Abstract Background Osteoarthritis (OA) is a common chronic disease in older adults. Currently, there are no effective therapies to reduce disease severity and progression of knee OA (KOA), particularly in mid- to late-stages. This study aims to examine the effect of methotrexate (MTX) on knee effusion-synovitis and pain in symptomatic patients with mid- to late-stage KOA. Methods/design This protocol describes a multicentre randomised placebo-controlled clinical trial aiming to recruit 200 participants with mid- to late-stage symptomatic KOA and with effusion-synovitis grade of ≥ 2. Participants will be randomly allocated to the MTX group (start from 5 mg per week for the first 2 weeks and increase to 10 mg per week for the second 2 weeks and 15 mg per week for the remaining period if tolerated) or the placebo group. Primary outcomes are effusion-synovitis size measured by magnetic resonance imaging (MRI) and knee pain assessed by visual analogue scale (VAS). Secondary outcomes are signal intensity alteration within infrapatellar fat pad (IPFP) and Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) total score and subscores, and the Outcome Measures in Rheumatology Arthritis Clinical Trials-Osteoarthritis Research Society International (OMERACT-OARSI) responders. Both intention-to-treat and per-protocol analyses will be performed. Discussion If MTX intervention can relieve symptoms and reduce inflammation in patients with mid- to late-stage KOA, it has the potential for significant clinical and public health impact as this low-cost and commonly used intervention would delay the time to knee replacement, leading to substantial cost savings and improve quality of life. Trial registration ClinicalTrials.gov NCT03815448 . Registered on 21 January 2019.