14 December 2015

Patenting Precision Medicines at the European Patent Office

In this article, Tanya Heare provides an update on patenting precision medicine inventions at the European Patent Office (EPO). As well as discussing matters as they stand before the Boards of Appeal at the EPO, she also provides an insight from the EPO examiners’ perspective and includes some practical advice for potential applicants with inventions in this field.

Precision medicine is an emerging field that promises to bring radical changes in healthcare.  By making use of genetic and/or non-genetic biomarkers (which correlate with a particular disease condition or absence of a certain drug-induced side effect, for example), the right therapeutic strategy can be tailored to the right patient at the right time.

However, while this new field of medicine is incredibly exciting and holds great potential for the future of healthcare, it poses some tricky issues for the patent industry.

Medicines and diagnostics are both eligible for patent protection at the European Patent Office (EPO).  A patent, being a legal right to stop third parties from using the patented technology, affords the patentee an opportunity to commercialise its invention without direct competition, as a reward for the time and money invested in research and development.  Patents thus play a key role in healthcare, as they encourage innovation and investment in this vitally important field.

In order to be patentable, an invention must be new, amongst other things.  This means that the invention cannot have been disclosed to the public prior to the date on which an application for a patent in respect of the same is filed.

Does a patentable invention, then, lie in the treatment of a group of patients having a particular biomarker, when the biomarker was previously unidentified, but nevertheless present, in at least some of the patients that had the associated disease condition and were already being treated with the indicated drug?  A similar question applies to a biomarker that was previously known to exist, when its role or significance was not known.  In other words, can the treatment of a known disease with a known drug be considered new, and therefore potentially patentable, by virtue only of the identification of the biomarker (or of its role or significance) and, therefore, a “new” group of patients to be treated?

The Boards’ View

A developing body of case law from the Technical Boards of Appeal at the EPO is now shedding some light on this issue and, on the whole, the results are encouraging for research and development in the precision medicine field.

Indeed, although an early case saw the Board adopt a very strict stance on what constitutes a “new” patient group, the Boards in subsequent cases have taken a much more lenient approach.

For example, in an early decision1 (dated May 2000) the Board held that, in order to be patentable, the new patient group must be distinguished from the known patient group by its physiological or pathological status.  In the Board’s view this meant that (1) the new patient group could not overlap with the known patient group (e.g. sero-positive vs. sero-negative or non-haemophilic vs. haemophilic subjects), and (2) the identified biomarker (e.g. being sero-positive or being non-haemophilic) must have a real impact on the therapeutic or pharmacological effect achieved.

However, the Boards in subsequent cases have not adopted this two-part test.  They have granted or upheld patents relating to the treatment of “new” patient groups that overlap with known groups.

For example, in one decision2 (dated October 2006) the Board found in favour of a patent relating to the treatment of a particular group of hepatitis C virus (HCV) patients (antiviral treatment naïve chronic HCV genotype 1 patients with a specific virus load) with a known active agent.  A very high percentage of all HCV infections were genotype 1 infections, which were known to be associated with a high virus load.  Thus, in general, by following the established teaching to treat chronic HCV patients with the indicated active agents, patients with the indicated biomarker (the specific HCV genotype and associated virus load) would inevitably have been treated.  However, the Board reasoned that, in the patent in suit, the data convincingly showed that it was exactly the patient group as recited in the claims of the patent that profited most from the recited treatment.  On this basis, the Board held that the recited biomarker was capable of distinguishing the claimed invention from that already known.

Other decisions3,4, issued around the same time (October 2003, August 2007), have similarly found in favour of claims relating to patient groups that overlap with those known in the art.

Moreover, the most relevant recent decisions5,6 show that the Boards are continuing to take a more lenient view on novelty for this issue.

Indeed, in T108/09 (dated February 2013), the Board held that the use of fulvestrant for the third-line treatment of breast cancer in patients who had first been treated with tamoxifen and subsequently with an aromatase inhibitor was novel because, although fulvestrant was known for treating breast cancer (and notably for second-line treatment after tamoxifen failure), the third-line patients had acquired resistance to tamoxifen and subsequently the aromatase inhibitor, such that the tumours had changed from a biological perspective.  In the Board’s opinion, this was enough to define a new subgroup of disease and, hence, a new subgroup of patients.

Similarly, in T734/12 (dated July 2013), the use of rituximab (a known inhibitory antibody against tumour necrosis factor alpha (TNFa)) for the treatment of rheumatoid arthritis in patients who experience an inadequate response to a TNFa inhibitor was considered novel because, although 30-40% of rheumatoid arthritis patients are inherently TNFa inhibitor-refractory (and, as such, it was statistically plausible that the treatment of rheumatoid arthritis patients with rituximab in the past had included the treatment of TNFa inhibitor-refractory patients), rituximab had never before been administered specifically to this subgroup.  In particular, it was noted that the TNFa inhibitor-refractory patients had increased Th17 cell numbers and increased IL-17 expression, and that treatment with rituximab reduced both of these phenomena.  As such, the claimed subgroup was considered distinguishable from the rheumatoid arthritis patients treated in the past with rituximab and, therefore, new.

Accordingly, whilst claims relating to patient groups that overlap with those treated in the past are clearly now allowable in the eyes of the Technical Boards of Appeal at the EPO, it nevertheless remains a requirement that the biomarker recited in the claims must be capable of distinguishing the claimed subgroup over those patients already known in the art, in terms of their physiological or pathological status, for the claimed subgroup to be considered “new”.

That the biomarker has a real impact on the therapeutic or pharmacological effect achieved by the claimed active agent must also still be true, particularly now with respect to an inventive step.

In this latter regard, as well as being new, an invention must also be associated with a surprising technical advantage in order to be patentable at the EPO.

The conclusion from recent decisions of the Technical Boards of Appeal (T734/12 in particular) is that experimental data demonstrating a nexus between the presence or absence of the biomarker and the improvement (efficacy or safety) in the treatment of the new subgroup in particular is helpful for establishing that a surprising technical advantage and, hence, an inventive step, exists.

Thus, an important consideration for obtaining a patent with claims relating to precision medicine at the EPO should now be whether supporting data can be provided, which establish that any identified and selected subgroup was not arbitrary, but rather a result of a functional relationship between the claimed biomarker and an improved effect of the claimed treatment.

This is also in line with the (unofficial) standard for assessment of inventive step as applied by EPO examiners at present.

The Examiners’ View

The EPO examiners’ approach to claims relating to the treatment of patient subgroups has been more restrictive than the Boards’ approach to date.

Indeed, in a meeting between the EPO and the Biotech Committee of the European Patent Institute (the professional institute for European Patent Attorneys)7 in November 2011, an EPO representative essentially said that, where a patent is based on the identification of a genetic marker to treat a disease, the claimed invention can lack novelty, as one patient with the marker will inevitably have been treated, even if the so-called “prior art” does not explicitly say so.  This message seemingly filtered down to EPO examiners who, unofficially at least, were assessing precision medicine inventions by indeed considering the inevitability of one patient with the biomarker previously having been treated.

In this regard, the standard of proof for lack of novelty at the EPO is “beyond reasonable doubt”8, which is interpreted in line with statistical standards in the art.  Consequently, later claimed precision medicine inventions were traditionally considered to lack novelty (and therefore patentability) by EPO examiners if it was beyond reasonable doubt that, in the successfully treated group of patients in the prior art, at least one of them had the identified biomarker.

All was not lost in that situation, however, as a European patent for the invention could seemingly still be obtained by making explicit reference in the claims of the patent specification to the actual step of diagnosing patients having the biomarker.  The diagnostic step thus formed part of the intended therapeutic use and consequently conferred novelty on the claimed invention.  Suffice to say, therefore, that precision medicines were also patentable in the eyes of EPO examiners, provided the claims of the patent specification were drafted appropriately.

No further guidance has been issued externally by the EPO in recent times regarding the examiners’ approach to claiming precision medicine inventions.  Unofficially, however, we have been informed by a biotech examiner at the EPO that guidelines have now been prepared for internal use, to help examiners understand how precision medicine inventions should be assessed.

Our source has indicated that the standard for novelty has been set “rather low” (seemingly more in line with the Boards’ view: novelty will be acknowledged where a correlation between the claimed biomarker and therapeutic indication has not been previously disclosed) and that, for inventive step, a pathological or physiological link must be demonstrated between the biomarker and the treatment being claimed in order to show the presence of a technical effect.  Our source has unofficially advised that some examiners believe a ‘risk factor’ should not be regarded as a novel therapeutic situation.  Rather, a functional relationship should be demonstrated between the biomarker and the therapeutic application, otherwise novelty cannot be derived from the intended therapeutic use (citing Enlarged Board of Appeal decision G2/889 and Technical Board of Appeal decisions T1437/0710, T1399/042, T734/126 and T108/095 in this regard).

The question remains, however, as to what data would be required to demonstrate such a link between the biomarker and the treatment being claimed.  Our source has provisionally advised that, if there is a functional link (e.g. the biomarker is part of a pathway known to be involved in the disease), data from animal models or cell cultures might be sufficient.  If the link is rather a simple correlation, data from individual human samples would most probably be required.  Statistical analyses should also be presented, according to our source, and any generalisation in the claims (as to the class of drugs, type of sample, type of disease, up/down-regulation of/caused by the biomarker, for example) should also be supported by working examples or prior art knowledge.


Thus, Europe continues to provide a favourable forum for patenting precision medicine inventions.  The majority of questions raised at the EPO concern novelty and (more recently) inventive step, but these can seemingly be overcome by ensuring that the claims recite a biomarker that distinguishes the new patient subgroup in terms of their physiological or pathological status and that also has a real impact on the therapeutic or pharmacological effect achieved by the claimed therapeutic agent and/or regimen.

Where applicants may find it challenging to obtain precision medicine patents in other jurisdictions (such as the US or Australia, for example), they may well have success in Europe and, as such, should be encouraged to seek European patent protection in respect of their inventions.

Naturally we will advise of any developments in this field as and when they arise.


  1. Board of Appeal Decision No. T233/96:
  2. Board of Appeal Decision No. T1399/04:
  3. Board of Appeal Decision No. T836/01:
  4. Board of Appeal Decision No. T1642/06:
  5. Board of Appeal Decision No. T108/09:
  6. Board of Appeal Decision No. T734/12:
  7. epi Information, June 2012, page 37, paragraph 8:
  8. Case Law of the Boards of Appeal of the European Patent Office, 7th Ed., 2013, I-C-2.2, first paragraph:
  9. Board of Appeal Decision No. G2/88:
  10. Board of Appeal Decision No. T1437/07: