Acute leukemia is extremely rare in the first year of life: about 150 cases are diagnosed each year in the United States. Clinical and molecular studies of infant acute leukemia (IAL) suggest a distinct etiology and provide strong evidence of an in utero origin. Little is known about IAL etiology; however, advanced maternal age (adjusted for paternal age) has been associated with an increased risk of childhood cancers diagnosed between 0-14 years of age including leukemia (both acute lymphoblastic (ALL) and acute myeloid leukemia (AML)), lymphoma, central nervous system tumors, neuroblastoma, Wilms’ tumor, bone tumors and soft tissue sarcoma, while associations with paternal age after adjusting for maternal age are less striking.
Linked birth and cancer registry data for the period 1980-2004 were pooled from five states (NY, WA, MN, TX and CA) to evaluate associations between birth characteristics and IAL. The pooled data contained information for 351 infants diagnosed with acute leukemia (220 ALL, 131 AML) in the first year of life and 57,966 infant controls born in the same years. Children with Down syndrome documented at birth were excluded. Unconditional logistic regression was used to calculate odds ratios (ORs) and 95% confidence intervals (CIs).
Overall, males had a slightly decreased risk of developing IAL (OR=0.8, CI 0.7, 1.0), but there was little evidence of association with plurality, birth weight or maternal race. After adjusting for paternal age, infants with mothers ≥40 years of age had a significantly increased risk of developing IAL (OR=2.9, CI 1.4, 5.9), which was confined to the AML subtype (OR=5.5, 95% CI 2.1-14.6). In contrast, young paternal age <20 years (adjusted for maternal age) was associated with a significantly increased risk of IAL (OR=2.7, CI=1.4, 5.3), which was confined to the ALL subtype (OR=4.4, CI=1.9-10.0).
Ours is the first study to demonstrate a strong significant association of paternal young age with infant ALL. Since this study ascertained case and control status by record linkage, selection bias-- where cases or controls determine extent of participation-- is of little concern. Recall bias is also unlikely, as parental age was recorded prior to cancer diagnosis. These intriguing associations provide further evidence of differences in underlying etiology for IAL. Of note, approximately 80% of infants with ALL and 60% of infants with AML have a rearrangement involving the MLL gene in their leukemia cells, which has been shown to arise in utero. Given our findings, parental age associations with MLL rearrangements in IAL need to be investigated. Further, while epidemiological studies of IAL have primarily focused on maternal factors that may increase risk, our data suggest that paternal factors also need to be considered. Future IAL research to help elucidate the biological basis should include collection of detailed epidemiological data and biospecimens from both parents and infants to determine whether parent-of-origin de novo mutations and/or carcinogenic exposures are involved.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.