Regents v Accord: The Court of Appeal Re-frames “Context” and Hindsight in Obviousness Analyses for Pharmaceutical Patents

1. Introduction

In Accord Healthcare Ltd & Ors v Regents of the University of California & Anor ([2025] EWCA Civ 936) the Court of Appeal (Arnold, Snowden and Zacaroli LJJ) revisited one of the most persistent battlegrounds in patent law: the assessment of obviousness where the prior-art disclosure is itself a near-complete drug candidate. The dispute concerns enzalutamide, a blockbuster prostate-cancer therapy marketed by Astellas. Six generic manufacturers (Accord, Sandoz and Teva group companies) sought to revoke European Patent (UK) No 1 893 196 and the associated Supplementary Protection Certificate (SPC/GB13/079) on grounds of obviousness over two prior-art disclosures—an academic poster (“the Poster”) and conference slides (“the Slides”)—which emanated from the patentees’ own scientists but were published within the priority gap.

Mellor J at first instance dismissed the revocation action. On appeal the generics narrowed their attack to pure obviousness, contending that a single molecular tweak (cyclobutyl → geminal dimethyl at “position X”) from the public disclosures to the claimed compound was technically de minimis and therefore obvious. Arnold LJ, giving the lead judgment, upheld Mellor J. In doing so, the Court crystallised two propositions of wider importance:

1. The assessment of obviousness must remain context-anchored; a court is entitled to inquire why the skilled team would wish to make the contested modification, even where the modification is structurally “small”.
2. Expert evidence can become impermissibly coloured by hindsight when it is framed around the Pozzoli steps without an articulated research scenario; courts will scrutinise such evidence and may discount it if the “target” compound is smuggled in through the questioning.

2. Summary of the Judgment

• The appeal was dismissed. Mellor J had made no error of principle; his multifactorial evaluation stood.
• Although the structural change between RD162 (prior art) and RD162′ (enzalutamide) looked trivial, the Court accepted the first-instance finding that the generics’ expert (Professor Westwell) had reached his conclusion via impermissible hindsight and without describing any real-world research programme that would motivate the change.
• The Court emphasised that an obviousness attack cannot rely solely on the “smallness” of the chemical step; motivation, alternatives and realism of the medicinal chemistry workflow remain critical.
• Instructing experts to slavishly follow the Pozzoli framework may inadvertently direct them toward the patented endpoint; parties should frame expert tasks in open, research-oriented terms.

3. Analysis

3.1 Precedents Cited

The Court drew on an established line of authority but sharpened its application:

• Actavis v ICOS [2019] UKSC 15 – Lord Hodge’s nine-factor matrix for obviousness provided the analytical backdrop, with particular weight on (i) multiplicity of research avenues and (ii) motivation.
• Pozzoli v BDMO [2007] EWCA Civ 588 – the structured four-step test remained the skeleton, but the Court warned against using it as a “paint-by-numbers” exercise that masks hindsight.
• Conor v Angiotech and Pharmacia v Merck – reaffirming that obviousness is fact-specific and that a step can be obvious even without commercial motivation, yet motivation is still a relevant factor.
• Lifestyle Equities v Amazon [2024] UKSC 8 and Iconix v Dream Pairs [2025] UKSC 25 – cited for the limited appellate remit over evaluative fact-findings.

3.2 Legal Reasoning

1. The “Context” Requirement
   Arnold LJ accepted that courts assume technicians may make trivial tweaks, but held that the judge was entitled to ask for the contextual research objective that would prompt the skilled team to replace cyclobutyl with gem-dimethyl. No such objective emerged from Professor Westwell’s reports.
2. Hindsight and Expert Evidence
   The Court upheld Mellor J’s scepticism: Westwell had first reviewed the Patent and Slides, then later the Poster, and his analysis was framed around the exact change guarded by the patent. That sequencing, combined with a lack of articulated TPP (Target Product Profile) or SAR plan, rendered his “immediately obvious” conclusion suspect.
3. Multiplicity of Research Avenues
   Even accepting that RD162 was a promising lead, the medicinal chemist had many alternative moieties to explore (RHS substitutions, hetero-atom tweaks, linker variations). Choosing the specific dimethyl swap was not a “one-way street” and therefore could not be presumed obvious.
4. Appellate Deference
   Applying Actavis v ICOS and the recent Supreme Court guidance, the Court of Appeal repeated that it will not interfere with a judge’s multifactorial evaluation absent an error of principle. None was found.

3.3 Impact of the Decision

• Expert Instruction – Litigants must ensure that experts explain the real-world project the skilled team would undertake. A bare Pozzoli walk-through, especially if preceded by review of the patent, risks being discounted.
• Pharmaceutical Obviousness – “Small-molecule” cases often revolve around seemingly minor substitutions. This judgment clarifies that courts will look beyond structural triviality to research motivation and competing paths.
• Patent Drafting & Competitive Monitoring – The Court recognised “competitive and patenting considerations” as potential motivators but did not elevate them into a legal test. Innovators should assume competitors will be wary of close analogues yet courts may still protect them if no credible, context-based route is shown.
• Procedural Strategy – Parties may rethink the order in which documents are shown to experts and the structure of expert reports to avoid accusations of hindsight.

4. Complex Concepts Simplified

• Obviousness – A patent claim is invalid if the skilled person/team would have found the invention obvious over the prior art at the filing date; it concerns technical, not economic, obviousness.
• SAR (Structure–Activity Relationship) – An iterative medicinal chemistry process where small structural changes are made to a lead molecule to map how each affects biological activity & drug-like properties.
• Target Product Profile (TPP) – A predefined “wishlist” of efficacy, safety and PK parameters that guides a drug-discovery project.
• PK/PD – Pharmacokinetics (what the body does to the drug) and Pharmacodynamics (what the drug does to the body).
• Geminal Dimethyl – Two methyl (CH₃) groups attached to the same carbon atom; here, compared with a cyclobutyl (four-membered carbon ring) as hydrophobic substituents at “position X”.
• Hindsight Bias – The cognitive fallacy where knowledge of the invention colours the assessment of what would have been obvious at the relevant date.

5. Conclusion

Regents v Accord does not radically rewrite the law of obviousness, but it sharpens two practical axes: the need for contextual motivation in expert evidence and the vigilance against hindsight when the road from prior art to claim appears short. By upholding Mellor J, the Court of Appeal signals that a “small” chemical modification will not be labelled obvious unless the challenger can narrate a plausible research path the skilled team would have followed—grounded in the realities of drug discovery rather than in counsel’s retrospective lens. For practitioners, the case offers both a warning and a toolkit: craft expert reports that mirror realistic laboratory decision-making, and scrutinise the sequencing of information to avoid the very hindsight the Court has, once again, placed under a bright appellate light.