Thank you very much, and thanks to all the prior speakers for terrific talks. So first, a bit of disclosure. So these are mine, a word about the GM-CSF and Akinra. I do receive study drug to support the clinical trials that you just heard about from these companies, but neither myself nor my research program receives income other than the study trials from these companies, and they had no role in the crafting of this presentation. So we're going to talk about the dynamic immune response to SARS-CoV-2 in this talk. We're going to understand the current state of immunomodulatory therapy in COVID-19, at least in as much as one can do in 15 minutes, because in two and a half years, a remarkable amount of work has been done, and to identify some opportunities to personalize that approach. Now, you all have been inundated with information about SARS-CoV-2 and its pathophysiology over the last two and a half years, and we understand that it has direct impacts upon parenchymal tissue, that it activates innate immune cells in part through the activation of pattern recognition receptors, it activates complement, and those activated innate immune cells go on to activate T-cells and produce pro-inflammatory cytokines which feed back and render ineffective many of those T-cells and innate immune cells as well. There is pro-inflammatory mediated endothelial dysfunction and hypercoagulability that results in lung and other organ injury as well. Now, when you're thinking about immunomodulation in this or any other critical ill phenotype, there's great debate about which things are important, and I think the parable of the blind people and the elephant is a very apt one for this discussion. If you look at one pathway, you're going to become convinced that that one pathway is very, very important. If you look at a different pathway, tell you what, that's going to be abnormal too, and you're going to think that that's the one that's really, really important. And so I think one of the key things throughout this discussion, whether it's SARS-CoV-2 infection, sepsis, trauma, cardiopulmonary bypass, postoperative inflammation, you have to look at the animal as a whole to try to understand where the signals are and where the targets should be. So I'm going to continue the parade of the red and blue figure and come back to this as an anchoring figure, and I'm going to start with the hyperinflammatory part of it. And this has been going on long enough that there are some nice meta-analyses of anti-cytokine therapies in COVID, and here is one that was published in JAMA discussing anti-IL-6 therapies. So as you know, IL-6 is a pro-inflammatory cytokine that's produced by both innate and adaptive immune cells. It has effects itself on the cellular machinery of both parenchyma and vascular endothelium and immune cells. It also induces the liver to make acute phase reactants like CRP. And you can see that in a remarkably short period of time, there have been a number of IL-6 blocking agents that have been evaluated in the context of adult sepsis. And this forest plot here, I don't intend for you to read every study, but to be impressed with the number of them that have been done, and to remark on the fact that there is a benefit in terms of all-cause mortality when these therapies are evaluated as a whole, and perhaps even a stronger benefit when the outcome is progression to invasive mechanical ventilation, ECMO, or death. We're going to take a little deeper dive into the Remap CAP paper that came out in 2021. This was a remarkable, and is a remarkable, research program that is a multi-center, multinational, open-label, adaptive platform trial to try to find the best combination of therapies for patients with severe pneumonia in both pandemic and non-pandemic settings. And boy, wasn't that nice to have when the pandemic for SARS-CoV-2 broke out. Its advantages include the fact that it's randomized, that it embeds these procedures in routine clinical care, that there is multifactorial modeling with sufficient power to allow for multiple interventions to be tested simultaneously, with frequent interim looks for, with a Bayesian approach to response-adaptive randomization, and a remarkably robust platform on which to permit continuous enrollment. So, there was a module within the Remap CAP study that evaluated immune modulators in SARS-CoV-2 infection, and this included IL-6 receptor blockers, IL-1 receptor antagonist, interferon beta, and, of course, control. The primary outcome variables, the number of respiratory and cardiovascular support three days in 21, with a number of rather predictable secondary outcome variables. So, the analysis of IL-6 respiratory, I'm sorry, receptor antagonists in the SARS-CoV-2 pandemic stratum was the subject of this analysis. And by this, by severe, they mean patients who were in an intensive care unit who were receiving respiratory and cardiovascular support. They were randomized early, within 24 hours of ICU admission. Patients with known immune suppression were excluded, and because of some potential toxicities with the IL-6 blockers, or receptor blockers, patients with elevated transaminases or severe thalamocytopenia were excluded. And rather remarkably, there was a dramatic benefit in terms of organ support three days, with one to two extra days of organ support free time for patients in the tocilizumab or sireliumab arms. And similarly remarkably, there was a significant mortality benefit to enrolled subjects in those arms as well. And here you can see probability of survival in the upper two panels, as well as probability of discharge from ICU or hospital in the lower panels, which, again, support the use of these IL-6 pathway blocking drugs. When you go to the appendix, or the supplemental material, I think it gets even more interesting, because these patients are not monolithic. This is a relatively heterogeneous population of patients, even within our ICUs, and what you can see here is when they evaluate the effect by tertile, tercile, of CRP levels, you can see that while there was a strong probability of superiority to control in all of the terciles, it was only in the highest tercile where the confidence interval didn't span one. In other words, it was the highest tercile of CRP, which would imply the highest tercile of IL-6, though that was not measured in this study, where the greatest benefit occurred. And that was in terms of the primary outcome and in the mortality outcome. So, it raises the question, did some of these folks with lower CRP and presumably lower IL-6 levels really need this drug? But the signal was high enough that it remained flamingly positive when the cohort as a whole was evaluated. Now, of course, IL-6 is not the only cytokine or pathway that has been the target of blockade in patients with SARS-CoV-2. The IL-1 pathway has as well. This is a relatively recent meta-analysis that looked at a number of clinical trials that used anakinra, or recombinant human IL-1 receptor antagonist, and strongly suggests benefit to adults with COVID-19 pneumonia. I want to bring this group, the Jim Relos-Barbulas group, has been prolific in the last two-plus years in understanding the immunobiology of COVID-19 in adults. And this study enriched the population of patients by only looking at those who had high soluble urokinase plasminogen activator receptor, or SUPAR, levels. So, this biomarker had been shown in preliminary work to place patients with a SARS-CoV-2 infection at higher risk for disease progression. And here, the authors, again, selected this population of patients with a primary outcome of preventing worsening by the clinical progression scale. For those of you, like me, who don't have SUPAR available in your clinical laboratory, they provide, in a supplementary table, some clinical factors that might predict a high SUPAR, and these are patients that you might imagine would have a high risk for adverse outcomes from COVID-19. And sure enough, patients in the anakinra arm had a dramatically faster improvement in their clinical progression, or failed to progress, to worsening disease compared to those in the placebo arm. And similarly impressive effects with the number of patients with respiratory failure and even survival. So, this is one of the few studies, again, by the same group, that attempted to use a truly personalized approach to SARS-CoV-2 immunomodulation. So, this was an open-label phase two trial of anakinra or tocilizumab based on immune phenotype. So, not unlike the phenom study that you heard, though the definitions are a bit different, they tried to break people down into those with macrophage activation-like syndrome, macrophage activation-like syndrome or MALS, and they used a ferritin threshold that they had derived from prior work, a higher one than the pediatric study did, here greater than 44 or 20. Those subjects would get anakinra. If they had immunoparalysis, in this case defined by an HLA-DR expression less than 5,000 on monocytes, and they didn't have severe hyperferritinemia, they would get tocilizumab because some in vitro work suggested that the use of tocilizumab could improve HLA-DR expression in cultured monocytes. If the patients had immunoparalysis but also had high transaminases, because, again, tocilizumab, one of its toxicities, can be elevation of transaminases, we're just going to give you anakinra, right? So, they also got anakinra. And importantly, there was no control group. Their primary outcome was a decrease in SOFA score or a 50% increase in the PF ratio in the second week. And when comparing these two approaches, you can see that the patients who received anakinra were more likely to have the decrease in SOFA score, and their primary endpoint was more likely to be met. Cumulative probability of being discharged was better in the anakinra group, and there were more ventilate-associated pneumonias and bloodstream infections in the tocilizumab group. So, again, I think the take-homes from this are relatively few from a practical perspective, but one of them is that real-time immunophenotyping in their hands was feasible. I love that. It also suggests that the treatment of immunoparalysis with normal transaminases of tocilizumab was associated with worse outcomes than the treatment of MALS or immunoparalysis with elevated transaminases with anakinra. Do with that what you will. I don't know what to do with that. The lack of a control group was a major weakness, and it necessitates the performance of placebo-controlled trials, and I do know what to do with that. So, I think what this does do is provide a preliminary data and impetus for performance of those key randomized controlled trials. Now, back to the immunoparalysis side of this equation. Does this happen in SARS-CoV-2 infection? Yes. Yes, it does. So, and that went away. So, what you're going to see here in a minute is that people have immunophenotyped patients with SARS-CoV-2 and has shown that monocyte HLA-DR expression by flow cytometry is lower in patients with COVID-19 than healthy controls, and that patients with more severe illness demonstrate a reduction of HLA-DR expression on both classical and non-classical monocytes as well. Separate groups showed something very similar. Patients in the ICU with COVID-19, these were all adults, had higher CRPs and higher IL-6 levels compared to the non-ICU patients. Not terribly surprising. But when you take their blood out and you take their peripheral blood mononuclear cells and you stimulate them ex vivo, the ones in the ICU don't work. They don't make pro-inflammatory cytokines like interferon gamma, IL-22. The non-ICU ones are a little bit suppressed compared to healthy controls, but look at the ICU ones. They don't make these pro-inflammatory cytokines. Neither do they express monocyte HLA-DR very well. And here's another one where the blood was taken out. It was stimulated with phytohemagglutinin. This is a lymphocyte stimulator. And again, across the board, both pro- and anti-inflammatory cytokines, mind you, the SARS-CoV-2 infected patients had suppression of their immune responses. So we're hearing all about how these patients have high levels of circulating cytokines in the plasma. They're not coming from the white blood cells in the plasma, ladies and gentlemen. They're coming from somewhere else. They're coming from the infected parenchyma. They're coming from the vascular epithelium and other sources. A little brief pediatric commercial because I have to throw some pediatric data in here. So Catherine Bline, one of my faculty members, looked at myeloid-derived suppressor cells, a relatively recently identified population of cells that we think drives an anti-inflammatory response in critically ill adults and children. And she indeed found higher levels of myeloid-derived suppressor cells in patients with SARS-CoV-2 positivity, including, to a lesser degree, those who were incidentally positive, who were only found upon screening, though were otherwise clinically asymptomatic from a respiratory perspective. Those who had a higher proportion of MDSCs in their bloodstream had lower proportion of CD4 cells, T cells, in their bloodstream, and they had a longer duration of hospitalization. So should we give an immune stimulant to these people? We heard about giving GMCSF to people with immunoparalysis. Well, there have been a couple of clinical trials, more than a couple of clinical trials, of blocking GMCSF in these patients. So giving anti-GMCSF antibodies to patients with COVID, and, in fact, there's a trend toward reduction in mortality on the basis of a relatively recent meta-analysis, and a statistically significant reduction in intubation risk by blocking GMCSF. Huh. Well, another group of investigators is wondering if that's the right approach. Because alveolar macrophages are known to be dysfunctional in patients with COVID-19, this appears to be due at least in part to insufficient GMCSF-induced alveolar macrophage differentiation. And the theory is that inhaled GMCSF could represent a treatment option for this patient population, and in this same report is a pilot study of an 81-subject open-label RCT of inhaled GM versus standard of care in non-ventilated patients, and there was a trend towards better oxygenation and no safety signal, or unpleasant safety signal, and no evidence of worsening inflammation. A similar trial was carried out nearly contemporaneously with that, in which inhaled GMCSF was given to 100, or standard of care was given to 122 hospitalized adults who were not mechanically ventilated but had COVID-19 pneumonia with a primary outcome variable of systemic oxygenation and intubation rates. And sure enough, there was faster improvement in oxygenation in the inhaled GMCSF group. It was not really powered to detect a mortality or, sorry, an intubation or mortality signal, so they'd have to stay tuned on that. So I'm not sure what to do with that either, exactly, except keep looking at it. Keep studying it. This is a nice review that came out, you know, about a year ago now by many of the people that you have seen on the slides to date, and it provides some guidance about when one might consider using immunomodulators in patients with COVID. These would include dexamethasone. These would include the anti-IL-6 drugs I talked about, the anti-IL-1 drugs. JAK inhibitors, which for a time I did not talk about but certainly are worth reading about, and monoclonal antibodies. The lighter blue represent research opportunities. They all represent research opportunities, but represent ones that might have less evidence to support clinical use now but certainly worth thinking about. So in summary, the immunobiology of SARS-CoV-2 is dynamic, and it can include both hyper inflammation and immune suppression, not unlike sepsis. There's substantial evidence supporting the use of targeted anti-inflammatory therapies, particularly anti-IL-6 and anti-IL-1 therapies, though it remains unclear in my view that all patients need the exact same approach. As has been said multiple times this morning, I think each patient should be evaluated differently, and even within those studies, you're going to see effects based upon that patient's baseline inflammatory state. The use of, or blockade of, immunostimulatory pathways remains deeply controversial, and immunophenotype-directed, randomized, placebo-controlled trials are badly needed. And studies of truly personalized immunomodulation in COVID-19 are still lacking. I haven't presented you very much about personalized immunomodulation in this 20-minute talk about personalized immunomodulation, other than we need to learn about it and do it. So, and I promise I did not steal this knowingly from Dr. Machinsky, but we need the right patient to get the right drug at the right time, because that's what this is all about. So, with that, thank you so much.

immune response

SARS-CoV-2

immunomodulatory therapy

IL-6 blocking agents

IL-1 receptor antagonists

personalized immunomodulation