Can patient self-evaluation of functional status be used for evaluation of impairment of motor function in Guillain-Barré syndrome? Mapping clinician-and patient-reported outcomes in a phase 3 study of eculizumab in Japan

Antoine Regnault ^1* Angély Loubert ¹ Stéphane Quéré ¹ Qun Lin ² Glen Frick ² Hirokazu Ishida ² Yuko Abeta ² Helene Chevrou-Severac ²

• ¹ Modus Outcomes, A THREAD Company, Lyon, France
• ² Alexion, AstraZeneca Rare Disease, Boston, MA, United States

Background and purpose: Guillain-Barré syndrome (GBS) is an autoimmune neurological disorder characterized by muscle weakness. In clinical trials, treatment benefit and disease severity are typically measured using clinician-reported outcome measures like the Hughes Functional Grading Scale (HFGS). However, patient-reported outcome measures, such as the Rasch-built Overall Disability Scale (R-ODS) may provide additional insight into the patient experience during treatment. In this study, exploratory analyses of clinical trial data were performed to investigate how existing ClinROs and PROs can help to assess disease progression by providing an accurate measurement of functional status. Methods: Data were collected as part of a phase 3 study to assess the safety and efficacy of eculizumab in patients in Japan with severe GBS. The association between HFGS score and R-ODS total centile score (linear measure of limitations; 0, most severe activity and social participation limitations and 100, no limitations) was assessed using the Spearman rank-order correlation coefficient. Threshold values of R-ODS total centile score that could differentiate between patients with an HFGS score of ≤ 1 and > 1 were determined using receiver-operating characteristic curve analyses and mapping (Rasch measurement theory). A triangulation approach was used to establish a proposed value for R-ODS total centile score equivalent to an HFGS score of ≤ 1 or > 1. Results: Overall, 57 patients were included in this analysis. These exploratory analyses revealed good correlation between R-ODS total centile and HFGS scores. Using the Rasch model, mapping of HFGS to R-ODS scores showed a good fit. Evaluation of the R-ODS threshold that could approximate the functional motor symptom categories based on HFGS (score of 0 or 1) revealed a range of values from 60 to 80. Based on a trial sample, a threshold of 60 was found to have 100% sensitivity and 87% specificity at week 4, and 93.8% sensitivity and 77.8% specificity at week 24. Conclusion: This study established thresholds for R-ODS total centile score that could approximate classification of functional impairment in GBS based on the HFGS. As the R-ODS reflects the patient perspective, it may be used to capture a more complete picture of GBS severity.