POS0905 CHARACTERISATION OF SWEDISH MYOSITIS PATIENTS WITH ANTI-MDA5 AUTOANTIBODIES AND CORRELATION OF CLINICAL FEATURES WITH AUTOANTIBODY LEVELS

Abstract

Background: The association between anti-melanoma differentiation association protein 5 autoantibodies (aMDA5) and rapidly progressive interstitial lung disease (RP-ILD) in clinically amyopathic dermatomyositis is well established in Asian population cohorts. In western cohorts, ILD has been strongly associated with aMDA5 but data regarding RP-ILD have been more conficting. It is also suggested that western cohorts have more pronounced myopathic features than Asian. Objective(s): To characterise the disease manifestations of a Swedish aMDA5 positive idiopathic infammatory myositis (IIM) cohort and to explore antigen reactivity of the MDA5 protein. Method(s): First available serum samples collected from 28 consecutive patients with IIM and positive aMDA5 ever tested by ELISA, Line Blot (LB) or Immunopre-cipitation, attending Karolinska University Hospital between 1999 and 2021, were included. Clinical data including presence of anti-SSA autoantibodies by ELISA or LB was retrieved retrospectively. An in-house ELISA was used to screen serum samples for reactivity against a recombinant MDA5 protein (rMDA5, aa A110-D1025, UniProt ID Q9BYX4) and seven MDA5-derived constructs containing different domains. Correlations between aMDA5 reactivity levels and clinical data were explored. Result(s): Nine patients showed no reactivity to any of the rMDA5 constructs by ELISA and were excluded from further analysis. Reactivity against rMDA5 was confrmed by ELISA in 19 patients (median 184.7 mu g/mL (interquartile range (IQR) 277.07). The cohort included 13 male and 6 female patients, 94% Caucasian, with mean age at diagnosis of 41.05 years (standard deviation (SD) 10.5). Median disease duration at time of sampling was 0 months (IQR 1). All patients except one had signs of muscle involvement (muscle weakness, elevated muscle enzymes, muscle oedema or muscle biopsy consistent with myositis). At diagnosis 63.2% of patients reported muscle weakness (21.1 % had a manual muscle test 8 score <75). Dermatological fndings were observed in 17/19 (89.7 %). During disease course nine patients (47.4%) had confrmed arthritis. ILD was diagnosed in 16/19 patients (84.2%), four of these (25%) developed a RP-ILD. One patient passed away due to RP-ILD and one required a lung transplant. Patients with ILD had a statistically signifcant higher mean age at diagnosis than those without (42.8.5 (SD 10.3) vs 31.3 (SD 4.7) years, p=0.02). Patients developing RP-ILD were not signifcantly older than patients with chronic ILD. Respiratory symptoms were reported by 75% of patients with ILD at time of diagnosis. The mean total lung capacity (TLC) of the ILD cohort was 68% (SD 17), mean diffusion capacity of carbon monoxide (DLCO) was 59% (SD 15) and mean forced vital capacity (FVC) was 62% (SD 19). There was a higher proportion of patients with CRP >= 3 times the reference range at diagnosis amongst patients with FVC <70 % than patients with FVC >70 % (88.9 % vs 16.7 %, p= 0.01). Ten patients (52.6%) had anti-SSA autoantibodies, all had ILD. Anti-SSA positive patients had a statistically signifcant lower TLC than those without (62% vs 79% respectively, p=0.04) and a lower FVC (57% vs 76% respectively, p=0.05). We found a weak non-statistically signifcant negative correlation between titres of aMDA5 and TLC, DLCO and FVC (Pearson coefficients-0.187,-0.289,-0.130 respectively). Frequency of ILD was higher in patients with aMDA5 titres >100 mu g/mL than those with titers <100, but not statistically signifcant (81.3% vs 18.8%, respectively). Conclusion(s): In this Caucasian cohort of aMDA5 positive IIM patients, ILD was present in over 80% of patients, of these, one quarter had RP-ILD. Older patients were more likely to present with ILD. Anti-SSA positivity and higher CRP levels were associated with worse lung function. We found a weak negative correlation between aMDA5 titres and lung function tests, as well as a trend of higher frequency of ILD in patients with higher aMDA5 titres. Muscle and skin involvement were found in a high proportion of patients.