Paul Tsoukas, MD
Vasculitis Research Fellow
Hospital for Sick Children
Toronto, Ontario, Canada
Background: This study describes a novel subgroup of Multisystem inflammatory syndrome in children (MIS-C) patients that develop a systemic juvenile idiopathic arthritis (sJIA)-like illness. Their clinical characteristics, treatment response and genetic data were compared to cohorts of MIS-C and sJIA patients.
Methods: Clinical data and biospecimens from treatment-naïve MIS-C patients (n=204) were collected prospectively, while sJIA patients’ clinical data was obtained retrospectively from the onset of the SARS-CoV2 pandemic (n=12). All hospitalized MIS-C patients were adjudicated by a multidisciplinary team, treated with immunosuppressive agents and fulfilled the Royal College of Paediatrics and Child Health case definition. 91% and 73% of the co-diagnosis patients satisfied the Paediatric Rheumatology INternational Trials Organisation sJIA, and Yamaguchi criteria, respectively. Clinical exomes (n=10; co-diagnosis patients) and whole genomes (n=249: 68 MIS-C, 181 acute COVID controls as part of the CanCoGen project) were performed using Illumina platforms and sequencers. Publicly available genome-wide association study summary statistics for COVID-19 severity (Host Genetic Initiative) and sJIA (Ombrello et al. 2017) were utilized for calculating the PRS via the Pruning+Thresholding method. Best models were selected per the highest variability explained using Nagelkerke R-squared values.
Results: 5.3% of our MISC cohort (11/204) were diagnosed with an sJIA-like illness. Overall, all groups are inflammatory. When compared to MISC (n=193) and sJIA (n=12), co-diagnosis patients share similar clinical characteristics and laboratory investigations to both groups, with each cohort varying in the intensity of the immune response along a hyperinflammation spectrum. Patients diagnosed with both sJIA and MIS-C show a more pronounced inflammatory response evidenced by the longest duration of fevers, as well as the highest ESR, leukocyte, and maximum platelet counts. (Table 1) Co-diagnosis patients resemble MIS-C in terms of KD features and cardiac involvement. Indeed, MIS-C and sJIA share clinical features, but per the higher prevalence of MAS in sJIA, cohorts with an sJIA diagnosis exhibit greater ferritin and D-dimer levels.
Variants known to be enriched in sJIA were found in 5/9 patients who underwent clinical genetic testing. To investigate shared genetic contributions with MIS-C, PRSs were calculated for COVID-19 severity and sJIA. No significant association between MIS-C and acute COVID-19 severity PRS (p=0.451, OR = 1.023, 95% CI [0.9645,1.0845]) was observed. However, the sJIA PRS was found to be significantly associated with MIS-C (p=0.014, OR = 1.063, 95% CI [1.0136,1.1182]) after Bonferroni correction.(Figure 1)
Conclusion: MISC and sJIA exhibit shared clinical features, treatments and genetic contributions, suggesting a shared immunobiology.