• Level of Pain on visual analogue scale [ Time Frame: 4 weeks following the last injection ] [ Designated as safety issue: Yes ]
  

• Number of tender joints [ Time Frame: at weeks 4, 6, 10, 14, 26 ] [ Designated as safety issue: Yes ]
  
• Number of swollen joints [ Time Frame: at weeks 4, 6, 10, 14, 26 ] [ Designated as safety issue: Yes ]
  
• Global assessment of handicap by patient [ Time Frame: at weeks 4, 6, 10, 14, 26 ] [ Designated as safety issue: Yes ]
  
• Global assessment of handicap by physician [ Time Frame: at weeks 4, 6, 10, 14, 26 ] [ Designated as safety issue: Yes ]
  
• Duration of morning stiffness [ Time Frame: at weeks 4, 6, 10, 14, 26 ] [ Designated as safety issue: Yes ]
  
• DREISER'S algofunctional index [ Time Frame: at weeks 4, 6, 10, 14, 26 ] [ Designated as safety issue: Yes ]
  
• Cochin hand index [ Time Frame: at weeks 4, 6, 10, 14, 26 ] [ Designated as safety issue: Yes ]
  
• Weekly pain recording, and recording of analgesic use (acetaminophen) [ Time Frame: at weeks 4, 6, 10, 14, 26 ] [ Designated as safety issue: Yes ]
  
• Percentage of non responders according to the number of patients requiring NSAIDS intake during the trial and/or rule out for lack of efficacy [ Time Frame: from week 0 until week 26 ] [ Designated as safety issue: Yes ]
  

Objective: To demonstrate the efficacy of two subcutaneous injections of anti-TNF alpha (adalimumab) in patients with DO refractory to usual treatments (NSAIDs - analgesics).

Design: Randomized placebo-controlled multicenter (19 sites) trial

Trial duration: 6 months follow-up per patient, total trial duration: 12 months

Number of patients: 84(up)

Dose of administration: Two subcutaneous injections separated by a 2-week interval of either adalimumab or placebo. Therapeutic precautions inherent to the prescription of anti-TNF alpha are based upon those recommended in rheumatoid arthritis. With only two injections of anti-TNF alpha, the risk of complications (infectious in particular) inherent to the anti-TNF alpha appears to be extremely slight.

Patient selection: Patients with digital osteoarthritis (DIP and PIP) according to American College of Rheumatology (ACR) criteria (with a recent X-ray, less than 6 months, of the hands, showing signs of digital osteoarthritis), symptomatic for more than 3 months (at least three finger joints) and scored more than 40 mm on a pain visual analog scale (VAS) despite use of a level 1 analgesic (acetaminophen: 4 g daily) andNSAIDs. Main Exclusion criteria are contraindications to anti-TNF alpha according to international guidelines.

Methods: Patients will be seen at a screening visit, then during treatment administration (W0 and W2) and at follow-up visits: W6, 10, 14 and 26. NSAIDs will be stopped at the time of screening.

Efficacy evaluation: The efficacy endpoint is improvement in pain on a pain VAS at different times of assessment (primary endpoint at W6). A more than 50% improvement, in comparison with baseline, will be considered to be significant. Secondary endpoints are: number of tender joints, number of swollen joints, morning stiffness, global assessment of handicap by patient, global assessment of handicap by physician, the DREISER'S algofunctional index, Cochin hand index, weekly evolution of pain ( patient record), and recording of analgesic use (acetaminophen or NSAIDs).

Statistical data: Statistical analysis will be performed on the inter-group difference in the intention-to-treat population. khi-2 will be used only for criteria number 1. Anticipated result: Significant superiority of adalimumab over placebo which would provide a therapeutic solution in difficult-to-treat patients with hand OA.