# Comparable area under the curve for three risk scores to detect interstitial lung disease in patients with rheumatoid arthritis: an external validation

## TL;DR

This study validates three risk scores for detecting lung disease in rheumatoid arthritis patients and finds no strong link between a gene variant and lung changes in northern Sweden.

## Contribution

External validation of three RA-ILD risk scores and investigation of MUC5B variant in a northern Swedish RA cohort.

## Key findings

- All three risk scores showed comparable AUC ROC values (0.70–0.75) in detecting RA-ILD.
- MUC5B promotor variant frequency was 26% but not significantly linked to subclinical lung changes.
- Risk scores performed well and could be used clinically for RA-ILD detection.

## Abstract

Rheumatoid Arthritis associated Interstitial Lung Disease (RA-ILD) is an extraarticular manifestation of rheumatoid arthritis (RA) associated with increased morbidity and mortality. Several risk factors for RA-ILD have been identified, one is the promotor variant of mucin 5 B gene (MUC5B). Juge et al., Wheeler et al. and Koduri et al. have developed risk scores for identifying patients with RA at risk of RA-ILD. We aimed to externally validate the three risk scores and further investigate the frequency of MUC5B promotor variant and its association with subclinical lung changes in patients with RA in northern Sweden. Our cohort consisted of 54 patients with RA. The risk score variables were evaluated in binary logistic regression and validated using area under the receiver operating characteristics curve (AUC ROC). The genetic material was purified and genotyped for MUC5B. The Juge et al. risk score performed an AUC ROC of 0.71 (95% CI 0.57;0.86), the Wheeler et al. risk score an AUC ROC of 0.75 (95% CI 0.59;0.90) and Koduri et al. risk score an AUC ROC of 0.70 ((95% CI 0.55;0.85) in our cohort. The differences in AUC were not statistically significant. The MUC5B promotor variant frequency was 26% (n = 14). In our cohort, MUC5B was not significantly associated with subclinical lung changes. The three externally validated risk scores for RA-ILD performed well in this cohort and could be used clinically. In patients with RA in northern Sweden, MUC5B was not found to be independently associated with subclinical RA-ILD.

The online version contains supplementary material available at 10.1007/s00296-025-06005-z.