Trial Report Exploring the role of galectin-9 and artemin as biomarkers in long COVID with [CFS]: links to inflammation and cognitive function, 2024, Elahi et al

Discussion in 'Long Covid research' started by Dolphin, Sep 5, 2024.

  1. Dolphin

    Dolphin Senior Member (Voting Rights)

    Messages:
    5,884
    See post #5 for published version

    https://www.frontiersin.org/journals/immunology/articles/10.3389/fimmu.2024.1443363/abstract

    Front. Immunol.
    Sec. Viral Immunology
    Volume 15 - 2024 | doi: 10.3389/fimmu.2024.1443363
    This article is part of the Research TopicImmunological consequences of viral infection on brain homeostasis and cognitive impairmentView all articles
    Discriminatory biomarkers Galectin-9 and Artemin in Long COVID with chronic fatigue syndrome: Correlation with inflammatory and cognitive markers
    Provisionally accepted
    [​IMG] Shokrollah Elahi *[​IMG] Maryam Rezaeifar[​IMG] Mohammed Osman[​IMG] Shima Shahbaz
    • University of Alberta, Edmonton, Canada


    This study aimed to assess plasma galectin-9 (Gal-9) and artemin (ARTN) concentrations as potential biomarkers to differentiate individuals with Long COVID (LC) patients with myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) from SARS-CoV-2 recovered (R) and healthy controls (HCs).

    Receiver operating characteristic (ROC) curve analysis determined a cutoff value of plasma Gal-9 and ARTN to differentiate LC patients from the R group and HCs in two independent cohorts.

    Positive correlations were observed between elevated plasma Gal-9 levels and inflammatory markers (e.g. SAA and IP-10), as well as sCD14 and I-FABP in LC patients.

    Gal-9 also exhibited a positive correlation with cognitive failure scores, suggesting its potential role in cognitive impairment in LC patients with ME/CFS.

    This study highlights plasma Gal-9 and/or ARTN as sensitive screening biomarkers for discriminating LC patients from controls.

    Notably, the elevation of LPS-binding protein in LC patients, as has been observed in HIV infected individuals, suggests microbial translocation.

    However, despite elevated Gal-9, we found a significant decline in ARTN levels in the plasma of people living with HIV (PLWH).Our study provides a novel and important role for Gal-9/ARTN in LC pathogenesis.

    Keywords: Long Covid, chronic fatigue syndrome, galectin-9, artemin, microbial translocation, HIV

    Received: 03 Jun 2024; Accepted: 03 Sep 2024.

     
    Last edited by a moderator: Oct 11, 2024
  2. SNT Gatchaman

    SNT Gatchaman Senior Member (Voting Rights)

    Messages:
    5,923
    Location:
    Aotearoa New Zealand
  3. Andy

    Andy Committee Member

    Messages:
    23,154
    Location:
    Hampshire, UK
    Full paper available from https://www.sciencedirect.com/science/article/pii/S089684112400101X

    "Our ME/CFS patients were selected from a pool of over 2000 patients exhibiting LC symptoms. Through a comprehensive evaluation process that involved clinical assessments, laboratory tests, and the administration of well-defined questionnaires. Specifically, we utilized the de Paul Symptom Questionnaire (PSQ) to identify which patients fulfilled the criteria of ME/CFS; then used the FACIT Fatigue (Version 4) and multidimensional fatigue inventory to identify the severity of fatigue as outlined in the Canadian Consensus criteria (CDC) for ME/CFS and WHO [15,16] as we have previously described [17,18]."
     
  4. Lindberg

    Lindberg Established Member (Voting Rights)

    Messages:
    52
    I find this paper interesting, but wondering if anyone with more knowledge and research experience has any thoughts on it?
     
  5. SNT Gatchaman

    SNT Gatchaman Senior Member (Voting Rights)

    Messages:
    5,923
    Location:
    Aotearoa New Zealand
    Now published as —

    Exploring the role of galectin-9 and artemin as biomarkers in long COVID with chronic fatigue syndrome: links to inflammation and cognitive function
    Elahi, Shokrollah; Rezaeifar, Maryam; Osman, Mohammed; Shahbaz, Shima

    This study aimed to assess plasma galectin-9 (Gal-9) and artemin (ARTN) concentrations as potential biomarkers to differentiate individuals with Long COVID (LC) patients with myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) from SARS-CoV-2 recovered (R) and healthy controls (HCs).

    Receiver operating characteristic (ROC) curve analysis determined a cut-off value of plasma Gal-9 and ARTN to differentiate LC patients from the R group and HCs in two independent cohorts. Positive correlations were observed between elevated plasma Gal-9 levels and inflammatory markers (e.g. SAA and IP-10), as well as sCD14 and I-FABP in LC patients. Gal-9 also exhibited a positive correlation with cognitive failure scores, suggesting its potential role in cognitive impairment in LC patients with ME/CFS.

    This study highlights plasma Gal-9 and/or ARTN as sensitive screening biomarkers for discriminating LC patients from controls. Notably, the elevation of LPS-binding protein in LC patients, as has been observed in HIV infected individuals, suggests microbial translocation. However, despite elevated Gal-9, we found a significant decline in ARTN levels in the plasma of people living with HIV (PLWH). Our study provides a novel and important role for Gal-9/ARTN in LC pathogenesis.


    Link | PDF (Frontiers in Immunology) [Open Access]
     
  6. Hutan

    Hutan Moderator Staff Member

    Messages:
    29,697
    Location:
    Aotearoa New Zealand
    Interesting things to look at
     
  7. Hutan

    Hutan Moderator Staff Member

    Messages:
    29,697
    Location:
    Aotearoa New Zealand
    I think there are questions about the definitions used to identify the LC patients with ME/CFS. For example, reference #40 reported the following and its abstract did not mention PEM

    They report
    Discovery cohort was 44 LC patients and 24 Recovered
    Validating cohort was 34 LC patients and 34 Recovered

    They say the study participants were age and sex matched. They report that all of the LC patients in their study met their ME/CFS requirements; people who did not meet the ME/CFS requirements were not included in the cohorts. So, the 'LC' label here is actually applied to people who have had LC for at least 12 months and who meet their ME/CFS criteria. The 'Recovered' label is applied to people who have been recovered for at least 12 months.

    The study size is adequate, especially with the validating cohort. So, some questions around the ME/CFS definition, but it's looking not too bad.

    They also had 63 people with HIV on anti-retroviral therapy and 25 healthy controls.
     
    Last edited: Oct 11, 2024
  8. Hutan

    Hutan Moderator Staff Member

    Messages:
    29,697
    Location:
    Aotearoa New Zealand
    GAL-9 levels

    Screen Shot 2024-10-12 at 7.50.38 am.png
    A
    so, first they identified a value of Gal-9 that separated the LC-ME/CFS Discovery cohort from the healthy controls (1725 pg/ml). The 97% sensitivity means that nearly every individual with their LC-ME/CFS label was identified by this cutoff. The 100% specificity means that every individual with the HC label was identified as negative.

    B They then compared the LC-ME/CFS Discovery cohort people's levels of Gal-9 with the Recovered Discovery people. They had to use a higher cutoff point (2779 pg/ml) but the sensitivity was still not bad at 78.4% - most of the people with the LC-ME/CFS label were selected (and so had higher Gal-9 levels), while none of the Recovered people were (100% specificity).

    C Encouragingly, things held up fairly well in the Validation cohort. A cut off point similar to the one used in A, identified 82% of the people with LC-ME/CFS and did not identify many of the Recovered people.

    I'm not sure that we could expect a better result if Gal-9 actually turns out to be a marker of ME/CFS, given the uncertainty around diagnosis and the possibility that ME/CFS is a spectrum rather than a binary thing. By that last comment, I mean that it is possible that some of the Recovered people may actually have subclinical levels of whatever ME/CFS is - perhaps their lifestyle just accommodates a small amount of impairment.

    It is helpful for the story holding together that higher levels of Gal-9 seem to be associated with conditions with perturbed immune systems.
     
    Last edited: Oct 11, 2024
    Turtle, Peter Trewhitt and forestglip like this.
  9. Hutan

    Hutan Moderator Staff Member

    Messages:
    29,697
    Location:
    Aotearoa New Zealand
    Screen Shot 2024-10-12 at 8.19.21 am.png

    So, top row Discovery cohort; bottom row validating cohort.
    SAA is serum amyloid A, charts on the left. IP-10 charts on the right.

    R2 aren't amazing, but the relationships look quite convincing. But, look at the scales. What is going on there? In chart D, the LC discovery cohort has levels up to 40 million pg/mL, while all the values in F (the validating cohort) are below 100,000.

    A paper on inflammatory rheumatic disease says:
    3 mg/L
    1 milligram/litre = 1 million picograms/millilitre
    So, we would expect healthy people to have blood levels of less than 3 million picograms/millilitre. And so, in D, where the y axis has numbers involving 10's of millions, there does seem to be abnormal levels of SAA, although mostly less than a 10-fold increase. I don't know what is going on with chart F, I assume it's a presentation error.
     
    Turtle, Peter Trewhitt and forestglip like this.
  10. Hutan

    Hutan Moderator Staff Member

    Messages:
    29,697
    Location:
    Aotearoa New Zealand
    It's interesting that serum amyloid A was found to be associated with worse fatigue and joint pain in a small study of Gulf War Illness.
    And it has been mentioned in relation to the microclots, I think the proponents of the microclot theory have found increased SAA in Long Covid.
     
    Peter Trewhitt and forestglip like this.
  11. forestglip

    forestglip Senior Member (Voting Rights)

    Messages:
    1,007
    I looked at their previous paper, 29.
    So it seems every ME/CFS participant met category II, PEM, in both papers.
     
    Turtle, Peter Trewhitt and Hutan like this.
  12. Hutan

    Hutan Moderator Staff Member

    Messages:
    29,697
    Location:
    Aotearoa New Zealand
    Super, thank you forestglip
     
    Peter Trewhitt and forestglip like this.
  13. forestglip

    forestglip Senior Member (Voting Rights)

    Messages:
    1,007
    I love correlations with actual symptom metrics:
    upload_2024-10-11_18-6-4.png

    It's strange that they don't explain what exactly they mean by cognitive failure in this paper, but they do mention it in the context of their previous paper:
    The Cognitive Failures Questionnaire (CFQ) and its correlates, 1982, Broadbent et al.
    upload_2024-10-11_18-19-45.png
     
    Peter Trewhitt and Hutan like this.
  14. forestglip

    forestglip Senior Member (Voting Rights)

    Messages:
    1,007
    I asked Claude to come up with all the reasons it could think of that could explain the correlation between Gal-9 and cognitive failure:
    The authors' suggested explanation:
    64 doesn't seem like the right citation for the HIV claim. I think it's 37:
    upload_2024-10-11_19-2-0.png

    And 34, but I can't access the full text:
     
    Last edited: Oct 12, 2024
  15. Hutan

    Hutan Moderator Staff Member

    Messages:
    29,697
    Location:
    Aotearoa New Zealand
    This is from a Canadian team, specifically from Edmonton.
    I've sent them an invitation to join us here, and a query about the y axis scale in Figure 1F.
     
  16. forestglip

    forestglip Senior Member (Voting Rights)

    Messages:
    1,007
    In reference 34, the two largest Gal-9 - cognitive correlations in HIV were with HVLT-R and digit symbol tests.

    HVLT-R
    Digit symbol
    Screenshot_20241011-202942.png
     
  17. Lindberg

    Lindberg Established Member (Voting Rights)

    Messages:
    52
    I tried to take the DSST test that @forestglip posted above. During the course of the test it took a longer time to react and draw the symbols, but certain symbols were more difficult than others. For instance number six. It’s like my mind freezes. It knows what to do, but it can’t decide on which way to go so it just freezes. I get trouble moving my eyes and my hand and it takes an unusually high amount of effort to get them moving again. All in all these two rows of numbers and symbols took me three minutes to complete. Anyone else having the same kind of trouble?
     
    Last edited: Oct 12, 2024
    forestglip and Peter Trewhitt like this.
  18. Peter Trewhitt

    Peter Trewhitt Senior Member (Voting Rights)

    Messages:
    4,151
    I just tried it using my finger to out line the shapes. For about five I thought this is easy, then froze having forgotten what I was meant to do. Mentally stopping and starting again worked until after the next five when I froze again.

    This is interesting and I wonder if it might be happening in other cognitive tasks. For example writing comments here I regularly have to stop recollect what I am trying to say and start again, which could be why my grammar falls apart when after a stop I recollect the intent of my post but not the specific language I had planned.
     
    forestglip and Lindberg like this.
  19. Lindberg

    Lindberg Established Member (Voting Rights)

    Messages:
    52
    Interesting! I found the symbols number 5 and 6 to be the most difficult ones, that’s when my mind froze. But you are right in that it might be the repetition in itself (after approximately five tasks) that the mind freezes. Perhaps also a combination of repetition “tiredness” and the type of symbol (cognitive task). I’ve noticed before that I can do certain types of cognitive tasks easily, like finding a linear pattern. But, whenever I have to choose between two directions my mind freezes and it’s very difficult to force it to act.
     
    Last edited: Oct 12, 2024
    forestglip and Peter Trewhitt like this.
  20. Peter Trewhitt

    Peter Trewhitt Senior Member (Voting Rights)

    Messages:
    4,151
    This is probably a tenuous link, but I can experience what I term ‘jamais vu’ particularly when in PEM. It used to happen more frequently when I was still working and had a daily commute, when I would suddenly have no idea where I was on a road I knew well and would have to wait until cognitive mapping kicked back in. When my ME is at its worst it can even happen in the house, but as there are only a few doors on the landing I can bypass the problem by looking through each til I find the bathroom or my bedroom.
     
    forestglip and Lindberg like this.

Share This Page