This is a common misconception. While the basic research that underpins the biological theory might happen in academic labs, turning basic research into an actual medicine that will a) work and b) not blow up your liver in the process is exclusively the province of drug companies currently. In fact, medicinal chemists (the people that actally figure out how to make the compound and scale up the manufacturing) are almost exclusively found in industrial drug development companies. They don't exist in any appreciable numbers In academia. They are analogous to the rocket scientists at NASA who worked on the moon missions that everyone worries can't be replaced because they have such specialized experiential knowledge.

Every once in a while an academic research lab will spin off a biotech to develop a new compound, but in general this is not how things have been done in the past. There is a current trend to in-license things from academia to lower the companies' exposure to early R&D risk. This ignores the fact that early research costs pale in comparison to the later clinical trials where most potential drugs go to die. Cancer is particularly bad with something like a 98% failure rate. In-licensing from academia isn't going to buy you much comfort when the thing doesn't work in a phase III clinical trial.