For the first time the Swiss Federal Patent Court (SFPC) referred to the reasonable expectation of success as a criterion to assess the inventive step. However, this test is not yet well settled. With this PharmaCircular we would like to inform you about the most recent case law on obviousness in Switzerland.

An inventive step is acknowledged if the invention is not obvious to a person skilled in the art when considering the state of the art. To assess the inventive step, the could-would approach would have been established in the case law of the European Patent Office.

According to this test it is not sufficient that the skilled person could have arrived at the invention based on the prior art. Rather, it is required that he would have done so. An invention, however, is not only obvious if the solution was predictable to the skilled person, but also if the success could realistically be expected.

Reasonable expectation of success

In its judgement in the matter of Novartis AG vs. Mepha Pharma AG regarding a fix dose combination with amlodipine and valsartan the SFPC adopted the could-would approach as developed by the European Patent Office. It held that the questions whether the skilled person would find the claimed solution depended on the skilled person's expectation of success. A reasonable expectation of success, however, should not be accepted lightly. Although a certainty of success is not required, a mere hope to succeed is not sufficient. Even an expectation of success of 50 percent is not more than a mere hope to succeed.[1]

The patent in suit concerns a pharmaceutical combination composition with valsartan and amlodipine and a pharmaceutical carrier for the treatment of hypertension. The patent acknowledges that the lowering blood pressure effect of valsartan was known in the prior art. In addition, different active ingredients with different mechanisms of action were combined for the treatment of hypertensive vascular diseases.

The scientific study of Corea et al reports on the reduction of blood pressure and side effects of a combined administration of amlodipine and valsartan. The purpose of this study was to compare the antihypertensive efficacy of valsartan, a then new angiotensin II antagonist, in a reference therapy with amlodipine. The study was made with 168 patients with light or moderate hypertension for 12 weeks. Patients whose blood pressure was uncontrolled after 8 weeks of mono­therapy with valsartan received for the remaining 4 weeks a combination treatment with 80 mg valsartan and 5 mg amlodipine once daily.


Based on the results of this study the SFPC held that the skilled person would have noticed no negative effect of this combination on the hypertension. In addition, there were less side effects associated with the combination therapy than with the treatment of valsartan or amlodipine as such. Consequently, based on Corea the SFPC held that the combination with amlodipine and valsartan had been used to treat patients with hypertension and that there were no fundamental prejudices against this treatment. Therefore, there was experimental evidence in the prior art for the efficacy of the combination treatment. Starting from Corea as closest prior art the inventive problem was at best to provide an alternative form of administration. In light of the general knowledge, this was not enough for the finding of an inventive step.

Under these circumstances it remains unclear what the court intended with its general remarks about the reasonable expectation of success of the skilled person. To adjudicate the inventive step, the court did not have to evaluate the expectation of success.


With its reasoning, however, the court raises the question whether it has the same understanding of the reasonable expectations of success as the European Patent Office. This although in its reasoning the SFPC explicitly refers to the case law of the European Patent Office. Unlike the assumption of the SFPC, however, the EPO originally introduced the requirement of the reasonable expectation of success to reduce the threshold for obviousness. According to the case law of the board of appeals it is not required to establish empirically what could be expected reasonably. It is obvious to try, if the skilled person can realistically expect a success.

This is different in Switzerland, where the SFPC imposes stringent requirements regar­ding the skilled person's expectations of success so that not even 50% would suffice.

Besides, the expectation of success does not answer how this 50% should be measured. If the probability of efficacy of a new pharma­ceutical ingredient is remote, such as in oncology, the consequence of the 50/50 rule would be that hardly ever an invention would be obvious. This is unlikely intended by the "inventor" of this rule.