Abstract| Volume 26, SUPPLEMENT 1, S26, April 2018

Synthetic transdermal cannabidiol for the treatment of knee pain due to osteoarthritis

      Purpose: Cannabidiol (CBD) has shown promise in reducing pain and inflammation in pre-clinical models of arthritis. However, there have been no well controlled studies evaluating CBD for osteoarthritis (OA) in humans. This study evaluated safety and efficacy of ZYN002 (transdermal synthetic CBD gel) for the treatment of knee pain due to OA in adults.
      Methods: A Phase 2A, randomized, double-blind, placebo-controlled, multiple-dose study assessed ZYN002 administered twice daily for 12 weeks to adults with knee pain due to OA. Patients met ACR criteria for OA of the knee and underwent a 1-week washout to stop current anti-inflammatory agents/other analgesics (except paracetamol) followed by a 7 to 10-day baseline period capturing daily worst pain ratings using a 0 to 10 numeric rating scale. Eligible patients were randomized 1:1:1 to ZYN002 250 mg daily in 2 divided doses, ZYN002 500 mg daily in 2 divided doses, or placebo. The primary efficacy endpoint was change from baseline in the weekly mean of the 24-hour average worst pain score at Week 12. A key secondary endpoint was a responder analysis, defined as average weekly improvement in worst pain score of > 30% and decrease in WOMAC physical function sub scale of at least 20% at last observation.
      Results: Three hundred and twenty patients were randomized. Mean age was 62 (41–78) years, baseline mean worst knee pain score was 6.9. ZYN002 was not statistically different from placebo on the primary endpoint; week 12 mean reduction from baseline in average worst knee pain was −2.64 for ZYN002 250 mg/day (n = 106), −2.83 for ZYN002 500 mg/day (n = 105) and −2.37 for placebo (n = 103). However, patients using ZYN002 250 mg/day (n = 93) significantly outperformed placebo (n = 88) for the responder analysis (52.7% vs 34.1%, p = 0.016). Post-hoc analyses revealed that men treated with ZYN002 250 mg/day (n = 43) had significantly greater reductions from baseline in average worst knee pain scores than placebo-treated men (n = 31; 2.68 vs 1.70, p = 0.049) and greater performance in the responder analysis (n = 45; 60% vs 26.7%, p = 0.003), as compared to men who received placebo (n = 45). Indeed, women on ZYN002 did not differ from placebo in changes in average worst knee pain scores or responder analysis. Patients with the least amount of variability in baseline pain scores had greater performance in the responder analysis in both the 250 mg/day (n = 34; p = 0.055) and 500 mg/day (n = 29; p = 0.046) compared to placebo (n = 37). There were 2 treatment emergent adverse events that exceeded 3% of patients on ZYN002 and were greater than placebo: application site dryness (3.8% vs 0.9%) and headache (3.3% vs 1.9%).
      Conclusions: After 12 weeks of blinded treatment, while ZYN002 250 mg/day produced numerically better mean reductions from baseline in average worst knee pain scores, it was not statistically different than placebo. In contrast, the responder endpoint showed statistically significant differences between 250 mg/day and placebo. Significant gender differences were noted with the responder analysis and pain scores. Both doses of ZYN002 were well tolerated.