LAUSR

VOLUME 35 NUMBER 12 DECEMBER 2017 NATURE BIOTECHNOLOGY cases meet the written description requirement. Later, in Noelle v. Lederman, a claim to antibodies, defined by binding affinity to an antigen, was rejected because the structure of the antigen was not adequately described and, as a corollary, would not apply to all antibodies binding that antigen. In a subsequent case involving patents on tumor necrosis factor-α (TNF-α)-targeting antibodies, Abbott’s patents on Humira (adalimumab) were upheld on the grounds that competitor Centocor Ortho Biotech’s claims to a class of antibodies that could bind to human TNF-α did not provide an adequate description of the invention. “It has been, at the least, hotly disputed that knowledge of the chemical structure of an antigen gives the required kind of structure-identifying information about the corresponding antibodies,” the Federal Circuit court noted in its opinion in the Sanofi case. The case will now go back for another trial, after which many court observers expect the Federal Circuit to make a clear statement about when it will be possible to invoke an antibody exception rule. Even with current synthetic technology, when given an antigen, the chances of making the same antibody to it, structurally or even functionally, are small. “As monoclonal antibody technology exists today and since its inception, it is essentially a trial-and-error process,” says Oskar Liivak, a professor at Cornell Law School in Ithaca, New York, who has filed several briefs in Centocor and other biotech patent cases discussing the Sanofi, and Regeneron, in Tarrytown, New York, from selling Praluent. The injunction was never enforced, however, and Sanofi and Regeneron appealed the ruling. The Federal Circuit court heard the challenge and on October 5 overturned the decision, It concluded that the District Court had erred because, although Amgen’s patents disclosed a functional description of how they generated and screened antibodies that bind PCSK9, the patents lacked detail of the antibody’s structure. The key question is whether a functional description of an antibody, as stated in the antibody exception rule, may be enough to support a claim. According to internal Patent and Trademark Office guidelines, it is enough if there is a “known or disclosed correlation between function and structure.” The PTO has even used the functional relationship between antibody and antigen as an example in training materials for examiners. But the Federal Circuit has never clearly outlined when it is acceptable to cite an antibody’s function rather than describe its structure as the basis for a patent that would block competitors attempting to develop drugs against the same biological pathway. The Appeals Court Federal circuit has wrestled with this issue several times. The first such case, Enzo Biochem, Inc. v. GenProbe Inc., involved claims to nucleic acid probes defined by hybridizing them to the genetic material of certain bacteria. There, the court observed that a functional description of the genetic material could in some A high-profile trial to decide whether the cholesterol-lowering drug Praluent (alirocumab) can stay on the market could have broad implications, not only for its owners Sanofi and Regeneron, but for antibody therapies generally, some of whose patents rely on a controversial rule. A decision by the Court of Appeals for the Federal Circuit, which oversees the application of US patent law, will allow Praluent’s drugmakers to present evidence challenging rival Amgen’s patent claims. The outcome could shape the reasoning the US Patent and Trademark Office applies when it considers antibody patent applications, making it considerably more difficult for a company to obtain antibody patents based on the “antibody exception rule,” which allows intellectual property claims to the target, rather than the antibody’s composition. The case could even have implications beyond antibody therapies to other drug classes, including chimeric antigen receptor technology (CAR)-T. At issue are two key patents owned by Amgen for its own cholesterol medicine Repatha (evolucumab). The Thousand Oaks, California–based biotech took Sanofi and Regeneron to court claiming that Praluent infringed its patents. The partners have admitted infringement, but argue that Amgen’s patent claims to Repatha are invalid (Nat. Biotechnol. 34, 452, 2016). Both medicines are monoclonal antibodies that lower cholesterol by binding to proprotein convertase subtilisn kexin type 9 (PCSK9). Although high cholesterol is mostly treated with statins, in some situations, due to side effects or because the statins are ineffective, patients need an alternative, which PCSK9 blocking agents provide (Nat. Biotechnol. 33, 785–786, 2015). PCSK9 is a natural protein that regulates the receptors on liver cells that clear cholesterol from blood. PCSK9 destroys and recycles these receptors. By binding PCSK9, both Praluent and Repatha allow the receptor to stay on the liver longer so it can clear more cholesterol. Praluent and Repatha have been in a close contest since 2015, when they were approved by the FDA a month apart. In March 2016 Amgen won a patent infringement court case against Regeneron and Sanofi when a US District Court upheld two key Amgen PCSK9 monoclonal antibody patents, refusing to hear key evidence that would have supported the Sanofi/Regeneron position (Nat. Biotechnol. 34, 452, 2016). The court issued an injunction to stop Paris-based Amgen cholesterol antibody feud could kill key IP rule