Background
In the late 1980s and early 1990s, scientists discovered that the enzyme cyclooxygenase (COX) existed in two forms: COX-1, which performs beneficial housekeeping functions in the body, and COX-2, which is associated with inflammation and pain. Existing non-steroidal anti-inflammatory drugs (NSAIDs) like aspirin and ibuprofen inhibited both enzymes, causing GI side effects because they suppressed COX-1 as well. The discovery that COX-2 was the relevant enzyme for inflammation suggested that a selective COX-2 inhibitor could treat pain without the GI side effects.
The University of Rochester held U.S. Patent No. 6,048,850, which claimed a method of selectively inhibiting COX-2 activity in a human patient by administering a “non-steroidal compound” that selectively inhibited COX-2. The patent disclosed an assay method for identifying compounds that would selectively inhibit COX-2, but it did not identify or disclose any actual compounds that performed this function. When Pfizer developed and sold Celebrex and Bextra — selective COX-2 inhibitors — Rochester sued, arguing these drugs performed the method claimed in its patent.
The district court dismissed Rochester’s suit on the grounds that the patent’s written description was inadequate: the patent described how to find a COX-2 inhibitor but did not describe one. Rochester appealed.
The Court’s Holding
The Federal Circuit affirmed, holding Rochester’s patent invalid for failure to satisfy the written description requirement of 35 U.S.C. § 112. The court held that a patent claiming a method of using a compound to achieve a result must adequately describe the actual compounds to be used. Rochester’s patent described only the functional goal (selective COX-2 inhibition) and a screening assay for identifying compounds that might achieve that goal. It did not describe any actual compound that would perform the claimed method.
The court characterized Rochester’s patent as claiming a “reach-through” claim — a claim that reaches through from a disclosed screening method to cover all future compounds anyone might discover through that screen. Such claims fail written description because the inventor has not actually possessed the underlying compounds that make the method work. An assay that tells you what properties to look for is not the same as having found a compound with those properties.
Judge Linn dissented, arguing that the majority’s approach was inconsistent with established law on written description and that the patent adequately described the method even if specific compounds were not identified.
Key Takeaways
- A method patent claim covering use of a compound to achieve a biological result must disclose actual compounds capable of performing the method, not just an assay for finding such compounds.
- “Reach-through” claims — those that attempt to cover future discoveries by claiming the functional result of a screening process — do not satisfy written description.
- Describing a problem and a method of searching for a solution is not the same as possessing the solution for written description purposes.
- The written description requirement is an independent condition of patentability, not just a formality merged into enablement.
- Universities and research institutions must have actual, working examples of the claimed compounds before filing broad method-of-use patents.
Why It Matters
The University of Rochester case was a landmark ruling for pharmaceutical and biotechnology patent law. The decision firmly established that patent applicants cannot patent the destination of a research program without having arrived there. Universities and research institutions, which often file patent applications early in a research program to protect potential commercial value, cannot claim methods of using compounds they have not yet discovered.
The case also had enormous practical significance for the COX-2 inhibitor market, which was one of the most commercially successful pharmaceutical categories of the early 2000s. By denying Rochester’s claim to a share of that market through a broad reach-through patent, the court preserved Pfizer’s exclusive rights in Celebrex and Bextra. The ruling influenced how biotech and pharma companies structure their patent portfolios, pushing them toward earlier filing with actual compound data rather than broad functional claims, and became a touchstone in debates about the proper scope of university technology transfer patenting practices.