The team is joined by Guest Kats Rosie Burbidge, Stephen Jones, Mathilde Parvis, and Eibhlin Vardy, and by InternKats Verónica Rodríguez Arguijo, Hayleigh Bosher, Tian Lu and Cecilia Sbrolli.

Thursday, 24 December 2015

Mr Justice Carr tackles plausibility in Actavis v Eli Lilly ADHD dispute (Part II)

The law on insufficiency and the role of plausibility

Putting a face to a name:
Lilly's Strattera drug for the treatment of ADHD
The most interesting part of the Mr Justice Carr’s decision was his take on plausibility. [“Way to bury the lead, AmeriKat”, criticizes Merpel]. The parties’ arguments are summarized as follows.

Actavis argued that plausibility meant that a patent provides a "reasonable prediction" that the invention will work based on the common general knowledge.  This argument broke down as follows:
  • In light of Hospira v Genentech [2014] EWCA Civ 1094 and Regeneron, the patent must make it possible to make a reasonable prediction that the invention will work across the scope of the claim – it must be plausible or credible. Because Lord Justice Kitchin equated “plausible” and credible” with “reasonable prediction the invention will work”, the test for plausibility, in the context of sufficiency, is the same as “reasonable expectation of success” for obviousness. 
  • Although the results of clinical trials are not required, the patent has to be more than a paper proposal to use drug A to treat disease B without any other basis. Showing a pharmaceutical effect in vitro or if there is a clear and accepted established relationship between the physiological activity and the disease may be sufficient.
  • The standard for assessing obviousness and sufficiency should be the same as they both focus on the technical contribution provided by the patent. A patentee should not be able to obtain a monopoly for an invention that does not add to the art either because it was obvious to reach the invention based on the common general knowledge/and or prior art or because it was implausible based on the patent’s teaching. 
  •  Therefore, the EPO and the courts require it to be possible to make a reasonable prediction that the invention will work based on the technical teaching of the patent and the common general knowledge.
Eli Lilly argued that pluasibility was a low threshold test.  Its arguemente was as follows:
  • Lack of plausibility is not a ground of invalidity and the imposition of such a requirement would be contrary to TRIPS, the EPC and the Patents Act 1977. 
  • Plausibility arises only as a check on overly-broad claims, for example where a class of compounds is alleged to work in a number of different conditions (as was the case in Regeneron and Salk). 
  • Accordingly, the threshold for plausibility is a low and flexible one, especially as a higher threshold would discourage innovation as it would either be too difficult to satisfy the inventive step requirement as patents would only be awarded for inventions that were “implausible” from the prior art or too difficult to satisfy the test for plausibility, particularly in the life sciences sector where regulatory approval takes years of testing. 
  • This test was supported in Conor where Lord Hoffmann held that the objection of lack of plausibility applies to speculative patents, whose breadth of claim meant that the alleged invention is inherently improbable. The patentee in that case offered very little about how or why taxol would be efficacious in preventing restenosis and offered no proof in support. However, the House of Lords did not consider that this would lead to implausibility, which was unsurprising as it would have put a squeeze on the “obviousness to try” case (i.e. if it was obvious to try with a reasonable expectation of success surely the patent also made it plausible). The fact that the patent offered a theory, but no evidence, that taxol would be effective for the claimed therapeutic purpose, did not render it implausible. If it did, the patent would be invalid for insufficiency. This is why, held Carr J, plausibility is referred to as a “threshold test” which is consistent with its application to prevent the filing of purely speculative patents. 
  • This low threshold, in the context of industrial applicability, was supported by the Supreme Court in HGS v Lilly [2011] UKSC 51 where “plausible” was used alongside “educated guess” and “reasonably credible”. Had the Supreme Court intended plausibility to be the same standard of obviousness they had opportunity to say so in Conor and HGS and would not have adopted the “educated guess” language. 
  • This is supported by the EPO case law, in particular T 578/06 Ipsen, where the Board emphasised the need for substantiated doubts about the suitability of the claimed invention to solve the technical problem in order to call into question the plausibility of the patent. 
  • The “reasonable prediction” language comes from Agrevo where the Board there explained that it is possible for a prediction to be “not obvious, but nevertheless reasonable”. This only makes sense if the hurdle for “reasonableness” is below that of obviousness. “Reasonable prediction” is therefore not the flip side of obviousness. 
  • Evidence obtained after the priority date can be adduced to defend an allegation of insufficiency/lack of support at the EPO. This could only be justified if the plausibility hurdle is a low one – if it was high there would be no point in adducing evidence. There has to be some basis in the application. 
  • Finally, Lilly flagged that under Rule 42(1)(e) EPC a patent is not required to include examples nor does the invention have to actually been carried out (see T 1437/07 Allergan).
Mr Justice Carr gives the middle
box a big red check
From this Mr Justice Carr held that:
  • Although plausibility is not a separate ground of objection to the validity of a patent, it must be borne in mind and is when the Boards of Appeal of the EPO and the national courts interpret the requirements of TRIPS, the EPC and the Patents Act 1977. 
  • Plausibility does not apply only to claims of wide scope, but all claims. Whether an invention is plausible is fact, invention, claim, specification and CGK dependent. Claims of narrow scope are likely to be easier to demonstrate as plausible than those of wide scope. 
  • There is no requirement in the EPC that a patent should contain data or experimental proof to support its claims. Salk’s reference to the provision of experimental data is by way of example, only. 
  • The policy considerations underlying plausibility for sufficiency are different from those underlying fair expectation of success for obvious. As to the latter, almost anything could be said to be obvious to try and would lead to a denial of patent protection (and therefore research in pharmaceuticals) if it was not for the fair expectation of success requirement. “Plausibility, on the other hand,” Mr Justice Carr held, “is to exclude speculative patents, based on mere assertion, where there is no real reason to suppose that assertion is true.” 
  • The test of plausibility is a threshold test which is satisfied by disclosure that is “credible”, as opposed to speculative. Such disclosure may be subsequently confirmed or refuted by post-published evidence. If it is shown that the invention does not work across substantially all products/methods falling within the scope of the claim, the monopoly will exceed the technical contribution and the patent will be invalid. This is why plausibility is only a threshold test and not the same test as expectation of success in the context of obviousness.
The plausibility of the Patent

The judge held that the Patent was plausible by in light of four factors:
  1. The disclosure in the Patent: The patent disclosed that atomoxetine is effective in the treatment of ADHD, is a selective NE reuptake inhibitor and cited the Gelhert paper in support. The skilled person reading Gelhert would also read Wong, which would further support atomoxetine as a selective NE reuptake inhibitor. 
  2. The common general knowledge: At the priority date, the skilled clinician would have considered the hypothesis that the TAs had efficacy in the treatment of ADHD as a result of selective inhibition of NE reuptake to be reasonable and that selectivity would reduce side effects. The position with ADHD could be more complex, but that does not detract from the conclusion that the skilled team would consider the invention credible based on the specification and common general knowledge. 
  3. Post-published evidence: The widespread administration of atomoxetine to ADHD patients has confirmed that the product is safe and efficacious, as disclosed in the Patent. 
  4. The expert evidence: All the experts considered that the disclosure of the Patent to be credible. Actavis argued that Lilly's experts, Professors Hill and Sharp, made a fundamental error in that they had a mistaken believe that the Patent was an authoritative document in which statements made in it had to be supported by evidence. The judge agreed that their assumptions were wrong. However, their mistaken belief that the Patent was some authoritative document was one of several reasons why the experts considered the Patent credible.
Accordingly the judge held that the disclosure of the Patent was plausible. He continued:
“It’s technical contribution has proved valuable in the treatment of this serious condition. The prior art lacks this technical contribution. Neither citation makes it obvious to try atomoxetine for the treatment of ADHD. Furthermore, neither citation gives rise to a reasonable expectation that atomoxetine would be effective in the treatment of ADHD. For these reasons, I reject the alleged squeeze between obviousness and insufficiency.”
So with that, Mr Justice Carr held the Patent valid and Lilly’s counterclaim for threatened infringement succeeded.

The plausibility threshold:  you will
will trip if you don't look
where you going (or your patent lacks

Mr Justice Carr's decision provides some reminders on the limits on invalidity challenges by confirming that:
  1. The court must come to the question of obviousness with the mindset of the skilled person.
  2. Obviousness must be assessed in reference to the scope of the claimed invention, not in reference to some vague paraphrase based upon the extent of the disclosure in the patent's description. 
  3. Plausibility is a threshold test that asks if the invention is "credible" based on the patent and common general knowledge.  It is not to be equated with the fair expectation of success test for obviousness.  
In ring fencing the scope-creep of plausibility, Mr Justice Carr's decision will provide some reassurance for patentees who have watched plausibility attacks seep into obviousness and sufficiency without any discernible bounds (especially by virtue of vague and fuzzy pleadings).  However, it is doubtful that this will be the final word on plausibility as if there is any area in patent law ripe for the Court of Appeal, this is one of them. 


Anonymous said...


"It is not to be equated with the fair expectation of success test for obviousness." Absolutely.

But, how about:
"It is to be equated with the fair expectation of success test for plausability."

LowHurdle said...

In Merck v Ono, Birss, following EPO case law said a prior art document that was not plausible was not novelty destroying. That will mean applicants will increasingly push for plausibility to be a high hurdle as they argue in examination that their claims are novel. So it will be interesting to see whether the Courts can keep it as a low hurdle test.

Subscribe to the IPKat's posts by email here

Just pop your email address into the box and click 'Subscribe':