• Phoenix Rising needs funds to operate: please consider donating to support PR

Lipkin and Hornig go hunting for ME/CFS pathogens

View the Post on the Blog

by Simon McGrath


For me, the star attraction of Nancy Klimas' recent CFS/GWI conference was always going to be Professor Mady Hornig and her talk.

Hornig might not be well known by ME/CFS patients - yet - but her boss is: Ian Lipkin, who so skillfully handled the XMRV 'dediscovery' study (which she worked on too). Despite disproving a link with XMRV, Professor Lipkin made clear his belief that ME/CFS was a serious disease that had not received the serious attention it deserved. Even more important - given his stellar record as a scientist - was his commitment to playing a serious role in trying to solve the illness. He's asked Mady Hornig to lead their hunt for the cause of ME/CFS, so I was hoping for an update on how exactly they were going about the work and when results were likely.

As well as that, we had a presentation of science that made my eyes pop, as Mady Hornig took us through some of her incredible work in other illnesses where brain and immune system take centre stage too. If she brings these kind of approaches to ME/CFS then I think we will benefit enormously - more on this next week.

First, though, to the world's largest ever ME/CFS biomedical study: the hunt by Hornig, Lipkin and colleagues for a virus or other pathogen that may cause our illness.


Huge initial studies look for pathogens or tell-tale signs of infection

Work has already started with around 400 patients plus controls, including 200 patients + controls from the CFI pathogen study (Professor Hornig is Principal Investigator for CFI's Pathogen Discovery and Pathogenesis). If that's not enough, they are hoping to get funding for up to a further 400 patients and controls. Assuming my maths are right, that makes a mind-boggling maximum of 1,600 subjects: 800 patients and 800 controls. This is a big step up for ME/CFS research.

The major part of the study is looking for pathogens - viruses, bacteria or protozoans both known and unknown. But they are also looking for protein/immune abnormalites in patients, and they even plan to take a look at gene expression too to see if that throws up any clues.

The pathogen hunt has three steps:
  1. Screen for a panel of 18 specific pathogens already implicated in ME/CFS, such as EBV.
  2. If no such pathogens are detected, they move to the heavy-duty phase, basically sequencing all DNA/RNA in the blood, which should identify both known and unknown viruses. This technique has been successfully used by Ian Lipkin in the past to discover new viruses.
  3. To be thorough (and these people are nothing if not thorough) any 'finds' from the first two steps will be confirmed by a smaller-scale but more accurate technique.


Finally, once they have confirmed candidate pathogens - assuming they find ones that have a statistically significant link with ME/CFS - they will develop new tests specifically for these pathogens.

As a bonus, the heavy-duty sequencing techniques in stage 2 makes it fairly easily to look at gene expression too. Previous attempts to identify unusual patterns of gene expression in ME/CFS patients have found differences, but these have not stood up to the test of replication. Hopefully the much larger cohorts used here will produce more reliable findings, and show which genes go awry with with ME/CFS.


Protein signature for ME/CFS?

The team will also try to define 'host profiles', looking for a unique protein signature associated with the illness. If the study does find a robust CFS signature, it could be used for diagnosis - and a validated diagnostic test is almost the Holy Grail of ME/CFS research. The signature could also be used to measure treatment progress.

Yet more high-tech approaches

They are using a fancy technique called 'multiplexed immunoassay' to look at over 50 specific proteins that are markers of immune/inflammatory changes, or oxidative stress. Again, the idea is to home in on plausible candidates for disruption in ME/CFS. For a smaller sub-sample, they will look more widely for any abnormality in protein profiles using "Target proteomics Mass Spectroscopy" - Nature's "2012 Method of the Year", no less. This approach could throw up proteins no one has yet considered playing a role in ME/CFS, giving new clues to the causes of the illness.

Another good reason to look for protein signatures is to detect general signs of infection. One possible scenario raised by Hornig is that lots of different bugs can trigger ME/CFS, and if that's the case then each individual pathogen might not show up as statistically significant. (This is to do with statistical power, but you really don't want to know). Finding the fingerprints of infection or inflammation instead might prove a general link between infection and ME/CFS, even if no individual pathogens can be pinned down for sure.


First results due later this year

The first result for these studies will be submitted to journals in the next few months, with more papers submitted later in the year. By the time the year is out we should have a fairly good idea if there is a strong link between pathogens and ME/CFS.

Professor Hornig concluded by saying that they may not find an infectious agent, it may be an immune change. It could even be something else. But if you look at the scale and cutting-edge nature of their work then, even if it does turn out not to be a pathogen, these are very good people to have on the case.


Aiming to look for genetic factors


Heading upstream to the source

In the Q&A section after her talk, someone asked if Mady Hornig and her team also plan to look for any genes that are associated with ME/CFS. It turns out they don't have enough samples to look at the whole genome, so instead they would like to use a targeted approach that Hornig calls 'swimming upstream'.

This involves identifying unusual patterns in proteins and/or gene expression using results from this study, then working backwards to identify which gene could be involved in producing these protein changes. It's an approach the group has used successfully before, and requires a smaller sample size for robust results.

A wider look at the whole genome might be also possible at a later date using meta-analysis. This would combine their data with data from other studies - giving sufficient 'statistical power' for robust results.

The studies I've covered above are hugely impressive, but what grabbed me even more was the rest of Mady Hornig's presentation, where she discusses some of the extraordinary work she's done on other hard-to-understand illnesses. She's very keen on bringing similarly innovative approaches to ME/CFS (assuming the current work doesn't solve the illness!) - as I describe in my next article, Mady Hornig: How do you solve a problem like CFS?.




View the Post on the Blog
 

Comments

Thanks Simon,

It's great to hear these details! Very well summed up. So pleasing that every time that I hear something more about the pathogens study it is something good. It's really great that they are looking at gene expression too and immune abnormalities too.

I like everything about this study; its the kind of thing we couldn't have dreamed of a few years ago. Can't wait for it to finish. 2013 really could be our year!
 
Thanks for this well-written, informative article. It's refreshing to read about a study that wll actually make a substantial contribution to the knowledge base, instead of the usual junk research whose main endpoint is to avoid the business end of 'publish or perish'.
 
Fantastic, Simon - I've been dying to know what kind of progress they've been making on this! Thanks for such a good summary.

I'm pleased that they're expecting to start to submit papers in the next few months.

You're right, we couldn't have expected anything like this a few years ago - XMRV, even though it was a dead end, bought us this opportunity.
 
Thanks snowathlete, jimmells and Sasha - glad you liked it!

The presentation iteself was quite complex but Mady Hornig was enormously helpful answering my many questions so hope it's all correct. And I really hope they can get funding for all the cohorts they would like to study
...I like everything about this study; its the kind of thing we couldnt have dreamed off a few years ago. Can't wait for it to finish. 2013 really could be our year!
... It's refreshing to read about a study that wll actually make a substantial contribution to the knowledge base, instead of the usual junk research whose main endpoint is to avoid the business end of 'publish or perish'.
You're right, we couldn't have expected anything like this a few years ago - XMRV, even though it was a dead end, bought us this opportunity.
I wholeheartedly agree: Mady Hornig and Ian Lipkin are class acts and whatever the findings from the study they will be robust conclusions that others (and hopefully they themselves) can build on.
 
I just wanted to say I love that cute pathogen image :D
I've recently been having bad reactions to probiotics. I've never had this problem before. I've been wondering if it's related to some pathogen. I stopped probiotics for a month or longer and then once I started is when I started having trouble.
 
Yeah!!! This is great news!! Thank you for posting this!!
 
What's the significance of gene expression and where can I read more about it? I've heard someone mention that methylation affects gene expression which I'm curious about since I'm following a methylation protocol. It's also been referenced in Life Extension magazine in relation to calorie restriction and mimetics. They seem to think it's important, but I don't understand any of it.
 
Thanks Simon! Looking forward to the results. I hope the cohort is well-defined too.

I wish everyone was getting the heavy-duty phase, not just the negatives from the first phase.

Edit: Simon later clarifies that "20% of those positive in stage 1 get sequenced in stage 2 regardless".
 
Thank you so much for this article! I've also been dying to know what was going on with this study.
 
Thanks for that info Simon.

When reading about this study, I often get a fear that the likely heterogeneous nature of CFS will lead to a misleading negative result. I know that I've seen Lipkin talk about CFS as being likely to have a number of different causes, but when there is reason to think that (for some) abnormal responses to common infections like EBV play some role, then if we do not understand the mechanism by which these responses occur (look at the trouble there's been understanding the role of EBV in MS) a well designed study which fails to find solid relationships could be taken as 'evidence of absence'.

From what I've read of his work with autism (another difficult area) and his brief work on XMRV, I think Lipkin seems like a great guy to have involved in the science of CFS, and he seems to have a cautious and understanding approach to the politics too... but I've still got a concern about how people at the Science Media Centre could present negative results.
 
Great article Simon and it's great to have the results of this study to look forward to! Even negative results (which I think are unlikely) are progress, as they narrow down the possibilities. As long as the study is well-run, it'll be useful :)
 
Are they going to be looking at CNS fluid in the studies? Or just blood?
I was just thinking that they must've not considered looking at cerebrospinal fluid or tissue biopsies because it costs lots more $ and is lots more cumbersome.

A member here brought my attention to the fact that, for instance, hhv6 may not be found in blood but in tissue such as in gut as I myself have shown (i've never shown hhv6 in blood but have super high levels of it in two gut biopsies). So what if this study shows false negatives?
 
Bit more on cohorts:
Are they going to be looking at CNS fluid in the studies? Or just blood?
So far, just bloods, but Mady Hornig enthused about Dan Peterson's Cerebro Spinal Fluid samples and hopes to get funding for these too. Peterson has 60 extremely-well defined ME/CFS patients, plus an equal number of healthy controls or 'comparators' i.e. patients with specific illnesses eg MS.
I hope the cohort is well-defined too.
I think they are. Here's a quote from CFI's Scott Carlson about their cohort in a CFS Central Interview:
The CFI-sponsored cohort recruitment protocol uses the updated CDC definition and the Canadian Consensus Criteria to identify well-characterized subjects. We also noted the new definition presented at the recent IACFSME conference [International ME Criteria].
So that's half of the pathogen samples in this study. I'm fairly sure (but not 100%) that most of the remainder are from the NIH XMRV study (that Lipkin & Hornig ran) which used Canadian Consensus Criteria only.
 
I would have thought Dr Chia's patients would have been of interest - stomach biopsies??
 
I didn't see in the article who is funding these studies. Is it the CFI? I imagine they would be very expensive!
 
...I often get a fear that the likely heterogeneous nature of CFS will lead to a misleading negative result.

I know that I've seen Lipkin talk about CFS as being likely to have a number of different causes, but when tthere is reason to think that (for some) abnormal responses to common infections like EBV play some role, then if we do not understand the mechanism by which these responses occur (look at the trouble there's been understanding the role of EBV in MS) a well designed study which fails to find solid relationships could be taken as 'evidence of absence'.

From what I've read of his work with autism (another difficult area) and his brief work on XMRV, I think Lipkin seems like a great guy to have involved in the science of CFS, and he seems to have a cautious and understanding approach to the politics too... but I've still got a concern about how people at the Science Media Centre could present negative results.
Reasonable points, and Mady Hornig and Ian Lipkin have designed the studies to take account of this, at least to some extent.

First off, they have huge samples (esp if they get funding for 800 patients) which will make it easier to detect a real association, even if there are several sub-groups within CFS (which most people expect there to be). But even more important is their work on protein profiles.

One possibility Mady Hornig discusses is where many different pathogens cause the illness (eg if it's a much down to the person's genetic make up, and the environment, as it is to any bug). Their specific search for markers of immune response, inflammation and oxidative stress should show if there are general signs of infection in CFS patients vs controls. But also this seach, plus the wider seach of proteins they are doing on a sub-sample, shuold also pick up any general disturbances in the immune system, or other systems in the body.

So I think it's unlikely that the study would completely find 'evidence of absence' even if no pathogens are definitively found. Mady Hornig's talk was mainly about her work in other hard-to-understand illnesses and the kind of innovative and imaginative work she's done elsewhere strongly suggestss she won't give up even with a negative result - blog coming very soon!

While there's know controlling how the SMC might spin 'no pathogens linked to CFS' results, dismissing Lipkin's work - let's say they finds no pathogens but do find signs of abmormalities - won't be easy: he's a major-league scientist.
 
Did she say anything in her talk about treatments, Simon, if any of this stuff comes up positive?

I know it's not her area and it would be hard to talk specifically about treatments before the bugs/host end has been done but I wondered if her experience in doing this sort of thing with other diseases had given her any experience about what happens next, once that research is done - does it tend to point to treatments that can be fast-tracked?
 
While there's know controlling how the SMC might 'no pathogens linked to CFS' results, dimissing Lipkin's work - let's say they finds no pathogens but do find signs of abmormalities - won't be easy: he's a major-league scientist.
And Lipkin has experience from XMRV of how ME/CFS stories get negatively spun, so he'll probably be carefull how anything is published, even potential negative aspects.