With this Notice, the National Cancer Institute (NCI) announces and seeks public input on NCI’s plan to promote and ensure public availability of results from all commenced, NCI-supported clinical trials.
NCI, a component of the National Institutes of Health (NIH), is dedicated to improving the health of Americans by conducting and funding biomedical research through an extensive portfolio of clinical trials and clinical trials-related research. A fundamental premise of all NIH-funded research is that the results of such work must be shared in order to contribute to the general body of science and ultimately, to the improvement of public health. Grantee institutions are expected to make the results and accomplishments of their activities available to the research community and to the public at large.
NIH funding recipients ensure the timely disclosure of their scientists’ research findings through publications, presentations at scientific meetings as well as by sharing research tools, depositing information into databases and materials into repositories and through other means. NIH has many policies in place to educate funding recipients about their responsibility to share the results of NIH-funded work, and to facilitate such sharing. For example, the NIH Data Sharing Policy (http://www.grants.nih.gov/grants/policy/data_sharing/data_sharing_guidance.htm), the NIH Public Access Policy (http://publicaccess.nih.gov/), the NIH Research Tools Policy, and the NIH Genome Wide Association Policy (http://gwas.nih.gov/03policy2.html) are all important examples of critical information and materials sharing policies that ensure that NIH research funding is used productively and to the best advantage of science and the public health. Such sharing is fundamental to biomedical research program performance.
NCI is aware that public dissemination of the results of NCI-supported clinical trials is particularly important to cancer patients since the results of such research, more so than with most preclinical research, could directly impact their care. Unfortunately, the results and findings of a sizeable number of NCI-supported trials are not published. This occurs most frequently when a trial has negative results or is not completed. Reasons for incomplete trials can include unanticipated toxicity, poor accrual or lack of drug availability. Trials that have negative results or are incomplete are harder to publish. Understandably, medical journals often seek to attract the attention of readers and positive trials do this more readily than incomplete trials or trials with negative results. This can lead to discouragement on the part of clinical investigators who may try to publish negative results or incomplete trials but often cease their efforts to publish after repeated rejections. This selective publication of certain trials and not others is the source of publication bias which can lead to inappropriate conclusions about particular therapies. See, e.g., Nissen S, Biomarkers in Cardiovascular Medicine, The Shame of Publication Bias, JAMA Intern Med 2013 March 25; doi:10.001/jamainternmed.2013.4074; Dwan K et al., Systematic review of the empirical evidence of study publication bias and outcome reporting bias, PLoS One 2008 Aug 28;3(8):e3081. doi: 10.1371/journal.pone.0003081; Begg C, Berlin J, Publication bias and dissemination of clinical research, J Natl Cancer Inst 1989, Jan 18;81(2):107-15.
NCI believes that it is critically important for scientific information from all NCI-supported clinical trials, whether or not completed or positive, to be widely available to all health care providers, as well as to the many millions of Americans who seek credible information on clinical trials supported by NCI. Public access to such data drives scientific progress, advances patient safety, promotes health and assures optimal return on the nation’s investment in cancer trials. Indeed, according to a recent commentary in JAMA on evidence-based medicine, “science advances only when the totality of available information is shared widely within the academic community,” and clinical decision making is correspondingly hampered by the absence of findings from all available studies. See Nissen S, Biomarkers in Cardiovascular Medicine, The Shame of Publication Bias, JAMA Intern Med 2013 March 25; doi:10.001/jamainternmed.2013.4074.
While studies with positive results often have the most immediate impact on clinical practice, it is nonetheless true that negative results and incomplete studies can also be informative. For example, information about side effects can help guide other studies; unexpected negative interactions between agents or between radiation and agents can prove useful to the development of newer approaches; and understanding the reasons for incomplete trials can reduce unnecessary duplication. The current situation, where many NCI-supported clinical trials fail to be adequately communicated to the scientific community and the public, wastes precious, limited resources (including human resources, time and money) and undermines the ability of NCI to fulfill its mission to provide evidence-based approaches to cancer therapy.
For these reasons, NCI expects funded grantees, contractors and supported principal investigators conducting interventional clinical trials to find other ways to make their results publicly available when the traditional route – peer-reviewed publication – is not available.
Although certainly not the only way, one way to achieve this is through a publicly available registry or database. In its April, 2010 Consensus Report, “A National Clinical Trials System for the 21st Century: Reinvigorating the NCI Cooperative Group Program” (the IOM Report), the Institute of Medicine identified the need for and promoted the use of such a registry to list and maintain comprehensive information about clinical trials for drugs, biologics, and other therapeutic modalities. Such a registry, the IOM Report states, “could improve both physician and patient awareness of the available trials” and facilitate and improve subject involvement. NCI believes that publicly available databases could likewise be a useful and important means of communicating the results of clinical trial research to the public; particularly in cases where no peer-reviewed publication resulted from the trial. Accordingly, as more fully described below, appropriate reporting of results of an NCI-supported clinical trial in a publicly available registry will be considered satisfactory compliance with the new policy.
Request for Information on the Proposed Policy
NCI seeks comments and input concerning its draft policy through which results from NCI’s investment in clinical research will be communicated and made available to scientists and the public as outlined in this Notice. Interested parties may wish to react to the draft policy as a whole or to on one or more specific issues. Your comments to NCI can include but are not limited to the following areas:
- The extent and effectiveness of the draft policy to address an important gap in reporting results of NCI clinical trials.
- Discussion of particular benefits or drawbacks with respect to the short-term and long-term effects of the draft policy.
- Suggestions for other options for making clinical trial results publicly available other than via publication in journals and registration with public registries such as ClinicalTrials.gov.
Submitting a Response
Persons, groups and organizations interested in providing input on NCI’s results reporting policy should direct their comments to the following email address with “Comments for NOT-CA-14-005” in the subject line: PIO@ctep.nci.nih.gov.
The draft of the proposed policy now follows:
NCI Clinical Trial Access Policy – Draft
Consistent with the mission of the National Cancer Institute (“NCI”) to provide evidence-based approaches to cancer therapy, NCI believes that the full value of NCI-Supported Interventional Clinical Trials can be realized only if Final Trial Results are published as rapidly as possible. Rapid and broad access to Final Trial Results by investigators, clinicians and patients is particularly important for cancer research studies because the results of such research, more so than with most preclinical research, has the potential to directly impact patient care.
Covered Trials. “Covered Trials” means all initiated or commenced NCI-supported Interventional Clinical Trials whether extramural or intramural. Extramural trials include research grants, cooperative agreements, and contracts to conduct Interventional Clinical Trials in all phases and disciplines (e.g., treatment, prevention, supportive care, diagnosis). “Covered Trials” excludes Observational Studies and any NCI-Supported Interventional Clinical Trial in which no subjects are enrolled, but includes any NCI-Supported Interventional Clinical Trial in which at least one subject is enrolled even if the trial is not completed.
Final Trial Results. “Final Trial Results” means summary data and information about the Covered Trial including, at a minimum, data and information that characterize the study design including allocation, drop-outs, and population included in the analysis; the baseline characteristics of the population studied; pre-specified primary and secondary endpoints; adverse events; and other relevant information, including study limitations.
Interventional Clinical Trials. “Interventional Clinical Trials” means studies involving human beings (or subjects) in which the investigator assigns study subjects (randomly or not randomly) to receive a specific intervention based on the applicable protocol. Such subjects may receive diagnostic, therapeutic, behavioral, or another type of intervention and the intervention may, but need not, be investigational or involve an investigational agent (e.g., clinical trials involving surgery, radiation or screening tests). The subjects are then followed and biomedical and/or health outcomes are assessed. The term “Interventional Clinical Trials” encompasses all types of trials in all phases including pilot trials, phase zero trials, and normal (or healthy) volunteer trials.
NCI-Supported. “NCI-Supported” means all trials financially supported – whether in whole or in part – by NCI. In the case of NCI-designated Cancer Centers, the Policy does not apply to the subset of trials which, although they may benefit from core support from a Center grant, are funded privately and in which the data from the trial belong to the private funder. However, all trials at NCI-designated Cancer Centers that are funded in whole or in part with NCI funds and for which the data resides with the academic investigator are considered “NCI-Supported” and subject to this Policy.
Observational Studies. “Observational Studies” means those studies in which investigators observe and analyze the outcomes of patients receiving care in a routine setting.
Primary Completion Date. “Primary Completion Date” means the date that the final subject was examined or received an intervention for the purpose of final collection of data for the primary outcome, whether the clinical trial concluded according to the pre-specified protocol or was terminated.
Publish. To “Publish” means to report in a publicly accessible manner.
The NCI Clinical Trial Access Policy (the “Policy”) applies to all Covered Trials and, with respect to NCI-funded grants and cooperative agreements that support a Covered Trial, will be incorporated as a term and condition of the grant award.
With respect to research contracts involving a Covered Trial, submission of this information will be incorporated as a contract deliverable and failure to comply with this Policy will be cause for penalty or contract termination.
IV. Results Reporting Expectation
For every Covered Trial, Final Trial Results are expected to be Published within twelve (12) months of the Trial’s Primary Completion Date. Primary Completion Date, as defined in Section II of this Policy, refers to the date that the final subject was examined or received an intervention for the purpose of final collection of data for the primary outcome, whether the Covered Trial concluded according to the pre-specified protocol or was terminated. Accordingly, under the Policy, data from incomplete trials are expected to be reported within twelve (12) months of the date that the last subject had data collected or was examined even if the Trial does not achieve its primary aim.
To comply with the Policy, Final Trial Results are expected to be Published (i.e., reported in a publicly accessibly manner). This expectation may be satisfied in several ways including, but not limited to, by publishing trial results in a peer-reviewed scientific journal (whether published in print or on line), or through on-line registration and reporting with a publicly accessible registry dedicated to the dissemination of clinical trial information such as ClinicalTrials.gov. Recently, some journals have indicated renewed willingness to publish small, incomplete or negative trials using an abbreviated format. This would be an acceptable option under the Policy even if the peer review process for these abbreviated publications is itself less rigorous than the full length articles in the respective journals. NCI will not mandate use of a single mechanism but any mechanism used is expected to enable ready access to the Final Trial Results not only by researchers and providers but by patients as well. Public access to the results generally, and patient access specifically, are important goals of the Policy. Accordingly, if publication in a journal is the mechanism selected to satisfy the Policy, the institution must comply with the NIH Public Access Policy (http://publicaccess.nih.gov) under which "investigators funded by the NIH must submit or have submitted for them, to the National Library of Medicine's PubMed Central, an electronic version of their final peer-reviewed manuscripts upon acceptance for publication, to be made publically available no later than 12 months after the official date of publication."
V. Responsible Institutions
With respect to each Covered Trial, the grantee or contract recipient, as applicable, is responsible for ensuring reporting of Final Trial Results in accordance with this Policy. For Covered Trials involving multiple grantees, the lead institution is responsible for ensuring compliance with the Policy and for any NCI-designated Cancer Center participating in a multi-center Interventional Clinical Trial that is NCI-Supported, the lead Center is responsible for ensuring Policy compliance.
CTEP Protocol & Information Office on Behalf of Jeff Abrams, M.D.