Patent lifecycle management in biotechnology

A patent’s nominal life is twenty years from filing, harmonised by Article 33 of the TRIPS Agreement. Almost nothing about a biotechnology patent’s commercial life follows from that number. Two independent constraints do: what the claim can lawfully cover, and how much of the term survives the years spent obtaining marketing authorisation.

Why a functional genus claim fails

The commercially valuable claim in biologics is a genus — not one antibody but every antibody that binds a target and blocks its function, not one sequence but every sequence sharing a stated identity. United States law tests such claims against two separate requirements in 35 U.S.C. §112(a). Enablement asks whether the specification teaches a skilled person to make and use the full scope claimed without undue experimentation. Written description asks something else, established as independent in Ariad Pharmaceuticals v. Eli Lilly, 598 F.3d 1336 (Fed. Cir. 2010) (en banc): whether the disclosure shows the inventor actually possessed what is claimed.

Amgen v. Sanofi, 598 U.S. 594 (2023), applied enablement to antibodies claimed by function — those binding a defined region of PCSK9 and blocking its interaction with the LDL receptor. Amgen had disclosed the amino acid sequences of twenty-six such antibodies and two screening roadmaps. The Supreme Court held the claims not enabled: the genus was potentially vast, and the roadmaps amounted to instructing others to repeat the original search by trial and error.

The reason this is a real constraint rather than a doctrinal preference is informational. A functional definition names a set whose membership can only be determined by testing. Sequence space is combinatorial — a claim to sequences with, say, ninety per cent identity over three hundred residues covers a number of variants beyond enumeration, the overwhelming majority of which do not fold or do not function. The specification cannot describe that set because the inventor never held it; the applicant found the functional members by screening, and the patent bargain requires teaching how to obtain them by something other than the same screen.

The practical consequence is that biologics patents narrow toward what was actually made and characterised — specified sequences, defined complementarity-determining regions, deposited material — and the broad blocking claim that portfolio strategy assumes is the least durable asset in it.

The term arithmetic

The second constraint is calendar. Clinical development and regulatory review consume patent term that began running at filing, so the exclusivity that matters is the remainder. Compensating regimes exist and are bounded: in the United States, 35 U.S.C. §156 restores part of the regulatory review period, capped at five years and at fourteen years of effective term from approval; in the European Union, the supplementary protection certificate under Regulation (EC) No 469/2009 grants at most five years, with a further six months available under the paediatric regulation. Both restore one patent per product, not a portfolio.

Term is also lost from inside. Continuation families that claim obvious variants of an earlier patent are exposed to obviousness-type double patenting, and the Federal Circuit held in In re Cellect (2023) that patent term adjustment does not shield a later-expiring family member from that doctrine — so a filing strategy built to stagger expiry dates can compress them instead.