T 0689/19 - (I) Dasatinib III (inventive step)

Key points

•  This opposition appeal concerns dasatinib. It is (at least) the third appeal decision about dasatinib. It is the second appeal in the opposition proceedings. In the first appeal, T 0950/13, the Board found the recited medical use to be sufficiently disclosed without examples in the application as filed.
• " The patent in suit aims at treating certain types of cancer, in particular chronic myelogenous leukemia, by oral administration of [dasatinib]" 
• The decision under appeal and all appellants [opponents] rely, inter alia, on either document (5) or (18) as closest prior art.
• It is common ground that the difference between the subject-matter of claims 1 or 3 of auxiliary request 2a and the disclosure of these two documents is the use of dasatinib as active agent.
• "  it is common ground between the appellants and accepted as one possibility by the respondent that the technical problem is the provision of an alternative treatment for CML.' 
• " The board draws attention to the primary finding of T 950/13, namely that the subject-matter of claims 1 and 4 of the then-main request, which is identical to the subject-matter of claims 1 and 3 respectively of auxiliary request 2a, is sufficiently disclosed. These claims define the treatment of CML with dasatinib. The board in T 950/13 found that there was a plausible technical concept to support this treatment. The fact that the board went on to find that such a plausible technical concept was not disclosed for the treatment of a specific patient group, i.e. patients suffering from a cancer resistant to imatinib treatment, does not change this general finding." 
• The Board in T 950/13: Claim 2 is directed to treatment of CML cancer that is resistant to imatinib. " The application as filed contains no information at all, neither in the form of experimental data nor in the form of a plausible technical concept, that dasatinib is suitable in the treatment of those patients with imatinib-resistant CML. The functional analogy to imatinib as BRC-ABL kinase inhibitor is not helpful in this context and cannot explain why dasatinib should be active, when imatinib is, or has become, inactive." "
• The opponents argue that it is a problem that claim 1 encompasses this embodiment.
• The Board " According to [G1/03, point 2.5.2], the inclusion of non-working embodiments is of no harm if there are a large number of conceivable alternatives and the specification contains sufficient information on the relevant criteria for finding appropriate alternatives over the claimed range with reasonable effort. In T 950/13, the board did not state that the imatinib-resistant patient could not be treated, rather, it concluded that there was no technical concept for such a treatment in the application as filed. In the context of the discussion of inventive step, the following applies. A lack of a plausible disclosure for one group of patients in the application as filed does not automatically lead to the finding that the treatment does not work for this group of patients. It seems to be paramount to distinguish in this respect between the concepts of lack of disclosure and non-working embodiments." 
• "While a plausible technical concept might not be disclosed in the application as filed for a specific embodiment (in the present case imatinib-resistant patients), this does not automatically imply that this specific embodiment has to be classified as a non-working embodiment, or, seen in the context of the problem-solution approach, as an embodiment that does not solve the technical problem." 
• " In the case at hand, the board identified (the disclosure in the application as filed of) a technical concept for the general treatment in T 950/13. No such technical concept could be identified in the application as filed for the specific embodiment concerning imatinib-resistant patients. However, the reasons underlying the failure of a treatment are not necessarily linked to the reasons for not accepting the disclosure of a technical concept. From this, it follows directly that a lack of disclosure of this concept does not automatically equate with a failure of treatment. " 
• " In sum, the lack of a plausible disclosure regarding the imatinib-resistant patients cannot be equated with a finding that the claimed subject-matter includes non-working embodiments. No non-working embodiments have been identified by the appellants. Consequently, the problem is to be considered as solved over the whole scope of the claims." 
• " It remains to assess whether its solution would have been obvious to the person skilled in the art."
• " Document (1) relates to certain cyclic compounds and their use in treating protein tyrosine kinase (PTK)-associated disorders such as immunological and oncological disorders " 
• " One of the cyclic compounds is dasatinib (example 455). In total, 580 compounds are exemplified and characterised by their HPLC retention time. For some of the compounds, information concerning the solid state and colour is given. Under the heading "utility", inhibition of members of the Src family is linked to diseases relating to immunological disorders, and inhibition of HER1 or HER2, which are receptor PTKs, is linked to anti-angiogenic uses such as the treatment of cancer and diabetic retinopathy (page 39, line 10 to page 40, line 9). A list of potential diseases to be treated is given on page 40, line 19 to page 43, line 25. It is stated that the compounds described in the examples have been tested in one or more of the enzyme assays (using Lck, Fyn, Lyn, Hck, Fgr, Src, Blk, Yes, Her1, Her2) and have "shown activity" (page 50, line 1 to page 51, line 30). The treatment of chronic myelogenous leukemia or the inhibition of Bcr-Abl is not disclosed. None of the exemplified compounds has been shown to have any activity for inhibiting any of the PTKs mentioned." 
• " Document (1) [WO00/62778] lists various PTKs, belonging to both main families of PTKs. A Markush formula and several (580) specific compounds are disclosed. However, none of the compounds has been linked via its inhibitory activity to any of the kinases mentioned. Consequently, the person skilled in the art would have been faced with a screening project for "pairing" any of the compounds with one or more of the kinases mentioned. In the absence of any guidance as to which structure/compound might inhibit which PTK, such a screening project goes beyond the routine activities performed by a skilled person when trying to find alternative compounds inhibiting specific (either Bcr-Abl or dual Bcr-Abl and Src) kinases for the treatment of CML. The disclosure of document (1) can at best be seen as an invitation to perform a research programme." 
• D1 was the PCT application leading to Dasatinib I, T0488/16, where the Board found the compound claim to lack inventive step for lack of plausibility. 
• The claims are found to be inventive.

EPO 

The link to the decision is provided after the jump, as well as (an extract of) the text of the decision.

https://www.epo.org/law-practice/case-law-appeals/recent/t190689eu1.html

5. Inventive step

The patent in suit aims at treating certain types of cancer, in particular chronic myelogenous leukemia, by oral administration of a compound of formula IV (paragraphs [0001], [0007] and [0079]). The compound of formula IV is known as "dasatinib".

5.1 Documents (5) or (18) as closest prior art

5.1.1 The decision under appeal and all appellants rely, inter alia, on either document (5) or (18) as closest prior art.

Document (5) is entitled "Bcr-Abl inhibition as a modality of CML therapeutics". The disclosure of this document thus focuses on the treatment of chronic myelogenous leukemia, especially in the context of Bcr-Abl inhibition. After discussing Bcr-Abl as a therapeutic target in the treatment of chronic myelogenous leukemia (page M11, right-hand column, first paragraph), the document goes on to describe the design of a potent, selective and orally active protein kinase inhibitor class based on a central phenylamino-pyrimidine template (see Figure 1A and paragraph bridging pages M11 and M12). Two further advantageous groups are identified, see Figures 1B and 1C, which are included in the most promising compound, STI-571 (imatinib). Then the in vitro and in vivo profile of imatinib and clinical studies (relying on oral administration) carried out with imatinib are discussed, followed by details concerning resistance to imatinib (chapters 4 to 6). The structural basis of imatinib specificity is linked to the distinctive inactive conformation of the activation loop of Abl (page M17, left-hand column, second paragraph). In the "Conclusions" hopes for a new generation of specific, targeted therapies in oncology following the development of imatinib are expressed.

Document (18), in the context of CML and other Bcr-Abl+ leukemias, discusses possible strategies to prevent or overcome imatinib mesylate resistance. To this end mutations at the inhibitor binding pocket of Bcr-Abl or mutations destabilising the inactive conformation of Abl are effected. Two compounds, PPI (a 4-amino, 3-aryl substituted 1H-pyrazolo[3,4-d]pyrimidine) and CGP76030 (a 4-amino, 5-aryl substituted 7H-pyrrolo[2,3-d]pyrimidine), known to be Src kinase inhibitors, were found to inhibit Bcr-Abl in a concentration-dependent manner (abstract). Dual-specific Src/Abl kinase inhibitors are thus identified as having a potential role in the treatment of advanced or imatinib mesylate-resistant Philadelphia chromosome-positive leukemias (last paragraph).

5.1.2 It is common ground that the difference between the subject-matter of claims 1 or 3 of auxiliary request 2a and the disclosure of these two documents is the use of dasatinib as active agent.

Furthermore, it is common ground between the appellants and accepted as one possibility by the respondent that the technical problem is the provision of an alternative treatment for CML.

5.1.3 As a next step, it needs to be determined whether this problem has been solved over the whole scope claimed.

The appellants pointed to the finding of T 950/13 regarding the lack of plausibility of the treatment of CML which is resistant to imatinib (T 950/13, reasons 3.14). The direct consequence of this finding was, according to the appellants, that the subject-matter of the claims encompassed variants which did not plausibly solve the technical problem.

The board draws attention to the primary finding of T 950/13, namely that the subject-matter of claims 1 and 4 of the then-main request, which is identical to the subject-matter of claims 1 and 3 respectively of auxiliary request 2a, is sufficiently disclosed. These claims define the treatment of CML with dasatinib. The board in T 950/13 found that there was a plausible technical concept to support this treatment. The fact that the board went on to find that such a plausible technical concept was not disclosed for the treatment of a specific patient group, i.e. patients suffering from a cancer resistant to imatinib treatment, does not change this general finding.

According to G01/03 (OJ EPO 2004, 413, point 2.5.2), the inclusion of non-working embodiments is of no harm if there are a large number of conceivable alternatives and the specification contains sufficient information on the relevant criteria for finding appropriate alternatives over the claimed range with reasonable effort. In T 950/13, the board did not state that the imatinib-resistant patient could not be treated, rather, it concluded that there was no technical concept for such a treatment in the application as filed. In the context of the discussion of inventive step, the following applies. A lack of a plausible disclosure for one group of patients in the application as filed does not automatically lead to the finding that the treatment does not work for this group of patients. It seems to be paramount to distinguish in this respect between the concepts of lack of disclosure and non-working embodiments.

While a plausible technical concept might not be disclosed in the application as filed for a specific embodiment (in the present case imatinib-resistant patients), this does not automatically imply that this specific embodiment has to be classified as a non-working embodiment, or, seen in the context of the problem-solution approach, as an embodiment that does not solve the technical problem.

In the case at hand, the board identified (the disclosure in the application as filed of) a technical concept for the general treatment in T 950/13. No such technical concept could be identified in the application as filed for the specific embodiment concerning imatinib-resistant patients. However, the reasons underlying the failure of a treatment are not necessarily linked to the reasons for not accepting the disclosure of a technical concept. From this, it follows directly that a lack of disclosure of this concept does not automatically equate with a failure of treatment. As such automatic equating is not possible, a finding that the problem has not been solved over the whole scope requires confirmation of the existence of non-working embodiments. In the case at hand, the appellants have not identified any non-working embodiments.

Regarding the issue of whether the problem has been solved over the whole scope, a closer look at the problem under consideration might be necessary in certain situations and/or technical fields. It requires a case-by-case assessment to establish whether a problem relating to the treatment of diseases, i.e. in the field of second medical uses, has been solved. In the case of the treatment of cancers, CML being one, a 100 percent "success" of the treatment would most probably not be expected by the medical practitioner. Even the term "success" might require some interpretation. "Success" in the treatment of cancer spans from total remission to slowing down of progression. Instead, in the case of cancer, the medical practitioner would expect to be confronted with a proportion of patients, possibly even with a substantial proportion of patients, who do not respond to a treatment that has been shown to "work" in principle. As a consequence, the medical practitioner would change the medication for the given patient, but would not doubt that the treatment might provide benefits to other patients. However, such considerations are not crucial for the present case.

In sum, the lack of a plausible disclosure regarding the imatinib-resistant patients cannot be equated with a finding that the claimed subject-matter includes non-working embodiments. No non-working embodiments have been identified by the appellants.

Consequently, the problem is to be considered as solved over the whole scope of the claims.

5.1.4 It remains to assess whether its solution would have been obvious to the person skilled in the art.

All the appellants referred to document (1) as rendering the claimed subject-matter obvious. They argued that the structures of the compounds of documents (5) and (18) that were responsible for their activity (pharmacophores) would provide an incentive for the person skilled in the art to seriously contemplate the compounds of document (1). Furthermore, document (18) identified Src kinases and pointed to dual or multiple kinase inhibitors. Screening activities were within the competences of the person skilled in the art.

Document (1) relates to certain cyclic compounds and their use in treating protein tyrosine kinase (PTK)-associated disorders such as immunological and oncological disorders (page 1, lines 9 to 12). The section discussing the background of the invention lists various families of PTKs. This section goes on to mention "a wide variety" of oncological and immunological disorders and finally focuses on Lck, a cytoplasmic/non-receptor PTK of the Src kinase family, and its activity in relation to T cells. The inhibitors of the PTKs are described as cyclic compounds of formula I, which is a Markush formula with Q, a 5-membered heteroaryl ring, a 6-membered heteroaryl ring or an aryl ring in central position (page 3, line 2 to page 8, line 7). Preferred compounds have a thiazole in central position (page 13, line 18). Methods of preparation of the compounds of formula I are described in Schemes A to E and I to XI (page 14, line 4 to page 39, line 8). One of the cyclic compounds is dasatinib (example 455). In total, 580 compounds are exemplified and characterised by their HPLC retention time. For some of the compounds, information concerning the solid state and colour is given. Under the heading "utility", inhibition of members of the Src family is linked to diseases relating to immunological disorders, and inhibition of HER1 or HER2, which are receptor PTKs, is linked to anti-angiogenic uses such as the treatment of cancer and diabetic retinopathy (page 39, line 10 to page 40, line 9). A list of potential diseases to be treated is given on page 40, line 19 to page 43, line 25. It is stated that the compounds described in the examples have been tested in one or more of the enzyme assays (using Lck, Fyn, Lyn, Hck, Fgr, Src, Blk, Yes, Her1, Her2) and have "shown activity" (page 50, line 1 to page 51, line 30). The treatment of chronic myelogenous leukemia or the inhibition of Bcr-Abl is not disclosed. None of the exemplified compounds has been shown to have any activity for inhibiting any of the PTKs mentioned.

A close structural similarity in the sense of a common, or at least related, pharmacophore has not been shown to be present between dasatinib and imatinib or between dasatinib and PPI or CGP76030. Steric aspects of the molecules and aspects relating to electronic distribution within the molecules that would lead to accommodation in the kinase binding site despite the presence of different ring systems and substituents have not been invoked by the appellants. Concerning the ring systems and substituents as such, the following has been found. Document (1) identifies the thiazole group of dasatinib as a preferable element. In addition, dasatinib has a benzamide group in its peripheral parts and non-fused heterocyclic rings in the form of piperazine and pyrimidine. Document (5) clearly identifies the phenylamino-pyrimidine core of imatinib and points to a (centrally located) benzamide group as being advantageous. PPI and CGP76030 share fused heterocyclic ring systems with amino and phenyl substitution. Neither a phenylamino-pyrimidine core nor a fused heterocyclic ring system with amino and phenyl substitution can be found in dasatinib. Thus the person skilled in the art would not have been pointed towards dasatinib for reasons related to structural similarity.

Concerning the functional aspects, the following applies. Document (1) lists various PTKs, belonging to both main families of PTKs. A Markush formula and several (580) specific compounds are disclosed. However, none of the compounds has been linked via its inhibitory activity to any of the kinases mentioned. Consequently, the person skilled in the art would have been faced with a screening project for "pairing" any of the compounds with one or more of the kinases mentioned. In the absence of any guidance as to which structure/compound might inhibit which PTK, such a screening project goes beyond the routine activities performed by a skilled person when trying to find alternative compounds inhibiting specific (either Bcr-Abl or dual Bcr-Abl and Src) kinases for the treatment of CML. The disclosure of document (1) can at best be seen as an invitation to perform a research programme.

As mentioned under point 5.1.1 above, document (18) relates to dual-specific Src and Abl kinase inhibitors. The two exemplified compounds are said to have been originally characterised as Src kinase inhibitors. The passage goes on to state that both compounds inhibited Bcr-Abl in a concentration-dependent manner by overlapping binding modes (abstract). It is thus clear that document (18) does not depart from the teaching that it is the Bcr-Abl inhibition as such that is responsible for the treatment of CML. The Src kinase specificity comes in (merely) in the context of imatinib mesylate resistance. The person skilled in the art, being thus made aware that a Bcr-Abl inhibition was paramount for successful treatment of CML, would not have further investigated the compounds disclosed in document (1) despite the vague mention of Src kinase inhibition in this document.

5.1.5 The subject-matter of auxiliary request 2a is a non-obvious alternative over the teaching of documents (5) and (18).

5.2 Document (1) as closest prior art

5.2.1 The disclosure of document (1) is discussed in point 5.1.4 above.

The most promising starting point in document (1) is dasatinib. However, this compound has not been linked to any activity (in the present case, inhibition of one or more specific PTKs) or the treatment of any disease.

5.2.2 Consequently, the technical problem when starting from document (1) can be seen as identifying a medical use for dasatinib.

5.2.3 In view of the many PTKs listed in document (1) and the very general nature of the disclosure failing to link any of these PTKs to diseases or classes of diseases, it amounts to a research programme to identify such a medical use.

5.2.4 Carrying out a research programme is beyond the routine tasks of the person skilled in the art. Consequently, the person skilled in the art would not have arrived at the subject-matter of auxiliary request 2a starting from document (1).

5.3 The subject-matter of auxiliary request 2a involves an inventive step (Article 56 EPC).

Having come to this conclusion, it is not necessary to examine the more ambitious technical problems identified as problems b) and c) by the respondent.

5.4 Referral

As post-published evidence is not crucial to the decision, issues relating to T 116/18 and a possible referral relating to such issues do not have to be dealt with.

6. Article 84 EPC

Despite the issue of whether aspects relating to Article 84 EPC are open for discussion (see G 3/14), the board provides herein, in favour of appellant 1, its conclusion on appellant 1's objection.

Article 84 EPC stipulates that the claims shall define the matter for which protection is sought. They shall be clear and concise and be supported by the description.

The claims of auxiliary request 2a differ from the claims as granted, inter alia, by the deletion of dependent claims relating to CML resistant to STI-571, which are numbered 2 and 9 in the set of claims as granted.

Support for the subject-matter of the independent claims of auxiliary request 2a and for the subject-matter of the dependent claims referred to above can be found in paragraph [0079], which is present in the patent as granted and in the description found by the opposition division to be compliant with Article 84 EPC for the set of claims of auxiliary request 2a.

The board fails to see how the deletion of a dependent claim can affect the support by the description for the independent claim (on which this deleted claim depended). Furthermore, the interpretation of the independent claims is not affected by the presence or absence of any dependent claims. This is in line with the board's conclusion, see point 5.1.3 above, that the problem of providing an alternative treatment for CML has been solved. The requirements of Article 84 EPC are met.

Having come to this conclusion, issues relating to the admission of the objection and the applicability of G 3/14 do not need to be addressed.

The objection by appellant 1 is rejected.

Order

For these reasons it is decided that:

The appeals are dismissed.