Randomised clinical trials for medicines and medical devices are progressively being monitored for safety and other interim results by both Sponsors and Contract Research Organisations (CROs). There are a variety of methods to aid in proactive monitoring and gauging of patient safety and risk, including the utilisation of independent panels of expert physicians and biostatisticians, or what are known as Data Safety Monitoring Boards (DSMBs) and Data Monitoring Committees (DMCs).
DSMBs/DMCs have as their top priority the ongoing assessment of subject safety during the course of a trial where subjects are exposed to heightened risks of trial participation due to the use of experimental medicines and/or medical devices.
The importance and growing popularity of monitoring panels has been recognised, evidenced by the FDA’s guidance and recommendations to Sponsors regarding when such committees should be convened as well as a general operational framework for each member’s role and responsibilities. In order to maintain an independent unbiased role in their oversight of a trial, the physicians and biostatisticians of a DSMB/DMC may not have any fiduciary association with the Sponsor and may not be operationally involved in the trial as steering committee members, investigators or as the active study biostatistician.
A common example of when a DSMB/DMC panel may be convened surrounds central adverse event adjudication. Another area where DSMB/DMC panels play a key role is concerning trial protocols. Before a trial protocol is finalised for regulatory approvals, the DSMB/DMC may participate in a review of the trial protocol to provide feedback to the Sponsor as to operational characteristics that may be problematic and lead to subject safety and/or trial scientific integrity concerns. If there are formalised interim stopping rules defined in the trial protocol, these will be reviewed so that all parties understand the timing and mechanisms by which these decisions will be made.
Following are some of the ongoing safety and scientific integrity questions that a DSMB/DMC may need to consider:
- Are subjects being exposed to reasonable risks given the scientific research that is being conducted?
- Are there unexpected events not initially envisioned at the outset of the trial?
- Are there evolving safety signals that might suggest that subjects in the trial are being exposed to an elevated level of risk and, with an eye toward the future, is it reasonable to continue the trial given the projected risks to subjects yet to be enrolled?
- If continuing the trial is reasonable, a DSMB/DMC may make recommendations to the Sponsor to communicate safety related issues to the study steering committee, the investigators and even the FDA to raise awareness
- If the safety issues are material and the risks to subjects are no longer acceptable, the DSMB/DMC can make a recommendation to stop the study early due to these concerns
- Are the a priori assumptions that led to the power and sample size calculations in the trial protocol still valid?
- Is the observed data deviating from these assumptions, perhaps leading to the need for an interim sample size re-estimation which would most commonly be an increase in the enrolment target?
- Can the new enrolment target be achieved?
- Is there new data from an external source (e.g. a similar or clinically relevant trial) where the results that have become available suggest that it would be unethical to continue the trial being monitored?
In trials where there are formalised interim stopping rules, these must be pre-defined in the trial protocol. These stopping rules would be derived through detailed statistical methods.
- Stopping for Safety
- Predefined criteria for the incidence of serious adverse events, especially if they are deemed to be treatment emergent/related or are indicative of treatment failure. These are typically expected events, which are occurring at rates that raise questions about the risk/benefit balance of the treatment(s) under study
- Stopping for Efficacy
- Predefined criteria, which would demonstrate that the benefit observed in the treatment of interest is sufficiently large, that to continue the study to its logical conclusion would not alter or reverse this finding
- Stopping the study for efficacy on an interim basis would have stricter than typical p value thresholds for statistical testing (e.g. p ≤0.001 as opposed to p ≤0.05) to control for false positive (Type I) errors, and would account for multiple looks at the data
- Ifthe criteria are met, stopping would reduce the required sample size for the study, reducing risks to subjects not yet enrolled and address the ethical questions, for example, of continuing to treat subjects with an alternative that is now demonstrably inferior. It would allow the Sponsor to proceed to regulatory submissions earlier and at a lower cost than originally budgeted
- Stopping for Futility
- Predefined criteria that would indicate that there is a lack of a clinically meaningful benefit attributed to the treatment of interest. If the study were to continue to its logical conclusion, the probability that this finding would be altered or reversed to demonstrate a benefit is sufficiently low (typically <10%) as to raise ethical concerns to continue.
Commercial Eyes’ experienced Clinical Development team can help you navigate the data and safety monitoring requirements of your clinical trials. Call us on +61 3 9251 0777 to discuss your needs.
This article was written by Ric DeGaris, a member of our Clinical Development team.