How near are we to a diagnostic biomarker? What is it likely to be?

You are free to hit me on the head with a foam bat if things don’t pan out.

 

My vote is it will be a pattern or collection of tests. If it was one gene I suspect it would have been found already. So interaction of several SNPs in slightly different possible configurations, microbiome testing, autoantibodies/immunosignature forming a bigger picture?

The exception is maybe the substance in the blood. That could be one simple blood test.

Researchers seem to be getting there with evidencing biological abnormalities, but everyone who has suspected ME will be feeling ill. Will something like 82% accuracy distinguishing ME from controls be any use in a doctor’s appointment? I’m suspicious of studies with such high results because isn’t it more likely to be picking up general illness? With the diagnostic mess we’re in I’ll eat my felt hat if 82% currently diagnosed pwme all have the same biomarker and it’s specific to ME (though machine learning of general patterns is very plausible).

BTW I’m thinking of https://www.ncbi.nlm.nih.gov/pubmed/27338587

“Using a machine learning approach trained on the data obtained from 16S rRNA and inflammatory markers, individuals were classified correctly as ME/CFS with a cross-validation accuracy of 82.93 %.”

 

I remember Nancy Klimas talking about 3 genes a few years ago, but that must have not been confirmed as we didn’t hear anything more.

 

As I understand it, lots of fullly recognised diseases don't have a single biomarker either. They have a clinical and a serological profile and you can have "classic" cases that tick all the boxes, and others that don't. Some diseases, like Crohn's can have a variable picture in both symptoms and blood tests.

What is more important than a 100% accurate biomarker is to find a biomarker in enough cases that we can use it to start building up a disease model. Once we have a disease model, its possible to diagnose even when the primary biomarker is absent - if all other features point to the disease.

 

I have no confidence in the genetic test option. Hell, they can even find a genome associated with depression now. Doesn't get you any closer to knowing what the more proximal causal factors are - the ones you'd need to address to treat the depression. And it certainly doesn't show the disease is "biological", since the genome could impact on personality factors too.

 

She's primarily using 23andMe data, so it's unlikely to find anything.

 

@Woolie
I am sorry you have little confidence, and can sympathise.

——-
Generally speaking.....(not you Woolie)
IMO some PwME have been lured into some kind of lull to almost have a betting pool mentality with a small handful of researchers/doctors. None of them are gurus, and thinking of them in that way, I feel is unhealthy for both ME research and PwME.

I know we want hope, and I fully understand the benefits. I am the point today of wondering what’s true and what’s not. So I almost think it better not to advertise/promote specific researchers names along with their promises. For me, it’s been too many ups and downs; a string of promises followed by a series of let downs.

 

Agreed. It should be about the research being produced, not celebrity status.

 

My memory might be letting me down but I think this was a few years ago, before the current study. Something about people with just one gene mutation of the three tended to be milder but in combination of more than one relevant mutations they had more severe ME.

 

Thanks @Ron for keeping us in the loop and providing the details.

 

The rare autoimmune disease doctors suspect I may have (relapsing polychondritis aka: RP) and that I am currently being evaluated for has no biomarker. In RP patients with severe inflammation and ongoing cartilage destruction, inflammatory markers are often completely within normal range.

 

Would you be willing to explain what the picture shows?

In the article Dr. Lombardi is mentioned. Is this the same Lombardi from the XMRV topic?

 

What about epigenetics and/or mutations (e.g. as in mastocytosis)?

Many rare illnesses are very hard to diagnose. In many cases genetic testing is very helpful or it ascertains a suspicion. In MS, there is no biomarker in itself, and often it is diagnosed in a later stage (although maybe it could have been seen earlier; e.g. some radiologists see in PET scans Alzheimer's disease in a very early stage, 10 years before any symptoms occurr).

What I want to say: The wish for a simple (and cheap) diagnostic tool for ME is understandable, but maybe not realistic.

 

There is an ongoing study (online only) by Nancy Klimas about genes, and you can upload 23andme-data (and another company was mentioned) only. There is a thread somewhere on s4me but don't know anymore where.

 

"Epigenetics" is just one way of saying that an external factor has more impact than the DNA itself does. Hence a standard genetic test wouldn't show epigenetic changes. The problem with novel mutations (not inherited) is that they can't really proliferate enough after childhood to cause problems, unless the mutation results in uncontrolled growth, in which case we'd be diagnosed with cancer instead.

Heteroplasmic mitochondrial mutations are a possibility, and I don't think there have been good studies involving muscle biopsy in ME patients yet, which is what is generally needed to detect such mutations. Though it sounds like the NIH on-campus study might be adding that.

 

These sorts of results are usually pretty meaningless, until they've been replicated. Generally a lot of comparisons are being made between patients and controls, and then even more are being made based on severity level. False positive results are guaranteed in such a situation, unless statistical corrections are made for making multiple comparisons.

Most SNP studies in ME/CFS have also been done with Fukuda, and the results usually contradict what can be seen in 23andMe results from ME patients on the forums.

 

Yes, this is the same Dr. Lombardi around during XMRV. My take away from what I posted is that a group of researchers at Arizona State University are developing an immune signature test and that it's still a work in progress, but that progress is being made towards a biomarker. I specifically like that this group has a spin off company, Health Tell that already provides these types of tests for other diseases. To me the graph shows that ME/CFS patients for the most part have a consistent immune signature. I mention any of this for encouragement purposes only that progress is being made.

http://www.healthtell.com/

 

Maybe I misunderstand you here.

As I understood epigenetics is, amongst others (mutations would be the other), something like a "switch".
The genome is like a big database with lots of "recipes". The cell reads parts of genes in order to produce proteins/amino acids. Epigenetics would be an external factor influencing which part(s) of the genome are read out (by a cell) or not. So, epigenetics is, amongst others, about which external factor will lead to which gene expression and in which amount.

"Epigenetic aberrations are also important for the development of immunological and neuronal diseases..."
http://epigenetics.uni-saarland.de/de/home/

I am certain there are aspects we don't know yet.

I don't know when the first cases of ME did appear - is it 200 years ago, 100 years...? Our environment changes constantly, and I'd expect changes in livings' DNA due to that - new illnesses, new skills, new looks and more.

Mastocytosis is a counter example, although I agree that in general mutations would lead to a (swift) destruction of the organism (death).

This is very interesting!
I had thought in that direction and think about getting active. At first I decided against it because an ME expert told me most patients report no findings. But I am not sure how thoroughly it was looked (many patients are categorized as "psycho" which leads to no or bad diagnostic, so...).