Pharmacia LLC v Juno Pharmaceuticals Pty Ltd

Pharmacia LLC v Juno Pharmaceuticals Pty Ltd [2022] FCAFC 167

In this case, the Full Court upheld Burley J’s conclusions that three of six batches of Juno Pharmaceutical Pty Ltd’s parecoxib products did not infringe Pharmacia LLC’s patent.  Their Honours also upheld Burley J’s conclusions that the other three batches of Juno’s products did infringe and that the patent was valid

Background

Pharmacia was the patentee, and Pfizer Australia Pty Ltd (the Pharmacia parties) was the exclusive licensee, of a patent relating to the formulation and administration of cyclooxygenase-2 (COX-2) inhibitors.  Parecoxib is a ‘prodrug’ – a molecule that is converted into a pharmacologically active agent (the selective COX-2 inhibiting drug valdecoxib) after administration to a patient.  Juno offers for sale and sells parecoxib in Australia.

The Pharmacia parties sought relief against Juno and its then-supplier, Neo Health (Australia) Pty Ltd, for infringement of the patent.  Juno and Neo denied infringement and alleged the asserted claims were invalid. 

The trial was conducted by reference to the batch records for six exemplar batches of allegedly infringing products.  Justice Burley concluded that three of the six exemplar batches infringed and that the asserted claims of the patent were valid. 

His Honour made declarations that three of the exemplar batches fell within the scope of the claims and the other three did not, refused injunctive relief, and deferred costs of infringement until quantum of any pecuniary relief owed to the Pharmacia parties had been resolved.  The Pharmacia parties appealed: (a) Justice Burley’s declaration that three of the exemplar batches did not fall within the scope of the claims; (b) his Honour’s decision allowing Juno to amend its defence to specify that the infringing exemplar batches were not sold or brought into Australia; and (c) the order deferring the costs of the infringement action. 

Juno appealed: (a) Burley J’s declaration that three of the exemplar batches were infringing; and (b) his Honour’s orders that the cross-claim be dismissed and that the patent’s validity be certified. 

The Pharmacia parties filed a notice of contention, asserting that Burley J ought to have found that: (a) the skilled addressee could not have been expected to ascertain the prior art; and (b) the problems addressed by the invention were not part of common general knowledge.

The claims and Justice Burley’s construction

The patent described certain problems in relation to formulating parecoxib.  In particular, it described that:

  • parecoxib converts to valdecoxib upon exposure to water (said to be solved by the patent);
  • where the therapeutic agent is parecoxib, partial conversion to valdecoxib can occur in a composition over time (said to be solved or greatly reduced by reduction, or preferably elimination, of bulking agents such as mannitol); and
  • parecoxib formulations containing mannitol and other excipients do not dissolve instantly (said to be solved by achieving a high degree of porosity, achieved by adopting a process which included identified steps in a lyophilisation cycle and process).

The claims were, relevantly, to (emphasis added):

  1. A pharmaceutical composition comprising, in powder form:

(a)    at least one water-soluble therapeutic agent […] in a therapeutically effective total amount constituting about 30% to about 90% by weight […]

  1. A process for preparing areconstitutable selective COX-2 inhibitory composition, comprising a step of lyophilizing an aqueous solution that comprises:

(a)    at least one therapeutic agent selected from selective COX-2 inhibitory drugs and prodrugs and salts thereof, in a therapeutically effective total amount constituting about 30% to about 90% by weight […]

(c)    other parenterally acceptable excipient ingredients in a total amount of zero to about 10% by weight of the composition, excluding water […].

There was a dispute before Burley J as to how to treat residual water in the lyophilisate in the calculation of the percentages of the constituents by weight.  As a matter of claim construction, the Pharmacia parties contended that any residual water must be taken into account in calculating the total weight of the composition.  Juno disagreed. 

Justice Burley observed that lyophilisation, although designed to remove all water, in practice will likely leave some residual water in the lyophilizate and the skilled addressee would appreciate that.  As a result, Burley J held that any residual water in the composition must be taken into account when considering the total weight of the composition of claim 1 (and dependent claims), but not claim 26 (and dependent claims) because it (and they) expressly excluded water.  Because of the water content in three of the exemplar batches, Burley J found they infringed.

Juno’s arguments on appeal

Infringement

Juno’s primary argument on appeal was that the construction by the primary judge was unworkable and uncertain because the amount of residual water in a lyophilisate can vary substantially between batches, so residual water is an “uncontrolled variable”.

Juno submitted that the skilled addressee would not understand the breadth of a claim for a composition to vary depending upon manufacturing efficiency (and that, on the primary judge’s construction, infringement of claim 1 could not be ascertained until after the lyophilisation and testing process).  Juno submitted this was supported by claim 26, which plainly excluded water.

The Full Court rejected Juno’s arguments.  Their Honours held that: (a) as a matter of plain language, if a composition contains some residual water, the “weight of the composition” includes the weight of that residual water; (b) “excipient” was defined to mean “all non-therapeutically active components of the composition except for water”, meaning water was a component but not an “excipient”; (c) the specification teaches that a powder composition will likely contain residual water; (d) claim 26 (unlike claim 1) explicitly excluded water, and claim 1 should be construed differently; (e) the variability of water content is expected to be within a tight range of variables that the skilled addressee can moderate; and (f) the specification confirmed that “comprising” was used exhaustively. 

The Full Court therefore affirmed Justice Burley’s conclusion that three of Juno’s exemplar batches fell within the claims. 

Validity

Their Honours also rejected Juno’s case that the invention was obvious in light of the prior art (noting pre-Raising the Bar law applied). 

Juno had advanced a development pathway of eight steps.  The primary judge had concluded that: (a) a pain specialist would identify one piece of prior art; (b) that prior art would be supplied to the formulation expert, who would identify another piece of prior art; (c) the formulator would decide to attempt to formulate parecoxib in light of that prior art; (d) the formulator would not obtain parecoxib sodium; and (e) in any event, the formulation of parecoxib sodium would involve dead ends and blind alleys, and the formulator would use mannitol as a buffer (and therefore, its formulation would fall outside the claims of the patent).

The Full Court concluded that Burley J did not err in drifting from the hypothetical to the practical in considering whether the formulator could have obtained parecoxib sodium, where Juno had asserted that the formulator would do so, and the Pharmacia parties contested that.

In any event, his Honour went on to consider subsequent steps in Juno’s posited development pathway and concluded that the case failed in relation to those steps, meaning that any error was immaterial.

Although the Full Court confirmed that a skilled addressee might take multiple pathways simultaneously, the fact that lyophilisation is a routine process does not mean that that it was routine to depart from the rule of thumb that a skilled addressee would likely add a bulking agent as about 10% solid material. 

For that reason, the Full Court concluded that his Honour was correct in concluding that, if a skilled addressee moved to lyophilisation, they would first try a formulation with mannitol (which fell outside the claims).  Abandoning the rule of thumb, and not using a bulking agent at all, would be one of a number of possibilities, but not one that would be taken as a matter of routine, nor one the skilled addressee would be directly led as a matter of course to try.

This case confirms that, although more than one thing may be obvious, it is still necessary to demonstrate that the skilled addressee would reach the invention as matter of routine.  A course which involves too much trial and error is unlikely to be held to be obvious.

In light of their conclusion that the claimed formulation was not obvious, their Honours did not reconsider whether the skilled addressee would have ascertained the prior art.

The Pharmacia parties’ appeal

Infringement

The Pharmacia parties submitted that the word “about” in the claims should be understood to allow for a 5% margin of error, so that “about 90%” extends to 94.5%, not the nearest whole number. 

The Pharmacia parties relied on Smith & Nephew Plc v ConvaTec Technologies Inc [2015] EWCA Civ 607 to argue that even without “about”, whole numbers would be understood to the nearest whole number.  The Full Court rejected this submission, which related to a different specification and to different expert evidence.   The Pharmacia parties also submitted that if “about” related to numerical rounding, one of the examples would be excluded.  However, the Full Court rejected this argument on the bases that: (a) a skilled addressee would not look to the formulations to determine the meaning of “about”; and (b) the expert evidence that the exemplified formulation fell outside the claim was based on unexplained assumptions. 

Their Honours held that Burley J was correct to conclude that “about” meant “near, close to”, and that the scope for variance was not intended to be large, particularly since the claims themselves provide variances, and since “about” was also used, in relation to pH ranges or the length of lyophilisation cycles.  The Full Court concluded that “about” was plainly intended as an approximating term, not one which was plus or minus 5%.  The authors consider that this outcome reflects the sensible principle that the construction of “about” will depend on the patent and expert evidence. 

Their Honours concluded that Burley J had not erred in allowing Juno to amend its defence to specify that the exemplar batches were not supplied for sale in Australia, and to open up the prospect that neither phosphoric acid nor sodium hydroxide was added to Juno products sold or supplied in Australia, in circumstances where the Pharmacia parties had not complained that it would be prejudiced or sought to adjourn the trial or adduce further evidence.

Declarations

The Full Court held that Burley J did not err in making declarations that some exemplar batches fell within the claims, rather than that parecoxib products corresponding to the infringing exemplar batches had been sold in Australia. 

The Full Court said declarations referring to goods “corresponding to” the infringing exemplar batches should not be made, as that would be ambiguous, and declarations must be clear, particularly where the manufacturing process meant only some exemplar batches.

Costs

The Full Court held that Burley J did not err in deferring the question of costs where the question of whether any of the batches which were imported and sold infringed. 

His Honour noted that the Pharmacia parties, who accepted the exemplar batches as representative, were unaware that potentially no batches imported would infringe, but noted that it was not irrelevant that the Pharmacia parties might fail to establish an entitlement to pecuniary relief.  In the circumstances, Burley J deferred consideration of costs until the issue of quantum had been resolved, so this issue could be explored.  The Full Court observed that this was a sensible course to adopt.   

Print Friendly, PDF & Email

You may also like...

Leave a Reply

Your email address will not be published. Required fields are marked *