Estimating the minimal clinically important difference for the Myasthenia Gravis Quality of Life revised scale (MG-QOL15R)

PURPOSE: To estimate the minimal clinically important difference (MCID) for the MG-QOL15R. METHODS: Data from two multi-country myasthenia gravis studies were used: the ADAPT RCT (n = 157) and the MyRealWorld-MG (MRW-MG) survey (n = 92). MCIDs were estimated using four anchor-based methods: change difference (CD), receiver operating characteristic (ROC) curve with Area Under the Curve (AUC), linear regression, and equipercentile linking. MG-Activities of Daily Living (MG-ADL) was the anchor. MCIDs were applied to the ADAPT RCT to assess the impact of different thresholds. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated to estimate the odds of achieving an MCID threshold at week 4 in the efgartigimod arm compared with placebo among patients positive for acetylcholine receptor antibodies. RESULTS: Baseline mean age was 46.7 years old in ADAPT and 49.8 years old in MRW-MG; women comprised 71.3% and 71.7%, respectively. ADAPT included only generalized MG, while MRW-MG included 12.0% ocular and 88.0% generalized MG. Average disease severity (MG-ADL 9.0 vs. 6.3) and HRQoL impairment (MG-QOL15R 16.3 vs. 12.0) were greater in ADAPT. MCID estimates were: CD (ADAPT 1.6; MRW-MG 2.8), ROC (both 2.5, AUCs 0.66; 0.71), linear regression (2.1; 1.6), and equipercentile linking (2.0; 3.0). Applying MCIDs of  ≥ 2.0 and ≥ 3.0 resulted in ORs of 3.31 (95%CI 1.65–6.87) and 4.14 (95%CI 2.11–8.38). CONCLUSIONS: MCID estimates ranged between 1.6 and 3.0 across both studies. MCID thresholds of 2- and 3-points may indicate a minimal clinically meaningful change for monitoring progress and guiding treatment. Future research should use patient global impression of change anchors to improve interpretability and clinical relevance of MCID estimates.