“Therefore, we wanted to know whether blocking of synovial inflammation by a well-known drug such as etanercept would also have a positive effect on outcomes in erosive hand OA,” Dr. Kloppenburg explained.
The primary outcome measure was the level of OA pain assessed on a visual analog scale (VAS) at 24 weeks.
Secondary endpoints included assessment of hand function, quality of life, the number of tender joints, and grip strength after 4, 8, 12, 24, and 36 weeks, and after 1 year. Radiographic progression of IP joints was scored blindly at baseline, 24 weeks, and 1 year following the quantitative Ghent University Scoring System (GUSS). VAS pain was compared between treatment groups at 24 weeks and 1 year in intention-to-treat analyses.
Although etanercept was not superior to placebo on VAS pain at 24 weeks, it was superior to placebo both on pain and structural damage assessed by GUSS in the symptomatic and inflammatory patients who completed the study. The drug was especially effective in joints with signs of inflammation.
Overall, VAS pain in all patients decreased by 24.8 mm (95% CI, –29.2 to –20.5; P less than .001) at 24 weeks. In intention-to-treat analysis, differences in pain between the groups were in favor of etanercept but did not reach statistical significance. The per-protocol analysis of GUSS showed a mean difference in favor of etanercept, indicating more remodeling in the etanercept group.
Additional analyses showed an interaction between soft swelling/erythema and etanercept treatment on GUSS, resulting in a statistically significant (P less than .05) mean difference between the two treatment groups. More patients dropped out on placebo than on etanercept (six vs. three) because of inefficacy, whereas more dropped out on etanercept than on placebo (six vs. one) because of adverse effects.
“Synovial inflammation is an interesting target for treatment in OA patients with an inflammatory hand osteoarthritis phenotype,” Dr. Kloppenburg said.
A separate analysis of the same multicenter study suggests that etanercept is effective in inhibiting bone marrow lesions (BMLs) in patients with erosive hand OA.
The researchers studied 20 participants with symptomatic erosive OA with clinical and ultrasonographic signs of inflammation in at least one IP joint. The patients underwent contrast-enhanced MRI of the eight distal and proximal IP joints of one hand at baseline and 1 year. Images were scored for synovitis and BMLs (0-3 per joint, total score 0-24), blinded for patient characteristics.
Radiographs of the same hand were scored according to the Verbruggen-Veys system. Logistic regression was used to associate the presence of an MRI feature in a joint with being in an erosive versus nonerosive anatomical phase.
“Although erosive hand OA is a condition with a high disease burden, no disease-modifying treatments are available yet,” explained Féline Kroon, a PhD student in the department of rheumatology at Leiden University Medical Center in the Netherlands, in an interview. “Research to find a new form of therapy is partly hindered by our limited knowledge on the pathophysiology of the disease.”
However, “new imaging modalities like MRI enable us to study the pathophysiology of erosive OA more closely,” she said. “This study also gave us the unique opportunity to investigate whether anti-TNF, which is known to lead to clinical improvement and improvement of inflammatory lesions on MRI in other rheumatic diseases like rheumatoid arthritis, might also be effective in erosive OA.”
The presence of BMLs, but not synovitis, was associated with the presence of an erosive anatomical phase in a joint, and treatment with etanercept appeared to be effective in inhibiting these lesions. That suggests a role for TNF in the pathophysiology of erosive OA.
The inhibitory effect of etanercept on BMLs was more pronounced in IP joints with severe synovitis at baseline, suggesting that inflamed synovial tissue could be a source of TNF production in erosive OA.
The total synovitis score was similar at baseline and 1 year in both study groups. For BMLs, the median total score at baseline was 5.4 (range 2-9) and 7.0 (0-9) at 1 year in the placebo group, versus 4.5 (3-9) and 3.7 (0-8), respectively, in the etanercept group.
The presence of BMLs was associated with being in the erosive and remodeling anatomical phases of erosive hand OA. Synovitis was not associated with those phases.
“We think that TNF-alpha plays a role in the pathophysiology of erosive OA via an effect on the subchondral bone,” Ms. Kroon said. “Because we saw that the beneficial effect of etanercept on BMLs was more pronounced in joints with synovitis at baseline, we think that, in an inflamed synovial hand joint, an interaction takes place between synovium and subchondral bone, which could be influenced by blocking TNF.”