Jeremy M. Schraw
CHICAGO — Children with nonchromosomal birth defects such as congenital heart disease and central nervous system defects demonstrated significantly higher risk for certain childhood cancers, according to study results presented at American Association for Cancer Research Annual Meeting.
“Currently in the United States, about 3% of children will be born with a birth defect. [They] can fall into three categories: chromosomal anomalies, known single gene syndromes and nonsyndromic/nonchromosomal defects,” Jeremy M. Schraw, PhD, postdoctoral fellow at Texas Children’s Cancer Hospital and Baylor College of Medicine, said during a press conference.
About 87% of birth defects are categorized as nonchromosomal. Little is known about cancer risk among these children, Shaw said.
Schraw and colleagues pooled registry data from Texas, Michigan, North Carolina and Arkansas from 1992 to 2013 and linked information from birth certificates, birth defects registries and cancer registries to assess associations between 31 childhood cancers and 60 birth defects.
The researchers identified 517,548 children with nonchromosomal birth defects and 14,775 children with cancer. Researchers assessed associations for all conditions with at least five comorbid cases.
Cancer risk appeared more than twice as high among children with nonchromosomal birth defects (HR = 2.6; 95% CI, 2.4-2.7) after adjustment for maternal age, child’s sex and state of birth.
Certain cancers had stronger associations with certain birth defects.
Children with ventrical septal defects (HR = 10.5; 95% CI, 5.8-19) and those with craniosynostosis (HR = 9.6; 95% CI, 4.2-21.9) demonstrated significantly elevated risk for hepatoblastoma. Those with craniosynostosis also had a threefold increased risk for neuroblastoma.
Children with right ventricular outflow tract defects (HR = 7.1; 95% CI, 3.5-14.1) and those with left ventricular outflow tract defects (HR = 7.7; 95% CI, 3.4-17.1) had an increased risk for neuroblastoma.
Central nervous system defects appeared associated with increased risk for astrocytoma (HR = 6.7; 95% CI, 4.6-9.8), ependymoma (HR = 7.4; 95% CI, 3.5-15.7) and extracranial germ cell tumors (HR = 22.5; 95% CI 10.9-46.4).
Children with obstructive genitourinary defects also had a significantly increased risk for extracranial germ cell tumors (HR = 32.4; 95% CI, 16.2-64.6). Children with hydrocephalus without spina bifida exhibited an increased risk for ependymoma (HR = 23.4; 95% CI, 9.6-56.8) and astrocytoma (HR = 10.5; 95% CI, 5.7-19.6).
“This study was not designed to evaluate mechanisms,” Schraw said. “There are a few potential mechanisms, and they may vary depending on the association we’re considering. It’s possible, for example, that there are genetic variants that cause both the birth defect and the cancer. In some cases, there may be common in-utero exposures to which these children are subject.”
Children with certain other birth defects, such as cleft palate or cleft lip, did not demonstrate increased risk for childhood cancer.
“Childhood cancers are rare events so, even though the risks I presented were pretty dramatic in some instances, keep in mind the majority of the birth defects studied will not lead to childhood cancer,” Schraw said.
Study limitations included lack of data on the children’s health between birth and cancer diagnosis, as well as the lack of biological samples from the children.
“We do see that there are some instances with some birth defects for which the risk of childhood tumors is increased quite significantly, but we need to better understand the ages of the children who are at risk,” Schraw said. “We [also] need to validate the findings and understand the mechanism of those associations better before we make recommendations about screening and treatment.”
Louis M. Weiner
Future research could focus on expanding registry linkage to study rare cancers, as well as sequencing informative families to investigate genetic causes, Schraw said.
“Should these associations be validated in future studies, this will be important information for affected kids, for their patients and for their health care providers,” said Louis M. Weiner, MD, director of Georgetown Lombardi Comprehensive Cancer Center and moderator of the press conference at which this study was presented. “This is an example of how we can use big data approaches to uncover important associations that can lead to changes in clinical practice.” – by Cassie Homer
Schraw JM, et al. Abstract LB-161. Presented at: American Association for Cancer Research Annual Meeting; April 14-18, 2018; Chicago.
Disclosures: Schraw reports no relevant financial disclosures. One researcher reports serving as on an advisory panel of Baylor Genetics Laboratory.