Could Methotrexate Work in Hand Osteoarthritis? Save

Lancet reports the results of the METHODS study from Australia where hand osteoarthritis (HOA) patients benefitted from 6 months of methotrexate (MTX) - as long as there was MRI synovitis!

This multisite, double-blind, randomised, placebo-controlled trial, enreolled adults (aged 40–75 years) with hand osteoarthritis (Kellgren and Lawrence grade ≥2 in at least one joint) and MRI-detected synovitis. Patients received either weekly placebo or MTX 20 mg for 6 months. The primary outcome was pain reduction at 6 months.

A total of 97 were randomized to MTX (n=50) or placebo (n=47), but only ~85% (82) had evaluable primary outcome data. The mean change in VAS pain at 6 months was

• MTX: –15·2 mm (SD 24·0)
• PBO: –7·7 mm (25·3) (between-group difference of –9·9 (95% CI –19·3 to –0·6; p=0·037)
• Adverse events occurred equally in both groups (62% MTX; 60% PBO)

This proof of concept trial suggests that MTX may have a a role in treating hand osteoarthritis with an Synovitis (by MRI). Nonetheless this study has numerous faults, limiting its potential utility without a repeat trial:

1. Significant dropouts with incomplete data at study end (only the good patients persisted?)
2. Intent to treat analysis (the most lenient and unreliable analysis method)
3. Having MRI synovitis questions the diagnosis of pedestrian hand OA
4. Many trials in OA, hand OA, and inflammatory Hand OA have failed to demonstrate efficacy of DMARDs, biologic or regenerative agents
5. Several other trials have tested the effect of weekly MTX in OA. Even with a positive reporting bias (negative trial results are unlikely to be reported), there are more suggesting MTX is ineffective in OA  (eg, Semin Arthritis Rheum. 2021 Aug;51(4):831-838) (Editors note: I performed a 20 patient knee OA trial with MTX nearly 20 years ago - the results were negative and we never reported it)

Thus, these results are preliminary and isolated; further research is needed.