In this exclusive video, Harlan Krumholz, MD, director of the Center for Outcomes Research and Evaluation at Yale University and Yale New Haven Hospital in Connecticut, discusses a recent preprint on distinguishing long COVID through immune profiling. Krumholz, a cardiologist, is also the Harold H. Hines, Jr. Professor of Medicine and a professor in the Institute for Social and Policy Studies at Yale School of Medicine.
The following is a transcript of his remarks:
My name is Harlan Krumholz. I’m a professor of medicine at the Yale School of Medicine, and I’m here today to talk to you about a recent preprint that came out that describes some immunologic features of people with long COVID.
Now, there have been many theories about what’s causing long COVID. Some people think it’s about viral persistence, some people think it’s about autoimmunity, some people think it’s about disruption of the microbiome, some think it’s about reactivation of other viruses, some think it’s tissue damage that initially started with the acute infection that persists, and some people just plain doubt the existence of it.
Those of us who have been talking to patients are convinced that this is real, that people are suffering, and it’s likely that there are many different clusters or groups of people who are experiencing things maybe with different mechanisms.
But what’s clear is that there are a lot of people who are affected. The CDC thinks that maybe one in four people with COVID actually ended up having symptoms that stretched beyond 4 weeks. The estimates of people who have severe symptoms is something like 3-5%, but we still need to learn a lot about this.
This study sought to take people who were infected fairly early in the pandemic and look at a group that had persistent symptoms, a group that had good recovery, and a group that were just simply healthy controls. The question is, what would their symptoms show us? Were there distinguishing features about what they were experiencing? And then, what would their immune system [look like] with deep immune profiling — looking at thousands of different features of the immune system — could we begin to understand this better?
This work was led by an MD/PhD student, Jon Klein, led by a two-year collaboration between labs of Dave Putrino at Mount Sinai and Akiko Iwasaki at Yale, and many other people who contributed. So what were the main findings of this preprint? Again, [it’s] under submission at a peer-reviewed journal, but it represents the science to date.
So first of all, the patients who were reporting the long COVID symptoms had a fairly specific group of symptoms. Now, their symptoms were really broad-based, but they ended up in some clusters that were very different from any symptoms that were reported by those who were ostensibly healthy or who had gotten through COVID without really having this long COVID syndrome. The most common things were brain fog, fatigue, dysautonomia, but there were a whole wide range of symptoms that tended to distinguish the group. That represents the patient-reported side.
Now on the immunophenotyping side, there were marked differences that started to become apparent, such that the group that was reporting long COVID was very different than the group that wasn’t. There were some overlaps, but there were very distinctive distributions for a wide range of things, such things as exhausted T cells, or some of the interleukins, or double positive T cells, activated B cells, DN [double negative] B cells, and non-classical monocytes.
Now, for those who aren’t in the field, these words may not mean a lot, but the important thing is that the immune system was reflecting a sort of very different activation profile, a very different what we might call immune signature than what we were seeing in the other group.
Now, the SARS-CoV-specific antibody responses were also elevated in those who were reporting long COVID after vaccination. So those who had been vaccinated were demonstrating a sort of different response, more elevated IgG and again, some differentiation between the groups. There was also evidence of herpes virus reactivation in the long COVID patients, as well as Epstein-Barr Virus. So, there were clues around here that maybe viral reactivation might be playing a role.
Interestingly, there was a long look at a big list of autoantibodies, and nothing really came out of that. A lot of us have been wondering, “Are autoantibodies at play here?” And at least in this group, that wasn’t what was suggested by the results.
It seemed like, using machine learning, that this profile of long COVID was distinctive. In fact, if you just looked at the immune signatures, you could predict the group who had long COVID, who was reporting long COVID, as a result of the way these immune signatures looked.
Then there was one really intriguing finding: low cortisol levels were reported more commonly in those with long COVID. And in fact, it was associated with disease severity, raising the possibility that this could be used as a screen or as a potential strong mechanism. Of course, cortisol is very strongly related to stress response, so it was a bit puzzling to see this finding.
Now, all these things are going to need to be replicated. There are issues with regard to the size of this study and with regard to what the controls were. It represents a very early study and it hasn’t been through peer review yet, but the most important thing is that when we took a bunch of people with long COVID symptoms, they were very distinctive with regard to their symptoms, but they were also very distinctive with regard to their immune signatures — suggesting, indicating, demonstrating that really this is a syndrome that can be described and is reflected in biological processes.
Now we just need to get going to understand it more, bring in more people, and be able to study this with greater fidelity. We’re doing this at Yale; we have a study called the LISTEN Study. We’re working with David [Putrino]; Akiko [Iwasaki]’s lab is at the forefront of this. You can email listenstudy@yale.edu and find out more about this study, where we’re trying to bring people in and enlarge what we’ve done in this study of immune profiling.
Lots of questions remain, but the important thing is that this supports the idea that people who are complaining of long COVID actually have physiologic issues. Many of their other tests have been normal — that’s because our tests within clinical practice are insensitive to pick this up. But using this expanded array of tests, thousands of tests, reflecting the immune system function and producing these immune signatures, we can see differences.
So more to come, but this is an important contribution in our view about getting us started and [something] to build on.
Please enable JavaScript to view the