“Individualised Disclosure” Re-defined: The Court of Appeal’s Clarification on List-Based Prior Art in Modernatx Inc v Pfizer Ltd & Ors [2025] EWCA Civ 1032

1. Introduction

The mRNA-vaccine wars reached the Court of Appeal (Civil Division) in 2025. Moderna (Modernatx Inc) alleged that Pfizer/BioNTech’s flagship COVID-19 vaccine Comirnaty infringed its European Patent (UK) 3 590 949 (“EP 949”) relating to 100 % substitution of uridine with N1-methyl-pseudouridine (m¹Ψ) in mRNA. Meade J had found EP 949 valid and infringed. Pfizer/BioNTech (“PBNT”) appealed on validity, asserting lack of novelty and obviousness over a University of Pennsylvania international application (“UPenn”; WO 2007/024708).

LJ Arnold (with LJ Moylan and LJ Snowden concurring) dismissed the appeal, handing down a decision that does far more than affirm Moderna’s patent: it clarifies the “individualised disclosure” test where prior art supplies lists or classes of alternatives, and resets the approach to the skilled person’s motivation and expectation of success in experimental biotechnology.

2. Summary of the Judgment

• Novelty: UPenn, although brimming with lists of modified nucleosides and percentages of substitution, did not “individualise” the combination of m¹Ψ and 100 % uridine replacement in mRNA. Consequently, claim 3 of EP 949 remained novel.
• Obviousness: The skilled person—defined as an RNA biologist interested in practical applications—would have focused on using Ψ itself, not embarked on a “blind” screen of every Ψ-like analogue. The putative experimentation lacked concrete expectation of success, rendering claim 3 inventive.
• Expert evidence: The Court endorsed a nuanced evaluation of expert witnesses: their distance from the patent’s “practical” field and presence of hindsight devalues their contribution on obviousness.
• Key doctrinal clarification: Lists (or classes) in prior art are not reservoirs from which features may be plucked at will. A disclosure only anticipates if it clearly and unambiguously directs the skilled person to the specific combination later claimed.

3. Detailed Analysis

3.1 Precedents Cited and Their Influence

1. Dr Reddy’s v Eli Lilly [2009] EWCA Civ 1362
   – Established the need for an “individualised description” for selection inventions. Arn­old LJ relied heavily on Jacob LJ’s exposition that simply naming a forest does not disclose every leaf within it.
2. SmithKline Beecham (Paroxetine) [2006] RPC 10 (HL)
   – Lord Hoffmann’s “flag has been planted” test for anticipation; reiterated that enablement matters only after a clear disclosure is located.
3. EPO Board of Appeal authorities – Hoechst/Enantiomers T 296/87, Spiro Compounds T 181/82, Di­astereomers T 12/81, Novartis/Antidiabetic T 783/09, and KCI Wound Dressings T 1878/19. These decisions distinguish between a generic class and a specific teaching, and consider combinations from multiple lists. Arnold LJ synthesised them into English law, confirming there is no rigid numerical rule—the question is always one of individualisation.
4. Actavis v ICOS [2019] UKSC 15
   – Provided the Supreme Court’s multifactorial checklist for obviousness, especially the “obvious to try” doctrine. The Court of Appeal applied this, stressing expectation of success and practical motivation.
5. Technip’s Patent [2004] RPC 46 & Mölnlycke [1994] RPC 49
   – Guidance on the role of expert evidence; cited to emphasise that substance of reasoning, not mere credentials, drives weight.

3.2 Court’s Legal Reasoning

a. Novelty – “Route 3” and “Individualised Disclosure”

PBNT’s principal novelty attack (“Route 3”) cherry-picked two separate lists in UPenn: (i) a definition of “pseudouridine” in ¶ [0056] (listing m¹Ψ among others) and (ii) a menu of substitution percentages in ¶ [0074] (including 100 %). Arnold LJ held:

• ¶ [0056] is only a definition—a legal “basket clause” capturing future variants—not a technical teaching preferring m¹Ψ.
• ¶ [0074] gives 33 percentages without signalling 100 % as preferred.
• Selecting m¹Ψ and 100 % replacement is a two-list combination without any pointer; prior art must furnish more than an “obvious” combination—it needs a clear and unambiguous one.

b. Obviousness – Motivation & Expectation of Success

Key findings defeating PBNT’s obviousness case:

• Skilled person: Interested in practical application (vaccination, immunotherapy, etc.), not a pure “basic” scientist on a curiosity-driven quest.
• Ψ’s success steers focus inward: Because UPenn never explained why Ψ worked, the logical next step was to test Ψ in one’s own system, not to embark on a broad analogue screen.
• Blind screening = no expectation: Any modification could yield “catastrophic” loss of function; there was no rational basis to expect m¹Ψ would retain or enhance the properties of Ψ.
• Effort vs reward: In vitro and animal testing were non-trivial; without a tangible prospect of success, the skilled person would not commit resources.

c. Treatment of Expert Evidence

• Dr Enright (Pfizer/BioNTech) was “a pure, basic scientist”. The Court found hindsight seeped into his focus on m¹Ψ and discounting of other analogues, partly due to his knowledge of the litigation target.
• Prof Rosenecker (Moderna) better represented the “practical” skilled person; his concession that one “could” test analogues did not equate to a conclusion that it was obvious to do so.

3.3 Impact of the Judgment

• Patent drafting & litigation: Parties relying on expansive lists or Markush formulae must ensure specific embodiments are clearly sign-posted. Otherwise, later patents that “pluck” specific items plus additional conditions may survive.
• Pharmaceutical R&D: Confirms that routine “variant hunting” is not always obvious; absence of mechanistic rationale and expectation of success can uphold patents on fairly small molecular tweaks.
• Expert selection: Litigants should align experts with the patent’s practical field; courts will discount testimony skewed by hindsight or too remote from application-oriented research.
• mRNA technology space: Moderna retains a potent weapon against competitors using m¹Ψ-based platforms, potentially influencing licence negotiations and freedom-to-operate analyses.

4. Complex Concepts Simplified

“Individualised Disclosure”

Imagine a cookbook listing “any fruit” plus a table of sugar quantities from 1 g to 100 g. The book does not specifically teach “100 g sugar with raspberries”. Likewise, UPenn’s lists didn’t teach “m¹Ψ at 100 % substitution”.

Markush Formula / Class vs List

A Markush claim: “R-COOH where R is C_1-6 alkyl”. That covers millions of molecules but doesn’t spotlight “n-propyl-COOH”. Whether expressed as a chemical diagram (class) or bullet list (items), the question is: has the document singled out the later-claimed molecule?

Skilled Person’s Motivation

Patent law’s “skilled person” is pragmatic. They conduct experiments with a purpose and reasonable hope of success, not random fishing expeditions “just to see what happens”.

5. Conclusion

The Court of Appeal’s judgment in Modernatx v Pfizer fortifies two pillars of UK patent law:

1. Prior art that merely lists possibilities does not destroy novelty unless it gives a clear, unambiguous road-map to the claimed embodiment. The judgment crystallises this as the “individualised disclosure” requirement, applicable regardless of whether the list is explicit or embedded in a chemical class.
2. In biotechnology, where small molecular tweaks can have unpredictable effects, the absence of a mechanistic rationale and concrete expectation of success will preserve inventiveness, even if an experiment is technically feasible. Motivation matters.

By anchoring the skilled person in practical reality and policing hindsight in expert evidence, the Court provides a robust framework for future disputes involving list-based prior art and variant selection patents—particularly salient in the rapidly evolving RNA therapeutics landscape.