OBJECTIVES: Psychometrically validate and derive meaningful change thresholds for the Inflammatory-Rasch-built Overall Disability Scale (I-RODS) and Modified Rasch-built-Fatigue Severity Scale (R-FSS) using clinical trial data on Riliprubart in Chronic Inflammatory Demyelinating Polyneuropathy (CIDP) patients.

METHODS: Interim data from a multinational, multicenter phase 2 study of adults with CIDP (NCT04658472, n=64) including subpopulations Standard Of Care (SOC) Treated (n=34), SOC-Refractory (n=18) and SOC-Naïve (n=12), were analyzed. Analysis investigated reliability (internal consistency, test-retest), construct validity (convergent and known-groups validity, discriminant validity for I-RODS), structural validity (Rasch analysis) and responsiveness to change. Clinically meaningful change thresholds for improvement were derived for individual level (Response Definition, RD) and for between and within group change (Clinically Important Difference, CID) using distribution- and anchor-based approaches.

RESULTS: I-RODS (baseline mean=51.52, SD=20.54): Floor/ceiling effects were found for some items, however, not on the score. Internal consistency and test-retest reliability were strong (Cronbach’s α>0.9; Intraclass Correlation Coefficients [ICC]=0.848). Moderate evidence supported convergent, discriminant, and structural validity and responsiveness (convergent: absolute correlations with Inflammatory Neuropathy Cause and Treatment Disability scale [INCAT], European Quality of Life Group 5 Level [EQ-5D-5L] Mobility and Usual Activities, Patient Global Impression-Severity [PGI-S], and Physician’s Global Assessment-Severity [PhysGA-S]: 0.46-0.79; discriminant: correlation with EQ-5D-5L Anxiety/depression dimension=-0.41; structural: mixed fit to Rasch model; responsiveness: improvement by PGI-S, effect size [ES]=0.532, and PGI-Change [PGI-C], ES=0.542). Ranges for RD were 3-16 and 8-10 for CID. R-FSS (baseline mean=14.73, SD=6.47): Few floor but several ceiling effects were observed for items, however, not on the score. Strong reliability, convergent and structural validity were found (reliability: α>0.9; ICC=0.861; convergent: correlation with PGI-S-Fatigue [PGI-S-F]=0.54; structural: R-FSS fits the Rasch model). Moderate responsiveness was found (improvement by PGI-S-F, ES=-0.746, and PGI-C-F, ES=-0.735). RD was 2-8 and CID was 3-4.

CONCLUSIONS: I-RODS and R-FSS are acceptable as endpoint measures in clinical trials in adult CIDP patients.