Not all trials are created equal: the unique role of the NCI National Clinical Trials Network
Year: 2025
Type: editorial
Abstract: Shortly after President Richard Nixon declared a “war on cancer” through the signing of the National Cancer Act in 1971, the National Cancer Institute (NCI) organized the nationwide coordination of clinical trials through the formation of specialized networks. Some of the components for these networks originated nearly 2 decades earlier, tracing their roots back to 1955, but were formally structured in the early 1970s to bring together the “cooperation” of academic institutions, community practices, and various oncology specialists. The goal was to facilitate the development and conduct of large-scale cancer clinical trials that would engage different patient communities across the country and accelerate progress against this complex disease through the generation of robust data critical for regulatory decision making and clinical practice changes. Over the ensuing years, this “Cooperative Group Program” has undergone maturation and modernization as the science, supporting technologies, and treatments of cancer have evolved in an iterative manner. This evolution included a major modernization of the program in 2014 as a result of a 2010 Institute of Medicine report.1 The revamp included renaming the program as the National Clinical Trials Network (NCTN) and consolidated all the previous cooperative groups into 4 adult groups and 1 pediatric group. The 4 adult groups now include NRG Oncology, the SWOG Cancer Research Network, the Alliance for Clinical Trials in Oncology, and the ECOG-ACRIN Cancer Research Group. The aptly named Children’s Oncology Group conducts clinical trials for the youngest of our patients with cancer. Funding for this program, like all publicly supported cancer research, is appropriated to the NCI annually through congressional legislation and budget processes. The appropriation process is a heavily scrutinized and annually debated activity, yet there have been decades of overwhelming bipartisan support for the NCI investments and the work being done to reduce the US cancer burden through research. Once annual funds are appropriated by Congress, the NCI then conducts the work through a series of competitive grants, peer review, and NCI oversight to support operations, statistical and data services, correlative biospecimen and imaging infrastructure, and support for the member sites where participating patients are enrolled and treated in accordance with the clinical trial protocols (Figure 1). Importantly, the network the NCTN created, supported through the Cancer Trials Support Unit, boasts more than 3000 participating clinical trial sites, helping ensure that these clinical trials are available to patients with cancer in almost every community across the country, regardless of whether care is delivered at public, private, academic, or community practices. Schematic of the NCI National Clinical Trials Network (source: National Cancer Institute). COG = Children’s Oncology Group; NCI = National Cancer Institute; NCORP = NCI Community Oncology Research Program; NCTN = National Clinical Trials Network. Clinical hypotheses tested through the clinical trials developed and deployed by the NCTN include refinements of and challenges to the standards of clinical care, pushing the envelope not only with new therapies (eg, new anticancer drugs) but also addressing critical questions related to optimizing multimodality treatments. Such questions involving surgical treatments, radiation use, type and duration of therapies, chemotherapy duration and sequencing with other treatment modalities, incorporation of specific supportive care elements, coordinated use of these treatment options, and even the potential to deescalate treatments without introducing patient harm are all hallmarks of NCTN clinical trials. Importantly, these types of studies are exactly the kind of questions that should be supported by public resources (ie, US tax dollars) because they have both important patient care and health-care financial implications yet will never be a priority of industry or private-sector partners, who instead appropriately focus on development of new therapies and technologies. Thus, without many of the clinical trials conducted by the NCTN, we would miss the opportunity not just to add new treatments to care but to determine optimal care for patients with cancer. In this issue of the Journal, Dr Noticewala and colleagues provide a comparative analysis of the effect size differences in phase 3 oncology trials sponsored by the NCTN vs industry.2 They looked at 385 superiority-designed, nonadaptive, 2-arm, phase 3 oncologic randomized controlled trials with a time-to-event endpoint sponsored by industry (88%) or the NCTN (12%) between 2002 and 2020. Most trials evaluated chemotherapy (53%) or targeted therapy (27%), as would be expected given the relative large proportion of industry-sponsored studies in this data set. When comparing predicted hazard ratios for effect size against the observed effect, industry-sponsored studies more frequently achieved target (expected) hazard ratios than did those from the NCTN. Their conclusion is that overly optimistic estimations of effect sizes were more common in the NCI-sponsored trials and relatively uncommon in industry-sponsored studies. In addition, they noted that observed hazard ratios in industry-sponsored trials were substantially lower than those from NCTN studies, suggesting a larger effect size demonstrated in industry-sponsored trials. The findings are provocative but not entirely unexpected. Simply summarized, NCTN-sponsored studies are different. Even when compared with similarly titled trials sponsored by industry partners that also bring forward a new drug or device as a therapeutic intervention in pursuit of a US Food and Drug Administration registration, NCTN-sponsored trials are still different. As previously stated, NCTN studies are conducted across a highly heterogenous clinical trial network population of clinical sites, investigators, and patients. Unlike industry-sponsored clinical trials, where the sponsor or a third party (eg, a clinical research organization) may vet and select a relatively customized set of clinical trial sites that go through a formal site application and selection process, NCTN studies are instead designed to be available to any site that is qualified and rostered with the Cancer Trials Support Unit. This process allows for self-selection by sites to choose or decline the studies with which they participate. In designing NCTN studies, it is therefore imperative to anticipate writing studies with a relatively broad and inclusive set of eligibility criteria, protocol conduct that can align with different institutional processes and variability, and oversight in the absence of a real-time monitor who may issue data queries every few months. This selection bias can most likely account for some of the tighter hazard ratios expected and higher effect sizes seen in industry-sponsored trials because the patient and site population are more homogenous. In fact, several new oncology therapies have lost Food and Drug Administration registration or withdrawn their filings after the magnitude of the clinical impact in the postapproval setting (in more generalized clinical settings) failed to validate the initial clinical trial results (conducted in more homogenous and controlled environments). This perhaps argues that NCTN studies reflect more “real-world” outcomes, with a resultant lower observed hazard ratio compared with industry-sponsored events. The impact of the Cooperative Group Program, now NCTN, on bringing forward evidence-based advancements to the field of oncology and the collective deliverable of establishing the standards of care in the clinic is nothing short of remarkable. Unger and colleagues3 have provided a contemporary and comprehensive assessment of the impact of NCTN trials on clinical advancements. In all, 163 NCTN treatment trials involving more than 130 000 patient participants produced the following findings: NCTN trials influenced treatment guidelines in 63% of reviewed cases. 41% of trials led to statistically significant treatment effects, and 45% of those changed the standard of care. These trials affected practice in several high-prevalence cancers, including breast, lung, colorectal, and prostate. More than 1.2 million patients with cancer in the United States per year were estimated to receive care influenced by NCTN trial results. As we scrutinize and appropriately debate the use of public resources to ensure that they demonstrate a meaningful return on investment for the good of the nation, support for the NCTN and clinical trials that prioritize answering questions outside the scope of the private sector remains critical. Without it, our biomedical discoveries and new technologies cannot be fully incorporated into new care standards with the goal of optimizing care for all. The funder had no role in the writing of this editorial. We thank all the patients, caregivers, and clinical research team members who support the pursuit of clinical advancements through cancer clinical research participation. Thomas J. George, MD (Conceptualization, Data curation, Formal analysis, Project administration, Supervision, Writing—original draft, Writing—review & editing) and Theodore S. Hong, MD (Writing—review & editing). T.G. and T.H. are supported through NCI UC0CA180868-08. T.G. is additionally supported through NCI R50CA281930-01. T.G. reports the following disclosures: data safety and monitoring board of Seagen, Nihon Medi-Physics, KAHR, and ArbeleBio; consultant for Avammune Therapeutics, BillionToOne, Exact Sciences, and Summit Therapeutics; and research support (to institution) from Bristol Myers Squibb, Merck, AstraZeneca, Lilly, Bayer, Incyte, Ipsen, Genentech, Astellas Pharma, GlaxoSmithKline, Amgen, OncoC4, BillionToOne, Jounce Therapeutics, Elicio Therapeutics, AMAL Therapeutics/Boehringer Ingelheim, Seagen, Regeneron, and Deciphera. T.H. reports the following disclosures: advisory board of NeoGenomics Laboratories, Merck, GlaxoSmithKline, Nanobiotix, Zola Therapeutics, and Agenus; consultant for NextCure and Zola Therapeutics; and scientific advisory board of PanTher Therapeutics (Equity) and Lustgarten. No new data were generated or analyzed for this editorial.
DOI: 10.1093/jnci/djaf114
Authors: Thomas J George, Theodore S Hong
Language: English
Field: Medicine
Domain: Health Sciences
Open Access status: bronze
Funder: National Cancer Institute
OpenAlex ID: W4411009840