. . . The Federal Circuit explained how these two cases provide important guideposts. . . .
In Cytiva BioProcess R&D AB v. JSR Corp., No. 2023-2074 (Fed. Cir. Dec. 4, 2024) (“Decision”), the Federal Circuit addressed obviousness challenges to multiple patents related to affinity chromatography, clarifying when a “lead compound” inquiry may be substituted with an “obvious-to-try” analysis and when the presence of inherent properties may obviate the need to demonstrate a reasonable expectation of success. This decision arose from inter partes review (“IPR”) proceedings initiated by JSR, challenging patents held by Cytiva, before the Patent Trial and Appeal Board (“PTAB”), which invalidated 79 composition claims across three patents while upholding four process claims. On appeal, the Federal Circuit affirmed the PTAB's invalidation of the 79 composition claims but reversed the PTAB’s decision on the four remaining process claims, concluding that all contested claims were unpatentable as obvious.
Background
The patents at issue are U.S. Patent Nos. 10,213,765; 10,343,142; and 10,875,007. These patents relate to affinity chromatography matrices and processes for isolating biomolecules, particularly antibodies.
Affinity chromatography is a biochemical technique that separates molecules based on specific binding interactions. The matrices at issue use ligands, which are small molecules or peptides designed to bind selectively to a target biomolecule. In this case, the ligands were derived from Protein A (SPA), a molecule found in the bacterium Staphylococcus aureus and widely used in the purification of antibodies. SPA contains several subunits, known as “domains” (labeled A through E), each capable of binding to antibodies. Researchers had previously engineered a synthetic variant of one domain (Domain B), introducing a specific mutation—replacing a glycine residue with alanine at position 29 (G29A)—to improve stability under alkaline conditions. See Decision at 6. Cytiva’s patents applied the same G29A mutation to another domain, Domain C, claiming both the matrices containing this modified ligand and the processes for isolating antibodies using the matrices. Id.
Chromatography matrices are packed into columns, which serve as the primary tools for separating and purifying biomolecules. During operation, a liquid mixture containing the target compound is passed through the column. The ligands within the matrix selectively bind to the target molecules, while non-target substances flow through the column. Subsequently, the target molecules are released—or eluted—from the ligands by altering the chemical conditions, such as pH or ionic strength, of the solution passing through the column. See Decision at 4. This cycle of binding and elution allows for efficient purification of biomolecules like antibodies. Id.
The problem these innovations sought to address was the degradation of SPA-based matrices during repeated exposure to the high-pH alkaline solutions used for cleaning. The cleaning process involves flushing the columns with the alkaline solution to maintain their functionality across multiple cycles. However, this high-pH environment can degrade the structural integrity of the ligands, shortening the column’s lifespan. The G29A mutation enhanced the structural integrity of SPA, reducing degradation and extending the functional lifespan of the chromatography matrices. Although the prior art had successfully applied this mutation to Domain B, its application to Domain C was not explicitly demonstrated. Cytiva argued that this extension to Domain C, coupled with its specific antibody-binding properties, constituted a nonobvious advancement of the art. See Decision at 5.
JSR challenged the validity of Cytiva’s patents in the IPR proceedings, asserting that the claims were obvious in light of prior art, referred to in the opinion as Linhult, Abrahmsen, and Hober. See Decision at 8. The PTAB ruled that the 79 claims to matrix compositions were unpatentable in view of that prior art but upheld the four process claims, which separately described isolating specific antibody regions. Both parties appealed to the Federal Circuit.
The Federal Circuit’s Analysis
In analyzing the issues on appeal, the Federal Circuit applied the well-established two-part inquiry for determining obviousness. First, the court evaluated whether there was a motivation to modify the prior art to arrive at the claimed invention. See Decision at 10-11 (citing Otsuka Pharm. Co. v. Sandoz, Inc., 678 F.3d 1280, 1291 (Fed. Cir. 2012)). Second, the court considered the sufficiency of evidence demonstrating whether a skilled artisan would have had a reasonable expectation of success in achieving the claimed invention. See Otsuka, 678 F.3d at 1292.
In its decision, the Federal Circuit confirmed that the so-called “lead compound” analysis need not be applied in circumstances where the prior art “expressly suggest[s] the proposed modifications.” Decision at 11 (citing SIBIA Neurosciences, Inc. v. Cadus Pharm. Corp., 225 F.3d 1349, 1359 (Fed. Cir. 2000)). The Federal Circuit further confirmed that demonstrating a reasonable expectation of success is not necessary where the claims recite known and previously demonstrated inherent features of a claimed composition. See Decision at 22-23 (citing Hospira, Inc. v. Fresenius Kabi USA, LLC, 946 F.3d 1322, 1332 (Fed. Cir. 2020)).
Composition Claims
Typically, the first step in an obviousness inquiry involves identifying a lead compound in the prior art and establishing the basis for selecting that lead compound. Here, the PTAB did not identify a lead compound in its analysis, and the Federal Circuit first addressed whether this omission constituted error. See Decision at 11. Cytiva argued that the Board’s determination that the claimed modifications to Domain C were obvious lacked a foundational analysis of whether a skilled artisan would have selected Domain C as a starting point for the modifications. The court rejected this argument, explaining that a lead compound analysis is not always necessary “where the prior-art references expressly suggest the proposed modification.” Decision at 11 (citing, SIBIA Neurosciences, 225 F.3d at 1359).
The court found that the Linhult and Abrahmsén references expressly disclosed mutating the SPA in Domain B with the G29A substitution to improve stability in alkaline conditions. Decision at 12. The references further taught that “the IgG binding domains of SPA – E, D, A, B, or C share many structural similarities.” Decision at 12. Under these circumstances where the prior art explicitly suggests the claimed modifications, the Federal Circuit concluded that a lead compound analysis is not required. Id. at 13. Instead, the court upheld the PTAB’s reliance on KSR Int’l Co. v. Teleflex Inc., 550 U.S. 398, 415 (2007), which allows for an “obvious-to-try” approach, especially when there is a finite number of predictable solutions to a known problem. See id. at 11-12.
The court further emphasized that even under a formal lead compound analysis, JSR would still prevail. Specifically, the court noted that the PTAB had expressly found that “the prior art ‘suggest[s] the use of any one of the SPA IgG binding domains E, D, A, B, or C as the starting ligand.’” Decisison at 13 (emphasis in the original). Moreover, the prior art demonstrated that all SPA domains shared significant structural similarities and that the G29A modification had already been successfully applied to Domain B. Under these circumstances, the court was satisfied that a skilled artisan would have reasonably selected Domain C as a starting point for further development. Id. at 12-13.
Having established that the G29A modification to Domain C was explicitly suggested by the prior art, the court turned to the issue of whether a skilled artisan would have been motivated to make the modifications indicated by the prior art with a reasonable expectation of success. Although the parties did not dispute that an inherent feature of the claimed invention need not be specified in the prior art as part of the obviousness analysis, the parties disputed whether an inherent property of an otherwise obvious combination negates the need to demonstrate a reasonable expectation of success. See Decision at 19.
In advocating its position, Cytiva relied on case law establishing that “‘unexpected properties [although inherent] may cause what may appear to be an obvious composition to be nonobvious.’” Decision at 21 (quoting Honeywell Int’l Inc. v. Mexichem Amanco Holding S.A. DE C.V., 865 F.3d 1348, 1354–55 (Fed. Cir.2017)). In contrast, JSR relied on case law establishing that “‘[i]f a property of a composition is in fact inherent, there is no question of a reasonable expectation of success in achieving it.’” Decision at 21 (quoting Hospira, 946 F.3d at 1332)).
The court reconciled these positions by explaining that in Honeywell, the court held that inherency could not substitute for a lack of motivation to modify the prior art, emphasizing that “Honeywell did not attempt to claim the inherent result.” Decision at 22. Instead, Honeywell's claims were directed to a composition which was not itself inherent and which exhibited unexpected properties. Id. By contrast, in Hospira, the claims were directed to the inherent property itself, and the court concluded that a property inherently present in an obvious combination obviates the need for a separate showing of reasonable expectation of success. Id.
The Federal Circuit explained how these two cases provide important guideposts: “When claims require prior knowledge of the inherent property—e.g., for motivation to combine—then a petitioner would still generally need to demonstrate a reasonable expectation of success.” Id. The court noted, however, that “the situation is different” when “simply claiming an inherent property of an otherwise obvious composition or process—i.e., one obvious without regard to the property at issue.” Id. at 22-23. The court emphasized that when the claim is directed to a property that inherently arises from an obvious combination, a reasonable expectation of success is presumed, and a separate showing of the expected success is unnecessary. Id. at 23.
Applying these principles, the court determined that the matrix binding claimed in Cytiva’s patents, as an inherent feature of the modified ligand, naturally flowed from the modifications suggested by the prior art. Because the prior art explicitly taught the claimed modifications, Cytiva’s reliance on Honeywell was misplaced, and the composition claims were unpatentable due to obviousness.
Process Claims
The Federal Circuit next considered whether the PTAB erred in finding the four process claims nonobvious. In doing so, the court rejected the PTAB’s conclusion that there were no substantive differences between the composition claims and the process claims. While the process claims contained additional language describing steps for binding and isolating specific antibody regions, the recited “Fab part of an antibody,” the court explained that these steps relied on the inherent properties of the modified ligand recited in the composition claims. The ligand’s binding ability, already deemed an inherent and obvious feature of the composition claims, necessarily enabled the binding and isolation processes described in the process claims.
The court next addressed the PTAB’s narrow construction of the claim term “Fab part of an antibody” as referencing only “Fab fragments.” The Fab part of an antibody refers to the “fragment antigen-binding” region, which is responsible for recognizing and binding to specific antigens. It consists of parts of the antibody's variable domains. Fab fragments, on the other hand, are smaller portions of antibodies that include the antigen-binding regions but lack other structural components of the full Fab region.
Because the PTAB had interpreted the term to mean only Fab fragments, rather than encompassing the broader Fab region of a full antibody, the PTAB concluded that JSR had failed to demonstrate a reasonable expectation of success in isolating “Fab fragments,” as opposed to the entire “Fab part of the antibody,” when using the claimed processes. The Federal Circuit disagreed, finding that the claim term “Fab part of the antibody” encompasses both Fab fragments and the Fab region of a full antibody. This broader construction was consistent with the patents’ specifications, which described ligands capable of binding to both full antibodies and fragments. By improperly limiting the scope of the claims, the PTAB’s analysis failed to consider the full breadth of the term and hence the obviousness of the claimed invention.
With the corrected claim construction, the court revisited the PTAB’s findings. The court concluded that the inherent Fab-binding properties of the ligand—established during the analysis of the composition claims—applies equally to the process claims. Because the process claims, like the composition claims, recited a property that naturally and inherently flowed from an otherwise obvious combination, the process claims were also unpatentable as obvious.
Conclusion
The Federal Circuit’s decision in this case clarifies when a lead compound analysis is or is not required. The court explained that a lead compound analysis can be avoided in favor of a KSR-type “obvious-to-try” analysis where the prior art explicitly teaches a finite number of predictable solutions. Additionally, the decision underscores that the inherent properties of those modifications satisfy the reasonable expectation of success requirement of an obviousness analysis where the inherent properties naturally result from the prior art combination and are themselves recited limitations of the claimed invention.