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

The Chronic Fatigue Initiative and Interview with Mady Hornig

View the Post on the Blog

In a follow-up article to the recent IACFS/ME conference presentation in San Francisco, and after speaking at length with Dr. Mady Hornig, 'searcher' delves deeper into the impressive work being completed by the Chronic Fatigue Initiative, and focuses in on those cytokine results ...

Members of the Chronic Fatigue Initiative (CFI) and Scott Carlson, the executive director of the Hutchins Family Foundation, gave a talk at lunch-time on the second day of IACFS/ME to share their research thus far and their future plans.

This presentation was one of the highlights of the conference to me because of the high quality research CFI is doing.

The CFI was established by the Hutchins Family Foundation in 2010 with a $10 million investment (Wall Street Journal), and their objective is to identify the causes of and treatment for chronic fatigue syndrome.

As far as I know, this occasion was the first time that CFI had officially revealed certain findings.

I think the CFI has already done a fantastic job furthering research into ME/CFS, especially since the U.S. government hasn’t shown any urgency toward researching the illness.

They have already spent much of this initial investment, but are working on identifying several more projects to which they will give additional funding.

The Panelists at the session were:

  • Nancy Klimas, Principal Investigator (PI) for CFI Cohort Recruitment
  • Gail Ironson, PI for CFI Cohort Database
  • Dana March, PI for CFI Epidemiology Data Analysis
  • Mady Hornig, PI for CFI Pathogen Discovery and Pathogenesis Project
  • Anthony Komaroff, Professor of Medicine at Harvard Medical School
The team's first goal was to recruit a cohort from experts in the field (see below).

CFI felt it was important to do extensive epidemiology work and they looked at the large dataset from the Harvard School of Public Health/Nurses’ Health Study, did a separate CFI Clinical Epidemiology study, and also analyzed the CFI cohort.

The CFI researchers are looking into the mechanism of illness. They broke the potential mechanisms down into three basic categories:

Virology, Mitochondrial Dysfunction, and Immune Dysregulation

Research into mitochondrial dysfunction and immune dysregulation was based on some initial research findings.

The CFI Cohort

Dr. Nancy Klimas (left) at the IACFS/ME conference
Principal Investigator for CFI Cohort Recruitment

Nancy Klimas explained that the CFI cohort consists of 203 ME/CFS patients and 202 closely matches controls.

It took exactly one year to assemble, which she felt would usually have required an approximately four-year long project. The complex data made it all the way to analysis within one year.

The cohort comes from five experts/sites: Donna Felsenstein (Boston, Massachusetts), Lucinda Bateman (Salt Lake City, Utah), Daniel Peterson (Incline Village, Nevada), Sue Levine (New York City, New York), and Nancy Klimas (Miami, Florida). It was important to have true experts in the illness recruiting patients to make sure that every single member of the cohort is a true case.

The biorepository includes serum, plasma, RNA extraction, DNA extraction, urine, tears, saliva, rectal swabs, and peripheral blood mononucleated cells (PBMCs)

The resulting clinical dataset and biorepository are available to the wider ME/CFS research community. They will need to get more fecal samples for the upcoming microbiome study (discussed in more detail below) but otherwise there are still more samples remaining.

Their goal is to provide access to the repository for the purposes of “accelerating science and attracting new investigators to the field.”


The dataset includes 4000+ variables including data from the prescreen, standard labs, serologies, chart reviews, physical exams, and medication history.

CFI deliberately oversampled for acute onset patients and also made sure that at least 25% of the patients were ill less than three years and that at least 25% were men.

The cohort consists of severely ill patients with a high level of cognitive and autonomic symptoms.

Dr. Klimas thinks this is very timely data that may help with the debate on case definitions and that the dataset really reinforced the Canadian Case definition.

The average SF-36 was very poor, and is similar to other chronic illnesses such as very severe chronic obstructive pulmonary disease.

Cognitive dysfunction was as important as fatigue.

The more acute the onset, the more the inflammatory symptoms were predominant. The shorter duration cohort (<3 years since onset) had greater autonomic dysfunction.

Women tended to have more severe disease on average (more inflammation, autonomic dysfunction, GI problems, etc.) Younger patients were more likely to have POTS.

CFI Clinical Epidemiology Study

Dr. March spoke about the aims of the CFI Clinical Epidemiology study.

They had five aims:

  • To establish a longitudinal database to document the natural history of CFS
  • To evaluate presentation and trajectory of symptom change over time
  • To determine co-morbid and new diagnoses
  • To identify factors that contribute to recovery and relapse
  • To identify successful treatments with long-term benefit
This study looked at 1430 patients. Patients had to be 18+ years old and had to have been diagnosed more than five years earlier, with an emphasis on patients who had been seen for more than 10 years.

They received 960 completed surveys. Fifty-nine people in the sample were deceased.

As mentioned in our earlier reporting, a majority of the deceased patients were from Dr. Peterson's clinic. The reason for this is likely because those patients go back the furthest, so it’s likely due to the fact that Dr. Peterson has been seeing patients since the Incline Village outbreak. Of these patients 37.8% died of cancer, which is 4 times the background rate of cancer. Another 19% died of heart disease and 19% more died of suicide.

The patients in the study were mostly white (95%), middle-aged (55.1+/- 12.3 years), and highly educated.

Of these patients 57% reported a sudden onset and, “for every one-point increase in severity of initial post-exertional malaise, odds of reporting sudden onset increased by 63%.”

The type of onset didn’t impact the severity of symptoms:

“Symptoms either stayed the same or improved slightly; viral symptoms and post-exertional malaise showed the most improvement across sites.”

Improvements were generally modest.
Not surprisingly, there were a large number of new and co-morbid diagnoses. The average number of co-morbid conditions was 3.5 +/- 2.2. The most common by far was fibromyalgia, followed by depression, anxiety, hypothyroidism and sleep-related issues.

I think the finding that surprised me the most was that 30% of the patients reported at least one remission. The average number of remissions was 2.5 and the median duration was 52 weeks. I don’t know how they defined remissions (i.e., was it getting back to 100% or a more limited definition?)

Also of interest is that the more severe a patient’s symptoms were, the more likely he or she was to self-report improvement in health status over time.

The most effective reported treatment was self-help strategies: 65% found those strategies helpful. Rest was the top treatment, with 37% of patients saying they were helped by rest. Diet helped 21% and exercise 19%. Self-help strategies in combination with traditional medicine were the most effective treatment approaches and use of this combination was reported by approximately 25% of the patients.
The big question is what will they do with all this epidemiological data? They are circulating two papers -- one on the natural course of chronic fatigue syndrome and one on co-morbid and consequent conditions. They are also looking more in depth at treatments to understand which are most effective. CFI intends to share data with other researchers to allow for further research.

Pathogen Discovery and Pathogenesis Study

Dr. Mady Hornig at the IACFS/ME conference
CFI Lead Researcher - Pathogen Discovery and Pathogenesis

Dr. Mady Hornig then presented on the CFI’s pathogen discovery and pathogenesis studies. She started by saying it was her first time at the conference and she hoped IACFS/ME could become an annual meeting, instead of a biennial meeting. We also had follow-up conversations with Dr. Hornig to clarify some questions on April 14th and April 16th.

Her team's goal is to find pathogenic triggers -- microbial or otherwise -- that contribute to the initiation, perpetuation, or exacerbation of chronic fatigue syndrome.

They also really want to find biomarkers that serve as vulnerability factors prior to getting ill and after getting ill. They'd like to help parse out the great heterogeneity that is seen clinically in the disorder and perhaps uncover the different biological triggers that cause the illness.

The researchers looked at many factors in the main cohort including geographical residence, as there were clusters in the beginning of the history of the illness in the United States. They even looked at season and time of day of sampling to make sure the patient samples were closely matched to the controls.

CFI has funded an extension study. The CFI extension study consists of 50 cases and 50 controls re-recruited from the five original sites. They are getting fecal samples beginning in May to survey the microbiome and will look at the bacteriome (16S rRNA), the mycobiome (18S rRNA/ITS), and the virome (viral particle preparation/HTS.)

They are repeating immune profiling for longitudinal analysis and will collect PBMCs for future studies. Dr. Lipkin and Dr. Hornig would also like to do a larger, more in-depth study which is currently raising funds through a crowdfunding effort.

CFI has completed a CSF study with Dan Peterson's patients co-funded by the Evans Foundation, and that dataset has 60 cases and 60 neurological/non-neurological controls.

The team is also doing multiplexing, blood metabolomics, and targeted proteomics (large-scale study of proteins and their interacting partners) -- the metabolomics and targeted proteomics are in progress now. Metabolomics is the study of the complete set of small-molecule metabolites (including molecules such as serotonin, amino acids, sugars and hormones), which will include metabolites from the microbiome that are naturally absorbed into the blood.

Dr. Hornig thinks that looking at the microbiome and the metabalome in conjunction with one another, and understanding how they correlate, could help lead to new blood tests that reflect disturbances in the microbiome -- including the presence of specific pathogens.

Such disturbances may be important in certain subsets in chronic fatigue syndrome, although the CFI investigators aren't currently funded to study the metabalome and microbiome at the same time, which is work that could help solidify a potential link. They have also looked closely at cytokine and immune signatures.

Fascinating cytokine results

Initial screening studies of 265 serum samples (214 cases and 51 matched controls) revealed very few pathogens. There were 2 cases with HHV-6B and 1 with parvovirus B19. But they found a lot more when they looked at cytokines.

The cytokine results showed very strong evidence of inflammation in ME/CFS patients. The researchers saw some very interesting phenomenon related to length of illness when they looked at 51 cytokines in patients from the CFI cohort. The reasons for this difference are unknown but immune exhaustion could be playing a role.

Patients who had been ill for less than three years had an increase in certain pro-inflammatory cytokines, especially IL1B, IL6, IL17A, and IL8. Patients who had been ill for greater than three years had similar results to controls for those cytokines.

Similar results were seen for Plasma Th2/counter-regulatory cytokines-- IL4, IL10, IL13, and IL1RA were increased in patients ill for less than three years, but patients ill for longer durations were on average similar to controls.

For earlier reporting on what cytokines are and how they may be related to ME/CFS, please see Andrew Gladman's article on cytokines.

TNFα and IFNγ were increased in early phase ME/CFS, whereas eotaxin and CD40L (which is involved in apoptosis) were decreased compared to controls.

After looking deeper at results, they saw that IFNγ and IL12p40 were significantly increased. IFNγ was the most striking result and was also associated with cognitive impairment. In his summary at the end of the conference Dr. Komaroff said about the extremely statistically significant interferon-gamma results: "IFNγ is thought to be commonly released by viral infection or infection with intra-cellular bacteria, so it suggests but by no means proves that in CFS there may be an underlying infectious agent."

There was a significant linkage between two of the cytokines and cognitive impairment. An increase in IL17A and a decrease in GROα was highly correlated with cognitive impairment in the early-stage patients.

When they looked at CSF the researchers found that inflammatory cytokines were greatly dysregulated in patients. Eotaxin was increased in CSF, unlike the decrease seen in plasma in early-stage patients. The reason for this is currently unknown.

Dr. Hornig mentioned that IL17A antibodies are already in use in rheumatoid arthritis due to IL17A's association with RA symptoms. She stated that there isn’t enough evidence to do an intervention yet, but it is definitely worth investigating. It is possible that this could help in early-stage patients in preventing a persistent illness since early-stage patients generally express high levels of IL17A.

Scott Carlson of the Hutchins Family Foundation concluded by stating that they want to continue to further research in CFS and would like their samples and data to be used by other researchers for rigorous research done in a collaborative manner. We should be seeing a lot more from CFI once their papers get published, the ongoing studies are completed, and the new projects are identified and funded.

Phoenix Rising is a registered 501 c.(3) non profit. We support ME/CFS and NEID patients through rigorous reporting, reliable information, effective advocacy and the provision of online services which empower patients and help them to cope with their isolation.

There are many ways you can help Phoenix Rising to continue its work. If you feel able to offer your time and talent, we could really use some more authors, proof-readers, fundraisers, technicians etc. We’d also love to expand our Board of Directors. So, if you think you can help in any way then please contact Mark through the Forums.

And don’t forget: you can always support our efforts at no cost to yourself as you shop online! To find out more, visit Phoenix Rising’s Donate page by clicking the button below.

View the Post on the Blog
Last edited by a moderator:


Thanks @searcher for a cracking article.

I am hugely excited by CFI's work and ambition. Their cohort of 200 patients and carefully matched controls, complete with rich clinical data and plenty of biological samples seems to me to be taking ME/CFS research into the major league: it's a huge resource they've made available and I do hope that other researchers are quick to use it too.

Nancy Klimas's comment the data collected already can help sort out the case definition date is tantalising, esp given she adds it backs up the Canadian Criteria. Hopefully it will dovetail with data coming out of the CDC multi-clinic study too.

The cytokine data from Mady Hornig was another highlight, esp the correlation with cognitive problems.. I see they looked at 51 cytokines, the same number covered by Jose Montoya's independent and equally large cytokine study - hope they are the same 51. Two huge studies both looking at the same thing has got to be a good thing.

Can't wait to see the papers when they start rolling off the production line.
Thank you, I however think it's important when these cytokine levels have been measured. Maybe they are higher in periods of PEM.
Thanks guys!

I agree @svetoslav80 - I think that one area of research that seems be missing in the CFI studies is looking at cytokines and the metabolome at many different time points to see how they fluctuate, especially with PEM. It is something I have thought about a lot-- when I participated in a study I actually purposely gave myself PEM so my results would show what wrong when I tried to do normal activities, but that is obviously reckless and still doesn't show the differences between my cytokine levels when I am at my worst vs my best.

That said, most of us still have symptoms even at our best so our microbiome/cytokines/metabolome should still be different than that of healthy controls at all times. There has to be something "off" that causes theoretically normal amounts of activity to trigger symptoms and the studies are already showing problems even without inducing PEM (other than the potentially serious PEM induced by going into the lab for blood draws.)

My hope is that CFI's findings can be looked at in conjunction with Dr Younger's study on day-to-day cytokine levels (http://www.translational-medicine.com/content/11/1/93) along with Dr Shukla's studies on the microbiome of CFS patients after exercise (http://solvecfs.org/research/cfidsa.../return-on-your-investment-sanjay-shukla-phd/). The added time-series info with the follow-up microbiome/cytokine study should also show potentially interesting variations in cytokines over time.
Last edited:
Thank you, I however think it's important when these cytokine levels have been measured. Maybe they are higher in periods of PEM.
Thanks guys!

I agree @svetoslav80 - I think that one area of research that seems be missing in the CFI studies is looking at cytokines and the metabolome at many different time points to see how they fluctuate, especially with PEM. ... The added time-series info with the follow-up microbiome/cytokine study should also show potentially interesting variations in cytokines over time.
Nancy Klimas was doing a 'Good day/Bad day' study a few years ago that I think looked specifically at cytokines - but I have never seen any results reported for it. Will try to find out what happened.
Last edited:
Nancy Klimas was doing a 'Good day/Bad day' study a few years ago that I think looked specifically at cytokines - but I have never seen any results reported for it. Maybe they didn't find anything? Will try to find out.
That would be interesting
Thanks Searcher for a great summary. I missed this talk and am so glad you were able to capture and relay all this information for the rest of us.
I can't wait for the results on the blood metabolomic and proteomic studies. I'm sure they will show abnormalities which will prove, even further, that we really are ill. I have had a similar type of testing done at Nutrichem, a testing lab here in Ottawa Canada, that tests for metabolites from all sorts of chemical reactions that go on in the body on a daily basis. I had a lot of very abnormal results, especially in the energy ( mitochondrial) portion of the testing panels. I think a lot ( or all of us) will show the same.
Hope you are OK after doing all that work at the conference. It is very much appreciated by everyone, including myself. Even though I was there in body, I can see now that my brain was napping a lot of the time.
@Gamboa - My brain was also half asleep a lot at the conference; I also had to miss some sessions to take some naps in the break room.

The CFI presentations were definitely a highlight for me and I am also especially looking forward to the proteomics and metabolomics results. It is great to see test results that reflect what we have all been experiencing.

Would you be up for sharing your Nutrichem results? I think that would be really fascinating to see.

I really hope you are recovering ok from the conference as well. It was an intense experience so I have had to rest even more than usual since then. I think it was worth it though.
Last edited:
What is the situation for markers in long-term patients?
also thanks so much for a fabulous article. :)
Hi @WillowJ - Patients who had been sick for greater than three years looked a lot more like controls. The researchers aren't sure why that's the case, but one theory is that the reduction in inflammatory cytokines is due to "immune exhaustion." I suspect this could mean that treatments for more recent-onset patients may turn out to be different than that for patients who have been ill for longer.
And thank you :)
Last edited:
@WillowJ I forgot to mention a key point from Dr Hornig- many, but not all, of the cytokines that are high in short-duration patients were actually lower than controls in the long-duration (>3 years) patients in this study. I think this will turn out to be very important for testing.