Non-steroidal anti-inflammatory drugs and the risk of melanoma

1027 Background: Non-steroidal anti-inflammatory medications (NSAIDs) have been shown to inhibit tumor growth in melanoma cell lines and mouse models of melanoma. We investigated the association between NSAIDs and melanoma in a case-control study comparing individuals with invasive melanoma, those with in situ melanoma, and melanoma-free controls. Methods: We recruited 327 subjects with incident melanoma diagnosed in 2000 (222 with invasive melanoma and 105 with in situ melanoma), as well as 117 melanoma-free spouse controls. All subjects completed a structured interview that assessed personal and family history of cancer, sun exposure, skin phenotype, and medication use. Current use of NSAIDs was assessed by subject self-report with a minimum duration of three months of daily use. Results: Use of NSAIDs, aspirin, and cyclo-oxygenase-2 (COX-2) inhibitors was reported by 7.66%, 9.23%, and 5.63% of subjects, respectively. The unadjusted odds ratio (OR) for use of any NSAID was 0.73 (95% CI 0.34–1.55) in a comparison of subjects with any melanoma to those without melanoma. After adjustment for age, sex, skin type, hair and eye color, and family history, the OR was 0.59 (95% CI 0.24 - 1.47). Similar results were found when restricting to cases with invasive melanoma (adjusted OR 0.64, 95% CI 0.27 - 1.73). Use of COX-2 inhibitors was also inversely related to any melanoma diagnosis (adjusted OR 0.44; 95% CI 0.14 - 1.35) and to invasive melanoma diagnosis (adjusted OR 0.58; 95% CI 0.18 - 1.9). Aspirin users had an unadjusted OR of 0.93 (95% CI 0.47–2.01), and an adjusted OR of 1.59 (95% CI 0.48–5.22). Conclusions: These results suggest that NSAIDs may be associated with a reduction in melanoma risk. This association, although not statistically significant, is consistent across classes of melanoma diagnosis and subtypes of NSAIDs. Because of potential overmatching of spouse controls, the odds ratios reported here are likely to be attenuated. Larger studies are required to evaluate potentially meaningful protective effects of NSAIDs. In addition, these findings need to be balanced with recent data showing that rofecoxib (a COX-2 inhibitor) increases the risk of cardiovascular events. No significant financial relationships to disclose.