Cigarette smoking and risk of acute leukemia: associations with morphology and cytogenetic abnormalities in bone marrow.

BACKGROUND: Cigarette smoking may be a risk factor for leukemia. No detailed biological mechanism has been proposed, but a causal link is made plausible by evidence of systemic effects of cigarette smoke and the presence in cigarette smoke of chemicals that have been associated with leukemia risk. PURPOSE: Our purpose was to investigate the leukemia risk associated with cigarette smoking in a multicenter case-control study of acute leukemias in adults. METHODS: Adults aged 18-79 with newly diagnosed leukemia were contacted to participate in this epidemiologic study when they entered a clinical trial to be treated under protocols sponsored by Cancer and Leukemia Group B. Smoking histories for 610 patients with acute leukemia and 618 population control subjects were obtained by telephone interviews. We examined bone marrow samples and classified patients by morphology of leukocyte precursor cells according to the French-American-British (FAB) classification system and, for 378 patients, by the presence or absence of specific clonal chromosome abnormalities. We calculated odds ratios (ORs) for risk of leukemia associated with smoking cigarettes. ORs were adjusted for age, race, and sex. RESULTS: Smoking was associated with only a modest increase in risk for leukemia overall (adjusted OR = 1.13; 95% confidence interval [CI] = 0.89-1.44). However, among participants aged 60 and older, smoking was associated with a twofold increase in risk for acute myeloid leukemia (AML) (OR = 1.96; 95% CI = 1.17-3.28) and a threefold increase in risk for acute lymphocytic leukemia (ALL) (OR = 3.40; 95% CI = 0.97-11.9). Among older persons, risks increased with amount and duration of smoking. Smoking was associated with increased risk for AML classified as FAB type M2 at all ages, with ORs of 1.70 (95% CI = 1.00-2.90) for those younger than 60 and 3.50 (95% CI = 1.53-8.03) for those aged 60 and older. Smoking was also associated with ALL type L2 at all ages, with ORs of 1.72 (95% CI = 0.90-3.27) for those younger than 60 and 5.34 (95% CI = 1.03-27.6) for those who were older. Smoking was more common among patients with specific chromosome abnormalities in AML [-7 or 7q-, -Y, +13] and in ALL [t(9;22)(q34;q11)]. CONCLUSIONS: Cigarette smoking is associated with increased risk for leukemia and may lead to leukemias of specific morphologic and chromosomal types. The association varies with age. IMPLICATION: Examining discrete subtypes of disease may permit more accurate assessment of risk. As standardized morphologic classification and cytogenetic and molecular evaluation of leukemia patients becomes more common, epidemiologic studies that take advantage of these advances will begin to contribute to the identification of additional risk factors and mechanisms in acute leukemia.