Accelerating pediatric Hodgkin lymphoma research: the Hodgkin Lymphoma Data Collaboration (NODAL)

Abstract

Data commons have proven to be an indispensable avenue for advancing pediatric cancer research by serving as unified information technology platforms that, when coupled with data standards, facilitate data sharing. The Pediatric Cancer Data Commons, the flagship project of Data for the Common Good (D4CG), collaborates with disease-based consortia to facilitate development of clinical data standards, harmonization and pooling of clinical data from disparate sources, establishment of governance structure, and sharing of clinical data. In the interest of international collaboration, researchers developed the Hodgkin Lymphoma Data Collaboration and forged a relationship with the Pediatric Cancer Data Commons to establish a data commons for pediatric Hodgkin lymphoma. Herein, we describe the progress made in the formation of Hodgkin Lymphoma Data Collaboration and foundational goals to advance pediatric Hodgkin lymphoma research.

Hodgkin lymphoma has become one of the most treatable malignancies affecting children and adolescents. Excellent outcomes can now be achieved across all risk groups, with 5-year overall survival exceeding 90% (1,2). With such high survival rates, the burden of morbidity is shifting from the disease itself to the toxic late effects of treatment, including those associated with anthracyclines, alkylating agents, and radiation therapy. As a consequence, contemporary clinical trials have focused on the use of less toxic therapy and incorporation of novel agents and approaches to maintain excellent outcomes. To date, there is no uniform standard of care for treatment of pediatric Hodgkin lymphoma (3).

Advances in pediatric oncology are in large part attributed to collaboration among international research cooperative groups. Myriad treatment regimens have demonstrated efficacy for treating pediatric Hodgkin lymphoma in trials conducted in parallel across the globe. One unfortunate consequence of these parallel efforts is that basic aspects of management—including staging, risk stratification, treatment assignment, treatment regimen, and response criteria—vary among cooperative groups (4-6). These differences make efforts to pool patient data in retrospective comparisons difficult and pose barriers to the conduct of prospective intergroup trials.

Several international disease-focused consortia have been organized to address these challenges in other diseases. Examples include the International Neuroblastoma Risk Group (7), International Soft Tissue Sarcoma Consortium (8), and Malignant Germ Cell International Consortium (9).

Seeking to advance collaboration and standardize aspects of diagnosis, staging, treatment, and response assessment, pediatric Hodgkin lymphoma researchers established the Hodgkin Lymphoma Data Collaboration (NODAL) consortium and partnered with the Pediatric Cancer Data Commons (PCDC) (10), led by Data for the Common Good (D4CG), to develop consensus data standards and realize a data commons for pediatric Hodgkin lymphoma.

Foundational goals

The foundational goal of NODAL was to harmonize pediatric Hodgkin lymphoma data to facilitate cross-trial comparisons of patients across all disease stages and subtypes. The first step of this process was the development of a consensus data dictionary that provides a standardized format in which data collected on completed clinical trials are transformed. As the data-harmonization process is labor intensive and can result in loss of data richness, the data dictionary elements provide a template for case report forms of future clinical trials so that newly collected data already conform to the consensus data standard and do not require additional harmonization. Data harmonization necessitates that staging, risk stratification, and response assessment are applied uniformly; thus, another goal of NODAL is to standardize data reporting in these domains by using evidence-based methods.

Because the vast majority of children and adolescents with Hodgkin lymphoma will become long-term survivors, a data commons that includes long-term follow-up data provides an opportunity for survivorship and late-effects research on a broader scale. By using data elements that are harmonized across diseases (eg, treatment received) and pooling data, data commons provide a means to conduct cross-disease research on topics such as anthracycline-related cardiomyopathy and radiation therapy–associated secondary malignancy. Using long-term follow-up data can expand our understanding of toxic late effects of treatment and form the basis for changes in the next iteration of clinical trials.

Another advantage of pooling multiple datasets is the ability to study rare subgroups. Nodular lymphocyte-predominant Hodgkin lymphoma is an uncommon subtype generally associated with favorable prognosis but for which no standard-of-care treatment exists. Because of its rarity, treatment is based largely on small numbers of patients who were included in classic Hodgkin lymphoma trials, a small handful of clinical trials, and retrospective studies. By pooling multiple small datasets to increase the sample size, statistical power to detect differences in outcomes of interest in patients with this rare subtype will increase. Moreover, as novel agents (eg, immune checkpoint inhibitors) are introduced into treatment regimens for Hodgkin lymphoma, our understanding of the unique and emerging adverse effects associated with these treatments in pediatric patients will improve when data from multiple trials are pooled.

Early milestones

Initial discussions on the formation of NODAL began with three planning meetings between June 2018 and December 2019, including members of the Children’s Oncology Group (COG) Hodgkin Disease Committee, Pediatric Classical Hodgkin Lymphoma Consortium, and D4CG. A memorandum of understanding was signed in February 2021, and NODAL was formed with membership from two data contributors: COG and the St Jude–Stanford–Dana-Farber Pediatric Hodgkin Consortium. COG is the National Cancer Institute’s clinical trials organization devoted to childhood, adolescent, and young adult cancers, with more than 220 member organizations in the United States, Canada, Australia, New Zealand, and Saudi Arabia (11). As a result of discussions about the initial research questions to address with the NODAL dataset, three working groups were established: toxicity, nodular lymphocyte-predominant Hodgkin lymphoma, and radiation oncology.

The first version of the data dictionary (available at https://commons.cri.uchicago.edu/NODAL) was developed and established in June 2020, with working groups activated to focus on developing different clinically focused aspects of the data dictionary: staging, response, outcomes, toxicity, and dose modification. Hodgkin lymphoma experts from North America met over several months with a data standards expert from D4CG. Using the case report forms from the clinical trials anticipated to be harmonized for the NODAL data commons and their knowledge of current and in-development trials, the groups developed the first NODAL data dictionary. After this data dictionary was balloted by NODAL leadership and representatives, it was circulated to data contributor groups as the template for data submission.

A formal publication policy (available at https://commons.cri.uchicago.edu/NODAL) was adopted by the NODAL Executive Committee. The policy describes the process for reviewing and approving or rejecting project requests, authorship guidelines, and acknowledgments. Before transferring data, data contributor agreements were signed.

Quality assessment and quality control scripts were developed to assess the validity of values and ensure conformance of data to the PCDC data schema. Upon submission to the PCDC, data are subjected to these quality assurance and quality control scripts. Detailed and summary data quality reports are provided to data contributors, who are asked to address any errors in their data and resubmit for subsequent reassessment. This iterative quality assurance and quality control process continues until data are of sufficient quality to be staged for future production data releases.

Current state and future goals

As of December 21, 2023, data on 2254 patients enrolled on the COG AHOD0031 and Pediatric Hodgkin Consortium HLHR13 (NCT01920932), HOD05 (NCT03755804), HOD08 (NCT03755804), and HOD99 (NCT00145600) studies are now available in the publicly accessible PCDC data portal (portal.pedscommons.org) for anyone in the world to explore (Figure 1; Table 1). Additional studies have been identified for harmonization and addition to the dataset, including COG studies AHOD03P1 (12), AHOD0831 (13), AHOD0431 (14), AHOD1221 (15), and APEC14B1. We continue to engage with researchers around the world and invite other pediatric Hodgkin lymphoma researchers to contribute clinical trial and registry datasets.

Figure 1.

Hodgkin lymphoma related data elements currently available (bolded) and anticipated (non-bolded) in the PCDC.

Table 1.

Hodgkin lymphoma studies included in the pediatric cancer data commons (as of September 12, 2023)^a

Study	Cooperative group	Study phase	Age of patients included, y	Stages included (Ann Arbor)	Intervention	Patients in the PCDC	ClinicalTrials.gov identifier
AHOD0031	Childrens Oncology Group	III	0-21	I-IIA with bulk, 1-IIAE, I-IIB, and IIIA-IVA	Doxorubicin, bleomycin, vincristine, etoposide, prednisone, and cyclophosphamide chemotherapy backbone, with elimination of involved-field radiation therapy in patients with rapid early response and the addition of chemotherapy (dexamethasone, etoposide, cisplatin, and cytarabine) in patients with slow early response	1712	NCT00025259
HLHR13	Pediatric Hodgkin Consortium	II	0-18	IIB, IIIB, and IV	Brentuximab vedotin, etoposide, prednisone, and doxorubicin followed by cyclophosphamide brentuximab vedotin, prednisone, and dacarbazine, followed by response-based radiation therapy for patients with an inadequate response after two cycles.	77	NCT01920932
HOD08	Pediatric Hodgkin Consortium	II	0-21	IA or IIA, with nonbulky mediastinal disease, <3 nodal regions involved, and no E-lesion	Reduced-duration Stanford V chemotherapy and response-based radiation therapy	88	NCT00846742
HOD99	Pediatric Hodgkin Consortium	II	0-21	All stages	Vinblastine, brentuximab vedotin, methotrexate, and prednisone chemotherapy, with risk-adapted radiation therapy for patients with a favorable risk profile, alternating vinblastine, brentuximab vedotin, methotrexate, and prednisone and cyclophosphamide, vincristine, and procarbazine, and low-dose involved-field radiation therapy for patients at intermediate risk and Stanford V chemotherapy with low-dose, involved-field radiation therapy for patients with an unfavorable risk profile	296	NCT00145600

We also invite pediatric Hodgkin lymphoma researchers to propose research projects that use the growing dataset. Any researcher interested in conducting research using the dataset is invited to complete a brief project request form (available at https://commons.cri.uchicago.edu/NODAL) for review by the Executive Committee (Figure 2). The Executive Committee aims to review all data requests at monthly meetings. During the review process, the Executive Committee may provide suggestions to strengthen the project and suggest content experts who can contribute expertise and mentorship to the project team. In addition to engaging researchers who are not formally affiliated with NODAL, the consortium expects retrospective studies using the newly harmonized dataset to originate from within the consortium, including studies focused on management of nodular lymphocyte-predominant Hodgkin lymphoma and on harmonization of staging and response criteria. To reduce the burden of data harmonization, case report forms for planned prospective clinical trials run by major cooperative groups, including the COG and St Jude–Stanford–Dana-Farber Pediatric Hodgkin Consortium, will use standardized data elements from the NODAL data dictionary, where possible. Other groups are welcome to use the data dictionary as a resource during development of future clinical trials.

Figure 2.

Using NODAL to enhance collaborative research in pediatric and adolescent Hodgkin lymphoma. Figure created with BioRender.com and published under license.

One challenge to data commons sustainability is securing long-term funding to support maintenance and ongoing operational expenditures. Operational expenditures include those related to updating data standards to keep pace with scientific advancements, onboarding additional datasets, long-term storage costs, and expenses related to fulfilling data requests. One potential sustainability model could resemble that of open-access journals, where operational expenses are covered by fees paid by depositors for the purpose of making the data available to researchers in perpetuity. Recent updates to the National Institutes of Health Data Sharing Policy (16) require all National Institutes of Health–funded research studies to be accompanied by a Data Management and Sharing Plan outlining how the data the research generates will be shared. It is conceivable that costs associated with open data sharing will become a standard line-item on research budgets in the near future.

As has been the case over recent decades, novel treatment regimens will undoubtedly continue to improve survival and reduce late effects for children and adolescents with Hodgkin lymphoma. Yet, treatment approaches across the spectrum of pediatric and adult oncology vary widely. Our hope is that collaborative development of data standards and data sharing will facilitate standardization of the care of children, adolescents, and young adults with Hodgkin lymphoma. With the formation of NODAL and the development of clinical data standards, another indispensable tool in the therapeutic armamentarium is now sharper than ever: collaboration.

Data availability

The dataset described can be freely explored at https://portal.pedscommons.org. For access to line-level data, a project request can be submitted at https://docs.pedscommons.org/PCDCProjectRequestForm/.

Author contributions

Kirk D. Wyatt, MD (Conceptualization; Visualization; Writing—original draft; Writing—review & editing), Suzi Birz, MScMI (Conceptualization; Methodology; Project administration; Writing—original draft; Writing—review & editing), Sharon M. Castellino, MD, MSc (Project administration; Writing—review & editing), Tara O. Henderson, MD, MPH (Project administration; Writing—review & editing), John T. Lucas, MD, MS (Project administration; Visualization; Writing—review & editing), Qinglin Pei, PhD (Project administration; Writing—review & editing), Yiwang Zhou, PhD (Project administration; Writing—review & editing), Samuel L. Volchenboum, MD, PhD (Methodology; Project administration; Writing—review & editing), Brian Furner, MS (Data curation; Methodology; Writing—review & editing), Michael Watkins, PhD (Data curation; Methodology; Writing—review & editing), Kara M. Kelly, MD (Conceptualization; Project administration; Supervision; Writing—review & editing), Jamie E. Flerlage, MD, MS (Project administration; Supervision; Writing—original draft; Writing—review & editing).

Funding

This work was funded by American Lebanese Syrian Associated Charities (ALSAC), the National Cancer Institute of the National Institutes of Health under award No. P30CA021765 and St Baldrick’s Foundation Research grant award No. 585226.

Conflicts of interest

T.H., who is a JNCI associate editor and co-author on this paper, was not involved in the editorial review or decision to publish the manuscript. The authors have no relevant conflicts of interest to disclose relating to this work.