T 1057/92 (Factor VIIa/BAXTER) 23-01-1997

Reasons for the Decision

Novelty

1. The issue in dispute is whether the intended use made public by Dr Hedner in her oral presentation at the Rome Symposium (see document (D3)) was use of a preparation that meets the technical feature recited in claim 1 that the therapeutic composition should contain factor VIIa as the sole effective activated haemostatic agent. This claim does not exclude the presence of activated clotting factors other than factor VIIa (see column 3, lines 17-19 of the patent in suit), and thus the presence of factor XIIa. However, such contaminant factors should not exhibit any clotting activity of their own (ibidem, lines 19-29).

2. It has not been disputed by the Appellant that in her presentation at the Rome Symposium (see document (D3)), Dr Hedner said that her factor VIIa preparation contained traces of factor XIIa used in a mass ratio of 1/200 XIIa to VII for activating factor VII to factor VIIa. She did not say that factor XIIa was removed or inhibited and, thus, there is no reason for expecting that the skilled person would do otherwise in repeating Dr Hedner's teaching.

3. A person skilled in the field of blood coagulation would not consider traces of clotting factor as meaningless. This would be contrary to the common general knowledge in this field that due consideration should always be given to traces of clotting factors because they might trigger important reactions. That traces of clotting factor may be active, transpires from many documents available to the Board, such as document (D9) (see contribution 14, page 112, lines 3-5): "minute amounts of activated factors so minute that they cannot be revealed by conventional assays, might exert their proteolytic action", and document (D13), page 394, lines 17-18, which recites: "Even infinitesimal amounts of thrombin appearing in the product may exert some action". This evidence is in keeping with the passage of document (A10) (see page 375, first two lines) stating that factor VII is a trace plasma protein but that it is nevertheless involved in blood coagulation, and with document (A25) (see legend to Table 4 on page 299) showing that traces of thrombin (factor IIa) at a concentration of 0.015. units/ml activate factor VIII. Further, owing to the enzymatic and hence catalytic nature of clotting factors, investigators in this field often take care to inhibit traces of extraneous clotting factors for avoiding false results (see for example, document (D8), page 25, lines 7-8 from the bottom).

4. Dr Hedner and Dr Kisiel stated in their publication (D11) (published after the priority date of the patent in suit) (see page 1837, r-h column, lines 5-7), which comments on the experiments with two dogs, orally reported by Dr Hedner in Rome in 1981 (document (D3)): "No attempt was made to remove the trace amounts of factor XIIa present in the factor VIIa preparation. Accordingly, these studies measured the thrombotic potential of Factor XIIa as well as Factor VIIa" (emphasis added). Thus, two experts in haematology such as Dr Hedner and Dr Kisiel lent due consideration to traces of factor XIIa and did not a priori exclude that these trace amounts of factor XIIa could trigger a thrombotic episode in the hosts.

5. In their published work, Dr Hedner and Dr Kisiel do not appear to endorse the appellant's proposition that traces of factor XIIa are meaningless. For the purpose of establishing whether trace amounts of factor XIIa might trigger a clotting reaction, it should be noted that the Board does not see any substantial difference between thrombotic potential and haemostatic potential. A thrombogenic effect can in substance be seen as an unwanted and uncontrollable systemic activation of the coagulation cascade as opposed to haemostasis, which is a therapeutic local clotting process. Depending on the circumstances, a given clotting factor may trigger either haemostasis or a thrombogenic episode. For instance, document (D7), page 1037, lines 1-2, states that factor VIIa might have "therapeutic or thrombogenic potential".

6. In view of this, the Board is satisfied that the composition, which Dr Hedner said she intended to use, would not meet the requirements of claim 1. Accordingly Dr Hedner did not even disclose an intention to use a preparation which would fall within claim 1. Her disclosure in Rome does not destroy the novelty of claim 1.

7. Regarding the alleged novelty destroying use on patients in Stockholm, this was a matter on which the Opposition Division refused to allow into the proceedings written evidence submitted only at the oral proceedings before them by the Other Party, and refused to hear evidence on this point from Dr Hedner, pursuant to their discretion to refuse belated material pursuant to Article 114(2) EPC. In the course of its appeal (since withdrawn) the Other Party again submitted this evidence as well as another affidavit from Dr Hedner dated 7 September 1992, in which she stated that she made the first treatment of a patient with the preparation she had disclosed in Rome "..on 23rd, 24th and 27th April 1981, as will be seen from the clinical record sheet from my hospital, a copy of which is exhibited hereto as Exhibit UH4. Many persons in my hospital knew what I had done and achieved. There was no need to keep the treatment secret. I had already announced at the Rome Symposium that I would be doing this and that I had suitable patients around all the time." The Appellant relied also on the arguments and evidence of the Other Party, so the Board considers this matter. Unlike for Dr Hedner's disclosure at the Rome Symposium, where there is evidence from witnesses of all parties that Dr Hedner's disclosure was made publicly available, the Board cannot accept that there is adequate evidence before it that the results of the trials at the hospital were made available to the public prior to the publication of Dr Hedner's results (document (D11)) in 1983 after the priority date. The trials at the hospital (unlike the disclosure at the Rome Symposium) were not made for the purpose of informing the public. In these circumstances, the Board cannot accept evidence that the person carrying out the trials believed that she was making the results available to the public already when carrying out the trials, and not only later when publishing her paper, as evidence that it was indeed made available to the public at the time of the trials. Just because Dr Hedner knew exactly what trials she was carrying out, she is not the appropriate witness to state what, if anything, was made available to the public as a result of the trials. She was not in the position of a member of the public, and thus could only make conjectures about what knowledge the public might have acquired. The Board thus does not take these trials in the hospital on human patients into account for the purpose of assessing novelty or inventive step.

8. The findings of points 6 and 7 above also allow novelty to be acknowledged for claims 2 to 15 on the same basis. Although none of method claims 12 to 15 explicitly requires the removal of any activated haemostatic agent which is not factor VIIa, this restriction is implicit by virtue of the direct or indirect reference in these claims to the therapeutic composition according to claim 1, which contains factor VIIa as the sole effective activated haemostatic agent. The method recited in these claims is said merely to "comprise" the explicit steps, so that the inactivation or removal of other haemostatic agent are not excluded.

Inventive step

Introduction

9. Before the priority date of the patent in suit, it was known that blood coagulation was an exceedingly complex process in which a number of blood coagulation factors interacted in a series of step-wise reactions, whereby an inactive proenzyme was converted to the active enzymatic form, which in turn converted another proenzyme to its active form (the so-called blood coagulation cascade), the cycle continuing until, as a final step, fibrinogen was converted into fibrin and the fibrin cross-linked. This process was even more complex owing to the presence of modulators and inhibitors taking part in the coagulation cascade, and feed-back effects (ie, some "downstream" clotting factors activating one or more "upstream" factors of the earlier coagulation phases). The coagulation cascade comprised two arms, namely the intrinsic pathway and the extrinsic pathway which converged at the activation of factor X, then merged into a final common pathway up to the end of the coagulation process. Thus, two enzymes were established as activators of factor X to factor Xa, one being the complex of factor VIII with factor IXa (intrinsic pathway) and the other being the complex between tissue factor and factor VIIa (extrinsic pathway). In the extrinsic pathway, factor VIIa activated factor X into factor Xa without participation of factor VIII (ie, by-passing factor VIII). It was also known that factor VIIa was only active in converting factor X into factor Xa when complexed with the so-called tissue factor, a lipoprotein exposed at the site of tissue damage. Thus, factor VIIa was inactive unless anchored to tissue factor at the site of injury. Therefore, factor VIIa was only active in the extrinsic coagulation system, which was intact in haemophilic patients. In fact, despite their having normal levels of factor VII, patients with haemophilia bled because clotting initiated by tissue factor (ie, the extrinsic pathway) could not compensate for their defect in intrinsic clotting (document (D6), page 837), and thus a normalization of the intrinsic coagulation system was highly desirable.

In the 1960s and 1970s, coagulation factor concentrates became available for treating haemophilia A and B patients. However, about 15% haemophiliacs turned out to develop antibodies against factor VIII or IX. Much effort was placed on finding agents capable of inducing haemostasis by activating the final common pathway of the coagulation cascade, independently of the presence of factor VIII. Activated prothrombin complex concentrates (APCCs) such as Autoplex® and FEIBA®, the latter being an acronym for Factor Eight Inhibitor By-passing Activity, were examples of said agents capable of by-passing factor VIII activity. They comprised a mixture of clotting factors (mainly factor II, ie, prothrombin, and factors V, VII, IX and X) in both activated and zymogen forms and other unknown components and provided some benefit to patients having deficiencies or inhibitors of blood clotting factors. Such a mixture the Board regards as the closest prior art.

Problem to be solved

10. However, such a prior art mixture besides loading the patient with unnecessary exogenous proteins, could give rise in some patients to dangerous thrombogenic side-effects such as disseminated intravascular coagulation (DIC) or thromboembolic complications. The problem therefore arose of providing a blood clotting preparation that was effective in inducing haemostasis in patients suffering from deficiencies of blood clotting factors or from the presence of inhibitors to blood clotting factors, and had no or fewer side effects than such a mixture.

Solution to the problem

11. The above problem is solved by the therapeutic preparation of claim 1 of the patent in suit comprising factor VIIa as the sole effective activated haemostatic agent. The examples of the patent in suit indeed show that when factor VIIa alone is administrated to haemophilic dogs with bleeding episodes, haemostasis is achieved without giving rise to unwanted thrombogenic side effects. Therefore, the Board is satisfied that the present invention solves the problem as previously set out.

12. Although published after the priority date (1982), document (D9), the proceedings of lectures and papers given at the Rome Symposium on Activated Prothrombin Complex Concentrates held on 30-31 March 1981, can reasonably be taken as truly reflecting what the skilled persons were thinking about the possible by-passing agent(s) in APCCs two months before the priority date of the patent in suit. This document comprises contributions by six experts or teams taking part in the Rome Symposium and reporting possible theoretical explanations for APCCs efficacy in by-passing factor VIII activity. It emerges from document (D9) that the scientific community had not yet identified the by-passing agent(s) responsible for effectiveness of APCCs in achieving haemostasis. This is shown by the statements made by the experts attending the Rome Symposium on pages 71, 95, 101 to 102 and 115 of the document, according to which there was wide disagreement as to which factors alone or in concert with other unknown serine proteases were responsible for the in vivo action. There were many possible mechanisms for haemostasis correction. These mechanisms involved a novel factor Xa-like activity (page 80), a factor VIIICAg-factor IXa-phospholipid complex (paragraph bridging pages 80 and 81), a protease acting through platelets (page 93), a factor VIII coagulant antigen (page 107, lines 3 to 4), an hitherto unidentified factor (or complex) acting directly on the late phase of blood coagulation and activating factor X (page 112, last two lines) or intermediary forms of coagulation factors such as prethrombin 1, prethrombin 2 or some form of factor IXa (page 115, second full paragraph). Document (D9) also mentions that some authors (ie, Dr Seligsohn's team) believed that factor VIIa might play a role in haemostasis correction (paragraph bridging pages 106 and 107). Attractiveness of factor VIIa as a candidate for the effective ingredient of APCCs is also emphasized by document (D15) (see page 118, second paragraph). However, the author thereof also underlines experimental facts against this theory. He came to the conclusion that it was impossible at that time to identify the in vivo active components (ibidem, last lines). A major problem to be overcome lay in fact with elucidating whether the activated factors assayed in the plasma after infusion of the APCC were an expression of interactions or simply persisted unmodified in the circulation after infusion. There was no line of action suggested for which there was any reasonable expectation of success, as opposed to a mere hope for success.

13. The Appellant has argued that as the haemostatic properties of factor VII were known since 1956, while the concept of the blood coagulation cascade with its intrinsic and extrinsic pathways, had been introduced in 1964, so the skilled person wishing to by-pass the extrinsic pathway would increase factor VII above physiological levels by added factor VII. The Board disagrees with this line of argument. First, the patent in suit is concerned with factor VIIa, not with factor VII. The use of factor VII is in no way suggestive of factor VIIa since the two molecules perform different functions. Secondly, in the skilled person's mind, the mechanism of factor VIII by-passing activity could have reasonably been explained by the presence in APCCs of one or more activated component(s) from below the block caused by the absence of factor VIII, ie, by "downstream" component(s). Thus, one could have thought of increasing any factor(s) down in the cascade, not necessarily factor VII.

14. APCCs comprised many clotting factors (mainly factor II, V, VII, IX and X) in both activated and zymogen forms in addition to unknown molecules. Given this plethora of components, the skilled persons took very seriously the possibility that more than one component of the APCCs was required for the by-passing effect. This common belief is substantiated by the evidence given hereafter (emphasis added throughout). The authors of document (D6) believed that an activity in the concentrate might activate factor VII to factor VIIa (page 835, line 2 from the bottom) and on page 836, last paragraph they use the terms "activities" and "contributes" suggesting that factor VIIa could require the participation of other factor(s) to achieve haemostasis. Document (D12) also uses the expression on page 7, r-h column, line 10 "the active principle(s)". This view is in line with the ones expressed by the experts taking part to the Rome Symposium (see page 79, line 6 from the bottom: "a normalization of the intrinsic system cannot be satisfactorily explained by factor VIIa alone"; page 98, lines 5 to 7 from the bottom: "interaction of phospholipids present in APCCs with factor VIIa"; page 121, lines 1 to 3: "factor VIIa and some intermediate form of factor IXa") and is consistent with Dr Hedner's intention to elucidate the haemostatic role (not "action") of factor VII (see document (D3)). Also document (D15) on page 118, last paragraph refers to "active components". To conclude this long list, Dr Fujimaki believed that the by-passing activity of FEIBA® required the cooperative effect between factor VIIa and factor IXa (see document (A26), page 228, fourth paragraph). Therefore, given the common belief that more than one component of APCCs was responsible for the by-passing activity, there was no incentive to prepare a composition according to claim 1 of the patent in suit with factor VIIa as the sole activated haemostatic agent. The Board has thus to acknowledge this fact as a contribution to the inventive step.

15. As Dr Hedner's presentation to the symposium in Rome (document (D3)) did not comprise any evidence showing success on haemophilic dogs, in the Board's judgement the skilled person who listened to her disclosure would not have had any reasonable expectation of success for a trial of what she suggested, let alone a reasonable expectation of success for a modification of this by excluding activated Factor XIIa.

16. Thus, in the Board's judgement the skilled person would not derive in an obvious manner a therapeutic composition comprising factor VIIa as the sole activated haemostatic agent, from the state of the art. The claims as granted all explicitly or implicitly rely on this feature. Therefore, they satisfy the requirements of Article 56 EPC.