Read on Twitter
Antibody #patents are dead. CAFC upholds JMOL finding Amgen's anti-PCSK9 antibody patents lack enablement. In doing so, CAFC focuses on "breadth" of claims, with particular focus paid to claims "functional elements." http://www.cafc.uscourts.gov/sites/default/files/opinions-orders/20-1074.OPINION.2-11-2021_1731739.pdf /1
Here's the court in its own words: /2
Amgen's defenses about "representative embodiments" fell flat. So did its defense—which is likely technically correct—that any skilled artisan could have used to teachings in the #patent to make another given antibody, if chosen. /3
Instead the uncertainty surrounding which embodiments *may* work, and which ones *won't* ran afoul of Factor 8 of the Wands factors. /4
The reality of antibody science—functional mapping, crystallography—means that a decent amount of work has to go into validating whether an antibody has its desired binding function. Perhaps doing this is routine—but what it's not is entirely predicable. /5
Unfortunately for patentees, the vast majority of antibody claims have this breadth in parallel with functional language. That's how scientists think of them! *Anti*-PCSK9; *anti*-TNFalpha, and so on. /6
Given that antibody patents are *the most valuable patents in the world* (see, e.g., the Humira patent estate, all of the other biologic drugs that are big sellers) this is a monumental decision likely to fall under the radar. /7
And since everything has a COVID-19 angle now, let me add that this applies equally to antibody therapies for #COVID! /8
Maybe the decision is doctrinally right, i.e., we shouldn't have an antibody exception. But important to note the real world impact of not having some technologically specificity in the area. 9/9
And oh, if for whatever reason you want to hear me talk about this case (with an awesomely shaggy haircut) here you go
Never actually watched myself giving this talk. Man, I'm really into it. I'm, like, method acting my own talk.
