Guidelines for Data and Safety Monitoring of Clinical Trials
The National Institute on Deafness and Other Communication Disorders (NIDCD) is establishing the following system for the appropriate oversight and monitoring of the conduct of clinical trials to ensure the safety of participants and the validity and integrity of the data for all NIDCD-supported clinical trials.
NIH policy stipulates that a system be in place for appropriate oversight and monitoring to ensure the safety of participants and the validity of the data in all NIH-sponsored or NIH-conducted clinical trials (http://grants.nih.gov/grants/guide/notice-files/not98-084.html). Investigators are required to submit a description of a monitoring plan for all clinical trials to the funding Institute or Center (IC) as part of the research application. The monitoring plan must include a general description of the mechanisms for reporting adverse events to the Institutional Review Board(s) (IRB), the Food and Drug Administration (FDA), and the National Institutes of Health (NIH). Oversight of the monitoring activity will be the joint responsibility of the investigator(s) and the NIDCD program director. The initial review group will review this plan, and any comments and concerns will be included in an administrative note in the summary statement. NIDCD staff will ensure that all concerns are resolved before any award is made. A detailed data and safety monitoring plan must be submitted to the NIDCD program director prior to an award being made. The plan should address how the investigator will implement the data and safety monitoring activities.
Responsibility for Data and Safety Monitoring
NIDCD-supported clinical trial monitoring activities should be commensurate with the nature, size, and complexity of the trial. The data and safety monitoring plan may vary from a safety officer to a committee known as a Data and Safety Monitoring Board (DSMB). This ongoing review of the data by an independent individual or DSMB assures the investigator(s) that the trial can continue without jeopardizing patient safety. These monitoring activities are distinct from the requirement for study review and approval by an Institutional Review Board (IRB).
The NIH already requires data and safety monitoring in the form of Data and Safety Monitoring Boards (DSMB) for Phase III clinical trials. For earlier trials (Phase I and Phase II), a DSMB may be appropriate if the studies have multiple clinical sites, employ particularly high-risk interventions, or involve vulnerable populations. However, other monitoring schemes for Phase I and Phase II clinical trials may be appropriate depending on the risk to participants, the population being studied, the research environment, etc.
The following provides illustrative examples of appropriate types of monitoring and oversight for different types of studies (http://grants.nih.gov/grants/guide/notice-files/not98-084.html). These examples are not exhaustive since monitoring activities should be appropriate to the study, population, research environment, and degree of risk involved.
Phase I: A typical Phase I trial of a new agent or procedure frequently involves relatively high risk to a small number of participants. The investigator and occasionally others may have the only relevant knowledge regarding the treatment because these are the first human uses. As an example, a study investigator would perform continuous monitoring of participant safety with frequent report to a safety officer and to the NIDCD staff with oversight responsibility for the trial. The investigator and the NIDCD should appoint the safety officer and agree on the frequency and contents of the monitoring report. The implementation of this approach should be part of the monitoring plan.
Phase II: A typical Phase II trial follows Phase I studies and provides more information regarding risks, benefits, and monitoring procedures. In Phase II trials, more participants are involved and the toxicity and outcomes are confounded by the disease process. Monitoring may be similar to that of a Phase I trial. However, it may be necessary to supplement this level of monitoring with individuals with expertise relevant to the study who might assist in interpreting the data to ensure patient safety or to appoint a DSMB.
Phase III: A Phase III clinical trial typically compares a new treatment or procedure to a standard treatment or procedure or to no treatment. Treatment allocations are randomly assigned, and the data are usually masked. Such trials involve several clinical sites and usually involve a large number of participants followed for longer periods of treatment exposure. In a Phase III trial, the investigators must consider the long-term effects of a study agent or procedure for safety and/or efficacy differences between the control and study groups. All Phase III clinical trials require monitoring by a DSMB (http://www.nih.gov/grants/guide/notice-files/not98-084.html). The implementation of this approach should be part of the monitoring plan. The organization, responsibilities, and operation of the DSMB are described below.
For studies co-funded with other ICs, the lead IC will be responsible for monitoring the study and establishing a DSMB if necessary. Oversight of the DSMB will be the collaborative responsibility of the lead IC and the NIDCD.
Data and Safety Monitoring Boards
The Data and Safety Monitoring Board (DSMB) serves as an advisory body to both the study investigators and the NIDCD. The DSMB is responsible for oversight of the activities related to implementing the clinical trial to ensure patient safety; conformance to the clinical protocol; overall performance of the trial components, such as the Study Chair office, Coordinating Center, and clinical sites; and integrity of the data being collected. Since each Phase III clinical trial supported by the NIDCD has gone through the NIH peer review process plus secondary review by its national advisory board, it is not the role of the DSMB to serve as a peer review group to redesign any portion of the trial, unless the DSMB feels that patient safety is compromised under the proposed design. In this case, the DSMB should communicate its concerns to the NIDCD and to the study investigators prior to the start of patient recruitment.
Responsibilities of the DSMB
- Review and approve the research protocol(s) and plans for data and safety monitoring. The DSMB, in collaboration with the study leadership, should establish specific guidelines for monitoring for safety. This should include a listing of events that should be reported immediately to the DSMB and the format of reporting cumulative data at intervals.
- Review interim analyses of outcome data and cumulative toxicity data for safety and efficacy to determine whether the trial should continue as originally designed, be changed, or be terminated. The DSMB reviews trial performance information such as patient recruitment and retention, clinical center and resource center performance, and proposals for ancillary studies. It provides advice to the investigators and the NIDCD on these topics. The DSMB also recommends whether and to whom outcome results should be released prior to the reporting of study results.
- Review the primary trial abstract(s) and manuscript(s) with regard to determining that the results are fairly presented and the conclusions appropriate.
- Review published reports of related studies submitted to it by NIDCD, the study leadership/investigators, or DSMB members to determine whether the monitored study needs to be changed or terminated.
- Review proposed modifications to the study prior to their implementation (e.g., increasing target sample size, dropping an arm based on other trial outcomes or toxicity results, modifying outcomes, etc.).
- As soon as possible but within 10 business days following each DSMB meeting, provide the study leadership and the NIDCD with DRAFT written recommendations along with justification related to continuing, changing, or terminating the trial.
- As soon as possible but within 10 business days following each DSMB meeting, provide the study leadership and the NIDCD with a statement, where appropriate, concerning the impact on the trial of individually observed or cumulative adverse events. The study leadership will provide this information to each clinical center director to be shared with their IRBs.
Responsibilities of the Study Leadership and the NIDCD Representative to the DSMB
The Study Chair and/or Coordinating Center will be responsible for providing overall management for the operation of the DSMB. Tasks include but are not limited to the following:
- Handle all logistics related to each meeting or conference call, such as booking a group of rooms at a local hotel, providing information to the DSMB, and providing timely reimbursement to each member following each meeting.
- Work with the DSMB to determine what information should be made available for review at each meeting to assist the DSMB in carrying out its primary charge related to patient protection oversight, study operation, and data integrity. This should be discussed at each meeting since different information or tables may be required from meeting to meeting.
- At least 14 days prior to each meeting, the Study Chair or Coordinating Center must provide a report to the DSMB on current status of the trial, interim analyses, adverse events, and problems encountered (see Meetings for more details).
- The Study Chair and Coordinating Center are responsible for assuring that adequate notes are taken during the open, closed, and wrap-up sessions of each meeting and during each conference call so that draft minutes from each meeting or conference call can be prepared by the Study Chair and Coordinating Center within 10 business days following the meeting, for review by the DSMB chair and the NIDCD representative. This review should be completed within 5 business days from receipt of the draft minutes. After receiving approval and/or comments from the DSMB chair and NIDCD representative, the final minutes should be prepared by the Study Chair and/or Coordinating Center and distributed to all voting and non-voting members of the DSMB. The final minutes of a meeting or conference call should be available no later than 20 business days following the end of the meeting or conference call. No minutes are required for the executive session for DSMB members only. The minutes should include general highlights of the discussions, general recommendations, action items, suggested protocol/study changes, and the rationale for each. Confidential data is not to be included in the minutes. The date for the next scheduled meeting of the DSMB should be specified at the end of the minutes.
- The NIDCD representative to the DSMB is responsible for: ensuring adequate and direct communication between the study leadership and the DSMB; providing general advice to the study leadership and DSMB related to operational issues; and keeping the NIDCD leadership apprised of the status of the trial and reviewing the recommendations presented by the DSMB so that informed decisions related to acceptance of recommendations and their potential impact can be assessed.
The NIDCD Director or designee will appoint the DSMB voting members, including the DSMB chair, based on recommendations from the study leadership and NIDCD program staff. The DSMB shall consist of a minimum of 5 voting members with expertise in biostatistics, clinical trial methodology, ethics, and specialty areas of medicine, and laboratory sciences as appropriate. Selection will be based on experience, knowledge of clinical trial methodology, participation on other DSMBs, and absence of apparent conflicts of interest including financial, propriety, or intellectual. NIDCD program staff shall serve as non-voting ex officio members. The Study Chair and representatives from the Coordinating Center may also serve as non-voting ex officio members at the discretion of the voting members of the DSMB.
Voting members of the DSMB should be independent of the trial(s) to be monitored. In exceptional circumstances, a voting member may be from an institution participating in the trial. In this situation, the member should view his or her role as representing the interest of the participants enrolled in the trial and not that of the institution.
The voting members of the DSMB will be reimbursed for expenses related to serving on the board through funds provided by the NIDCD to either the Study Chair's cooperative agreement or the Coordinating Center's cooperative agreement. These funds will be used to reimburse each member for travel expenses (coach fare only), hotel costs, per diem at US Government rate for the city where the meeting is held, and a daily participation fee at the rate allowed by the US Government for each day in attendance at the meeting.
The frequency of DSMB meetings will depend on the nature of the trial. A DSMB should meet at least annually. Each meeting may be divided into four parts.
- An open session at which members of the clinical trial team, representatives from the NIDCD, and others may be present, at the request of the DSMB, to review the conduct of the trial and to answer questions from members of the DSMB. The focus in the open session should be on accrual, protocol compliance, and general toxicity issues. Outcome results by treatment group should not be discussed during this session.
- A second, closed session involving the voting and appropriate non-voting DSMB members should be held. The outcome results and toxicity information, if any, will be presented to the DSMB by treatment group.
- The third part of the meeting may be an executive session involving only voting DSMB members to allow them the opportunity to discuss the general conduct of the trial and all outcome results, including toxicities and adverse events; develop recommendations; and take votes as necessary. At the request of the DSMB, some non-voting members may attend all or portions of the executive session to respond to questions or to clarify issues. Votes may be done by voice/show of hands or by ballot. A majority vote is required to carry any recommendation.
- A final wrap-up session is held, at which the DSMB Chair, with input from other DSMB members, presents and discusses the committee's recommendations, along with the rationale/justification for each, with the study leadership and the NIDCD representative.
Interim meetings and/or conference calls may be held at the request of DSMB members, the study leadership, or the NIDCD. If the NIDCD or the study leadership requests that the DSMB vote on issues using a ballot, the ballots will be returned to the NIDCD program officer managing the trial. Only the aggregate vote on each issue will be reported to the study leadership and the DSMB by the NIDCD program officer. A majority vote is required to carry any issue.
A written report containing at least the current status of the trial, performance and data quality, interim outcome data, and any toxicity data should be sent to DSMB members by the Study Chair or Coordinating Center at least two weeks prior to the meeting to allow sufficient time for the DSMB members to review the report. This report should address any specific concerns about the conduct of the trial. The report may contain recommendations for consideration by the DSMB concerning clinical center performance, whether to continue accrual and/or follow up, whether to close the trial, and whether the results should be reported.
Recommendations from the DSMB
DSMB recommendations should be based on results from the trial(s) being monitored as well as on published data from other studies. It is the responsibility of the Coordinating Center, Study Chair, trial investigators, NIDCD staff, and individual DSMB members to ensure that the DSMB is kept apprised of non-confidential outcome results from other related studies as they become available. It is the responsibility of NIDCD staff to keep the DSMB apprised of any programmatic concerns related to the trial being monitored.
DSMB recommendations are advice provided to the NIDCD and to the study investigators. DSMB recommendations will be provided to the study leadership and to the NIDCD as soon as possible but no longer than 5 business days following the DSMB meeting. No action on the recommendations is to be taken by the study leadership until the recommendations have been reviewed and accepted by the NIDCD. The NIDCD representative to the DSMB will notify the study leadership and the DSMB of NIDCD's decision(s) on the recommendations.
DSMB recommendations that a study change for patient safety or efficacy reasons, or that a study be closed early due to slow accrual or other reasons should be communicated to the study leadership and the NIDCD immediately following the DSMB meeting. It is the responsibility of the Study Chair and Coordinating Center, in collaboration with the NIDCD, to review the recommendation(s) and to implement the change(s) as expeditiously as possible. If the study leadership and/or NIDCD do not concur with the DSMB's recommendation, it will be the responsibility of the NIDCD Director or designee, study leadership, and DSMB Chair to reach a mutually acceptable decision about the trial. If such an agreement can not be reached, final authority will rest with the NIDCD. Confidentiality must be maintained during these discussions.
If a recommendation is made to change a protocol for reasons other than patient safety or efficacy or for slow accrual, the DSMB will provide an adequate rationale for its decision. In the absence of disagreement, the Coordinating Center will be responsible for amending the protocol and notifying the clinical centers as expeditiously as possible. It will be the responsibility of the local clinical center directors to notify their local IRBs of any protocol changes.
Release of Outcome Data
Confidential outcome data should not be made available to individuals outside of the DSMB. Any release of outcome data to individuals outside of the DSMB must be reviewed and approved by the DSMB, the NIDCD, and the study leadership.
No communication, either written or oral, of the deliberations or recommendations of the DSMB will be made outside of the DSMB except as provided for in these guidelines. Outcome results are strictly confidential and must not be divulged to any non-member of the DSMB. Each member of the DSMB, including non-voting members, must sign a statement of confidentiality.
Conflict of Interest
Individuals invited to serve on the DSMB, as either a voting or non-voting member, will disclose any potential conflicts of interest, whether real or perceived, to the NIDCD program director on an annual basis. The Study Chair or Coordinating Center will provide each member with a form to be completed and returned to the NIDCD program director. Conflict of interest can include financial interest, professional interest (in the sense of the trial outcome benefiting the individual professionally), proprietary interest, and miscellaneous interest as described in the NIH Grants Policy and 45 CFR Part 94. NIDCD and the study leadership will make collaborative decisions regarding service by individuals with potential conflicts of interest or the appearance of conflicts of interest.