Steroid dynamics in myalgic encephalomyelitis / chronic fatigue syndrome: a case-control study [...], 2025, Thomas, Armstrong, Bergquist et al

Could you guys be thinking of this Klimas study

Leveraging Prior Knowledge of Endocrine Immune Regulation in the Therapeutically Relevant Phenotyping of Women With Chronic Fatigue Syndrome
Abstract
Purpose

The complex and varied presentation of myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) has made it difficult to diagnose, study, and treat. Its symptoms and likely etiology involve multiple components of endocrine and immune regulation, including the hypothalamic-pituitary-adrenal axis, the hypothalamic-pituitary-gonadal axis, and their interactive oversight of immune function. We propose that the persistence of ME/CFS may involve changes in the regulatory interactions across these physiological axes. We also propose that the robustness of this new pathogenic equilibrium may at least in part explain the limited success of conventional single-target therapies.

Methods
A comprehensive model was constructed of female endocrine–immune signaling consisting of 28 markers linked by 214 documented regulatory interactions. This detailed model was then constrained to adhere to experimental measurements in a subset of 17 candidate immune markers measured in peripheral blood of patients with ME/CFS and healthy control subjects before, during, and after a maximal exercise challenge. A set of 26 competing numerical models satisfied these data to within 5% error.

Findings
Mechanistically informed predictions of endocrine and immune markers that were either unmeasured or exhibited high subject-to-subject variability pointed to possible context-specific overexpression in ME/CFS at rest of corticotropin-releasing hormone, chemokine (C-X-C motif) ligand 8, estrogen, follicle-stimulating hormone (FSH), gonadotropin-releasing hormone 1, interleukin (IL)-23, and luteinizing hormone, and underexpression of adrenocorticotropic hormone, cortisol, interferon-γ, IL-10, IL-17, and IL-1α. Simulations of rintatolimod and rituximab treatment predicted a shift in the repertoire of available endocrine–immune regulatory regimens. Rintatolimod was predicted to make available substantial remission in a significant subset of subjects, in particular those with low levels of IL-1α, IL-17, and cortisol; intermediate levels of progesterone and FSH; and high estrogen levels. Rituximab treatment was predicted to support partial remission in a smaller subset of patients with ME/CFS, specifically those with low norepinephrine, IL-1α, chemokine (C-X-C motif) ligand 8, and cortisol levels; intermediate FSH and gonadotropin-releasing hormone 1 levels; and elevated expression of tumor necrosis factor-α, luteinizing hormone, IL-12, and B-cell activation.

Implications
Applying a rigorous filter of known signaling mechanisms to experimentally measured immune marker expression in ME/CFS has highlighted potential new context-specific markers of illness. These novel endocrine and immune markers may offer useful candidates in delineating new subtypes of ME/CFS and may inform on refinements to the inclusion criteria and instrumentation of new and ongoing trials involving rintatolimod and rituximab treatment protocols.
We have a thread here
 
I just can't agree with your objections to the introduction. The writing looks to me like the standard ways of describing disease severity, what we know about disease physiology and drawing on respected authorities like WHO. I can't see saying that the ICD-10, even on a little known disease like ours, isn't evidence based. I don't think this intro should draw more ire than the intros that band on about ME/CFS being "unexplained", subtext psychosomatic. And I think it would be better to focus on the results.
The reason the problems in the introduction are worth discussing is that it is in a paper on research supported by OMF. Scientists funded by OMF and with plenty of experience in ME/CFS should be getting the introduction right by now. This introduction is a long way from 'right', and so many of the introductions we see from OMF funded scientists are poor.

I don't think making a comparison with introductions that 'bang on about ME/CFS' being psychosomatic is setting the bar at the right level. Yes, of course the introduction in this paper is better than one that suggests that ME/CFS is psychosomatic. It doesn't mean that it can't be more accurate and more persuasive. The ME/CFS community is not funding the psychosomatic proponents, it is contributing to the funding of Bergquist and his team.

It seems, wabi-sabi, that you feel that you are not qualified to evaluate what these scientists have written. I think you are capable of evaluating at least some of what they have written if you take the time to give some attention to it.

If you keep reading the critiques of research here, I think you will come to realise that any scientist can get something wrong, and many of the ME/CFS researchers held in high esteem by many in the ME/CFS community get many things wrong. And that motivated people with ME/CFS who take the time to read these papers can make worthwhile observations.


They may seem silly to you (and hey, maybe they are), but people in healthcare take them seriously. If we are going to be understood, we need to speak the language they understand. Papers are meant to start off with the seriousness of the illness to grab the reader's attention. Typically you do that either with numbers of affected or severity of illness. It's a legit tactic for the authors to use. It really, really doesn't make us look hysterical.
How many 'people in healthcare' have you tried to tell 'ME/CFS is worse than cancer and stroke'? I disagree with 'it really, really doesn't make us seem hysterical'.

There are many ways to give an impression of severity that would work better. I gave an example of one that would not give a knee-jerk impression of over-claiming among readers familiar with the awfulness of significant cancers and stroke.

We routinely see ME/CFS papers on biochemistry devote too many paragraphs to trying to justify ME/CFS as a problem worthy of consideration. Working too hard to convince readers that the disease is worthy of attention can actually give the opposite impression, as the introduction in this paper does. Papers on diseases like multiple sclerosis don't usually feel the need to say that it is coded by the WHO as a disease of the nervous system.


However, I can't focus on the results since I don't have the expertise to evaluate them properly. But I am glad the authors are doing this research, even if it is awfully preliminary. I feel like there's some biting the hand that feeds us going on here.
You seem to feel that you have the expertise to dismiss what I am saying and suggest that I am biting the hand that feeds us though. I'm assuming that you, like the authors, understand how fundamentally hormone levels change at different stages of the menstrual cycle, and with hormonal contraceptive use and between females with regular menstrual cycles and without. It only requires a very basic understanding of statistics to understand how even quite minor differences in the life stage and contraceptive use of the participants in the small samples could result in different hormone ratios between the groups.

If you want to compare people with ME/CFS to pets, dependent on their owners to feed them, let us then extend the metaphor. What should those pets do if what they are fed, instead of something sustaining, is as nutritious as sawdust? The pet can appreciate the intent of its owner to care, be deeply grateful even, but, at some point if it is to survive, it becomes necessary to note that things need to be better. There's no biting going on here, just a pointing out that this paper doesn't help and a request for things to be better.



The authors acknowledge all of the menstrual cycle and birth control and menopause stuff you bring up, so they do understand that it is an issue and are not accidentally ignoring it.
It's great that the authors acknowledge the issues, but they have still made claims as if these problems do not make their findings probably worthless and at best highly questionable. Here are the abstract's conclusions.
abstract said:
Despite no significant differences in absolute steroid levels, network analysis revealed profound disruptions in steroid-steroid relationships in ME/CFS compared to controls, suggesting disrupted steroid homeostasis.​
Collectively the results suggest dysregulation of HPA axis function and progestogen pathways, as demonstrated by altered partial correlations, centrality profiles, and steroid ratios.​
These findings illustrate the importance of hormone network dynamics in ME/CFS pathophysiology and underscores the need for more research into steroid metabolism.​

See how definitive those statements are? But, different percentages of people on hormonal contraceptive and different percentages of people at different points in their menstrual cycle will have an enormous impact on ratios of hormones found. These factors have such a big impact that it means that it is not reasonable to draw any conclusions from this data. Especially when the conclusions have a lot to do with progesterone.

This paper muddies the waters, contributing to misinformation rather than moving the field of ME/CFS forward. It's impossible to know what, if anything, is a finding with any relevance to ME/CFS.

Great that the authors do understand that the problem is an issue and that they aren't accidentally ignoring it. But, that means that, when it comes to them making the conclusions in their abstract, they have deliberately ignored the problem. And that is arguably worse.


"Research on cortisol, one of the most studied steroid hormones and potential biomarkers in ME/CFS, generally supports the presentation of mild hypocortisolism inpatients [37] which is in contrast with the current findings.
Research on cortisol does not support the presentation of 'mild hypocortisolism'. I think it's probably accurate to suggest that mean cortisol levels are often found to be slightly lower in ME/CFS than in the control groups, but the means are still almost always found to be within normal ranges, and there is range of cortisol levels among the patient cohort. A slightly reduced cortisol level is to be expected in people not doing vigorous physical activity.

It's just not correct to claim that mild hypocortisolism (i.e. cortisol slightly below normal ranges) is a characteristic of ME/CFS - the literature does not support it. It's particularly disappointing, because I directed Chris Armstrong to our thread discussing the problem with the claims around cortisol in ME/CFS a while back. @MelbME
 
Last edited:
Thank you for the analysis of the paper on that thread. I could not have asked for a better example of what I see as the questionable research that has come from the Klimas team.

That study used no measurements of hormones in people with ME/CFS. It took some measurements of cytokines and some relationships between cytokines and hormones in general literature to build 26 possible models of what is happening in ME/CFS. It then used those models to predict who might benefit from rituximab and rintolimod treatments.
 
I think you will come to realise that any scientist can get something wrong
my guess is that most of us are constantly wrong and that a key difference maker lies in how receptive to argument and vigilant against bias one is

this is to say: don't shy away from papers unless it's for a health reason! In theory everybody should be capable of getting at least the gist of most papers, it's part of our job to make them readable by a non expert.
 
Last edited:
Seems like this paper wasn't well received. It was a small study, not a world changer. No indication of low or high steroids in ME/CFS patients. Limitations as the data was older (as pointed out by Wiggle).

Something is unusual about the steroid pathway. All these steroids are governed by a handful of enzymes and their relationship to each other should be strong as observed in healthy. This is not the case in ME/CFS cohort. That was the main finding from the study. It seemed like it was worth publishing this information.
 
Something is unusual about the steroid pathway. All these steroids are governed by a handful of enzymes and their relationship to each other should be strong as observed in healthy. This is not the case in ME/CFS cohort.
But how do you know that, given you don't know crucial things about the participants, things that have major impacts on steroid levels?

That's the point I and others have been making. You can't know that claimed main finding is true. And, in that case, how is it useful to publish something confidently making a claim based on inadequate evidence? We've had too many myths get created and sustained about ME/CFS, based on poor evidence.
 
But how do you know that, given you don't know crucial things about the participants, things that have major impacts on steroid levels?

That's the point I and others have been making. You can't know that claimed main finding is true. And, in that case, how is it useful to publish something confidently making a claim based on inadequate evidence? We've had too many myths get created and sustained about ME/CFS, based on poor evidence.

Are you asking if this is apparent in all ME/CFS patients? When I say "ME/CFS cohort", I'm talking about the cohort in the study. No, we didn't look at this in thousands to get a sufficient scale to be widely representative of most people with ME/CFS. We have a separate study we've been trying to publish for a year that highlights HPA to cortisol as important in ME.

The point of relationships in-person is that regardless of the time of day, point in life, etc they should be maintained. 3-4 steroids of the biosynthesis pathway use the same enzyme to be created at the same time.

The point of publishing is to let people know what we looked at and what we found. The useful part here was the analytical workflow, to point out to other scientists that looking at the full steroid pathway allows you to look at relationships within the steroid pathway, and something here is off in the one ME/CFS cohort we've ever looked at. Perhaps others have this data and will now think to look at it in their data and they can publish corroborating or contradicting findings.

I understand your concerns about people in this area getting the wrong idea. But if they are actually working in this space then chances are they understand what we did and why. They are the people we have in mind when communicating at this level. The alternative is sitting on information and never sending that message.

Yes myths get created and used by random companies trying to sell vitamins or claiming a treatment, etc. That happens unfortunately. I don't know what to do about that though. I'd be keen to see this information not misused.

It's also helpful to publish what you did in science, it marks what you spend your time on, which is the only proof one has of productivity when applying for governmental grants. We're trying to build to get more governmental funding for ME/CFS research.
 
Last edited:
The point of relationships in-person is that regardless of the time of day, point in life, etc they should be maintained. 3-4 steroids of the biosynthesis pathway use the same enzyme to be created at the same time.
So the relationships should be present in every sample taken at any timepoint because they are produced at the same time? Do they have the same half-life (or the equivalent) as well? Are they constantly produced at the same rate and always equally dispersed in the entire body?

I know nothing about this, so I might be asking dumb questions.
 
So the relationships should be present in every sample taken at any timepoint because they are produced at the same time? Do they have the same half-life (or the equivalent) as well? Are they constantly produced at the same rate and always equally dispersed in the entire body?

I know nothing about this, so I might be asking dumb questions.

Here is a link to a diagram that shows the pathway, in blue are the enzymes:

Note how the same enzyme is used for the production of different steroids. It's like a crosshatch of a few enzymes allows you to produce many more steroids. It means that you would expect statistically significant relationships between the steroids produced of the same enzyme, which is what we see so clearly in control cohort.

They are produced at different rates and not dispersed all the same. It's possible to break the relationships, which might be what is happening in the ME cohort.
 
Last edited:
View attachment 27512
@MelbME just wanted to quickly draw attention to figure 1--is it possible that this is the wrong image? The references to Fig 1 in the text and the figure legend are both discordant to what the image is actually showing.

Yes, we're in contact with the journal about it, the figure order is all messed up and unaligned with the legends, the figure number and legend are correct but all the images are on the wrong spots. They messed it up initially, we fixed this issue in proofs and they appear to have published the messed up version.

Pour out a sip of coffee for the hours wasted dealing with the publication process
 
Last edited:
Here is a link to a diagram that shows the pathway, in blue are the enzymes:

Note how the same enzyme is used for the production of different steroids. It's like a crosshatch of a few enzymes allows you to produce many more steroids. It means that you would expect statistically significant relationships between the steroids produced of the same enzyme, which is what we see so clearly in control cohort.
Thank you for linking the figure.

I understand that you’d expect constant ratios of output from the production process. But aren’t there loads of things that might happen between the production and whatever ends up in a blood sample? Other things than these from the abstract?
Abstract said:
Collectively the results suggest dysregulation of HPA axis function and progestogen pathways, as demonstrated by altered partial correlations, centrality profiles, and steroid ratios.

These findings illustrate the importance of hormone network dynamics in ME/CFS pathophysiology and underscores the need for more research into steroid metabolism.
 
For what's it's worth, "something is weird in the steroid hormone biosynthesis pathway" seems to be corroborated in both my previous LC study looking at metrics of post-COVID physical function in the IMPACC cohort and in the comprehensive metabolomics paper by Maureen Hanson's group. Nearly the exact same list of androgenic steroids were downregulated in our study and the Hanson study--all metabolites downstream of pregnenolone. I actually spent a while trying to dig into the ratios between metabolites and possible interpretations, but unfortunately we were already facing limitaitons on space in the manuscript so it's likely to all be left out of the final version.

My interpretation (which was included in the preprint, so I can discuss it here) was that the conversion of cholesterol to pregnenolone seemed to be the step with the biggest discordance between LC and recovered ppts. There was significant variability thereafter, though our cohort had a big enough sample size that we did see statistically significant decreases in all the measured androgenic steroid metabolites downstream of pregnenolone in LC. And this trend had consistent directionality between males and females in a subset analysis, so it is not likely to be driven by contraceptive use. A caveat is that we did not do a targeted screen like in this paper, so we were only able to confidently detect a handful of metabolites from that pathway, and only the sulfated conjugates.

To @Utsikt's question--you can generally expect a partial correlation, I agree with @MelbME there. The fine details do tend to get hairy, though, since every diagram of this pathway is oversimplifying everything to a massive degree. There are a number of alternative pathways--and a lot of enzymes that usually aren't listed--for example, androstenediol doesn't only come from DHEA like the chart linked by MelbME might imply.

Even people with congenital disorders that render certain important enzymes defunct are still able to produce some level of each of those hormones by some combination of "backdoors." All this is to say that I wouldn't trust ratios of metabolites to tell me the exact culprit for which enzyme isn't doing it's job (with one or two notable exceptions), but it can provide insight that something in the whole system is not working as expected.

This particularly study is underpowered, I agree with Hutan on that. Though viewed in light of other findings, it does provide additional corroboration that something odd is happening here.
 
Thank you for linking the figure.

I understand that you’d expect constant ratios of output from the production process. But aren’t there loads of things that might happen between the production and whatever ends up in a blood sample? Other things than these from the abstract?

Steroids are lipophilic and the tissues that generate them are highly vascularised. They enter blood stream very rapidly.

There are potentially other less likely reasons than we point to in abstract. I think our use of the word dysregulation could have been better. Perhaps "altered regulation compared to controls" is more accurate.
 
For what's it's worth, "something is weird in the steroid hormone biosynthesis pathway" seems to be corroborated in both my previous LC study looking at metrics of post-COVID physical function in the IMPACC cohort and in the comprehensive metabolomics paper by Maureen Hanson's group. Nearly the exact same list of androgenic steroids were downregulated in our study and the Hanson study--all metabolites downstream of pregnenolone. I actually spent a while trying to dig into the ratios between metabolites and possible interpretations, but unfortunately we were already facing limitaitons on space in the manuscript so it's likely to all be left out of the final version.

My interpretation (which was included in the preprint, so I can discuss it here) was that the conversion of cholesterol to pregnenolone seemed to be the step with the biggest discordance between LC and recovered ppts. There was significant variability thereafter, though our cohort had a big enough sample size that we did see statistically significant decreases in all the measured androgenic steroid metabolites downstream of pregnenolone in LC. And this trend had consistent directionality between males and females in a subset analysis, so it is not likely to be driven by contraceptive use. A caveat is that we did not do a targeted screen like in this paper, so we were only able to confidently detect a handful of metabolites from that pathway, and only the sulfated conjugates.

To @Utsikt's question--you can generally expect a partial correlation, I agree with @MelbME there. The fine details do tend to get hairy, though, since every diagram of this pathway is oversimplifying everything to a massive degree. There are a number of alternative pathways--and a lot of enzymes that usually aren't listed--for example, androstenediol doesn't only come from DHEA like the chart linked by MelbME might imply.

Even people with congenital disorders that render certain important enzymes defunct are still able to produce some level of each of those hormones by some combination of "backdoors." All this is to say that I wouldn't trust ratios of metabolites to tell me the exact culprit for which enzyme isn't doing it's job (with one or two notable exceptions), but it can provide insight that something in the whole system is not working as expected.

This particularly study is underpowered, I agree with Hutan on that. Though viewed in light of other findings, it does provide additional corroboration that something odd is happening here.

Yes underpowered, we highlight the low sample size in the manuscript. I think our limitations section was quite lengthy from memory. We did a power analysis and I think it was 35-40 in each group that would have identified a significant decrease.

Yeah, we haven't pinned down exactly what is off with this pathway. Its not the same for each person with ME/CFS or LC. More detailed in-person studies are needed.
 
Steroids are lipophilic and the tissues that generate them are highly vascularised. They enter blood stream very rapidly.
So you’d expect a random blood sample to be highly correlated to the production? Would it be possible that the steroids are consumed differently, so the differences are not related to the production?
There are potentially other less likely reasons than we point to in abstract. I think our use of the word dysregulation could have been better. Perhaps "altered regulation compared to controls" is more accurate.
If I’ve understood the concerns correctly, I think the issue is more that one interpretation of the data is highlighted in the abstract when there are other plausible (but maybe less likely in your eyes) explanations - and when there are so substantial limitations like old samples and very small groups:
Collectively the results suggest dysregulation of HPA axis function and progestogen pathways, as demonstrated by altered partial correlations, centrality profiles, and steroid ratios.
The last part is also a too strong to my mind. To say that this illustrates something’s role in the pathophysiology seems to be going beyond the evidence. There’s not even any «might», «suggests», etc.:
These findings illustrate the importance of hormone network dynamics in ME/CFS pathophysiology and underscores the need for more research into steroid metabolism.
 
Yeah, we haven't pinned down exactly what is off with this pathway. Its not the same for each person with ME/CFS or LC. More detailed in-person studies are needed.
We observed a lot of intra-group variability re: exact metabolites too. If it is some signaling molecule inducing differences in the transcription of these enzymes or interfering with their function, I'd also expect it to be highly influenced by how much pwME have exerted themselves prior to giving the sample, which seems like a logistic nightmare to try to control for.

I'd definitely be interested to see further investigation on this (partially for selfish reasons, since some of the enzymes might be relevant gene targets for a particular working hypothesis of mine). Best of luck on future studies!
 
So you’d expect a random blood sample to be highly correlated to the production? Would it be possible that the steroids are consumed differently, so the differences are not related to the production?

If I’ve understood the concerns correctly, I think the issue is more that one interpretation of the data is highlighted in the abstract when there are other plausible (but maybe less likely in your eyes) explanations - and when there are so substantial limitations like old samples and very small groups:

The last part is also a too strong to my mind. To say that this illustrates something’s role in the pathophysiology seems to be going beyond the evidence. There’s not even any «might», «suggests», etc.:

In a healthy cohort you'd expect correlations to exist yes. There are many potential reasons, we landed on the most likely but not to say you should rule out any reason for what we saw.

Yes degradations and half-life do vary. Production will vary. You still expect a relationship across a cohort.

The samples weren't old, the data was old. Data was collected on fresh samples at the time. Data had been sitting around but luckily it doesn't degrade. I think 25 vs 25 is on smaller side but still reasonable to publish. I don't not what scale is being used here for very small, we may have different opinions. What would you call 12 vs 12? Or 6 vs 6?

The last paragraph of abstract may be not clear, that was just meant to highlight that more research in to steroid dynamics is needed.
 
Are you asking if this is apparent in all ME/CFS patients? When I say "ME/CFS cohort", I'm talking about the cohort in the study.
I think you may have missed my point. I thought I was clear. I am saying that the researchers here did not gather crucial information about the participants in this study e.g. menopausal, having regular periods, taking hormonal contraceptives. They did not ensure that samples were taken at the same time in the menstrual cycle.

The point of relationships in-person is that regardless of the time of day, point in life, etc they should be maintained. 3-4 steroids of the biosynthesis pathway use the same enzyme to be created at the same time.
I don't think that is true. We know that oral contraceptives fundamentally change hormonal levels. They do that because they have to change an aspect of the way the female body works, to make it not fertile.

We know that hormonal levels vary greatly during the menstrual cycle.

So, if in one sample you have 50% of the females taking oral contraceptives, and 30% were sampled during their period, and, in the other sample, only 10% of the females were taking oral contraceptives and only 10% were sampled during their period, then the average ratio of various hormones for the cohorts will be substantially different.

AI said:
Oral contraceptives (OCs) primarily work by influencing the ratios of estrogen and progestin hormones, disrupting the normal menstrual cycle to prevent ovulation. In a typical combined oral contraceptive, the ratio of progestin to estrogen is significantly higher than what is naturally found in a woman's body during the ovulatory cycle. This altered ratio, along with the synthetic nature of the hormones, prevents the natural hormonal fluctuations that trigger ovulation.



It's also helpful to publish what you did in science, it marks what you spend your time on, which is the only proof one has of productivity when applying for governmental grants. We're trying to build to get more governmental funding for ME/CFS research.
I do understand this, and I am sure that securing more funds was a major motivation for publishing the study. And, that is a big part of my concern here.

The researchers have said 'Hey, here is a problem! Please give us money to study it further.' And yet, you have not established that there really is a problem, because the participants were not adequately characterised. You have drawn conclusions that may or may not be true. We cannot know if the finding is true at this point. This study's abstract makes the case for studying hormones in ME/CFS more compelling, and may result in other studies that perhaps are more likely to bear fruit being delayed.
 
Last edited:
The samples weren't old, the data was old.
My bad, I misremembered. Thanks for correcting it.
I think 25 vs 25 is small, I don't not what scale is being used here for very small, we may have different opinions. What would you call 12 vs 12? Or 6 vs 6?
Fair enough. Let’s go for small for this one, I’m not in a position to give exact cut-offs.
The last paragraph of abstract may be not clear, that was just meant to highlight that more research in to steroid dynamics is needed.
The issue isn’t that it isn’t clear, but that it goes beyond the evidence. You could have just said that more research is needed to further explore this, without saying that the findings shows that this in particular is a part of ME/CFS. (Ignoring that «more research is needed» is redundant even though I recognise that it has become the norm for non-scientific reasons).
 
Back
Top Bottom