Analysis of Clinical Cytokine Panel Profiles As Diagnostic and Prognostic Biomarkers in Pediatric Hyperinflammatory Conditions

Abstract

Background: Cytokines are secreted intracellular signaling peptides that execute a wide variety of essential physiologic functions including cellular growth, differentiation, inflammation, immune regulation, and wound healing. Cytokine storm syndromes are characterized by uncontrolled inflammation and cytokine hyperexpression, often leading directly to disease pathology. Patterns of cytokine expression have been increasingly used as clinical biomarkers for disease diagnosis, risk assessment, disease progression, and selection of targeted therapy. Although cytokine panels developed for clinical use have been utilized at multiple centers, literature on their clinical utility and diagnostic accuracy is lacking. Methods: Our interdisciplinary group has developed clinical cytokine testing in a CLIA/CAP certified lab to help with the diagnosis and management patients with cytokine storm syndromes including during the SARS-CoV-2 pandemic. Greater than 360 cytokine panels were collected over a 2-year period. Thus far the electronic medical record of 170 patients between the ages of 0 - 28 years who received one or more cytokine panel during that time were manually reviewed. Complete analysis of all cytokine panels during this period will be presented. Demographic, diagnostic, prognostic, and laboratory data were collected from each patient. Cytokines measured for each patient include interleukin (IL)-2, IL-4, IL-1beta, IL-13, IL-12p70, IL-10, IL-8, IL-6, tumor necrosis factor alpha (TNFa) and Interferon gamma (IFNg). Results: Preliminary results demonstrate 81% of patients underwent cytokine testing only once, with 19% receiving two or more cytokine tests. In 18.5% of cases the cytokine panel changed the diagnosis or redirected management. Of the times the cytokine panel changed management 5.6% added a JAK inhibitor, 22.5% added anti-cytokine biologic therapy, 10% added or broadened antimicrobials, 25% started steroids, 7% started other immunosuppression, 2.8% attempted cytokine removal via PLEX, and 24% did not start or decreased cytokine-directed/steroid therapy. Cytokine profiles were also analyzed by diagnosis including primary and secondary hemophagocytic lymphohistiocytosis (HLH), COVID-19, MIS-C, ARDS, septic shock, and cytokine release syndrome from immunotherapies including CAR-T. Three cross-comparisons were highlighted in our results - shock vs acute respiratory distress syndrome without shock, multi-system inflammatory syndrome in children (MIS-C) vs. severe COVID-19/SARS-COV-2, and shock vs. infection without shock. Additional analyses comparing HLH and septic shock and CART with septic shock are ongoing. Shock and acute respiratory distress syndrome (ARDS) without shock are often difficult to differentiate clinically. IL-6 was found to be significantly different between these two disorders (p-value: 0.008); both having a wide range of values, but ARDS with an interquartile range between 10 - 225 pg/mL, whereas shock had broader range from 0 - 2000 pg/mL. MIS-C and COVID-19/SARS-COV-2 had a large representation in the first half of our study. It can be difficult to differentiate these two entities, therefore reliable biomarkers are needed to direct disparate treatment regimens. IFNG values for MIS-C range from 0 - 9000 pg/mL, with interquartile range of 225 - 1250 pg/mL, whereas the values for COVID-19 range from 0 - 1300 pg/mL, with interquartile range of 20 - 250 pg/mL (p. value: <0.0001). Finally, shock and infection without shock can present with similar symptoms, but have vastly different treatments, making it very important to be able to differentiate in clinical settings. Again, a significant difference was found in IL-6 (p-value: 0.001). Shock has a broad interquartile range from 0 - 2000 pg/mL, whereas infection ranges from 0 - 5700 pg/mL, with an interquartile range of 10 - 600 pg/mL. Conclusion: This retrospective review of cytokine panel usage at a single center is consistent with prior reports suggesting that cytokine profiles have important clinical utility. Our study supports the conclusion that cytokine profiles can be used as biomarkers to distinguish between similarly presenting disease entities, assess risk of disease progression, and help guide management decisions around when to start or stop immunosuppressive therapy, including cytokine-directed therapy.