Diagnosis of myositis-associated interstitial lung disease: Utility of the myositis autoantibody line immunoassay

Objectives: The detection of myositis autoantibodies (MA) in patients with interstitial lung disease (ILD) has major implications for diagnosis and management, especially amyopathic and forme frustes of idiopathic inflammatory myositis-associated ILD (IIM-ILD). Use of the MA line immunoblot assay (MA-LIA) in non-rheumatological cohorts remains unvalidated. We assessed the diagnostic performance of the MA-LIA and explored combined models with clinical variables to improve identification of patients with IIM-ILD. Methods: Consecutive patients referred to a specialist ILD clinic, with ILD-diagnosis confirmed at multidisciplinary meeting, and MA-LIA performed within six months of baseline were included. Pre-specified MA-LIA thresholds were evaluated for IIM-ILD diagnosis. Results: A total 247 ILD patients were included (IIM-ILD n = 12, non-IIM connective tissue disease-associated ILD [CTD-ILD] n = 52, idiopathic interstitial pneumonia [IIP] n = 115, other-ILD n = 68). Mean age was 64.8 years, with 45.3% female, mean FVC 75.5% and DLCO 59.2% predicted. MA were present in 13.8% overall and 83.3% of IIM-ILD patients. The most common MA in IIM-ILD and non-IIM ILD patients were anti-Jo-1 (prevalence 40%) and anti-PMScl (29.2%) autoantibodies respectively. The pre-specified low-positive threshold (>10 signal intensity) had the highest discriminative capacity for IIM-ILD (AUC 0.86). Combining MA-LIA with age, gender, clinical CTD-manifestations and an overlap non-specific interstitial pneumonia/organising pneumonia pattern on HRCT improved discrimination for IIM-ILD (AUC 0.96). Conclusion: The MA-LIA is useful to support a diagnosis of IIM-ILD as a complement to multi-disciplinary ILD assessment. Clinical interpretation is optimised by consideration of the strength of the MA-LIA result together with clinical and radiological features of IIM-ILD.

Automated summary

The study evaluated the diagnostic performance of MA-LIA for IIM-ILD and explored models combining clinical variables to improve patient identification. Results showed that combining MA-LIA with age, gender, clinical manifestations and HRCT features can enhance discrimination for IIM-ILD.