Hello, I'm Dr. Kenneth Remy from Case Western University in Cleveland, Ohio, University Hospitals of Cleveland, and Rainbow Babies and Children's. Today I'm going to be talking about COVID-19, a cytokine storm, or immune exhaustion, very controversial topic over the past two years, especially among those of us that study immunology. And so I hope to shed a little bit of light, add to the confusion, and hopefully at the end we could come to some conclusion that this is a very detailed topic that certainly may in fact have different answers. So just for the purposes of today's talk, there is the evil offender, COVID-19 reflects the disease, SARS-CoV-2 is the virus. And this is our evil offender, and each of these aspects of the structure of the coronavirus, whether it's the nucleocapsid protein, the membrane glycoprotein, spike protein, envelope protein, or the RNA, all individually have different immunologic aspects by which they are involved with internalization of the virus, and then eventually wreak havoc on different aspects of both innate and immune, adaptive immune modifiers. So let's just take a quick lesson in history back into, walk back into time. So when the pandemic originally struck, there were three key papers that came out of Wuhan, China, and in JAMA, Lancet, and New England Journal of Medicine, the clinical characteristics of individuals with this virus were published. And they were very interesting because they did unearth something key. The most common laboratory abnormality observed in the JAMA, 138 hospitalized patients were depressed total lymphocytes, and that the lymphocyte count continued to decrease until death occurred. And the authors in this paper said that these abnormalities suggest that the 2019 infection may be associated with cellular immune deficiency. The JAMA article also demonstrated in non-survivors versus survivors that neutrophil counts increased in those that over time that did not live in comparison to those that had a relatively steady neutrophil count that lived. And you can see here this profound lymphopenia that occurred, and it continued to decline all the way through the 19 days that this study was published. And then in the New England Journal of Medicine paper, they found the same thing. Lymphopenia presented 83.2% of all patients, leukopenia and 33.7%, with lymphocytopenia being the most common abnormality. And if you looked at the Lancet paper, those that succumbed from COVID-19 disease certainly had in this red lines here, worsened lymphopenia in comparison to those that survived. Lymphopenia was the most common abnormality and continued to decrease until death. And they also recapitulated the JAMA data as well, which demonstrated that there was a rise in polymorphic nuclear cells that potentially could be secondary to secondary infections. So lymphopenia, very interesting for most of us in the field, especially for individuals that study sepsis. So in two separate papers in both JCI and ICM, you can see here that there is a relative deficiency and inclusion of lymphopenia when you look at also subsets. So if you look at CD3, CD4, CD8, and K cells compared to healthy controls, B cells, you can see that in the GenA paper, there's undeniably over time, a suppression of the immune system with these adaptive and innate markers. And then the Chen article demonstrated that there's certainly with severity of illness, some worsened lymphopenia and suppression of CD4, CD8 cells, as well as NK cells, although not statistically significant on that latter point. And what did we hear about? In the beginning of the pandemic and subsequently, there's a cytokine storm. This disease is cytokine storm. This isn't a suppression, this is cytokine storm. And so this continued in the beginnings of the pandemic and mimicked a lot of what we knew for sepsis, because it was initially thought that there was a hyperinflammatory phase in sepsis and subsequently, the development of an immune suppressive phenomenon over time. But maybe COVID actually mimics sepsis a lot more than we think, because understanding this interplay between both the innate and adaptive immune systems certainly could unearth pathways that could be helpful in development of new therapeutics for COVID. And so if you think about it, in hyperinflammatory diseases where you have endothelial interactions like sepsis and you get leaky blood vessels up front, we do see early deaths. But many of those individuals with sepsis over time will have suppressive and T cell exhaustion leading towards secondary infection and subsequent demise. And it's possible that both of these phenotypes do exist in COVID disease. So what is a cytokine storm? Well, cytokine storm and cytokine release syndrome, certainly made familiar through CAR T therapy, are life-threatening systemic inflammatory syndromes involving elevated levels of circulating cytokines and immune cell hyperactivation that can be triggered by therapies, pathogens, cancer, autoimmune conditions, and monogenic disorders. Simplistically, there's an overactivation or a hyperactive immune response that's characterized by release of interferons, type 1 and type 2, interleukins, tumor necrosis factors, chemokines, and several other mediators. In this wonderful review by a New England Journal of Medicine in 2010, not in regards to COVID obviously, but regards to just cytokine storm, it was reported that cytokine storm is an umbrella term and encompasses several disorders of immune dysregulation characterized by constitutional symptoms, systemic inflammation, and multi-organ dysfunction that can lead towards multi-organ failure and death. But it's nonspecific. But if you follow the pathway, which I think they brilliantly described in this review, you have these pathogen-induced triggers that, and here this is a paper that's taken from 2021, COVID-19 was included. You can have iatrogenic causes or monogenic or autoimmune disorders. And you can see that these leads towards driver cells, adaptive T cells, CD4, CD8. It has alterations specifically in activating these driver cells, including macrophages, dendritic cells, and NK cells. And this leads towards immune hyperactivation, either from an inappropriate triggering or danger sensing, inappropriate or ineffective amplitude of the response, or a failure if there is a cytokine storm to resolve inflammation and return to homeostasis. And these driver cells subsequently will lead towards a cytokine storm, which subsequently then leads towards excessive circulating cytokine levels. And these include things such as TNF, interferon gamma, interleukin-6, interleukin-17. And with prolonged activation of these signaling pathways, you can see with MAPK, NF-kappa-B, the JAK-STAT3 pathway, certainly a barcitinib target, and mTOR. And as these subsequent storms of these cytokine production or overactivation of the underlying innate and adaptive immune systems occurs, you have collateral-induced damage. And that includes acute systemic inflammatory respects and secondary organ dysfunction. Furthermore, it's not just unique, as mentioned, to just the innate cells, but Th1, Th2 differentiation, Th9, Th17, and cytotoxic T cell lymphocytes are certainly influenced by this. And in cytokine storm, as aptly put in this table, you can see the differences in the main cell source by which you have alterations and hyperactivation development of these cytokine production through a number of these and the type of function of what each of these are used for. So what about COVID-19? Well, Davali and Neitscher in 2020 published what I think is a really, really nice piece of the literature, where it was evaluated cytokines between healthy donors, CAR-T, no cytokine release syndrome, CAR-T cytokine release syndrome in COVID-19 patients, and what you can see here is interleukin 6, interleukin 8, TNF alpha, and IL-1 beta. And just to take for one second, you compare IL-6 and COVID as compared to healthy volunteers, you can see that there is certainly a robust increase in this population. Less of a response or increase in IL-1 8, less of a response in TNF alpha, and a slight increase when you look at IL-1 beta. But what's most interesting is that when you looked at IL-6 production and cytokine release syndrome, that certainly is the highest levels that were reported in this Nature paper. And certainly less in COVID-esque in comparison, but elevated certainly from those that have no cytokine release syndrome that have endured CAR-T cell therapy. So from this paper, it looks like in isolation, IL-6 is certainly elevated in COVID patients, although the severity of illness, in fact, was not well described. Prateek Sinha at Washington University subsequently published a nice paper in JAMA Internal Medicine looking at the large heterogeneity in IL-6 levels from COVID compared to other ARDS studies. And he pooled that data and looked at the different manuscripts that came out of COVID as well as the large trials that have come out for ARDS. And if you look here at the picograms per mil of those IL-6 levels, the total population, as well as those with severe disease, what's interesting is that there is certainly a heterogeneity of those levels. And in fact, other ARDS studies have reflected even higher levels of IL-6. So it's certainly feasible that it could be elevated high in some patients with COVID, but in fact, maybe lower than other ARDS studies that previously have been published. So let's just shift gears for one second. What's immune exhaustion? T cell exhaustion is a state of T cell dysfunction that arises when many chronic infections and cancer is defined by poor effector function, sustained expression of inhibitory receptors and a transcriptional state distinct from that of the functional or memory T cells. It prevents optimal control of infection and is non-specifically defined by lymphopenia. So just think back to what we were talking about. So in the pandemic, based off of all the information that was coming certainly out of our Chinese cohort in those three major papers, myself and a number of other individuals wanted to certainly put this in our own hands and study this with our severe COVID patients. And so given the fact that cytokine storm was becoming reported in both the lay and the scientific literature, we decided to embark upon a number of studies, which were subsequently published in JCI Insights and Science Advances and soon to be in PLOS One. And in our group, we've demonstrated and compared COVID ICU patients versus septic patients versus critically ill non-septic patients against healthy volunteers. And there were similarities between all of the groups and what I'll present is those comparisons with a number of different descriptive immune readouts. So the first thing that we wanted to understand was, could we recapitulate the data that was coming out of the Chinese cohort? And what you see here in panel A, B, and in C are survivors versus non-survivors. You can see that lymphopenia persisted throughout the duration of time for those that had COVID-19. And then you can see that this is the survival over time from days after ICU admission and days after symptom onset. And this work was done through our laboratory, including with Monty Mazur, Isaiah Turnbull, and of course, Richard Hotchkiss. What you're looking at here is the comparison of COVID-19 against septic patients. And this is aisle six levels as measured via an ELISA. We did this over time. Non-survivors are in red, survivors are in black. And as you can see here, in comparison to sepsis, at least in our groups, aisle six levels were approximately two-fold higher than in septic patients. So one might surmise that certainly there is a cytokine storm of interleukin-6, and that this is certainly higher in COVID patients than in septic patients. So we wanted to understand at least with functional readouts. And so we resort to an assay that we've used commonly in our laboratory called enzyme-linked immune absorbent spot, or ELISPOT, where you're able in a 96-well plate to stimulate with either CD3, CD28 to look at T cell function, looking at interfering gamma versus stimulating monocytes or the innate immune readout, which is via lipopolysaccharide to form production of TNF-alpha. And what you can do is count the number of spots and the robust size of those spots to demonstrate cellular response and function. And then you can evaluate this against different therapies. And so if you've got a robust or cytokine storm, you'll have lots of spots, and if you've got a suppression, you'll have very minimal spots after stimulation with a positive stimulus like CD3, CD28, or lipopolysaccharide. And so we compared our critically ill non-septic patients, our septic patients, and our COVID. Remember that those that died are in red dots. And what was most interesting was that there was a depressed T cell function and interfering gamma production when you compared COVID patients against septic patients, and certainly compared to critically ill non-septic patients. And it was clear in our data that there was a suppression of this function. We next wanted to evaluate this same finding against TNF-alpha to look at our innate marker. And so you're looking at COVID-19 again, septic, critically ill non-septic, and healthy. And as you can see across the board, there were cytokine production decreases with interfering gamma production and TNF in our COVID-19 as compared to sepsis. And it appears that those that are non-survivors are at the bottom levels of those productions. So they have pretty profound T cell exhaustion, as well as potentially monocyte inability to form TNF-alpha after LPS. And then when we looked over time, you can see that both of these suppressions persisted, especially in non-survivors. Another group that we collaborated with at Washington University, both Ali Elabadi and Philip Mudd, excellent scientists and clinicians, who published another paper with the same patient population, but now they've compared it to their work in influenza. And you're looking here at B cell, CD8 T cells, CD4 T cells, classic monocytes, intermediate non-classicals, as well as activated CD4 cells, CD8 cells, and plasma blasts. And they compared SARS-CoV-2, which is in green, against influenza against healthy controls. And what was really interesting was that the amount of suppression seemed across all cell types was more profound, yet very similar to influenza in the COVID or SARS-CoV-2 group. Definitely demonstrating the same findings that we saw in ELI spot. And Phil Mudd went on to, I think, publish a really exquisite publication, and I'm not going to go into a lot of this in depth, but looked at single gene cell gene expression of PBMCs from these COVID positive patients against influenza patients and healthy subjects, and found differences in the relative abundance and transcriptional cell subsets across both conditions. He further went on to calculate odds risks of COVID-19 versus influenza. And as you can see here, it's reflective that there's a number of statistically significant cytokines that certainly were suppressed in comparison to their influenza cohorts. And as we publish this data, Zeng in Cellular Molecular Immunology in 2020 demonstrated similarly that COVID-19 had T cell exhaustion and lymphocyte depletion in both severity of mild versus severe compared to healthy controls. So these COVID-19 patients had cytotoxic lymphocytic exhaustion, which worsened with severity. And in fact, those that had the decreased level of cytokines in peripheral T cells had the most severe disease. And so it beckoned the question that understanding all of this information, whether or not evaluation of these sorts of patients may in fact deem different targeted therapies for virus elimination, rather than potentially using use of anti-inflammatory therapies. Isaiah Turnbull at Washington University in our group as well, looked at cytoproteomic experiments. Although I'm not going to show that data, he did demonstrate that COVID-19 with increased severity compared to healthy donors actually did have a systemic inflammatory response. In fact, across IL-6, IL-8, IL-1RA, MCP-1, MIG-1 and IL-1 beta with severity of illness, there was an increase in those cytokines. So now it's becoming more clear that there certainly is, there are groups that have this cytokine storm, if you will, as most predominant with IL-6, especially as compared to those with sepsis. But also these are the same patients that demonstrated suppressive phenomenon. So they have these two divergent yet potentially telling phenotypes. And so it's certainly possible, this is a paper that Phil and I published in JCI, that when you compare the response to human virus respiratory infection in panel A with influenza as compared to COVID-2, that in COVID-19, excuse me, the cellular immune system remains activated despite viral clearance. And this persistence and development of T-cell exhaustion and development of secondary infections may set SARS-CoV-2 infection apart from other viruses. So you get this over activation, but you get this ongoing suppression and exhaustion. And each of these may in fact lead towards prolonged hospitalizations and or subsequent death. John Weary's group at UPenn, I think, did a wonderful job in doing deep immune profiling, and they demonstrated these three COVID-19 immune phenotypes, type 1, type 2, and type 3, which may in fact reflect differences in severity. In one group, you had highly activated CD4 and CD8s that were increased. You had altered CD8 eMRA, PB responses, and altered CTFH. This could be a very hyperactivating group. You also had a group that had low or no activated T-cells or B-cells. This could be a very suppressed group. And then you have this group in the middle that sort of reflects both conditions, but three very distinctive COVID-19 immune phenotypes. So what does this mean? Well, cytokine storm involves an immune response that causes collateral damage. There is an overactivation of some immune modifiers that leads towards ongoing and worsened adoptive exhaustion. But as John Weary's group, and perhaps as I've just demonstrated, could it be possible that you've got three different immune phenotypes that could underlie why there's differences what we're seeing at the bedside? You've got high versus pulmonary compliance patients. You've got that happy hypoxemic versus the classic ARDS patients. You have those that have and require intubation very early in their course versus those that linger on noninvasive positive pressure for weeks on end. You have those patients that recover quite quickly and others that are intubated for weeks upon weeks, at which point then they either succumb to the underlying illness or subsequently they recover. Is it possible that understanding that these patients may in fact have both a cytokine storm and temporally also may have T cell exhaustion or immune suppression may in fact unearth perhaps new ways to think about this disease and also allow us to really decide how we can functionally immune phenotype them so we can deliver precisely the therapies that they may best benefit? Because invariably there is an interaction with internalization via ACE2 with innate cytokines attracting and activating the inflammatory cells, inducing that cytokine storm, utilizing hyperinflammatory macrophage activation, but then there's an exhaustion of T cells, CD8s, CD4 cells in macrophages, let alone what could happen potentially in complement. And then subsequently these lead towards exhausted NK cells. So in conclusion, SARS-CoV-2 infection is complicated. It provides a heterogeneous set of endotypes, including profound immune exhaustion and a subset that endures a storm of cytokinemia with potentially very different clinical phenotypes. Therapies should be tailored to each of these endotypes and not apply to a one-size-fits-all approach. And remember, therapies that may alter the host could potentially cause differences in immune suppression or inactivation. So I don't think I've answered the debate, but I think I hedged my bet to say that perhaps both sides are correct and further understanding and delineating these differences may in fact be informative as we move forward with this virus and certainly understanding others. I appreciate the time and thank you to Society of Critical Care Medicine for the opportunity to talk about this really exciting topic.

COVID-19

immune system

cytokine storm

immune exhaustion

SARS-CoV-2 virus

T cell dysfunction

Sepsis

Infection

Cytokine

Presentation

Select

2022

Intermediate

Advanced

Sepsis Resources