IP ingredients: The not-so ‘Impossible’ burger

With meat alternatives facing growing scrutiny – amid labeling disputes and intense funding competition – robust intellectual property (IP) protection is more crucial than ever for a competitive edge.

Late last year, Impossible Foods appealed the European Patent Office’s (EPO) revocation of its patent EP2943072B1. Although oral proceedings were set for January 2025, the EPO’s Board of Appeal issued its decision (T 0425/23) without a hearing after the strawman opponent withdrew.

The patent concerns using heme proteins and flavor precursors to create a meat substitute that cooks, smells, and tastes like meat. However, the EPO’s Opposition Division ruled it was not sufficiently disclosed, as some described compositions failed to achieve the desired meat-like taste, indicating insufficient information for a skilled person to replicate the invention across its full scope without undue effort.

In Europe, an invention is sufficiently disclosed only if someone skilled in the field can reproduce its advantageous effects across the claim’s scope without undue burden. The Opposition Division found that identifying effective combinations from at least two flavor precursors, chosen from a list of 40, would impose an unreasonable burden.

The takeaway for patent applicants is clear: sometimes, less truly is more

On appeal, Impossible Foods submitted a revised claim set (AR1A), shifting claim 1’s focus to a method of making the meat substitute and narrowing the flavor precursor list. The opponent argued the claim still lacked sufficient disclosure, citing reliance on leghaemoglobin and claiming that selecting suitable heme proteins would be overly burdensome. The Board of Appeal disagreed. After reviewing examples 3 and 5, which demonstrated meat-like taste with leghaemoglobin, and example 9, showing that heating various heme proteins with ribose and cysteine produced meat-like flavor compounds, the Board found sufficient disclosure.

The Board rejected the opponent’s interpretation of key patent passages, clarifying that while different heme proteins can yield varying flavors and aromas, this variation does not prevent the overall achievement of a meat-like taste. Consequently, there was no evidence that it would be impossible to replicate the claimed effect using the specified heme proteins.

The opponent also raised concerns about whether all combinations of flavor precursors within the claim’s scope would yield a meat-like flavor under any concentration or pH. The Board dismissed this argument as well, citing examples 3 and 5, which showed that specific concentrations and pH levels did achieve the desired taste. The Board reasoned that a skilled person would naturally avoid conditions shown not to yield the intended results.

Beyond sufficiency of disclosure, the Board rejected objections related to added matter, clarity, novelty, and inventive step. As a result, the patent was upheld based
on the auxiliary claim set AR1A.

This outcome may surprise life sciences patent applicants, who are typically expected to provide data proving consistent technical effects. However, the Board’s decision is clearer when comparing the narrowed appeal claim with the broader original claim. EPO case law requires sufficiency objections to be based on serious doubts supported by verifiable facts. The patent’s broad scope – covering billions of combinations – raised concerns, especially as some described compositions produced ‘bready’ or ‘potato’ flavors instead of a meat-like taste.

However, the appeal’s narrowed claim, supported by examples, eliminated these doubts. The opponent lacked experimental evidence showing the specified combinations failed to produce a meat-like taste. In EPO opposition proceedings, such evidence is crucial, with a typical submission deadline of nine months post-grant.

Although this decision is a win for Impossible Foods – especially after a related patent was revoked in the USA – it comes at the cost of a significantly reduced claim scope. The outcome raises an important question: if Impossible had excluded non-working examples from the original filing, would the opposition have had
enough verifiable facts to raise serious doubts? Likely not, shifting the burden onto the opponent to provide evidence.

The takeaway for patent applicants is clear: carefully consider which experimental data to include, as non-working embodiments can undermine sufficiency and inventive step. Sometimes, less truly is more.

Jennifer Bailey is a Patent Director at HGF. She is an experienced European Patent Attorney, having practised in the field of IP for more than 10 years. Her technical expertise spans a wide range of sectors including life sciences, chemistry, materials and food science. This article is republished from the Q1 2025 edition of Protein Production Technology International, the industry's leading resource for alternative proteins. To subscribe to all future editions, please click here