Genentech’s patent relating to a dosage regime for Herceptin® has been found invalid following a revocation action by Hospira.


Hospira are interested in selling the breast cancer drug trastuzumab, marketed by Genentech under the trade name Herceptin®. Trastuzumab is a monoclonal antibody that targets the HER2 receptor (a protein on the surface of a cell that affects the cell’s growth) and is approved for use in the treatment of HER2+ breast cancer (i.e. breast cancer in which the breast cancer cells have a high number of HER2 receptors causing the cells to divide and grow quickly).

Genentech’s SPC for its basic trastuzumab patent is due to expire in July 2014, but Genentech had additional patent protection for trastuzumab in the form of EP 1210115, which relates to dosage regimes, and EP 1308455, which relates to a composition of trastuzumab with low levels of impurities. Hospira brought a revocation action against both EP 1210115 and EP 1308455. The following discussion focusses on EP 1210115.

EP 1210115 claimed an initial dose of 8 mg/kg of body weight and a plurality of subsequent doses of 6 mg/kg of body weight; the doses being administered every three weeks, summarised as 8 + 6 q3w. Hospira asserted that this claim was invalid for obviousness and/or lack of sufficiency.


The skilled person was considered to be a team comprising an oncologist and a pharmacokinetics expert, each member of the team having their own common general knowledge. The US Food and Drug Administration (FDA) label for Herceptin® shows that the drug had been approved for a dosage regime of 4 + 2 q1w (i.e. an initial dose of 4 mg/kg of body weight and a plurality of subsequent doses of 2 mg/kg of body weight; the doses being administered every week) in combination with paclitaxel (another breast cancer drug), which is administered on a three weekly schedule. The FDA label also contains pharmacokinetic data that shows that the serum half-life for trastuzumab is dose dependent.

The judge considered that the oncologist would consider the possibility of administering trastuzumab every three weeks to make it easier to administer trastuzumab at the same time as paclitaxel. Since the half-life for trastuzumab is dose dependent, the judge considered that the pharmacokinetics expert would consider trialling higher doses of trastuzumab than the FDA approved doses every three weeks. Furthermore, the judge considered that choosing 8 mg/kg for the initial dose and 6 mg/kg for the subsequent doses would be obvious in view of the serum half-life data and hence the patent was found to be obvious.


For a patent to be found sufficient it must be plausible or credible to the skilled person that the invention will work across the scope of the claim. In this case, where the effectiveness of the drug in treating breast cancer had already been established, it had to be plausible or credible to the team comprising an oncologist and a pharmacokinetics expert that the claimed dosage regime would be effective in treating breast cancer. It was noted that the patent did contain pharmacokinetic data for the approved dosage regime of 4 + 2 q1w, but it did not contain pharmacokinetic data from trials using the claimed dosage regime.

Genentech’s technical expert, Professor Boddy, considered that a pharmacokinetics expert could use the pharmacokinetic data derivable from the FDA label and in the patent to model the claimed dosage regime of 8 + 6 q3w. Professor Boddy submitted models of the pharmacokinetic data and stated that, on the basis of the models he had prepared, a pharmacokinetics expert would advise an oncologist to enrol patients on a trial of the claimed dosage regime. Hospira pointed to apparent inconsistencies between Professor Boddy’s models and the FDA label. A model submitted by Hospira’s technical expert, Dr Earhart, was also inconsistent with the FDA label. The judge concluded that, partly due to problems and inconsistencies between the available models and the FDA label, the skilled person would not conduct a trial of the claimed dosage regime based on the information in the patent. Therefore, even if the patent had not been found to be obvious, it would lack sufficiency.


The judge briefly commented on priority, stating that he was satisfied that the claimed dosage regime was disclosed in the priority document but that the priority claim was invalid because the priority document was not an enabling disclosure since it lacked sufficiency for the same reasons that the patent lacked sufficiency.


This decision demonstrates that although dosage regimes are not excluded from patentability, it may be difficult to defend the inventive step of a claimed dosage regime in court. In particular, for an approved drug, any data published as a consequence of the approval process (e.g. an FDA label) may prove to be highly relevant prior art. In addition, where a drug has been approved for administration with another drug, common dosage regimes for the co-administered drug may be considered highly motivating to the skilled person.

This decision also raises concerns regarding how much information needs to be included in a priority application for a dosage regime to be considered sufficiently disclosed. Since clinical trial data is often made public, trial data could be novelty destroying prior art for a patent application and yet this decision suggests that such data may be required in a priority application. Hence the timing of a trial and the filing of a priority application may need to be very carefully managed.

Following this decision, if an applicant wishes to obtain patent protection for a dosage regime for an approved drug, they should carefully consider the inventive step of the dosage regime itself and try to obtain trial data for the regime to help bolster inventive step. Ideally, the trial data should be included in the priority application to ensure sufficiency of disclosure and a valid priority claim.