Nothing yet in long COVID, or PASC, has better illuminated for me the difference between the chronic fatigue syndrome (ME/CFS) and fibromyalgia (FM) field than the recent RECOVER Research Review talk on biomarkers and long COVID. (Transcript here.)
The participants in the hour-and-a-half talk were impressive. An infectious disease doctor at the University of California San Francisco, Dr. Peluso, was, prior to the emergence of long COVID, focused on “long HIV”. Dr. Abdel-Mohsen of the Wistar Institute is a specialist on how the immune system responds to viral infections. Dr. David Walt is an award-winning nanoscience and nanomedicine inventor. Dr. Grace McComsey at Case Western Reserve University is an internationally known researcher in the field of the metabolic and cardiovascular complications of HIV. All appear to have glommed onto the possibility of long COVID quite early.
With long COVID, we’re in a different world – a bigger, more complex and more precise world. We’ve talked about biomarkers for chronic fatigue syndrome for years, but after Peluso’s talk I wondered if we have, really.
Dr. Peluso on Biomarkers…and an Explosion in Therapeutics
Peluso began one of the first long-term long-COVID studies in April – yes, April 2020. The man clearly knew what was coming. Since then, he’s enrolled hundreds of long-COVID patients into his LIINC study – which is now part of the RECOVER Initiative.
Symptom Assessments Don’t Cut it
The first takeaway from his talk was that using symptoms to characterize patients just doesn’t work. Too many variations in symptom presentation – even in this single virus-triggered illness – are present to use them. There’s lots of variability over time within individuals, and even more variability between individuals. That makes it hard to use “patient-reported outcomes” – those symptom assessments we see so often in ME/CFS and FM – as core outcome measures for long COVID.
Of course, this problem has been recognized in ME/CFS for years and is regularly trotted out as the reason for low pharmaceutical company interest. This is a big deal – so big that Peluso said, “we really desperately and urgently need other measurements that we can make to supplement the clinical data that people are telling us”; i.e. they need biomarkers!
Obviously, we’ve hardly scratched the surface of biomarkers in ME/CFS, as four different types of biomarkers exist – none of which I’ve ever seen differentiated in this disease.
- Mechanistic biomarkers – measure what’s gone wrong – as in high levels of the IL-6 cytokine. IL-6 wouldn’t be a diagnostic biomarker because high IL-6 levels are found in other diseases, but it would indicate the extent of the illness.
- Diagnostic biomarkers – a biomarker that’s unique to long COVID; if someone has this, they have long COVID.
- Predictive or prognostic biomarkers – biomarkers that suggest someone is at risk from long COVID, or that someone who has long COVID is at risk of having a severe case of long COVID.
- Surrogate biomarkers – the holy grail of biomarkers – a biomarker which, when you change it, the patient gets better or worse. Surrogate biomarkers pretty much make clinical trials a snap; all you need to do is measure the biomarker to see if the trial is successful.
The HIV Saga Tells the Tale
Peluso said that for most of the ’80s and part of the ’90s, diagnostics and therapeutics for HIV really struggled because of the lack of biomarkers. People were trying everything under the sun – almost always unsuccessfully – for one reason and one reason only – the field lacked accurate biomarkers. Clinical trials, that were based on the same types of inaccurate endpoints that are now available for long COVID and ME/CFS, had to be enormous to tease out even small positive effects over time.
Everything changed, though, when the biomarker for HIV – a surrogate biomarker (plasma HIV RNA levels) – was finally discovered in the mid-90s. With that, they had a target they could use in clinical trials, and bam – the field just exploded and one treatment after another started popping up.
Peluso compared the situation with long COVID now with the early stages of the HIV epidemic. They have some ideas about mechanistic biomarkers that can demonstrate how severe the illness is but no diagnostic, predictive, or surrogate biomarkers.
Possible mechanistic biomarkers in long COVID:
- A persistent virus – Dr. Maureen Hanson is studying exosomes in ME/CFS and Goetzl and Waltz at UCSF found – apparently to their great surprise – the SARS-CoV-2 coronavirus proteins months after infection in the exosomes of long-COVID patients. People with more severe neurological symptoms had more proteins in their exosomes.
- Inflammation – a 120-person study that used an ultrasensitive assay found that people with long COVID had higher levels of inflammatory markers such as interleukin 6, TNF-alpha and IP-10. The sicker they were, the more markers of inflammation they had. Plus, people with higher levels of inflammation had reduced exercise capacity. As these findings have been replicated in other studies, it’s now becoming clearer that inflammation, as suspected, is playing a role in long COVID.
- Other possible mechanistic biomarkers include autoantibodies, coagulation, cortisol, microbial biomarkers in the gut, and exercise study results.
It took 15 years to find a surrogate marker for HIV, but Peluso believed that all these biomarkers – mechanistic, diagnostic, predictive, and surrogate – will be developed more quickly with long COVID (as well they should given the enormous growth of technology since then.)
He predicted that the studies (remember he’s in the RECOVER Initiative) are going to “really increase in complexity over the next year” and when they do, the biomarkers are going to become clearer.
The big news, though, is that the production of biomarkers – particularly surrogate biomarkers – should unleash an explosion of treatment trials and ultimately, good therapeutics for long COVID and, I suspect anyway, ME/CFS (see Conclusion).
Dr. David Walt – A Persistent Virus (or Not?)
Talk about using cutting-edge technology. Dr. David Walt (an inventor in the US National Inventors Hall of Fame) described the technique he’d developed which allows him to “literally count the number of molecules present”. Say bye-bye to all the background signals that cause chaos in standard Elisa studies, and say hello to being able to detect much more – like 1,000 to 10,000 x more – with this new technique.
He followed 37 people diagnosed with PASC and 26 patients who had come down with COVID. Trying to determine if the virus persisted, they looked for four aspects of the virus (the nucleocapsid portion of the spike protein, the S1 fragment of the spike protein, and the full spike protein) and did ultrasensitive assays of 10 cytokines. None of the patients they followed had been hospitalized.
To their surprise, they found evidence that the spike protein persisted in 65% of the long-COVID patients – but not in the healthy controls. That indicated that a persistent viral reservoir of actively replicating virus must be present in a large number of the long-COVID patients. It also indicated that a viral reservoir was not causing problems in many people with long COVID.
The cytokine assays – as they have often been in ME/CFS – were totally unrevealing; i.e. no difference was found even in the supersensitive assays between the long COVID patients and the controls.
Beta Testers Requested!
Then came the fly in the ointment. They’d just tested 600 post-Omicron patients from the RECOVER study and found much less evidence of the spike protein; i.e. much less evidence of viral persistence! Was it because of the different variants? Or god forbid, did the reagents they used in the two studies have “a slightly different” specificity? They should know more shortly as they’re continuing to test different batches of patients from different groups – and that’s the good news. In ME/CFS, a problem like this might have ended this effort, but they have the resources to figure this out and they will.
Meanwhile, the presence of the spike protein is being used to assess effectiveness in the Paxlovid trial.
Dr. Abdel-Mohsen – The Gut-Brain Connection in Long COVID (and, of course, ME/CFS/FM)
The next presentation asked whether the virus had altered the flora of the gut to such an extent that leaky gut had occurred, that evidence of systemic inflammation was present, and most importantly, that the metabolism had been affected.
We don’t often think of the gut in terms of metabolism, but the gut – with its production of many metabolites – plays a huge role in our metabolism. If crucial metabolites (such as tryptophan) are not being metabolized properly, big problems can result.
Just as in the severe COVID-19 patients, they found evidence (zonulin or tight junction permeability) that the gut walls of the long-COVID patients had become leaky. Increased beta-glucan levels indicated that gut bacteria had made their way into the bloodstream. The fact that people with more gut bacteria in their blood had more symptoms indicated that leaky gut was clearly contributing to long COVID! A correlation analysis suggested that inflammation was higher in these patients as well.
- Symptom assessments just won’t cut it! So said the first presenter in the NIH’s RECOVER Review series. We need biomarkers – and plenty of them: mechanistic, diagnostic, prognostic, and above all “surrogate” biomarkers. (Who knew there were so many different kinds…)
- The HIV/AIDS field struggled for years with a lack of biomarkers. Only when the right biomarker was found did the treatment options for HIV/AIDS explode. Dr. Peluso predicted the same thing will happen for long COVID – it will just happen more quickly.
- Several different “mechanistic” biomarkers that are able to chart the severity of the illness are being investigated including inflammation, autoantibodies, coagulation, cortisol, microbial biomarkers in the gut, and exercise study results.
- Peluso predicted that more complex studies are on the way that will bring us closer to finding biomarkers.
- Dr. Walt – a renowned inventor – has invented a many, many times more sensitive way to find proteins, molecules, cytokines, etc. His early long COVID study found, to his surprise, evidence indicating that an actively replicating virus was still present in many people with long COVID but not in people who had recovered.
- His latest study has, thus far, not found such an incidence of an active virus in post-Omicron patients and it’s not clear why. We should know more about what’s happened soon.
- Dr. Abdel-Mohsen found evidence that a) leaky gut was present, b) that gut bacteria was making its way into the bloodstream, c) that it was causing inflammation, and d) most importantly of all, that it appeared to be interfering with the metabolism.
- It appeared that the coronavirus had interfered with tryptophan metabolism in the gut and gave rise to a host of metabolites that can trigger the NMDA receptor in the brain to produce glutamate which can produce neuroinflammation.
- Simply putting plasma from long COVID patients was enough to trigger inflammation – apparently because of the bacteria present – in cell cultures People with more neurological problems tended to have more symptoms and more gut bacteria in their blood.
- Finally, another study didn’t find evidence of endothelial dysfunction but did find evidence of arterial stiffness in long COVID.
- While the presenters didn’t mention it, with the exception of viral persistence, all these potential biomarkers have been found in ME/CFS/FM. Leaky gut is found, problems with tryptophan metabolism, and arterial stiffness have all been found – some over a decade ago – in these diseases as well.
- The research environment, however, has never been robust enough for these findings to reach “culmination” – to get to a point where they trigger studies that provide definitive results. Given the resources dedicated to long COVID those studies should occur and biomarkers should be found. Given the strong overlap between ME/CFS and long COVID, one has to ask why those biomarkers wouldn’t apply to ME/CFS as well.
Then came the really interesting part. Having increased inflammation was an outcome no one would desire, of course, but things potentially got quite a bit nastier when the researchers considered the metabolic implications of the altered floral bloom the coronavirus had triggered in the long-COVID patients’ guts.
Tryptophan may be an essential amino acid, but if it’s not broken down properly in the gut, bad things can happen. Toxic metabolites such as quinolinic acid that can affect brain functioning can build up. These can activate the NMDA receptor which, in turn, turns on the excitatory amino acid glutamate. (Think “brain on fire”). High glutamate levels can damage neurons and cause inflammation and have been associated with neurological diseases.
Three studies have found at least modestly increased quinolinic acid levels in long COVID, but the NMDA story doesn’t end there. Abdel-Mohsen found increased levels of other metabolites (e.g. S-sulfocysteine) that can turn on the NMDA receptors as well. Plus, the more neurological symptoms a person had, the higher levels of NMDA-associated metabolites they had.
Abdel-Mohsen, then, provided a potentially clear link between a virus-triggered gut flora alteration and neuroinflammation.
A theme showed up during these presentations – multiple large laboratories working together – and so it was with Abdel-Mohsen, who stated:
“we are working with many others in current and future studies to investigate both the diagnostic and prognostic potential as a biomarker or function as a mechanism of this marker of intestinal permeability, microbial translocation, and also host metabolite during PASC”.
It’s this mass of research groups that are working together on long COVID that’s going to make the difference. It’s not, after all, as if tryptophan issues and tryptophan metabolites haven’t shown up in ME/CFS and FM before. Interest in tryptophan in fibromyalgia dates back to 1986! A 1998 paper concluded that “The underlying mechanism (in FM) seems to be defective absorption of the precursor amino acid tryptophan from the gut”. Almost 20 years ago, Blankfield proposed both FM and ME/CFS were tryptophan-kynurenine pathway disorders.
More recent evidence suggests the same. A 2021 study, a 2022 study, a 2021 review, and others have explored problems with tryptophan metabolism in ME/CFS. Tryptophan metabolism is also the key pathway featured in Robert Phair’s Metabolic Trap hypothesis for ME/CFS. You could probably have substituted long COVD for ME/CFS in this paragraph from the 2022 study.
“Changes in tryptophan metabolites… in ME/CFS could both be compatible with anomalies in the sphere of energy metabolism. Overall, clinical traits together with serum biomarkers related to inflammation, intestine function, and tryptophan metabolism deserve to be further considered for the development of personalized medicine strategies for ME/CFS.”
So we see yet again an apparent overlap (leaky gut, tryptophan metabolism) in physiological problems in ME/CFS/FM and long COVID. The problem with ME/CFS/FM has been the scattershot nature of the field. Instead of large laboratories working together to track down the clues, ME/CFS and FM researchers’ efforts generally have consisted of mostly small studies that have slowly been building a case for, in this instance, tryptophan metabolism. They may have been right all along – in fact, this long-COVID study suggests they were – but they never gathered enough oomph to concentrate the field and give us the studies that would provide some definitive answers.
It’s this kind of scattershot, inefficient, and ultimately not very revealing approach that permeates so much disease research that caused the RECOVER Initiative to focus so heavily on building an infrastructure that can methodically collect data and test hypotheses, discard those that don’t work and move with on with increased focus with ones that do.
Dr. Grace McComsey – Blood Vessel Biomarker
Dr. McComsey emphasized that multiple biomarkers associated with different types of long COVID are probably going to show up and that the field needs stable, reproducible biomarkers that have been validated by multiple studies. Then she introduced a surprise.
Her studies did not find problems with endothelial functioning (whoops!) but did find large changes in arterial plasticity or stiffness (nice!). That’s actually a pretty nice finding for fibromyalgia and ME/CFS as multiple studies have found increased arterial stiffness in FM, in particular. Lynn Hodges’s fascinating New Zealand exercise study even found evidence that exercise increases arterial stiffness in ME/CFS. That increase in stiffness could make it difficult to get enough blood to the muscles during the recovery period.
Dr. McComsey then brought things full circle when she reported that several markers of inflammation, as well as markers of gut integrity, were associated with arterial stiffness in long COVID.
On a last note, she reported that people with low vitamin K status ended up with much more severe diseases than those with good vitamin K levels.
All in all, the RECOVER Initiative provided a fascinating, as well as encouraging, review session. Viral persistence is a possibility for long COVID but nothing more than that at the moment. It was the gut and arterial stuff – and the continuing rather astonishing overlap that continues to show up between ME/CFS/FM and long COVID – that really stood out. ME/CFS/FM wasn’t mentioned at all, but it was like the elephant in the room that everyone either didn’t see or was ignoring.
Gut microbiome alterations, leaky gut, and arterial stiffness have been found in all three diseases – and some of the findings in ME/CFS/FM date back decades. The ME/CFS/FM fields have never had the resources, however, to bring those findings into the limelight and get them definitively studied. Long COVID, on the other hand, appears to have the resources to do that.
The inability of the small ME/CFS field to come up with any recognized biomarkers at all over the past 40 years has dramatically stunted the production of therapeutics. The HIV field was in the same fix for years until a surrogate biomarker was found – and the race for therapeutics was on. After that, treatment possibilities exploded.
Dr. Peluso believes the same thing will happen more quickly with long COVID, which brings up the question: “What about ME/CFS, FM, and similar diseases?”. Given the close physiological overlap between ME/CFS/FM and long COVID, my question is: why would a similar biomarker not apply to these diseases as well? Dr. Nath may be right – if you solve one of these mystery diseases, you’ll probably be well on the way to solving the others. Let’s hope so.
BIG (Little) Drive Update
Thirty K is a small number for most organizations but it’s a big number for us and we’ve just about reached it – thanks to the over 350 people who’ve helped keep Health Rising on the web for another year.
Biomarkers have been whispered around the ME/CFS field for decades. Everyone knows we need a biomarker but this blog really laid out why – and explained why we may very well get one – not just for long COVID but possibly for all three diseases. That, in turn, could result in an explosion in therapeutics…
That’s what digging into this stuff revealed – and digging is what Health Rising does. If that enriches you please support us.
HEALTH RISING IS NOT A 501 (c) 3 NON-PROFIT