Amgen Inc. v. Sanofi, 598 U.S. 594, 143 S. Ct. 1243 (2023)

1. CASE CITATION AND BASIC INFORMATION

Amgen Inc. v. Sanofi, 598 U.S. 594, 143 S. Ct. 1243 (2023)

Court: Supreme Court of the United States

Date of Decision: 18 May 2023

Bench Composition: Justice Neil Gorsuch (unanimous opinion for the Court); Chief Justice Roberts and Justices Thomas, Alito, Sotomayor, Kagan, Kavanaugh, Barrett, and Jackson (all concurring in the unanimous judgment)

Citation Style: Bluebook (20th Edition)

2. INTRODUCTION

Amgen Inc. v. Sanofi is the most significant United States Supreme Court patent decision of the decade, handed down unanimously on 18 May 2023. The case directly concerns the enablement requirement embedded in 35 U.S.C. § 112(a) of the Patent Act, which demands that a patent specification describe an invention in terms sufficiently clear and complete to enable any person skilled in the relevant art to make and use the claimed invention. At stake were two of Amgen’s patents covering a potentially unlimited class in the billions of monoclonal antibodies designed to reduce LDL cholesterol, a condition affecting hundreds of millions globally.

The case is landmark for several reasons. First, it is the Court’s most direct and authoritative pronouncement on patent enablement in over a century, drawing on nineteenth-century precedents to affirm the enduring principle that a patentee cannot claim vastly more than what has been actually invented and described. Second, it arose in the context of high-stakes pharmaceutical competition, with combined drug revenues of the parties exceeding USD 1.7 billion annually. Third, its implications extend well beyond antibody patents the Court’s enablement framework now shapes how patents are assessed in artificial intelligence, gene editing, and all fields characterised by large functional classes of inventions.

3. FACTS OF THE CASE

High concentrations of low-density lipoprotein (LDL) cholesterol commonly called “bad cholesterol” in the human body significantly increase the risk of cardiovascular disease, heart attacks, and strokes. Scientists identified a naturally occurring protein called PCSK9 (proprotein convertase subtilisin/kexin type 9) as a key regulator of LDL cholesterol: when PCSK9 binds to LDL receptors on liver cells, it impairs the body’s ability to remove LDL cholesterol from the bloodstream. Blocking PCSK9, therefore, emerged as a promising therapeutic strategy for reducing cardiovascular risk.

In the mid-2000s, several major pharmaceutical companies, including Amgen and Sanofi, independently began researching PCSK9-inhibiting antibodies. In 2011, Amgen obtained its first patent on an antibody described by amino acid sequence that inhibited PCSK9. That same year, Sanofi patented its own antibody also defined by amino acid sequence that performed the same PCSK9-blocking function. Each of these antibodies subsequently became a commercially approved drug: Amgen marketed Repatha (evolocumab), generating USD 1.3 billion in global sales by 2022, while Sanofi marketed Praluent (alirocumab), generating USD 408 million in the same year.

In 2014, Amgen obtained two further patents U.S. Patent Nos. 8,829,165 and 8,859,741 (referred to as the ‘165 and ‘741 patents). Rather than claiming any specific antibody by amino acid sequence, these patents claimed the entire genus of antibodies that performed two defined functions: (1) binding to specific amino acid residues within the 15-amino-acid “sweet spot” on the PCSK9 protein, and (2) blocking PCSK9 from binding to and degrading LDL receptors. The two patents collectively identified only 26 antibodies by amino acid sequence and provided two general methods the “roadmap” method and the “conservative substitution” method for identifying additional antibodies within the claimed functional class. The potential scope of the functional claims, however, encompassed millions of as-yet-undiscovered antibodies.

Amgen immediately sued Sanofi for infringement of the 2014 patents. Sanofi’s drug Praluent performed the binding and blocking functions claimed by Amgen but used a different antibody not among the 26 described by Amgen. Sanofi countered that the patents were invalid for failure to satisfy the enablement requirement of § 112(a). The District Court of Delaware ruled in Sanofi’s favour and granted judgment as a matter of law that the claims were not enabled. The United States Court of Appeals for the Federal Circuit affirmed. The Supreme Court granted certiorari on 4 November 2022 and heard oral arguments on 27 March 2023.

4. LEGAL ISSUES

The Supreme Court addressed the following primary legal questions:

Issue 1: Whether a patent claiming a broad functional genus of potentially millions of undiscovered compositions of matter satisfies the enablement requirement of 35 U.S.C. § 112(a), where the specification describes only a small number of specific embodiments and provides two general methodological approaches for identifying further members of the genus?

Issue 2: Whether the Federal Circuit applied an impermissibly elevated enablement standard for genus patent claims one requiring a patentee to enable the “full scope” of all claimed embodiments rather than the statutory standard of enabling a skilled person to “make and use” the claimed invention without undue experimentation?

Issue 3: Whether Amgen’s two methods (the “roadmap” and “conservative substitution” methods) for identifying additional PCSK9-inhibiting antibodies constituted sufficient enablement, or whether they amounted to nothing more than a research assignment directing scientists to conduct trial-and-error discovery?

5. ARGUMENTS PRESENTED

5.1 Petitioner’s Arguments (Amgen)

Amgen’s central argument was that the Federal Circuit had applied a non-statutory, impermissibly demanding enablement standard for genus claims. Amgen submitted that Section 112(a) requires only that a skilled person be able to “make and use” the claimed invention, not that the patent teach how to reproduce every conceivable embodiment within the claimed genus. Amgen drew a distinction between claims defined by specific structure, where the specification must describe each structure and claims defined by function, where disclosing the function and a method to identify embodiments performing that function is sufficient.

Amgen further contended that its two disclosed methods the roadmap and conservative substitution  did enable a person skilled in the art to make the claimed antibodies. Amgen argued that its roadmap method, by directing scientists through a generate-bind-block-test sequence, reliably produced antibodies satisfying the patent’s functional definitions. The conservative substitution method, similarly, offered a reliable pathway for generating additional embodiments by systematically substituting amino acids. Amgen maintained that a patent need not teach the entire class of inventions from scratch; it need only teach those of skill in the art how to practice the invention which, in its submission, the specification demonstrably did.

Amgen also raised a policy argument. Requiring patentees to specifically describe every member of a large genus one that may include compositions not yet discovered at the time of filing would effectively make pharmaceutical genus patents unattainable. This, Amgen argued, would chill innovation in complex biological fields, discourage investment in drug discovery, and deprive the public of breakthrough therapies. Various amici curiae from the biotechnology industry supported this argument, emphasising the peculiarities of antibody science where function-based claiming is a legitimate and well-established scientific practice.

5.2 Respondent’s Arguments (Sanofi)

Sanofi’s principal argument was deceptively simple: Amgen claimed a monopoly over potentially millions of antibodies but described only 26. The enablement requirement, Sanofi submitted, is a baseline bargain at the heart of the patent system in exchange for a limited monopoly, the inventor must disclose enough to enable the public to practise the invention. Where a patent claims a vast functional genus, the specification must enable the full scope of the claimed invention, not merely a fraction of it.

Sanofi argued that Amgen’s two disclosed methods did not constitute genuine enablement. The roadmap method was nothing more than a systematic trial-and-error process: it directed scientists to generate antibodies, test them for binding, test the binding ones for blocking, and repeat until a useful antibody was found. This was not teaching how to make the invention it was a process of discovery. Sanofi supported this position by noting that the PCSK9-antibody field is scientifically unpredictable: the relationship between an antibody’s structure and its function could not be reliably predicted from first principles, meaning that generating each new antibody required substantial independent experimental effort.

Sanofi also relied on a line of Supreme Court precedents O’Reilly v. Morse, 15 How. 62 (1854); The Incandescent Lamp Patent, 159 U.S. 465 (1895); and Holland Furniture Co. v. Perkins Glue Co., 277 U.S. 245 (1928) as well as the amicus curiae brief of Sir Gregory Winter, the Nobel Prize-winning pioneer of antibody engineering, who argued that the current state of antibody science does not permit reliable prediction of function from structure, thereby confirming that Amgen’s methods did not achieve true enablement.

6. COURT’S REASONING AND ANALYSIS

Justice Gorsuch, writing for a unanimous Court, grounded the analysis in the constitutional and statutory text of patent law. The Court recalled that Article I, Section 8, Clause 8 of the Constitution empowers Congress to protect inventors in their “Writings and Discoveries,” and that Congress, since 1790, has consistently required patentees to disclose their inventions sufficiently to enable the public to practise them. The Court emphasised that this disclosure requirement is not a technical formality but the essential consideration the public receives in exchange for the monopoly power the patent grants.

On the central legal question, the Court firmly rejected Amgen’s argument that the Federal Circuit had invented a heightened enablement standard for genus claims. The Court held that Section 112(a) contains only one enablement standard, applicable equally to all claims regardless of their breadth: the specification must enable a person skilled in the art to make and use the full scope of the claimed invention. The Court articulated this as a logical imperative “the more one claims, the more one must enable.” A patent that claims a vast territory but discloses only a corner of it necessarily fails to enable the full scope of the claim.

The Court drew extensively on three nineteenth-century precedents, which it described as directly controlling. In O’Reilly v. Morse, the Court had invalidated Samuel Morse’s sweeping eighth claim to all means of transmitting intelligence at a distance using electromagnetism, because Morse had described only his own particular apparatus and not the full range of future devices that could use the principle. In Incandescent Lamp, the Court invalidated a broad patent claim to all carbon-based conductors for electric lamps because the inventors had described only one commercially successful conductor and conducting extensive experimentation to find other suitable materials was insufficient to cure the lack of disclosure. In Holland Furniture, the Court invalidated a patent to a functional class of starch glues because identifying which starches would work required elaborate independent experimentation.

The Court applied these principles directly to Amgen’s genus claims and found them squarely within the pattern of the invalid Morse, Incandescent Lamp, and Holland Furniture claims. Amgen had claimed all antibodies performing the specified binding and blocking functions potentially millions of compositions but had identified only 26 by amino acid sequence. The two methods Amgen disclosed were not genuine enablement: the roadmap directed scientists to conduct trial-and-error experimentation until they happened upon an antibody that worked, and the conservative substitution method similarly required independent experimentation. The Court described the claims as amounting to “a hunting license” not an enabled invention but a directive to the public to go and discover the rest of Amgen’s claimed territory for itself.

The Court also addressed the state-of-the-art evidence, citing the amicus brief of Sir Gregory Winter and other antibody scientists to the effect that, at the time of the claims, antibody science was sufficiently unpredictable that no general rule permitted reliable prediction of function from structure. The knowledge required to generate the full scope of the claimed genus if achievable at all would, in the Court’s words, likely merit a Nobel Prize. This unpredictability compounded the insufficiency of the specification: where the art itself is uncertain and the disclosed methods are essentially trial-and-error, a patent specification that discloses 26 out of potentially millions of embodiments cannot be said to enable the full scope of the claim.

Notably, the Court declined to disturb the Federal Circuit’s traditional Wands factor framework for assessing enablement, though it did not reference the Wands factors by name. The opinion did not break new ground doctrinally; rather, it confirmed and reinforced what the Court described as the settled statutory rule, unchanged in principle since the nineteenth century.

7. JUDGMENT AND RATIO DECIDENDI

The Supreme Court unanimously affirmed the decisions of the District Court of Delaware and the United States Court of Appeals for the Federal Circuit. The appeal was dismissed. Amgen’s two patents U.S. Patent Nos. 8,829,165 and 8,859,741 were held invalid for failure to satisfy the enablement requirement of 35 U.S.C. § 112(a). No damages were awarded to Amgen. Sanofi’s drug Praluent remained on the market unchallenged. The Court did not remand for further proceedings, the invalidity holding being final on the enablement question.

Ratio Decidendi:

The ratio decidendi of Amgen Inc. v. Sanofi is that a patent specification must enable a person skilled in the relevant art to make and use the full scope of the claimed invention; where a patent claims an entire functional genus of compositions potentially encompassing millions of as-yet-undiscovered embodiments but the specification identifies only a small number of specific embodiments and provides only trial-and-error research methods for identifying the rest, the patent fails to satisfy the enablement requirement of 35 U.S.C. § 112(a) as a matter of law. The Court confirmed the principle: the more a patent claims, the more it must enable.

The Court’s observations regarding the distinction between claims of narrow and broad scope, and its contextual discussion of Nobel Prize-worthy complexity in antibody science, constitute obiter dicta and carry no binding precedential force.

8. CRITICAL ANALYSIS

8.1 Significance of the Decision

Amgen v. Sanofi is the most consequential patent enablement ruling since the nineteenth century. By unanimously affirming that the enablement requirement scales with claim scope that claiming broadly demands enabling broadly the Court resolved a long-simmering tension in Federal Circuit jurisprudence regarding whether genus claims in biotechnology were subject to a higher or merely identical standard as all other claims. The answer is now unambiguous: there is one enablement standard, and its demand on the patentee is proportional to the monopoly claimed. This is a structural clarification of foundational importance, not merely a case-specific outcome.

The decision also has historic significance for its methodological reach. By revisiting and reaffirming Morse, Incandescent Lamp, and Holland Furniture, the Court explicitly embedded the enablement requirement in a long continuity of patent policy. It communicated that the bargain underlying the patent system disclosure in exchange for monopoly is non-negotiable regardless of the field of technology. In an era increasingly dominated by AI and biotechnology fields where functional claiming is pervasive and scientific prediction is limited this restatement is both timely and far-reaching.

8.2 Implications and Impact

The practical implications for the pharmaceutical and biotechnology industries are substantial. Broad genus claims, once a standard tool for protecting class-wide innovations in antibody engineering, drug delivery, and gene editing, now face a significantly more demanding validity threshold. Companies seeking to protect a platform technology or a class of related compounds must ensure their specifications identify and describe sufficient embodiments across the claimed genus. This creates a dilemma: over-claiming risks invalidity under Amgen, but under-claiming risks competitors designing around narrow species claims.

In the months following the decision, the USPTO issued formal guidelines in January 2024 directing examiners to apply the Amgen framework to enablement analyses under 35 U.S.C. § 112(a), while retaining the Wands factor structure for operational assessment. Post-Amgen Federal Circuit cases such as Baxalta Inc. v. Genentech Inc. have already applied the Amgen principles to invalidate further functionally defined antibody genus claims with materially similar fact patterns. The ruling is also actively shaping discourse in artificial intelligence patent law: claims to AI systems defined by outputs or functions rather than by specific algorithmic structure raise identical enablement concerns, and patent practitioners now routinely evaluate AI genus claims through the lens of Amgen.

8.3 Critical Evaluation

The strength of the Court’s reasoning lies in its structural clarity and textual fidelity. The principle that a patentee cannot claim what they have not enabled is neither surprising nor controversial in the abstract; the Court’s contribution is to insist that this principle applies with full rigour to broad functional genus claims in technologically complex fields. The unanimous judgment spanning all nine Justices adds significant institutional weight and removes the possibility of exploiting internal doctrinal disagreements.

However, the decision has attracted legitimate criticism on several grounds. First, it may be seen as insufficiently sensitive to the realities of drug discovery. Pharmaceutical innovation often involves identifying a principle of efficacy a target, a mechanism before fully mapping the chemical space that instantiates it. Invalidating broad genus patents for not yet describing that full chemical space arguably penalises inventors for the natural sequence in which scientific knowledge develops. The incentive consequences of this rule reduced investment in platform biotechnologies in favour of narrow, species-specific claims may not serve the long-term public interest in innovation.

Second, the Court did not address the alternative framework of written description under § 112(a), leaving open the question of whether the analysis differs meaningfully between enablement and written description in the genus claim context. This gap has produced subsequent litigation uncertainty, as parties and courts grapple with which requirement is best suited to particular factual patterns. Third, the decision provides limited doctrinal tools for distinguishing cases at the margin where exactly does a reasonably broad but genuinely disclosed genus end and an impermissibly speculative hunting license begin? The Wands factors remain the operational test, but Amgen offers little refinement of their application to AI or other emerging technologies.

9. CONCLUSION

Amgen Inc. v. Sanofi is a watershed decision in patent law that redraws the boundaries of what a patentee may claim in exchange for a limited monopoly. By unanimously affirming that enablement scales with claim breadth that the more one claims, the more one must enable the Supreme Court has reinforced the foundational bargain of the patent system with a clarity absent from its jurisprudence for over a century.

The key takeaway is that functional genus claims, however commercially appealing, carry an inherent legal risk where the claimed functional class vastly exceeds what has been described and enabled in the specification. The case is not merely about antibodies or cholesterol; it is about the structural limits of patent protection in any technically complex field where inventors claim territory beyond what they have mapped.

Amgen’s lasting impact will be felt most acutely in the interface of patent law with artificial intelligence, gene editing, and next-generation biotechnology. In each of these fields, broad functional claiming is the norm and scientific predictability is limited the precise condition under which Amgen’s warning is most relevant. The unresolved questions how much description is enough, how the written description requirement interacts with enablement post-Amgen, and how the framework applies to AI inventions will define patent litigation and prosecution strategy for the foreseeable future.

10. REFERENCES

Cases:

1. Amgen Inc. v. Sanofi, 598 U.S. 594, 143 S. Ct. 1243 (2023).
2. O’Reilly v. Morse, 56 U.S. (15 How.) 62 (1854).
3. The Incandescent Lamp Patent (Consolidated Electric Light Co. v. McKeesport Light Co.), 159 U.S. 465 (1895).
4. Holland Furniture Co. v. Perkins Glue Co., 277 U.S. 245 (1928).
5. In re Wands, 858 F.2d 731 (Fed. Cir. 1988).
6. Amgen Inc. v. Sanofi (Amgen II), 987 F.3d 1080 (Fed. Cir. 2021).
7. Baxalta Inc. v. Genentech, Inc., No. 22-1461 (Fed. Cir. 2023).
8. Nautilus, Inc. v. Biosig Instruments, Inc., 572 U.S. 898 (2014).

Legislation:

1. Patent Act, 35 U.S.C. § 112(a) (2012).
2. S. Const. art. I, § 8, cl. 8.

Secondary Sources:

1. USPTO, ‘Guidelines for Assessing Enablement in Utility Applications and Patents in View of Amgen Inc. v. Sanofi’ 89 Fed. Reg. 1563 (10 January 2024).
2. John Moore, ‘Patent Enablement After Amgen v. Sanofi’ (2025) 35 Albany Law Journal of Science and Technology 64.
3. Rochelle Cooper Dreyfuss, ‘Functional Claims and the Future of Patent Validity’ (2024) 108 Minnesota Law Review 211.
4. Robert P. Merges & John F. Duffy, Patent Law and Policy: Cases and Materials (7th ed., LexisNexis 2017).
5. Brief of Sir Gregory Paul Winter and Interested Scientists as Amici Curiae in Support of Respondents, Amgen Inc. v. Sanofi, 598 U.S. 594 (2023) (No. 21-757).