A case-control study was conducted to investigate the role of a familial history of cancer in the etiology of childhood acute leukemia. The history of cancer in the relatives of 472 cases was compared with that of 567 population-based controls. Recruitment was frequency matched on age, sex and region. The familial history of cancer in each child's relatives was reported by the mother in response to a standardized self-administered questionnaire. A familial history of solid tumor in first or second-degree relatives was associated with an increased risk of acute lymphoblastic leukemia (odds ratio (OR)=1.6 [95% confidence interval, 1.2-2.1]), while a familial history of hematopoietic malignancies in first or second-degree relatives was associated with an increased risk of acute myeloid leukemia (OR=4.3 [1.4-13]). The ORs for the histories of cancer increased with the number of relatives with cancer (OR=1.5 [1.1-2.0] for one relative and OR=2.3 [1.3-3.8] for two relatives or more; Ptrend<0.0001). Significant associations between childhood acute leukemia and familial history of genital cancers and brain tumor were also observed (OR=2.7 [1.2-5.8] and OR=10.7 [1.3-86], respectively). This study supports the hypothesis that a familial history of cancer may play a role in the etiology of childhood acute leukemia. It also evidences some specific associations that require further investigation.