The Federal Circuit today released an unusual opinion in Abbott Labs v. Sandoz, which I noticed after seeing this post on patentlyo late this evening. It was not on the CAFC website mid-day today (Pacific time), which is when I checked the site and saw only Autogenomics, Inc. v. Oxford Gene Tech., Inc., and Dennis Crouch didn’t comment on the case until almost 10pm (Pacific). I assume the case was released at the end of the day on the East Coast.

This timing information wouldn’t be of the least interest if it were not for the fact that it is such an unusual opinion. It features an en banc section on product-by-process claims which makes a sweeping bright line rule, overturning a line of precedents on an issue that was not even addressed by the parties to the case. And, as carefully pointed out by Judge Newman in her excellent dissent (joined by Judges Mayer and Lourie), the en banc portion of the opinion departs from normal en banc practice in significant ways. There was no public notice of the decision to take a portion of the case to the full Federal Circuit, and no opportunity for amicus briefs to be filed to provide the Court with input from the community of patent holders, practitioners and reformers. Normally, when a case is taken en banc, it is to focus, as here, on an important question of patent law. In this case, not only was no comment sought from the interested patent community, but there were no oral arguments and apparently even the parties to the case were unaware that an en banc opinion was being prepared. So all in all, the whole affair has the feel of “guerilla judging”, almost as if the Federal Circuit intentionally has pulled a fast one.

While I personally have not been involved in biotech cases, it seems that the import of the opinion is likely to be immense. The opinion flatly overturns the “rule of necessity” which has been used for over a century. This rule allows applicants to claim new products using product-by-process claims, and to claims them as products, when there is no other way to properly describe what is invented; this is common with complex chemical and biotech inventions when detailed structural information is not yet available at the time of filing (as indeed is the case in the patent in suit, which is for a pharmaceutical known as cefdinir, which has a complex structure). The rule only applies when the product so claimed is itself new; in other cases (including several cited by the majority opinion), product-by-process claims are used to claim a known product produced by a new process, and are therefore treated as process claims.

The importance of this rule has been that, when it is invoked, an accused product produced by a different process still infringes the patent. On the other hand, in the second case of a new process used to make a known product, an accused product produced by a different process is clearly not infringing, since the final product in these cases was already known, and it was only the new process which was in fact claimed (and hence the product-as-process claims were in effect process claims). With this ruling, however, the majority summarily executed the product-by-process claim as a product claim. According to this new rule, whenever a process step is recited as an element of a product claim, then only products produced using at least the cited process steps will be held to infringe.

Again, as I am not a biotech specialist, I will only speculate, but it seems likely that a large number of valuable patents have been weakened or basically gutted by this ruling, which makes the unusual way in which it was reached and promulgated all the more surprising.

A further very interesting and disturbing aspect of this ruling, noted by both Judge Newman and Dennis Crouch, is that now, for the first time, the principles of claim construction for validity are different than for infringement. As pointed out at length in the dissent, there is a long line of cases clearly stating that it is a bedrock principle of patent law that that which would infringe if invented after a critical date will anticipate if invented before the date, or in other words that claim construction is the same for validity determinations and infringement determinations. Now, for product-as-process claims, this is no longer true (unless later cases make it true as an assumption, and then pursue a complex path to reconciliation along the lines, suggested by Dennis Crouch, that product-by-process claims to a known product using a new process are not anticipated by the existing product).

All in all, this “guerilla judgment” seems to be the harbinger of more confusion and debate, much as the Bilski case has, and it will likely unsettle a number of highly valuable property rights.