Big Data for Our Smallest Patients

How NCI Is Connecting Kids to Find Cures for Cancer

BY THE NIH CATALYST Staff

Every year, roughly 15,000 children and adolescents in the United States are diagnosed with cancer, a diagnosis no parent ever wants to hear. Although treatments have improved over the decades, many pediatric cancers remain stubbornly difficult to manage or poorly understood—and the data to study them are scattered like puzzle pieces under a couch.

That’s where the Childhood Cancer Data Initiative, or CCDI, comes in. Launched by the National Cancer Institute in 2019, the CCDI—a 10-year, $500 million effort—aims to do something audacious: Learn from every child diagnosed with cancer in the United States and do so by turning fragments of laboratory, genomic, and patient-reported data into a connected, national research framework.

The goal is deceptively simple: Gather enough high-quality data to better understand rare cancers and ultimately improve treatment and survival. The execution depends on effectively engaging the entire childhood cancer community, including patients and advocates, and an efficient collaboration between NIH extramural and intramural researchers.

Much of the data comes from extramural investigators across the country who enroll patients, collect biospecimens, and submit clinical and genomic data to centralized platforms. These investigators include large cooperative groups, pediatric hospitals, and national registries. NCI supports these efforts through contracts, grants, and technical infrastructure.

NIH’s intramural research program (IRP), mainly through the NCI Pediatric Oncology Branch, contributes its own clinical trial data. The IRP also helps define the rare cancers included in the CCDI and tests early-stage therapies that may later move into cooperative group trials, some open at the NIH Clinical Center. The broader IRP also helps design the data models and pipelines that ensure information coming into CCDI is usable, searchable, and shareable.

Together, researchers are not merely gathering the missing puzzle pieces of pediatric cancer data but finally fitting them into an accessible and cohesive picture.

The CCDI’s progress and current challenges will be on display at its annual symposium, with the theme “Collaborate. Innovate. Transform,” October 6–7, at the Natcher Conference Center (Building 45) on the NIH Bethesda campus. This year, CCDI will recognize six years of progress with expert speakers, networking opportunities, and science poster sessions. View the agenda or register to attend.

What follows below provides a snapshot of the symposium’s program with a focus on IRP’s role.

A White House call to action

The CCDI was born from a national call to action, publicly announced during the 2019 State of the Union address, when the White House committed to a new federal investment in childhood cancer research. This announcement was the culmination of efforts to address a critical need—the lack of a centralized, accessible system for sharing childhood cancer data (PMID: 37267580).

Then–NCI director Ned Sharpless took the lead in conceiving and developing the initiative. Sharpless recognized that although significant progress had been made in treating many childhood cancers, advancements for others were limited due to data being siloed in individual hospitals and institutions. He believed that by harnessing the power of data from every child, adolescent, and young adult with cancer, researchers could accelerate discoveries and improve outcomes.

With the promise of an annual budget of $50 million for 10 years, the CCDI set off to create a community-centered data ecosystem. This ecosystem would not only collect and harmonize diverse data—from clinical care and research to genomic sequencing—but also provide a platform and the tools to make it accessible to the broader community of researchers, clinicians, and advocates.

The initiative, however straightforward, represents a novel approach to data sharing, according to Brigitte Widemann, chief of the NCI Pediatric Oncology Branch and special advisor to the NCI Director for Childhood Cancer. Widemann helps to coordinate the intramural component of the CCDI. She said that Sharpless, who left in 2022, believed that if the plan could work at all, it would be in the realm of pediatric cancer, because pediatric oncologists tend to be highly collaborative, and because among cancer patients, a higher proportion of children and young adults are enrolled in cancer clinical trials than older adults.

“If you can do this for pediatric cancer, this could be a model for adults,” Widemann said.

Enter the conductor

Just before Sharpless left the NIH, he and James Doroshow, an NCI senior investigator, recruited Gregory Reaman as the CCDI scientific director. Reaman trained as a fellow at NCI in the 1970s. In his own career before NIH, he directed the Center for Cancer and Blood Disorders at Children’s National Medical Center; served on the faculty at George Washington University; and worked at the FDA as associate director for Oncology Sciences and later as associate director for Pediatric Oncology in the Oncology Center of Excellence. He also was the founding chair of the Children’s Oncology Group (COG), which is the world’s largest clinical trials network dedicated to childhood cancer. 

Reaman is now responsible for developing, directing, and coordinating the CCDI programs. Described as the CCDI symphony conductor because of his vast familiarity with all the players, his expertise and scientific oversight have been crucial to driving the CCDI’s vision and ensuring that its progress remains aligned with the needs of the childhood cancer community.

He also has been instrumental in the initiative's core mission of aggregating, harmonizing, and broadly sharing data that would have otherwise remain siloed—which, he said, is still picking up momentum. “There are still plenty of opportunities for people to use data and to contribute data,” Reaman said. He noted the need for adding more genomic data, for example, and developing large language models to mine data, particularly unstructured data from medical records, assignments well suited for the IRP.

Molecular Characterization Initiative

One element of the intramural–extramural CCDI effort is the Molecular Characterization Initiative (MCI), which provides advanced molecular testing to children, adolescents, and young adults newly diagnosed with cancer. One of the co-leaders is Jack Shern, a physician-scientist in the Pediatric Oncology Branch, and the project represents a collaboration among the NCI, the COG, and many community partners.

The goal is to collect and analyze diagnostic tumor samples and germline material to obtain a comprehensive genomic profile of every child’s tumor to refine diagnoses, guide treatment, and inform future research. Officially launched in March 2022, the initiative began by offering molecular characterization to patients with newly diagnosed central nervous system tumors. It has since expanded to include soft tissue sarcomas, neuroblastomas, and other rare cancers.

To highlight the initiative’s impact on this specific cancer, Widemann said that, since opening, the CCDI has enrolled nearly half of the people who are diagnosed with pediatric soft tissue sarcomas each year in the United States and sequenced their tissue samples.

According to Shern, who studies the genetic and molecular drivers of rhabdomyosarcoma, this effort has, for the first time, allowed researchers to build cohorts of people with these rare tumors that can be studied. “There are groups of patients who due to the rarity of their tumors, we really do not know the best way to treat,” he said. “This effort will change that.”

One of the initiative’s major achievements has been its quick turnaround time. The program aims to return molecular testing results to a patient’s physician within 21 days of receiving samples, allowing for timely, informed treatment decisions. Moreover, by performing assays such as whole-exome sequencing, RNA fusion analysis, and DNA methylation profiling, the MCI provides a detailed genetic and molecular blueprint of each patient’s tumor, resulting in a more precise classification of tumors and a deeper understanding of the genetic changes that drive these pediatric cancers.

For example, in brain tumors, the MCI has helped to identify over 75 unique tumor categories, allowing for more targeted therapies, according to COG.

The real paradigm shift has been the enhanced data sharing. Data collected through the MCI is de-identified and made available to the broader research community through the CCDI ecosystem in near-real time. This sharing creates an invaluable resource for scientists worldwide to study pediatric cancers and accelerate the development of new treatments.

A critical component of this work has been the close collaboration between NCI and COG (PMID: 40794906). This coordination ensures that high-quality samples are delivered efficiently, vital for the rapid turnaround time the MCI promises, and allows the initiative to receive data from a network of more than 200 institutions.

Key to this collaboration has been Erin Rudzinski, a professor of clinical pathology and laboratory medicine at Indiana University School of Medicine (Indianapolis), lead pathologist for COG’s participation in the MCI. Rudzinski oversees the accuracy of pathologic diagnoses of patients enrolled and specimens submitted, which is essential for ongoing analyses of possible associations between gene variants and specific pediatric cancers and their potential relevance to treatment outcomes and the discovery of new actionable targets for drug development.

Patient partners

Widemann, as both a member of the CCDI Steering Committee and co-lead of the CCDI Engagement Committee, has been at the center of the initiative’s strategic planning and community outreach. Her background as a pediatric oncologist and her success in developing the first FDA-approved medical therapy for neurofibromatosis type 1-related tumors has informed the CCDI’s mission to address the most difficult-to-treat cancers.

Central to the mission, Widemann said, is the CCDI’s commitment to integrating the perspectives of patients and patient advocates, which aligns the initiative to the real-world experiences and priorities of the people they aim to help. The CCDI actively seeks input from patients and their families about how rare tumors affect their lives. The CCDI inquires about daily challenges, quality-of-life issues, treatment burdens, and other types of information that is not systematically collected.

By participating in workshops and committees, advocates and patients help guide researchers to the questions that are most important to answer and the outcomes that matter most to those living with cancer, Widemann said. For example, patient groups play a fundamental role in communicating the purpose and progress of initiatives, such as MCI, to the wider patient community, encouraging participation in studies and clinical trials.

Such direct involvement helps to ensure that research is not only scientifically sound but also patient-centered, addressing the needs and concerns of the childhood cancer community at every stage of the process, from designing studies to disseminating results.

A registry for the rarest of the rare

The CCDI is attuned to all child and adolescent cancers. “Rare” cancers, by NCI’s definition, are those that affect fewer than 15 in 100,000 young individuals. Reaman said that the MCI has enrolled more than 6,000 patients between ages 0 and 25.

However, with community help, the CCDI is about to launch an even rarer “rare cancer registry” for pediatric, adolescent, and young adult populations (or AYA), defined as between the ages of 0 and 39 years. “Rare” here refers to cancer types thought to be diagnosed in fewer than 2 per 1 million people per year.

Mary Frances Wedekind Malone, an assistant research physician in the Pediatric Oncology Branch, will serve as the principal investigator for this registry. The goal is to gather more clinical data on these very rare cancers, which include chordoma and malignant peripheral nerve sheath tumor, both studied at the NIH Clinical Center.

Wedekind Malone said that the CCDI is building an infrastructure to process clinical data from the medical records, extract important information from those records, add genomic data and molecular profiling attained through sample collection, and place this information in the CCDI database for all researchers to access.

Data collection can be challenging for many reasons, particularly for the AYA population given this time in human life spans many transitions, Wedekind Malone said.

“Teens and those in their 20s are extremely hard to, first off, get a hold of and, second off, have them enroll in clinical studies…especially for clinical trials where they may not feel they are getting a huge benefit.” Also hard to reach are individuals in rural areas seen by community physicians and hospitals. In both cases, “we rely heavily on our advocacy organizations,” Wedekind Malone said.

Scaling for transformation

As the CCDI passes the halfway mark of its 10-year journey, its leaders emphasize that the work ahead may prove even more transformational. The infrastructure is now largely in place, from molecular characterization pipelines to shareable databases, registry frameworks, and patient-engagement platforms.

The next five years are about scaling, Reaman said—enrolling more patients, bringing in harder-to-reach populations, layering in new data types, and ensuring that researchers across the country—and around the world—can give to and take from the CCDI.

“It’s a resource,” said Reaman.

“[The CCDI] has begun to transform how we approach pediatric cancers, I think,” Widemann added. “It will be important if we can work toward excellent documentation of the value of CCDI, for example, of the MCI so that health insurances will cover this analysis [after CCDI funding ends]. I also hope we will have created well-characterized cohorts of patients that can serve as control [subjects] for interventional clinical trials.”

These ideas and hopes will be on display at the October 6—7 symposium. Beyond highlighting scientific milestones, the CCDI team plans to challenge its stakeholders to press forward on key fronts, including ensuring molecular diagnostics in childhood cancer as part of standard of care; leveraging big data and real-world evidence in childhood cancer research; and engaging industry, regulators, and advocates to drive transformation.

Perhaps most crucially, they aim to foster a shared sense of urgency among investigators, clinicians, and advocates, so that the data ecosystem truly accelerates progress, rather than simply documenting it.

If successful, CCDI may not just change the trajectory of pediatric oncology; it could also fundamentally redefine how we use data to tackle all rare diseases and all cancers.