Like many industrial scientists, I’ve been dealing with the patent literature for so long that I’m used to its (many) idiosyncrasies. There are large sections of any patent that I just page through as rapidly as possible because they are utterly not worth reading. The parts where the various dosage forms and potential dosing combinations are recited, for example, is the densest boilerplate – you wonder when the last time was that this stuff wasn’t just copy-and-pasted from another application before being lightly edited to fit the current one. The figures are most certainly worth study, the claims, of course, to see what’s actually being claimed, and then I work my way back on any specific parts of those to see how well those claims are actually exemplified. That’s especially important in dealing with a patent application, and most of the time I’m looking at applications (since I’m keeping up with the latest news available from 18 months ago). You can put all kinds of stuff into your application; what comes out the other end as a granted patent is often another matter entirely (or should be!)
This article in Science points out something important to keep in mind while you’re doing that, though. It’s about “prophetic examples”, which are especially common in US applications and patents (and in filings in other jurisdictions that originated with a US application’s text). It’s the author’s impression, and mine as well, that many people don’t catch on to these. A prophetic example is an experiment that has not been carried out yet. It’ll probably do what it says (that’s why it’s in there, to show that the inventors have anticipated that this might be another use of the invention), but it hasn’t been actually accomplished yet. That goes for procedures, actual chemical compounds, whatever.
How do you know that you’re reading a prophetic example and not a real experiment with real data? The first key is the verb tense. Things that have actually been done are in the past tense. This was added, this was isolated. If it’s in the present tense (“is added”) or future tense (“will be added”), you are not reading about something specific that has actually been performed yet. Now, you’ll see those tenses used (along with the conditional) when describing various aspects of the invention or general procedures (“Compounds of the present invention may be prepared by condensation of an appropriate aldehyde XVI. . .“), but when you see them applied to something specific, it’s prophetic. Even if it’s an apparent procedure with amounts of reagents, etc.
The second tip-off is the lack of real data. For chemical compounds, that should be pretty obvious: there’s no such thing as a prophetic NMR or melting point. What you will see, though, are compound tables full of structures (sometimes page after page of them), but the numbers attached to the compounds are only things you can calculate, like molecular weight or cLogP. If there are no “hard” numbers associated with such compounds anywhere in the patent (but only stuff like that) you’re dealing with a bunch of “Yeah, we thought of these too” prophetic examples, not things that have actually been made. If they had had actual physical data, they’d have put the numbers in there to make the application stronger.
Could you have a successful patent application with only prophetic examples? Interestingly, the answer is yes (see 2164.02 of the Manual of Patent Examining Procedure). But it had better be good. A 1987 court decision stated that “The mere fact that something has not previously been done clearly is not, in itself, a sufficient basis for rejecting all applications purporting to disclose how to do it“, and if the application discloses things well enough that (patent language) someone “skilled in the art” would be able to carry out the invention without “undue experimentation”, then fine. But if you’re claiming chemical matter (as in a drug patent), you’d be well advised to have actually made some compounds before filing that application.
A problem with this system, as the article notes, is that there are people who don’t understand it and who have cited such prophetic examples as if they were real results. I like the proposed solution: keep prophetic examples, sure, but label them forthrightly as “Predicted Results” or “Hypothetical Compounds”. The USPTO already requires that such things be labeled by avoiding the past tense, so this would just be a clearer label and not some totally new requirement. The only benefit (let’s put that in quotes) that I can see of the present labeling system is that it causes confusion about what’s real and what’s hypothetical, which can occasionally be of advantage to those filing the patents. That, we can lose.