Quarterly Journal Volume 53, Issue 2 June 2025

ALPHAFOLD 3, AI, ANTIBODY PATENTS, THE FUTURE OF BROAD PHARMACEUTICAL PATENT CLAIMS, AND DRUG DEVELOPMENT 

Artificial intelligence (AI) will have an enormous impact both on pharmaceutical development and patent protection, particularly for antibody therapeutics. In Amgen Inc. v. Sanofi, the U.S. Supreme Court limited the scope of Amgen’s therapeutic antibody patent to only those antibodies that were specifically described in Amgen’s patent application and that had been shown to bind to a particular region of the target antigen, blocking the activity of the antigen that caused disease. The reason for this limitation was the patent requirement of enablement: that potentially millions of antibodies could be generated to the target antigen but that not all would bind in a way that produced the therapeutic effect. The Court concluded that Amgen’s patent had not enabled other scientists to produce antibodies with the desired activity without “undue” experimentation, concluding a decades-long shift in their caselaw limiting the permissible scope of monoclonal antibody patents. Our primary conclusion is that artificial intelligence has the power to overcome the problem of enablement that currently limits the scope of antibody patents. We also conclude that the rapid pace of improvement in AI is likely to bring about significant changes in pharmaceutical patents generally, with the potential to transform the future of drug development and the pharmaceutical industry.

PRIMA FACIE LACK OF NOVELTY: WHEN PRIOR ART RANGES GIVE RISE TO REBUTTABLE ANTICIPATION 

The Court of Appeals for the Federal Circuit has long- and well-established standards for determining whether a claimed numerical range is prima facie obvious over a prior art reference disclosing a similar range, or whether the claimed range is conclusively anticipated by a narrower range or a point. It is fair to say that this area of the law is stable. However, until relatively recently, such has not been the case with the doctrine of rebuttable anticipation. This article provides a contextual and historical survey of decisions in the area of claimed and prior art ranges. It then focuses in detail on the younger doctrine of prima facie anticipation and reaches conclusions as to its application and scope.

The law regarding prima facie anticipation of numerical ranges by overlapping prior art ranges is nuanced. After Atofina v. Great Lakes Chem. Corp. (Fed. Cir. 2006) the court does not treat endpoints of prior art ranges as exemplified values. When a prior art endpoint falls within—or abuts the endpoint of—a claimed numerical range, it does not conclusively anticipate. Instead, it raises a case of prima facie anticipation. That was not always the case, as shown by Atlas Powder Co. v Ireco Inc. (Fed. Cir. 1999) and Perricone Corp v Medicis Pharmaceutical. (Fed. Cir. 2005), two decisions preceding Atofina. In those two cases, the court was still treating prior art endpoints as conclusively anticipatory.

The doctrine of rebuttable anticipation arose almost sub silentio in Atofina. The reason for the lack of clarity is that the concept of rebutting the presumption of anticipation by demonstrating criticality of the claimed range did not appear in the court’s reasoning until six years later in ClearValue v Pearl River Polymers (Fed. Cir. 2012). Both Atofina and ClearValue together established what we call the Atofina/ClearValue framework for prima facie anticipation and its rebuttal.

In a pair of post-Atofina/ClearValue decisions, the Federal Circuit clarified what it does or does not consider convincing evidence of criticality to rebut prima facie anticipation. In OSRAM Sylvania v. American Induction (Fed. Cir. 2012) the court found the proffered evidence of criticality convincing, yet in Ineos v. Berry Plastics (Fed. Cir. 2015) it did not. The difference resides in that successful evidence needs to be related to the operability and functionality of the claimed invention (OSRAM) but not to side issues, such as decreased manufacturing costs (Ineos).

Atofina (Fed. Cir. 2006) also analyzed a scenario when an overlapping prior art range fully subsumes the claimed range. This is a situation where neither of the prior art endpoints falls into—or abuts the endpoints of—the claimed range. If the subsuming prior art range is “small,” it may conclusively anticipate under the doctrine of “immediately envisaging” every point of the prior art. However, if a fully subsuming prior art range is “large” for purposes of “immediately envisaging” every point, it does not even lead to prima facie anticipation. At best, it then raises prima facie obviousness. And such prima facie obviousness may be rebutted by classical evidence of secondary considerations.

Furthermore, in recent cases such as Genentech, Inc. v. Hospira, Inc. (Fed. Cir. 2020) and UCB, Inc. v. Actavis Laboratories UT, Inc. (Fed. Cir. 2023), the court has started confronting situations that raise both prima facie anticipation and prima facie obviousness. The court then analyses different evidences of rebuttal: criticality for prima facie anticipation and secondary considerations for prima facie obviousness. And, at times, convincing evidence of criticality can rebut both presumptions.

In the conclusions of the article, we provide a flowchart, which will help practitioners navigate the many issues that arise when confronting disclosures of ranges in the prior art.

The adoption of prior user right in U.S. patent law was a big deal, largely aligning the U.S. with the rule in other countries. Simultaneous or near-simultaneous invention is extremely common, so prior use should be as well.

But a surprising thing has happened in the last thirteen years: virtually nothing. Only three decisions in those thirteen years involve substantive claims of prior use. All are by district courts, and two of those decisions were made on largely procedural grounds. Only a single case actually finds prior use by another under the statute.

I document the surprising absence of prior user right litigation. I consider a number of reasons for the missing caselaw, none entirely satisfactory. I consider and (mostly) reject explanations that are a function of when and how cases are litigated, the difficulty of detecting non-public uses, and the existence of prior art as an alternative to the defense. Nor does the explanation seem to be that simultaneous use without publication is no longer as important as it once was. Instead, the real reason no one uses the prior user right is likely a function of the significant limits Congress put on the exercise of the right. It may be time to revisit those limits.

CHINA’S UTILITY MODEL SYSTEM: A FRAMEWORK TAILORED TO THE COUNTRY'S SPECIFIC ECONOMIC AND DEVELOPMENTAL CONTEXTS AND 

China's utility model system, which, intentionally aligned with national economic goals, incorporates considerations of inventor profiles and technological landscapes to drive both innovation and economic advancement strategically. In the face of rapidly evolving industrial landscapes, China's utility model system stands out for its ability to adapt, thereby propelling technological progress and significantly contributing to the overall economic growth of the nation. Particularly noteworthy is its role in fostering innovation within Micro and Small Enterprises (MSEs) and individual inventors. However, the effectiveness of China's utility model system hinges on continuous efforts to address challenges and enhance the caliber and authenticity of utility model patents. A persistent commitment to refinement is imperative in maintaining a balanced and effective intellectual property framework.