Antibody Patents

Antibody patent claiming strategy fundamentally divides into structural and functional approaches, each with distinct scope and validity considerations: WHAT ANTIBODIES ARE: antibodies (immunoglobulins) are Y-shaped proteins produced by B cells to identify and bind specific targets (antigens); the binding specificity is determined by the complementarity-determining regions (CDRs) in the variable domains of the heavy and light chains; each antibody has 3 CDRs in its heavy chain variable domain (CDR-H1, CDR-H2, CDR-H3) and 3 in its light chain variable domain (CDR-L1, CDR-L2, CDR-L3); monoclonal antibodies (mAbs) are single-specificity antibodies produced from identical immune cells; STRUCTURAL CLAIMING — DEFINITION BY SEQUENCE: structural antibody claims define the antibody by its specific amino acid sequences; most commonly, claims include: the full heavy chain variable domain (VH) sequence; the full light chain variable domain (VL) sequence; the six CDR sequences (3 heavy + 3 light); or just the three heavy chain CDR sequences (which are most critical for antigen binding); EXAMPLE STRUCTURAL CLAIM: 'An isolated antibody comprising a heavy chain variable region comprising the sequence of SEQ ID NO:1 and a light chain variable region comprising the sequence of SEQ ID NO:2'; ADVANTAGES OF STRUCTURAL CLAIMS: clearly enabled by specific examples; strong written description support from sequence listing; not subject to broad genus enablement concerns under Amgen v. Sanofi; narrow but definite scope; competitor knows exactly what is and is not covered; DISADVANTAGES OF STRUCTURAL CLAIMS: competitors can design around by selecting different CDR sequences; covers only a specific antibody or closely related variants; FUNCTIONAL CLAIMING — DEFINITION BY WHAT IT DOES: functional claims define the antibody by its binding properties, therapeutic effects, or competitive relationship to a known antibody; examples: 'An antibody that binds to PCSK9 and inhibits PCSK9 binding to LDL receptors with IC50 ≤ 10 nM'; 'An antibody that competes with antibody X for binding to antigen Y'; ADVANTAGES OF FUNCTIONAL CLAIMS: very broad scope — covers ALL antibodies with the functional property; hard to design around because competitors must achieve the same function; DISADVANTAGES: subject to genus enablement challenges under Amgen; may lack written description support; THE AMGEN REVOLUTION: Amgen v. Sanofi (S.Ct. 2023) held that purely functional antibody genus claims are not enabled unless the specification enables POSITA to make and use the full scope — which requires more than just examples at the extremes of the claimed functional class.

Amgen Inc. v. Sanofi (S.Ct. 2023) is one of the most important patent cases for the biopharmaceutical industry since Mayo and Myriad: THE PATENTS: Amgen's patents covered antibodies that bind to PCSK9 (a protein that degrades LDL receptors, causing high LDL cholesterol) and block PCSK9 from interacting with LDL receptors; Amgen's drug Repatha (evolocumab) and Sanofi's Praluent (alirocumab) both work through this mechanism; the claims covered ALL antibodies that bind to one of 15 specific amino acid residues on PCSK9 AND block PCSK9 from binding to LDL receptors; THE CLAIMS AT ISSUE: broadly functional genus claims: any antibody that binds to specific residues on PCSK9 AND blocks PCSK9-LDL receptor interaction; the specification disclosed 26 specific antibody examples (working examples) and a general method (shotgun mutagenesis + screening) for identifying additional antibodies in the class; THE SUPREME COURT'S DECISION: unanimous 9-0 decision; THE COURT HELD: to enable a genus, the specification must enable POSITA to make and use the full scope of the claimed genus; the two guideposts for enablement are the quantity of experimentation required and the predictability of the field; for unpredictable fields (like protein biology), the burden of enablement is higher; providing only 26 working examples from a potentially vast genus of millions of antibodies with the claimed functional properties, plus a 'roadmap' for searching for additional antibodies, was NOT sufficient enablement; the court applied the Wands factors to reach this conclusion; WHAT AMGEN SAID ABOUT THE GENUS: the claimed genus — all antibodies with a specific binding/blocking function — could include millions of antibodies; the specification did not enable POSITA to make and use all those antibodies without undue experimentation; the roadmap of 'try the mutation method and screen for function' leaves POSITA to engage in the same extensive development work that Amgen itself did; IMPACT ON ANTIBODY PATENT STRATEGY: (1) PURELY FUNCTIONAL GENUS CLAIMS ARE VULNERABLE: broad 'any antibody that does X' claims without extensive enabling disclosure are now extremely difficult to maintain; (2) MORE REPRESENTATIVE EXAMPLES REQUIRED: the specification must provide representative examples that enable the full scope — not just examples at the extremes; (3) STRUCTURAL DEFINITION OF THE GENUS: defining the genus structurally (by CDR sequences or structural features common to the class) is safer than defining it by function alone; (4) COMBINATION CLAIMS: combining functional and structural limitations may narrow the genus to a degree that is actually enabled; WHAT AMGEN DID NOT DECIDE: the case did not invalidate all functional antibody claims; it did not say structural claims are always better; it requires that whatever is claimed must be enabled across the full scope.

CDR-based claiming is the primary alternative to purely functional claiming that provides meaningful scope while surviving post-Amgen enablement scrutiny: CDR STRUCTURE BASICS: each antibody has 6 CDRs (3 heavy chain: CDR-H1, CDR-H2, CDR-H3; 3 light chain: CDR-L1, CDR-L2, CDR-L3); the CDR3 of the heavy chain (CDR-H3) is the most variable and most important for antigen binding specificity; small changes in CDR sequences can eliminate binding or improve affinity; TYPICAL CDR CLAIMING STRATEGY: level 1 — claim specific CDR sequences of a lead antibody (SEQ ID NOs for CDR-H1 through CDR-L3); provides narrow but strong protection for the specific antibody; level 2 — claim CDR sequences with conservative substitution language (1-3 conservative amino acid substitutions in CDR-H1, CDR-H2; CDR-H3 specified exactly because it is most critical); level 3 — claim the VH and VL domain sequences (all 6 CDRs within the context of the framework regions); level 4 — claim classes of CDR sequences defined by common structural features or binding mode analysis; CONSERVATIVE AMINO ACID SUBSTITUTIONS: patent claims often include 'or conservative substitutions thereof'; conservative substitutions = replacing an amino acid with another of similar properties (e.g., hydrophobic for hydrophobic; charged for charged); the specification must show that conservative substitutions do not substantially alter binding or efficacy; if not shown in the specification, this language may not be enabled; HUMANIZED AND CHIMERIC ANTIBODY CLAIMS: chimeric antibodies (human constant regions + non-human variable regions) and humanized antibodies (human framework regions + donor CDRs) are common drug forms; claim the humanized version specifically (with human framework sequences identified); also claim the donor (typically murine) CDR sequences that are carried into the humanized antibody; SPECIFICATION DRAFTING BEST PRACTICES: (1) PROVIDE SEQUENCE LISTINGS: include a formal sequence listing (ST.25 or ST.26 format) with all antibody sequences; assign SEQ ID NOs to each sequence; (2) CHARACTERIZE MULTIPLE ANTIBODIES: isolate and characterize multiple lead antibodies (ideally 10-50 or more); include CDR sequences, VH/VL sequences, binding affinity data, functional data for each; (3) SHOW VARIANTS WORK: for each position where conservative substitutions are claimed, provide data showing variants retain activity; (4) STRUCTURAL DATA: if available, crystal structure data showing the binding epitope and contact residues provides strong support for structure-activity relationship claims; (5) CROSS-COMPETING ANTIBODY DATA: data showing antibodies that compete for the same epitope (competitive binding assays) can support broader claims about antibodies binding the same epitope; (6) DEFINE SCOPE STRUCTURALLY: define the genus by structural features common to all members, not just by the functional outcome.

Modern therapeutic antibodies come in many formats, each with distinct IP considerations: FULL-LENGTH MONOCLONAL ANTIBODIES (mAbs): IgG format (IgG1, IgG2, IgG4 most common for therapeutics); Fc region interacts with immune effector cells; Fc engineering (neonatal Fc receptor (FcRn) engineering for extended half-life; Fc region modifications to enhance or eliminate ADCC/CDC functions); CLAIMING: specific VH/VL sequences + specific Fc modifications; Fc engineering patents are distinct from binding domain patents; ANTIBODY-DRUG CONJUGATES (ADCs): antibody linked to a cytotoxic drug payload via a chemical linker; three components: antibody (targeting); linker (specific cleavable chemistry); payload (drug); CLAIMING: can claim the antibody-linker-payload combination; can claim the linker chemistry separately; can claim the conjugation method (site-specific vs. random conjugation); site-specific ADCs (conjugation at specific engineered cysteine residues or unnatural amino acids) have cleaner IP than random conjugation; BISPECIFIC ANTIBODIES: two different binding domains in one molecule; can bind two different antigens simultaneously or two different epitopes on the same antigen; FORMATS: knob-in-hole (engineered Fc heterodimer); CrossMab (domain crossover to prevent mispairing); BiTE (bispecific T cell engager: one anti-CD3 + one tumor antigen binding domain); CLAIMING: the specific two binding domain sequences + the bispecific format architecture; the design principle of bringing two specific cell types together can support method claims; ANTIBODY FRAGMENTS: Fab (antigen-binding fragment); scFv (single-chain variable fragment); Nanobodies (single domain antibodies from camelids; VHH domains); diabodies; CLAIMING: same VH/VL or VHH sequence strategies apply; single-domain antibodies (nanobodies) have simpler structure and simpler IP; BIOBETTERS: biosimilar competition is a major driver of antibody patent strategy; biobetters are improved versions of existing antibodies; CLAIMING BIOBETTERS: specific VH/VL sequences with improvement (different CDR combination); specific Fc modifications (LALA mutations to reduce effector function); extended half-life modifications; improved formulation (specific pH, excipients, concentration); FORMULATION PATENTS: pH; buffer; excipient; concentration; subcutaneous formulation (vs. IV); co-formulation with other drugs; these are often the last line of protection as the core antibody patent approaches expiration.