A phase 2, randomized, double-blind, placebo-controlled study of senolytic molecule UBX0101 in the treatment of painful knee osteoarthritis

Purpose: To assess the efficacy and safety of single-dose, intra-articular (IA) administration of UBX0101 in patients with painful knee osteoarthritis (OA). UBX0101 is a p53/MDM2 interaction inhibitor that can induce apoptosis of senescent synoviocytes. Results from a recent Phase 1 study (NCT03513016) suggested that IA UBX0101 had dose-dependent, clinically meaningful effects on pain and function in patients with knee OA. Methods: This was a Phase 2, randomized, double-blind, placebo-controlled, parallel group study in OA patients randomized 1:1:1:1 to IA UBX0101 0.5 mg, 2.0 mg, 4.0 mg or matched placebo. Key eligibility criteria included knee OA by ACR criteria, Kellgren-Lawrence grade (KLG) 1-4, and mean daily pain between 4 and 9 on a Numeric Rating Scale (NRS, 0-10). Study duration was 24 weeks. Clinical outcomes included WOMAC pain (WOMAC-A) and function (WOMAC-C) sub-scores (each on a 0-4 Likert scale), daily pain NRS, patient global assessment (PGA) and patient global impression of change (PGIC). The primary endpoint was the change from baseline (CFBL) in WOMAC-A at Week 12. Key secondary efficacy endpoints included the CFBL to Week 12 of the WOMAC-C and the weekly mean of the daily pain NRS scores. With 45 patients per group, the study had 90% power and an alpha of 0.10 for a two-sided comparison of UBX0101 versus placebo with an effect size of 0.50 and an assumed standard deviation of 0.75 on the WOMAC-A item score least square (LS) mean change at Week 12. Results: A total of 183 patients were randomized. The study population was balanced regarding patient characteristics and baseline outcome measure values. Mean age was 62.9 years, 64% of the population was female, and 78% was white. Mean WOMAC-A sub-score at baseline ranged between 2.05 and 2.20. Decreases in WOMAC-A from baseline to Week 12 were similar for all treatment groups. The LS means CFBL in WOMAC-A at Week 12 were -0.924, -1.52, -1.019, and -1.017 for UBX0101 0.5 mg, 2.0 mg, 4.0 mg, and placebo, respectively. Secondary endpoints were not met. An historically high placebo response through Week 12 confounded the ability to discern a UBX0101 treatment effect. Possible reasons for this were the IA dosing route, patient and/or investigator expectedness, and gender dimorphism in pain reporting. Single IA doses of UBX0101 up to 4 mg were associated with an acceptable safety profile and were well-tolerated. Most adverse events (AEs) were mild. Seven, non-related, serious AEs occurred during the study; one was a death due to coronary artery disease. Conclusions: This Phase 2 study failed to demonstrate efficacy of single IA doses of UBX0101 in patients with painful knee OA. The encouraging symptomatic improvements observed in the previous Phase 1 study were not confirmed in this well-controlled, adequately powered study.