Development and expert refinement of a stratified framework for progression independent of relapse activity (PIRA) in multiple sclerosis

INTRODUCTION: Progression Independent of Relapse Activity (PIRA) is a critical measure of disability progression in multiple sclerosis (MS) independent of relapses but lacks a standardized definition. Current reliance on the Expanded Disability Status Scale (EDSS) limits sensitivity to non-motor domains, necessitating a stratified framework to enhance detection and guide management. OBJECTIVES: To develop a novel seven-level stratified PIRA definition and assess its validity, enhanced sensitivity, clinical relevance, and feasibility through expert consensus, and propose a simplified framework based on feedback. METHODS: A two-stage study: (1) A four-expert panel developed a seven-level PIRA framework (PIRA 1: EDSS-based; PIRA 2: EDSS-plus measures; PIRA 3: stress tests; PIRA 4: patient-reported outcomes [PROs]; PIRA 5: conventional MRI; PIRA 6: advanced MRI; PIRA 7: biomarkers) via literature synthesis. (2) A survey of 90 MS experts (26 responded, 28.9%) from nine countries evaluated each level's validity, sensitivity (vs. EDSS), relevance, and feasibility (1-5 scale), with open-ended comments on barriers and suggestions. High agreement was defined as a score ≥ 4. Qualitative feedback on barriers and improvement suggestions was thematically analyzed. RESULTS: Strong support (24/26; 92.3%) endorsed a stratified PIRA definition. Respondents included primarily clinician-researchers (22/26; 84.6%), with 11/26 (42.3%) reporting more than 20 years of MS experience. High agreement for validity ranged from 15/26 (57.7%) for PIRA 1-23/26 (88.5%) for PIRA 6. Agreement regarding enhanced sensitivity was highest for PIRA 2 (25/26; 96.2%), followed by PIRA 6 (22/26; 84.6%) and PIRA 5 (19/26; 73.1%). Clinical relevance was rated highly for PIRA 1, PIRA 2, and PIRA 6 (each 25/26; 96.2%). Feasibility was highest for PIRA 1 (24/26; 92.3%) and declined for higher levels, particularly PIRA 7. Barriers included inter-rater variability (PIRA 1, 30.8%), subjectivity (PIRA 4, 23.1%), and cost/expertise (PIRA 6-7, 26.9% each). Suggestions included digital tools (PIRA 2-3), AI for MRI (PIRA 5-6), biomarker validation (PIRA 7), and combining PIRA 5-6. PIRA 1-2 were preferred for clinical practice, PIRA 5-7 for research. A three-tier framework was proposed: Probable PIRA (clinical), Definite PIRA (clinical + MRI), and Definite PIRA Plus (clinical + MRI + biomarkers). CONCLUSIONS: The proposed seven-level PIRA framework demonstrates strong expert support for its conceptual validity and clinical relevance but highlights feasibility challenges for advanced assessments. A simplified three-tier model may provide a practical structure for standardized evaluation of relapse-independent progression in MS. Further empirical validation in diverse clinical cohorts is required.