Like any blog owner, I check the traffic on my site. It follows the working day, with peaks during on weekday lunchtimes, in case you're wondering: I can see both the East and West coasts kicking back with sandwiches and some blog time. And I can see when some post has really revved up the readership, or brought in hordes of outside links. (Yep, "Things I Won't Work With" is the champion in that category). I can also see when a topic has failed to do either of the above.
Patent law is the champion there. That's both understandable and sort of a shame, because it is, of course, of huge importance to the drug industry. And it can also be quite interesting, once you get into it a bit. But there's no doubt that it can also make you wish that you'd listened to Momma and gone to truck-driving school like she always wanted. Many chemists just try to avoid dealing with patent questions, or rely on a few rules of thumb that they've picked up over the years, accurately or not.
Well, there's now a book that might do a lot of us a lot of good. I've been looking over The Chemist's Companion Guide to Patent Law, by Chris Miller and Mark Evans, and I think that the field has been needing something like this for a long time. It was published just last month, and one of the authors had Wiley send me a copy. I'm in the process of reading it cover to cover, and it's staying on my reference shelf.
The title is accurate; it's a top-to-bottom look at the major features of patent law as it applies to the business of chemistry. Freedom to operate, patentability (two very different concepts), claim structure, prior art, enablement, obviousness, inventorship, infringement - all the key concepts that, frankly, almost all working chemists turn out to be a bit hazy on when you get down to details. (And law, inevitably, always gets down to the details). It's illustrated with numerous examples from recent cases, structures and all, and with plenty of very realistic hypotheticals. For example:
Imagine that you are a chemist who has been laboring to find a compound that is capable of inhibiting a very important pathway in a human disease state. After many years of hard work and false leads, you find a compound - compound 4 in Figure 7.9 - that appears to possess all of the necessary attributes. However, your information scientist reports that there are prior art references that disclose a total of three different Markush structures, each of which encompasses your compound as shown in Figure 7.9. The prior art references that contain these three genera provide the general methods of making the compounds, and the preparation is enabled for one skilled in the art. A few specific examples have been made that fall within genus 1 (and hence genera 2 and 3 as well) of the prior art, but your exact compound has not been specifically disclosed. The question is now whether compound 4 is anticipated and rendered nonpatentable per se. . .
Sound familiar? For most experienced drug discovery chemists, it sure will. As the authors go on to say, a Markush that has so many variables that it can be expanded to eight wazillion compounds isn't something to really worry about - you're supposed to be able to "at once envisage" the later invention if it's going to wipe out patentability. On the other end, a direct claim of only one compound - yours - is clearly a direct hit. But what about that huge area in between? If you don't know what it means to reference the Petering case in this area, you should.
There's a lot of good stuff in this book. It's not always light reading, but it's the most readable treatment of some very complex patent issues that I've seen. Patent attorneys know everything in it (or they flippin' well should), but if you're a chemist, you probably don't. I've learned quite a few interesting things myself in the few days I've been looking it over. Every industrial chemistry department should have a copy.