Biomolecule Patents

Association for Molecular Pathology v. Myriad Genetics, Inc. (S.Ct. 2013) fundamentally changed the landscape for DNA and genomic sequence patents: THE CASE BACKGROUND: Myriad Genetics had patents on the BRCA1 and BRCA2 gene sequences — genes associated with significantly elevated risk of breast and ovarian cancer; Myriad's patents covered: isolated naturally occurring BRCA1 and BRCA2 DNA sequences; cDNA versions of these genes; methods of diagnosing cancer risk by analyzing the BRCA genes; THE SUPREME COURT DECISION: the Court held 9-0 that: NATURALLY OCCURRING GENOMIC DNA IS NOT PATENTABLE even when isolated from the genome; isolation alone does not transform a naturally occurring product into a patent-eligible invention; rationale: the nucleotide sequence is determined by nature, not by the inventor; the isolated DNA has the same information-bearing sequence as the naturally occurring chromosomal DNA; THE cDNA EXCEPTION: the Court specifically held that cDNA (complementary DNA) CAN be patent-eligible if it differs from naturally occurring sequences; cDNA is synthesized in the laboratory from messenger RNA (mRNA) using reverse transcriptase; cDNA contains only the exon sequences (the protein-coding regions) — the intron sequences (non-coding regions present in genomic DNA) are absent; because human cells do not naturally produce cDNA, and cDNA does not exist naturally, it is NOT a product of nature; HOWEVER: if a naturally occurring gene happens to have no introns (as some short genes do), the cDNA would be identical to the genomic DNA and would NOT be patent-eligible; THE METHODS: the Court did NOT address method claims in Myriad (they were dismissed on other procedural grounds); methods of using gene sequences — such as diagnostic methods — remain potentially patentable subject to Mayo/Alice analysis; THE BROADER SIGNIFICANCE: Myriad invalidated thousands of previously issued DNA sequence patents covering the human genome; gene testing companies could now offer BRCA testing without Myriad licenses; it clarified the natural phenomena exception for biological molecules; THE MARKEDLY DIFFERENT TEST: for biomolecule patents generally, the key inquiry is whether the claimed molecule has markedly different characteristics from what exists in nature; isolation alone is not sufficient — there must be a functional or structural difference.

Myriad established a framework for thinking about nucleic acid patent eligibility that continues to shape biotechnology patents: CLEARLY PATENTABLE NUCLEIC ACIDS: (1) cDNA SEQUENCES: as confirmed in Myriad, cDNA with exon-only sequences that differ from any naturally occurring genomic sequence is patent-eligible; the patent must disclose a specific cDNA sequence, not merely claim an abstract class; (2) SYNTHETIC/ARTIFICIAL NUCLEIC ACIDS: nucleic acids with: modified backbones (phosphorothioate; phosphorodithioate; peptide nucleic acid (PNA)); unnatural nucleotides (not A, T, G, C/U); locked nucleic acids (LNA); other non-natural chemical modifications; these are not products of nature because they do not exist in nature — they are chemical entities created by human ingenuity; ANTISENSE OLIGONUCLEOTIDES: antisense drugs (which hybridize to and silence mRNA) often use phosphorothioate-modified backbones — these are synthetic and not naturally occurring; the chemical modification is sufficient to distinguish them from natural nucleic acids; siRNA AND miRNA APPLICATIONS: naturally occurring siRNA and miRNA sequences may not be patentable as compositions of matter (they exist in nature); however: specific synthetic siRNA constructs with chemical modifications; specific delivery system combinations; specific targeting sequences selected from among many possibilities; methods of using these for therapeutic purposes with specific technical steps; — these may be patentable; mRNA VACCINES AND THERAPEUTICS: the mRNA in COVID-19 vaccines (Moderna, Pfizer/BioNTech) included modified nucleosides (N1-methylpseudouridine substitutions) that are not naturally present in mRNA; this modification made the mRNA more stable and less immunogenic; such chemically modified mRNA is a synthetic molecule with markedly different characteristics from naturally occurring mRNA — likely patent-eligible; the specific sequence design choices and lipid nanoparticle delivery systems also add patentable subject matter; WHAT REMAINS PROBLEMATIC: genomic DNA sequences even when isolated from the body; naturally occurring siRNA sequences without modification; naturally occurring gene expression patterns used as diagnostics; correlations between naturally occurring genetic variants and disease without inventive technical application.

The intersection of biomolecule discovery and diagnostic patent law creates significant patent eligibility challenges under Mayo and Alice: THE MAYO DECISION: Mayo Collaborative Services v. Prometheus Laboratories, Inc. (S.Ct. 2012): Prometheus had patents on methods of optimizing thiopurine drug dosing by measuring metabolite levels; claim steps included: (1) administering thiopurine; (2) measuring metabolite levels; (3) 'wherein' a high level indicates a need to decrease the drug and a low level indicates a need to increase; THE COURT'S ANALYSIS: the correlation between metabolite levels and drug efficacy was a LAW OF NATURE; the additional steps (administering, measuring) were conventional/routine; adding routine steps to a natural correlation does not create patent-eligible subject matter; the claims essentially attempted to patent the natural phenomenon itself; THE MAYO TEST FOR DIAGNOSTIC CLAIMS: (1) is the claim directed to a law of nature, natural phenomenon, or abstract idea?; if yes: (2) does the claim add an inventive concept — something significantly more than routine, conventional, well-understood steps?; if the additional steps are routine, conventional, and well-understood, the claim is not eligible; WHAT MAKES DIAGNOSTIC CLAIMS FAIL: claims that: identify a natural correlation between a biomarker and a condition; specify only generic, routine measurement steps (well-known assays); use only conventional data analysis; essentially claim the natural relationship itself by preempting the correlation; STRATEGIES FOR DIAGNOSTIC PATENT ELIGIBILITY: (1) CLAIM SPECIFIC NOVEL TECHNICAL METHODS: not just 'measuring the level of biomarker X'; claim a specific novel measurement technique with technical steps that are not routine; (2) CLAIM SPECIFIC THRESHOLD VALUES WITH EVIDENCE: if a specific threshold is non-obvious and not previously known, the threshold selection may be an inventive concept; (3) CLAIM COMBINATIONS OF BIOMARKERS: a multi-biomarker combination that does not exist as a natural phenomenon may be more defensible; (4) CLAIM SPECIFIC TECHNICAL APPLICATIONS: methods that include specific treatment steps (not just observing the correlation) tied to clinically actionable results; (5) TECHNICAL SOLUTION TO A TECHNICAL PROBLEM: frame the claim as solving a specific technical problem (e.g., reducing false-positive rates) using novel technical means; NEXT-GENERATION SEQUENCING DIAGNOSTICS: diagnostic methods based on whole-genome sequencing require specific bioinformatic tools; these tools may add enough inventive technical content to satisfy Mayo; claim the bioinformatic algorithm and its specific non-obvious implementation alongside the diagnostic correlation.

Proteins and antibodies occupy a complex space in biomolecule patent law — natural proteins have eligibility issues while engineered proteins and antibodies raise written description and enablement challenges: NATURAL PROTEINS — ELIGIBILITY ANALYSIS: naturally occurring proteins (e.g., insulin, erythropoietin, growth hormone as they exist in the body) are products of nature and are NOT patentable as isolated naturally occurring proteins; however: MODIFIED PROTEINS: proteins with non-natural amino acid modifications (substitutions; additions; deletions; non-standard amino acids) that change the functional or structural characteristics; FUSION PROTEINS: proteins combining domains from different proteins; GLYCOSYLATED PROTEINS: specific glycosylation patterns not found in nature; these can be patentable as compositions of matter with markedly different characteristics; THE ANTIBODY PATENT LANDSCAPE: antibodies (monoclonal antibodies; bispecific antibodies; antibody-drug conjugates (ADCs)) are a major category of pharmaceutical patents; ANTIBODY CLAIMING STRATEGIES: (1) STRUCTURAL CLAIMS: specific CDR (complementary determining region) sequences; specific variable domain sequences; specific constant region modifications (Fc engineering for enhanced effector function); (2) FUNCTIONAL CLAIMS: claims based on binding activity (binds to epitope X with Kd ≤ 1 nM) — subject to written description and enablement challenges (Amgen v. Sanofi); (3) COMBINATION CLAIMS: antibody + conjugate (ADC); antibody + formulation; WRITTEN DESCRIPTION AND ENABLEMENT FOR ANTIBODIES: Amgen Inc. v. Sanofi, Aventisub (S.Ct. 2023): the Supreme Court held that Amgen's genus claims to antibodies that bind PCSK9 were not enabled because the specification did not teach how to make and use the full range of antibodies claimed; purely functional claiming without adequate structural disclosure cannot support broad genus claims; practitioners must provide: representative antibody examples across the claimed genus; structural features (CDR sequences; variable domain sequences); methods for producing additional antibodies in the claimed class; experimental evidence that the claimed functional properties are achievable across the genus; PROTEIN ENGINEERING PATENTS: proteins engineered through directed evolution (Nobel Prize 2018 to Frances Arnold); these are clearly novel human creations with markedly different characteristics; the evolved sequence selection and specific functional properties add inventive content.