Product-by-Process Patent Must be Directed to New Product

SmithKline Beecham v. Apotex (Fed. Cir. 2006, 04–1522).

SmithKline’s U.S. Patent No. 6,113,944 is directed to paroxetine (Paxil) made through an allegedly novel process (product-by-process). In 1998, Apotex filed its ANDA and paragraph IV certification asserting that the patent was invalid — thus provoking action from SmithKline.

On summary judgment, the district court found the patent invalid as anticipated by prior knowledge of paroxetine. Specifically, the lower court determined that the claimed process steps were not limiting.  Thus, any prior disclosure of the end-product was anticipating.

Product-by-Process: On appeal, the CAFC affirmed, finding that product-by-process claims cannot recapture the end-product from the public domain.

[O]nce a product is fully disclosed in the art, future claims to that same product are precluded, even if that product is claimed as made by a new process.

This rule of law follows the MPEP, which states that:

[Even] though product-by-process claims are limited by and defined by the process, determination of patentability is based on the product itself. The patentability of a product does not depend on its method of production. If the product in the product-by-process claim is the same as or obvious from a product of the prior art, the claim is unpatentable even though the prior product was made by a different process.

MPEP § 2113 (8th ed., Rev. 2, May 2004) (quoting In re Thorpe, 777 F.2d at 698).

Differences from Prior-Art Product: In a footnote, SmithKline’s attorneys noted that the product made through the claimed process is actually different (lacked a pink hue and did not contain spherical granules).  The district court ignored these differences because they were “not required by the patent claims or specification.” 

However, the appellate panel the differences at least interesting — “If those product-by-process claims produced a different product than that disclosed by the [prior art], there would be an argument that the [prior art] disclosure did not anticipate.” In the end, the panel refused to consider the issue because it was only raised in a footnote:

This footnote is the only statement that even approaches a substantive argument on novelty in the entire Argument section of SmithKline’s opening brief. . . . We find that these mere statements of disagreement with the district court as to the existence of factual disputes do not amount to a developed argument. . . . Further, arguments raised in footnotes are not preserved.

Invalidity affirmed.

Comments

[Rob Robinson]

What's there to say but that this chemical chimera should have never escaped from the lab to begin with.....

[Blaise Mouttet]

As a former U.S. Patent Examiner I'm surprised that this issue does not come up more often. There are many patent applicants who routinely use the strategy of including dependent claims such as "The product produced by the process of claim 1" or "The article of manufacture made by the apparatus of claim 1", etc. These claims often end up allowed when the independent claim is allowed either because examiners are used to simply allowing dependent claims when the independent claim is allowed or because production pressure makes examiners reluctant to hold out continued rejections when the majority of the claims are allowable.

The problem is enhanced by the fact that many mixed statutory claims such as these are extremely difficult to evaluate by a patent examiner even when an allegedly identical product is found in the prior art. The rejection would necessarily be based on inherency arguments which patent attorneys will typically argue around and which are (in my experience) frowned upon by USPTO culture.

MPEP 2173.05(p) gives some insight on this issue (see link).

http://www.uspto.gov/web/offices/pac/mpep/documents/2100_2173_05_p.htm#sect2173.05p

[Kathryn Lappegard]

I didn't realize we needed a CAFC ruling on this.

In my practice, I'm pretty careful about making sure my product is novel before including a product-by-process claim or dependent claim to a product. My inventors tend to push for these claims and I give the "product must be novel" speech on a regular basis.

[Dennis Crouch]

Kathryn, As you can see from the dissent in this case, there is still some debate.

I think that you should seek product+process claims when the product is distinct, but perhaps not patentably distinct. The combination of a slightly different product and a novel process may be sufficient.

[Bart van Wezenbeek]

This ruling confirms the views that are also maintained in Europe. However, would the opposite also be true in the US: does protection of a -new- product from a product-by-process claim encompass also products made by other processes or is it limited to only those products made by that single process? As you may know, in Europe the first option (general protection of the -new- product) is held to be true.

[Dennis Crouch]

Bart. Great comment. Ultimately, I think that this is the point of the dissent. Although it is fine to have an exception for product-by-process (must be a new product), that exception should not mean that claims should be construed to entirely disregard the process limitations.