Blood DNA methylation in post-acute sequelae of COVID-19 PASC: a prospective cohort study, 2024, Balnis et al.

[SNT Gatchaman]

Blood DNA methylation in post-acute sequelae of COVID-19 PASC: a prospective cohort study
Joseph Balnis; Andy Madrid; Lisa A. Drake; Rachel Vancavage; Anupama Tiwari; Vraj J. Patel; Ramon Bossardi Ramos; John J. Schwarz; Recai Yucel; Harold A. Singer; Reid S. Alisch; Ariel Jaitovich

BACKGROUND
DNA methylation integrates environmental signals with transcriptional programs. COVID-19 infection induces changes in the host methylome. While post-acute sequelae of COVID-19 (PASC) is a long-term complication of acute illness, its association with DNA methylation is unknown. No universal blood marker of PASC, superseding single organ dysfunctions, has yet been identified.

METHODS
In this single centre prospective cohort study, PASC, post-COVID without PASC, and healthy participants were enrolled to investigate their symptoms association with peripheral blood DNA methylation data generated with state-of-the-art whole genome sequencing. PASC-induced quality-of-life deterioration was scored with a validated instrument, SF-36. Analyses were conducted to identify potential functional roles of differentially methylated loci, and machine learning algorithms were used to resolve PASC severity.

FINDINGS
103 patients with PASC (22.3% male, 77.7% female), 15 patients with previous COVID-19 infection but no PASC (40.0% male, 60.0% female), and 27 healthy volunteers (48.1% male, 51.9% female) were enrolled. Whole genome methylation sequencing revealed 39 differentially methylated regions (DMRs) specific to PASC, each harbouring an average of 15 consecutive positions, that differentiate patients with PASC from the two control groups. Motif analyses of PASC-regulated DMRs identify binding domains for transcription factors regulating circadian rhythm and others. Some DMRs annotated to protein coding genes were associated with changes of RNA expression. Machine learning support vector algorithm and random forest hierarchical clustering reveal 28 unique differentially methylated positions (DMPs) in the genome discriminating patients with better and worse quality of life.

INTERPRETATION
Blood DNA methylation levels identify PASC, stratify PASC severity, and suggest that DNA motifs are targeted by circadian rhythm-regulating pathways in PASC.

FUNDING
This project has been funded by the following agencies: NIH-AI173035 (A. Jaitovich and R. Alisch); and NIH-AG066179 (R. Alisch).


Link | PDF (Lancet: eBioMedicine) [Open Access]

 

[Hutan]

103 patients with PASC (22.3% male, 77.7% female), 15 patients with previous COVID-19 infection but no PASC (40.0% male, 60.0% female), and 27 healthy volunteers (48.1% male, 51.9% female) were enrolled.

I don't know of course, but this study seems really underpowered, given the numbers, especially of the post-Covid-19 controls, and the variable percentages of males in the groups. I would have thought that epigenetics would have a huge amount of noise, with sex right up there as a source of that. Circadian rhythm is mentioned as something regulated by identified transcription regulators, but surely it's possible that, on-average, different lifestyle factors (not working> waking later; more physical activity challenges) might alter the transcription regulators?

 

[Hutan]

A Lancet paper - New York authors

PASC involves many chronic deficits that undermine individuals’ quality of life, reinsertion to work, and other relevant outcomes 4, 5, 6 Given the diversity of PASC-associated deficits, this entity has been extraordinarily difficult to characterize accurately. 7,8 Indeed, biomarkers of PASC are nonspecific or relevant to single-organ dysfunctions. 9 There are currently no global PASC biomarkers encompassing multi-organ deficits. The identification of global PASC molecular signatures could facilitate diagnosis and severity scoring; and provide important insights into the process pathophysiology.

The authors don't seem to be understanding that PASC is a collection of different things. It's not looking good for a homogenous sample.

Self-referred adult patients coming to the Outpatient Pulmonary Clinic at Albany Medical Center in Albany, NY complaining of symptoms of PASC were considered for enrolment. While the clinic operated at the Pulmonary Medicine outpatient's facility, we offered global and comprehensive clinical evaluation to patients who sought advice regarding diverse symptoms associated with PASC, even those exceeding pulmonary conditions. Definition of PASC was ongoing, relapsing, or new symptoms or conditions present 30 or more days after COVID infection.30

Nope.

Whole genome methylation sequencing (WGMS), which interrogates all possible CpG dinucleotides in each patient (>25 million unique CpGs), provides a substantially higher resolution (∼30-fold) to capture persistent COVID-19-induced DNA methylation changes in comparison with microarray technologies. 22 We postulate that the use of WGMS to quantify DNA methylation levels would have a greater chance of identifying specific CpGs that discriminate PASC from healthy individuals.

 

[Hutan]

Table 2: Only around 15% of the PASC people reported fatigue and lack of energy as a reason for attending the clinic. Around 20% of the PASC people reported dyspnea and respiratory issues as a reason for attending the clinic.

Quite a lot of co-morbidities too.

We found 39 DMRs that significantly discriminate PASC versus healthy groups; and 476 DMRs that significantly discriminate PASC versus post-COVID patients. These DMRs were constituted by an average of 15 similarly disrupted CpGs per region (i.e., all hypo- or all hypermethylated) and were distributed as shown in Fig. 2-B and Supplementary File S2 and S3, with about 20% located at promoter and exonic regions, and 80% located at intronic and intergenic regions.

CpGs = groups of consecutive positions

The 39 regions that were found differentially methylated in PASC versus healthy participants were also part of those regions that made PASC different from post-COVID patients, suggesting that changes evoked by PASC are specific to this condition and not found in healthy volunteers or asymptomatic individuals

We noticed that the number of identified DMRs in PASC versus patients without PASC (non-PASC) was substantially larger than those found versus healthy volunteers. We speculate that these differences were due to the fact that patients without PASC (non-PASC) had originally been hospitalized during their acute infection, and thus suffered from more severe COVID-19 disease burden leading to more pronounced changes in their methylome.
22

Figure 4 suggests that there are some genes expressed differently in the leukocytes. But see, for example the data for this gene:

Each dot is the data from one person. There does look to be some people with markedly different expression of this gene. And, if we look up FRG2C, it sounds a bit interesting
FSHD Region Gene 2 Family Member C
It is associated with

Spondyloarthropathy (SpA) is a chronic rheumatic disease encompassing related disorders such as ankylosing spondylitis (AS), psoriatic arthritis (PsA), reactive arthritis, and arthritis associated with inflammatory bowel disease. These disorders may occur simultaneously or sequentially in the same patient, suggesting various phenotypic expressions of the same disease.

And, just maybe 'reactive arthritis' has some relevance.

BUT, 7 of the PASC people have a reported diagnosis of "rheumatic disease" - so a lot of the people with increased expression of the gene probably actually have one of those rheumatic diseases, and it's nothing much to do with PASC at all.

 

[wabi-sabi]

The authors don't seem to have a clear understanding of what long COVID is. However, given that all of the people who were still symptomatic after COVID had a different methylation pattern, doesn't that show something of interest?

Also, from figure 4 above- are they making hay of that result? The PASC mean looks a bit higher to me, but only a tiny bit. And there's so much variability in this data! If we get rid of the outliers, is there a mean difference at all? And what's the difference between the PASC people who look really about the same as the healthy controls and the recovered people and the ones who really do have a different methylation pattern in figure 4?

Overall, I haven't seen much research on epigenetics and long COVID. Is this a viable area of research or are these people just grasping at straws?