How does something like this happen?
Drugmakers talk to the Food and Drug Administration regularly about the design of their clinical trials. Roche has a ton of experience bringing new drugs to market. And yet, the drugmaker received a "refuse to file" letter from the FDA.
Having the FDA refuse to accept an application is about the worst thing that can happen to a drugmaker. It means the agency gave the application a onceover and decided it wasn't even worthy of a review. Drugmakers sometimes get sent back to the drawing board because of missing paperwork, but in this case the FDA refused to look at the application for Roche's trastuzumab-DM1 (T-DM1) because the agency says Roche didn't test the cancer drug in the right population of patients. Roche was seeking an accelerated approval, which requires less data, but can only be used for patients who don't have any other options; the FDA didn't think that was the case for the patients who went through Roche's phase 2 trial.
I don't know where the communication broke down. Did the FDA just spring this on Roche, or has the agency been saying this all along, and Roche just refused to listen to the advice? Either way, it shouldn't have happened.
The big loser in all this is Roche's partner, ImmunoGen
While the delay in getting paid is disappointing, it's not disastrous for ImmunoGen. There's no reason to think the drug doesn't work, and the phase 3 trial needed to be completed in order for the drug to stay on the market after an accelerated approval anyway. Patient investors willing to wait for T-DM1 and for ImmunoGen's partnerships with sanofi-aventis
Run on over and check out Tim Hanson's pitch for Adidas.