Objectives To compare the efficacy and safety of chondroitin sulfate plus glucosamine hydrochloride (CS+GH) versus celecoxib in patients with knee osteoarthritis and severe pain. Methods Double-blind Multicentre Osteoarthritis interVEntion trial with SYSADOA (MOVES) conducted in France, Germany, Poland and Spain evaluating treatment with CS+GH versus celecoxib in 606 patients with Kellgren and Lawrence grades 2–3 knee osteoarthritis and moderate-to-severe pain (Western Ontario and McMaster osteoarthritis index (WOMAC) score ≥301; 0–500 scale). Patients were randomised to receive 400 mg CS plus 500 mg GH three times a day or 200 mg celecoxib every day for 6 months. The primary outcome was the mean decrease in WOMAC pain from baseline to 6 months. Secondary outcomes included WOMAC function and stiffness, visual analogue scale for pain, presence of joint swelling/effusion, rescue medication consumption, Outcome Measures in Rheumatology Clinical Trials and Osteoarthritis Research Society International (OMERACT-OARSI) criteria and EuroQoL-5D. Results The adjusted mean change (95% CI) in WOMAC pain was −185.7 (−200.3 to −171.1) (50.1% decrease) with CS+GH and −186.8 (−201.7 to −171.9) (50.2% decrease) with celecoxib, meeting the non-inferiority margin of −40: −1.11 (−22.0 to 19.8; p=0.92). All sensitivity analyses were consistent with that result. At 6 months, 79.7% of patients in the combination group and 79.2% in the celecoxib group fulfilled OMERACT-OARSI criteria. Both groups elicited a reduction >50% in the presence of joint swelling; a similar reduction was seen for effusion. No differences were observed for the other secondary outcomes. Adverse events were low and similarly distributed between groups. Conclusions CS+GH has comparable efficacy to celecoxib in reducing pain, stiffness, functional limitation and joint swelling/effusion after 6 months in patients with painful knee osteoarthritis, with a good safety profile.