Biotech Values

[DewDiligence]

Re: AD clinical_trials_in_a_post-Adu-approval_environment (cont'd)

I would favor a staggered start design. Lets say there are 1000 patients in a 2 year study. 250 start drug at entry and are on drug x 24 months in blinded fashion, 250 are placebo x 6 months then start treatment for 18 months, 250 are placebo x 12 months then drug x 12 months and 250 on placebo x 18 months then drug x 6 months. This allows true blinded placebo comparison x 18 months and also allows determination whether the drug is just a treatment or is disease modifying (if just a treatment the delayed start group will be about the same at 24 months as the initial group; if disease modifying, the delayed group does not catch up). It also allows earlier treatment for many patients. Of course, all should be offered at least another year or more of OLE.

Clever indeed, but I would advise any company not to employ such a complex design for a registrational trial because these kinds of trials tend to produce results that are hard for regulators to interpret. BMY’s CHECKMATE-227 study is a case in point.