May 24, 2023Client Alert

Amgen v. Sanofi – Enabling Inventions in the 21st Century

On May 18, 2023, the United States Supreme Court issued its decision in Amgen v. Sanofi, 598 U.S. ____, the Court’s latest interpretation of the enablement requirement of Section 112 of the Patent Act.  As the unanimous decision crafted by Justice Gorsuch explains, the Court has been interpreting enablement since the earliest days of the original Patent Act of 1790.  In every generation, revolutionary technology pushes the limits of  enablement.  Antibody technology has revolutionized the treatment of disease, but given the almost infinite variety of antibodies, unprecedented complexity results in defining the scope of claims based on function as opposed to structure.  Although a relatively straightforward application of enablement, Amgen is useful in understanding how to obtain broader patent claims in the life sciences field and how to apply enablement principles beyond the life sciences. 

In 2014, Amgen obtained two patents that broadly claimed antibodies that both bind to and block the activity of a protein known as PCKSK9.  Such antibodies are effective in controlling LDL (bad) cholesterol.  Amgen patented one such antibody used in its Repatha drug, and Sanofi, Amgen’s competitor, patented a different antibody marketed as Praluent.  But conflict ensued when Amgen asserted two additional patents with broader claims that included all antibodies that performed the functions of binding to and blocking the activity of PCKSK9. 

Attempting to enable the claims, Amgen’s patent specification identified the amino acid sequences of 26 qualifying antibodies and then described only two methods for identifying other such antibodies.  The first method, “the roadmap,” consisted of generating candidates that might have the desired properties and then testing the candidates to see which ones performed the desired functions.  The second method, “conservative substitution,” worked backward from antibodies known to perform the desired functions by selectively substituting amino acids in those antibodies with amino acids having similar properties and then testing the substituted antibodies to see if they performed the claimed functions. 

Both the trial court and the Court of Appeals for the Federal Circuit accepted Sanofi’s contention that the patents were not enabled.  The Supreme Court affirmed. 

Justice Gorsuch begins the analysis by noting that, from the earliest version of the Patent Act, it required the equivalent of “a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art . . . to make and use the same.”  

Justice Gorsuch then recounts the key enablement decisions over the years, beginning with O’Reilly v. Morse (1853).  His analysis is notable for at least two reasons.  First, it is an informative and entertaining history of enablement illustrative of Justice Gorsuch’s signature writing style.  Second, he makes a compelling case that, not only has enablement been part of the Patent Act from the very beginning, the wording and import of the requirement has been amazingly constant for over 200 years.  Each decision he describes recognized the importance of the enablement requirement to the patent bargain, and the Court has always required that the specification inform the public about how to identify and practice each embodiment within the scope of the claims, even if the specification does not necessarily describe every embodiment.  In other words, the patent specification must enable a person skilled in the art to make and use the entire class of claimed compounds. 

After laying this foundation, Justice Gorsuch unequivocally rejects that Amgen’s patents enabled all antibodies that bind to and block the PCKSK9 protein.  The 26 antibodies sequenced in the patents were an infinitesimal sample of the millions of potential antibodies that contained no common characteristic that distinguished them.  The two methods of identifying antibodies within the scope of the claims were nothing more than an invitation to engage in the same trial and error that gave rise to Repatha and Praulent.  Justice Gorsuch then dismissed Amgen’s contentions that the Federal Circuit had improperly raised the bar for enablement in a way to stymie innovation. 

So where does the decision leave an inventor seeking to obtain broad coverage for their life sciences innovation?  There are two key criteria that an inventor can identify to protect a broad range of embodiments using functional claim language.  First, the inventor needs to find a common characteristic in the compounds they wish to protect.  That characteristic must be objective and evidence-based and must allow others to distinguish the protected compounds.  Examples might include a critical subsequence of amino acids or motifs that provide the desired functionality.  To the extent possible, the significance of this functionality should be illustrated in examples in the specification.  This requires inventors as they are developing their technology to intensify their focus on factors beyond functionality such as mechanisms of action and structure. 

Second, to the extent inventors rely on methods to identify embodiments within the scope of the claims, the methods must go beyond the trial and error that gave rise to the original embodiments.  The methods must identify surrogate characteristics that are readily identified and provide a basis for concluding that, if the surrogate characteristic is present, the functionality will follow.  In this sense, the analysis folds into the first criteria, namely, identifying a common characteristic that distinguishes the claimed embodiment.  The methods must be straightforward ways of measuring characteristics that are narrower than the ultimate function of the invention. 

Reading Amgen more broadly provides useful guidance for protecting non-life science inventions, which have recently been more vulnerable to enablement rejections from the Patent Office.  The common thread is identifying the characteristic that distinguishes embodiments within the scope of broad claims and providing examples and other content within the specification to demonstrate what elements of structure give rise to such characteristics.  For example, in an electrical invention, the specification can identify what aspects of a circuit or other components of the invention generate functionality and how that functionality arises from those elements.  In a mechanical invention, the specification can describe what structure creates the desired functionality and how other structures create such functionality. 

In this age of fractious, contentious Supreme Court decisions outside the patent realm, Amgen is a comforting reminder that some areas of the law are more consistent over the centuries and that principles of reason still can provide a way to apply old doctrines to modern developments that the drafters of the first Patent Act could never have imagined. 

back to top