The following questions suggest ways for a Data and Safety Monitoring Board (DSMB) to view a protocol from the perspective of improving the quality of monitoring and implementation.
The first three questions relate to the primary outcome measure, with intent to clarify how close to an actual clinical benefit or estimate of risk of harm the protocol outcome measures will be. The greater the precision of the measurements, the more meaningful a determination of acceptable risk (for likely benefit) becomes.
- Does the primary objective and proposed endpoint, also known as the outcome measure, describe a direct clinical benefit or risk of harm?
- A biomarker is an endpoint that is postulated but not formally established to represent a clinical outcome for either benefit or risk of harm. If the primary outcome measure is not a direct measure of clinical benefit or risk of harm, does the protocol document that the outcome measure is qualified as a biomarker either by the Food and Drug Administration (FDA) or by literature citations? What is the biomarker intended to represent?
A surrogate marker is an endpoint that has been formally established to represent either a clinical benefit or risk of harm. If the primary outcome measure is a surrogate marker, does the protocol document what the outcome measure is a surrogate for? Also, does the protocol document the precision of the correlation between the surrogate marker and actual clinical benefit or risk of harm? Does the surrogate marker have both positive and negative correlation with the intended clinical outcome?
The next two questions relate to the clarity of the protocol text and the appropriateness of the implementation plan with regard to screening participants and reacting to and managing adverse events.
- Are the protocol participant eligibility criteria, also known as the inclusion and exclusion criteria, clearly defined and appropriate for the study population? Do the eligibility criteria also include adequate safety parameters such as taking into account the safety profile of the intervention and the age or phenotypic characteristics of the participants, including pregnancy status for women?
Are the clinical and adverse event management plans clear and adequate to ensure study participant safety? If applicable, are the individual participants’ escape criteria clear and adequate to ensure participant safety? If applicable, are overall study stopping criteria clear and adequate to ensure study participant safety?
The next question addresses one of the most common reasons for failing to complete a clinical study within the allocated time and resources.
What is the target enrollment? Are risks that could affect enrollment rate and enrollment total included in the protocol, along with proposed mitigation plans? Are there additional risks (not otherwise described in the protocol) to meeting accrual targets that should be addressed?
The next two questions focus on data quality and monitoring prior to study completion, while there is still the opportunity to make adjustments.
- Does the protocol identify the operational, infrastructural, or procedural study risks with regard to delivering a high-quality, well-characterized analytic data set? If not, what do you recommend should be included?
Is the statistical analytic plan sufficiently described such that progress toward capturing enough high-quality data to implement the plan can be assessed?
The last two questions relate to the overall monitoring and reporting schedule, for which annual DSMB review is a minimum requirement. The DSMB in a high-risk study may wish to meet and discuss the study team reports at more frequent intervals, such as every six months or even quarterly.
- Given the overall and specific risks, did the study team indicate in the protocol or supporting documents the type of monitoring for participant safety and data integrity they plan to implement and the means for implementation? Is the monitoring plan sufficient for the anticipated study risks? Do you recommend any additional monitoring that could improve the study implementation or diminish the risks?
- How frequently do you recommend the study team provide structured updates and reports to the DSMB? How frequently should the DSMB discuss the study, with options to adjust the discussion schedule based on progress and events?