A brief report sets forth a “new normal,” lets NCI start to streamline clinical trials

Given the urgent need addressed by this initiative, the recommended standard practices are presented to the NCI in this interim report rather than wait for a comprehensive report addressing all the tasks assigned to the Working Group.

This strategy will allow NCI to consider the recommended standard practices and implementation strategies, including stakeholder consultations for further refinement as appropriate, identification of technical requirements for operationalization of the proposed standards, and revision of NCI guidance documents to reflect the standards.

The recommended standard practices presented in this interim report do not represent the final set of Working Group recommendations. Once the Working Group completes deliberations for all assigned tasks, a final report will be issued.

We must modernize clinical trials. We’ve got to be nimbler, faster. I know we’ve been saying this forever, but it is more imperative than ever. And we’ve set a goal to double clinical trial accrual. 

We can’t leave important work on the table because we lack the bandwidth to conduct our studies. We have to reinvent how we approach, design, and conduct clinical trials and report our results. We’ve learned some things from COVID, but there’s a lot more that we can think about. Innovation in clinical trials is absolutely essential. 

On that note, I really look forward to Dr. [Neal] Meropol [vice president of research oncology, scientific and clinical lead, clinical research at Flatiron Health] and Sumithra J. Mandrekar [professor of biostatistics at Mayo College of Medicine and oncology group statistician at Alliance for Clinical Trials in Oncology] session at noon today, which focuses on streamlining clinical trials. 

I think that’ll be very illuminating for us all and is a wonderful place to start. I think all of you have the knowledge we need to achieve these kinds of goals, and you also know that there are resources needed to make this happen, and that’s why your input is so important.

I’ve been watching this off and on for a number of years, and it’s always good to think about getting what we need and not trying to do the things we don’t really need. Did you make any effort to try to think about what would happen if you applied this to an NCTN trial or two? 

If you took what we did for that trial and you took what we might have done if we used these guidelines, would we have missed anything big? Would we have missed concerns? I don’t know, toxicity concerns, or things like that, because we use these more streamlined guidelines?

Nancy, that is an excellent question. Have we done that in these areas? Probably not, because it’s not very easy to identify studies that collected comments versus, I mean, those are the easiest ones; right? We probably never utilize comments. And so that’s probably an easy way to say we don’t, there was not a lot of value for these data that were collected. But it’s kind of hard to do that in a very systematic way, but this is a great idea. I think one of the things [we] did was to go through our case report forms across a number of large phase III trials. 

And that’s kind of how we landed on these particular categories because it was very inconsistently collected. It wasn’t standard across trials. It was kind of very not systematic and standardized, I will put it that way. And so that’s what brought us together to think through these specific data elements and say, okay, why are we collecting it in some trials, and not on all trials? Why are we collecting it a certain way in certain trials and not comprehensive otherwise? And so, that’s where we landed.

And then the second one I will say is in terms of have we tried to see, if we didn’t collect this, what would’ve happened? Would we have missed a signal? I don’t think we have done that exercise, but I will put a plug in for an Alliance trial that’s coming up, where we are attempting to not collect attribution on that study, because it is on approved agents and it’s a randomized study. 

And we got buy-in from the NCI and the study team to say, okay, if we only collect whatever the study team is reporting and we have information on both arms, we can probably figure out what, just by subtracting the rates of adverse events, we would probably be able to figure out what is attributed to the treatment.

So, we are making some progress while we are working on this. But yeah, yours is a great question and I think we should take it back to the working group to see if there is a way to address that. I don’t know, Neal, if you had other comments.

I’ve been watching this off and on for a number of years, and it’s always good to think about getting what we need and not trying to do the things we don’t really need.

Nancy Davidson

I would just underscore exactly what you said, and that is that the process that led to this compendium of categories basically asked the group to pressure-test what if we eliminated this, and were these data that were collected, analyzed in any way at all? 

And we found that a lot were not. Also, the reliability of certain parameters was low. So, for example, attribution. The committee highlighted that work had been done to show inconsistencies in attribution from evaluator to evaluator, which led us to conclude that attribution is really not useful. And at least that’s our hypothesis for why it found its way into the recommendations.

Marc R. Theoret, an ex officio member of the working group and deputy director of the FDA Oncology Center of Excellence, said FDA would be willing to discuss streamlining data collection requirements for IND studies, which can be used to support new indications. Said Theoret:

Because these are IND-exempt, meaning that they’re not trials that would be intended to support a new indication or meaningful changes to the labeling, there’s much different considerations from the FDA perspective, and we’d absolutely be happy to participate in a more integrated and longitudinal way when you start thinking about the translation to IND studies.

One of the things that we were certainly thinking about is situations in general where collection of limited data would be appropriate to interest very specific science-based day questions. I just wanted to make, just highlight even in the IND-exempt kind of phase, we’re talking about marketed products. 

So, there is a difference between collection of safety information and then the reporting requirements, of course, for marketed products. But no, I just wanted to flag that we’re happy to, as this moves forward into thinking about ways to limit data collection in the IND space, happy to provide additional input.

The text of the interim report follows:

Introduction

In November 2020, the Clinical Trials and Translational Research Advisory Committee (CTAC) ad hoc Strategic Planning Working Group (SPWG) released its report, which envisioned the development of flexible, faster, simpler, and less expensive high-impact clinical trials that seamlessly integrate with clinical practice. 

The SPWG developed 15 recommendations and 3 operational initiatives that span the following themes:

• Trial complexity and cost
• Decentralized trial activities
• Promoting accrual and access
• New data collection approaches
• Patient-report outcomes (PRO) data for clinical trials
• Operational burden
• Statistical issues
• Workforce outreach and training

Even as the SPWG’s report was being brought to completion, the COVID pandemic was imposing extensive operational, workforce, and budgetary challenges on key partners in the NCI clinical trials program: the NCI-Designated Cancer Centers and their affiliates, which are central to NCI’s National Clinical Trials Network (NCTN), the community hospitals and oncology practices that participate in NCTN trials through the NCI Community Oncology Research Program (NCORP), and other participants in NCTN studies. These challenges have persisted even as the acute phase of the pandemic has passed.

In July 2022, in response to the urgent need to mitigate the operational burden of NCI-sponsored clinical trials without compromising trial objectives or patient safety, NCI convened the CTAC ad hoc Streamlining Clinical Trials Working Group to advise the NCI Director and CTAC on the implementation of three of the SPWG recommendations:

• Limit clinical trial data collection in late phase trials to data elements essential for the primary and secondary objectives of the trial (Recommendation TCC1)
• Resolve the logistical and data quality challenges of extracting clinical trial data from electronic health records (EHR) (Recommendation NDCA1)
•  Engage EHR and Clinical Trial Management Systems (CTMS) vendors to create mechanisms for automatically integrating study-specific documents into local
• implementations of their products (Recommendation OB1)

The Working Group is co-chaired by Dr. Sumithra J. Mandrekar, professor of biostatistics and oncology, Group Statistician, Alliance for Clinical Trials in Oncology, Department of Quantitative Health Sciences at the Mayo Clinic and Dr. Neal Meropol, vice president of research oncology, scientific and clinical lead, clinical research at Flatiron Health. The full membership of the Working Group is provided [below]. 

The process that led to this compendium of categories basically asked the group to pressure-test what if we eliminated this, and were these data that were collected, analyzed in any way at all? And we found that a lot were not.

Neal J. Meropol

The Working Group’s initial efforts have focused on the SPWG recommendation to limit data collection in late phase trials. In preparation for the Working Group’s deliberations, the NCI Coordinating Center for Clinical Trials (CCCT) reviewed study protocols and case report forms for recent NCTN phase III adult treatment trials to characterize the scope and extent of data collection in relation to study objectives. Based on this analysis and in consultation with CTAC members and other expert informants, CCCT developed a list of potential opportunities for reduction of data collection supported by empirical findings and expert judgment.

At its first plenary meeting on July 13, 2022, the Working Group discussed these potential opportunities in light of the findings of the trial analysis and identified several to be pursued. During the summer of 2022, via an iterative process of comment and review, a draft recommended set of proposed standard practices for data collection for adult, late phase, IND-exempt trials was developed, reflecting the consensus of the Working Group. At its second and third plenary meetings, held on Oct. 4 and Oct. 25 respectively, the Working Group discussed and refined this draft resulting in a set of proposed standard practices for recommendation to NCI.

Given the urgent need addressed by this initiative, the recommended standard practices are presented to the NCI in this interim report rather than wait for a comprehensive report addressing all the tasks assigned to the Working Group.

This strategy will allow NCI to consider the recommended standard practices and implementation strategies, including stakeholder consultations for further refinement as appropriate, identification of technical requirements for operationalization of the proposed standards, and revision of NCI guidance documents to reflect the standards. 

The recommended standard practices presented in this interim report do not represent the final set of Working Group recommendations. Once the Working Group completes deliberations for all assigned tasks, a final report will be issued.

The Working Group’s initial recommendation and associated set of proposed standard practices is presented in the following section of the report.

Recommendation

A set of standard practices for data collected in NCI phase III and phase II/III adult, IND-exempt, treatment trials should be established.

To realize this recommendation, the Working Group proposes an initial set of standard practices for trials that meet the following criteria:

• Managed by the Clinical Investigations Branch of NCI’s Cancer Therapy Evaluation Program (CTEP)
• Phase III or Phase II/III
• Interventional
• Focused on treatment
• Adult
• IND-exempt¹

The proposed practices are intended to define a “new normal” for data collection that is less burdensome, more efficient, and more sustainable. It is important to note that the practices are not intended to be applied rigidly at the cost of compromising key study objectives. The intent is that investigators may depart from these standards, but for each proposed departure, justification specific to the clinical details and scientific objectives of the trial should be provided and undergo review according to established protocol review processes.

The following data collection practices are recommended:

1. Adverse Events (AEs)

1. Collect only AEs of grade 3 or higher, unless assessment of tolerability related to lower-grade AEs is a stated objective with a prespecified analysis plan
2. For each AE, collect only CTCAE² term and CTCAE grade
3. Do not collect AE attribution or AE start/stop times
4. Solicited AEs³, regardless of grade, should be limited to those that would result in dose modification, treatment discontinuation, or non-adherence

2. Medical History⁴

1. Collect only those medical history items that are relevant to trial inclusion/exclusion criteria

3. Concomitant Medications

1. At baseline, collect concomitant medications only if their use requires modification of the study treatment
2. During the trial, collect only changes in concomitant medications that cause modification or discontinuation of the study treatment

4. Physical Exam

1. Physical exams should be conducted according to standard of care, augmented by any trial-specific requirements
2. Only the following physical exam findings⁵ are collected:
   • Findings that are protocol-specified endpoints or are required to assess protocol-specified endpoints
   • Findings that represent AEs (per section 1 above)
   • Findings that result in dose modification or treatment discontinuation

5. Laboratory Tests

1. Laboratory tests should be conducted according to standard of care, augmented by any trial-specific requirements
2. Only the following laboratory test results are collected:
   • Test results that are protocol-specified endpoints or are required to assess protocol-specified endpoints
   • Test results that represent AEs (per section 1 above)
   • Test results that result in dose modification or treatment discontinuation

6. Imaging and Other Assessment Procedures (e.g., bone marrow biopsies)

1. Limit imaging and other assessment procedures to those required to meet specified trial objectives (e.g., determining treatment assignment or modification, assessment of clinical outcomes)
2. The cost of any imaging or other assessment procedures not covered by insurance must be covered by the research study

7. Patient-Reported Data

1. Patient-medication diaries should not be required unless the protocol defines how the data will be analyzed to address specified trial objectives
2. Data collection plans for patient-reported outcomes (PRO) must address how PRO instruments⁶ will be chosen and data collection scheduled⁷ to achieve specified trial objectives while minimizing patient burden

Although the Working Group has not focused on the operational mechanics of data collection, one aspect did emerge during its deliberations: the need to assure that data collection requirements are specified consistently across study protocols, case report forms, and data collection calendars/schedules, and that this consistency is maintained when protocols are amended.

Conclusion

Timely implementation of a set of standard practices for data collected in NCI phase III and phase II/III, adult, IND-exempt, treatment trials is expected to have a material impact on the operational burden of these trials and will provide important insights that will inform future Working Group discussions addressing the development of data collection standards appropriate to the requirements of late phase IND trials and of pediatric trials, respectively.

These subsequent discussions will draw on lessons learned in advancing to implementation the proposed practices outlined in this report for IND-exempt trials. The Working Group notes that early phase trials have distinctive scientific and clinical objectives that will require careful reconsideration of each of the topics covered in the current set of proposed practices.

Implementation of these standards is also synergistic with ongoing efforts to assess and address operational frictions and inefficiencies in data collection processes experienced by clinical research associates and other front-line operations staff working on NCI-sponsored clinical trials. Changes in standard operating procedures at NCI and the NCTN Groups should address required changes in data collection scope and improvement in data collection procedures in an integrated, transparent, and efficient manner.

End notes

1. IND-exempt studies are those that meet criteria defined by the Food and Drug Administration for exemption from the requirement to submit an Investigational New Drug application to the FDA. In general, studies that are IND-exempt involve agents that have received marketing approval in the United States and that are not intended to be reported to the FDA in support of a new indication or other significant change in the labeling of the agent. Approximately 40% of NCI’s NCTN phase II/III and phase III trials are IND-exempt. 
2. Common Terminology Criteria for Adverse Events.
3. “Solicited AEs” are defined as protocol-specified AEs that are required to be assessed and reported on a regular schedule (e.g., with each treatment cycle) as present or absent.
4. “Medical history” is defined as medical events or ongoing conditions identified at trial baseline either via patient report or via review of the patient’s medical record.
5. Performance status assessed during the trial is considered a physical exam finding and should be collected if it meets criteria specified in 4b.
6. “PRO instruments” may include complete instruments, selected questions and/or rating scales, as appropriate to the trial’s scientific objectives.
7. “Scheduling” includes timing, frequency, duration of follow-up, and coordination with other data collection activities.

Ad hoc Working Group on Streamlining Clinical Trials

CO-CHAIRS

Sumithra J. Mandrekar, PhD 
Professor of biostatistics,
Oncology Group Statistician, 
Alliance for Clinical Trials in Oncology; 
Quantitative Health Sciences,
 Mayo Clinic College of Medicine 

Neal J. Meropol, MD
Vice president of research oncology,
Scientific and clinical lead,
Clinical research,
Flatiron Health

MEMBERS

Charles D. Blanke, MD 
Chair,
SWOG Cancer Research Network;
Professor, Knight Cancer Institute, 
Oregon Health and Sciences University 

Gary C. Doolittle, MD
Capitol Federal Masonic Professor, 
Division of Medical Oncology,
University of Kansas Medical Center 

Michael V. Knopp, MD 
Professor of radiology,
Department of Radiology,
Novartis Chair of Imaging Research, 
The Ohio State University

Seth P. Lerner, MD, FACS 
Vice chair, faculty affairs, 
Beth and Dave Swalm Chair in Urologic Oncology,
Professor, Scott Department of Urology,
Baylor College of Medicine 

Robert S. Mannel, MD
Director,
Peggy and Charles Stephenson Cancer Center,
College of Medicine,
University of Oklahoma Health Sciences Center 

Suresh S. Ramalingam, MD, FASCO
Executive director,
Winship Cancer Institute,
Roberto C. Goizueta Chair for Cancer Research,
Emory University School of Medicine

Victor M. Santana, MD
Associate director, clinical research,
Vice president, clinical trials administration,
St. Jude Children’s Research Hospital 

Julie M. Vose, MD 
Neumann M. and Mildred E. Harris Professor,
Chief, Division of Oncology/Hematology,
Department of Internal Medicine,
University of Nebraska Medical Center  

George Wilding, MD
Professor emeritus,
Director emeritus,
University of Wisconsin, Madison 

Executive Secretary

Sheila A. Prindiville, MD, MPH
Director,
Coordinating Center for Clinical Trials,
Office of the Director,
National Cancer Institute