NIH intramural research program update

[Forbin]

Below is a link to the February 22, 2011 NIH videocast "Demystifying Medicine - Chronic Fatigue Syndrome: Is there a virus?"

Dr. Gill interviews a patient at about 20:00 and then starts his main presentation at 34:40.

http://videocast.nih.gov/Summary.asp?File=16477&bhcp=1

 

[Scarecrow]

phase 1 is a broad investigation of biological abnormalities. Phase 2 is a validation of potential biomarkers found in phase 1.

With only 40 patients in the initial sample (some of whom, because of short illness duration, may turn out not to have ME/CFS), could something crucially important be missed? Can anyone put an estimate on this?

 

[viggster]

With only 40 patients in the initial sample (some of whom, because of short illness duration, may turn out not to have ME/CFS), could something crucially important be missed? Can anyone put an estimate on this?

Six months is not long enough to determine if someone has ME?

 

[halcyon]

With only 40 patients in the initial sample (some of whom, because of short illness duration, may turn out not to have ME/CFS), could something crucially important be missed? Can anyone put an estimate on this?

They could lose the bogus control groups and recruit 40 long term ME patients in their place.

 

[Scarecrow]

Six months is not long enough to determine if someone has ME?

Most cases of post infectious fatigue would be expected to resolve within six months but not all. I have a close friend who had glandular fever / infectious mononucleosis in his late 40s and it was close to twelve months before he started to improve.

I think that the short illness duration is one of the weaknesses of this study but I was reassured when we learned it was longitudinal because it should become clear if there are any patients who just took longer than usual to get over their post infectious fatigue. I would not consider these patients to have ME/CFS but I don't object to a clinical diagnosis after six months (or even earlier).

 

[Scarecrow]

@viggster If you have a look at page 17 of this PDF, you'll see what I'm getting at. It's the page headed:

Post-infective fatigue (PIFS) outcomes

https://www.rcpa.edu.au/getattachme...w-Lloyd-Chronic-Fatigue-Syndrome-gone,-f.aspx

 

[viggster]

Most cases of post infectious fatigue would be expected to resolve within six months but not all. I have a close friend who had glandular fever / infectious mononucleosis in his late 40s and it was close to twelve months before he started to improve.

I think that the short illness duration is one of the weaknesses of this study but I was reassured when we learned it was longitudinal because it should become clear if there are any patients who just took longer than usual to get over their post infectious fatigue. I would not consider these patients to have ME/CFS but I don't object to a clinical diagnosis after six months (or even earlier).

Right, that makes sense. Also, we (or I) know ME patients who go into remission or recover or whatever you want to call it after 2, 3, 5, 10 years. And there's no way to predict who those people will be. Depending on how long the NIH follows patients, this study might give us some info on how to make better prognoses early in the illness.

 

[voner]

The animal model stuff sounds suspiciously like animalmodellobabble to me, although very rarely animal models of this sort do actually show something surprising - like the B27 rats for ank spond. So maybe worthwhile.

I worry that nothing is said about getting patients from an unselected primary population based cohort. If this is not done the recruits could easily be people who are unrepresentative - especially if derived from a clinic or the internet.

The cohort number also looks small - 40? I would start with 200 or 500. Everybody who has tried to get biomarkers with ME has struggled with sizes like 40.

@Jonathan Edwards raises 3 very fundamental problems with this proposed study.

1. animal models. I have followed the well funded "pain" research for many years. Using animal models they have been analyzing cytokines, receptor pathways, etc. and still have not developed any new pharmaceuticals or methodologies to reduce pain partially because pain is very complex, as I suspect is "fatigue" (i dislike the term). I cannot see how they can use a small pre-selected group of 40 patients to create a animal model.

2. The patients will be funneled into the study by experienced ME/CFS clinicians, not from an unselected primary population based cohort. this can certainly add a bias to the population being studied. of course, I suspect that the reason that the NIH study is proposing to use their methodology of funneling is because the disease is so poorly or loosely described that they're afraid that only very experienced conditions can accurately identify a patient.

3. The patient sample size is way too small. If they're finally going to put some money into researching this disease, they should put money into getting a appropriate sized patient sample that has been selected correctly. Otherwise the study will be like so many other studies "- garbage in, garbage out".

I want to point out one other sampling bias they have in the study. For some reason, they are selecting a patient population who has a specific infectious onset and eliminating the patient population that cannot point to a specific infectious onset. further by biasing the patient population they're going to use.

 

[viggster]

@Jonathan Edwards raises 3 very fundamental problems with this proposed study.

1. animal models. I have followed the well funded "pain" research for many years. Using animal models they have been analyzing cytokines, receptor pathways, etc. and still have not developed any new pharmaceuticals or methodologies to reduce pain partially because pain is very complex, as I suspect is "fatigue" (i dislike the term). I cannot see how they can use a small pre-selected group of 40 patients to create a animal model.

2. The patients will be funneled into the study by experienced ME/CFS clinicians, not from an unselected primary population based cohort. this can certainly add a bias to the population being studied. of course, I suspect that the reason that the NIH study is proposing to use their methodology of funneling is because the disease is so poorly or loosely described that they're afraid that only very experienced conditions can accurately identify a patient.

3. The patient sample size is way too small. If they're finally going to put some money into researching this disease, they should put money into getting a appropriate sized patient sample that has been selected correctly. Otherwise the study will be like so many other studies "- garbage in, garbage out".

I want to point out one other sampling bias they have in the study. For some reason, they are selecting a patient population who has a specific infectious onset and eliminating the patient population that cannot point to a specific infectious onset. further by biasing the patient population they're going to use.

I wouldn't characterize the mouse model as a 'problem.' It's one small part of a very broad + deep investigation into biological abnormalities. If the mice turn out to be useless, it's not really much of a loss. And like Edwards says, mouse models helped with a very difficult-to-treat autoimmune condition. Mouse models have also been useful in MS.

The NIH is relying on expert clinicians in the US to refer patients to the trial. I think this is very wise for a few reasons. First, no one at the NIH Clinical Center has experience diagnosing ME. And it can be tricky to diagnose. So leveraging the expertise of those who have made many many diagnoses is smart. Two, by doing so, NIH is avoiding the outrage and suspicion in the patient community that would have occurred if NIH clinicians had done all of the choosing themselves. And three, selecting patients from the entire US population via a laborious screening process would have added a lot of time, expense and hassle to this study. The point of phase 1 of the NIH project is to do a deep investigation of a few patients who definitely have ME. And we already have a lot of robustly diagnosed ME patients at the specialty clinics in the US.

As for choosing patients who develop ME after an acute infection, I think this is also smart to try to reduce the heterogeneity of the group. I know it does not cover everyone who develops ME, but it at least takes one variable off the table.

40 patients is a good start for this kind of study. Ron Davis is spending $1 million on 20 patients & the NIH will be doing even more testing per patient. More is always better, and there may be a chance to add patients to the protocol later. NIH can be flexible when they're doing these types of investigations at the Clinical Center. Phase 2 (validating biomarkers) and 3 (testing treatments) of the NIH project will require more patients.

 

[Kati]

I wouldn't characterize the mouse model as a 'problem.' It's one small part of a very broad + deep investigation into biological abnormalities. If the mice turn out to be useless, it's not really much of a loss. And like Edwards says, mouse models helped with a very difficult-to-treat autoimmune condition. Mouse models have also been useful in MS.

The NIH is relying on expert clinicians in the US to refer patients to the trial. I think this is very wise for a few reasons. First, no one at the NIH Clinical Center has experience diagnosing ME. And it can be tricky to diagnose. So leveraging the expertise of those who have made many many diagnoses is smart. Two, by doing so, NIH is avoiding the outrage and suspicion in the patient community that would have occurred if NIH clinicians had done all of the choosing themselves. And three, selecting patients from the entire US population via a laborious screening process would have added a lot of time, expense and hassle to this study. The point of phase 1 of the NIH project is to do a deep investigation of a few patients who definitely have ME. And we already have a lot of robustly diagnosed ME patients at the specialty clinics in the US.

As for choosing patients who develop ME after an acute infection, I think this is also smart to try to reduce the heterogeneity of the group. I know it does not cover everyone who develops ME, but it at least takes one variable off the table.

40 patients is a good start for this kind of study. Ron Davis is spending $1 million on 20 patients & the NIH will be doing even more testing per patient. More is always better, and there may be a chance to add patients to the protocol later. NIH can be flexible when they're doing these types of investigations at the Clinical Center. Phase 2 (validating biomarkers) and 3 (testing treatments) of the NIH project will require more patients.

To hoping.

 

[Snow Leopard]

And like Edwards says, mouse models helped with a very difficult-to-treat autoimmune condition. Mouse models have also been useful in MS.

Mouse models can be useful, but often aren't. The problem is where the human disease differs too much from the mouse model - which I'd have to say is the case in most animal model papers I have read and this includes MS.

I also have ethical issues about animal experimentation - avoid unless it is at least 99% representative of human disease and there is no other way to gain the information.

 

[Ecoclimber]

No single animal model can capture the entire spectrum of heterogeneity within this human ME/CFS disease in clinical research. A model can be developed that represent selected aspects of the human disease but there needs to be an understanding on their limitation within research. EAE models do have their limitations within MS research.

 

[Never Give Up]

Dr. Bateman just mentioned on Facebook that the number of grand rounds views matters, so, even though we now have the slides and audio posted here, it might help our cause to actually view the presentation. It should be up tomorrow.

 

[Ecoclimber]

Someone at the CDC chose the speakers today. It was a CDC event.

NIH is doing protein analysis of CSF, so I'm guessing they've seen this study.

Not necessaily. Researchers in one specialized area or field of work would not be focusing their attention on research in another field of work like ME/CFS which does not apply to their field of speciality. We (Miller, I & his team) believe that Natelson research work was robust, enough to run a mass sepctromety using a VeriChip against xmrv at Pacific Labs. The results were negative for xmrv.

 

[Jonathan Edwards]

If you look at Nath's slides you'll see the 40-patient phase 1 is a broad investigation of biological abnormalities. Phase 2 is a validation of potential biomarkers found in phase 1. Phase 3 is treatment. There will be ever larger numbers of patients as this research program advances. The 40 patient phase 1 is just the start.

The problem is, as Scarecrow, says, that it is very easy to miss something important with a 40 patient cohort - just for statistical reasons, considering that it is very likely that the important thing may only apply to 30% or so of your cases (maybe even only 10%) with a heterogeneous pathophysiology.

I would use a 40 patient cohort for a look see study costing $100,000 targeting one particular biomarker - where you have no Bonferoni problems, which makes it easier to avoid type 2 errors. For a high cost study looking at lots of variables I find it hard to see how you are going to get reliable stats out of 40. And the key thing is that you may well miss the crucial markers, so going on to study what showed up on bigger populations takes you further and further down a blind alley.

I think concentrating on post-infective cases is a very good idea, because it reduces heterogeneity and post-infective ME/CFS is accepted as at least part of the population of significant disease. But I would have thought it would be better to do a Dubbo type study prospectively looking for cases in a defined population. Trawling from clinicians is almost certainly going to bias the type of patient. That may work out well but it may work out badly and you are unlikely to know why if it does. I have personal experience of doing trials using patients sent to me by other clinicians and my general feeling is you just do not want to go there. Clinicians refer patients to trials for reasons that have pretty little to do with them being representative cases. Usually it is because they are very unrepresentative and the clinician cannot think what to do with them.

Strangely, the NIH is actually funding a population based cohort (I think 200) already and this does not seem to appear in the study.

 

[viggster]

Strangely, the NIH is actually funding a population based cohort (I think 200) already and this does not seem to appear in the study.

Are you referring to the CDC 5-site study now in its fourth or fifth year? NIH is working with those sites to recruit for the new study.

I'm not sure what other 200-person cohort you could be referring to.

 

[Sasha]

Most cases of post infectious fatigue would be expected to resolve within six months but not all. I have a close friend who had glandular fever / infectious mononucleosis in his late 40s and it was close to twelve months before he started to improve.

I think that the short illness duration is one of the weaknesses of this study but I was reassured when we learned it was longitudinal because it should become clear if there are any patients who just took longer than usual to get over their post infectious fatigue. I would not consider these patients to have ME/CFS but I don't object to a clinicaldiagnosis after six months (or even earlier).

The small sample size worries me in relation to this question. I don't know if PWME have ever been compared to PWPIF (post-infectious fatigue) in their first few months and how distinguishable they are symptomatically (it's so long since I've been normal that I can't remember what normal recovery from an illness looks like). But it PIF is a similar syndrome to ME and if it's also rare that PWPIF go on to be PWME, then perhaps this tiny sample of 40 will be cluttered up with PIF for no good reason. The sample is too small to do lots of comparisons between subgroups.

 

[Scarecrow]

The small sample size worries me in relation to this question. I don't know if PWME have ever been compared to PWPIF (post-infectious fatigue) in their first few months and how distinguishable they are symptomatically (it's so long since I've been normal that I can't remember what normal recovery from an illness looks like). But it PIF is a similar syndrome to ME and if it's also rare that PWPIF go on to be PWME, then perhaps this tiny sample of 40 will be cluttered up with PIF for no good reason. The sample is too small to do lots of comparisons between subgroups.

Exactly, Sasha. PIF is most definitely very similar to (if not indistinguishable from) ME. The Dubbo study tracked people who came down with EBV, Ross River virus and Q fever and it was this study on which the presentation linked to in this post was largely based on.

If the Dubbo figures are representative, then roughly 50% of people who are ill at six months will have recovered by twelve months (but the chance of recovery declines with duration so that someone who is ill at nine months is less likely to recover than someone who was ill at six months. I think the chance of recovery falls to about 10% at twelve months but don't quote me on that).

I suppose the key question is, does PIF = ME. My gut says no...........but my gut isn't a scientific instrument! If they are the same, then perhaps the only issue is that two or three people may recover shortly after the study begins depending on the distribution of illness duration in the sample. What meaningful information will that give us?

 

[aimossy]

They may be able to see what changed in these people longitudinally re recovery? This could also be useful especially of your taking a real in depth look at them? I mean this comes down to something with respect to study design and why these questions are good questions and useful to bring up.

 

[duncan]

Longitudinal studies require patients remain in them.

This study better pull in the 40 from locals, or they may lose participants due to attrition. I think patients will join willingly, but the rigors may levy a toll.

I think losing even locals over time may prove a worrisome factor.