Issue 200 | May 19, 2004
SmithKline Beecham Corp. v. Apotex Corp.,
Effect on future pharmaceutical patenting strategy

A recent decision by the U.S. Court of Appeals for the Federal Circuit in a patent infringement case between innovator and generic drug manufacturers, SmithKline Beecham Corp. v. Apotex Corp., 2004 U.S. App. LEXIS 8107 (Fed. Cir. Apr. 23, 2004), provides clarification of “accidental” infringement by the unintended presence of minor components in pharmaceutical products, and a decision of first impression on a “public use” bar to patentability arising from clinical trials.

Facts at issue in the case

Apotex, a generic drug manufacturer, filed an Abbreviated New Drug Application for the anhydrate form of paroxetine hydrochloride (PHC). Subsequently, SmithKline accused Apotex of patent infringement based on claim 1 in a SmithKline patent (“the ‘723 patent”), which was directed to the PHC hemihydrate compound (currently marketed by SmithKline as Paxil®).

Apotex’s product containing trace amounts of the patented product infringes

SmithKline charged, inter alia, that Apotex’s anhydrate product would necessarily contain at least trace or undetectable amounts of PHC hemihydrate and would therefore literally infringe the ‘723 patent. To prove infringement, SmithKline presented evidence that, based on a “disappearing polymorph” theory, the previously known anhydrate form inevitably converted to the patented hemihydrate by seeding. The District Court and the Federal Circuit found that this evidence supported the inference that Apotex’s anhydrate contained at least trace or undetectable amounts of the hemihydrate.

The lower court held that the SmithKline patent was not infringed because it found that Apotex’s anhydrate product did not contain “commercially significant” amounts of the hemihydrate, that the formation of trace amounts of the hemihydrate was undesirable and that the presence of the hemihydrate did not provide Apotex with any commercial advantage. The Federal Circuit reversed, holding that the SmithKline claim to the hemihydrate was not limited to commercially significant amounts of that compound, and held the patent infringed.

SmithKline’s clinical trials did not constitute an experimental use

Apotex argued that the ‘723 patent was invalid because PHC hemihydrate was present in SmithKline’s PHC clinical trials more than a year before the filing date of the ‘723 patent. 35 U.S.C. §102(b) bars patenting where, before the critical date, an invention is ready for patenting but was involved in a “public use” by a third party not under any obligation of confidentiality. SmithKline did not question the public disclosure of its clinical trials and admitted having administered PHC hemihydrate to patients without any apparent confidentiality restrictions. However, SmithKline contended that the experimental use doctrine negated application of the §102(b) public use bar, asserting that the clinical trials were conducted to establish the safety and efficacy of PHC hemihydrate. After reaching the threshold determination that the non-confidential clinical trials constituted a public use bar under §102(b), the court agreed with Apotex, holding that “an experimental use only negates a statutory bar when the inventor was testing claimed features of the invention.” Id. at *31.

Here, SmithKline’s claim was directed to the compound, without regard to its efficacy, commercial use, or pharmaceutical viability. Thus, the court reasoned that clinical trials designed to establish safety and efficacy of the compound as an antidepressant for FDA approval were not experimental uses related to the features of the claimed invention, the chemical compound. The Court held that the compound invention had been completed (reduced to practice) when the compound was first made and that its efficacy as an anti-depressant was irrelevant. Consequently, further testing did not qualify as experimental use for purposes of negating the §102(b) public use bar, and claim 1 of SmithKline’s ‘723 patent was found invalid.

The court went on to caution future patentees that “testing the properties, uses, and commercial significance of a compound claimed solely in structural terms may start the clock under §102(b) for filing a claim that is not limited by any property, commercially significant amount, or other use of the compound.” Id. at *39.

Effect of the decision on future pharmaceutical patenting strategy

Therefore, according to the ruling in SmithKline Beecham Corp., a product containing trace or even undetectable amounts of a patented compound may be held to infringe a claim directed to the compound where the evidence establishes that the patented compound is inevitably present in the accused product. Additionally, SmithKline Beecham Corp. indicates that non-confidential clinical trials may not constitute a permissible experimental use that negates the §102(b) public use bar unless those trials are designed to test the claimed features of the invention.

In order to avoid infringement of compound patents after SmithKline Beecham Corp., generic manufacturers will need to ensure that the patented compound is not inevitably formed in their products, whether identifiable or not.

This case also serves as a warning to innovators to file patent applications claiming chemical compounds potentially useful in pharmaceutical products prior to (or, in the United States, at least within one year after) initiation of clinical trials related to such products.

More information

If you have any questions about the above information or its ramifications for intellectual property practice, please contact Darby & Darby P.C. Principal Paul Fehlner at pfehlner@darbylaw.com or 212 527-7665.

The information contained in this email is provided for informational purposes only and does not represent legal advice. Neither the APLF nor the author intends to create an attorney client relationship by providing this information to you through this message.

APLF · PO Box 7418 · Washington, DC · 20044-7418