When is a DSMB Needed?

It may seem that every clinical trial should have a DSMB, but is that correct?

While there are no strict requirements on when to establish a DSMB, the FDA and the EMA guidelines generally agree on several recommendations for when to create a DSMB, including:

Any controlled clinical trial examining mortality or major morbidity.
Randomized trials with vulnerable populations.
Multi-center clinical trials evaluating treatments to prolong life or reduce the risk of major adverse health outcomes.

While both outline similar recommendations, the FDA guidelines provide a slightly more detailed outline with additional specific recommendations.

Recommendation criteria for DSMB creation

	FDA⁷	EMA⁸
Study endpoints	The study endpoint is such that a highly favorable or unfavorable result, or an obvious finding of futility, at an interim analysis might ethically require termination of the study before its planned completion	The study endpoint(s) should be taken into consideration in the context of preplanned interim analyses for early stopping (either for futility or for positive efficacy) or in case of complex study designs where the sponsor intends a possible modification of the study design based on unblinded interim data
Safety and toxicity	There are a priori reasons for a particular safety concern – for example, if the procedure for administering the treatment is particularly invasive There is prior information suggesting the possibility of serious toxicity with the study treatment	Prior knowledge or strong suspicion that a treatment under consideration has the potential to harm patients (even though being eventually more effective than other treatments already available)
Vulnerable population	The study involves a potentially fragile population such as children, pregnant women, very elderly people, or other vulnerable populations, such as those who are terminally ill or of diminished mental capacity	The clinical trial involves a pediatric population, even in a non-critical indication; similar considerations apply to clinical trials in mentally disabled patients
Life-threatening disease	The study is being performed in a population at elevated risk of death or other serious outcomes, even when the study objective addresses a lesser endpoint	Usually recommends the implementation of a DMC in cases of life-threatening diseases from an ethical point of view

The key point is that neither the FDA nor EMA requires a DSMB, but they recommend it in specific kinds of trials. The only exceptions are trials in critically ill subjects where obtaining the Informed Consent Form is impossible. While there are no strict requirements on when to establish a DSMB, the number of clinical trials that use them is steadily rising. In 1990, only 10% of trials published in medical journals reported using DSMBs, compared to 25% in 2000.³ In addition, 35.2% of multi-center industry-funded and 83% of multi-center National Institutes of Health-funded clinical trials registered in 2007-2010 used DSMBs.⁹

Additionally, while DSMBs are not typically considered for early-stage (Phase 1 and 2) studies, many medications have received expedited approval from the FDA based on the results from such early-phase trials. From 2011 to 2020, 410 new drugs were approved by the FDA, with a steady growth of 6.5% per annum in the number of approvals in the US.¹⁰ An independent DSMB is worth considering for such trials as it is not uncommon for confirmatory Phase 4 clinical trials to fail to demonstrate clinical benefit after receiving accelerated approval. For instance, sponsors voluntarily withdrew 20 accelerated approval indications between December 2020 and January 2023.¹¹

DSMBs can provide benefits regardless of study size or phase, and the study sponsor must ultimately decide whether to establish a DSMB in a clinical trial. During our bidding process, PSI provides a thorough and detailed analysis of the study protocol to determine whether a DSMB will benefit the study and the patients, considering factors such as:

Overall study risk.
The intervention’s complexity.
The study population's vulnerability.

Next up: Considerations for Establishing a DSMB