Infectious Disease

Half of long COVID patients form ‘inflammatory subset,’ may benefit from immunomodulation

October 06, 2023

6 min read


Published by:

healio rheumatology logo

Torgerson reports being a member of a data safety monitoring board for a Takeda sponsored clinical trial unrelated to COVID.


Receive an email when new articles are posted on

Please provide your email address to receive an email when new articles are posted on .


We were unable to process your request. Please try again later. If you continue to have this issue please contact [email protected].

Back to Healio

Half of patients with long COVID demonstrate persistent inflammation and may potentially benefit from immunomodulation, according to data.

The researchers additionally proposed a serum panel of three marker proteins — CCL7, CD40LG and S100A12 — that could potentially differentiate between inflammatory and noninflammatory long COVID, and thus which patients may benefit from immunomodulatory therapies.


In a paper published in Nature Communications, Talla and colleagues noted that the etiologies of post-acute sequelae from SARS-CoV-2 infection may include persistent inflammation, unresolved tissue damage or delayed clearance of viral protein or RNA.

However, the biology of those sequelae remains poorly understood, according to Troy Torgerson, MD, PhD, director of experimental immunology at the Allen Institute for Immunology, based in Seattle, and an investigator on the study.

In the current analysis, the group assessed serum proteome samples from 55 patients with post-acute sequelae from COVID-19. Eligible participants experienced symptoms for at least 60 days after onset of acute infection. The findings for this cohort were compared with data from patients who either had recovered from COVID-19 or were not infected with the virus.

The results suggested that subsets of patients with post-acute syndromes demonstrated signatures of persistent inflammation. Specifically, type II interferon signaling was observed, along with canonical NF-B signaling, which the researchers noted is “particularly associated with TNF,” according to the researchers. A persistent neutrophil activation signature was also reported in some patients.

These signals are the “most differentially enriched signaling pathways,” the researchers wrote.

“These findings help to clarify biological diversity within [post-acute sequelae of SARS-CoV-2 (PASC)], identify participants with molecular evidence of persistent inflammation, and highlight dominant pathways that may have diagnostic or therapeutic relevance, including a protein panel that we propose as having diagnostic utility for differentiating inflammatory and non-inflammatory PASC,” they added.

Healio sat down with Torgerson to discuss the associations between COVID-19 — both long and acute — and inflammation, as well as potential treatment strategies for patients with inflammatory markers of PASC and who is best positioned to manage this group.

Healio: What was the context of this research in terms of the ongoing reported associations between acute COVID-19 and rheumatology?

Torgerson: Many inflammatory or autoimmune diseases have been linked to previous infections. Some of those links have been well validated. We suspect that, similarly, COVID may initiate an inflammatory process in some individuals that may persist. One challenge for diseases that are not well understood at a mechanistic level is that they are first described clinically, but we know that there can often be different paths that lead to similar clinical symptoms.

Our question was whether individuals who were clinically defined as having long COVID had different reasons for having similar symptoms, and to determine whether some of them had evidence for ongoing inflammation that could possibly be treated with an immunomodulatory drug.

Healio: What are the links between rheumatology — specifically fibromyalgia — and long COVID?

Torgerson: The association between long COVID and fibromyalgia is unclear. There have been suspected links between previous infections and the onset of fibromyalgia as well, but to my knowledge those have not been clearly substantiated.

Like long COVID, fibromyalgia was first described clinically and likely has at least a handful of molecular drivers. Some studies have focused on understanding the various subsets of fibromyalgia. There is evidence that some patients may have ongoing inflammation as well, although it is not yet clear how that inflammation overlaps with what we observed in the inflammatory subset of long COVID.

Hopefully, all of these efforts will help to define which patients would be good candidates for treatment with an immunomodulator in both diseases. They could be used to select patients most likely to benefit into clinical trials of these drugs, and will hopefully help point the field to mechanisms of disease.

Healio: Regarding the specific proteins in the panel, why do they differentiate inflammatory from non-inflammatory long COVID?

Torgerson: Just over 50% of the individuals with clinical features of long COVID had evidence of circulating inflammatory proteins. About half of those had a set of elevated proteins in the blood that are well known inflammatory cytokines that act as stimulators of immune responses. Interferon-gamma was the most dominant of these, but others like TNF-alpha were also present.

In the immune system, these are known to be big kahunas in driving other inflammatory responses. We saw evidence of many of these as well. It has the flavor of what you would expect the immune system to do if it were still being exposed to some part of the virus. We know from recent studies done in tissue biopsies or autopsy samples that proteins or viral nucleic acids were detectable in various non-lung tissues of some patients for at least 6 to 8 months or more after infection.

The other inflammatory subset in our cohort had evidence of persistent neutrophil activation. Neutrophils are a type of immune cell involved in both early responses to infection and ongoing inflammation. We hypothesize that individuals that fall into these categories may be good candidates for treatment with an immunomodulator.

Healio: Does this mean that the virus persists in these patients, or that the separate entity of long COVID has taken hold?

Torgerson: That is one of the big questions that we still need to answer. It is possible that both could be true. Maybe one subset of individuals has a hard time clearing the virus after an acute infection, so it continues to be made at a low level in the tissues before it is fully eradicated. The immune system would view this as foreign garbage and make an ongoing immune response to try to clear it.

In contrast, some people’s immune responses to the acute infection may be so robust that they kick off an exaggerated inflammatory response that is difficult to control or resolve, so they have ongoing inflammation. We are trying to sort this out by taking a deeper look using other experimental tools.

Healio: Is there a threshold of inflammation a patient may have to cross before being categorized or stratified as having inflammatory long COVID?

Torgerson: I don’t think we know that threshold. For example, I can’t give you a blood concentration of interferon-gamma below which you feel fine, and above which you are going to feel terrible. It probably varies somewhat from individual to individual.

For this study, we used an experimental method that allows us to measure more than 1,400 plasma proteins at the same time so we can see a more complete picture. When one inflammatory protein appears to be increased, we can also check whether some of its known effects on the immune response are also present, by evaluating whether other proteins that it is known to regulate are increased or decreased. It gives us increased confidence that the inflammatory protein is not only present but is having an effect on the system, and that what we are seeing is real.

Healio: For patients who do have inflammatory long COVID, who is best suited to treat them? Where do rheumatologists fit into the care team?

Torgerson: For the groups with signs of persistent inflammation, I would recommend that rheumatologists and immunologists who are well acquainted with the use of immunomodulators be key members of a care team. If the scientific community discovers that the ongoing inflammation is driven by persistence of viral particles, then we may need to also consider longer courses of antivirals like Paxlovid (nirmatrelvir/ritonavir, Pfizer), but that would only help if there were ongoing viral replication.

It is one of the reasons that this study is important because it may provide another tool to get patients routed toward therapies most likely to make a difference. For example, if a patient has no evidence of ongoing inflammation and their symptoms can be linked to organ damage sustained during the acute infection, they may benefit most from therapies focused on rehabilitation, while others with evidence of ongoing inflammation may benefit most from drugs that would reduce that.

Healio: Is it too early to talk about stratifying patients based on inflammation in the clinic and in clinical trials?

Torgerson: I don’t think it is too early to talk about this. That said, we do not yet have an easy, clinically validated protein panel for testing this. We need to use the data from our set and boil it down to a smaller panel of three to five proteins that are reflective of the overall inflammatory signature.

We propose a three-protein panel as part of our study, but the size of our validation cohort was limited. A larger study is needed to clinically validate a clinically useful panel. In addition, we tried to determine whether the presence of an inflammatory protein signature was associated with particular clinical features, but our cohort was too small to do intra-cohort stratification.

Healio: What do your findings mean for the average practicing rheumatologist? What can they offer in the management of these patients?

Torgerson: It suggests that about 50% of patients with long COVID have persistent inflammation and may benefit from trying an immunomodulator. One caveat is that these data were generated from patients with long COVID resulting from the ancestral strain of the SARS-CoV-2 virus. It is possible, with vaccines and natural immunity, that things have changed somewhat.

The incidence of long COVID after acute infection seems to be smaller now but it is unlikely that these measures would have completely prevented the occurrence of persistent inflammation in a subset of patients. Even though a clinically validated protein panel does not exist to differentiate inflammatory long COVID, there may be stop-gap measures. For example, there are clinically available assays to measure interferon-gamma activity in the blood — CXCL9 level for instance. Although this has not been clinically validated yet in long COVID, it may be useful for determining which patients have elevated levels of this cytokine, and thus ongoing inflammation.


Talla A, et al. Nat Commun. 2023;doi:10.1038/s41467-023-38682-4.


Receive an email when new articles are posted on

Please provide your email address to receive an email when new articles are posted on .


We were unable to process your request. Please try again later. If you continue to have this issue please contact [email protected].

Back to Healio

COVID-19 and Rheumatology

Related Articles