Association between use of non-steroidal anti-inflammatory drugs (NSAIDs) and statins and the risk of cutaneous melanoma (CM): A case-control study
8500 Background: CM accounts for more than 77% of skin cancer deaths. Chemoprevention strategies have been hampered by the lack of supporting epidemiological data. Methods: A case control study examined the association between statins and NSAIDs use and CM in people over 40 years of age. A total of 400 histologically confirmed CM cases were recruited within 3 months from diagnosis between 3/04 and 12/06. Controls included 600 individuals without history of CM matched for age, gender and neighborhood in a ratio of 1:1.5. All completed a standardized telephone questionnaire that captured demographic characteristics, CM risk factors, and drug exposure history (length and frequency of use). Odds ratios were calculated with testing for statistical significance based on Chi square analysis. Results: To date, all 400 CM cases and 547 controls have been recruited. Interview results out of 387 cases (192 F/195 M) and 505 controls (254 F/251 M) were included for data analysis. Mean age was 58.4 and 58.5 years. The most significant CM risk factors included: red hair phenotype OR: 2.29, history of non-CM skin cancer OR: 2.28, family history of CM OR: 1.76, light complexion OR: 2.7, and history of more than 4 sunburns in childhood OR: 4.06. 100 cases vs. 190 of the controls had been exposed to statins OR 0.58 (CI:0.43–0.78); no greater protective effect was detected with increased duration of statin use (p=0.87 for heterogeneity of ORs for =5 vs. <5 yrs). 238 cases vs. 397 controls reported the use of NSAIDs, OR 0.44 (CI 0.32–0.59). 43 cases reported the use of aspirin-ASA for = 5 years vs. 92 controls OR: 0.37 (CI: 0.20–0.68). Similar results were observed with exposure to other NSAIDs with 38 cases reporting the use for = 5 years vs.109 controls, OR: 0.35 (CI: 0.24- 0.50). Conclusions: This study suggests that extended use of NSAIDs decreases the risk of CM development. Although any statin use demonstrated a decreased risk, the lack of an association between duration of use and protective effect raises questions about a true causal relation. Further analysis with respect to the CM subtype and specific exposure (by dose, and frequency of use) to these groups of drugs may lead to the design of CM chemopreventive clinical trials. Final analysis will be presented at ASCO. No significant financial relationships to disclose.