Assessment timing and choice of outcome measure in determining treatment response in chronic inflammatory demyelinating polyneuropathy: a post-hoc analysis of the PRISM trial

Abstract

Introduction/Aims: Treatment response and its timing are variable in chronic inflammatory demyelinating polyneuropathy (CIDP). In this study we assessed the variability using multiple outcome measures. Methods: We performed a post hoc analysis of the PRISM trial, a 24-week prospective, multicenter, single-arm, open-label, phase III study of a 10% intravenous immunoglobulin preparation for CIDP. We ascertained timing of response with primary/secondary outcome measures. Results: At 6 weeks after treatment initiation, 13 of 40 subjects (32.5%) were defined as responders on the primary outcome measure, the adjusted Inflammatory Neuropathy Cause And Treatment (INCAT) scale. This increased to 20 of 41 (48.8%) at 12 weeks and to 32 of 42 (76.2%) at 24 weeks. Use of minimal important difference (MID)-determined amelioration of the inflammatory Rasch-built Overall Disability Scale (I-RODS), or of the Medical Research Council sum score (MRCSS), or of dominant hand-grip strength, in addition to the adjusted INCAT, indicated a sensitivity of 41.7% in identifying adjusted INCAT nonresponders at week 12 who subsequently responded at week 24. Specificity was 60% vs INCAT nonresponders at week 24. Consideration of amelioration of any amplitude on any secondary outcome measure indicated a 75% sensitivity, but only 30% specificity vs adjusted INCAT nonresponders at week 24. Discussion: Immunoglobulin treatment continuation may be justified for up to 24 weeks in CIDP. Additional outcome measures may help in the early treatment stages to predict delayed response on the adjusted INCAT. However, their use is limited by high false-positive rates. More robust, reliable, and relevant outcome measures are needed to detect early improvement in immunoglobulin-treated CIDP.