During clinical development of a drug it is not unusual to obtain approval for treatment of a subset of patients suffering from a condition (as opposed to the entire patient population). Difficulties can arise when you try to patent treatment of such a sub-population if you have previously disclosed treating the condition generally. A case in point is the recent EPO Board of Appeals decision in T0734/12 in relation to a patent (EP1613350) in which Genentech claimed use of rituximab (Rituxan®) to treat patients who were suffering from rheumatoid arthritis (RA) who were refractory to tumor necrosis factor a (TNFa) inhibitors (e.g., Enbrel®, Remicade® or Humira®). However, previous disclosures (including Genentech’s own disclosures) had suggested treating RA with rituximab.


The main claim of EP1613350 under consideration was:

Use of an unconjugated antibody which is rituximab in the manufacture of a medicament for treating rheumatoid arthritis by intravenous administration of two doses of antibody of 1000mg to a human who experiences an inadequate response to a TNF-a inhibitor, wherein the first dose is administered on day 1 of treatment and the second dose on day 15.

Other relevant issues to this case are:

  • Rituximab is approved in the EU for treating patients who are suffering from RA who are refractory to TNFa inhibitors and the approved dosage is 1000mg separated by 14 days.
  • An earlier press release by Genentech disclosed successful treatment of a population of RA patients with the claimed dosage of rituximab (with no reference to TNFa inhibitors).
  • An earlier disclosure suggested treatment of RA patients (including those refractory to TNFa inhibitors) with a different dosage regime of rituximab. The results of this study were unsatisfactory.

The opposition decision

At first instance, the EPO Opposition Division held the patent invalid over Genentech’s press release. While the disclosure does not mention TNFa inhibitors, the Opposition Division held that the population of “human[s] who experience an inadequate response to a TNF-a inhibitor” was too vague resulting in a large or complete overlap with the population described in Genentech’s press release or the population lacked a common physiological and pathological status in view of the manifold and divergent reasons causing it (T233/96, T1399/04 and G002/08). The Opposition Division then revoked the patent for lack of novelty.

The Appeal

Genentech appealed the decision and filed evidence that 30%-40% of RA patients are refractory to TNFa inhibitors and that this sub-population of RA patients:

  • is a distinct population (T19/86, T233/96, T836/01 and T1399/04) since it could be distinguished from responding patients by clinicians and/or on the basis of biomarkers;
  • has a distinct physiological or clinical situation (T233/96); and
  • is not an arbitrarily defined population since there is evidence of a functional link between the characteristics of the population and the therapy taught in the patent (T1399/04).

On a positive note for Genentech, the Appeal Board agreed and found that the claims were novel because they defined treatment of a patient group that is distinguishable from the prior art by its physiological and pathological status.

However, the Board went on to consider inventive step and their findings were not so positive for Genentech. The Board considered the problem addressed by the patent to be the identification of an alternative method for treating RA patients refractory to TNFa inhibitors (i.e., the same problem as addressed by one of the prior art documents). They went on to find that:

  • The prior art disclosed the treatment of RA patients refractory to TNFa inhibitors, thus showing that rituximab could be used to treat this patient sub-population. However the results of that study were unsatisfactory. The Board considered that this provided a motivation to change the dosage of rituximab.
  • In contrast, Genentech’s press release indicated successful treatment of all RA patients treated with the claimed dosage regime (and the success of this treatment had also been considered as “revolutionary” in an article by the British Arthritis Research Campaign).
  • Genentech’s press release discussed treating 31 RA patients and since 30%-40% of RA patients are refractory to TNFa inhibitors it was “highly probable” that at least some of the treated population fell within the claimed sub-population.

Given Genentech’s success in treating RA patients generally, the Board considered that the skilled person would have been motivated to use the successful dosage regime to treat RA patients refractory to TNFa inhibitors with an expectation of success. On this basis, the Board revoked the patent for obviousness.

It is also worthwhile noting that the Board explicitly noted in their decision that the patent provided no evidence of an improvement in the previously disclosed treatment of RA patients refractory to TNFa inhibitors (in fact, as the Board noted, the patent discloses no data showing the claimed woks, instead including a prophetic description of how such patients would be treated).

What can we learn?

While this decision applies previous EPO and Appeal decisions, it is a reminder to us of how difficult it can be to protect methods of treating new patient sub-populations using a known drug in a known dosage regime. On a positive note, the Board was willing to consider the claims as novel provided certain substantial requirements were met, and the outcome may have been different if the prior art was not so close or if the specification as filed included evidence of a surprising effect of rituximab in the claimed sub-population.

To place your application in a good position for successful prosecution, it is advisable to include at least the following in the application as filed:

  • evidence that the patient sub-population is not an arbitrary selection (i.e., they represent a population that can be distinguished from the larger population, they have a distinct physiological or clinical situation and there is a functional link between the characteristics of the sub-population and the treatment claimed);
  • evidence (i.e., data) showing that the treatment works surprisingly well in the claimed sub-population or at least a good reason as to why that the skilled person would not have applied the known method of treatment to the sub-population.

Despite the difficulties in obtaining patents for treating sub-populations in EP, the value realised from an approved therapy may make the expense worthwhile. Moreover, such difficulties are not always faced in other jurisdictions.