In this JCO Article Insights episode, Emily Zabor interviews Dr. Gregory H. Reaman, the Scientific Director of the Childhood Cancer Data Initiative at the National Cancer Institute, on their paper titled "The Childhood Cancer Data Initiative: Using the Power of Data to Learn From and Improve Outcomes for Every Child and Young Adult with Pediatric Cancer". Dr. Reaman introduces us to the initiative, its goals and structure, and what has already been achieved since its launch.

TRANSCRIPT

The guest on this podcast episode has no disclosures to declare.

Emily Zabor: Welcome to this JCO Article Insights episode for the August issue of JCO. This is Emily Zabor, JCO's Biostatistics Editorial Fellow. And today I am interviewing Dr. Gregory Reaman, the Scientific Director of the Childhood Cancer Data Initiative at the National Cancer Institute, on their paper titled "The Childhood Cancer Data Initiative: Using the Power of Data to Learn from and Improve Outcomes for Every Child and Young Adult with Pediatric Cancer."

Dr. Reaman, welcome to our podcast.

Dr. Gregory Reaman: Thanks very much, Emily. Appreciate the invitation.

Emily Zabor: Dr. Reaman, could you start by introducing yourself and describing your involvement in the Childhood Cancer Data Initiative?

Dr. Gregory Reaman: I'm Gregory Reaman. I'm a Pediatric Oncologist. And I guess my involvement with the CCDI began shortly after the initiative was announced at the State of the Union media address in 2019, which was followed shortly thereafter by the formation of a working group by the NCI's National Cancer Advisory Board Board of Scientific Advisors. Given my role at the FDA at that time as Associate Director for Pediatric Oncology in the Oncology Center of Excellence, and the fact that I was the founding Chair of the Children's Oncology Group, I was an ex-officio member of this working group. So from very early on, I had involvement. I formally joined the NCI in November, left the FDA to assume the position as Scientific Director of CCDI.

Emily Zabor: That's great. So you've really been involved from the start. I had not really been familiar with the initiative until I read this paper. And as a cancer biostatistician, I was really excited to learn about this initiative, which sounds like it will ultimately create a very valuable data resource to be used for research purposes, among other things. So I think it's a really interesting project. So for our listeners who may not be familiar, could you describe the motivation for and goals of the Childhood Cancer Data Initiative?

Dr. Gregory Reaman: As I mentioned, this really is a very unique initiative, venture, if you will, on the part of the Cancer Institute and in large part driven by this fortunate infusion of funds to support childhood cancer research. And given the fact that pediatric oncology is very much a collaborative enterprise, it really does sort of follow that data sharing and using the power of data, its ability to be used by multiple investigators, irrespective of the source, aspirationally can improve outcomes for children cancer.

The three primary objectives– actually, this working group that I mentioned earlier put together a white paper that had 24 specific recommendations to the NCI. But there were three foundational objectives or goals. One was to learn from every child diagnosed with cancer, irrespective of the institution where they were diagnosed to receive therapy, to develop an ecosystem that would enable the submission, aggregation of data, and harmonization in a federated system that could then be accessed and used by investigators and analyzed to ultimately improve outcomes.

And then one objective, which was a little bit more specific, and that was to really focus on the opportunity to genomically classify tumors from newly diagnosed pediatric cancer patients, because this was something that obviously is much more widespread in the adult population, given the advent of targeted therapy and precision oncology and its more widespread use in medical oncology than pediatrics. And although many large academic institutions do have resources, the majority of smaller institutions do not. And when it's necessary and preferable to accurately and timely identify or diagnose a child's cancer that may actually provide information on treatment recommendations, the ability to do that and have it covered by insurance is sometimes problematic. So developing a program that would not cost patients or institutions anything and then make that data available to patients, families, and providers, as well as making it available for secondary research use, was a major goal and objective.

Emily Zabor: Yeah, that sounds like such an important initiative. The Molecular Characterization Initiative, which I understand has already enrolled and characterized the genomics of 751 participants just in the first year, I think is what the paper reported?

Dr. Gregory Reaman: That's correct. That was in the first year. We're now beyond the first year and we're approaching 2000 patients that have had their tumors genotyped and about 1500 results that have been returned to patients and providers.

Emily Zabor: That's fantastic. So with this linkage to the clinical data, that's going to be an extremely important data source. And I understand that the participation is currently limited to members of the Children's Oncology Group, which consists of over 200 children's hospitals, universities, and cancer centers. Can you describe in more detail what information it currently provides and how this initiative is going to be advertised and implemented to ensure complete participation across all of these member sites? Because that sounds like a really big challenge.

Dr. Gregory Reaman: Limiting the participation to the Children's Oncology Group initially was in no way meant to be exclusionary, but really provided an opportunity for linkage to clinical data. Since the Children's Oncology Group really represents nearly all of the pediatric cancer programs in the United States and some programs even outside the US, in Canada, and a couple of European sites, Australia and New Zealand, it was felt that given the resources that currently exist within the COG for specimen procurement, specimen submission, and then DNA and RNA extraction through the COG's Biopathology Center at the Nationwide Children's Hospital would really facilitate having the sequencing done at a single site, single institution, using a single platform. And also it provided an opportunity for some clinical data, including demographics, diagnosis, radiographic data, and treatment data that could be collected somewhat longitudinally from patients enrolled on the MCI.

Looking to make this as broad as possible since the objective of the CCDI is to learn from every patient, and every patient that we're concerned about not being able to capture adequately within the Children's Oncology Group are older adolescents and young adults with cancers that are more frequently seen in the pediatric population. So we are looking at ways to work with the COG's Biopathology Center to see if we can create systems that we can actually have specimens submitted from patients seen at institutions outside of the COG and molecularly characterized the same way. And that will be important as we launch another new planned initiative called the Coordinated National Initiative for the Treatment of Rare Pediatric and Young Adult Cancers.

Emily Zabor: Okay, that makes sense. So those adolescents and young adults are harder to capture since they're not being seen at those COG member institutions. Okay, well, that sounds like a big challenge to find those patients at their institutions and get them involved, but I think it's an important piece of this for sure.

Dr. Gregory Reaman: I should also point out that there were opportunities for some of the larger well, for all of the NCI-designated cancer centers, the pediatric programs associated with those cancer centers, to submit genomic data on newly diagnosed patients. That was something that actually transpired early on in the history of CCDI. So those data are in the CCDI's ecosystem.

Emily Zabor: Oh, that's great. So you collected the existing data.

Dr. Gregory Reaman: Right.

Emily Zabor: That kind of leads into my next question about aggregating data sources. With these disparate sources of pediatric cancer data, it seems like the aggregation is a lofty and important goal, but once that's complete, you're going to have this data ecosystem, which you said was one of the main goals of this initiative. I was wondering if you could tell us who will have access to this data ecosystem and what will be required for individuals to gain access.

Dr. Gregory Reaman: All of CCDI was predicated on this really being a community initiative if you will, so multidisciplinary and community-based. So patients, families, advocates, clinical researchers, physician providers, basic and translational researchers, researchers in public health and epidemiology. So there will be different levels of data that will be available to specific individuals. Patient-level data will be deidentified through a system of APIs that will be used that will enable the association of clinical data to existing molecular data and outcome data that might be available in the ecosystem. Those data will be- there are many data in the ecosystem that will be open source and available to anyone who is interested. This includes data from the NCCR in the Childhood Cancer Data Catalog, which is basically a listing of some close to 300 pediatric cancer databases that are available. The patient-level data will be sort of a controlled access. So there will be a requirement for individuals, investigators who wish to access that data, to sort of be certified, if you will, utilizing NCI and NIH data sharing requirements.

Emily Zabor: That makes sense. Yeah, you mentioned deidentification, but especially when we're dealing with these kinds of rare diseases, patient privacy does seem like it could be a concern. So what exactly are you doing to ensure that that is not something that gets violated through this process?

Dr. Gregory Reaman: I think there's every attempt to eliminate any PPI, HPI, obviously. So, again, most of the clinical data that are being provided currently are data that's coming from the Children's Oncology Group, where for every patient enrolled or registered through the COG and enrolled on a clinical trial, there is a COG ID number that is associated and that will be available only to the NCI and the CCDI to link it to unique specimen identification numbers, which are the only numbers that will be available to any investigator. So no one will be able to make the connection from the specimen identifier to the unique patient identifier in the COG.

Emily Zabor: That's great. And that way, you can really get access to all of the detailed data without concerns about privacy.

Dr. Gregory Reaman: Correct. And then being able to link all of these disparate data sets will really require the identification or the development, I should say, of a participant index. So that is one of our highest priorities right now in developing a CCDI participant index so that we would be able to link the identifier or clinical data with any research data or biologic data that may be available on patients to facilitate research plans and programs.

Emily Zabor: And through that process, is there also some method involved for identifying duplicated data? Because I assume some of these patients may get seen at different institutions over time, and that could be a concern that they end up in the database multiple times.

Dr. Gregory Reaman: That's exactly why I think developing the participant index is so critical to, number one, link, and number two, to avoid, prevent duplication, because you're absolutely right. There may well be the same patient data in multiple data sets, which are, of course, disparate. And the only way that they're going to be really utilizable and made interoperable is by linking them to the specific patient or individual patient.

Emily Zabor: Great. And do you have an idea of the timeline when that part would be complete and this data ecosystem would be available to researchers?

Dr. Gregory Reaman: The ecosystem is already available to researchers. We launched several months ago the CCDI hub, which is sort of the entryway or entry point, if you will, for access to the ecosystem. We hope to actually have the participant index up and running, and it's something that we've been working on for over a year, but actually available and utilizable within the next several months.

Emily Zabor: That's fantastic. We'll have to go check out the CCDI hub that's already out there then. Before we end, is there anything you'd like to share with our listeners that we haven't already discussed?

Dr. Gregory Reaman: Well, I think the one program that I mentioned just briefly, the Coordinated National Initiative for Rare Pediatric and Young Adult Cancers, we see there's a real opportunity to address a major unmet need. Fortunately, all pediatric cancer is rare, but there are some cancers that are extremely rare and for which there are, in many cases, no defining standard of care, and in many cases, there are no treatment protocols because of the difficulty mounting studies with such small patient numbers. So we see this as an opportunity to actually develop a registry that will provide, hopefully, natural history data that will inform clinical trials. All of these patients will be enrolled on the Molecular Characterization Initiative. So there will be the opportunity to hopefully learn if there are specific molecular drivers of some of these cancers that could inform the use of targeted drugs in a therapeutic approach to some of these. And we're looking to do this international collaboration with colleagues in the EU as well. So that is something that we just launched a task force to develop a listing of core critical data elements to collect on patients and then developing the registries for a number of these rare cancers.

Emily Zabor: That sounds like it's going to be a really valuable resource for planning and designing future clinical trials, so I'm glad to hear about that.

Dr. Gregory Reaman: And we would invite anyone who's interested to find out about the CCDI, to find out more about the CCDI, which they can do through cancer.gov/ccdi. There is an opportunity for people to register for newsletters. We have a series of webinars, many of which are designed now to actually provide training on some of the resources and platforms that are available currently through the ecosystem and things that we have all planned for future developments and use. So as I said, this is a community venture and we look to expand the community in every way possible.

Emily Zabor: That sounds great. So hopefully our listeners will take note of some of those resources in addition to this paper being out there, which will guide some people in the right direction to learn about this really great initiative for childhood cancer.

So, Dr. Reaman, it has been a pleasure speaking with you. And thank you so much for joining me today on this episode of JCO Article Insights.

Dr. Gregory Reaman: Thank you very much. It's been great to be here. Appreciate the opportunity.

Emily Zabor: This concludes this episode on the article, "The Childhood Cancer Data Initiative Using the Power of Data to Learn from and Improve Outcomes for Every Child and Young Adult with Pediatric Cancer." Thank you all for listening and please tune in for the next issue of JCO Article Insights.

The purpose of this podcast is to educate and to inform. This is not a substitute for professional medical care and is not intended for use in the diagnosis or treatment of individual conditions.

Guests on this podcast express their own opinions, experiences, and conclusions. Guest statements on the podcast do not express the opinions of ASCO. The mention of any product, service, organization, activity, or therapy should not be construed as an ASCO endorsement.

JCO Article Insights: Introducing The Childhood Cancer Data Initiative

Journal of Clinical Oncology (JCO) Podcast

JCO Article Insights: Introducing The Childhood Cancer Data Initiative