NIH Funds Pediatric Data Resource Center

The NIH has granted funding—up to $14.8 million over 5 years—to create the Kids First Pediatric Data Resource Center, a centralized portal for the largest dataset to be amassed for childhood cancer and rare disease research. Pooling genomic and phenotypic data from tens of thousands of patients, the resource should help researchers identify genetic underpinnings of pediatric cancer as well as structural birth defects, such as congenital heart problems.

Clinicians, bioinformaticians, and data scientists at Children's Hospital of Philadelphia (CHOP) in Pennsylvania will lead the initiative—a collaborative effort with the Ontario Institute for Cancer Research (OICR) in Toronto, Canada; University of Chicago in Illinois; Children's National Health System in Washington, DC; Oregon Health and Science University in Portland; and the data analysis company Seven Bridges in Cambridge, MA.

One of the biggest challenges in pediatric cancer and birth-defect research is finding and combining small numbers of cases, says Adam Resnick, PhD, who directs the multidisciplinary team at CHOP's Center for Data Driven Discovery in Biomedicine. Although cancer remains the top cause of disease-related mortality in children in the United States, pediatric cases account for less than 1% of total cancer deaths.

To pool data from scattered cases, the CHOP team partnered with Seven Bridges to build a cloud-based environment for storing, sharing, and analyzing genomics data. This platform, called Cavatica, launched in early 2016 with data from several pediatric brain tumor cohorts. (For studies of adult cancers, OICR and the University of Chicago developed the NCI's Genomic Data Commons, which launched last year.)

Funding for the project was made possible by the Gabriella Miller Kids First Research Act, which Congress passed in 2014. It authorized the NIH to spend $126 million over a decade for pediatric research. From that sum, the NIH awarded $12.6 million in 2015 and 2016 for whole-genome sequencing of more than a dozen cohorts of children with cancer and/or structural birth defects. The data will populate the new data resource portal supported by the latest Kids First award. By 2018, the database is expected to include more than 25,000 patient cases—more than double the size of The Cancer Genome Atlas, a data-driven NIH initiative launched in 2006 to study genomic changes across adult cancers.

In addition, the database will include sequencing data from first-degree relatives so that researchers can investigate familial connections.

The new resource “will leverage everything we've learned in the context of building Cavatica and what our partners learned in building the Genomic Data Commons,” says Resnick, “and extrapolate beyond that to develop a portal for end users that are as diverse as clinicians, patients, and researchers, and that integrates data across cancer and noncancer genomic datasets.” –Esther Landhuis