Concentration isn't that hard - Methotrexate patent Accordingly found invalid

This moggy has just finished reading his copy of Mr Justice Birss’s decision Accord Healthcare Limited v. medac Gesellschaft [2016] EWHC 24 (Pat).  Accord, a generics manufacturer, initiated proceedings, seeking revocation of medac’s patent (EP 2046332), which related to the use of methotrexate for subcutaneous administration in the treatment of inflammatory autoimmune diseases, wherein the methotrexate is administered in a pharmaceutically acceptable solvent at a concentration of about 50mg/ml. The patent contained Swiss and second medical use claims.

After 132 paragraphs of discussion, Birss J ultimately held that the subject matter was obvious in view of “Russo”; a journal article published 6 years before the priority date.  To understand the reasoning behind the decision requires a certain understanding of the facts, so please bear with me dear reader. 

Methotrexate was known at the priority date to be used in high doses for cancer treatment and low doses in the treatment of autoimmune diseases such as rheumatoid arthritis (RA) and juvenile chronic arthritis (JCA).  Methotrexate was known to have potentially serious side effects.  On the evidence before him, Birrs J considered that these were known to the skilled person to be systemic in nature and thus dependent on the total dose administered (rather than localised in nature, in which case they would be dependent on the manner of administration).  Subcutaneous administration of up to 25mg was the highest known dose for the treatment of autoimmune ailments (where longevity of treatment is a key factor and minimising side effects is important).

The prior art, Russo, refers to “parenteral” administration of methotrexate and reports a study in which several patients reported pain at the injection site, but does not indicate the methotrexate concentration in the injected fluid.  Birss J found that the skilled person would see the reference to “parenteral” as disclosing subcutaneous and/or intramuscular administration and that any and all of these administration routes are obvious in view of Russo.  Medac argued that:

• the proportion of patients reporting pain in Russo (20%) is too low to incentivise the skilled person to create a new formulation;
• even if he were seeking to address the reported pain, several aspects of an injection are known to influence pain including the needle size and shape, the injection technique and the volume of injected fluid, and so it would not be obvious to increase the concentration to lower the dose volume; and
• the skilled person would have no expectation of success if the concentration were increased because of anticipated side effects, and so it would not be obvious to try increasing the concentration.

Medac also pointed to the 6 year interval between the Russo publication and the priority date, asking “if the invention was obvious, why was it not done before?”.

Accord’s clinical expert, whom Birss J found to be more persuasive than medac’s, explained that given an increase in concentration, the side effects which might be relevant would be local side effects (rather than systemic side effects) but that the skilled clinician would have no specific concerns about local side effects of subcutaneous methotrexate.  Accord’s formulation expert’s evidence was that part of the formulator’s common general knowledge was that a smaller injection volume caused less pain and that he would target 0.5 ml.  The average dose described in Russo is about 26 mg, which requires a 52 mg/ml concentration in a 0.5 ml volume (i.e. close to the 50 mg/ml figure recited in the claims).  The expert evidence indicated that, whilst such a concentration would need to be tested before administration, there would be a high expectation of success and thus it would be obvious to try such a formulation.  Birss J therefore saw a “powerful obviousness case over Russo”, subject to consideration of the “secondary” evidence.

Birss J then went on to consider medac’s “secondary” evidence (why had it not been done before?) but found several reasons why the higher concentration subcutaneous preparation might not have been developed between Russo’s publication and the priority date.  These included:

• since methotrexate is a generic drug, there was limited commercial interest in investigating new formulations, with research being primarily focussed on new biologics;
• simply that clinicians (and/or nurses depending on the identity of the person administering the drug) who form part of the skilled team might not have been aware of the problem of pain, since most methotrexate administered in the UK before the priority date was taken orally; and
• that Russo might simply not have been read by a skilled person looking to address the issue of the invention, and that this could be the sole reason for it not having been done before.  In order for such a “why had it not been done before?” position to have weight, the patentee would need to show that the Russo document had been read and its content appreciated by the skilled person, who had nevertheless not pursued the route taken by the inventors.

Overall, Birss J held that medac’s “secondary” evidence did not provide sufficient weight to overcome the strong obviousness argument over Russo.  Therefore, he found that the claims lacked an inventive step.  A further citation, Jansen, was not considered in detail by Birss J because he considered the Jansen obviousness attack to be weaker that the attack starting from Russo.

Having found the claims invalid, Birss J then commented on Accord’s argument that the invention was obvious in view of the common general knowledge.  Indeed he was quite disparaging:

The problem with arguments over common general knowledge alone is that the combination of features relied on is always and necessarily one created with hindsight knowledge of the invention, and worse, is one which the person attacking validity has not been able to find as a pre-existing combination in the concrete prior art.  If they had they would have relied on that concrete prior art.  Either the combination has not been made in the concrete prior art at all or it only appears with additional inconvenient details.  If an invention is not obvious over the concrete prior art which is relied on, the court is entitled to be sceptical that an argument that it is nevertheless obvious over common general knowledge alone is correct.

The problem is illustrated in this case. Sometimes an invention belongs to a field which is not well documented but in this case Accord did not lack possible starting points. It has pleaded two documents and could easily have pleaded others…. However the documents contain what might have been thought of as "inconvenient" details. Russo does not mention subcutaneous administration by name and is aimed at JCA rather than RA. Russo is also a small study and was published six years before the priority date. Jansen does mention subcutaneous and is for RA but it does not mention pain…. That the "inconvenient" details in Russo have not led to a finding of non-obviousness is not the point. To invent as a starting point in the prior art an amalgam of the best bits of the two cited documents while leaving out the inconvenient aspects, which is in effect what the argument was, created a combination which did not hitherto exist

Birss J also discussed an insufficiency / inventive step squeeze argument proposed by Accord.  The patent was entirely silent regarding the side effects of methotrexate and simply stated that the reduced injection volume was intended to reduce pain.  Accord argued that, had the argument about side effects materially contributed to a finding that the claims involve an inventive step (i.e. if it was not obvious to try the higher concentration preparation due to the risk of side effects), then the avoidance of side effects was not plausibly demonstrated in the patent.  The judgement states at paragraph 131:

If it was not obvious to a skilled person to make a 25 mg dose using a 50 mg/ml concentration in a 0.5 ml volume in order to reduce pain, such a finding would not reveal a sufficiency problem not least because the patent explains that the reduced volume reduces pain.  However the position with side effects is different.  If it was not obvious to administer a 25 mg dose using a 50 mg/ml concentration in a 0.5 ml volume subcutaneously because of a concern about the risk of side effects, then the patent does not give such a skilled person any comfort at all about that risk.  If that skilled person would not administer the formulation to treat [rheumatoid arthritis] due to the risk, they would still not do it after reading the patent, and the claim to the use of that formulation would be insufficient.

Finally, Birss J dismissed the proposition that the identity and reputation of the patentee could make a disclosure sufficient (as was suggested by Medac’s expert):

[medac’s expert] explained in his evidence that he regarded the invention as plausible because the holder of the patent, medac, was a well respected pharmaceutical company. Medac relied on this evidence. It is not relevant. As a matter of law, the identity of the inventor or patentee cannot have anything to do with sufficiency in general and plausibility in particular. The issue is an objective legal standard. It is based on the technical disclosure only.

This Kat would like to thank his colleague Tim Belcher for authorial assistance with this post.