Streamlining the conduct of cancer clinical trials: new standard data collection practices for National Cancer Institute late-phase clinical studies

Abstract

The increase in the complexity of cancer clinical trials over the past several decades has led to a dramatic growth in trial cost and operational burden. The extent and frequency of data collection, particularly in late-phase trials that enroll many participants, have been major contributors to this problem. The Clinical Trials and Translational Research Advisory Committee of the National Cancer Institute (NCI) recently assessed the effect of these stressors on the NCI National Clinical Trials Network (NCTN) and recommended that data collection in late-phase NCTN trials be limited to data elements essential to address the primary and secondary objectives of the trial. The purpose of this commentary is to describe the rationale for this recommendation, progress toward implementation, and the development of new streamlined standard practices for data collection for late-phase NCTN trials effective January 1, 2025.

Clinical trials have been central to the remarkable advances made in cancer therapeutics and patient outcomes in the past half-century. They generate the critical efficacy and safety data needed for regulatory approval of new therapies as well as the evidence for informed decision making by patients and clinicians. Yet, the dramatic growth in the cost and complexity of executing these trials threatens the sustainability of the entire enterprise (1-3). The COVID-19 pandemic further worsened this situation through research staff shortages at clinical trial sites causing significant backlogs in data reporting (4). To address these growing strains on the National Cancer Institute– (NCI-) supported clinical trials system, the NCI charged its Clinical Trials and Translational Research Advisory Committee (CTAC) with developing a strategic vision for NCI’s clinical trials system for 2030 and beyond and with making recommendations to achieve that vision. In its November 2020 report, CTAC envisioned a system that supports “flexible, faster, simpler, accessible, and less expensive high-impact clinical trials that seamlessly integrate with clinical practice” (5). Fifteen recommendations were made by CTAC spanning several themes, including trial complexity and cost, operational burden, decentralized trial activities, promoting accrual and access, and new data collection approaches. One of these recommendations called for an assessment of potentially unnecessary data collection in late-phase trials, and to refocus data collection on only those data elements essential for addressing the primary and secondary objectives of the trial. In response to this recommendation, a new set of standard practices for streamlined collection and submission of data in late-phase NCI-sponsored clinical trials was developed. We present these standard practices for data collection to the oncology clinical trials community in this commentary and discuss implementation plans along with considerations for the future.

Data collection has long been seen as a promising target for improving the efficiency of cancer clinical trials. Several factors have led to the increase in the burden of data collection over time: the increasing complexity of cancer trial designs and the associated goal of enhancing scientific insight by means of correlative studies; the proliferation of secondary and exploratory trial objectives, many of which require additional data collection; the desire for extended follow-up to illuminate long-term outcomes; and actual or perceived regulatory requirements, especially with respect to safety-related data in studies that may be submitted in support of product labeling. The pressure to expand data collection is understandable because once the substantial fixed cost of launching a clinical trial is incurred, the marginal cost of adding data elements seems small, and their value in producing potentially important scientific or clinical insights is easy to imagine.

However, the cumulative effect of many such incremental additions to data collection contributes substantially to trial operational burden and cost. Each change can result in a snowball effect of effort downstream including protocol amendments, revisions to case report forms, database changes, and increased data management, monitoring, and quality control for the newly added data elements. Worse, it has become apparent that much clinical trial data end up never leading to published analyses or other forms of dissemination that could have a scientific or clinical impact (6).

One area of data collection that is widely perceived to be especially burdensome and, accordingly, has received special scrutiny, is adverse event (AE) reporting, including the range of severity (grade) for which reporting is required, the parameters that must be reported for each event, including investigator attribution of causality, and the timing and periodicity of regular and expedited reporting. The Alliance for Clinical Trials in Oncology, one of the NCI National Clinical Trials Network (NCTN) Groups, as well as its predecessor organizations, the Cancer and Leukemia Group B (CALGB), the North Central Cancer Trials Group, and the American College of Surgeons Oncology Group, have conducted a series of empirical investigations of the reliability of investigator attribution of AEs within CALGB and Alliance trials, demonstrating that attribution is difficult, unreliable, and of questionable value (7-10). A working group convened by the American Society of Clinical Oncology, with the collaboration of 4 pharmaceutical companies and CALGB, organized a reanalysis of AE data from 8 clinical trials to explore the consequences of reporting sampled rather than exhaustive toxicity data and developed recommendations for reducing AE data reporting on that basis. The project also examined the potential savings from more selective reporting of concomitant medications (11,12). This analysis concluded that for trials supporting supplemental indications, no significant safety signals would be missed if AE reporting was limited to serious AEs in all patients and grade 3 and greater AEs for a subset of patients.

The NCI formed a CTAC Streamlining Clinical Trials Working Group (hereafter, Working Group) in July 2022 to identify actions that the NCI could take toward implementing the strategic plan recommendation to limit data collected in late-phase trials to data elements that are essential for the primary and secondary objectives of the study. The Working Group’s first course of action was to identify potential opportunities for reducing data collection in late-phase trials. A sample of current NCTN late-phase (ie, phase 3 and phase 2/3) study protocols including case report forms was reviewed to characterize the extent of data collection and develop a list of categories of data that could be streamlined. Through an iterative consensus building process, the Working Group developed a set of standard practices for streamlining data collection and submission to clinical trial databases for late-phase NCTN trials without compromising safety or study objectives. These new standard practices for data collection are intended to be applied during protocol development to reduce unnecessary data collection. The set of standard practices was presented initially to CTAC as an interim report in November 2022, further refined based on input from the NCTN Groups, and accepted by CTAC as part of the Working Group’s final report on March 14, 2024 (13).

New standard practices for streamlined data collection

The new NCI standard practices for streamlined data collection and submission to the clinical trial database are intended to apply to phase 3 and phase 2/3 NCI-supported NCTN interventional trials conducted without registrational intent and not under a Food and Drug Administration (FDA) Investigational New Drug application (IND-exempt) (14). This group of trials makes up about 35% of late-phase clinical trials conducted under the NCTN cooperative agreement managed by the Cancer Therapy Evaluation Program (CTEP). The Working Group selected this subset of NCTN trials for the development of streamlined standard practices because these trials generally involve FDA-approved drugs with well-established safety profiles marketed in the United States and are not intended to support a new indication or a significant change in the labeling of the drug. As a result, NCI and the NCTN have greater latitude to define the data collected for these trials. Additionally, NCI and the NCTN Groups can immediately start to implement these standard practices for this subset of trials without requiring regulatory approval.

Streamlined standard practices for data collection and submission to the clinical trial database were developed for 7 categories of data, as summarized in Box 1. The recommended practice for AEs is to submit only AEs of grade 3 or higher to the trial database unless there is a stated study objective and analysis plan for the use of lower grade or solicited AEs prespecified in the protocol. For example, lower-grade AEs may be collected as part of a protocol objective to assess patient tolerability or treatment discontinuation. AE attribution and start and stop dates are not recommended to be submitted. The NCI Common Terminology Criteria for Adverse Events (CTCAE) term and grade (severity) are to be submitted for each AE submitted, as specified in the protocol (15). The CTCAE term refers to any unfavorable and unintended sign (including an abnormal laboratory finding), symptom, or disease temporally associated with the use of a medical treatment or procedure that may or may not be considered associated with the medical treatment or procedure.

Box 1.

SUMMARY OF NCI’S NATIONAL CLINICAL TRIALS NETWORK (NCTN) STREAMLINED DATA COLLECTION STANDARD PRACTICES FOR IND-EXEMPT PHASE 3 OR PHASE 2/3 TREATMENT TRIALS¹

ADVERSE EVENTS (AEs)

• Submit to the trial database only the following AE data:

  1. AEs of grade 3 or higher, unless there is a stated objective for the use of lower grade AEs in analyses prespecified in the statistical plan

  2. CTCAE² term and CTCAE grade for each AE

• Do not submit AE attribution or AE start or stop times

OTHER CATEGORIES OF DATA

Medical History³, Concomitant Medications, Physical Exam⁴, Laboratory Tests, Imaging and Other Assessment Procedures⁵, Patient-Reported Data

• Submit to the trial database only the following data:

  1. Data needed for analyses pre-specified in the statistical plan

  2. Data needed to document patient characteristics for publication or other reporting purposes

  3. At NCTN Group discretion, data needed to determine:

     • Eligibility

     • Treatment assignment or dosing

¹Adapted from the NCI CTAC Streamlining Clinical Trials Working Group report, March 2024, https://deainfo.nci.nih.gov/advisory/ctac/0324/Mandrekar2.pdf.

²NCI Common Terminology Criteria for Adverse Events (Version 5.0), November 2017, https://ctep.cancer.gov/protocoldevelopment/electronic_applications/docs/CTCAE_v5_Quick_Reference_5x7.pdf.

³Medical events or ongoing conditions identified at trial baseline either via patient report or via review of the patient’s medical record.

⁴Performance status assessed during the trial is considered a physical exam finding.

⁵A bone marrow biopsy is an example of other assessment procedures.

Streamlined clinical trial data collection practices for the 6 other categories of data (medical history, concomitant medications, physical exam, laboratory tests, imaging and other assessment procedures, and patient-reported data) focus on submitting only data needed to address a stated study objective and for analyses prespecified in the protocol or to document patient characteristics for publication or other reporting purposes. Data to determine eligibility or treatment assignment or dosing may be collected if necessary for central oversight of the trial. The frequency and duration of data collection should be limited to what is required to meet pre-specified trial objectives. Data collection plans for patient-reported data should include a justification of the patient-reported outcome instrument included in the trial with the frequency of data collection aligned with the prespecified trial objectives while minimizing patient burden.

These new streamlined standard practices apply only to data submitted for purposes of the clinical trial. They do not replace or override requirements to record data in the local electronic medical record as part of documenting medical care or other requirements imposed by local institutions (eg, local clinical practice guidelines) or regulatory authorities. These standards are also not intended to require submission of data elements that a NCTN Group does not currently require. For example, depending on the study objectives, there may be situations when not every data element specified needs to be collected. Although the Working Group also considered developing standard practices for data associated with the administration of therapeutic agents, it was determined that substantial additional discussion would be needed to develop guidance for submission of treatment data that aligns with the standard practice principles given the wide range of treatment-related data elements currently submitted among NCTN Groups.

The Working Group also noted that, in principle, automated data extraction from electronic health records (EHRs) into clinical trials databases could mitigate some of the burden of data submission. However, it will be some time before such automation has advanced to the point of having meaningful operational impact on NCI trials. Moreover, many of the data elements required for complex trials may not be part of routine care and so are likely not captured within the EHR without modification to EHR workflows. The Working Group recommended that NCI support the timely testing and implementation within NCI clinical trial networks of tools for extracting clinical trial data from EHRs. NCI recently released a Request for Information (NOT-CA-24-021) and is planning a workshop to understand current capabilities for the automated entry and extraction of data from EHRs.

Discussion

Although the recommended streamlined standard practices for data collection in late-phase NCTN treatment trials are straightforward, creating a culture of change and systematic implementation of these standard practices across the NCTN will be critical for meaningfully reducing the operational burden of these trials. NCTN Groups and their investigators will need to adopt behaviors and mindsets to systematically apply these standard practices and require the submission of only essential data. For example, investigators will need to assess the utility of each proposed data element to ensure it is essential to the trial and adheres to the standard practices. Exceptions are expected to be uncommon, but requests to deviate from the standard practices will be considered by CTEP during the protocol review process if there is compelling scientific or medical rationale in support of scientific objectives or patient safety.

It is expected that all NCTN phase 3 and phase 2/3 IND-exempt treatment trial protocols activated on or after January 1, 2025, will adhere to these new standard practices for data collection. Many of the principles underlying the standard practices could be applied to non-treatment late-phase, IND-exempt NCTN and NCI Community Oncology Research Program (NCORP) trials across the cancer control spectrum. Although not required, we encourage NCTN and NCORP investigators to apply these principles as appropriate for these trials.

To expedite implementation, NCI and the NCTN formed a Streamlining Clinical Trials Implementation Committee in April 2023 with representation from each NCTN Group Operations and Statistics and Data Management Center as well as the NCTN Imaging and Radiation Oncology Core. Led by Dr Sumithra J. Mandrekar, Professor of Biostatistics and Oncology, Department of Quantitative Health Sciences at the Mayo Clinic and Ms Andrea Denicoff, Head, NCTN Clinical Trials Operations, CTEP, Division of Cancer Treatment and Diagnosis, NCI, the committee has reviewed and provided operational guidance on the standard practices, discussed each NCTN Group’s implementation plans to identify best practices, and advised on system-wide changes that are necessary. The committee is currently working on defining metrics for evaluating the impact of the standard practices on operational burden. Potential metrics include assessing the number of data elements and queries issued by category before and after implementation of the standard practices, the number and proportion of NCTN trials to which the practices apply, and summary of the type of exceptions to the standard practices.

Implementation of the standard practices is well underway with the initial focus on approved trials currently in development. An April 2024 analysis revealed that there are 13 IND-exempt trials across the NCTN Groups that are in development and expected to launch on or after January 1, 2025. Widespread changes such as these require training of site personnel and study investigators on the rationale for the new streamlined data collection practices and the changes to the data submission process. NCTN Groups have been providing education sessions at their group meetings and working with study investigators as protocols and case report forms are developed. CTEP has posted the standard practices on its website along with other policies and guidelines for NCI trials and will be reviewing protocols as they are submitted for compliance (16). Additionally, CTEP’s Medidata Rave clinical data management system and the CTEP Adverse Event Reporting System (CTEP-AERS) are being updated to accommodate elimination of the requirement to submit AE attribution and AE start/stop times for IND-exempt trials while maintaining the integration of these systems for other trials.

Although the initial implementation of the standard practices for reducing unnecessary data collection focuses on IND-exempt trials, many of the principles developed could also be extended to trials conducted under an IND in the appropriate clinical and regulatory context. The FDA published guidance in 2016 regarding the extent of safety data collection needed in late-stage pre-market and post-approval clinical investigations (17). The document highlighted the types of data that may not be necessary to collect or could be collected less frequently in certain settings such as non-serious adverse events not associated with dose modification, drug discontinuation, or withdrawal from the trial; routine laboratory monitoring; information on concomitant medications; and patient history and physical exams. Examples of the types of clinical investigations that may be considered for these selective safety data collection procedures include 1) clinical investigations of new indications of an approved drug; 2) post-approval clinical studies and trials conducted to fulfill post-marketing requirements and commitments; 3) late-stage pre-market and post-approval outcome clinical trials; and 4) post-approval clinical investigations in a different patient population or with different doses or other conditions of use. This guidance provides industry sponsors with the flexibility to design and implement protocols with selective data collection when a drug’s safety profile has already been well established. In addition, the FDA similarly calls for the application of pragmatic operational elements in the post-marketing setting, including streamlined data collection (18).

Some of these principles were applied by NCI, the FDA Oncology Center of Excellence, and SWOG, an NCTN Group, in the collaborative development of the NCI Pragmatica-Lung Trial (NCT05633602), a study with registrational intent being conducted under an IND that applies a highly streamlined, pragmatic approach to data collection. This phase 3 trial is designed to confirm promising results of a phase 2 study in patients with advanced non-small cell lung cancer comparing the overall survival of patients treated with the combination of pembrolizumab and ramucirumab vs standard of care. No imaging, biospecimens, or laboratory studies are required by the protocol, and only grade 5 and unexpected grade 3 and 4 adverse events are required to be reported. The hope for very streamlined trials such as this one is that the reduced burden of participation for sites and patients will result in rapid enrollment, broadened patient access, less attrition, and faster delivery of new therapies to patients. The initial brisk enrollment since study launch in March 2023 is encouraging with more than 70% of the planned 700 patients enrolled within the first 15 months of the study (19). Although not all cancer clinical trials involve agents and questions that lend themselves to such a pragmatic approach, in the right clinical and regulatory context, along a continuum from explanatory to pragmatic, trials conducted under an IND can incorporate streamlined procedures and data collection (20-22).

NCI is committed to the timely implementation of these new streamlined standard practices for data collection, which are expected to substantially reduce the burden for sites and patients to participate in IND-exempt NCTN trials, potentially incentivizing greater community site participation, facilitating more rapid enrollment, and providing greater access to clinical trials. We look forward to working with our partners in the broad oncology community to further assess ways in which these data collection principles could be extended to trials conducted under an investigational new drug application in the right clinical and regulatory context. Efforts such as these to modernize and reduce the complexity of clinical trials, and to include more people from diverse backgrounds, align with several of the goals outlined in the National Cancer Plan framework for realizing the Cancer Moonshot goal of reducing the cancer death rate by 50% in 25 years and improving the experience of people and their families living with and surviving cancer (23,24).

 

This is a U.S. Government work and there are no restrictions on its use. The content of this publication does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the U.S. Government.

Data availability

No new data were generated or analyzed in this article.

Author contributions

Sheila Prindiville, MD, MPH (Conceptualization; Project administration; Writing—original draft; Writing—review & editing), Sumithra Mandrekar, PhD (Conceptualization; Writing—original draft; Writing—review & editing), Neal Meropol, MD (Conceptualization; Writing—review & editing), Andrea Denicoff, RN, MS (Writing—review & editing) Oren Grad, MD, PhD (Conceptualization; Writing—review & editing), Judith A. Hautala, PhD (Writing—review & editing), James H. Doroshow, MD (Conceptualization; Writing—review & editing).

Funding

This work was supported in part with federal funds from the National Cancer Institute, National Institutes of Health, Contract NSFOIA040860 to Institute of Defense Analyses and Grant U10CA180882 to Alliance for Clinical Trials in Oncology.

Conflicts of interest

Dr Mandrekar is the Group Statistician for the Alliance for Clinical Trials in Oncology and is on the advisory board for Harbinger Health. Dr Meropol reports employment at Flatiron Health, Inc, which is an independent member of the Roche Group, and stock ownership in Roche.

Members of the CTAC streamlining clinical trials working group

Sumithra J. Mandrekar, Mayo Clinic College of Medicine, Rochester, MN (Co-Chair)

Neal J. Meropol, Flatiron Health, New York, NY (Co-Chair)

Charles D. Blanke, Oregon Health and Sciences University, Portland, OR

Gary C. Doolittle, University of Kansas Medical Center, Westwood, KS

Michael V. Knopp, University of Cincinnati, OH

Seth P. Lerner, Baylor College of Medicine, Houston, TX

Robert S. Mannel, University of Oklahoma Health Sciences Center, Oklahoma City, OK

Suresh S. Ramalingam, Emory University School of Medicine, Atlanta, GA

Victor M. Santana, St Jude Children’s Research Hospital, Memphis, TN

Julie M. Vose, University of Nebraska Medical Center, Omaha, NE

George Wilding, University of Wisconsin School of Medicine, Madison WI.