: Compared with currently available NSAIDs (which inhibit COX-1 and COX-2 isoforms of cyclooxy- genase), MK-0966 (a specific COX-2 inhibitor) is expected to cause less gastrointestinal toxicity.


: To compare the effect on the upper gastrointestinal mucosae of a high dose of MK-0966 with that of conventional doses of ibuprofen and aspirin.


Healthy subjects (n = 170; age range 18–54 years) with endoscopically normal gastric and duodenal mucosa were randomized to either MK-0966 250 mg q.d. (n = 51), ibuprofen 800 mg t.d.s. (n = 51), aspirin 650 mg q.d.s. (n = 17), or placebo (n = 51) in this 7-day, double-blind, parallel-group study. The mucosae were evaluated by endoscopy using a predefined scale; scores could range from 0 to 4. The primary end-point was the percentage of subjects who developed a mucosal score ≥ 2 (i.e. the development of one or more erosions). To evaluate COX-1 activity, serum thromboxane B2 levels were determined in a subset of the population.


The percentage of subjects who developed a mucosal score ≥ 2 in the MK-0966 group (12%) was significantly lower (< 0.001) than that in the ibuprofen (71%) and aspirin (94%) groups, and was similar to that in the placebo group (8%). Only ibuprofen and aspirin significantly (< 0.0001) reduced baseline thromboxane B2 levels. All treatments were generally well tolerated.


In this acute short-term endoscopic study, MK-0966 250 mg q.d. (a dose at least 10 times higher than that demonstrated to reduce the signs and symptoms of osteoarthritis) produced significantly less gastrointestinal mucosal damage than either ibuprofen 800 mg t.d.s. or aspirin 650 mg q.d.s. and was comparable to placebo in this regard.