I see that Against Intellectual Monopoly by Michele Boldrin and David Levine is about to be released. This is a provocative work, parts of which have appeared in articles over the last couple of years. The first few chapters are available online (still with some typos, I've noticed).
What makes this a bomb-throwing sort of book, from the perspective of a drug company researcher like me, is that the authors recommend abolishing the patent system as we know it. They have a chapter (PDF) on the drug industry specifically, as you'd expect. Looking it over, I find it a peculiar mix of compelling argument and things that make me hold my head and moan. For instance, the authors, in all their discussion of the German and Italian drug industries, don't seem to take into account the way these industries have been able to make money over here.
I'm open to the idea that there might be other ways to encourage innovation than the patent system we have now. But Boldrin and Levine seem to have bought into the whole it's-the-government-that-finds-drugs idea, which never fails to drive me insane. (See this category for more on the same topic here).
Here's just one example of a hold-my-head section:
Much of the case for drug patents rests on the high cost of bringing drugs to market. Most studies have been sponsored by the pharmaceutical industry and are so quite suspect. The Consumer Project on Technology examined the cost of clinical trials for orphan drugs – good data are available for these drugs because they are eligible for special government benefits. A pharmaceutical industry sponsored study estimated the average cost of clinical trials for a drug at about $24.5 million 1995 dollars. However, for orphan drugs where better data are available, the average cost of clinical trials was only about $6.5 million 1995 dollars – yet there is no reason to believe that these clinical trials are in any way atypical.
There aren't? How about the fact that they're often first-in-class therapies, and thus can have a lower efficacy standard (better than nothing!) to meet? Or the fact that they're orphan drugs, meaning that there's a limited patient population available to start with? There are cardiovascular drugs that have been through trials in more patients than the total market size of some orphan therapies. No, there are some deficiencies here.
For a worried take on Boldrin and Levine's work, see Arnold Kling, and for its application to the drug industry, this post at Samizdata. Here's some more discussion on the same topic.