Benefits Found for Hand OA Drug Treatments

MILAN -- Disease-modifying drugs may have a role in treating osteoarthritis (OA) of the hand after all, results from two separate studies reported here indicated.

Patients with radiologically confirmed hand OA obtained significantly greater pain relief when treated with methotrexate versus placebo in a randomized 6-month trial, according to Flavia Cicuttini, MBBS, PhD, MSc, of Monash University in Melbourne, Australia.

And in the other study -- also a randomized trial -- progression of erosive OA was substantially reduced with denosumab (Prolia) for 48 weeks relative to placebo, reported Ruth Wittoek, MD, PhD, of Ghent University in Belgium.

Both presented the results at the European Alliance of Associations for Rheumatology (EULAR) annual meeting.

The findings are remarkable because so-called disease-modifying anti-rheumatic drugs (DMARDs) have almost always failed to help patients with OA, whether in the hand or other joints. Pharmacotherapy currently relies on agents such as oral painkillers and corticosteroid injections that have no effect on the underlying disease process.

However, said Cicuttini, no study had ever examined methotrexate for the specific population of hand OA patients "of the inflammatory subtype," which her group hoped to capture for their trial. They enrolled 97 patients ages 40-75, with self-reported pain scores of at least 40 on a 100-point scale, with synovitis confirmed with MRI scans and other radiological evidence of OA in at least one hand joint. Participants were randomized 1:1 to methotrexate or placebo, with the drug started at 10 mg/week for the first 4 weeks and then 20 mg/week thereafter.

Actual pain scores averaged about 63 at baseline and half the patients had Kellgren-Lawrence radiological grades of 3/4. Mean age was about 62, with disease duration averaging 7 years, and about two-thirds of patients were women.

As is typical in OA trials, the placebo group also showed some degree of pain relief -- mean scores fell by 7.7 points -- but the decline was significantly greater with methotrexate at 15.2 points by month 6 (P<0.05).

Evaluation of overall disease burden with the Australian/Canadian OA Hand Index (AUSCAN), which also covers stiffness and physical function as well as pain, showed an even greater advantage for methotrexate. With mean scores at baseline of about 240, only a slight improvement was seen in the placebo group, whereas methotrexate led to a decline of some 60 points.

Overall, said Cicuttini, the trial demonstrated "a moderate effect" in favor of methotrexate.

Adverse event rates were similar between groups, with moderate elevations in liver enzymes being the only type to be more common with methotrexate (17 vs 10 patients).

The results were less persuasive than they might have been if the COVID-19 pandemic had not intervened, Cicuttini noted. Conceived in 2017, the trial had originally been planned to span 2 years and include radiological outcomes. But when March 2020 came along, the investigators believed it wise to stop methotrexate dosing, due to fears it might worsen COVID; when the study recommenced in November 2020, the treatment period was limited to 6 months and outcomes were only tracked remotely with no imaging performed.

Wittoek, who served as co-moderator at the session where the two researchers spoke, asked whether there was a plan to redo the study, this time with full radiological data and covering the full 2 years. Cicuttini responded, "That's the logical next step," but didn't say it was definitely in the works.

Denosumab for Erosive Disease

Wittoek discussed a similarly sized trial of denosumab, the bone-building biologic drug currently used to treat osteoporosis (and, in a different formulation, bone effects in cancer). Previous studies had suggested that the drug's target, receptor activator of nuclear factor-κb ligand, or RANKL, helps mediate development of hand OA erosions, so her group turned to denosumab to test their hypothesis that inhibiting RANKL might prevent further joint degradation.

They recruited 100 patients showing radiological signs of erosion (i.e., joint space narrowing and/or subchondral erosion) along with inflammatory activity in at least one joint. Their primary endpoint was change in Ghent University Scoring System (GUSS) score through week 48, with secondary endpoints including number of new erosive joints and assessments of pain and function.

Patients' mean age was 61, close to 80% were women, and disease duration averaged 6 years. Mean baseline 16-joint GUSS score was 248, with pain rated at 4.7 on a 10-point scale. The overall AUSCAN score averaged 68 and another index, the Functional Impairment in Hand OA (FIHOA), yielded a mean score of 10.4. Patients were randomized 1:1 to denosumab at 60 mg every 4 weeks (twice the dose normally used in osteoporosis, Wittoek noted) or placebo.

Effects of denosumab were profound. At week 48, mean GUSS scores with the active drug actually rose 10 points, reflecting less erosion that at baseline, whereas a decline of 7.9 was observed with placebo (P=0.003). An average of 1.8% of joints had new erosions with denosumab, versus 7.0% with placebo (P<0.001). However, there was little difference in pain or AUSCAN or FIHOA scores at week 48.

The trial also included a further 48-week extension in which the placebo group was switched to open-label denosumab, while those originally assigned to the drug stayed on it. Improvements in radiological parameters grew even greater in those on denosumab for the full 96 weeks, and those who switched to it from placebo quickly obtained the same improvements. In addition, Wittoek reported, some advantages for denosumab in pain and functional outcomes emerged during the extension.

As for next steps, Wittoek agreed with an audience member who suggested that the challenge now is to identify patients at risk for erosions who can be started on therapy before they actually develop. That is not currently possible, she acknowledged, and establishing cost-effectiveness in a non-lethal disease is also a potential barrier.

Another audience member asked what was happening to patients who finished the trial and were then taken off denosumab. It could be problematic for them, as a "rebound effect" is common in osteoporosis patients who stop denosumab. Wittoek explained that they were offered to go on the bisphosphonate agent alendronate (Fosamax) as part of a follow-up study; evaluations are now underway, she said, which appear encouraging at this point.