(In the second of a series of blogs of presentations on the UK CFS/ME Research Collaborative (CMRC) Conference in September of this year, Simon tackles Dr. Lloyds fascinating talk on the how infections trigger chronic fatigue syndrome.
Thanks for Simon’s and Action for ME for allowing us to post his summary. You can find more of their summaries here. I added many of the images and a few of my comments are in italics)
Prof Andrew Lloyd, University of New South Wales- Acute infection & post-infective fatigue as a model for chronic fatigue syndrome.
An innovative study that tracked the development of CFS/ME after several different initial infecctions has revealed surprising findings about the role of the immune system and role of genetic variations in the susceptibility to post-infective fatigue.
The sheep and cattle-farming outpost of Dubbo, four hundred miles inland from Sydney, Australia, was the surprising setting for a key CFS/ME study that put the development of the illness under the microscope. Prof Andrew Lloyd, an infectious diseases physician and immunology researcher from the University of New South Wales In Sydney, has documented the onset of CFS /ME in a tightly-defined subset of patients – and in the process has begun to unravel how CFS/ME develops after an infection.
One of the main difficulties in CFS/ME research, Prof Lloyd argued is ‘heterogeneity.’ All the evidence points to CFS/ME being a mixed bag of patients with potentially differing disease processes giving rise to a similar set of symptoms in all of them.
This diversity has scrambled research findings making it difficult to find a consistent set of abnormalities. In the 1990’s, Lloyd tried to explore this problem by systematically analysing patient symptoms. And indeed, he was able to isolate two broad groups — but even then it was clear that the bigger group was a mixed bag.
Lloyd concluded that the only way to make progress in the field was to zero in on a tightly-defined group of patients to see what went wrong as they fell ill.
Cue the seminal “Dubbo” studies of post-infective fatigue.
The basic idea was simple: track patients from the start of an infection and then compare those that develop post-infective fatigue with those that have exactly the same infection at onset but recover promptly. The goal was to see what was different about those who became fatigued and those who did not.
The team deliberately selected three very different infections known to be able to trigger prolonged illnesses:
- Epstein-Barr Virus or EBV – a ‘great big DNA monster’ of a virus that causes the fever, sore throat and tender lymph nodes typical of glandular fever and infectious mononucleosis.
- Ross River Virus – a tiny RNA virus spread by mosquito bites which causes fever, rash and joint pain.
- Coxiella burnettii – a bacterium that causes Q fever, usually picked up from sheep and cattle, which causes a nasty illness with high fever, and drenching sweats as well as inflammation of the liver.
Prof Lloyd had two working hypotheses: the chronic fatigue would be either due to an infection that doesn’t go away, or to an immune system that remains stuck on after the infection is gone. He said that at the time he thought: “It’s going to be one or the other – or just possibly a combination. I’ll sort it out quicktime, don’t you worry!”, a comment that drew much laughter. He also suspected genes played a role.
The key to the study was catching people early. They established good relationships with the local doctors so as to be notified as soon as there was a new case of glandular Fever, Ross River virus infection, or Q fever. Self-report questionnaires and blood samples were collected regularly and those who remained unwell at 6 months were formally evaluated for CFS (Fukuda criteria) using medical and psychiatric exams as well as laboratory tests.
All three infections led to a similar ‘natural history’. The proportion of subjects with ongoing symptoms started high in the acute phase and gradually reduced over time with fatigue being a particularly persistent symptom.
Regardless of the specific infection, around 10-15% met Fukuda criteria for chronic fatigue syndrome at six months, with approximately 6% still fatigued at twelve months, and around 2% remaining fatigued for many years. Overall, this looked like a good model for studying CFS/ME.
Severity of the Acute Infection is Key
Prof Lloyd and his team then started trying to identify factors that played a role in the development of long-term fatigue. It turned out that neither age, demographics, nor the presence of a psychiatric disorder prior to getting sick predicted fatigue. Nor, surprisingly, did the type of infection – with a similar illness duration seen for all the three very different types of infections.
What did predict long-term fatigue (and CFS/ME status) was the severity of the initial illness. For example, the 10% or so of patients who were ill enough initially to be hospitalized had a 200-fold higher risk of having CFS at six months. The graph below shows the huge difference in recovery rates between the third of patients who were severely affected (not just the hospitalised) at the onset or acute phase of the illness and those who were least affected:
Cytokines Can Drive Symptoms
To begin tracking down the differences in what was happening in the bodies of those who developed CFS/ME, Prof Lloyd started thinking about the ‘acute sickness response’, just as Prof Dantzer had spoken about earlier in the day.
What Prof Lloyd calls the ‘sickness response,’ and Prof Dantzer calls ‘sickness behaviour’ are the same. They refer to a biologically-driven, evolutionarily conserved process that occurs across species because it enhances survival value by helping animals or humans to focus on recovery from infection. Symptoms include fever, loss of appetite, muscle and joint pain, general malaise (feeling unwell), concentration problems and, of course, fatigue.
Much as Prof Dantzer wondered if a sickness response gone wrong was a factor in cancer-related fatigue, Prof Lloyd wondered if it could be a factor in post-infective fatigue
Lloyd’s group wanted to revisit the early work on sickness response in animal studies “in the flesh – in humans”, using Dubbo patients. They examined the link between two key pro-inflammatory cytokines, IL-1 beta and IL-6, and symptoms during the early acute phase Q fever patients. And they found the expected correlation between these cytokines and symptoms such as fatigue, muscle pain, poor concentration and malaise.
Looking for Chronic Immune Activation
So then they tracked levels of three key pro-inflammatory cytokines (IL-1b, IL-6 and TNF alpha) over time in the blood samples of their post-infective CFS/ME patients and individuals who recovered promptly from the same infection.. Each was known to cause symptoms associated with “sickness behavior”.
Despite that fact that the CFS/ME patients were exhibiting symptoms associated with sickness behavior, this hypothesis didn’t pan out: at three, six and twelve months after the initial infection, levels of all three cytokines were no different between fatigued cases and recovered controls. ‘The chronic immune activation hypothesis” Prof Lloyd said “was looking wobbly,’.
Prof Lloyd pointed out that pro-inflammatory cytokine levels that his group had found in patients during the acute phase were present in nanograms per millilitre, which is the level at which they are biologically active.
While some other studies have found elevated pro-inflammatory cytokine levels in CFS/ME patients (unlike his own study), those levels have been in picograms per millilitre, that is a thousand times lower, adding “it is not clear at all that that has biological meaning”.[Some researchers believe cytokine levels may not be the issue so much as the brains response to those cytokines. Miller with his basal ganglia and Younger with his leptin studies suggest small amounts of cytokines could be triggering a higher than expected microglial response in the brain. In his vagus nerve hypothesis vagus nerve infection hypothesis Van Elzakker proposed a small localized infection at the vagus nerve could be doing the same. ]
Looking for Chronic Infection
To test the idea of chronic infection, they simply tracked the persistence of the micro-organisms in the blood over time. They found that EBV, which causes glandular fever, did persist at measurable levels in the blood — but that’s normal for EBV, and crucially there was no difference in EBV levels between fatigued cases and recovered controls.
The Ross River virus and Coxiella burnettii (Q fever) pathogen were cleared rapidly, and again there was no difference in viral loads between cases and controls: the chronic infection hypothesis wasn’t standing on firm legs either.
Cytokine Genes Play a Key Role
If neither chronic infection nor high cytokine levels were causing the CFS/ME patients to remain stuck in their “sickness response” what was? Could the answer lie in their genes. They suspected some people might be genetically programmed to have their immune systems respond more vigorously to an infectious insult. These people, they reasoned, would be likely to become more severely affected in the acute phase of infection and thus have a greater likelihood of post-infective fatigue.
They focused on the inherited variations in genes which encode the immunological and neurological proteins likely to be involved in the sickness response. Specifically, they looked at the genes for five cytokines: the three they’d studied before, plus interleukin (IL)-10 (which inhibits inflammation) and Interferon (IFN)-gamma (which promotes inflammation).
Since high- and low-producing variants of each gene are found, a predominance of high producing gene variants in the CFS/ME patients could help explain why they remained sick while the others recovered.
The results were striking. Patients with the high-producing version of IFN-gamma or the low-producing variant of IL-10 were much more likely to have severe symptoms. The effect was even more prominent in those who had a combination of these ‘risky’ genes. The study showed that:
- those with one copy of the high-producing IFN-gamma gene were more likely to develop severe symptoms than those with only the low-producing IFN-gamma gene, with an odds ratio of 2.6 (where an odds ratio of 1 means no difference between two groups)
- those who had both copies of the high-producing version of the gene were even more likely to have severe symptoms: the odds ratio was 2.9
- those with a double whammy of both high-producing IFN-gamma genes and low-producing IL-10 genes had an odds ratio of 6.8, which is a really big effect.
The team also showed that those with the ‘high-producing’ gene version really did produce more cytokine during the acute phase of the illness. Overall, the type of IFN-gamma and IL-10 genes significantly affected illness severity, cytokine protein levels, and the duration of illness —independent of age, gender and even the specific infection type.
“How long you are going to be remain unwell after these acute infections is at least partly determined by your genetic makeup, in particular those two genes.”
However, these genes only explained some of the difference in illness severity between patients, so while they were part of the story, there must be other factors at play too. The team also looked at neuro-behavioural genes (for example, those that affect levels of neurotransmitters) and reported promising results on one, snappily called P2X7.
P2X7 is a neuro-immune gene that has been linked with psychiatric and neurological disorders. It’s also very important immunologically, affecting, for example, recovery prospects from tuberculosis. And Lloyd’s groups has now found that it may play a role in fatigue.
A cell-surface receptor that appears on both white blood cells and on microglia, the brain’s immune cells, P2X7 helps regulate inflammation, playing a specific role in the release of pro-inflammatory cytokines, including our old friend IL-1 along with IL-18, which stimulates release of IFN-gamma from T-cells. When it’s activated, P2X7 also acts as a pore, allowing transport of the energy molecule, ATP, across the cell membrane.
Prof Lloyd summed up by returning to his original hypotheses that post-infective fatigue was driven by chronic infection or chronic immune activation. This didn’t pan out: there was no difference between fatigued cases and recovered controls in either pathogen levels or cytokine levels.
The new hypothesis is that “stuff happens” in the acute infection, he said to laughter, though exactly how it happens isn’t clear. Severity of the acute infection predicts post-infective fatigue and our genetic make-up plays a role in both. He said if he had to choose a site where those genes play out, it might be in the body in the acute phase, but in the prolonged phase it’s going to be in the brain. Prof Lloyd added “we need to get over the blood” – the answer, he thinks, probably lies “upstairs”.
A new post-infective fatigue study under way (more conveniently located in Sydney) will be looking at autonomic factors and circadian rhythms. And, said Prof Lloyd, they wanted “to really get to the money” – by imaging activated microglia in the brain. (In answer to a later question from Prof Ian Lipkin, Prof Lloyd pointed to evidence in mice that an initial acute inflammation in the body can lead to long-term activation of microglia in the brain. Perhaps something similar is happening in CFS/ME cases that develop after infection.)
He also stressed the importance of using challenge testing. “If we are to continue to look at the blood, we need to use studies in which we carefully make the fatigue worse – transiently, to look at post-exertional exacerbation”, which he described as being “tough for the patients but good for the scientists!”
They are using a cycling exercise as a physical challenge and, intriguingly, they are also using a driving simulation as a cognitive challenge. Both trigger a symptom exacerbation which comes on over hours and lasts a day or two and then resolves. This offers real opportunities, Dr. Lloyd stated, to look for biological factors which correlate with the changes in fatigue.
It looks like Prof Lloyd’s group is well set up to reveal yet more about how CFS/ME develops after an infection.
- Simon’s first blog on the conference – Dantzer on “The Neuroimmune Basis of Fatigue” – and Chronic Fatigue Syndrome