Although nonsteroidal anti-inflammatory drugs (NSAIDs) are widely prescribed globally, their chronic use increases the risk of upper gastrointestinal (GI) damage. Cyclooxygenase-2–selective NSAIDs are considered to reduce this risk. Current guidelines in Japan recommend loxoprofen sodium (loxoprofen), a pro-drug in the propionic acid class of nonselective NSAIDs, as first-line therapy in rheumatoid arthritis.


To confirm the superiority of celecoxib, a cyclooxygenase-2-selective NSAID, to loxoprofen in the incidence of gastroduodenal (GD) endoscopic ulcers.


A randomised, multicentre, placebo-controlled, double-blind, phase IV clinical trial in healthy Japanese volunteers [mean age 57.5 (range: 40–74) years; >70% female], stratified by Helicobacter pylori status at screening (∼40% positive) and randomised 2:2:1 to receive celecoxib 100 mg b.d., loxoprofen 60 mg t.d.s. or placebo. Primary end point was incidence of any GD endoscopic ulcers after 2 weeks of treatment.


Of 190 randomised subjects, 189 received at least one dose of celecoxib (n = 76), loxoprofen (n = 76), or placebo (n = 37). Incidence of GD ulcers was 1.4%, 27.6% and 2.7% in the celecoxib, loxoprofen and placebo groups respectively (P < 0.0001 in favour of the celecoxib group); incidence of adverse events (AEs) was 34.2%, 51.3% and 21.6% in the celecoxib, loxoprofen and placebo groups respectively. No serious or severe AEs were reported.


Celecoxib 100 mg b.d. was superior to loxoprofen 60 mg t.d.s. regarding the incidence of gastro-duodenal endoscopic ulcers over 2 weeks. Celecoxib was well tolerated and no major safety concerns were observed. NCT00994461.