T 0091/98 (Antiviral nucleosides/WELLCOME) 29-05-2001

Reasons for the Decision

1. The appeal is admissible.

Main request Article 56 EPC; claim 1

2. The Appellants argued that, as there was no pharmaceutical formulations to treat diseases caused by retroviruses before that which they produced, there was no closest prior art to the claimed invention. The Board, although agreeing to the facts of the matter, does not consider that, in the given circumstances, they warrant the conclusion, the Appellants drew from them. Indeed, the etiologic agent causative of the diseases which the claimed formulation is intended to fight was identified as being retroviruses about three years before the first priority date of the patent in suit (see document (10), page 11, left-hand column for an historical perspective of the discovery of retroviruses as causative of diseases). As there is a practical impossibility for a pharmaceutical formulation against a given agent to be produced before this agent is identified, the fact that the claimed formulations were the first of the kind undoubtedly reflects the celerity with which the Appellants searched for a remedy against the diseases caused by the retroviruses. Yet, it is not a convincing proof of inventive step because, in such a situation, inventive step rather depends on the state of the art on retroviruses, which may or may not point out in an obvious manner to a way in which these viruses can be eliminated. In the Board's view, the ordinary skilled person would have looked in the literature for any known substances reported to have an effect on retroviruses because of the new and urgent need for a medicament against them. By doing so, he/she would have become aware of document (7) which is considered by the Board as the closest prior art.

3. Document (7) is a study of the effect of, in particular, azidothymidine, on the release of a retrovirus from transformed mouse spleen cells. The data are shown in Figure 1 and analysed under the heading "Inhibition of viral replication by...azidothymidine", in the first paragraph on page 4981.

It is disclosed that in the presence of azidothymidine, the transforming activity of the virus is decreased to less than 5% of that obtained from virus released from untreated cells and that, at the same time, the growth rate of the mouse spleen cells is changed only minimally compared to that of uninfected cells. In the discussion of their results, the authors mention the inhibition of virus replication by azidothymidine, as well as the low toxicity of this compound for the cells but high toxicity for the virus. Finally, looking into the future which included treating diseases known to be caused by DNA viruses but, of course, did not include treating diseases caused by retroviruses, as these had not yet been discovered, they suggest that azidothymidine might favorably replace another compound, BrdUrd, in a medical treatment against diseases caused by said DNA viruses.

4. Starting from the closest prior art, the problem to be solved is to find a medical treatment against diseases caused by a human retrovirus.

5. The solution proposed is a pharmaceutical formulation comprising a pharmaceutically acceptable derivative of azidothymidine and a pharmaceutically acceptable carrier therefor.

6. The derivatives of azidothymidine were not argued to present unexpected features compared to azidothymidine. They include such compound as the disodium salt of azidothymidine monophosphate (patent in suit, page 5, lines 9 to 16), the synthesis of which had already been achieved well before the first priority date (document (1), page 4300, right -hand column). Thus, in the assessment of the inventive step of the claimed subject-matter over document (7), no other features than that relating to azidothymidine per se need be taken into account.

7. In the Board's judgment, the skilled person aware from document (7) that azidothymidine had the property of inhibiting the replication of a mouse retrovirus and was little toxic to the mouse cells would find it obvious to test its activity or that of its derivatives on human retroviruses, because, as admitted by both parties, no differences were known to exist between retroviruses depending on their specific hosts (mouse or human). As the use of azidothymidine in a medical treatment is also envisaged in document (7), producing it or its derivatives in the form of a pharmaceutical formulation against diseases caused by human retroviruses would readily come to mind, said producing not being argued to require inventive step. It is, thus concluded that the subject-matter of claim 1 is derivable in an obvious manner from the teachings of document (7).

8. In this context, it should be pointed out that the approach to inventive step developed in genetic engineering cases, according to which inventive step is not denied on the sole basis that a project is obvious to try but in cases where there is a reasonable expectation of success that said project can be put into practice (cf T 296/93, OJ EPO 1995,627) does not apply here. The rationale behind this approach is that one may easily conceive of inventions to be made by genetic engineering, yet realising them may cause problems in view of difficulties known or experienced when starting the project. Here, to find out whether derivatives of azidothymidine have an activity against human retroviruses while remaining non toxic to cells, it is enough to perform well-known, routinely carried-out in vitro tests of viral infectivity (such as in document (7)) so it is rather a "try and see" approach which applies.

9. The Appellants argued that document (7) would not be taken into account because the experiments which it disclosed were carried out on cells which were chronically infected whereas the retroviruses which caused human diseases had to be eliminated while the diseases were in the acute phase. This argument is not convincing because document (7) explicitly teaches that azidothymidine inhibits viral replication. Taking into account the urgency that there was at the time to find a drug against such diseases (patent specification page 1), any compound having shown potentialities at stopping the replication of retroviruses, irrespective of the conditions of their use, would be tested for its efficiency as a medicament.

10. The Appellants also argued that the skilled person would not think of using derivatives of azidothymidine as medicaments because Figure 1 of document (7) showed that azidothymidine was toxic to cells. Yet, the authors of document (7) interpreted their results otherwise (see point 3). They must have been convinced that azidothymidine was not so toxic to cells that it could not be used as a medicament since they proposed such a use. There is no reason to doubt that the skilled person would have taken their conclusions at face value.

11. It was also pointed out that at the filing date azidothymidine had become known for having no appreciable activity against any DNA viruses. These viruses, however, are quite distinct from RNA viruses and, in the Board's judgment, the results obtained with them would not have been likely to alter the conclusion which the skilled person would draw from document (7) itself that azidothymidine has an effect on the latter.

12. Finally, the Appellants drew the Board's attention to document (17) which discloses that A-type particles are induced in the presence of azidothymidine during Friend's cell differentiation, implying that this would lead the skilled person away from using the compound as a drug. In document (17), it is stated on page 35 that the "intracisternal virus-like A-type particles might be precursors of the Friend virus or related to the Friend virus complex". The Board does not consider this statement as a possible deterrent from testing azidothymidine derivatives as medicaments.

13. The main request is rejected for lack of inventive step of claim 1.

First auxiliary request Article 56 EPC; claim 1

14. Claim 1 of this request corresponds to granted claim 2. Granted claim 2 is dependent on granted claim 1 which fails for lack of inventive step (points 2 to 12, above); it was not explicitly opposed by the Respondents. However, according to the Enlarged Board of Appeal decision G 9/91 (supra; point 11 of the decision), "even if the opposition is explicitly directed only to the subject-matter of an independent claim of a European patent, subject-matters covered by claims which depend on such an independent claim may also be examined as to patentability, if the independent claim falls in opposition or appeal proceedings provided their validity is prima facie in doubt on the basis of already available information".

15. Claim 1 (granted dependent claim 2) differs from granted claim 1 by the feature that the carrier contained in the claimed pharmaceutical composition is other than water. No technical facts were presented nor have they been any submissions in the proceedings specifically in relation to using carriers other than water in the pharmaceutical preparation. Thus, the information available to assess the inventive step of claim 1 is the same as the one which led the Board to conclude that claim 1 of the main request failed for lack of inventive step. The patentability of claim 1 of this request is, thus, prima facie in doubt and said claim may be examined for patentability.

16. The Appellants suggested that in case the Board would come to this conclusion, the case should be remitted to the first instance to safeguard them the chance of having two instances considering the matter. In accordance with Article 111(1) EPC, the Board of Appeal may either exercise any power within the competence of the departement which was responsible for the decision appealed or remit the case to that department for further prosecution. Remittal is, therefore, at the Board's discretion. The present case being more than twelve years old, the Board decides for sake of expediency not to refer the case back to the first instance but to consider the matter of inventive step.

17. In view of the findings (points 2 to 12, above) that document (7) is detrimental to the inventive step of pharmaceutical formulations of azidothymidine derivatives in general, inventive step could only arise from the added feature that the carrier is other than water. Yet, it is a basic fact of pharmacology that each and every pharmaceutically active substance need to be formulated in a different way which depends on the mode of administration which is envisaged. If, for example, it is envisaged that the pharmaceutically active substance is to be administered as a solid such as tablets, then, of course, the carrier may be a solid carrier (povidone, gelatin and hydroxypropylmethylcellulose are mentioned in this respect on page 6, first paragraph of the patent in suit ). No evidence was provided that obtaining azidothymidine derivatives in the form of a pharmaceutical formulation containing a carrier other than water was in any way difficult nor that the properties of said formulation were in any way surprising. Accordingly, the added feature (carrier other than water) does not impart inventive step to the rest of the claimed subject-matter (pharmaceutical formulations containing azidothymidine derivatives) which, as shown in points 2 to 12 above, is not inventive.

18. The first auxiliary request is rejected for lack of inventive step of claim 1.

Second auxiliary request Article 56 EPC, claim 14

19. Claim 14 relates to specific ester derivatives of azidothymidine. No advantageous properties or surprising effect were demonstrated for any of them. The reasoning developed in points 2 to 12 above in relation to pharmaceutically acceptable derivatives of azidothymidine in general thus equally applies to the ester derivatives. The Appellants argued that, if using derivatives at a unit dose which inhibited viral replication but was not toxic to cells (such a unit dose being included in claim 1 of this request) was inventive, then the subject-matter of claim 14 was also inventive because the ester derivatives would de facto be administered at the "inventive" unit dose. However, as the feature argued to be inventive is not claimed in claim 14 here at issue, it cannot support any inventive step argument. Auxiliary request 2 is refused as claim 14 does not fulfill the requirements of Article 56 EPC.

Third auxiliary request Article 56 EPC, claim 1

20. This claim differs from claim 1 of the main request in that the claimed pharmaceutical formulation is said to be adapted for oral administration. The Appellants argued that to be able to ingest the drug was a definite advantage for the patient compared to receiving it in any other form such as subcutaneously, intravenously, intradermally etc... In the Board's judgment, it is a matter of general common sense that oral administration is the least invasive means of taking up a drug and, thus, the feature as such cannot impart inventive step to the subject-matter of claim 1. No evidence was provided that in case of azidothymidine derivatives, this mode of administration has unexpected advantageous properties. Thus, the added feature (adapted for oral administration) does not impart inventive step to the rest of the claimed subject-matter (pharmaceutical formulation of derivatives of azidothymidine) which was found not to be inventive (points 2 to 12, above). Accordingly, the third auxiliary request is rejected for lack of inventive step of claim 1.

Fourth auxiliary request Article 123(2)(3) EPC; claim 1

21. A support is found in claim 12 of the application as filed for the pharmaceutical formulation comprising a unit dose of between 5 to 1500 mg of the active ingredient.

22. The scope of the claim is narrower than that of the corresponding granted claim 1 which relates to pharmaceutical formulations irrespective of their quantitative content.

23. The requirements of Article 123(2)(3) EPC are fulfilled.

Priority issue

24. In cases such as the present one, when multiple priorities are claimed, it is generally necessary to define the priority date of the subject-matter to be assessed for inventive step because only the documents pre-dating the priority date can be taken into consideration. Here, however, of the three documents cited in respect of inventive step, document (7) was published in 1974 ie before the first priority date (16 March 1985), the date of publication of the information contained in document (8) cannot be established (see points 25 to 31 below) and the argument raised on the basis of document (22) may not be taken into account (see points 32 and 33, below). Accordingly, the priority date of the claimed invention has no bearing on the assessment of inventive step and need not be ascertained.

Article 56 EPC; claim 1 Document (8)

25. The Respondents challenged inventive step on the basis of, in particular, document (8). A prerequisite for the information contained therein to be taken into account for the assessment of inventive step is, of course, that its date of availability to the public be known. Document (8) is an entry of the Lexis-Nexis data base which, as accepted by both parties, was printed out in 1996, ie some ten years after the filing date of the patent in suit. The heading of this entry reads: "10th Story of Level 1 printed in Full format. Copyright 1985 U.P.I. September 3, 1985, Tuesday AM cycle".

26. The fact that document (8) was retrieved in paper form from the Lexis-Nexis data bank in 1996 implies, of course, that the information, it contains was entered in the data bank before the date of printout. Yet, it does not provide any evidence as to when this information was entered into the data bank ie as to when it was made available to the public. Neither can the date of availability be taken as the date mentioned in the heading of the entry (3 September 1985) as this latter date cannot be equated to the distribution date of the information and need not even be right. How easily a document can be wrongly dated was shown during the appeal proceedings when the Appellants submitted a newspaper article dated by a hand-stamp 14 September 1985 although the Respondents provided evidence convincing the Board and the Appellants that this article was in fact published on 14 September 1986.

27. The Respondents argued that, on the balance of probabilities, the information contained in document (8) must have been available to the public on or shortly after the 3 September 1985. In accordance with the case law of the Boards of Appeal (T 750/94, supra) "when an issue of fact is being examined and decided by the EPO on the balance of probabilities, the more serious the issue the more convincing must the evidence be to support it." Document (8) discloses the use of azidothymidine as an experimental drug capable of stopping the virus causing AIDS and being tested inter alia at the National Cancer Institute in Bethesda, Md. In the text are included citations of the director Dr Sam Broder and the spokesman Frank Mahaney of the National Cancer Institute about the positive results of experiments carried out with this drug. As this information could have a decisive impact on the Board's conclusions on inventive step, the submissions by the Respondents about the date at which this document was made available to the public must be supported by unequivocal evidence.

28. The Respondents filed a declaration of Connie Landis (document (12)), an affidavit of Monica Kitts (document (13)) and a letter of Tobin Beck (document (14)) to support their position that document (8) must have been available to the public shortly after the 3 September 1985. Document (14) is from the managing Editor of United Press International (U.P.I.) who certifies "that to the best of my knowledge, the Nexus copy of the UPI story dated Sep.3, 1985, is a true and accurate copy of the story UPI moved at that time". In the Board's judgment, this statement does not amount to a clear and unequivocal statement that the information contained in document (8) was then available to the public because it is impossible to understand what kind of action the term "moved at that time" might involve. Furthermore, there is no indication why the managing director on August 27, 1997, the date of his letter, can exactly remember the story moved at that time ie twelve years before. He gives no explanations having special circumstances to remember that very article or having found a record of the article and the date of publication in the archives of UPI. The Board is convinced that the testimony in this letter is only influenced by the date written in document (8) and is not based on true recollection. Therefore, the expression in the said letter "to my best knowledge" is merely relative and cannot be considered as a factual argument.

29. Monica Kitts, the author of document (13) stated in her affidavit that "from the printout of the data base, it is obvious to me that its content was distributed as a press-wire article to various newspapers in North Carolina and Tenessee on September 3, 1985". She also recalls a conversation, she had in 1997 with the scientist cited in document (8) during which he mentioned having disclosed "the use of AZT as a treatment for AIDS to the press in the mid 1980s". Yet, as the printout (ie document (8)) does not disclose that the information it contains was ever published in newspapers, the Board does not consider the earlier statement as bringing any level of certainty as to the date when this information was made available to the public. As for the latter statement, the reference to the mid 1980s is insufficient to prove that the disclosure took place before the filing date of the patent in suit (14 March 1986).

30. Document (12) stemms from the actual product manager of the Lexis-Nexis data base who states that "to the best of her knowledge", document (8) "was available in the LN's NEXIS service from and after September 4, 1985". What kind of knowledge brought her to this conclusion is not supported by any facts. If the declaration was really made to the best of her knowledge, these facts should be mentioned. The Board cannot imagine that a document saved in a data bank is not recorded with the day of entry and this could not be verified at any time. It is also not stated whether she was employed by Lexis-Nexis in 1985 or has experience of the functioning of the data base in those days.

31. In view of these findings, the Board concludes that the date at which the information contained in document (8) was made available to the public cannot be unambiguously defined and, that, in consequence, this document cannot be taken into consideration to evaluate inventive step.

Document (22)

32. Document (22) is a declaration filed by the Appellants on 21 May 1998. In point 5 of said declaration, it is disclosed that a scientific meeting took place in Minneapolis from 29 September to 2 October 1985. The declarant states: "I attended the Minneapolis meeting and I was one of the ...scientists involved in making the presentation about our first results with zidovudine*. This presentation created considerable excitement both in the meeting itself and beyond..." (*The term zidovudine was linked to the term azidothymidine once azidothymidine had become a known antiretroviral drug; it was first listed in the 11th edition of the Merck index (1989), see declaration of J.Partridge, point 12, filed by the Appellants on 23. May 1998). For the first time at oral proceedings, the Respondents argued on the basis of this statement that the use of azidothymidine as a medicament was known to the skilled person as from the date of the meeting and, therefore, producing effective pharmaceutical formulations of it or its derivatives would be obvious.

33. The Board is not prepared to accept this argument into the proceedings because it has been submitted too late. Three years have passed from the moment the declaration containing the above mentioned statement was available to the Respondents. They, thus, had ample opportunity to make their objection known in good time for the Appellants to have a fair chance to comment on it.

Document (7)

34. As a consequence of the Board's finding with regard to documents (8) and (22), document (7) remains the only prior art document to be discussed. It discloses on page 4981, left-hand column, "Results", that azidothymidine is able to inhibit the replication of a mouse retrovirus while being little toxic to mouse cells when used at a concentration of 250 µM. It also suggests the use of azidothymidine in a medical treatment.

35. Starting from this closest prior art, the problem to be solved can be defined as producing a means to fight diseases caused by human retroviruses which would be effective against said virus while remaining non toxic to the host cells.

36. The solution provided is a pharmaceutical formulation comprising a pharmaceutically acceptable derivative of azidothymidine and a pharmaceutically acceptable carrier comprising a unit dose of between 5 and 1500 mg of active ingredient.

37. In points 2 to 13 above, it was established that the pharmaceutical formulations of azidothymidine derivatives in general were not inventive over the teachings of document (7). The question which remains to be decided is whether pharmaceutical formulations at the claimed unit dose would be.

38. In the patent in suit, page 5, lines 27 to 39, it is disclosed that the claimed unit doses are those which should be administered to achieve peak plasma concentrations, these being of about 1 to 75 µM, ie some 3 fold to 250 fold lower than the concentration disclosed in document (7) as inhibiting the replication of the mouse retrovirus while being little toxic to mouse cells. In the Board's judgment, it is an unexpected as well as advantageous result that azidothymidine derivatives are effective against the human retroviruses at such low concentrations.

39. The Respondents argued that it only required routine work to find out which unit dose would be appropriate. This may well be, yet it does not affect inventive step, which, as was just mentioned, is not due to finding out the relevant dosis but to the fact that this dosis is substantially lower than that which had been found effective against the mouse retrovirus.

40. The fourth auxiliary request fulfills the requirements for patentability.