All right, thank you for coming today. The title of my talk is Serum Protein Levels of the CXCL9, 10, and 11 chemokines reflect sarcoidosis disease severity in the grads cohort. My name is Nicholas Arger. I'm an assistant professor of medicine at UCSF, and I have no disclosures. My area of interest is sarcoidosis, which is the disease of systemic granulomas inflammation without a known cause. And some of the main challenges we have in this disease are assessing its activity in the lungs as compared to other organs, and predicting long-term outcomes, especially because we lack ways of differentiating patients who will have favorable disease courses, i.e., resolution of their disease, as compared to patients who have persistent or progressive disease requiring aggressive immunosuppression treatment. Here I show a CT scan of one of my patients from clinic, who is manifesting this granulomas inflammation in a typical pattern of peri-lymphatic nodules and adenopathy. Now for a while now, we've known that CD4 T cells and the production of interfering gamma are essential in granuloma formation and perpetuation. And today, I'm going to be talking about CXCL9, 10, and 11. These are proteins that are what we call chemotactic cytokines, or chemokines. They are produced by a variety of cells, including immune cells and interstitial cells, in response to interfering gamma. And when they travel to the tissue and into the blood, they recruit other inflammatory cells, for example, other CD4 T cells from the blood into areas of inflammation. And thus, they are a potential useful biomarker to assess the severity and burden of granulomas inflammation in diseases such as sarcoidosis. And in fact, these levels in the blood have been shown to be higher in sarcoidosis. And in other granulomas diseases like tuberculosis. In my prior work using the UCSF longitudinal sarcoidosis cohort, I found that these chemokines were indeed reflective of disease severity. For example, CXCL10, I found that higher levels of CXCL10 was associated with lower values cross-sectionally of lung function, here FEC, FBB1, total lung capacity, and diffusing capacity. And in this graph, I show that patients with higher CXCL10 levels were more likely to have a decrease in their lung function over time as compared to those with lower CXCL10 values. And in comparison, those with higher CXCL9 values were more likely to have more organs involved, either with having more than one organ versus one, or as a continuous variable modeled here. And CXCL11 had a mix between these two patterns. So the goal of this study that I'm showing you today was to validate these findings in a separate cohort by, again, determining the association of these serum cubic kind levels with clinical outcomes, specifically pulmonary function measures measured cross-sectionally as assessed by force vital capacity and FBB1 percent predicted using GLI global equations, as well as diffusing capacity, and total organs affected by sarcoidosis using the WASAG organ assessment tool. I did this using serum samples from the GRADS cohort, which was a multi-center study across nine different sites that measured the transcriptomic and microbiomic signatures in the blood in the lung. Now, I was able to take these serum samples and use techniques that we previously used to measure these chemokine levels using quantikine sandwich ELISA assays. Here's a table of the clinical characteristics of this cohort. I want to draw attention to the fact that above 40 percent of patients had immunosuppression use, and so I had to include that in my models for outcomes. The outcomes specifically I looked at were pulmonary function, and although the average pulmonary function across the cohort was normal, there was a wide range. And similarly for the number of organs involved, although most had one or two organs, there was a range all the way up to 10 organs. And so in terms of the results, these are the findings from multivariate linear regression models that I did after measuring these chemokine values. Here I show the outcome, which is percent predicted FEC, and the beta coefficient from these equations for CXCL9, showing that with higher CXCL9 levels, the lower the FEC percent predicted was, and that was statistically significant. And indeed for all three measures of the pulmonary function values, they were negatively associated with all three of the chemokines. And this is in fact similar to findings from our UCSF cohort, where there was this negative association with these PFT values with these chemokines. And the fact that I saw more statistically significant associations in the grads cohort is probably because I had a higher N and thus a higher power to detect these differences. But overall, the associations were negative in terms of higher chemokine values, lower pulmonary function. In terms of organ involvement, when I modeled the number of organs involved using Poisson regression, which is the count of the number of organs involved, I found that for CXCL9 and 11, the higher the chemokine value, the more likely an additional organ was to be involved with sarcoidosis. And here I show that with the margins plot that has the predicted organ number here adjusted for age, sex, and immunosuppression abuse at each level of chemokine. And here we again see this positive association with the chemokine and the number of predicted organs involved for CXCL9 and 11. And using a logistic regression where I compare those with one or more, two or greater organs, I found that those with higher CXCL11 values were more likely to have an additional or more likely to have more than one organ involved. So what does this mean? Well, it shows that these chemokines are negatively associated with these disease outcomes in sarcoidosis. Specifically, all three of the chemokines were negatively associated with lung function. And CXCL9 and 11 were, again, positively associated with the number of organs involved. And these are, again, similar to the findings we saw in our UCSF cohort. So this really gives credence to the idea that these chemokine levels, when we measure them in the blood, which is where they're supposed to be going to traffic more cells to the sites of inflammation, can be used to assess the severity of granulomas inflammation and thus the disease activity. My next step is to measure these chemokine levels in longitudinal cohorts of newly diagnosed sarcoidosis patients, which is the subject of my K award. And eventually, I want to see how these chemokine values vary after immune suppression, as well as determine if there's unique tissue sources of these chemokines that could make them be useful for assessing specific types of organs involved with the disease. I'd like to, there are many people to thank. I would especially like to thank Laura Koth, who's my clinical and research mentor and heads our sarcoidosis research program at UCSF. I'd like to thank the study participants and the grads' leadership that enabled me to use the serum samples for this study, and especially the Chess Foundation, who awarded me the John Idritzo Research Grant in sarcoidosis, which enabled me to actually do this work. And with that, I thank you and would like to take some questions. Hi. Thank you to the Chess organizers for giving me the opportunity to share our research. We received some grant support, otherwise no other disclosures. So the background to this study, there's been a lot of interest in omega-3 fatty acids, and particularly these long-chain omega-3 acids. They're thought to have potential protective effects in chronic diseases, notably cardiovascular disease, potentially malignancy, and as well as chronic lung disease. It's thought that their downstream metabolites may actually attenuate fibrogenesis and neutrophilic-driven inflammation. A lot of this work has come from Bruce Levy's lab at Bergen Women's Hospital, as well as Patricia Simon at VCU. And the two most notable long-chain omega-3 fatty acids are DHA and EPA, and they've been shown to be protective in bleomyosin urine models. Notably DHA and EPA, they're primarily obtained from either diet, which includes fish and nuts, or supplementation. So there's been recent prior work examining these omega-3 fatty acids with at least lung function. So there's one study from the Blue Journal a few years ago where they pulled a bunch of NIH cohorts, they measured omega-3 fatty acids in the blood, and they found that higher levels of these omega-3 fatty acids, particularly DHA, was associated with higher FUBC. This is a force plot from that figure. Box estimates here to the right indicating higher force vital capacity for every one increment in the DHA level. And then just a few months ago, the same author group, they published studies looking at longitudinal changes in lung function and their relationship with omega-3 fatty acid levels, and they found it very similar. So here, actually, this force plot on the left shows that higher levels of DHA was actually associated with a lower hazard ratio for obstructive lung disease defined by a ratio of less than 0.7. And a few years ago, we actually published a paper in the American Journal of Epi where we examined associations of these omega-3 levels in the blood with interstitial lung disease-related phenotypes amongst community-dwelling adults. And so what we found was that higher levels of DHA and EPA were associated with a lower prevalence of ILA on CT. And then here in one of the cohorts, MESA, we found that adults who had higher levels of DHA, which is the x-axis, actually had a lower risk of ILD-related death and hospitalization. And so this background data led us to the hypothesis that potentially higher plasma levels of these omega-3 fatty acids may be associated with slower disease progression and longer survival in pulmonary fibrosis. So this was an observational prospective study in which we measured omega-3 fatty acids from stored plasma samples, and these were measured at the Wayne State Lipidomic Center. We measured this across three independent cohorts, the Pulmonary Fibrosis Foundation Patient Registry, the University of Chicago, and the University of Virginia. And we used mixed-effect modeling to examine associations of these omega-3 fatty acid levels in plasma with changes in FUVC and DLCO over time. We also used Cox proportion hazard models to examine five-year transplant-free survival. And our primary exposure interest was this N3 index, or omega-3 index, which is the sum of DHA, EPA. Notably, these plasma levels of omega-3 fatty acids indicate about weeks to months of dietary intake. So these are some baseline characteristics of the three independent cohorts. The PFF registry, it was 100% of them were IPF, and IPF was the majority of the UVA-Chicago cohorts. And notably, the UIP pattern on CT was also the predominant CT pattern. And then here we found more never-smokers amongst the UVA cohort compared to the other two. Here, which is probably relevant, we also captured cardiovascular comorbidity. And notably, Chicago had a higher prevalence of concomitant pulmonary hypertension compared to the pulmonary fibrosis registry as well as UVA. And then baseline percent predicted FUVC was also similar across, maybe a little bit higher at UVA. And then DLCO was also notably higher in Chicago in the UVA cohorts. And then antifibrotic usage, as you see there below. So these were our main findings. And so not shown here is that we did not see a strong association between omega-3 levels and force vital capacity, but we did find a strong association with DLCO. So this is a force plot. So the x-axis indicating increments in DLCO, so boxes to the right of that line indicating higher DLCO over time. And what we found was that for every increment in log transform omega-3 index, it actually was associated with a higher DLCO over 12 months. And so you could think of that as slower progression. And notably, this was a mixed effect model that was adjusted for age, sex, smoking, BMI, as well as for the cohort as a random effect. And then this was predicted output from our mixed effect model here. The blue line indicating those with higher omega-3 index levels compared to the orange group, which is lower index 3. This was using a median cutoff. For overall transplant-free survival, we found that in those patients who had higher omega-3 levels, so again, this is indicated by the blue line above the median cutoff. We found that they had better survival compared to those with lower omega-3 index levels. And notably, in the Cox proportional hazards model here, after adjusting for age, sex, smoking, BMI, and baseline lung function, these associations held up. So for every increment in the omega-3 level, a lower risk of transplant or death. And then in the Chicago cohort, we also had leukocytelomere length measured by PCR. And so here we found in this first survival curve, those with lower telomere length actually had better survival if they had higher omega-3 levels. And this was not as strongly seen amongst those with higher telomere length. So in summary, we found that higher levels of omega-3s in blood was associated with a slower decline in DLCO and longer transplant-free survival. Notably, some of these associations were stronger amongst those with shorter leukocyte telomere length. We don't have repeat omega-3 measurements. So potentially, this is just a surrogate marker of overall good health, good behavior. Alternatively, you could think of it as maybe a potential therapeutic target. We probably need mechanistic and intervention studies to further elucidate this. And then notably, there was no significant difference among those with and without cardiovascular comorbidities. So a lot of people to thank, particularly Deji, because he's in the audience here from the University of Chicago. We got a lot of data from his cohort. And then in particular, thanks to Krishna Rao Madhupati at Wayne State who measured these lipids. And it was funded by the K23 award of the PFF. Thanks. Thank you, Dr. Kim. So hello, everyone. My name is Sean. I'm actually a fourth-year medical student at UT Health San Antonio. I'm presenting our project, A Low FEC to DLCO Ratio Helps Increase Clinical Suspicion for Fibrotic HP over IPF. And I worked with this under my mentor, Dr. Nambiar. I have no financial disclosures. And for some background information, fibrotic hypersensitivity pneumoniasis, or fibrotic HP, and idiopathic pulmonary fibrosis, or IPF, are common interstitial lung diseases that can be challenging to differentiate because they do have overlapping presentations of similar symptoms and findings on imaging and histopathology. So we believe there's really a need for a readily available and non-invasive tool to guide diagnosis. And observations we saw in clinics suggested that fibrotic HP patients might actually have a lower baseline ratio of the FEC percent predicted value to the DLCO percent predicted value, or FEC to DLCO, or what we call it the FD ratio, than IPF patients at the time of the confirmed diagnosis. So our objectives were to determine if there was a significant difference in the baseline FD ratios in fibrotic HP patients compared to IPF patients. And we hypothesized that fibrotic HP patients would have lower baseline FD ratios compared to IPF patients due to the more small airway involvement and what we saw in clinic. So for our methods, we performed a retrospective chart review at the UT Health San Antonio Center for ILD. Our key inclusion criteria was a probable or definite diagnosis of fibrotic HP or IPF per ATS guidelines. And key exclusion criteria were poor quality PFTs and a lack of available PFTs within six months of the confirmed diagnosis. And we did perform a secondary analysis, excluding patients with evidence of pulmonary hypertension seen on right heart cath or echo and arresting hypoxemia defined by supplemental oxygen use at rest. Because we believe these would be confounders that would affect the DLCO value and therefore the FD ratio. And key information we collected were just demographics, diagnosis modality, FVC and DLCO values within six months of their confirmed diagnosis, and hemoglobin within three months of their PFTs. And so we calculated each patient's baseline FD ratio using the following formula of the FVC percent predicted value over the DLCO percent predicted value, adjusted for hemoglobin when available, and then compared each group's mean FD ratios. And here's just our main demographics table. Key points are we had 39 patients in the primary analysis for fibrotic HP and 50 patients with IPF. And a majority of them were diagnosed at the end by multidisciplinary discussion, 74% for fibrotic HP and 86% for IPF. And here are the results for our primary analysis with a box and whisker plot. In orange is fibrotic HP and in green is IPF. We had 39 fibrotic HP patients with a mean FD ratio of 1.24, and then 50 patients for IPF with a mean FD ratio of 1.44. And the p-value between these two means was 0.014, indicating that fibrotic HP patients had significantly lower baseline FD ratios than IPF patients. And then in our secondary analysis where we reviewed patients with pulmonary hypertension and or evidence of resting hypoxemia. We then had 33 fibrotic HP patients with a mean FD ratio of 1.22, and then 40 IPF patients with a mean FD ratio of 1.37. And the p-value for these two means, again, was 0.01. So fibrotic HP patients still continue to have lower FD ratios than IPF patients after excluding the confounders. So for our conclusions, inpatients with probable or definite fibrotic HP or IPF, the baseline FD ratio, or FVC to DLCO ratio, at the time of confirmed diagnosis was significantly lower in patients with fibrotic HP than IPF. And we believe that a lower baseline FD ratio could possibly increase clinical suspicion for fibrotic HP over IPF, and this could help reduce the need for invasive procedures like lung biopsies, and potentially help guide initial management, as we know that treatment for IPF and FHP are separate and distinct and could be harmful to one another. And we believe that future larger prospective studies are needed to confirm our findings and see if FD ratios can actually accurately differentiate patients with FHP from patients with IPF. And that's it. Thank you. Good morning. My name is Jiwon Yoo from Bucheon St. Mary's Hospital, College of Medicine, the Catholic University of Korea. Today, I'm going to talk about the validation of progressive pulmonary fibrosis as a predict marker of acute suspicion. I have no financial disclosure. Non-IPF ILD had a heterogeneous phenotype. There are a lot of therapeutic and diagnostic challenges. Criteria for progressive pulmonary fibrosis was proposed recently. In the previous study, they showed the patient with FVC decline of more than 10%, and additional PPF criteria increased the risk of death or lung transplantation. Looking into ILD acute suspicion, ILD-AE is acute morbid deterioration of patient respiratory function leading to hospital admission. We divide this into IPF-AE and the other types of ILD-AE. There is a possibility that ILD-AE patient may present a progressive fibrosis phenotype. The data of risk factor of PPF-AE were insufficient. So our study, we planned in our study to try to prove that the patient who present with PPF had a higher risk of ILD-AE, and to confirm a relationship between PPF-AE and mortality. So our study was conducted in a single center at Bucheon St. Mary Hospital from July 2007 to September 2021, retrospectively. The patient who were diagnosed with non-IPF ILD and at least one PFT performed after the baseline PFT were included. The patient who did not attend their follow-up appointment without baseline HRCT were excluded. So of 183 non-IPF ILD patients, 109 of these patients were divided into the PPF group 32 patients and the non-PPF group 77 patients. We defined PPF according to ADS-ER and JRS guidelines. In order to meet the criteria for PPF, patients had to meet at least two following criteria within the past year with no alternative explanation. These are worsening respiratory symptoms, absolute decline in FVC of more than 5%, or absolute decline in DLCO of more than 10% within one year of follow-up, and radiological evidence of disease progression. Radiological evidence of disease progression included in new GGO and new or increased honeycombing and increased lower volume loss. ILD-AE were defined by an International Working Group report as acute clinically significant respiratory deterioration by evidence of new widespread alveolar abnormality. Patients were diagnosed with ILD with acute worsening or development of dyspnea within one month. The chest imaging of patients showed new bilateral GGO and or consolidation superimposed on a background of UIP pattern. The deterioration was not fully explained by cardiac failure or fluid overload. Let's look at the result. There are no differences between the PPF group and the non-PPF group in the baseline demographics, such as sex, age, BMI, smoking, comorbidity, gap score, and ILD family history. The percentage of ILD classification and initial FVC, initial DLCO, and the percentage of the patient treated with immunosuppressant and or antifibrotics were not distinctive between the two groups. We compared acute expiration and mortality between the two groups. The ratio of acute expiration was higher in the PPF group than in the non-PPF group. Mortality, hypoxia, and the time between diagnosis and onset of AE were not distinctive between the two groups. The mortality was higher in the PPF group than in the non-PPF group. However, this is not significant statistically. We analyzed the risk factor for ILD-AE by univariate and multivariate analysis. PPF was not a risk factor for ILD-AE in multivariate analysis. Higher gap score was a risk factor for ILD-AE. So this study had several limitations. First of all, the number of subjects was quite limited. And this study was conducted in a single center. And we did not apply the other PPF criteria for examination. In conclusion, pulmonary progressive fibrosis patients suffered from acute expiration more. Patients with PPF had higher mortality. For further study, we would like to increase the number of subjects so that we will be able to draw further conclusion. Thank you. immunology background, so therefore I would, my study or presentation will be focused on immunologist's perspective into this disease, sarcoidosis. We were triggered by our work in tuberculosis where we always wanted to look for more than 10 years, infiltrating T-cells and macrophage, their interactions and different molecules which regulates the inflammation, immune inflammation. So similarly in sarcoidosis, similar disease, non-caseating granulomatous inflammations, so there would be involvement of the T-cells as we know and there are positive stimulator, and there are, sorry, there are positive signals and negative signals, we want both of these. However, OX40 is one important positive signal for the T-cells in general, this is memory T-cell, memory effective cell expansion and survival facilitating immune molecule. So we also know, and from our study we have seen that there is indeed an attempt by the host to send the suppressive cells and suppressive molecules, we did some study in immune checkpoint inhibitors as well in the lung infiltrating cells in the bulb. So there is indeed enrichment of T-cells in the lung of sarcoidosis patients. And this T-rex cells, and OX40 actually stimulates, co-stimulates most of these different functional T-cell subsets. However, their impact on regulatory T-cells are not conclusively decided. Some study they said it is positive, some study it says that it inhibits the T-rex cell function and that goes very well with the fact that in spite of tremendous enrichment of T-rex cells in the sarcoidosis lung, there is still perpetual inflammation going on. So some T-rex cells must be failing to mediate its suppressive function in the local milieu, as we showed in our rheumatoid arthritis patients and synovial fluid. So as you see here, there is enrichment of inflammatory cytokines, especially where I think this is, I have not seen before that IL-9 to TH9 cell are quite enriched and we'll show that TH9 indeed is hierarchical cytokines that facilitates other inflammatory cytokines in that. We compared basically peripheral blood of the pulmonary sarcoidosis versus bowel. So there is always enrichment of all these inflammatory cells, including TH9. Now TH9 is a very interesting cytokine. Initially we used to believe that it is a TH2 because TH2 likes cytokine because it is tightly associated with allergic and parasitic conditions. However, we showed in rheumatoid arthritis and similarly here we show as well that if we block, actually TH9 cells, they actually augment the inflammatory T cells in many diseases, including pulmonary sarcoidosis. So if we block the TH9 signaling by blocking this TH9 receptor, which are expressed on the affected T cells, what you find that there is a reduction of the percentage positive cells of these cytokines. On the other hand, if we, in the cells from the bowel, if we stimulate them in presence of these inflammatory cytokines, in presence of IL-9, these inflammatory cytokines are elevated. However, vice versa is not true. The other cytokines, they cannot affect the frequency of the TH9 cells. So therefore, we propose, and we proposed in our previous paper on rheumatoid arthritis, that IL-9 is probably a hierarchic TH1-like cytokines. It regulates the other inflammatory cytokine-producing cells but others cannot regulate it. So therefore, controlling IL-9 would be a better candidate, blocking IL-9 would be a better candidate to inhibit the inflammatory diseases. So as we see here, there is indeed enrichment of regulated, sorry, regulated T cells in the sarcoid recess bowel. And these regulated T cells, they fail to produce when we stimulate them in vitro after retrieving them from the bowel. These regulated T cells, when they are stimulated in vitro, they are, if you see, relative to the peripheral blood, bowel-derived synovial Treg cells, they are always deficient in terms of their IL-10 and TGF beta production, which are actually immunosuppressive cells. So this tells us that there is indeed an enrichment. Body is trying to send the controlling cells to control the inflammation. However, in that milieu, they are failing to execute their suppressive function. And this can be attributed to the inflammatory cytokines. This slide shows that these cytokines, if it is blocked, then it can restore the Treg cells, suppressive cytokine production capacity, as we observed that there is gain of frequency of IL-10 positive and TGF beta positive Treg cells derived from the bowel. So this suggests that there is indeed failure in incapacitation of Treg cells. Those are accumulated in the bowel of PS patients. And we also see that, we tested actually a variety of suppressive, co-stimulatory and co-inhibitory molecule. We see that OX40 expression, percent positive cells for the OX40 on the Treg cells, actually both in effector and regulatory T cells, OX40 positive T cells are very high. And OX40 ligand is also very high, relatively higher in number in PS bowel fluid. So suggesting that OX40 axis is actually elevated, and which here we show that this inflammatory cytokines, which are present in the disease milieu, they indeed can both can actually give rise to the higher number of OX40 positive cells, both in effector arm as well as regulatory T cell arms. So this cytokine actually helps in expression of OX40, and it is understandable because these OX40 molecules are supposed to generate the effector memory cells through their survival and expansion. So interestingly, we observed that OX40, even though it is memory generation molecule, so if we stimulate the cells, effector cells in presence of agonistic antibody, which is stimulating antibody, we find that both the cytokines, their numbers are increased positive cells. However, this will be positive. On Treg cells, actually, it is just reverse. It reduces the cytokine production, which are suppressive cytokine in case of Treg cells, significantly, both IL-10 and TG-beta producing cells, suggesting that this molecule actually has a differential impact in terms of stimulation or suppression, if we consider two different subsets of functional subsets of T cell effector and Treg cells. OX40 actually, rather than stimulating the function, cytokine, suppressive cytokine production function, it actually reduces or decreases or inhibits the suppressive cytokine production by the Treg cells, which are accumulated. So that's what is our finding. And we did some, what should I, so silencing study. Here, we show that through silencing study, we demonstrate that if we silence the OX40 on Treg cells, actually, we get gain of percent positive cells, which are positive both for these two cytokines, which are suppressive in nature. So therefore, inflammation, inflammatory cytokine is actually increasing the OX40 positive cells. And this OX40, however, is having inhibitory impact on the Treg cells, which are supposed to inhibit the inflammations. Therefore, in spite of enrichment, the Treg cells are failing to execute their suppressive function and to control the inflammation in this case. So what we add a little bit in this field? I think perhaps for the first time, we show in PS that IL-9 plays an important role and OX40 pathway plays some role in the pathogenesis. But it's trying to send the good cells to inhibit the inflammation of the disease, but it's failing. And indeed, here we demonstrate again, like RA and tuberculosis, as we have studied, that IL-9 actually a hierarchic cytokine in terms of inflammatory cascade. Therefore, controlling them or regulating them would be better approach in controlling several inflammatory disease. And OX40 signaling, I'm not presenting those signaling data, mTOR and others. OX40 signaling actually renders the B cells, which are infiltrating the active disease site in PS. It renders those cells functionally incapacitated, rendering them actually incapable of controlling the inflammation for which actually they are being recruited. And so blocking OX40 may be an interesting way to restore the immune suppression of the Treg cells, which are accumulated in the lung. And at the same time, OX40 inhibition can inhibit the inflammatory cells. So it works from both the angle. It inhibits the inflammatory cells and it restores the inhibitory function of the regulatory T cells. It may be very effective potential therapy for the treatment resistant cases. Thank you, Dr. Mitra. Yeah, we're done. So this is the graphic representation. So I'm not going into detail. So these are our group. She's a graduate student. She was a Commonwealth fellow, came from Heidelberg. And she was another graduate student in our department, Professor Anand Mohan and myself. Thank you very much. It is published already in the church. Thank you.

sarcoidosis

chemokines

CXCL9

CXCL10

CXCL11

lung function

GRADs cohort