You all know MEGA and the U.K. situation much better than I do so please don't hesitate to tell me where I am not understanding something...
As MEGA are now talking about a biorepository,, which sounds a lot like a biobank to me, I'd hope the would consult with the UK mecfs biobank on how to go about this. But the mecfs biobank, like MEGA, draws from NHS clinics (I didnt spot anyone complaining about that, but maybe I missed it) and MEGA will define patients according to multiple criteria. All biobank samples meet Fukuda and most meet Canadian; for MEGA all will meet NICE (I'm not sure which is 'better' between Fukuda and NICE, neither are ideal in my book) and will also diagnose according to other criteria..
So ultimately they will be very similar - as I understand from the Q&A, researchers could pull out a sample of CCC-only patients from both.
From what I have read, the London biobank is reaching out to severe ME patients to collect samples. Is that true and has MEGA implemented a strategy for this?
Re Fukuda versus CCC - totally agree that Fukuda is not good for research. Advocates in the U.S. have been pushing NIH to use a criteria that requires hallmark symptoms as seen in CCC and ME-ICC. One of the concerns raised with the initial plans for the NIH Intramural study was that it was using the Reeves 2005 definition - Reeves uses the Fukuda inclusion and exclusion criteria but then uses a set of assessment tools that resulted in a 6 fold increase in prevalence over Jason's study and also the inclusion of more mental illness. NIH has since said they are using CCC
Worth mentioning that the uk biobank selected on Fukuda, but most of those still met Canadian. I think Lenny Jason has done a similar study finding the same. So I don't think there's any good evidence that using looser criteria (Fukuda or NICE) will exclude patients with a tighter criteria such as Canadian.
I wonder to what extent that depends on how NICE is applied in actual clinical practice by the NHS clinics. By definition, CFS is
medically unexplained chronic fatigue. To what extent do the clinics exclude patients from a diagnosis of NICE CFS who have evidence of e..g. neurological or immunological impairment? It would be important to know how they apply exclusion criteria both at the initial visit and over time. The same is/has been true of Fukuda. The CDC CFS website used to say that a diagnosis by CCC was not the same as a diagnosis of CFS since CCC included neurological symptoms that were considered exclusionary for CFS.
Yes, we need more detail. Eg willl they use Lenny Jason's DePaul symptom questionnaire to diagnose Canadian, as several groups do. The biobank used its own unpublished questionnaire to categorise patients as meeting Canadian criteria (that's not ideal either - but then many of the issues raised about MEGA apply to many other studies).
As for PEM, I think that's a very important question. Jason has done some of the best work on this, but I'm not totally convinced by his questionnaire: his analysis concludes that the single best q for defining PEM is "a dead heavy feeling in the muscles". I wonder how many people here would agree with that.
So I think these are all important issues, and we need to know how MEGA are doing this - but I'm not sure there are lots of perfect solutions out there already.
Completely agree we don't yet have perfect solutions for assessing critical factors in this disease and we need to have these in all these studies. Some thoughts... NIH has announced a
focus group to collect information on PEM which will be used as interview tool for assessing post-exertional malaise in its intramural study. (Patient selection will be key there as well and ideally, the outcome will be compared to the DePaul questionnaire.)
NIH also has an initiative for common data elements that is intended to drive consistency across studies on how symptoms are assessed and what data is collected. They are running a session on this at the IACFS/ME conference - abstract below) Perhaps an opportunity to drive some international consistency.
My one concern with the common data elements approach is that CDC has in the past stated that as long as you collect common data elements, it doesn't matter what definition you use. While I appreciate the value of common data elements, I wonder how well we will be able to harness the future evidence base if every published study continues to use the same disease label for all these different definitions. Assuming I got it right, I think that's one of the things the London biobank is doing right - samples that meet only Fukuda and not CCC are labeled differently than those that meet CCC.
One final thought - MEGA has set a goal of a large number of patients and then is applying a broad definition to achieve those numbers. I've read the NICE criteria but I can't tell whether PEM is mandatory or not and how its assessed. Based on concerns raised here, it sounds like NICE's focus, like Oxford and Fukuda, is still on medically unexplained chronic fatigue. Is that true?
Assuming for the moment that NICE does not require PEM as we understand PEM... If researchers were running a big data study for multiple sclerosis, would they use a catch-all diagnosis of medically unexplained chronic fatigue to boost the recruitment numbers, knowing that some of those patients would not have MS? I'd assume they'd take their best stab at defining an MS cohort, knowing up front that that would dictate how many patients they get and that if they were to add additional groups, they'd carefully decide what groups to add and treat them as comparator groups, not part of the MS group.
So why not start with a research definition that requires the hallmark symptoms of ME? If one of the goals is to compare the underlying biology of ME patients to fatigued patients who do not have ME, what about including some patients with
medically explained fatigue, like MS?
I get the point that Dr. Shepherd made about proving once and for all that the NICE criteria contain unrelated groups that need to be separated. That's a really important political problem in the U.K. It may be necessary, but this seems like an expensive and time-consuming way to solve that problem especially given all the reports and science that have come out in the last 2-3 years. I imagine the value delivered by the study will hinge on how well it selects and characterizes ME patients and how many and how broad a cross section of ME patients it is able to recruit.
IACFS/ME Conference session
Common Data Elements (CDEs) for Standardized Testing and Clinical Studies
Chair: Vicky Whittemore, Ph.D.
Program Director, Channels, Synapses and Circuits
National Institute of Neurological Disorders and Stroke
The National Institute of Neurological Disorders and Stroke (NINDS), National Institutes of Health and the Center for Disease Control and Prevention (CDC) will partner to develop common data elements (CDEs) for standardized testing and common data elements to be recorded in clinical studies/trials of individuals with myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS). The development of CDEs for ME/CFS will facilitate the comparison of results across studies and help to standardize analysis. The session will be led by NINDS and CDC Program Staff to discuss the timeline and process for developing the CDEs and to obtain feedback and input from ME/CFS stakeholders.