Background/Purpose: Osteoarthritis (OA) is the most common chronic joint disease and OA pain is a leading cause of work disability. p38 inhibition suppresses the production of PGE2 and pro-inflammatory cytokines and may directly modulate nociception. A study was conducted to evaluate the efficacy and safety of ARRY-797, an orally administered, selective p38 inhibitor, in patients with OA knee pain refractory to NSAID treatment.

Methods: A randomized, double-blind, placebo (PBO) and active controlled 4-week Phase 2 study was conducted to evaluate the efficacy and safety of ARRY-797 vs PBO in patients with moderate to severe pain despite ongoing treatment with NSAIDs. Oxycodone extended release (Oxy ER) was used as an active control. 157 patients were randomized 1:1:1 to receive ARRY-797 (400 mg q12h), Oxy ER (10 mg q12h, Week 1; 20 mg q12h, Weeks 2 – 4) or PBO while maintaining use of NSAIDs. The primary endpoint was the change in WOMAC pain (0 – 10 NRS) for the index knee from Baseline to Week 4 compared to PBO using Baseline Observation Carried Forward (BOCF) for discontinuations due to adverse events (AEs) and Last Observation Carried Forward (LOCF) for other reasons.

Results: ARRY-797 provided statistically significantly greater improvement (‑0.8) for the primary endpoint of WOMAC pain at Week 4 (‑2.4 vs ‑1.6; 1‑sided p = 0.0247). ARRY‑797 treatment showed continual improvement throughout the 4-week treatment period (table). ARRY‑797 also demonstrated numerical improvement (ns) vs PBO on secondary endpoints including WOMAC physical function and stiffness, patients achieving ≥30% or ≥50% reduction in pain, Patient’s Global Impression of Change and Patient Treatment Satisfaction. Improvement in WOMAC pain for Oxy ER (‑0.71) was comparable to that seen with ARRY‑797 (‑0.68) when LOCF was used for imputation of missing data.

Biomarkers of cartilage (COMP) and bone (CTX‑I) degradation were also assessed. ARRY‑797 treatment resulted in statistically significant decreases in COMP and CTX‑I at Week 4 (10% and 38% vs PBO, respectively). The decrease in CTX‑I was rapid, sustained, and returned to Baseline by the Follow‑up visit.

The incidence of withdrawals due to AEs was similar for ARRY-797 (6%) and PBO (8%) and both were lower than Oxy ER (34%). The most common AEs in patients treated with ARRY-797 included mostly mild or moderate skin-related disorders, dizziness, diarrhea and stomatitis. Transient increases in CK and mild prolongations in the QTc interval were also noted. One serious AE occurred in the oxy ER group while none occurred with either ARRY‑797 or PBO treatment.

Conclusion: ARRY‑797 treatment resulted in durable, statistically significant improvement in OA pain and in reduction of circulating biomarkers of both cartilage and bone degradation in this 4‑week study. Further evaluation of the efficacy of this non-opioid analgesic and the potential for disease modifying activity are warranted.