Randolph J. Noelle v. Seth Lederman, Leonard Chess, and Michael J. Yellin

Citations: 355 F.3d 1343; 69 U.S.P.Q. 2d (BNA) 1508; 2004 U.S. App. LEXIS 774; 2004 WL 77931Docket: 02-1187

Court: Court of Appeals for the Federal Circuit; January 19, 2004; Federal Appellate Court

SummaryOriginalCytator Case TreatmentBackward CitatorAnalysis

EnglishEspañolSimplified EnglishEspañol Fácil

Randolph J. Noelle appeals a decision from the United States Patent and Trademark Office's Board of Patent Appeals and Interferences, which found no interference-in-fact between his United States Patent Application Serial No. 08/742,480 ('480 application) and the United States Patent No. 5,474,771 ('771 patent) owned by Seth Lederman, Leonard Chess, and Michael J. Yellin. The Board also rejected specific claims (51, 52, 53, 56, 59, and 60) of the '480 application under 35 U.S.C. 102(b). The Federal Circuit, led by Circuit Judge Gajarsa, affirmed the Board's decision, citing substantial evidence supporting the ruling and its compliance with the law.

The case involves antibodies and their function in the immune system, specifically the CD40CR antibody, which inhibits cell signaling between helper T-cells and B-cells. This antibody binds to the CD40CR antigen on activated helper T-cells, preventing their interaction with the CD40 receptor on resting B-cells, thereby blocking B-cell activation and antibody production. CD40CR antibodies are significant in treating conditions related to an overactive immune response, such as allergies and autoimmune disorders.

Noelle's '480 application, filed on November 1, 1996, is a continuation of a prior application and claims various forms of the CD40CR monoclonal antibody, including murine and humanized versions, as well as the hybridomal cell lines that produce it. Lederman's '771 patent, effective from November 15, 1991, describes the human form of the CD40CR antibody, specifically the 5c8 antibody, which inhibits T cell activation of B cells. An interference was declared by the USPTO on September 3, 1999, with Noelle identified as the junior party and Lederman as the senior party based on filing dates. The interference established one count regarding the monoclonal antibodies specified in both parties' claims.

Claim 1 of Lederman's patent pertains to a monoclonal antibody binding to the 5c8 antigen, while Claim 42 of Noelle's application addresses a monoclonal antibody binding to an antigen on activated T cells, with Claim 51 concerning antibodies specifically binding CD40CR. Claims 1 and 52 are described as relating to the 'human' form of the CD40CR antibody, whereas Claims 42 and 51 pertain to the 'mouse' and 'genus' forms, respectively.

During a hearing on June 28, 2001, the Board addressed preliminary motions, denying Lederman's requests to reject Noelle's mouse claims and redefine the count. The Board found Lederman did not prove that Noelle's mouse claims lacked written description, enablement, or definiteness. However, it determined that Noelle's human and genus claims did not meet the written description requirement as of February 14, 1992, referencing the case of Regents of the University of California v. Eli Lilly Co. to emphasize that a precise definition is necessary for adequate written description of claims.

The Board concluded that Noelle's claims regarding genus and human antibodies from the '480 application lacked written description support from the '799 application due to the absence of structural details for the antibodies or antigens. Consequently, these claims were deemed to constitute new matter, resulting in the denial of the '480 application the benefit of the earlier filing date of the '799 application. The Board found that the human and genus claims were anticipated by prior art, specifically Lederman '771 and Armitage '974, as Noelle had not distinguished his claims from this prior art and conceded that they would be anticipated if the earlier filing date was not applied. As a result, the Board rejected the claims under 35 U.S.C. § 102(b) and 37 C.F.R. § 1.641.

Additionally, the Board addressed the existence of an interference-in-fact between Noelle and Lederman's claims. After Lederman withdrew his motions and requested a finding of no interference-in-fact, the Board determined that a person of ordinary skill in the art would not have had a reasonable expectation of successfully obtaining the claimed invention from the other party due to the state of the art at that time. Although Dr. Edward A. Clark, Noelle's expert, presented three methods for isolating human CD40CR antibodies from the mouse version, the Board ultimately found these methods inadequate. The Board emphasized that references to Noelle's own specification as prior art were improper for the interference analysis, which should focus on the claims rather than specifications.

The Board determined that prior art immunization techniques would not effectively identify activated T-cells producing the required CD40CR antigen or isolate the antigen itself, as there was no reasonable expectation of success at the relevant time. It also found that isolating CD40-Ig would be 'extremely difficult' for a person skilled in the art, which Noelle claimed could be used to obtain the CD40CR antibodies. The Board referenced statements from the prosecution of application 07/969,703, asserting that expression cloning would not allow a skilled artisan to isolate the CD40CR antibody with reasonable likelihood of success. Consequently, the Board concluded that a person of ordinary skill would not likely isolate the human CD40CR antibody based on Noelle's claimed mouse CD40CR antibody, resulting in no interference-in-fact between Noelle's murine CD40CR claim and Lederman's human CD40CR claim. Noelle appealed, and jurisdiction lies under 28 U.S.C. 1295(a)(4)(A) (2000).

The court will review the written description requirement of 35 U.S.C. 112, paragraph (1), as a factual question under a substantial evidence standard. This requires examining the record as a whole to determine if a reasonable factfinder could support the agency's conclusion, acknowledging that conflicting evidence does not negate the finding's support. The written description must clearly convey that the applicant possessed the claimed invention as of the filing date of the earlier application. A prior application merely stating the function of later-claimed genetic material may not satisfy the written description requirement.

A precise definition of DNA is required for adequate written description in patent claims, including structural, formulaic, or chemical details, rather than vague plans for obtaining the invention. Statements regarding functional characteristics or isolation methods do not fulfill this requirement. Each case regarding written description is fact-specific, limiting the precedential value of previous cases. The court has noted that linking functional characteristics to a known or disclosed structure can meet the written description requirement. Additionally, depositing biological material with a public repository can satisfy this requirement. An example provided indicates that claims for antibodies can be valid if they reference well-characterized antigens. In the case of Noelle, the claims concerning the human CD40CR antibody were unsupported because Noelle did not disclose the necessary structural details of the human CD40CR antigen in the prior application, and his assertion that the antibody's binding affinity sufficed was insufficient without detailing the antigen.

Noelle's claims to human forms of CD40CR antibodies in his '480 application are invalid because he did not adequately describe the human CD40CR antigen; he only referenced the mouse form by citing the ATCC number of the hybridoma. Had he sufficiently described the human antigen, he could have claimed its antibody based on its binding affinity. Consequently, Noelle's attempts to define unknowns by their binding affinities do not merit the earlier filing date benefit from his '799 patent application. Furthermore, Noelle cannot claim the genus form of CD40CR antibody merely by describing the mouse form. He cites Staehelin v. Secher to support his genus claim, but this decision is non-binding and does not apply to his case. The Staehelin ruling allowed for genus claims without exhaustive species details, a principle later limited by this court's decisions in Enzo Biochem II and Regents, which emphasize that not all species can be generalized due to unpredictability in results. The Board found that Noelle's human and genus claims were anticipated by prior art (Lederman '771 and Armitage '974), with substantial evidence supporting this finding. Noelle conceded that without his '799 application’s earlier filing date, his claims were indistinguishable from existing prior art. Additionally, interference proceedings are governed by specific regulations aimed at determining the first inventor among competing patent applications for similar inventions.

To assess whether two parties claim the same patentable invention, the USPTO employs a 'two-way' test endorsed by the court. In this context, an invention 'A' is the same as invention 'B' if 'A' is either the same as or obvious in light of 'B,' assuming 'B' is prior art. Conversely, 'A' is considered a separate invention from 'B' if it is new and non-obvious in relation to 'B.' For an interference-in-fact to be established, both inventions must anticipate or render each other obvious, satisfying both criteria of the two-way test.

In the current case, Lederman could succeed in its motion for no interference-in-fact by demonstrating either that its claims are not anticipated or made obvious by Noelle's remaining claims, or vice versa. Noelle's assertion that the Board incorrectly required a two-way test is unfounded, as prior rulings support the validity of this approach. The Board found that a skilled individual would lack a reasonable expectation of success in obtaining Lederman's human subject matter from Noelle's mouse subject matter and confirmed this conclusion through additional analysis, despite not needing to conduct the second prong.

Both parties acknowledged that the anticipation aspect of the analysis is not in dispute, given the distinct claims related to mouse versus human antibodies. Thus, the focus shifts to the obviousness analysis under 35 U.S.C. 103, which requires evaluating whether prior art suggests making the claimed invention and if it provides a reasonable expectation of success in doing so, both of which must be grounded in the prior art rather than the applicant's own disclosures.

The parties agree that a skilled artisan would be motivated to pursue the human CD40CR antibody if the mouse CD40CR antibody were available, but they disagree on the reasonable likelihood of success in doing so based on prior art. The key issue is whether substantial evidence supports the Board's conclusion that an ordinary skilled artisan would not have a reasonable expectation of success in isolating the invention. A reasonable expectation does not require absolute predictability (Yamanouchi Pharm. v. Danbury Pharmacal, Inc., 231 F.3d 1339, 1343). 

Noelle contends that his '799 patent application provides a reasonable likelihood of success for isolating human CD40CR antibodies using mouse CD40CR antibodies, specifically by employing the CD40-Ig fusion protein as a screening tool. He argues that the Board improperly disregarded this isolation method simply because it was included in the written description rather than in the claims.

The Board found no interference-in-fact between Noelle's and Lederman's claims, correctly stating that Noelle's methods for isolating human CD40CR antigens using CD40-Ig were not part of the claimed inventions nor considered prior art. According to USPTO rules, interference-in-fact requires that both parties claim the same patentable invention (37 C.F.R. 1.601(n)). A patentee's invention is defined only by their claims unless specific means-plus-function language is used. Noelle's claims do not include a method of isolating CD40CR antigens or related components.

While ambiguous terms in claims could be clarified using specifications or other sources, this was not necessary here as all terms were clear. Thus, Noelle cannot invoke the CD40-Ig isolation method to demonstrate obviousness against Lederman's invention since it is not claimed.

The Board's determination is further supported by substantial evidence indicating that, given the prior art at the time of the '799 filing, a skilled artisan would not have a reasonable likelihood of success in isolating human CD40CR antibodies from mouse CD40CR antigens. Noelle's assertion that a skilled artisan could manufacture hybridomas secreting monoclonal antibodies to activated human helper T-cell surface antigens relies on three screening methods from his specification, two of which utilize CD40-Ig. Expert testimony indicated that it would have been unpredictable and unreasonable to expect a skilled artisan to produce CD40-Ig at that time.

Dr. Clark, Noelle's expert witness, asserted that a person of ordinary skill in the art would reasonably expect to isolate the human form of the CD40CR antigen using the mouse form and existing expression cloning methods. Conversely, Armitage claimed during the prosecution of his '703 application that expression cloning could not have reasonably led to the successful isolation of the human CD40CR antigen. After reviewing the entire record, it was determined that substantial evidence supported the Board’s conclusion that in the early 1990s, a person of ordinary skill would not likely succeed in isolating the human antigen based on the mouse version. Consequently, the Board’s rejection of claims 51, 52, 53, 56, 59, and 60 from Noelle's U.S. Application No. 08/742,480 was affirmed, along with the Board's ruling granting Lederman's preliminary motion of no interference-in-fact. No costs were awarded. Additional notes provide context on antibody functions, terms associated with CD40CR antigen, and methods for creating hybridomas.