Support IPWatchdog with an individual sponsorship: Click here
“[O]n the facts here, binding is not disputed. Thus, if the ligand binds, the ligand is capable of binding.” – Federal Circuit
On Wednesday, December 4, the U.S. Court of Appeals for the Federal Circuit (CAFC) issued a precedential opinion in Cytiva Bioprocess R&D AB v. JSR Corp. affirming the Patent Trial and Appeal Board (PTAB) invalidation of Cytiva’s patent claims to chromatography compositions for isolating target compounds. The Federal Circuit also reversed the PTAB findings that Cytiva’s process patent claims were not unpatentable, holding that a reasonable likelihood of success need not be shown when the challenged claim language claims an inherent property of the technical subject matter.
Cytiva’s appeal stems from a series of 10 inter partes review (IPR) proceedings challenging the validity of claims from three patents: U.S. Patent No. 10213765; U.S. Patent No. 10343142; and U.S. Patent No. 10875007. Each of these patents shares the common title Chromatography Ligand Comprising Domain C from Staphylococcus Aureus Protein A for Antibody Isolation. These patents claim compositions and processes for the preparation of chromatography matrices, used in chemical processes like pharmaceutical manufacturing, including antibody binding properties and protein modifications to improve matrix stability.
The PTAB found Cytiva’s composition patent claims to be obvious in light of the prior art, noting that the G29A modification claimed by Cytiva, a technique for improved alkaline stability known since the 1980s and involving the substitution of glycine for alanine at position 29, was expressly suggested by the prior art. However, Cytiva’s process claims from the ‘142 and ‘007 patents survived these final written decisions. According to the Board, although the particular antibody binding feature claimed by Cytiva was an inherent property obviating the composition claims, JSR had not established that a person of ordinary skill in the art would have had a reasonable expectation of success that the prior art antibody binding process would have achieved improved binding capability for the modified chromatography matrix.
Flexibility of Obviousness Under KSR Means Lead-Compound Analysis Not Required
On appeal, Cytiva challenged the PTAB’s invalidity determinations on the grounds that the Board did not conduct a lead-compound analysis to support its findings that a person of ordinary skill would have selected the claimed binding region over other domains exhibiting unexpected antibody-binding properties. Lead-compound analysis is a two-step process that first asks whether a chemist of ordinary skill would have selected the prior art compound as a lead compound or starting point, and then whether the prior art provides a motivation to modify the lead compound to make the claimed compound with a reasonable expectation of success.
While lead-compound analysis is useful for obviousness challenges to chemical claims, the Federal Circuit underscored that such analysis is not always required, pointing to the flexibility in obviousness analyses required under KSR International v. Teleflex. Here, the appellate court found that the asserted prior art references expressly suggested Cytiva’s claimed modification. Under those circumstances, there was no issue with the Board’s reliance on KSR’s obvious-to-try rationale in its decisions.
Even under a formal lead-compound analysis, the Federal Circuit still found substantial evidence supporting the PTAB’s determination that any binding domains of the claimed staphylococcus protein A (SPA) could serve as lead compound. In its decisions, the Board found the prior art suggested the use of any SPA domain as a starting ligand, meeting the first prong of the lead-compound inquiry. Further, the prior art established that amino acid sequences of each domain were homologous, and that those domains share many structural similarities.
‘No Question of Reasonable Expectation of Success’ in Achieving Inherent Property
JSR cross-appealed, challenging the PTAB’s findings that Cytiva’s process claims were not unpatentable. The Federal Circuit responded first by concluding that there was no material difference between the composition claims and process claims at issue. While there were differences in claim language regarding the binding capability of the chromatography matrix’s ligand to the SPA, the appellate court noted that “on the facts here, binding is not disputed. Thus, if the ligand binds, the ligand is capable of binding.”
The court next found that the PTAB used an improperly narrow construction of the claim term “Fab part of an antibody.” The Board concluded that JSR had not shown a reasonable expectation of success in binding a modified SPA region to “Fab fragments” rather than the Fab region of the antibody. The Federal Circuit, noting that both parties agreed to a definition that encompassed both Fab fragments and the Fab region of the whole antibody, the Board was wrong to require JSR to show that a person of ordinary skill would know that the ligand binds to Fab fragments separated from the antibody before performing the claimed process.
Unexpected properties in an otherwise obvious composition can lead to a nonobviousness finding, a premise Cytiva cites from the Federal Circuit’s 2017 ruling in Honeywell International v. Mexichem Amanco Holding. However, the appellate court noted that Honeywell did not involve the claiming of an inherent property but rather the combination of two compounds leading to unexpected properties. Citing to Hospira v. Fresenius Kabi (2020), JSR argued that “there is no question of a reasonable expectation of success” in achieving inherent properties of claimed compositions. Because the process claims only claimed the inherent property of Fab binding, the Federal Circuit found that the same conclusion in Hospira held true here.
Finally, the Federal Circuit nixed Cytiva’s argument that the PTAB improperly ignored secondary considerations of the modified chromatography matrix’s retained ability to bind to the antibody’s Fab region. Although the PTAB could have more specifically stated that Fab binding was not unexpected, the Board was not required to recognize the inherent property in its analysis. As in In re Couveras (2023), the Federal Circuit concluded that results flowing naturally from a claimed compound cannot overcome obviousness, even if the nature of the mechanism is unexpected.
The opinion was authored by Judge Prost and Eric Dittmann of Paul Hastings LLP represented JSR.
