• dermatomyositis
• polymyositis
• autoantibodies

We are grateful for the interest in our work1 shown by Drs Yang and Liang. In their correspondence regarding this work,2 they raise concerns about (1) the association between anti-Ro52 autoantibodies and interstitial lung disease (ILD) in juvenile polymyositis (JPM) and juvenile connective tissue disease–myositis (JCTM), (2) the appropriateness of adjusting for duration of follow-up instead of length of time from onset to diagnosis and (3) the lack of statistical power to draw some conclusions.

First, as the prevalence of ILD between anti-Ro52-positive and anti-Ro52-negative patients was not statistically significant in the JPM and JCTM subgroups, we agree that larger studies will be necessary to confirm these tentative associations.

Second, as shown in table 2, the time from onset to diagnosis was very similar in anti-Ro52-positive (0.55 years) and anti-Ro52-negative patients (0.75 years, p=0.3). In contrast, the duration of follow-up trended towards being longer in anti-Ro52-negative patients (6.0 vs 4.3 years, p=0.09). For this reason, we chose to include duration of follow-up as a covariate in the multivariate analysis.

Third, the number of anti-Ro52 patients was large enough to detect highly significant differences in the multivariate analysis. For example, the prevalence of ILD in anti-Ro52-positive patients was 36%, while it was just 4% in anti-Ro52-negative patients, independent of the duration of follow-up, year of onset and the presence of myositis-specific autoantibodies (p<0.001). The low number of anti-Ro52-positive patients in some of the autoantibody groups did not affect these key findings.