retroviral Sequences in Chronic Fatigue Syndrome". That's the whole ballgame - that's all that anyone cares about right now - its ALL about detection - this was a timely article and it was nice to see someone tackle this publically. Dr. Weiss's Rumor Virus article gave a good historical overview of the sometimes extraordinary difficulties pathologists can face now given the immensely powerful technology available to them but Dr. Singhs article actually delved into the specific issues surrounding XMRV.

Splitter or a Lumper? Dr. LeGrice and others looked narrowly at the pMLV finding and stated that, until we know more, they should be treated as unrelated finding - which has its own logic. Dr. Singh took a broader view and stated the opposite; both have their own logic. For Dr Singh, considering that the pMLV's and XMRV share 95% sequence similarity sequences, the same receptor and, get this, are actually closer to each other than different variants HIV (85-95%), it appeared it was easy for Dr. her to place the pMLV findings by Alter/Lo firmly within the context of the Science papers XMRV finding. It's all about emphasis.

One Big 'Happy' Family? Is Dr. Singh implying these are all one family of of viruses that interact with each other like HIV does? Dr. Mikovits hinted, if I remember correctly, that more than one type of virus may be needed for them to survive in the body and Dr S. Ruscetti is reportedly looking at how these viruses combine with each other. There is also section of the pMLV's that we know can jump into other viruses...Are the pMLV's and mPMV's and XMRV's all one family????

Contamination - Dr. Singh felt contamination of the Lo/Alter samples was unlikely because the instruments used to search for it were 100 x's more powerful than those used to look for XMRV and the wide discrepancy in prevalence between the healthy controls and CFS patients. She noted, however, that with the exception of the 8 samples from present day patients that samples for most of the controls and patients were handled differently D) evidence of a replicating virus using antibody tests or actually finding the virus are critical E) (The same thing applies for the original Science study)

The 'Perfect' XMRV Study - But how to do a study that helps to clear up the confusion? That, according to Dr. Singh is certainly doable (and not easy) and definitely has not been done. That study needs to (A) collect blood from a large # of healthy controls and patients from the same area using the same techniques (B) the investigators must be blinded (a horrible fate but if it's necessary, it's necessary the Alter/Lo study researchers were not and the WPI study did not report it was in the Science paper) (C) negative controls on XMRV should be checked for contamination prior to amplifying the samples (D)detailing the limits of detection is important E) replicating one element of the positive study using blood from positive patients collected by a phlebotomy lab.

Note that Dr. Singh does not believe that it's necessary to replicate every procedure from the first paper - one is enough. We can guess that Dr. Singh's XMRV study probably )) closely replicates her advice and we can get prepared for a very precise methodological section - probably the most comprehensive yet as she details exactly what she did and didn't do - which is what the National Cancer Institute, in fact, is asking XMRV researchers to do right now. Dr. Singhs autopsy study should be at the publishers now and her XMRV study is believed to be finished as well. Hopefully we will hear from her in another scientific journal soon.

 

[Cort]

This makes her study - which I believe has been finished for some time all the more something to look forward. I was particularly enchanted with her statement that the MLV family differs less in sequence variation that HIV! Who knows what we may learn from these viruses? Are they all intermingling? Do they travel together? Pitch tents together? Work together in humans to create disease? Are we looking a bunch of human never before discovered group of retroviruses?

 

[Cort]

And what does this mean?

The guidance comes at an opportune moment. W. Ian Lipkin, the Columbia University-based virus hunter tasked by the government with resolving some of the XMRV questions, is convening a meeting this week. Representatives from HHS and the academic and advocacy communities are in New York to develop a strategy for a study to determine whether XMRV is found at higher rates in CFS patients.

Does this mean they found the discrepancy and its not contamination and the path is clear for new studies? Or does it mean they've simply determined the right way to find it and they will start a new round of testing? .

First sample preparation et al and then quickly - by the end of the year - the actual assay. Lipkin starts collecting his samples NOW and begins testing them by the end of the year.....things ARE moving.

 

[xrayspex]

Here is some summarizing from what I have read recently regarding Singh's latest info and WSJ article and interpretations of Lipkin's style. My *concern* is that Lipkin will not want to follow Singh's advice, ironically this guy DHHS gets to do this big xmrv study, Lipkin, also happens to have the personal philosophy of not wanting to do things way Mikovitz and Singh advising:

In his recent paper "Microbe Hunting", he gives his argument against resolving research conflicts with the use of positive samples from labs that have 'found something'.
But Singh says researchers should do just that: "Finally, an ideal study would use at least one method that has resulted in detection of XMRV in a previous study, preferably using a set of samples that were analyzed in that same previous study and are ‘known’ positives or negatives."

Its critical that the Lipkin study's design be such to include positive samples from the WPI and from Lo/Alter, and that its test samples be collected by well-known ME/CFS clinicians such as:Cheney, Klimas,Bell, Peterson, Komaroff etc.. Right now I don't think it is.

Perhaps we should write Collins and Sebelius about this, Mangan....others aware of this and have thoughts?

 

[akrasia]

xrayspex,

I don't think she published this advice believing that Lipkin would follow it. It sounds more to me like she wanted it out there so that if his project fails to find XMRV/MLV in PWME, the methodological argument will go on, depriving him of the final word on the best approach to the problem

Assuming that Singh's studies are coming up with some evidence of infection in the blood of people with cfs, she will have a high investment in proving Lipkin mistaken. I bet she and Lipkin are "friends", given the Columbia University connection; she could just as easily have dropped him a note explaining her misgivings. Instead, she laid it out quite publicly. As someone noted, this very well might be in preparation, an apologia, for her own impending publications on the subject.

Here is some summarizing from what I have read recently regarding Singh's latest info and WSJ article and interpretations of Lipkin's style. My *concern* is that Lipkin will not want to follow Singh's advice, ironically this guy DHHS gets to do this big xmrv study, Lipkin, also happens to have the personal philosophy of not wanting to do things way Mikovitz and Singh advising:

In his recent paper "Microbe Hunting", he gives his argument against resolving research conflicts with the use of positive samples from labs that have 'found something'.
But Singh says researchers should do just that: "Finally, an ideal study would use at least one method that has resulted in detection of XMRV in a previous study, preferably using a set of samples that were analyzed in that same previous study and are ‘known’ positives or negatives."

Its critical that the Lipkin study's design be such to include positive samples from the WPI and from Lo/Alter, and that its test samples be collected by well-known ME/CFS clinicians such as:Cheney, Klimas,Bell, Peterson, Komaroff etc.. Right now I don't think it is.

Perhaps we should write Collins and Sebelius about this, Mangan....others aware of this and have thoughts?

 

[xrayspex]

thx for info akrasia, I hadn't seen it addressed at PR yet, is there a link to a thread of what you said or perhaps I just missed it.....more scattered than usual lately

 

[WillowJ]

interesting... I didn't know about Lipkin's article "Microbe hunting." That would be cool if Singh is working on publishing an ME/CFS link, too. I loved that she included the Canadian criteria. And I was really interested in the part about comparing the similarity to that of the variations in HIV-1.

Why are we calling viruses "microbes", though? They aren't alive. I feel so sad to be at home instead of working on science. I'd much rather be helping figure this stuff out than waiting for others to do it!

 

[Otis]

Here is some summarizing from what I have read recently regarding Singh's latest info and WSJ article and interpretations of Lipkin's style. My *concern* is that Lipkin will not want to follow Singh's advice, ironically this guy DHHS gets to do this big xmrv study, Lipkin, also happens to have the personal philosophy of not wanting to do things way Mikovitz and Singh advising:

In his recent paper "Microbe Hunting", he gives his argument against resolving research conflicts with the use of positive samples from labs that have 'found something'.
But Singh says researchers should do just that: "Finally, an ideal study would use at least one method that has resulted in detection of XMRV in a previous study, preferably using a set of samples that were analyzed in that same previous study and are ‘known’ positives or negatives."

Its critical that the Lipkin study's design be such to include positive samples from the WPI and from Lo/Alter, and that its test samples be collected by well-known ME/CFS clinicians such as:Cheney, Klimas,Bell, Peterson, Komaroff etc.. Right now I don't think it is.

Perhaps we should write Collins and Sebelius about this, Mangan....others aware of this and have thoughts?

I think all we know about Lipkin's study approach, outside of this rather objectionable opinion about not using "validated" positive controls comes from the following paragraph from Amy Dockser Marcus' blog 'World Class Virus Hunter’ To Head Up the Latest XMRV Study.

The new study, Lipkin says, will involve fresh blood samples from 100 CFS patients and 100 similar, but healthy people — 25 of each group from four different sites around the country, to provide geographic diversity. The samples will be processed, blinded and sent to the FDA, the CDC and the Whittemore Peterson Institute, which led the team that published the original Science paper. If a lab finds a sample is positive for XMRV, further tests will be needed to confirm the result. If one lab finds a positive sample but another lab doesn’t, the same samples can be shipped again, with a new blinded code, to be tested again. “If you get the same result, it is valid,” Lipkin says.

This leaves lots of questions including patient selection. I agree the source of positive controls is of concern and I'm all for reminding him of that fact - I just prefer to have more info before implying a researcher is doing it wrong in a letter that I would write but with the current lack of information I'll probably be forced to reconsider and take it at his word (from Microbe Hunting).

On the plus side, as was noted earlier in the thread there were some big names (I can vouch for one that Cort didn't guess) in NY last week and if Lipkin follows his own "inner voice" over that of many smart people in the field, including Dr. Singh's excellent commentary we're probably just hosed anyway.

That said there's a LOT riding on Lipkin's study and I don't want to doomed to failure before it starts. We're living in an information vacuum which makes me feel like a mushroom.

 

[xrayspex]

Otis, I appreciate your measured approach and it wouldnt be scientific to jump to too many conclusions, yet, perhaps? but if anyone can find out more that would be great. If this is a concern, I don't know if writing lipkin is genius on our part, but to call attention to it in some measure with DHHS and press perhaps?

 

[Otis]

Thanks XRS.

I'm not adverse to reminding Lipkin of Singh's positive control recommendation as it's contradicted by his recent work in Microbe Hunting which would seem to be a caution, I may very do that myself. I (speaking for myself) just want avoid shouting him down for taking the wrong approach when I just don't know that.

I would love someone like Mindy Kitei to dig into this and give us some better details.

 

[Sasha]

I would love someone like Mindy Kitei to dig into this and give us some better details.

Would you consider contacting her and asking her to do this?

 

[Otis]

Well that seems perfectly practical and something I can do. You bet Sasha, great idea.

 

[akrasia]

I don't think "pressure" brought by anyone would be useful here, simply because Lipkin's role has been described as being above the fray. If Singh really thought he would change the protocol, she could have mentioned it backchannel and not as publicly as she has. One thing it does serve in my view is to allow sympathetic colleagues to add their voices to her disquiet. We'll see if it happens.

xray, I was alluding to Cort's statement, where he asserts that he believes publication of both Singh me/cfs study and autopsy study is imminent and draws the inference that a comprehensive description and concern for methodology is a central concern of hers.

With this paper, Singh is positioning herself so she can mount a powerful critique of Lipkin's results, if need be.

Thanks XRS.

I'm not adverse to reminding Lipkin of Singh's positive control recommendation as it's contradicted by his recent work in Microbe Hunting which would seem to be a caution, I may very do that myself. I (speaking for myself) just want avoid shouting him down for taking the wrong approach when I just don't know that.

I would love someone like Mindy Kitei to dig into this and give us some better details.

 

[Otis]

I don't think "pressure" brought by anyone would be useful here, simply because Lipkin's role has been described as being above the fray. If Singh really thought he would change the protocol, she could have mentioned it backchannel and not as publicly as she has. One thing it does serve in my view is to allow sympathetic colleagues to add their voices to her disquiet. We'll see if it happens.

xray, I was alluding to Cort's statement, where he asserts that he believes publication of both Singh me/cfs study and autopsy study is imminent and draws the inference that a comprehensive description and concern for methodology is a central concern of hers.

With this paper, Singh is positioning herself so she can mount a powerful critique of Lipkin's results, if need be.

That's an interesting perspective. If that's the case she must have some info telling her this is going to go off track. We have so little info on the study design it's hard to assess.

 

[Sasha]

I don't think she published this advice believing that Lipkin would follow it. It sounds more to me like she wanted it out there so that if his project fails to find XMRV/MLV in PWME, the methodological argument will go on, depriving him of the final word on the best approach to the problem.

Having re-read Amy's piece, I don't see any evidence that Dr Singh's paper is aimed specifically at Lipkin rather than being for the purpose of sorting the field-wide confusion about how to detect XMRV. I agree that the former would be one interpretation but I think her paper would have been timely and valuable at any time since the 0/0 studies started coming out.

I'm glad she has sent it to Lipkin, though!

 

[Cort]

I don't know why Lipkin wouldn't just check out positive samples?

On the other hand I think he will be using the protocols and assays developed by the Blood Working Group - right now he has the sample collection protocol and by the end of the year, or thereabouts, he will have the assays and he will start testing them. Those assays are being developed using the WPI's samples. I don't think they have positive samples from anywhere else - so, if I have this right, he will be using tests whose accuracy was checked using positive samples.

If that's the scenario it does make me wonder where HE comes in. He is known for his technical expertise in developing new and cutting pathogen detection techniques. Is he going to use those is he just going to rely on the BWG's findings? I would feel better if he was doing both - using their protocols and checking using his own stuff.

 

[Cort]

Having re-read Amy's piece, I don't see any evidence that Dr Singh's paper is aimed specifically at Lipkin rather than being for the purpose of sorting the field-wide confusion about how to detect XMRV. I agree that the former would be one interpretation but I think her paper would have been timely and valuable at any time since the 0/0 studies started coming out.

I'm glad she has sent it to Lipkin, though!

I think it was such a timely paper - and refreshingly rigorous after the Weiss's kind of historical 'essay' on rumor virus's.

 

[urbantravels]

I would guess it's more of a response to the zero/zero studies, especially the more quick and dirty of the lot, than any kind of message aimed at Lipkin.