Thanks, John. So again, my name is Pavan Bhattacharya. I'm an assistant professor in the Division of Pulmonary and Critical Care at the University of Washington, and director of our Sepsis Center for Research Excellence. So the title of my talk is Molecular Measurements in our Cohort here at SariPrep, a Multicenter Prospective Study. These are my funding disclosures. So the learning objectives are to describe the first 800 patients enrolled in SariPrep, and these were patients with severe acute respiratory infection, including COVID-19, as well as other viral infections. And then to briefly highlight some of the analyses that we can do with this type of cohort and what are some of the scientific discoveries we can find. So John talked a little bit about this. But SariPrep was really a collection of different regional as well as national collaborations and bringing them all together, including NETEC, the Society of Critical Care Medicine, and then the CDC Foundation was a primary funding entity. And the protocol paper was published in the Critical Care Explorations in the RADU-Postenico as well as other authors. So some of the goals of SariPrep are to advance our understanding of Sari and potentially do a deep dive into the epidemiology, the pathophysiology, and the molecular phenotyping. And then the overall planned enrollment includes 1,200 patients in a three-year prospective observational cohort study. We talked a little bit about the different sites, but I think one thing we wanted to highlight was when we're including sites to enroll patients, we're really trying to have some geographic diversity. As was seen early in the pandemic and even still in the pandemic, there's different places of the U.S. that are affected by the pandemic and affected by a number of cases of different viral respiratory infections. So this slide is a little outdated, but this was enrollment to date up to December 2022. So enrollment started early in the pandemic in April of 2020. And then currently we've enrolled approximately 820 patients. There's been 2,600 individual plasma aliquots, 800 different nasopharyngeal swabs, urine specimens, and peripheral blood mononuclear cells. And we're really taking a holistic approach to looking at these different biospecimens and leveraging the expertise within the consortium. So plasma and urine proteomics was done at the University of Washington, and I'll talk a little bit more about our initial findings with that. Humoral profiling is done at Nebraska with David Brett-Major as well as Jana Broadhurst. Peripheral blood transcriptomics is done by Leo Siegel at NYU. Viral sequencing for all the swabs, all the patients that were enrolled had viral sequencing completed. And then we did human genotyping also at University of Washington. And then George Inessi later is going to be talking about the hospital stress and strain surveys that were completed by all the sites. So then briefly I wanted to talk a little bit about the proteomics analysis. And so what we did with the proteomics analysis was that we wanted to do a discovery proteomics approach to understand what potential pathways are dysregulated in COVID-19, what pathways are upregulated that explain the poor outcomes that we're seeing. And so we started with 101 COVID-19 patients. We did a proteomics screen using a platform called Somascan, which is a somologic platform. It uses short oligonucleotides that bind to different proteins. And the value of Somascan is we were able to measure 5,000 proteins at once in one single blood specimen. The other benefit is the coefficient of variation is quite low on these measurements. And so you can have good generalizability and reproducibility of what you're finding. And then the final thing is we can potentially see what we're seeing in our cohort and apply it to other studies that use a similar platform. So the primary outcome we looked at in these analysis was 28-day ventilator-free days, as well as hospital mortality. And this is a volcano plot of our primary results. On the figure on the left, so on the x-axis are risk of ventilator-free days. So numbers right of zero, or each dot is a protein. And then so proteins right of zero are associated with higher or better outcomes, more ventilator-free days. Proteins left of zero are associated with fewer ventilator-free days. And so this is out of 101 patients. You see that one of the top proteins was angiopoietin-like 4. That's a protein that's highlighted in red there. I think the other thing to really highlight here is there are a number of proteins that are differentially regulated between the outcome of ventilator-free days, many that were associated with improved ventilator-free days and others associated with worse. Next what we did was, while we did this proteomic screen, we found that AngPTL4 may be one of these top proteins. We wanted to see, can we measure this using an ELISA, and can we do it in a multicenter cohort? And so using patients that were enrolled early in the pandemic in SariPrep, 225 patients, we found that higher AngPTL4 concentrations were associated with hospital mortality. The next thing we looked at is we had serial blood specimens collected during hospitalization. And so what happens over time? On hospital admission we found that the concentrations of this protein differentiate who's at development of poor outcomes, but what happens during the first two weeks of hospitalization. And we find that patients that ultimately go on to dying in the hospital have higher concentrations and those rise over the first two weeks. And then finally we looked in vitro in cell culture, in distal airway epithelial cells and saw, if we expose these epithelial cells to SARS-CoV-2, do we find upregulation of AngPTL4? And we found here on the figure on the left, you see that at 24 and 48 hours, there's increased AngPTL4 expression. And what's particularly exciting about this pathway is that this protein has been implicated in other viral infections, and I'll talk a little bit about that. But there's two domains, and the C-terminal domain is the one that's associated with endothelial leak capillary permeability. And then here, early in the pandemic, there were two papers that suggested that the receptor for AngPTL4, this is a receptor called Neuropilin, is upregulated in COVID-19 and a co-receptor for SARS-CoV-2 entry into different epithelial cells. So potentially arguing that if a patient has SARS-CoV-2, maybe they have upregulation of the receptor where AngPTL4 binds, and then AngPTL4 is more biologically active. And then finally, there's been two preclinical studies that have shown that you can potentially modulate this pathway. So an inhibitor to AngPTL4 in mice models improved outcomes in influenza as well as streptococcal pneumonia. So again, highlighting that potentially what we're finding in SARS-CoV-2 is maybe a drugable target. And then here is just a list of some of the other work that's come out of the consortium. So looking at how do kidney epithelial cells, how do they get infected with SARS-CoV-2 and what happens with kidney outcomes in the ICU? Another one looking at the endothelial pathways in COVID-19. And then finally, looking at the microbiome and secondary respiratory infections in COVID. And so in conclusion, SARS-CoV-2 is an ongoing observational cohort study. We're going to be enrolling until the end of this year. And then current enrollment includes different respiratory viruses, SARS-CoV-2, influenza, as well as RSV and others. And the early molecular features of COVID-19 illness in the ICU, we're able to compare this with other viral infections.

SariPrep study

severe acute respiratory infection

COVID-19

proteomics analysis

AngPTL4