CHICAGO-A recent study suggests that children with non-chromosomal birth defects had a significantly higher risk of developing childhood cancer than those who did not have birth defects, according to findings presented at the AACR Annual Meeting 2018, April 14-18 (Abstract LB-161).

"Although we generally think of cancer as a disease of aging, it is in fact the leading cause of disease-related mortality in children as well," said the study's lead author, Jeremy M. Schraw, PhD, a postdoctoral fellow at Texas Children's Cancer Center, Texas Children's Hospital, and Department of Medicine, Section of Epidemiology and Population Science, Baylor College of Medicine in Houston. "It is a fairly rare event in children with a combined incidence of about 18 in 100,000 kids, and for the most part there are few established risk factors for pediatric cancers.

"Currently, in the U.S. only 3 percent of children will be born with a defect," he noted during a press conference. Birth defects can be grouped broadly into three categories, according to Schraw, including chromosomal anomalies, known single gene syndromes, and non-syndromic, which occur in the absence of an identifiable chromosomal or genetic cause.

"While we know that children with certain chromosomal conditions, like Down syndrome, have an increased risk of cancer, the majority of birth defects have no known chromosomal or genetic cause, but less is known about cancer risk in these children," Schraw noted in a statement. "There is growing evidence that non-chromosomal birth defects may predispose children to cancer, and we are trying to learn more about this connection so that we can potentially identify children who may benefit from early cancer detection."

Study Details

Researchers conducting the GOBACK (Genetic Overlap Between Anomalies and Cancer in Kids) study pooled statewide registry data from Texas, Michigan, North Carolina, and Arkansas, and linked information from birth certificates, birth defects registries, and cancer registries. Depending on the state, data collection began as early as 1992 and continued as recently as 2013, according to Schraw.

"By linking data retrospectively from these registries over a period of years, we were able to form a retrospective birth cohort of more than 10 million live births," he explained. Among this group, researchers identified 517,548 children with non-chromosomal birth defects and 14,774 children with cancer. Additionally, 1,787 children with a birth defect and cancer were identified.

"Our goal was to determine specific birth defect/childhood cancer associations among children who have these non-chromosomal or non-syndromic birth defects. The significance of that work is two-fold," Schraw said. "Firstly, this will help us identify novel cancer predisposition syndromes. Secondly, this might potentially help us improve surveillance strategies for childhood cancer in the future by helping us identify cancers at their earliest, most curable stages."

Cox proportional hazard models were utilized to evaluate associations between 60 birth defects and 31 childhood cancers, according to investigators. "This allowed us to estimate the risk of childhood cancers in association with particular birth defects, taking into account other factors that might affect the relationship like the age of the mother, the sex of the child, and the state in which the child was born," Schraw noted.

Key Findings

Researchers found that the risk of any cancer was approximately 2.6 times higher in children with non-chromosomal birth defects compared to those without a deficit, Schraw reported during his presentation.

Additionally, investigators determined that certain cancers were strongly associated with certain birth defects. "We found, for example, that the children with several forms of congenital heart disease, including ventricular septal defects, were at an increased risk of childhood cancer, notably hepatoblastoma or neuroblastoma, and that the magnitude of the increasing risk ranged from 7-fold to 20-fold depending on the defect and the cancer being considered," Schraw explained.

"In addition, children with several craniofacial or CNS defects such as craniosynostosis, hydrocephalus, or spina bifida were at an increased risk of cancer," he continued. "Often, but not exclusively, these associations were with CNS tumors such as ependymoma or astrocytoma. Again, the magnitude of the risk increase is quite large ranging from up to 75-fold in the case of children with spina bifida and non-RMS soft tissue sarcomas.

"For the first time, we were able to identify a number of specific non-chromosomal birth defects as being associated with specific childhood cancers, often at magnitudes of association that exceed those that have been reported previously."

Schraw noted that when considering these findings it is important to remember that the risk of childhood cancers is rare. "Even though the increases in risk that I present are pretty dramatic in some instances, do bear in mind that the majority of children who are affected by the birth defects that I referenced will not in fact go on to develop childhood cancer."

Strengths & Limitations

The unique strengths of this study, according to Schraw, include that it is a very large, population-based estimate of these risks. "We were able to look at them in a much more specific way than has previously been possible," he noted.

However, Schraw acknowledged there are important limitations to consider. "This is an observational study," he noted. "It does not establish a causal link between birth defects and childhood cancers. Secondly, we are constrained by a lack of data on the time period between the child's birth and the date of their cancer diagnosis."

Additionally, validation of these results could prove challenging. "We would absolutely like to see these findings, particularly the findings being presented for the first time, validated," Schraw said.

"Unfortunately, the rarity of pediatric cancers, and particularly in this setting they are compounded by the rarity of some of these birth defects, makes it very difficult [and] perhaps not feasible to validate these findings in a prospective study. We can attempt to study these same associations in independent registry-based populations, and case control approaches also might be applicable here."

At this time, Schraw noted that "it is too soon to be making clinical recommendations based on these findings. We need to better understand the ages of which these children are at risk, we need to validate the findings, and we need to understand the mechanism of that association better before we start making decisions about screening and treatment."

Looking Forward

Commenting on these findings, Louis M. Weiner, MD, Co-Chair of the Clinical Trials Committee for the AACR Annual Meeting 2018 and Director of Georgetown Lombardi Comprehensive Cancer Center in Washington, D.C., noted the importance of the research.

"Should these associations be validated in future studies it would be important information for the affected kids, for their parents, and for their health care providers," he said. "It is an example of how we can use big data approaches to uncover important associations that can lead to changes in clinical practice."

Schraw and his colleagues plan to expand this research into additional states. Also, he noted, they are "conducting genomic sequencing in families where a child has both a birth defect and cancer to explore whether there are shared genetic origins underlying these associations."

Catlin Nalley is associate editor.