A prognostic scoring model (ALARN) was developed and validated to predict all-cause mortality in 1,194 patients with antisynthetase syndrome (ASyS) across three Chinese institutions, using Cox proportional hazards and random survival forest methods, with external validation in 431 patients.
The ALARN score — built on age at onset, lactate dehydrogenase, albumin, respiratory failure, and neutrophil-to-lymphocyte ratio — achieved time-dependent AUCs of 0.914 (1-year), 0.867 (3-year), 0.839 (5-year), and 0.854 (10-year) for mortality prediction; scores 0–2, 3–5, and 6–8 defined low, intermediate, and high risk with significantly different survival (log-rank P < 0.0001).
- Retrospective design limits causal inference. - All patients were from Chinese institutions, which may limit generalizability to other populations. - Specific autoantibody subtypes and treatment effects were not incorporated into the model.
Clinicians managing ASyS can use the five-variable ALARN score (all routine labs/clinical data) to stratify patients into low, intermediate, or high mortality risk at diagnosis. High-risk patients (score 6–8) may warrant closer monitoring and earlier escalation of therapy.
