Uncovering the Genetic Connection Between Birth Defects and Childhood Cancers

Researchers co-led by Deanne Taylor, PhD, are analyzing extensive genomic data from cohorts of children with birth defects, cancer, or both, which potentially could lead to uncovering shared genetic origins.

Children who are born with specific structural birth defects have a higher risk of certain childhood cancers. For example, children born with heart defects have a higher incidence than expected of neuroblastoma or hematologic malignancies as they grow older. 

However, limited and siloed genomic data from affected children has previously constrained a deeper dive into cancer epidemiology in children with congenital defects, so it is not well characterized.

Now, with support from the NIH Common Fund, researchers at Children's Hospital of Philadelphia are leveraging a data integration platform called the Common Fund Data Ecosystem (CFDE) Data Distillery Knowledge Graph (DDKG) and advanced machine learning techniques. They will analyze extensive genomic data from cohorts of children with birth defects, cancer, or both, which potentially could lead to uncovering shared genetic origins.

This study, co-led by Primary Investigator Deanne Taylor, PhD, has the potential in the long term to inform more effective diagnostic and therapeutic strategies, thereby improving health outcomes and reducing the burden of illness and disability in affected children and adults.

The study team also aims to enhance the accessibility of the project's findings and extend the utility of the DDKG for the broader research community.

"We wanted to build something that could jumpstart researchers into utilizing large data aggregation engines," said Dr. Taylor, Director of Bioinformatics in the Department of Biomedical and Health Informatics (DBHi).

Curated Resource Informs Complex Research Questions

The DDKG, created by Dr. Taylor and her team with another Common Fund grant, intelligently and comprehensively integrates 40 million data points and 300 million relationships. It is organized as a curated resource for researchers to easily make hypotheses around large-scale genomics and genetics datasets, combined with metadata on diseases, compounds, phenotypes, and developmental stages.

"With the Data Distillery, we have all these human diseases, a lot of genes, proteins, and data from the Common Fund — all in one database," said Dr. Taylor, who is also a Research Associate Professor in Pediatrics at the Perelman School of Medicine in University of Pennsylvania. "We can ask very targeted questions like, 'What drugs would be most likely to affect gene expression in the liver?' And we can ask that of our database because all that data is in there already."

By importing and analyzing large-scale pediatric cohort genomics data, the team will expand the DDKG's information capacity to gain insights into the genetic drivers of childhood cancers and structural birth defects. This will pave the way for future research and clinical applications.

Datasets chosen from this project are based on epidemiological observations on the relationships between congenital heart defects and neuroblastoma or blood cancers, as well as between brain or central nervous system congenital defects and brain tumors. Germline and tumor data from the Kids First Data Resource Center will be obtained from representative cohorts of children with birth defects, children with cancer, and children with both birth defects and cancer.

For the children who have birth defects and then get cancer, the research team will be looking for commonalities, pathways, and mechanisms that could link them together during early development.

"We are using a resource that we built to integrate large amounts of data to study a complex question on genetics and development," Dr. Taylor said. "By integrating childhood cancer and birth defects data into the DDKG, we will fine-tune its ability to support machine learning studies on these complex relationships."

Additional primary investigators include Sharon Diskin, PhD, and Elizabeth Goldmuntz, MD, of CHOP, and Jonathan Silverstein of the University of Pittsburgh.