Smoking and Risk of Acute Myeloid leukemia in Adults

Abstract

In recently published results from a Canadian case–control study on lifestyle factors and risk of adult leukemia, Kasim et al. reported a significantly increased risk of acute myeloid leukemia (AML) associated with cigarette smoking, based on 307 AML cases [1]. Specifically, they observed an odds ratio (OR) of 1.4 [95% confidence interval (CI) = 1.1, 1.8] for current smokers compared to neversmokers and positive trends relating smoking status (never, ex, current) and total pack-years of smoking with AML risk. Smoking was not associated with other forms of adult leukemia (acute lymphoid, chronic myeloid, chronic lymphoid, hairy cell). In a case–control study of AML conducted in Los Angeles, California, we observed increased risks associated with smoking that were particularly strong for French–American–British (FAB) subtype M2; OR for ‘ever smoked’ was 1.2 (CI = 0.9, 1.6) among all cases and 2.3 (CI = 1.1, 4.4) among M2 cases [2]. Among all cases, we observed weak evidence of dose-response. However, among M2 cases we observed significant doseresponse for total number of years smoked, average cigarettes/day, and type of product smoked (filtered versus non-filtered). In general, results were stronger for older cases (‡60 years). We hypothesized that the age effect may be due to decreased ability to repair DNA damage in older subjects, higher exposure levels among older subjects due to smoking habits that differed from those of younger subjects, and long latency for the smoking effect