In re Chapman, 595 F.3d 1330 (Fed. Cir. 2010) (DYK, Gajarsa & Clevenger)
MAJOR ISSUES: obviousness; PTO rejection based on erroneous interpretation of prior art reference; reference: no teaching or teaching away
CROSS-REFERENCE: on teaching away as an indicator of non-obviousness, see Chisum on Patents 5.03[3][a][i][G]; Chisum Patent Law Digest 1546.
1. DIVALENT ANTIBODY FRAGMENTS. An applicant claimed “divalent antibody fragments.” For cited prior art, see U.S. Pat. No. 6,025,158; U.S. Pat. No. 5,436,154. USPatNo6025158 USPataNo5436154
a. Antibodies play a key role in an animal immune system. Antibodies are “Y” shaped proteins that engage potentially harmful antigens. An antibody has variable regions at the Y’s branches (arms), the viable regions being specific to specific antigens. The arms consist each of a light chain and heavy chain. The arms are linked by disulphide bridges (bonds). Fragments are produced with enzymes that digest a whole antibody. A pepsin enzyme removes the stem from the “Y”, leaving a F(ab’)[2] fragment. A papain enzyme breaks the disulphide bridges between the arms, leaving two Fab’ (monovalent) fragments. Compared to whole antibodies, fragments circulate faster and have a shorter half-life.
b. APPLICANT: POLYMER LINKING TWO MONOVALENT FRAGMENTS; DUMBBELL-SHAPED. The applicant’s claimed invention entailed, inter alia, using a polymer to bridge two Fab’ fragments with their light chains removed. The result was a “dumbbell-shaped structure.” The bridge links the sulphur atom of the cysteine residue in the heavy chain of each Fab’ fragment. This bridge substitutes for the natural disulphide bridge. The claimed invention purportedly increased half-life to a point intermediate between a single fragment and a whole antibody.
2. OBVIOUSNESS REJECTIONS BASED ON REFERENCE (GONZALEZ). An examiner and the PTO Board rejected the claims as obvious over the Gonzalez reference, which disclosed, inter alia, attaching a polymer to a F(ab’)[2] fragment and attaching two fragments with a polymer.
3. UNOBVIOUSNESS AS A MATTER OF LAW; NO MOTIVATION TO MODIFY? The applicant argued that the claimed invention was not obvious over Gonzalez as a matter of law because, although Gonzalez described generally linking two fragments with a polymer to form a “dumbbell-shaped structure,” Gonzalez did “not specify how and where the antibody molecules are linked by the polymer molecules, or what fragments are to be used.” To the extent that the applicant argues that “there is no motivation to modify Gonzalez to arrive at [the applicant’s] claimed invention,” the issue is best addressed by the Board on remand, as ordered below.
4. NEITHER AN EXPLICIT TEACHING NOR A TEACHING AWAY. The applicant argued that the reference taught away from the claimed invention and, therefore, did not render the invention obvious. The Federal Circuit disagreed.
a. The reference did not explicitly teach the claimed invention, which entailed using the polymer to bridge two Fab’ (monovalent) fragments by attaching to the hinge cysteine residue of both fragments. In its only working example, which was a F(ab’)[2] fragment, the reference only taught attaching a polymer to the fragment. As an alternative, it taught attaching two fragments with a polymer but did not specify the fragments to be used or the mode of attachment.
b. SAME PROBLEM; SAME ATTACHMENT POINT PREFERRED. The reference did not teach away from the claimed invention. The reference taught attaching a polymer to solve the same problem (increasing antibody fragment half-life to a point intermediate between fragments and a whole antibody). The reference taught multiple locations for attachment but disclosed a preference for the hinge cysteine heavy chain attachment point, the same used in the claimed invention.
5. PTO: THREE ERRONEOUS STATEMENTS ABOUT REFERENCE TEACHING. As the government conceded on appeal, the Board made three erroneous statements about what the reference (Gonzalez) taught.
a. TWO Fab’ FRAGMENTS LINKED BY POLYMER. The Board stated that the examiner had found that the Gonzalez reference taught an antibody structure with two monovalent Fab’ fragments linked by a polymer. The reference did not teach such a structure.
b. LINKING OF CHAINS. The Board stated that the reference taught linking light and heavy chains in a F(ab’)[2] fragment. The reference did not teach such linking, only attaching a polymer to the chains in the F(ab’)[2] fragment.
c. NUMBER OF FRAGMENTS DISCLOSED. The Board stated that the reference taught that three fragments could be used, including Fab’. The reference disclosed six different fragments.
6. ERRORS NOT HARMLESS. Contrary to the government’s arguments, the erroneous statements were not harmless error. The errors increased the likelihood that the PTO erred in rejecting the claims for obviousness.
a. LESS LIKELY TO BE OBVIOUS. On the second error, the claimed invention (polymer linking two F(ab’) fragments) was “less likely to be obvious” because the reference did not teach use of a polymer to link light and heavy chains in a F(ab’)[2] fragment
b. MORE POSSIBILITIES FROM WHICH TO CHOOSE. On the third error, the claimed invention was less likely to be obvious if the reference disclosed “more possibilities from which to chose.”
7. REMAND; OBVIOUSNESS BASED ON A CORRECTED UNDERSTANDING OF THE REFERENCE; DISCLOSURE OF JOINING TWO FRAGMENTS WITH A POLYMER TO MAKE A DUMBBELL-SHAPED STRUCTURE. On remand, the PTO Board must revisit its obviousness conclusion based on a “corrected understanding” of the Gonzalez reference. “The Board is in no way precluded from, and indeed may be correct in, finding the claims to be obvious, particularly in the light of Gonzalez’s disclosure of joining two antibody fragments together with a polymer to make a dumbbell-shaped structure.”
