2022: Looking back on a year of ME/CFS research

As 2022 is closing, it is time for our annual tradition where we look back on the most interesting scientific studies on myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) of the past year.

The yield for 2022 is far from impressive but there were a couple of interesting ME/CFS studies that are worth discussing. A recurrent theme seems to be problems with the transport (or utilization) of oxygen into tissues. It will be interesting to see if this hypothesis will receive further support in 2023.

The largest genetics ME/CFS study to date

Let’s start with the largest genetics ME/CFS study to date. The Norwegian research team of Riad Hajdarevic and colleagues screened the entire genome for genetic differences between ME/CFS patients and controls. They collected data from 2532 patients which came from three different sources. The initial analysis was performed on patients from Norway and the authors also had a Danish ‘replication cohort’ to test if their findings could be confirmed in another patient group. Most of the data, however, came from patients who were registered in the UK Biobank as having a doctor’s diagnosis of chronic fatigue syndrome.

The results can be summarized briefly: not a single genetic difference between ME/CFS patients reached the significance threshold scientists use for genome-wide association studies (GWAS). The strongest signal was found in the Tubulin Polymerization Promoting Protein (TPPP) gene region. TPPP refers to a family of proteins that help assemble microtubules, the thin filaments that form the cytoskeleton of cells. Hajdarevic and colleagues were intrigued by TPPP because it is highly expressed in brain tissue and seems to play a role in the myelination of nerve cells. It will be interesting to see if this finding can be replicated in future genetic studies of ME/CFS.

Nonetheless, the main finding of this study was that none of the genetic differences reached statistical significance. This was not exactly a surprise because the consensus is that GWAS needs tens of thousands of participants to reliably detect genetic clues. In the discussion section of their paper, Hajdarevic and colleagues admit that their study was not adequately powered. They estimate that “up to 10 times more patients” are needed to detect risk variants with a small effect size. Fortunately, the long-awaited British GWAS named DecodeME has launched this year. It aims to collect saliva samples from 25.000 ME/CFS patients and will give us a higher-resolution picture of the genetics of ME/CFS.

Another big hurdle is heterogeneity. Because a diagnosis of ME/CFS is still largely based on symptom reports, it likely groups multiple unknown pathologies into one label. This heterogeneity makes it even harder to find genetic risk factors. Luckily, a company called PrecisionLife has developed a new (patented) analysis method that focuses on identifying subgroups in genetic analyses. While studies usually look for differences in single DNA letters, this approach can screen for combinations of features, so-called disease signatures. In 2022, PrecisionLife published an analysis of patients in the UK Biobank that identified 14 genes as likely associated with ME/CFS. The authors hope to replicate their results with the bigger dataset that DecodeME will provide.

Slow but deep breathing

This year also saw the best study to date on cardiopulmonary exercise testing in ME/CFS. In the past, numerous studies have reported differences between ME/CFS patients and (sedentary) controls when put on an exercise bike. Pretty much all these studies, however, suffered from the same shortcomings. They did not use blinded assessments, they failed to use standardized criteria for peak effort, and, most importantly, they did not use fitness-matched controls. Without these, we can’t know if differences in exercise test results are due to ME/CFS or a secondary consequence such as a lower fitness level or failure to reach peak effort during the test.

The study by Dane Cook and his colleagues is the first to use all three methodological safeguards mentioned above. The data came from 214 ME/CFS patients and 189 controls who participated in the Multi-site Clinical Assessment of ME/CFS (MCAM) project coordinated by the Centers for Disease Control and Prevention (CDC).

When Cook and colleagues analyzed the data as previous studies did, they found many of the differences that were previously reported. ME/CFS patients produced fewer watts, had reduced oxygen consumption, and their heart rates were lower. But when his team controlled for differences in fitness by expressing the data relative to each individual’s peak oxygen consumption (and by removing results that did reach peak effort) a different picture emerged.

Now many of those differences disappeared. As the authors explain, the results “indicate that many of the cardiopulmonary differences that have been reported in previous studies are explained by differences in aerobic fitness, and consequently exercise time, and are not pathophysiologic characteristics of ME/CFS.” There was, for example, little evidence for the idea that ME/CFS patients’ heart rate does not rise sufficiently with increasing effort, what others called chronotropic intolerance. Cook and colleagues also found no difference in the lactate response at rest, during exercise, or during the recovery phase.

Only a few small differences between ME/CFS patients and controls remained after adjusting for fitness but they are rather interesting. Cook and colleagues found a special breathing pattern characterized by higher tidal volumes and lower breathing frequency. In other words, ME/CFS patients were breathing slower and deeper even though they were doing the same exercise as controls. Cook and colleagues do not think this breathing pattern reflects a problem with the lungs or heart and pumping oxygen to the rest of the body. Instead, they suspect it might have something to do with poor oxygen extraction in skeletal muscle.

The Mestinon trial

This brings us to Harvard researcher David Systrom. His team is specialized in invasive cardiopulmonary exercise testing (iCPET), a procedure that allows researchers to measure gas exchange and blood flow through the arteries while patients are doing an exercise test. Systrom has been doing iCPET on patients with unexplained exercise intolerance for a while. He discovered that a subgroup had an abnormally low biventricular filling pressure or “preload failure” which could not be explained by a known heart condition.  As our research review of 2021 noted, most of these patients met ME/CFS diagnostic criteria. Some had high pulmonary blood flow but were less efficacious at extracting oxygen. Much like Cook’s study, Systrom’s research seems to be pointing towards a problem with getting oxygen delivered to the muscle.

In 2022 Systrom followed up on his findings with a randomized trial of the drug pyridostigmine (better known under its trade name Mestinon). Mestinon works by inhibiting the breakdown of acetylcholine, a neurotransmitter that is important for muscle contraction. Systrom thought that Mestinon would improve the cardiac output and exercise performance of his ME/CFS patients.

The results of the trial support this view but the changes were small and likely not clinically relevant. The trial also only looked at an acute effect. Patients did an iCPET, received a 60-mg dose of Mestinon or a placebo, then waited for 50 minutes and did another iCPET. The trial compared the differences between the first and second iCPET. In other words, it can’t tell us much about the long-term effects of Mestinon on ME/CFS symptoms.

Plasma metabolites

There were also a couple of interesting studies from the ME/CFS collaborative research centers in the United States.

The team of Ian Lipkin at Columbia analyzed 888 metabolic compounds in the plasma of 106 ME/CFS cases and 91 healthy controls. Patients had lower levels of carnitine and decreased levels of phospholipids, especially plasmalogens and phospholipid ethers. According to the authors, this points towards a dysregulation of “peroxisomal metabolism”. Peroxisomes are small, membrane-bound organelles that are involved in various metabolic processes such as the breakdown of fatty acids and amino acids. They are much like lysosomes with the main difference that enzymes of peroxisomes require oxygen to function. In addition, Lipkin and colleagues found elevated levels of dicarboxylic acids in ME/CFS patients, stating that “consistent with previous literature our results suggest dysregulation of peroxisomal metabolism and the tricarboxylic acid (TCA) cycle.”

Another impressive study came from the research team of Maureen Hanson from Cornell University. They tested 1157 plasma metabolites in 60 ME/CFS patients and 45 matched healthy controls. The cool thing about this study is that they measured metabolites before and after a double cardiopulmonary exercise test. The 2 exercise tests, taken 24 hours apart, were intended to invoke post-exertional malaise so that researchers could see what effect this had on patients’ metabolism.

It did seem to have made a difference. The number of metabolites that were significantly different between ME/CFS patients and controls increased substantially after the exercise tests. The problem was that none of the differences got large. There was always a considerable overlap between the ME/CFS and control group. Hanson, therefore, said that none of these can be used as a biomarker to diagnose patients. Instead, the differences should mainly be seen as clues into the underlying pathology of ME/CFS.

Unfortunately, it is not clear what the clues are telling us partly because 224 of the metabolites tested in this study have not been identified yet (It is not clear what they do and what bodily processes they are involved in). Overall, however, the data pointed toward a disruption in lipid metabolism. In the paper, Hanson and colleagues also highlight that numerous of the altered pathways depend on glutamate metabolism. Glutamate is an amino acid that has various functions in the body. It plays, for example, an important role in the transmission of signals between nerve cells. Glutamate itself, however, was not statistically different between patients and controls in the current study so it remains to be seen if this is a valuable clue or a dead end.

The research team of Hanson published another interesting study this year. It was posted as a preprint in October and has not yet been peer-reviewed and published in a scientific journal. In this smaller study of 30 ME/CFS patients, the researchers used single-cell RNA sequencing (scRNA-seq), a relatively new technique that allowed them to study the gene expression of individual cells. This analysis was also done before and after an exercise test. This time, however, the exercise test did not make much of a difference. The only significant change was seen in the transcriptomes of platelets. They were abnormal before the exercise test in ME/CFS patients and became normal after exercise. According to the authors, this indicated either a loss of platelets with defective transcriptomes or a large infusion of new, normal platelets after exercise. The main abnormalities found in this study, however, were not affected by exercise and were seen in monocytes. According to the authors, “the transcriptomes of classical monocytes from ME/CFS patients are biased towards a profile that promotes migration of monocytes to tissue and increased progression towards a macrophage fate.” The monocytes also had genes activated that are associated with the anti-inflammatory cytokine IL-10 which seems to be a conflicting message. While the advanced techniques used at the collaborative ME/CFS research centers have given us lots of interesting data, it is not always clear what they mean.

Endothelial dysfunction

While some ME/CFS research groups try to collect a large set of measurements and then see where the data leads them, others have a particular hypothesis in mind. Several groups suspect, for example, that ME/CFS patients suffer from endothelial dysfunction. Endothelial cells form a layer that lines the blood vessels and controls exchanges between blood and surrounding tissue. By releasing nitric oxide (NO), a gas that relaxes the muscle cells in the walls of blood vessels, they can regulate how much blood and oxygen flow to tissues.

Last year, the research group of Francisco Westermeier reported that several microRNAs that reduce the production of NO are increased in ME/CFS patients compared to healthy controls. This year, they followed up on these findings with a new experiment. They cultured endothelial cells in the lab and then found that these were less able to produce NO when they were put in the plasma of ME/CFS patients compared to the plasma of healthy controls. 

A German research group, of which Carmen Scheibenbogen is a member, is pursuing a similar hypothesis. They tested the reactive hyperemia index (RHI) in 14 patients who developed ME/CFS following COVID-19. RHI is a measure of vascular function that assesses the ability of blood vessels to dilate in response to an increase in blood flow. 5 Out of 14 ME/CFS patients had a reduced RHI, while none of the controls did. A reduced RHI was also found in 5 out of 16 Long Covid patients who did not have ME/CFS. This German research group also reported that concentrations of Endothelin-1 (ET-1) were significantly increased in ME/CFS and Long Covid patients compared to controls.

This is interesting because this finding was also reported by the Spanish study of 67 ME/CFS patients and 48 healthy controls. In contrast to NO, ET-1 is a potent vasoconstrictor, meaning that it causes blood vessels to constrict. This Spanish group also found increased levels of vascular cell adhesion molecule-1 (VCAM-1), a protein that is expressed on the surface of endothelial cells. It plays a role in the recruitment of white blood cells to sites of inflammation. In contrast to ET-1, however, the levels of VCAM-1 did not correlate well with the severity of the symptoms in ME/CFS patients.

Blood clots and ATG-13: overstated research?

This brings us to our next topic: blood clots!

You have probably heard about it because this line of research received a lot of attention in the media. South-African physiologist Resia Pretorius and British biochemist Douglas Kell argued that Long Covid patients suffer from an excess of tiny blood clots and that these explain lingering symptoms following COVID-19. Their research team went on to claim that Long Covid symptoms can be successfully treated by anticoagulant and antiplatelet therapy to prevent microclotting. While some researchers saw value in this theory, others reacted with skepticism, saying they could not find any evidence of it in their Long Covid patients, for example in capillary-rich organs such as the lungs and kidneys. Out of desperation and lack of better treatment options, many Long Covid patients have already undergone experimental testing and treatment for the supposed microclots in their blood. 

This year Pretorius and Kell published a study on microclots in ME/CFS patients. In their paper, they argue that the area of plasma images containing microclots were more than 10-fold greater in the plasma of ME/CFS patients than that of healthy controls. Image interpretation, however, is highly subjective and it is unclear if researchers were properly blinded to ME/CFS status. They also performed thromboelastography (TEG), a more objective method to test the efficiency of blood coagulation. An abnormality was found in the rate of clot formation (alpha and MRTG) but there were no significant differences in clotting time, maximum amplitude, or total thrombus generation. The authors note that “the extent of hypercoagulability of the ME/CFS group was not as severe as we have previously reported in diabetes, acute SARS-CoV-2 infection, or Long COVID/PASC.” This makes it rather unlikely that these blood clots are the cause of characteristic ME/CFS symptoms such as post-exertional malaise. Kell has also claimed that blood clots are present in multiple other conditions, not just Long Covid or diabetes, but also rheumatoid arthritis and lupus, making the whole theory rather questionable.

Another researcher who thinks to have found evidence of clotting is Avik Roy, the chief scientific officer at Simmaron Research. His team found protein aggregation in a small sample of 7 ME/CFS patients. They suspected this pointed towards a defect in autophagy, the clean-up and recycling process of cells. They investigated this further and found that the autophagy-related protein ATG13 was upregulated in the serum of ME/CFS patients. Rather than confirming this in a bigger sample, Roy and colleagues continued further experiments. They noticed that the serum of ME/CFS patients evoked the production of reactive oxygen species (ROS) and nitric oxide (NO) in microglial cells. The cool part of this study is when they neutralized ATG13, the production of ROS and NO in microglial cells greatly diminished. All these experiments, however, were done in only a handful of ME/CFS patients and so the results are not that robust.

Roy has continued his experiments on ATG13 with a mouse model which was announced on social media with strong statements. In November Simmaron Research tweeted: “We’re onto something big: a treatable pathway for PEM. The implications are huge. Not only do we believe the chemical pathway involving the ATG-13 protein is a culprit in PEM, we believe it can be targeted for drugs. There is hope for treatment!” Unfortunately, the data do not support these claims. We hope that Roy and Simmaron will be able to substantiate these bold claims with new publications in 2023.

Post-infectious syndromes

Post-infectious syndromes have historically been overlooked in medicine but are now receiving more attention during the ongoing COVID-19 pandemic. Jan Choutka wrote a nice overview about this topic in Nature Medicine. They highlight how long-term symptoms have not only been reported after COVID-19 but after various infections including Polio, Lyme disease, Ebola, Dengue, Epstein-Barr Virus etc. This year saw new publications highlighting “an under-appreciated burden of post-infection chronic sequelae” following dengue and other viral respiratory illnesses. A Dutch study that used the Lifelines cohort, found that approximately 12% of patients have lingering symptoms that could be attributed to COVID-19. In the US, a prospective study by John Aucott and colleagues found that participants with a history of Lyme Disease had 2-3 times higher odds of reporting moderate or severe fatigue compared to those without.

Florence Brelier and colleagues took a different approach but arrived at the same conclusion. They looked at UK primary care records and found that patients with acute Lyme disease subsequently were 2-3 times more likely to be diagnosed with fatigue and 16 times more likely to be diagnosed with ME/CFS. The number of patients with a diagnosis of ME/CFS in the study was only 17, making it difficult to draw strong conclusions, even though the effect size found was quite large. Another issue is the reliability of diagnoses in primary care records. The authors, however, were able to match Lyme disease patients and controls per general practitioner (GP). Therefore, underdiagnosis of fatigue or ME/CFS would be less of a problem as it would not explain a difference between groups that went to the same GP. The most likely explanation is that chronic Lyme and ME/CFS both describe similar health problems following an acute Borrelia infection.

On to Long Covid research. One of the most impressive studies of the year was published by a collaboration between the teams of David Putrino at Mount Sinai and Akiko Iwasaki at Yale.  They compared 99 Long Covid patients against healthy, uninfected controls and controls that were infected with SARS-CoV-2 but had no lingering symptoms. Putrino & Iwasaki used complex immunological tests which showed various (subtle) differences between Long Covid patients and the other groups. They also used a new approach to test for antibodies called Rapid Extracellular Antigen Profiling (REAP) but here no meaningful differences were found. The most important result came, rather surprisingly, from the stress hormone cortisol. In Long Covid patients, levels of plasma cortisol were roughly half of those found in healthy or convalescent controls. Cortisol has frequently been studied in ME/CFS and the results have been conflicting to say the least. It is also a hormone that is not very specific and can be easily influenced by different behavioral patterns.

Null results

Then there are a couple of studies that deserve a mention, not because they provided intriguing findings, but because they reported null results that might be helpful for further ME/CFS research.

The first is a study of the ELISA Autoantibody Assay produced by the German company Celltrend. This assay has been used in previous ME/CFS research, for example to argue that patients have increased autoantibodies to beta-adrenergic and muscarinic cholinergic receptors. Many patients have also paid to get these tests done. Dysautonomia International was skeptical of the assay and funded a study to test its reliability. Samples of POTS-patients and controls were blinded and sent to Celltrend. The results that the researchers got back were curious. The proportion of POTS patients and healthy controls who fell above the diagnostic thresholds was not different for any of the tested autoantibodies. In addition, 98.3% of POTS patients and 100% of controls had α1 adrenergic receptor autoantibody concentrations above the seropositive threshold provided by the manufacturer. We hope that this problem with the CellTrend assay will be sorted out before it is used in future ME/CFS research.

An Israeli research group used a relatively new method called phage immunoprecipitation sequencing (PhIP-Seq) to test for antibodies against gut microbiota and viral antigens. They received a sample of 40 patients with severe ME/CFS from the UK ME/CFS Biobank (UKMEB). The number and diversity of overall antibody-bound peptides did not show any significant differences between patients and healthy controls. The researchers did find, however, that antibody responses against flagellins were over-represented in ME/CFS patients. Flagellins make up the tail of certain types of bacteria to increase their mobility (the Latin word flagellum means “whip” presumably because the tail makes a whip-like movement).

There was also a French study that looked at predictors of recovery and substantial improvement in 168 ME/CFS. The patients who recovered or improved were a bit older when they got the illness, and they were diagnosed sooner than the other patients. None of the other measurements including fatigue, pain, PEM severity, a sudden onset of ME/CFS etc, turned out to be a significant predictor. A possible explanation is that the sample size of the improved group was too small as recovery and improvement rates were only 8.3% and 4.8%, respectively.

Honorable mentions

Lastly, there are some honorable mentions that did not make our top tier and that we will only list briefly below.

Orji et al. published a prevalence study in Australia based on primary care data from 2015 to 2019. The prevalence of ME/CFS in patients aged 13 years or older was estimated to be approximately 1 in 1000 which translates to approximately 20.000 ME/CFS patients in Australia. The authors caution, however, that this is likely an underestimate of the true prevalence rate as not all cases of ME/CFS get picked up in primary care.

Vyas et al. conducted what is probably the largest study on the impact of ME/CFS on family members. Their online survey was completed by 1418 patients and their family members from 30 different countries. Family members were most impacted emotionally by worry, frustration, and sadness and personally by family activities, holidays, sex life, and finances.

The research team of Bhupesh Prusty, a virologist at the University of Würzburg, was able to search for herpesviruses HHV-6 and EBV in 3 autopsies of ME/CFS patients. They reported having found “abundant viral miRNA in various regions of the human brain and associated neuronal tissues including the spinal cord that is only detected in ME/CFS patients and not in controls.”

Brooke Scoles of the London School of Economics published an interesting study as part of her doctoral thesis. She extracted comments made by physicians on Reddit and analyzed in which terms they discussed various illnesses. The results for ME/CFS included over four times more negative language words than the results for other stigmatized illnesses such as depression.

Conclusion

So that is a wrap! If there are important ME/CFS studies that we missed, feel free to post them in the comment section below. We already look forward to reading the ME/CFS studies of 2023 and hope they will bring many interesting findings.

Best wishes for the New Year to all!