Effective Treatments for Hand Osteoarthritis: The Search Goes On

Hydroxychloroquine was not effective for inflammatory erosive osteoarthritis (OA) of the hands, a multicenter randomized study confirmed.

At week 52, pain scores on the Australian Canadian Hand Osteoarthritis Index (AUSCAN) were 26.7 (95% CI 23.9-29.4) among patients receiving hydroxychloroquine and 26.5 (95% CI 23.9-29.1) among those given placebo, according to Claudia Kedor, MD, of Charité Universitatsmedizin in Berlin, Germany, and colleagues.

That represented a nonsignificant adjusted mean between-group difference of 0.2 (95% CI -3.5 to 3.9, P=0.92), the researchers reported in the study online in RMD Open: Rheumatic & Musculoskeletal Diseases.

Hand OA is a common ailment characterized by pain and limitations in function and grip strength. Erosive hand OA is a severe subtype of the condition with subchondral erosions and cortical damage that may be accompanied by deformity of the joint itself.

There are no approved disease-modifying agents for hand OA; treatment is largely symptomatic and focuses on nonsteroidal anti-inflammatory drugs (NSAIDs) and analgesics. In many cases, current treatment is inadequate.

Hydroxychloroquine is a disease-modifying drug commonly used in many rheumatic diseases such as systemic lupus erythematosus and rheumatoid arthritis.

However, recent studies in the United Kingdom and the Netherlands found no benefit for hydroxychloroquine in symptomatic hand OA. The current study, known as OA-TREAT, "provides important confirmatory and new data, since it focused exclusively on hand erosive OA and included measurement of radiographic progression in this subgroup of patients that is especially at risk for structural damage," Kedor and colleagues wrote.

OA-TREAT was a phase IIIb double-blind trial that took place at 47 centers in Germany from 2013 to 2017. Participants had pain, swelling, and redness and/or warmth in more than three finger joints for at least 3 months despite the use of NSAIDs and analgesics. Patients also had to have at least one erosion present on radiographs.

The individual hand and finger joints were assessed for sclerosis, lateral deformities, narrowing of the joint space, and cortical collapse, with radiographs assessed by the Kallmann score.

Patients were randomized to placebo or hydroxychloroquine (200 mg/day, 200 and 400 mg on alternating days, or 200 mg twice/day, depending on body weight). Stable doses of NSAIDs and COX-2 inhibitors were permitted.

The study included 75 patients in the hydroxychloroquine group and 78 in the placebo group. Mean age was 51, disease duration averaged 10 years, and most participants were women.

The co-primary endpoints were AUSCAN pain and AUSCAN hand disability after 1 year of treatment.

For AUSCAN function, the adjusted mean score at week 52 was 48.1 (95% CI 43.0-53.2) in the hydroxychloroquine group and 51.3 (95% CI 46.6-56) in the placebo group, for a nonsignificant between-group adjusted difference of -3.2 (95% CI -10 to 3.6, P=0.36).

On secondary week 52 endpoints, the only differences were in erythrocyte sedimentation rate favoring hydroxychloroquine (8.2 vs 11.7, P<0.01) and morning stiffness favoring placebo (16.3 vs 30.2 minutes, P=0.001). There were no significant differences in tender and swollen joint counts, C-reactive protein, Health Assessment Questionnaire scores, physician and patient global assessments, or Short-Form 36 physical and mental scores, the investigators reported.

For radiographic changes, total Kallmann scores at week 52 were 47.1 (95% CI 46.0-48.2) in the hydroxychloroquine group and 46.8 (95% CI 45.7-47.8) in the placebo group, for a nonsignificant difference of 0.3 (95% CI -1.2 to 1.9, P=0.71).

There were seven serious adverse events in the hydroxychloroquine group and 15 in the placebo group. Five patients in the hydroxychloroquine group were hospitalized for conditions such as peripheral arterial occlusive disease, nausea, and pneumonia. In the hydroxychloroquine group there also was a life-threatening generalized rash and a death from hyperglycemic coma (not thought to be drug-related).

Eleven patients in the placebo group were hospitalized for conditions such as acute renal failure, atrial fibrillation, and foot operations.

A weakness of the study was its rather short duration, Kedor and co-authors noted. "Since OA is a slowly progressing disease, the observed time of 52 weeks for radiological progression may have been too short." In addition, MRI might have been more sensitive in detecting inflammatory changes, "but this was beyond the means of our trial," which was publicly funded by the German Federal Ministry of Education and Research.

The lack of superiority of hydroxychloroquine over placebo suggested that the drug should not be recommended as a disease-modifying treatment for hand OA, the team concluded. "Thus, the search for the underlying mechanism that causes the disease and for disease-modifying treatments must continue."

Comment