Category Archives: Licensing

Astex v AZ: second thoughts

Yesterday, IP Draughts reported on the English Court of Appeal decision of Astex Therapeutics Ltd v Astrazeneca AB [2018] EWCA Civ 2444.

Having reflected overnight, he is keen to understand what lay behind this litigation. He wonders whether there was a disconnect between the technical, drafting point that was decided in the case, and the commercial realities of the dispute.

Without speaking to the parties who drafted and negotiated the deal, this is inevitably an exercise in speculation. But several points jump out at IP Draughts from the case reports that merit further attention.

For once, IP Draughts appreciates the great detail of Arnold J’s judgment at first instance. Although he cannot claim to have read all of its 43,000 words, a skim-reading and some word-searching reveal some interesting background to the dispute.

The agreement

Unfortunately, IP Draughts has not been able to find the full agreement on the internet. He has done some searches on Edgar, the database of the US Securities and Exchange Commission, but the agreement does not seem to have been lodged there as a material contract by either of the parties (or their parents). Seeing the full agreement would be really helpful to understand the drafting issues in their full context, and to see whether he agrees with the interpretation decided upon by Arnold J and Floyd LJ, respectively. If anyone can find a copy for IP Draughts, he would be very grateful.

As part of the drafting process, IP Draughts would have expected the parties to think through what would happen if AZ developed a compound after the collaboration had come to an end, but drawing on the output of the collaboration, and draft specific terms (particularly financial terms) to address this possibility. But it seems that the agreement doesn’t include provisions of this kind, or the court reports would have referred to them.

It seems rather an indirect way of looking at this issue, as the court was forced to do, to look at the various definitions of Program, Collaboration Term, Lead Compound, etc, and work out from these whether post-collaboration developments by AZ could be subject to the terms of the agreement, including milestone payment obligations.

Why didn’t the parties consider and address this important issue during the negotiations? Did they simply not address their minds to the point? Or did one party think about it but not want to raise the point explicitly in negotiations, preferring to slip in some wording and hope the other party wouldn’t notice it?

Perhaps there was enought complexity in the agreement already, and no-one had the energy to think about and address hypotheticals that were not covered. IP Draughts was amused by the following extract from Arnold J’s judgment, reporting on the actions of one of the AZ scientists:

Mr Berg read the Agreement once, but found it very complicated. He focussed on the science, and left the Agreement to his Business Development colleagues.

Perhaps the agreement text was too complicated for any of the participating parties to understand fully, whatever their professional backgrounds. And perhaps there wasn’t enough inter-disciplinary discussion of its terms. IP Draughts is strongly of the view that preparing a good collaboration agreement involves lawyers, business development people, scientists and others (eg regulatory specialists, accountants) working closely together. As Arnold J noted about another witness in the case:

Dr Angst’s evidence was that he did not have the authority to authorise payments to third parties. But he was clear that the Legal department was not able to make its own assessment of the science, and he did not know whether Business Development and Legal attempted to consider the issue of CD1’s contractual status independently or whether they just adopted and implemented the RAMT’s recommendation to pay the milestone payment.

The decision to declare CD1 and CD2 as collaboration compounds

AZ decided to pay milestone payments to Astex on the basis that CD1 and CD2 were compounds that were “caught” by the terms of the agreement. Why did they do this and why did it take them so long to change their minds?

As noted above, it was an AZ committee, the RAMT (Research Area Management Team) that, in effect, took this decision. Reading Arnold J’s judgment, it seems that this decision may have been based on an assumption by a member of the RAMT, Dr Angst:

Dr Angst recalled having a disagreement in about 2007 with Dr Nordstedt in respect of whether Södertälje’s work on the BACE project was within the Agreement, which may have been as a result of the discussions considered in the preceding paragraphs. At that time, Dr Angst was the person on the LGT and the RAMT with the most detailed knowledge of the Astex collaboration. Dr Angst’s view was the BACE project was covered by the Agreement. His view prevailed, and Dr Nordstedt never brought the matter up at the RAMT.

postcard from Södertälje

At the CD1 nomination meeting of the RAMT in August 2010, or shortly before, Dr Angst recalled that a question was raised by a Södertälje scientist about whether CD1 was a Collaboration Compound. Dr Christer Köhler, who headed both the RAMT and the iMed Leadership Team, asked Dr Angst for his view. Dr Angst’s view was that CD1 was evolutionarily linked to Collaboration Compounds and therefore was itself a Collaboration Compound. Accordingly, he recommended payment of the $1m for Program Milestone 3.

If this was what led the RAMT to decide that AZ should make the payments to Astex, it is a rather flimsy basis for a significant commercial decision. But as noted above, Dr Angst was not ultimately the decision-maker on this issue.

It seems that a business development director, Dr Farmery, notified Astex that it would be paying the first milestone payment for CD1. He declined to be a witness in the proceedings. Was it his decision based on the RAMT’s recommendation? In fact, did anyone carefully scrutinise the agreement to see whether the payment was due? Reading between the lines, it looks as though no-one did, until much later.

AZ’s change of mind

It seems that in 2012, an AZ scientist queried (within AZ) whether CD2 was a collaboration compound and subject to the agreement. Her query was referred to a recently-appointed in-house solicitor, Conor Johnston, who investigated the matter and discussed it with other AZ staff. (IP Draughts doesn’t know Mr Johnston, but sees that they have 6 connections in common on LinkedIn. Small world.)

In 2014, AZ licensed the CD2 compound to Eli Lilly, who also queried whether any payments were due to Astex. If payments were due, Eli Lilly agreed to pay 50% of them.

In 2015, Mr Johnston notified Astex of AZ’s change of view, ie that CD2 was not subject to the agreement.

Reflections

Arnold J’s judgment reveals a long line of people at AZ who thought that CD1 and CD2 were subject to the agreement. It was only when a skilled in-house lawyer examined the issue in detail, and discussed it with relevant parties, that AZ’s official position changed. This new position was ultimately supported by the High Court and the Court of Appeal.

Why did those people take a different view? It seems to have been a mixture of:

  • making wrong assumptions about the scientific issues
  • obstinacy, having made a decision
  • being put off by the complexity of the agreement
  • the imperfections of decisions by committee and going along with someone else’s view, however ill-informed
  • not wanting to “rock the boat” or prejudice relations with outside collaborators
  • failure to consult lawyers until a very late stage
  • not wanting to take a final decision until prompted by the pressure of a new business partner not wanting to share the burden of the payments

IP Draughts is left feeling that the merits of the case were really about the extent of the scientific connection between the two compounds and the earlier work done in the Astex/AZ collaboration.

Yet when the issues came before the court, at first instance and on appeal, its decisions were taken partly on the basis of interpreting complex definitions and contract wording that had several inconsistencies as identified by the court.

It seems that the drafting of the collaboration agreement sometimes has to do the “heavy lifting” when a dispute reaches the court, and that drafting may not be up to the task.

IP Draughts takes some comfort from the fact that it was not only the contract drafting that was sub-optimal in this case. As revealed by the case reports, the work of the scientists, business development managers and others also left a lot to be desired.

 

 

 

 

 

 

 

 

Leave a comment

Filed under Contract drafting, Intellectual Property, Licensing

Interpreting complex collaboration and licence agreements

The English Court of Appeal’s decision in Astex Therapeutics Ltd v Astrazeneca AB [2018] EWCA Civ 2444, reported earlier this week, should make us all pause and think about our contract drafting.

Drafting style of pharma agreements

Pharmaceutical collaboration and licence agreements tend to be drafted in a very “thorough” or at least detailed way. There are several reasons for this, in IP Draughts’ view:

  • they tend to deal with long term relationships – both scientific collaboration to generate new pharmaceutical drugs, and later commercialisation of those drugs
  • they contemplate the possibility of millions of dollars (or other currency) of payments being made, including so-called milestone payments and royalties on the sale of products.
  • the mindset of the individuals involved in such collaborations tends to favour a thorough and accurate approach – partly, IP Draughts suspects, because of the presence of chemists and other scientists in the team
  • the agreements that are used – many based on US templates – tend to be very detailed. In other words, the choice of document is partly based on habit and industry practice.

Which brings us to the Astex v AZ case cited above, and the collaboration and licence agreement that was in dispute. It is clear from the clauses that are quoted in the Court of Appeal judgment that the agreement is in a typical detailed format, and uses US-style drafting.

A peeve

<peeve> In an era of electronic documents, it would be much easier for practitioners to follow the arguments in the case, and decide whether they agree with the court’s interpretation, if the text of the agreement were attached to the end of the judgment in its entirety and in its original format (with whatever redactions the judge may agree to, to protect confidentiality). The traditional court practice of simply quoting relevant sections from the agreement, unformatted, is a very clunky approach, which may have been suitable for an earlier era of short agreements and high printing costs for law reports, but doesn’t make sense when the number of clauses quoted is large and the judgment is provided online. </peeve>

Key interpretation question in Astex v AZ

A central question in the case was how one should interpret the following definition (emphasis added), which formed clause 1.32 of the agreement:

1.32 “Program” means the research program described in the Research Plan, to be performed in collaboration by the Parties during the Collaboration Term as part of the Project, which thereafter may be continued by or on behalf of ASTRAZENECA alone.

Does the wording that IP Draughts has highlighted in red mean that product development activities conducted by AZ after the active collaboration with Astex has come to an end, but following on from the research conducted during that collaboration, come within the definition of Program?

This was one of the key questions of interpretation before the court. Reading clause 1.32 in isolation, IP Draughts’ first reaction is “yes” – those post-collaboration activities are within the definition of Program.

Then his second reaction is, did the parties really intend that any developments made by AZ downstream would automatically come within the scope of the agreement? That seems rather open-ended.

But let’s go back a few steps.

Context of the Astex v AZ dispute

As explained in the judgment of Floyd LJ in the Court of Appeal:

  1. The parties collaborated from 2003-2005.
  2. Thereafter, AZ continued the Project on its own.
  3. Some years later, two compounds known as CD1 and CD2 were nominated by AZ as Candidate Drugs under the agreement, and AZ made milestone payments to Astex. It seems that a total of US$2M was paid.
  4. CD1 progressed to Phase I trials but was then dropped.
  5. In 2014, AZ announced a collaboration with Eli Lilly under which the latter would conduct Phase II/III trial on CD2.
  6. In 2015, AZ informed Astex that it had reviewed its position and now considered that neither CD1 nor CD2 were Collaboration Compounds under the agreement.
  7. At the time of the trial of the present action, CD2 was in Phase III trials.
  8. By the time of the appeal, development of CD2 had been discontinued.

The main questions before the court at trial and on appeal were (in IP Draughts’ paraphrase):

(a) whether the Program ended when the collaboration ended, such that compounds generated by AZ after the end of the collaboration (eg CD1 and CD2) were automatically outside the scope of the agreement.

(b) if the answer to question (a) is no, were the compounds CD1 and CD2 discovered as a direct result of activities performed as part of the Program.

Arnold J in the High Court decided that the Program ended when the collaboratoin ended. And that the answer to question (b) was (for reasons that he set out at length) no. AZ won on all counts.

The Court of Appeal agreed with Arnold J’s conclusions on the definition of Program, and found it unnecessary to opine on the latter question, which didn’t arise in light of the limited scope of the Program.

Closer scrutiny of the Program definition

So, why did both courts decide that the Program definition should be interpreted in a way that, in IP Draughts’ view, wasn’t the obvious way of interpreting the words in isolation?

Arguments were presented by each side for the interpretation that they preferred, many of which were based on looking at the wording of other definitions and clauses to see whether it was intended that the Program would survive the end of the collaboration.

As Arnold J noted, Astex’s strongest argument was to look at the definition of Lead Compounds in clause 1.23, which referred to structures selected during the collaboration “or, after the Collaboration Term, by ASTRAZENECA as candidates for Lead Optimisation.” This definition seems to suggest that compounds discovered by AZ after the end of the collaboration could be caught by the agreement.

But Arnold J thought that other clauses, such as 1.6 and 1.17, indicated that the phrase “after the Collaboration Term” above could have been a drafting error. The wording of 1.6, the definition of Candidate Drug, and 1.17, the definition of Hits, made no reference to post-collaboration developments.

In the Court of Appeal, each side’s counsel played the game of “look at the other clauses” for all it was worth. Astex’s counsel referred the court to clauses 14.1, 2.9, 1.23, and 1.38. AZ’s counsel referred to the wording of the Research Plan and to clauses 3.7. 3.8, 3.1, 4, 5.1, 5.2, 14.1, and 8.

Floyd LJ assessed these arguments and ultimately came out in favour of AZ’s interpretation, ie that the Program ended when the Collaboration Term ended.

Award of costs

The case report reveals that, before the trial, AZ had offered to settle the dispute on terms that it wouldn’t ask for the US$2M back that it had already paid to Astex, and would bear its own costs in the litigation, but would require Astex to accept that CD1 and CD2 were not Collaboration Compounds. Astex rejected this offer. With hindsight, they were unwise to do so. Arnold J thought they were so misguided in their conduct that he ordered them to pay AZ’s costs on an indemnity basis (ie awarding more costs that would normally be awarded to the winner). But the Court of Appeal overturned this part of Arnold J’s order.

Concluding thoughts

As is often the case with pharmaceutical collaboration and licence agreements, the wording of this agreement is very detailed and appears to have been carefully thought out. But when you look at a set of facts, and try to interpret the detailed wording, problems of interpretation can emerge.

IP Draughts’ approach, when his client is willing to pay for the time involved, is in several stages:

  1. To try to start with a draft agreement with which you are familiar. But that depends partly on having the bargaining power to propose your own draft, or being cute enough to have a draft ready for circulation when the parties have finalised their heads of terms.
  2. To spend time carefully going through the draft, check one’s understanding of the commercial intention with business and scientific colleagues, and review the draft for consistency. Sometimes, errors can be spotted by noting subtle differences in wording between clauses dealing with similar issues. IP Draughts is a great believer in consistency as an important drafting technique.
  3. To stress-test the wording by using several real-life examples, to try to see how robust the wording is and whether it works for those examples. This may involve other members of the negotiating team, who bring their own professional expertise to bear on the stress-testing exercise.

This is a good argument, if one is needed, for erring on the side of simplicity rather than complexity; the longer the agreement, the more scope for drafting errors and unintended conseequences. IP Draughts has just checked the last agreement of this kind that he worked on, which happens to have been drafted by a Californian company. The word count was close to 30,000 words, or 70 pages.

Leave a comment

Filed under Licensing, Intellectual Property, Contract drafting

Unwired Planet win FRAND case appeal

The English Court of Appeal’s judgment in the case of Unwired Planet International Ltd & Anor v Huawei Technologies Co Ltd & Anor [2018] EWCA Civ 2344 has now been published. In brief, Kitchin, Floyd and Asplin LJJ wrote a joint judgment which in all material respects upheld the decision of Birss J at first instance. The first instance decision was discussed several times on this blog, including here. It seems that Huawei intends to try to appeal again to the UK Supreme Court.

A quick and oversimplified reminder: this case is partly about when the terms (and associated negotiation) of a licence agreement for standards-essential patents (SEPs) should be considered fair, reasonable and non-discriminatory (FRAND). Patent owners who declare (to standards-setting bodies) that their patents are SEPs are required by the rules of the standards-setting body to offer those patents for license on FRAND terms.

In very brief summary, the Court of Appeal decided:

  1. Birss was right to say that the licence had to be global, rather than limited to the UK as Huawei sought.
  2. It was not discriminatory for the terms of the licence to be different to those UP had previously provided to Samsung.
  3. UP had not abused a dominant position by starting legal proceedings against Huawei.

One of the topics that comes up in FRAND licensing is that parties tend to work out the financial terms by counting the number of standards-essential patents that the patent owner has, without looking too closely at whether each of them is valid. This is a crude but sometimes cost-effective approach for the parties to take. Reading through the Court of Appeal’s judgment, IP Draughts was struck by the number of arguments that they addressed and decided, and he wondered whether, in the interests of efficiency, litigation should be reduced to counting the number of arguments that each party makes.

In total the Court of Appeal addressed around 50 detailed sub-arguments that fell within the three numbered issues mentioned above. The Court of Appeal agreed with Birss J on virtually every point discussed in the CA judgment. The most notable area where they disagreed with him was that they held he was wrong to say that there is only ever one set of licence terms that is FRAND in any particular case, but this made no difference to the outcome of the case.

Commentators have today suggested that the Court of Appeal’s judgment will tend to make the UK a jurisdiction of choice for future FRAND licensing disputes. If that is correct, it should benefit both IP litigators and IP contract drafters based in the UK.

 

 

 

2 Comments

Filed under Intellectual Property, Licensing

Article on drafting FRAND licences now published

Back in December, IP Draughts had an article published in the online version of the Journal of Intellectual Property Law and Practice (OUP). The article’s title is How to draft a licence agreement that is fair, reasonable and non-discriminatory: a ten-point plan.

IP Draughts learnt at the weekend that this 10,000-word article has now reached the paper version of the journal. For non-subscribers to the journal, it can be purchased here.

The abstract reads as follows:

The English High Court case of Unwired Planet International Ltd v Huawei Technologies Co Ltd & Anor is of great interest to IP lawyers who advise on standards-essential patents (SEPs), on associated questions of competition law, and on licensing patents on fair, reasonable and non-discriminatory (FRAND) terms. The case is also of interest internationally, partly because there have been relatively few cases anywhere in the world, to date, that provided judicial guidance on legal issues relating to SEPs and FRAND and because this is the first time that an English court has determined a FRAND royalty rate.

This article focuses narrowly on the FRAND aspects of the Unwired Planet case, and on what the case teaches us about how to draft a patent licence agreement that is FRAND. But the discussion is broadened to reflect not only on the licensing of SEPs but also of other types of licence where there is an obligation to negotiate fair and reasonable licence terms, with particular reference to the terms of grant of the European Commission for Horizon 2020 research projects.

This focus results in the author omitting discussion of some points that would be of interest to the lawyer who is running litigation against a user of an SEP (perhaps in parallel with negotiating a FRAND licence with that infringer), or who may be advising on compliance with competition laws in the context of such litigation and negotiations.

Leave a comment

Filed under &Law Updates, Legal Updates, Licensing