Table 1.
Comparison of 3 Phase III trial design options with estimated duration.
| Option | Duration |
Pros and cons |
||||
|---|---|---|---|---|---|---|
| If stopped |
If continued |
|||||
| FPI ∼ LPI (Enrollment duration) | FPI ∼ Trial end | FPI ∼ LPI (Enrollment duration) | FPI ∼ NDA | Pros | Cons | |
| Option 1: Fixed design with a total sample size of 225 patients for 3 arms | 3 months | 9 months | 18 months | 35 months | ●Statistical power: high. Ideal statistical power (>90%) for a fixed design trial based on an assumed SD of 10. ●Trial implementation/Regulatory risk: low. |
●Enrollment risk: high. 225 patients considered to be the maximum feasible sample size, imposing the highest risk of not achieving the enrollment target. ●Development time: unfavorable. Enrollment period and trial duration almost one year longer than our desired. |
| Option 2: Fixed design with a total sample size of 150 patients for 3 arms | 3 months | 9 months | 11 months | 26 months | ●Enrollment risk: low. ●Development time: favorable. The shortest duration, fitting desired development time frame. ●Trial implementation/Regulatory risk: low. |
●Statistical power: low. Statistical power lower than 80% based on an assumed SD of 10, leading to the highest risk in trial success. |
| Option 3: Adaptive statistical design with a minimum sample size of 150 patients and a maximum sample size of 225 patients for three arms. | 3 months | 9 months | 11–18 months | 26–35 months | ●Statistical power: varies between those for Options 1 and 2. Avoids drawbacks of Options 1 and 2 with an opportunity to make mid-trial adjustments to accommodate uncertainty in the SD up to 10. ●Enrollment risk: varies between those for Options 1 and 2. ●Development time: varies between those for Options 1 and 2. |
●Trial implementation/Regulatory risk: high. More complex design compared to Options 1 and 2; lack of experience with implementing an adaptive design trial. Requires intensive upfront planning and special measures to prevent operational bias. |
FPI: First patient in; LPI: last patient in; NDA: new drug application.
Durations are rough estimates at the planning stage.