Complex Clinical Trial Protocol Designs – Part II
Impact on the Data Monitoring Committee (DMC)
Impact on the Data Monitoring Committee (DMC)
Complex clinical trials can accelerate drug development, offering patients timely access to transformative therapies. At the same time, these studies can present challenges for trial management. Part I of this blog series explored data around complex studies plus various complex trial designs. Then, Part II addressed the site experience and the importance of having a central institutional review board (IRB). In this segment, we look at the impact of complex clinical trials on the data monitoring committee (DMC).
Agencies want to ensure that the correct decisions are made regarding patient safety, independent of study sponsors. The more complicated the study, the more critical it is to have an independent statistician who understands complex study protocols and can act as a liaison with the DMC. Why? Because the sponsor’s statistician should avoid direct interface with the DMC to preclude bias.
There are three main considerations regarding complex clinical trials and DMCs:
In a standard study, the DMC is focused on typical patient safety outcomes – looking at adverse events (AEs), straightforward tables, and perhaps a basic efficacy summary to assess risk versus benefit. A complex trial protocol may require the DMC to review an interim analysis for an early look at the efficacy or futility of a primary endpoint. Therefore, for complex studies, it is essential to have an independent statistician who understands efficacy plus safety.
Further, a sponsor may want to combine the DMC meeting for a futility analysis with an already scheduled safety DMC, requiring extra effort for scheduling between the sponsor and DMC. Complex studies may require multiple looks at study data – perhaps involving sample size re-estimation, which introduces more complicated statistics and requires a higher level of expertise.
A major pharma client asked WCG for DMC support on a lung cancer treatment study. As part of independent reporting, we were responsible for producing interim analyses (IA), which had the potential to stop the study early for overwhelming benefit or futility. Typically, when an IA meets the statistical criterion for declaring superiority, the DMC communicates the result to the sponsor, whose study team unblinds itself to create and submit the application package to the FDA. This process can take 4-5 months, with the FDA then requiring 6-12 months to complete the review.
Since the drug showed superior survival, the FDA felt compelled to ensure rapid patient access. The agency requested WCG’s closed DMC report and derived analysis datasets to support the initial efficacy labeling. The FDA completed its review in only 2.5 months – more than 3 months ahead of the PDUFA date. WCG’s work in generating a clear, succinct, and interpretable DMC report was critical to the speed with which stakeholders could mobilize, understand the results, and make this new therapy available to patients.
There may be an AE of particular interest, which requires complicated analyses by an experienced independent statistician. Also, if the DMC reviews study data regularly (e.g., every six months), they may identify an unexpected safety signal. In this case, the DMC may request additional information or analyses from the independent statistical group, as they cannot confer with the sponsor.
Ensure that you have an experienced independent statistician who can provide your DMC with the necessary information without going to the sponsor. Remember, the objective of the DMC is to safeguard patient safety while reducing bias and maintaining the study blind.
Best practices for DMCs include regular transfers of all data sources from the sponsor/data vendor to the independent statistical group (ISG). This process enables the ISG to respond to ad hoc DMC requests without going back to the sponsor/data vendor for an off-cycle transfer, which could alert the sponsor that the DMC is noticing something unusual. The DMC may not be concerned to the point of raising the issue with the sponsor. If periodic transfers are arranged at the study onset, the ISG will have continual access to up-to-date data.
Both timing and content are essential; periodic transfers should be as complete as possible. The DMC may request additional analyses for data domains not originally planned for inclusion in its reports. For example, the DMC for an Alzheimer’s trial may be concerned about an emerging safety signal related to ocular events. If the ISG receives ophthalmology data on an ongoing basis, it can produce the additional analyses requested by the DMC, even if the DMC did not plan to review ophthalmology data at study onset.
Real-world considerations: How often is the sponsor/data vendor realistically able to transfer data? Do they have the mechanisms or resources to support such transfers? A WCG independent statistician experienced with DMCs will be aware of these considerations and prepared to discuss them with the sponsor at study onset – before patients are treated and before a potential safety signal arises.
Rather than having one DMC per study, we see sponsors implementing a portfolio-level or therapeutic-area DMC. In these cases, single umbrella DMCs may have oversight of multiple protocols. For example, a compound may have a specific mode of action, but the sponsor applies it across therapeutic areas. The sponsor will want an all-encompassing DMC to review the data if a specific safety concern arises due to the compound’s mode of action.
Does it make sense to combine these protocols into an integrated analysis for the DMC? If a safety signal is rare, it may be helpful to pool data from all protocols for a unified view. What about timing? Multiple protocols are likely enrolling at different speeds, so the logistics of scheduling the DMC and organizing the data may be challenging. Engage an independent statistician who is agile in pulling together relevant, timely data for your DMC.
WCG began working with a biotech client in the early stages of its development program. While Phase I DMCs are not common, this was a novel cellular therapy, so the client did use a DMC in Phase I. We worked with the biotech to expand the DMC’s purview as the sponsor grew its program. When the program became too large and the burden too heavy for a single committee, we advised the client on how to transition from a single DMC to multiple DMCs, each responsible for overseeing 2-5 studies. In doing so, the individual DMCs could perform a better, more focused assessment of safety.
Consider these real-world perspectives on using a portfolio-level DMC.
As a reminder, essential DMC qualities include:
A poorly managed DMC may miss safety and efficacy signals, putting patients at risk and leading to costly delays.
As you can see, complexity can be a significant factor in engaging DMC statistical support. WCG is a strong, trusted partner for DMC services , offering the expertise, experience, and depth of services needed to boost DMC quality and efficiency. Our statisticians focus on DMC work across therapeutic areas, with the majority of our full-service DMC projects contracted to support complex clinical trials. Benefits include reduced burden on the study team, clearly defined study outcomes, and an independent voice to protect patient safety.
Serving DMCs for complex clinical trials:
Serving DMCs across therapeutic areas: