Patent law Summerslam for sufficiency

Patent law Summerslam – one appeal judge overturns another and pins down new tests for insufficiency in patent law

In what are usually quieter summer months, the UK Court of Appeal handed down quite a loud patent decision in Akebia v Fibrogen this year (available here).

After a characteristically rigorous first instance decision by Arnold LJ (sitting in the High Court), who revoked Fibrogen’s patents on almost every ground, the Court of Appeal (Birss LJ presiding) has reversed the outcome in part. Fibrogen now has one family of valid and infringed patents.

The wholesale reversal under insufficiency is the most significant turnaround, which will reverberate on the UK’s approach to this ground of invalidity, particularly via the application of yet further tests. The first of these continues to wrestle with the concept and impact of plausibility, which continues to make headlines in UK and European patent law (with perhaps the EPO’s Enlarged Board of Appeal having their say on it next).

At the core of the reversal is also a fundamental change in the construction of Fibrogen’s claims. In this respect, the Court of Appeal took a more wholistic and positive view regarding the impact of multiple structural and functional integers, rather than an arguably more divisive and destructive approach at first instance.

The reverberations of the reversal are felt across all 3 approaches to insufficiency in the UK:

1. Excessive breadth/of plausibility/technical effect - this is approached as a matter of “reasonable prediction” by Birss LJ, who formulates a new 3-step test requiring an assessment of: (i) claim scope; (ii) what the invention is for; and (iii) whether the skilled person can reasonably predict that you get (ii) using “substantially all” of (i). This does not seem to be a significant change in the law, but it is an interesting formulation. This case (concerning a medical use claim) illustrates the potential value of having both structural and (in vivo) functional features: as the functional integers were held to limit the scope of products that then have the purported therapeutic effect. This seemed to be a critical contribution to the claim being held not to be excessively broad in this case.

2. Undue burden - - this is considered separately by Birss LJ, who formulates another new, this time 2-step test, requiring the skilled person to be enabled, without undue burden, to (i) make “some more” (and not substantially all) embodiments within the scope of the claim; and (ii) work substantially anywhere across the scope. The latter in particular gives the impression of needing only a sampling or spot checks. This might be said to dilute the significance of “making”, which was recently addressed by the Supreme Court and held to be fatal for the patent in Regeneron v Kymab.

3. Uncertainty - Birss LJ continues to demonstrate the UK court’s willingness to find that a considerable body of research can sometimes be routine. In contrast to the first instance decision, the outcome may buoy patentees with broad or conceptual claim integers. In this case the Court of Appeal held that the skilled person could readily consider and decipher what a “structural mimetic” meant, including with regard to the different types of direct/indirect/allosteric cross-competition. While this turned on the facts (here small molecule enzyme kinetics), this approach may impact on the currently swathe of cases concerning antibody binding properties, including in particular epitopes and cross-competition. Of perhaps greatest significance is that this case shows the difficulties of successfully proving uncertainty insufficiency when you have a product that has been firmly found to infringe (which in this case Akebia had not even appealed).

Overall, the complexity of these issues is perhaps best illustrated by the wrestle between two of the UK’s leading patent judges in the Court of Appeal (who disagreed again more recently, in the now infamous Thaler/Dabus/AI case).

A final reflection on this appeal is the quite different and distinct tests for each of the UK approaches to insufficiency, in particular with respect to parts (1) and (2) above. These distinctions are despite the fact that each approach is under the same and single ground of invalidity: enablement (or undue burden). This seems to help the patentee navigate a positive outcome in this case, by pitting only the low threshold of plausibility against “substantially all” of the claim under (1), while the higher threshold of “making” only needs to match “some” of the claim and spot checks under (2).

As with any insufficiency case, this is a fact heavy decision but (for now) is on any reading a win for the patentee, adding some balance to some recent high-profile revocations before UK courts. The “for now” caveat is deliberate, as while this could be the final count for one family of Fibrogen’s patents (Family B), Akebia will almost inevitably seek permission to appeal to the Supreme Court for a rematch against the other (Family A). This may stand a chance of getting to the UK’s highest legal arena, not least given the complexity of the reasoning and disagreement between such eminent patent judges. And all the while, there are also pending and parallel oppositions and appeals before the EPO/TBA. It seems there may be quite a rumble to come.