Epidemiologic data regarding the chemopreventive potential of nonsteroidal anti-inflammatory drugs (NSAIDs) against oral cancer are sparse. We found a relative risk for oral cancer of 1.2 (95% CI, 1.0 -1.6) among 169 589 Danish NSAID users (X2 prescriptions), with no apparent trends in subgroups. Our study provided no clear evidence that NSAIDs may protect against oral cancer. There is substantial experimental and epidemiological evidence that aspirin and other nonsteroidal anti-inflammatory drugs (NSAIDs) protect against colorectal adenomas and cancer (Baron, 2003;Ulrich et al, 2006). Epidemiologic studies have also consistently pointed to inverse associations between NSAID use and cancers of the stomach and oesophagus, whereas the results are mixed for other cancer sites (Gonzalez-Perez et al, 2003). Studies of animal models and human cancer cell lines have indicated that the potential chemopreventive effect of NSAIDs might extend to oral cancer (Goodin and Shiff, 2004;Wang, 2005), and this year a large phase III prevention trial of COX-2 inhibitors in patients with premalignant oral lesions (leukoplakia) is scheduled to be launched (Nelson, 2006). Interestingly, only a few epidemiological studies have provided data on the relationship between NSAID use and oral cancer, and the data are conflicting (Thun et al, 1993;Bosetti et al, 2003;Sørensen et al, 2003). This paucity of epidemiologic data prompted us to examine the incidence of oral cancer in a large cohort of NSAID users in Denmark.
MATERIALS AND METHODSThe study was carried out within the population of North Jutland County, Denmark (population approximately 490 000 inhabitants), during the study period 1 January 1991 to 31 December 2002. From the files of the Danish Civil Registration System, established in 1968(Frank, 2000, we identified all individuals in the county who were 16 years or older during the study period and resident in the county on 1 January 1991. We then excluded all individuals with a history of cancer (except nonmelanoma skin cancer) before study entry (1991 or age 16 years) by linkage to the Danish Cancer Registry, which has had accurate and almost complete nationwide ascertainment of cancer cases since 1943 (Storm et al, 1997). The final study population comprised 442 654 individuals.The Danish National Health Service provides tax-supported health care for all inhabitants and refunds part of patient expenditures on a wide range of prescribed drugs, including nonaspirin NSAIDs. In Denmark, non-aspirin NSAID can be obtained only by prescription, except for low doses of ibuprofen which account for approximately 14% of the total use (Sørensen et al, 2003). All health-related services are registered to individual patients by use of a civil registry number assigned to all Danish citizens, which encodes gender and date of birth. In North Jutland County, pharmacy data are transferred to a research prescription database, which was initiated in 1989 and covered all pharmacies by 1991 (Gaist et al, 1997). The database holds key information on ...