Epidemiological characteristics of childhood acute lymphocytic leukemia. Analysis by immunophenotype. The Childrens Cancer Group.

While a number of epidemiological studies of childhood acute lymphocytic leukemia (ALL) have been conducted, separate analysis of risk factors for ALL subtypes has generally not been possible. We report the results of an analysis of data obtained from parents of children with ALL (and a control group of children without cancer), linked to a clinical database. Cases were classified into four ALL subtypes, and odds ratios (OR) were determined for each subtype for a broad range of factors. Numerous significant associations were found, some across all subtypes and others that were subtype-specific. Factors with elevated and/or significant ORs included: (i) for common ALL (n = 286): Down syndrome; family history (FH) of bone/joint diseases; postnatal jaundice; birthweight; MMR vaccination; exposure to gases and insecticides; and parental occupational exposure to insecticides. (ii) for pre-B ALL (n = 38): FH of gastrointestinal, hematological or bone/joint diseases, or allergy; cat ownership; exposure to solvents, fumes, petroleum products, cleaning agents and farm animals; and parental exposure to farm animals, fumes and solvents; (iii) for T-cell ALL (n = 158): FH of gastrointestinal disorders, maternal age, male gender, and parental occupational exposure to metals; (iv) for null-cell ALL (n = 65): FH of congenital heart disorders; measles; and parental occupational exposure to fumes, metals or solvents. This analysis should be considered as a hypothesis-generating process for future case-control interview studies.