162 - A HOST PROTEIN SIGNATURE TO DISTINGUISH KAWASAKI DISEASE FROM OTHER INFECTIOUS AND INFLAMMATORY DISEASES

BACKGROUND/

Aim:   

There is an urgent need for a test to diagnose Kawasaki Disease (KD) as the clinical criteria are shared by many other infectious and inflammatory diseases. It is essential to achieve a diagnosis quickly as early treatment can prevent acute and long-term cardiac complications. The emergence of the COVID-19 pandemic further complicated diagnosis as approximately 40% of patients with Multisystem Inflammatory Syndrome in Children (MIS-C) met the diagnostic criteria for KD due to similarities in clinical manifestations. We therefore aimed to identify a protein signature that distinguished KD from MIS-C and other diseases as the basis of a diagnostic test.

Methods:

We compared the plasma protein abundance profiles of children with KD (n = 39), MIS-C (n = 79), definite bacterial (DB; n = 40) and viral infections (DV; n = 43), and healthy controls (HC; n = 24) using the SomaScan^® aptamer-based proteomic analysis of over 7000 proteins. Patients were recruited to the NIH-funded PREVAIL and the EU-funded PERFORM and DIAMONDS studies. Significantly differentially abundant (SDA) proteins between KD and each comparator group were identified. Feature selection was used to establish diagnostic signatures that could optimally discriminate KD from other diseases. The findings were validated using antibody-based assays on an independent validation cohort. The accuracy of the models was assessed with the area under the receiver-operating characteristic curve (AUROC).

Results:

736 proteins were SDA when comparing KD against combined comparator disease groups (MIS-C + DV + DB), with 388 proteins found to increase in KD and 348 decreased. Two sparse 4-protein diagnostic signatures were identified. Using all SDA proteins, the best-performing 4-protein signature had an AUROC of 98.1% (95% CI: 95.1%-100%). To enable easier detection of the proteins on platforms more suited for clinical use, we identified an additional signature based only on SDA proteins that were highly abundant and increasing in KD. This translational signature had an AUROC of 94.3% (95% CI: 89.6%-99%). 

Conclusion:

Our study establishes that as few as 4 proteins can distinguish KD from other paediatric infectious and inflammatory syndromes. The small number of proteins comprising both signatures makes translation into a point-of-care diagnostic test feasible.