Data monitoring committees (DMCs) for clinical trials have been used for many years and are vital tools for ensuring that patient safety is carefully reviewed in an unbiased manner. Historically, these committees were most often utilized for larger and relatively long-term Phase III trials with mortality or serious morbidity primary endpoints. There is broad consensus that independent DMC review by experts with no bias or conflict of interest with the sponsor company is critical to ensure patient safety during the conduct of these pivotal trials. More recently, despite the increased complexity of adding DMC oversight to trials, there has been an increase in their use. This expansion has primarily occurred into earlier stage trials that may be supportive of a registration. These non-pivotal trials may include adaptive design elements or protocol defined interim analyses allowing flexibility in the data gathering process.
Monitoring committees for these types of trials usually consist of either expert researchers completely independent from the company, or internal sponsor employees who are not in communication with study investigators and are separated from the clinical study team conducting the trial. Often the purview of committees for these types of trials is not only safety monitoring but also an evaluation of efficacy. Modifications to trial design as allowed by the protocol, such as adding dosage groups, adding patients to an existing dosage group, changing dose duration, and modifying inclusion/exclusion criteria may be included in the committee’s charter. The committee may recommend or decide to stop the trial based on the data review. Even in this new paradigm not all trials should have a DMC, many are too short in duration or enroll too few patients to allow efficient use of an independent review committee. Once the decision to have a DMC has been made, however, should there be more consideration in using a committee composed of both internal and external experts? In circumstances where the DMC is being put in place to only monitor safety independent of the study team it is probably more expedient to decide between a committee of internal experts, or form an external committee before establishing the appropriate oversight process. In adaptive trials, however, where the DMC is charged with making decisions or recommendations on future trial conduct then consideration of a mixed committee is reasonable.
Internal experts for DMCs are often selected because they are independent from the study team and bring deep drug development and global regulatory experience to the committee as well as therapeutic expertise. Since these experts are internal they possess a detailed knowledge of the company’s decision-making criteria. They have full access to data on the development candidate from prior or current work that is being conducted outside of the trial being monitored, and an ability to communicate with internal departments to prioritize rapid data analyses that maybeneeded for decision-making. Independent non-sponsor associated experts are often chosen because of their knowledge of the disease state, experience in treating the targeted patients, and their ability to act as patient advocates. Many of these experts also have extensive experience with regulatory agencies in connection with the disease under study. This blend of different perspectives and abilities may be highly valuable in adaptive trial situations where speed and efficiency are essential for making rapid decisions. This is especially the case in earlier stage trials where incorrect decisions can have severe consequences on the development program.
During the conduct of an adaptive trial it is critical that decisions are made after a rigorous discussion of the data and its impact on the future course of the study. This type of discussion is most productive when as many different perspectives as possible are included. By utilizing a mixed committee the blinded data can be reviewed with many of the consequences for future development being considered from a patient, treating physician, regulatory, and sponsor perspective. This type of just-in-time discussion is not possible when only internal or external experts are able to review blinded data and the course of the trial is decided. When this type of review during a trial does not occur, changes in the direction of the clinical development program for future studies can be made later after the trial is completed but this may cause a loss of valuable time and effort. Confidence that the committee has fully evaluated the severity of adverse events, the extent of efficacy, and the contribution of various inclusion/exclusion criteria can be increased when a mixed committee is reviewing the data. These considerations become even more important when the patient population being studied is relatively small, or is poorly studied, as in the case of many orphan diseases. In these instances, data from every patient is extremely valuable and decisions on the future conduct of the study require expertise that can only be provided when internal and external experts are both participating in the data review.
Once the decision has been made to utilize a mixed committee, there are several logistical considerations crucial for success. The charter for the committee must be very explicit in determining the decision-making process. Is the committee making the final decision on the trial’s conduct or does it make a recommendation to the sponsor? Does the committee decide by consensus or by vote? If it is by vote, do all members have full voting rights? The composition of the committee and the selection of the committee chair are always carefully considered for a DMC but this decision is even more critical when the committee is mixed in composition. In many circumstances, if the effort to include outside expertise is desired it is reasonable to have one of the external members chair the committee. The size of the committee should remain reasonably small, but is probably best served when there is a balance of internal and external experts.
There are many different strategies for designing clinical development programs, but the ultimate goal in early stage trials should always be to determine as quickly as possible whether or not a program should be continued or halted. If the program is to be continued, what can be done to learn as much about the new potential therapy in the most efficient manner? Given the tremendous effort and cost of bringing a development candidate into the clinic, it is crucial decisions are made rapidly but with the best information available, and with as many perspectives included as possible. An adaptive trial design allows us to learn from the data as it is being collected, but it also leads to the potential of making mistakes due to insufficient analysis. Given the criticality of making rapid and correct decisions during the conduct of a trial, as well as the need to maintain the trial’s integrity, a mixed DMC is a tool that can be utilized to maximize the probability of success.