Methotrexate Not Seen to Have Effect in Knee Osteoarthritis Pain or Effusion-Synovitis

A new study has found that 1 year of low-dose methotrexate did not reduce pain or effusion-synovitis in patients with knee osteoarthritis (OA) and effusion-synovitis.1

“[Previous] findings2 suggest an effect of methotrexate for inflammatory hand OA and a potential effect for knee OA. However, they also highlight the importance of phenotyping participants with knee OA to determine which specific populations, if any, will respond favorably to methotrexate. To date, no randomized clinical trial has investigated the efficacy of methotrexate in patients with inflammatory knee OA to our knowledge,” lead investigator Zhaohua Zhu, PhD, Clinical Research Centre, Zhujiang Hospital, Southern Medical University, Guangzhou, Guangdong, China, and colleagues wrote.1

Zhu and colleagues conducted a multicenter, placebo-controlled randomized clinical trial at 11 sites in China between July 2019, and January 2023. The trial enrolled participants with inflammatory knee OA with effusion-synovitis on magnetic resonance imaging, randomly assigned (1:1) to receive up to 15 mg methotrexate weekly or placebo using block randomization, stratified by study site. The trial primarily assessed knee visual analog scale (VAS) pain change and effusion-synovitis maximal area change over 52 weeks in the intention-to-treat population.

The trial screened 278 participants and randomized 215 participants to methotrexate (n = 108) or placebo (n = 107). Participants had a mean age of 60.4 years (standard deviation [SD], 7.4) and 89% (n = 191) were female. Overall, 175 participants (81%) completed the trial.

The investigators found that there was no significant different between methotrexate and placebo in VAS pain (between-group difference, 0.3 mm [95% CI, −6.7 to 7.3 mm]) or effusion-synovitis maximal area (0.1 cm2 [95% CI, −0.8 to 1.0 cm2]). Additionally, there were no significant between-group differences for any of the prespecified secondary outcomes, including mean change from baseline on the infrapatellar fat pad (IPFP) signal intensity, WOMAC total score, and response rate to therapy, as per OMERACT-OARSI criteria.

Zhu and colleagues did observe significant improvements in VAS pain (between group difference, -40.69 [95% CI, -70.48 to -10.90]), WOMAC pain (between group difference, -176.35 95% CI, -299.08 to -53.61]), function (between group difference, -715.28 95% CI, -1120.02 to 310.55]), and total score (between group difference, -917.26 95% CI, -1457.42 to 377.10]), and general health status (between group difference,-40.08 [95% CI, -64.79 to -15.37]) in participants with severe pain (VAS score ≥ 80 mm) at baseline.

Safety profiles were also comparable between groups, with at least 1 adverse event occurring in 32 participants (29.6%) in the methotrexate group and 26 participants (24.3%) in the placebo group.

Zhu and colleagues acknowledged several limitations of the study, including COVID-19-related recruitment delays, and a small, predominantly female, racially homogeneous population, but stated that follow-up loss and adherence rates were acceptable and balanced between groups.

“In this randomized clinical trial, we found that, compared to placebo, low-dose methotrexate did not relieve pain or reduce the size of effusion-synovitis over 52 weeks in patients with inflammatory knee OA. Further studies are required to confirm if methotrexate has an effect on inflammatory knee OA with severe knee pain,” Zhu and colleagues concluded.1

REFERENCES

1. Zhu Z, Yu Q, Leng X, et al. Low-Dose Methotrexate for the Treatment of Inflammatory Knee Osteoarthritis: A Randomized Clinical Trial. JAMA Intern Med. Published online June 02, 2025. doi:10.1001/jamainternmed.2025.1359

2. WangY, JonesG, KeenHI,et al.Methotrexate to treat hand osteoarthritis with synovitis (METHODS): an Australian, multisite, parallel-group, double-blind, randomised, placebo-controlled trial.Lancet. 2023;402(10414):1764-1772. doi:10.1016/S0140-6736(23)01572-6