The Model-Based Approach: A Better Way To Forecast Enrollment

The clinical phases of drug development represent the eagerly awaited period where, after several years of research and development, promising treatments become available for volunteer patients. The stakes at this stage are high: While less than two-thirds of Phase 3 trials are successful, they represent the most costly and time-consuming period of drug development. According to a paper published in the Journal of the American Medical Association, the most common cause for incomplete Phase 3 trials is related to enrollment. This is not surprising when as many as 37% of trial sites miss discontinuity enrollment targets, and 11% fail to enroll a single patient.

A vast majority of clinical trials experience delays in enrollment, and in the worst cases, these delays lead to discontinued trials. Even for successful trials, however, enrollment delays impose a substantial cost on the sponsor by increasing clinical operations expenses and loss of revenue due to delayed submissions. On the other hand, over-enrolling patients can also waste resources. Thus, optimal planning of patient enrollment – a key component of trial success - remains a difficult balancing act.

One of the critical questions facing trial planners is: “Can we recruit the required number of patients (e.g., 1000) within the assigned time (e.g., 25 months)?” A trial planner’s experience and judgment is certainly useful for addressing this question. However, as demonstrated below, intuitive judgments alone may be ill-suited to manage all the complex uncertainties and interdependencies between relevant variables.