Clinical features associated with macrophage activation syndrome in juvenile dermatomyositis: implications for early recognition

Abstract

This study aimed to characterize the clinical and laboratory features associated with juvenile dermatomyositis-associated macrophage activation syndrome and to identify warning features that may facilitate recognition of concurrent MAS at JDM presentation. The distribution and potential discriminatory contribution of myositis-specific antibodies were also explored. This study retrospectively reviewed patients with JDM diagnosed and treated at Beijing Children’s Hospital between January 2015 and January 2025. MAS was clinically diagnosed during routine care and retrospectively adjudicated by pediatric rheumatologists, with the 2016 EULAR/ACR classification criteria for systemic juvenile idiopathic arthritis-associated MAS used as the principal reference framework. Least absolute shrinkage and selection operator regression combined with 10-fold-cross validation and Firth penalized regression were used to identify features independently associated with MAS. Model performance was evaluated using the precision–recall area under the curve (PR-AUC) with 1,000 bootstrap resamples for internal validation. The incremental value of MSA status was further evaluated in patients who underwent MSA testing. Among 461 patients with JDM, 15 (3.3%) had MAS. Patients with MAS had a shorter symptom duration before JDM diagnosis than those without MAS ( p = 0.011). V-neck rash (aOR 169.65), dyspnea (aOR 11.38), lower hemoglobin (aOR 0.92), elevated C-reactive protein (aOR 15.24), and lower albumin (aOR 0.82) were independently associated with concurrent MAS. The study model yielded a PR-AUC of 0.875, with a mean bootstrap PR-AUC of 0.876. Among 238 patients with MSA testing, anti-NXP2 was the most common MSA subtype among patients with MAS. Anti-NXP2 positivity was not independently associated with MAS, although its inclusion increased the PR-AUC from 0.753 to 0.789. JDM associated MAS was uncommon but was usually present at the time of initial JDM diagnosis in this cohort. V-neck rash, dyspnea, lower hemoglobin, elevated CRP, and lower albumin should be interpreted as complementary warning features of concurrent MAS. Clinicians should maintain a low threshold for further MAS evaluation when newly diagnosed patients with JDM exhibit disproportionate systemic inflammation or multiorgan involvement. The potential discriminatory contribution of anti-NXP2 antibody remains exploratory and requires validation in larger prospective cohorts.