Development and Validation of the ALARN Model: A Predictive Mortality Score for Antisynthetase Syndrome

Area of Science:

• Rheumatology and Immunology
• Clinical Prognostics
• Epidemiology

Background:

• Antisynthetase syndrome (ASyS) is a rare autoimmune disease associated with significant morbidity and mortality.
• Accurate prognostic models are crucial for managing ASyS patients and improving outcomes.

Purpose of the Study:

• To develop and validate a reliable prognostic model for predicting all-cause mortality in patients diagnosed with ASyS.
• To create a simplified scoring system for practical clinical application.

Main Methods:

• A retrospective study involving 1194 ASyS patients from three Chinese institutions.
• Development of a mortality risk prediction model using Cox proportional hazards (CPH) and random survival forest (RSF) methods in a derivation cohort (n=763).
• Validation of the optimal model, simplified into the ALARN score, in an external cohort (n=431).

Main Results:

• The CPH-derived model identified age at onset, lactate dehydrogenase, albumin, respiratory failure, and neutrophil-to-lymphocyte ratio as key prognostic factors.
• The simplified ALARN score demonstrated strong predictive performance, with time-dependent AUCs for 1-, 3-, 5-, and 10-year mortality ranging from 0.839 to 0.914.
• The ALARN score effectively stratified patients into low, intermediate, and high-risk groups with significantly different survival outcomes (log-rank p<0.0001).
• Both internal and external validation confirmed the robustness of the ALARN score.

Conclusions:

• The ALARN score is a simple and practical tool for predicting mortality risk in ASyS patients.
• It incorporates five readily available clinical parameters, facilitating its use in routine clinical practice.
• The ALARN score aids in risk stratification and personalized management of ASyS.