Prognostic scoring system for pediatric Epstein-Barr virus-associated hemophagocytic lymphohistiocytosis based on baseline characteristics: A multicenter retrospective study

Background: The prognosis of pediatric Epstein-Barr virus (EBV)-associated hemophagocytic lymphohistiocytosis (EBV-HLH) varies. This study aimed to identify high-risk children early.

Procedure: Data from 264 children (0-14 years of age), diagnosed with EBV-HLH at six centers in China between January 2016 and December 2021, were analyzed. Patients were randomly divided into derivation (n = 185) and verification (n = 79) cohorts. A Cox regression model was used to explore risk predictors and establish a prognostic scoring system for death events that occurred during the follow-up period.

Results: Chronic active EBV infection (CAEBV) history (hazard ratio [HR] 1.82 [95% confidence interval, CI: 1.02-3.26]; p = .0441), plasma EBV-DNA more than 10⁴ copies/mL (HR 2.89 [95% CI: 1.62-5.16]; p = .0003), pulmonary infection (HR 2.24 [95% CI: 1.06-4.75]; p = .0353), digestive tract hemorrhage (HR 2.55 [95% CI: 1.35-4.82]; p = .0041), and hypoxemia (HR 3.95 [95% CI: 2.15-7.26]; p < .0001) were independent risk factors. Accordingly, the CAEBV history, plasma EBV-DNA copy number, pulmonary infection hemorrhage of digestive tract, hypoxemia prognostic scoring system (CEPHO-PSS) were developed, which separated patients into low- (0-1 points), middle- (2-3 points), and high- (4-8 points) risk groups. Survival curves for the three groups exhibited statistically significant differences (p < .0001). Internal and external verification of CEPHO-PSS was performed using receiver operating characteristic (ROC) and calibration curves in the derivation and verification cohorts, respectively, confirming good accuracy and applicability.

Conclusions: The CEPHO-PSS identified three risk groups with statistically significant differences in survival curves. It was based on the baseline characteristics, and can give clinicians a convenient check for risk prediction.