ObjectiveTo assess the long-term neurological safety of tanezumab, a monoclonal antibody against nerve growth factor. MethodsPatients with osteoarthritis of the hip or knee received stable doses of oral nonsteroidal anti-inflammatory drugs (NSAIDs) before study entry and during a ≤ 37-day screening period. Patients were randomized 1:1:1 to double-dummy tanezumab (2.5 mg or 5 mg, subcutaneous every 8 weeks) or oral NSAIDs (twice-daily) for 56 weeks, with a 24-week follow-up. Neurological safety evaluation focused on peripheral and sympathetic adverse events (AEs), neurologic examinations, and consultations with blinded, external diagnostic reviews. ResultsDuring the treatment period, 6.2%, 9.0%, and 4.6% of patients experienced AEs of abnormal peripheral sensation (APS) in the tanezumab 2.5 mg, 5 mg, and NSAID groups, respectively. Hypoesthesia, paresthesia, and carpal tunnel syndrome were the most common AEs of APS. Clinically significant worsening on examination occurred in <1% in any treatment group at the last study assessment. Diagnoses following external neurological consultation included mononeuropathy (1.3%, 2.1%, and 1.0%), radiculopathy (0.9%, 0.4%, and 0.5%), and polyneuropathy (0.3%, 0.5%, and 0%) in tanezumab 2.5 mg, 5 mg, and NSAID groups, respectively. AEs potentially associated with sympathetic neuropathy were reported for 1.8%, 2.3%, and 2.9% of patients in the tanezumab 2.5 mg, 5 mg, and NSAID groups, respectively. No patient was diagnosed with sympathetic neuropathy. ConclusionTanezumab had an increased incidence of AEs of APS versus NSAID; these were typically mild/moderate in severity, resolved during the study, and rarely resulted in discontinuation. Tanezumab was not associated with peripheral neuropathy and did not adversely affect the sympathetic nervous system. Trial registrationClinicalTrials.gov Identifier NCT02528188 (https://clinicaltrials.gov/ct2/show/NCT02528188).