A population-based case-control study of 649 leukemia cases and 1771 controls carried out in 11 Italian areas, offered the opportunity to evaluate the relationship between alcohol consumption and leukemia risk. For all leukemias, acute lymphocytic leukemia (ALL), acute myeloid leukemia (AML), and chronic lymphocytic leukemia (CLL), we found a non-significantly inverse association for moderate levels of total alcohol and wine intake, but increased risks at high levels, with, in most cases, significant trend effects (odd ratios (OR) for all leukemias in the lowest quartile of total alcohol consumption [0.1-9.0 g/day of ethanol] versus never-drinker = 0.73; 95% confidence intervals (95% CI) = 0.51-1.03; OR in the highest quartile [> 31.7 g/day] = 1.15; 95% CI = 0.82-1.63; p of the linear trend test = 0.007). For chronic myeloid leukemia (CML), we found a non-significantly positive association for all levels of total alcohol and wine intake, and a significant positive linear trend effect (p = 0.03) for wine intake (OR for 0.1-9.0 g/day of ethanol intake from wine = 1.34; 95% CI = 0.61-2.94; OR in the highest quartile of wine intake [> 27.7 g/day] = 2.13; 95% CI = 1.01-4.50). No consistent dose-response was detected analysing duration of alcohol consumption for any leukemia subtypes. In conclusion, even though our study did not show a clear association between alcohol intake and leukemia risk, some of the patterns of the risk estimates (a possible J-shaped dose-response curve between alcohol intake and ALL, AML, and CLL risks, and the positive association between alcohol and CML), may be suggestive.