Table 2 Additional variables, years 2009 to 2012
Description | N (%), | |
|---|---|---|
Number of scientific advice (SA) and protocol assistance (PA) letters with questions regarding adaptive phase II or phase III designs | 41 (100%) | |
Number of pivotal studies the company plans to conduct. | 1 | 33 (80%) |
2 | 5 (12%) | |
>2 | 1 (2%) | |
No information | 2 (5%) | |
Adaptive design was preplanned in advance of the conduct of the study | Yes | 40 (98%) |
No | 1 (2%) | |
The planned interim analyses will be performed in a blinded fashion | Unblinded | 33 (80%) |
Blinded | 3 (7%) | |
No information | 3 (7%) | |
Blinded and unblinded | 2 (5%) | |
Interim analyses are to be performed externally (for example, by an external data safety monitoring board, an independent statistician or a contract research organization) | No | 5 (12%) |
Yes | 25 (61%) | |
No information | 11 (27%) | |
Issues raised in the Committee for Human Medicinal Products (CHMP) answer regarding the proposed adaptive design (the issues raised relate both to designs that were conditionally accepted and those not accepted, multiple answers possible) | Adaptation strategy not sufficiently justified | 12 (29%) |
Potentially biased results | 12 (29%) | |
Too many interim analyses | 2 (5%) | |
A single pivotal trial which is adaptive is not recommended | 6 (15%) | |
Control of type I error rate | 12 (29%) | |
Other issues | 5 (12%) |