PBI19 Replicating Clinical Trial Exclusion Criteria in a Claims-Based Cohort of Patients with Chronic Inflammatory Demyelinating Polyradiculoneuropathy (CIDP) Initiating Intravenous Immunoglobulin (IVIG) Treatment

Abstract

CIDP is a debilitating, progressive, or relapsing disorder, affecting the peripheral nervous system, for which immunomodulatory IVIG is considered a first-line treatment. We replicated the exclusion criteria from a randomized controlled trial (RCT) in a real-world, claims-based cohort in this population to evaluate how a real-world population differs from RCT populations. Patients initiating IVIG after diagnosis with CIDP were identified from commercial, Medicare, and Medicaid claims, from 2008 through 2018, at their first use of IVIG following 6 months without use of immunoglobulins. Comorbidity- and treatment-based exclusion criteria from an ongoing RCT (NCT02549170) in CIDP patients were defined in claims data and applied to the IVIG cohort (using 6–12 months of available patient data prior to enrollment) to observe the impact on the study population. Not all RCT exclusion criteria could be defined in the claims data; those that could be defined included other neuropathies, multiple sclerosis, Parkinson’s disease, diabetic neuropathy, high-dose systemic corticosteroid use, plasma exchange/plasmapheresis, and immunosuppressant use. Of the 3975 IVIG initiators with CIDP identified, 75.4% would have met the RCT exclusion criterion of diagnosis of a different neuropathy, 26.5% had corticosteroid use in the 90 days prior to eligibility, and 15.9% had diabetic neuropathy or uncontrolled diabetes. Other RCT exclusion criteria were each present in <7.0% of the study population. After cumulatively applying all RCT exclusion criteria, 86.1% of the original study population was excluded. Applying exclusion criteria used in an RCT would have excluded most of the identified IVIG users with CIDP, primarily resulting from one criterion: other neuropathy. Although some RCT exclusions may be necessary to reduce misclassification of CIDP status, the resulting RCT-replicated population does not accurately reflect real-world users of IVIG.