Leveraging Prior Knowledge of Endocrine Immune Regulation in the Therapeutically Relevant Phenotyping of Women With CFS, 2019, Morris et al

[Andy]

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.

Paywalled at https://www.clinicaltherapeutics.com/article/S0149-2918(19)30112-2/fulltext

 

[Lisa108]

sci-hub link

 

[wigglethemouse]

Paywalled at https://www.clinicaltherapeutics.com/article/S0149-2918(19)30112-2/fulltext

FYI - This paper is not paywalled now.

For everyone's benefit this is a paper describing the exercise testing that Klimas has been doing in ME/CFS, taking blood at 8 timepoints, and measuring hormones and cytokines, and how they fitted a computer model to it. This is the one she has raved about in her videos that requires supercomputers to run.

Some quotes from the paper describing what they did.

Patients and Methods
A total of 88 female subjects (43 with ME/CFS and 45 healthy control subjects) were selected without exclusion for ethnicity from the patient population of the Institute for Neuro Immune Medicine at Nova Southeastern University (Fort Lauderdale, Florida), (directed by N.G.K.).

Study Design
Subjects were challenged with a supervised symptom-limited maximum graded exercise test performed under the McArdle protocol on a fully automated cycle Model 95Ri (Life Fitness, Rosemont, Illinois) and the Oxycon Mobile ergospirometry testing device (Vyaire Medical, Mettawa, Illinois). Subjects pedaled at an initial output of 60 W for 2 min, followed by an increase of 30 W every 2 min. This was continued until one of the following endpoints: (1) maximal oxygen consumption was reached; (2) respiratory exchange ratio was >1.15; or (3) the subject discontinued the challenge. Blood samples (8 mL) were collected before the test after a 30-min rest period, at maximal effort, and at 10, 20, 30, and 60 min' poststress, with additional blood draw at ∼12 h and 24 h’ poststress. Peripheral blood mononuclear cells were isolated by using Ficoll–Paque extraction and stored in liquid nitrogen; plasma was stored at −80C.

Here is what they measured

Peripheral blood mononuclear cell samples from each time point were analyzed by flow cytometry on a Beckman Coulter FC500 (Brea, California) using commercially available antibodies to record frequencies of B cell (CD19+) and natural killer (NK) cell (CD3-CD56+) populations. Plasma concentrations of interferon-γ (IFN-γ), IL-1α, IL-1β, IL-2, IL-4, IL-6, IL-10, IL-13, IL-15, IL-17, IL-23, and TNF-α were measured by using a Q-Plex multiplex ELISA (Quansys Biosciences, Logan, Utah). Details of the protocol and assay variability have been reported previously by our group.3, 29 Finally, serum samples were analyzed for concentrations of the predominant estrogen estradiol and progesterone by immunoelectro-chemiluminescence assays on a Roche Cobas 6000 analyzer (Roche Diagnostics, Basel, Switzerland), following all manufacturer's instructions for instrument maintenance and assay calibration and test procedures with interassay %CVs that are consistently <4%.

They extended their previous 2013/2014 model by using data generated from a machine learning approach. Don't know if you have seen this @mariovitali

Mechanistic Modeling of Endocrine–Immune Signaling
The model assembled and reported previously by our group27 has been extended in the present study to include additional regulators (nodes) of HPG and HPA axis function, as well as regulators of the hypothalamic-pituitary-thyroid axis and a much more detailed description of the immune signaling. Regulatory interactions (edges) between these entities were drawn from the Pathway Studio (Elsevier, Amsterdam, the Netherlands) knowledge database, a repository extracted from the published scientific literature using the MedScan32 natural language processing engine. Edges were verified independently by using our implementation of a Bayesian sentiment analysis classifier.33 Disagreements between MedScan and this platform were reviewed and adjudicated by the authors.

And yes @mariovitali they used machine learning yet again

Regulatory Network Structure and Parameterization
A survey of published literature on elements of the HPA, HPG, and immune systems implicated in ME/CFS identified 28 biological markers, including hormones, neurotransmitters, cytokines, and cell populations, with physiological stress as an input stimulus. Automated text mining of the Pathway Studio literature database and validation of statements about regulatory interactions between these entities identified 214 interactions (Figure 1). The structure of this regulatory circuit model was supported by a total of 21,146 references, with a median of 16.0 references and a mean of 58.9 references supporting each interaction (see Supplemental Figure 1 in the online version

Then I get a bit lost (understatement of the year) so I wanted to look at the data they measured shown in Supplementary Figure A3. From what I can see there is huge overlap between patients and controls, and the mean differences are pretty small...........

Experimental data for measured cytokines and cell populations over the time course of exercise responses, with maximum exertion at T1

They then did their magic and came up with 26 models to explain what is happening and tested two drugs "in silicon" models. You can read the details in the paper (watch that your brain doesn't explode trying to understand )

The question I have is, if you create a model that is selected/chosen using patient data (along with the literature), don't you need a second cohort of new data to test your model against to see that your model is real? Especially when there is so much overlap between healthy and patient experimental data and the mean differences between the data is pretty minimal. Otherwise, how can you trust the model?

EDIT: Posting an image of the experimental data didn't turn out so well. Here is a link to it
https://marlin-prod.literatumonline...9004e-dbda-422a-bd91-c0f92f18c11b/fx3_lrg.jpg

 

[mariovitali]

@wigglethemouse

I will have to read this very carefully , it appears that they use a number of techniques which also includes elements of network analysis (betweeness score) which i also use in my methodology.

The standard way of assessing a machine learning model is to use three subsets of data : Train, Validation and Test set. The Train set is used to have the algorithm(s) "learn" about the problem and you get to tune the algorithm(s) by looking how it performs on the validation set. The actual performance of the classifier though takes place on the test set (recall that the algorithms are being trained and optimized using the train and validation sets so the validation set is not a representative performance evaluation since it is used for the optimization).

As discussed , this paper describes a very complex work but i must say it is interesting because i can confirm the -possible- involvement of IL-6 and TNF-α from the methodology i use. The -possible- involvement of HPA and HPG axis is also there and -interestingly- the tool suggests that cases of ME patients with hypothyroidism may be a cause of this HPA dysregulation. Of course this is a hypothesis and i do not know if this stands medically speaking.

EDIT for correction : The hypothesis suggests that HPA Dysregulation may be the cause of hypothyroidism in some ME patients, not the other way around as it is suggested above.

 

[mariovitali]

Here is an example run , executed just today. Observe how il-6 and tnf-α are being selected along with hypoperfusion, hypoxia and vasoconstriction. Note that these results have to do with the symptoms of ME/CFS and not the cause.




I do not know why angiotensin, albumin are selected as being important (the same for norepinephrine - not shown). ckd = chronic kidney disease. Glutamate -as discussed- is considered a key topic according to Machine Learning AND Network Analysis.

BDNF was recently being mentioned as result of a research effort whose title i do not recall.

Also , important is caloric restriction. What this suggests (as a hypothesis) is that caloric restriction may either have a positive or a negative effect to the symptoms of ME/CFS. Unfortunately we do not know the context suggested (i.e if the certain feature is associated with symptom or non-symptom state)

 

[wigglethemouse]

I do not know why angiotensin, albumin are selected as being important (the same for norepinephrine - not shown). ckd = chronic kidney disease.

This could be why some of them are linked - to do with blood pressure/blood vessel control

Angiotensin is a peptide hormone that causes vasoconstriction and an increase in blood pressure. It is part of the renin–angiotensin system, which regulates blood pressure. Angiotensin also stimulates the release of aldosterone from the adrenal cortex to promote sodium retention by the kidneys.

https://en.wikipedia.org/wiki/Angiotensin

Albumin seems related to electrolyte balance & water in blood control and also blood volume.

Serum albumin is the main protein of human blood plasma.[6] It binds water, cations (such as Ca2+, Na+ and K+), fatty acids, hormones, bilirubin, thyroxine (T4) and pharmaceuticals (including barbiturates): its main function is to regulate the oncotic pressure of blood.

https://en.wikipedia.org/wiki/Albumin

Oncotic pressure, or colloid osmotic pressure, is a form of osmotic pressure induced by proteins, notably albumin, in a blood vessel's plasma (blood/liquid) that displaces water molecules, thus creating a relative water molecule deficit with water molecules moving back into the circulatory system within the lower pressure venous end of capillaries. It has the opposing effect of both hydrostatic blood pressure pushing water and small molecules out of the blood into the interstitial spaces within the arterial end of capillaries and interstitial colloidal osmotic pressure. These interacting factors determine the partition balancing of total body extracellular water between the blood plasma and the larger extracellular water volume outside the blood stream.

https://en.wikipedia.org/wiki/Oncotic_pressure

 

[wastwater]

Cytokines are the subject that always seems to come back and I could understand mecfs as the side effect of cytokines,even if these individual cytokines aren’t very elevated taken as a whole a little bit on each cytokine could add up to a lot

 

[Saz94]

Cort's just published an article about this on Health Rising:

https://www.healthrising.org/blog/2...ENJ57NHigfkBmYoGTMzEmq_0_HoPH3fG-Vm986xY71JEE

 

[Saz94]

It seems like this team's work on hormones in ME/CFS could prove to be extremely important.

But that's just my amateur impression - what do other S4ME members think?

 

[wigglethemouse]

It seems like this team's work on hormones in ME/CFS could prove to be extremely important.

But that's just my amateur impression - what do other S4ME members think?

If you believe their model then yes. But this is the raw data from actual measurements. Did they really only measure 2 hormones and model the rest based on their literature search???? Here is Supplementary figure 5.

Here are some quotes from the paper on ACTUAL measurements :-

In a 2-way ANOVA of estrogen and progesterone measurements over time, we found significant variation in estrogen according to health condition with elevated levels in patients with ME/CFS throughout the exercise response (P ¼ 0.002); t tests at each independent time point consistently showed a marginally significant increase in patients with ME/CFS (P < 0.1) for this hormone. A 2-way ANOVA of progesterone measurements indicated a marginally significant difference in progesterone levels across groups (P ¼ 0.070); however, individual t tests at each independent time point did not support these differences at this level of resolution.

Individual t tests at each time point does not support a conclusion that there is a difference in progesterone. Estrogen was elevated at all time points but again, refer to the actual data - there is a huuuuge overlap of datapoints between the two groups, so while the mean may be different, it is possible that could just be by chance.

It is noteworthy that the complexity of the regulatory model exceeds the coverage supported by the available data. In this case, measurements for important hormones were unavailable, and the range of conditions was limited to exercise challenge of a specific type. This outcome results in a parameter identification problem that is highly underconstrained or in which many model solutions exist that satisfy the data equally well.

This confirms that measurements for important hormones were unavailable..........

Despite this observation, other studies have failed to find significant groupwise changes in sex hormones in ME/CFS.52,53

And others don't notice a difference in their data in ME.

So if I read correctly, hormone differences exist in the model, but they do not have data to confidently back that up yet.