Background Hand osteoarthritis (HOA) is a frequent painful polyarticular disease which may not respond to any classical therapeutics, leaving the practitioners without any solution.

Objectives To evaluate anti TNF blocker in patients with painful HOA refractory to analgesics and NSAIDs

Methods The digital osteoarthritis in refractory hand OA study (DORA), is a phase 3 randomized superiority, double-blind (patients and outcome assessors), parallel, placebo controlled, 26 weeks, multicenter trial (conducted in 16 French clinical site) using TNF blocker adalimumab (2 sub cutaneous injections at week 0 and week 2) Patients meeting the American College of Rheumatology for hand OA with pain over 40mm on a 100mm VAS, (involving at least 3 interphalangeal joints) with at least 3 OA joints at Kellgren Lawrence (KL) grade > 2 on a recent X Rays, and who do not respond to analgesics and NSAIDs, were recruited. The primary endpoint was the proportion of patients with at least 50% improvementin their baseline painscore at week 6. Secondary outcomes were number of spontaneous painful joints, number of painful joints on pressure, number of swollen joints, morning stiffness, patient and practitioner global assessments, functional index of Dreiser and Cochin hand functional index, consumption of analgesics were recorded ( acetaminophen up to 3g/D was the only rescue medication allowed until week 6). Serum markers (COMP, PIINP, AH, usCRP, cytokines level of TNF, IL-6, IL-1) and urine level of CTX-II were measured at W0 and W6.

Results On the 99 patients selected, 85 were randomized (42 in the placebo group, 41 in the adalimumab group and 2 unknown treatment received). 37 patients in the placebo group and 41 in adalimumab group received at least one injection and were evaluated at week 6 (n=78) (mITT). 35 patients in the placebo group and 38 in the adalimumab group received the two injections. Mean (SD) age was 62.5 (6.9), 85 % of women, mean (SD) level of pain was 65.4 (12.9) mm. At week 6, primary outcome was achieved by 31.7% of adalimumab treated patients compared with 24.3% of placebo-treated patients (relative risk, 1.05; 95% confidence interval, 0.93-1.17). No statistically significant differences were foundfor any of the secondary outcomes. Post hoc analysis in sub group of patients with more than 3 swollen IP at baseline did not shown any difference between groups. Consumption of analgesics was not different between groups. There were no safety concerns. There was none variations of any biological markers between the 2 groups. TNF alpha serum level was not correlated with clinical outcome in the group of patients treated with adalimumab.

Conclusions In a group of patients with refractory hand OA, TNFalpha blockers (adalimumab, 2 sc injections) failed to demonstrate any clinical improvement

Acknowledgements Funding academic study supported by The French Section on Osteoarthritis Labortaory Abbott, Inserm Pro A

Disclosure of Interest None Declared