Sandoz v Teva UK: revocation of UK formulation patents for blockbuster drug, Apixaban
Eliquis®, the Bristol Myers Squibb (BMS) and Pfizer apixaban product, was recently ranked in the top five pharmaceutical products sold worldwide, with a significant increase in demand resulting from the Covid-19 pandemic. Enforceability of BMS and Pfizer’s patent portfolio will clearly be critical to the ongoing success of this blockbuster drug.
Following revocation proceedings brought by Sandoz and Teva Pharmaceutical Industries (Teva), earlier this year the High Court of England and Wales invalidated BMS’s basic European (UK) apixaban patent and supplementary protection certificate (SPC), for lack of plausibility (that is, Agrevo obviousness and/or insufficiency).
Recently, the High Court has also assessed the obviousness of four of BMS and Pfizer’s European (UK) patents apixaban formulation patents, following further revocation proceedings brought by Sandoz and Teva. The European patents belong to the same family, and have all recently been revoked by the EPO’s opposition division. However, these decisions are either under appeal or awaiting appeal.
The key claim to be addressed in the UK proceedings was claim 1 of EP3246021B (EP’021), in a proposed amended form. This claim was directed to a tablet comprising up to 5 mg crystalline apixaban particles having a specified particle size (D90 < 89 µm), and a diluent/carrier; the formulation having a particular dissolution rate (≥ 77% drug dissolved within 30 minutes, measured under standard conditions). It was accepted that the validity of the other three patents would stand or fall with EP’021.
On that basis, the skilled team was taken to include a clinician, who would define the desired formulation; and a formulator, who would aim to prepare it.
At the effective date, apixaban was well-known to be a promising anti-coagulant alternative to warfarin, and had reached advanced stage clinical trials.
The parties agreed that a review article concerning apixaban’s development would motivate the clinician to recommend preparation of 2.5 and 5 mg immediate release tablets, which had been used in the described clinical trials.
Starting from the review article, the necessary steps to the claim were taken to be the choice of particle size and dissolution rate. The remaining features were not considered to contribute to inventive step.
For most tablet-formulated drugs to reach the systemic circulation, the tablet must first disintegrate in the stomach; then, the active ingredient must dissolve into gastrointestinal fluids, and permeate across tissue membranes at the gastrointestinal tract’s absorption site.
The parties agreed that, at an early stage in development, the formulator would routinely assess the equilibrium solubility of apixaban at the recommended doses. Accordingly, they would understand apixaban to be a class III drug (high solubility and low permeability) under the biopharmaceutics classification system (BCS).
Although equilibrium solubility tends to correlate with dissolution rate, they are different; sometimes a soluble drug is slow to dissolve, risking limited drug absorption.
BMS and Pfizer’s expert believed that the skilled team would be so optimistic about the dissolution rate of a class III drug that it would not be tested. However, Mr Justice Meade disagreed. Given the complexity and expense of drug development, and risk to future process changes and biowaivers, the judge decided that the formulator would not leave poor bioavailability up to chance. Instead, the formulator would aim to ensure a fast dissolution rate (such as 85% dissolution in 15 or 30 minutes) at an early stage.
Although a problem with the dissolution rate could not be predicted, assuming the formulator discovered a problem, the judge concluded that they would arrive at the claimed formulation on the basis of their common general knowledge alone. The formulator would first check the disintegration rate, and make any necessary improvements. An obvious way to improve any remaining dissolution issue was to reduce the particle size; and the claimed size (D90 < 89 µm) was well within the range typically used (~10 to 100 µm).
Other obvious solutions were available, such as excipient optimisation. However, this was not considered to render the claimed solution any less obvious.
Conversely, if there was no problem with the dissolution rate, the claim makes no technical contribution.
Therefore, Mr Justice Meade found the UK designations of the four European patents to be invalid.
The judge was aware of the opposition division decisions. He acknowledged that different prior art had been taken into consideration at the EPO, but recognised that his reasoning is broadly consistent with that of the opposition division.