Merck & Co. Inc. v Sankyo Company Limited

[1991] APO 27

PATENTS ACT 1990

DECISION OF A DELEGATE OF THE COMMISSIONER OF PATENTS

Re:Patent Application No 527532 by MERCK & CO. INC. and Opposition thereto under S.59 of the Patents Act 1952 by SANKYO COMPANY LIMITED

Background
Merck & Co Inc (Merck) lodged application 527532 on 11 December 1978. The application claims priority from two earlier applications both filed in the USA on 19 December 1977. Acceptance of the application was advertised in the Official Journal on 10 March 1983 and on 9 June 1983 Sankyo Company Limited (Sankyo) lodged a s.59 notice of opposition to the grant of a patent on the application.
         Evidence in support, consisting of two declarations and four exhibits, was served by the opponent on 6 January 1984.  Following the grant of a number of extensions of time the applicant eventually served evidence in answer, consisting of three declarations and six exhibits, on 7 December 1987.  The opponent was granted three extensions of time to serve evidence in reply, but on 5 December 1988 advised that no evidence in reply would be served.
         A hearing on the matter was originally set down for October 1989.  At the request of the applicant this hearing was deferred so that the matter could be heard at the same time as the hearing on patent application no 574695 by Sankyo Company Limited and opposition thereto by Merck & Co Inc.
         The matter was heard in Sydney on 11 and 12 March 1991.  At the hearing the applicant was represented by Dr A. Bennett of Counsel, instructed by Mr J. O'Connor, patent attorney of Spruson &

Ferguson and the opponent was represented by Mr J. McCormack, patent attorney of Griffith Hack & Co.
The notice of opposition lists those grounds of opposition as specified in paragraphs (a)-(i) of s.59(1) of the 1952 Act. However the evidence and the submissions on behalf of the opponent at the hearing were confined to the opposition grounds of anticipation and compliance with s.40 of the 1952 Act.
The Specification as lodged
         The specification indicates the invention relates to 13-halo and 13-deoxy derivatives of C-076 compounds.  C-076 is a series of macrolides with potent antiparasitic activity which are isolated from the fermentation broth of Streptomyces avermitilis.  The C-076 compounds are isolated as a mixture of four pairs of compounds, each member of a given pair differing from the other member of that pair in the nature of the C-25 position substituent.  The compounds of the invention are prepared by a series of reactions which converts the C-076 starting materials from 13-disaccharide compounds to the aglycone compound followed by conversion to the 13-halo or 13-deoxy compounds.  These compounds are stated to have significant parasiticidal activity.
The C-076 starting materials are described at page 2 of the specification in terms of a structural formula in which a radical R2 is shown at the 25-position. At page 3 R2 is defined as "n-propyl or sec-butyl".  There then follows a table which identifies the eight individual C-076 compounds with reference to the structural formula.  In this table all R2 radicals are shown as sec-butyl or n-propyl. The specification at page 4 shows another structural formula which is stated to represent the compounds of this invention. This formula differs from the formula of the C-076 starting materials at the 5, 13, 23 and 25 positions. The radical at the 25-position is shown as R3 and is defined in the passage following the structural formula as "n-propyl or sec-butyl".  From the subsequent description and from the evidence it is apparent that the substituent at the 25-position remains unchanged during the reactions to convert the starting materials to the compounds of the invention.  The radical shown as R2 in the structural formula of the C-076 starting materials is therefore the same as the radical R3 shown in the structural formula of the end products.
         The specification describes a number of methods by which the starting materials are converted to the end-products and a number of Examples of the preparation of specific compounds are given as well as Preparations describing the production of starting materials.  My attention was directed by Counsel for the applicant to the passage at page 23 immediately preceding the examples, which reads as follows:

"In the isolation of the C-076 compounds, which serve as starting materials for the instant processes, from the fermentation broth, the various C-076 compounds will be found to have been prepared in unequal amounts.  In particular an "a" series compound will be prepared in a higher proportion than the corresponding "b" series compound.  The weight ratio of "a" series to the corresponding "b" series is about 85:15 to 99:1.  The differences between the "a" series and "b" series is constant throughout the C-076 compounds and consists of an n-butyl group and a sec-propyl group respectively at the 25-position.  This difference, of course, does not interfere with any of the instant reactions.  In particular, it may not be necessary to separate the "b" components from the related "a" component.  Separation of these closely related compounds is generally not practiced since the "b" compound is present only in a very small percent by weight, and the structural difference has negligible effect on the reaction processes and biological activities."

The reference in the middle of this passage to "an n-butyl group and a sec-propyl group" was a typing error according to the Angyal declaration (paragraph 21.2), and this conclusion was supported in paragraph 19 of the Fisher declaration.  The references should therefore be to n-propyl and sec-butyl for consistency with the other definitions in the specification.
         The Examples in the specification are all, from their titles, directed to specific "a" compounds, but Dr Bennett submitted, in view of the above passage, the products of those Examples were in fact unseparated mixtures of the "a" and "b" compounds.
         The specification ends with 21 claims of which claims 1-19 are directed to compounds per se.  Claim 1 repeats the definition of the compounds of the invention given at page 4, and reads as follows:
    "1.  A compound having the formula:

wherein the broken line indicates a single or a double bond;

R1 is hydrogen or halogen;
         R2 is hydrogen, methyl or loweralkanoyl;
         R3 is n-propyl or sec-butyl; and
         R4 is present only when the broken line indicates a single bond and represents hydrogen, hydroxy, loweralkanoyloxy, loweralkylthio, loweralkylsulfinyl, loweralkylsulfonyl or loweralkoxy".
         Claims 2-19 are dependent claims, claim 2 being directed to compounds in which R3 is n-propyl only, while claims 3-19 are limited to compounds in which R3 is sec-butyl.
         Claim 20 is a process claim to prepare compounds of the identical structural formula to that given in claim 1 by a specified treatment of a compound of that structural formula where R1 is hydroxy.
The Specification as accepted
         The main changes from the specification as lodged are:

.In the definition of the starting materials at pages 2-3 the radical R2 at the 25-position is defined as iso-propyl or sec-butyl.

.In the table at page 3 which identifies individual C-076 compounds all R2 radicals are defined as iso-propyl or sec-butyl.

.In the definition of the compounds of the invention at page 4 the radical R3 at the 25-position in the structural formula is defined as iso-propyl or sec-butyl.

.In the passage at page 23 quoted above the groups at the 25-position are stated to be sec-butyl and iso-propyl.

.The insertion of an extra paragraph at page 24 immediately preceding the Examples, which reads as follows:

"As indicated hereinbefore, in the isolation of the C-076 compounds which serve as starting materials compounds of the "a" series are produced together with compounds of the "b" series.  Thus in the foregoing examples where preparation of the C-076 compounds of the "a" series of the present invention is described, C-076 compounds of the present invention of the "b" series are also produced."

.The insertion of a statement at the end of each Example to the effect that the corresponding "b" compound is produced in like manner.

.The insertion of a statement at the end of each Preparation that the corresponding "b" compound is produced in like manner.

.The insertion of Preparations A-J immediately preceding the claims which describe in some detail how individual "a" and "b" compounds are separated.

.In claim 1 the compounds of the invention are defined in terms of a structural formula in which R3 at the 25-position is iso-propyl or sec-butyl.

.In claim 2 R3 is defined as iso-propyl only.

.New claim 20 is a per se claim to a specific "b" compound.

.New claim 21 is an omnibus claim to the compounds per se with reference to the Examples.

.New claims 22-31 are claims to compounds obtainable by specified reactions of C-076 starting materials.

.    In claim 32, which is equivalent to original claim 20, the      definition of R3 is iso-propyl or sec-butyl.

.New claims 33 and 34 are directed to a process of treating specific starting materials in a specified way.

.New claim 35 is an omnibus claim to the preparation of

C-076 derivatives with reference to the Examples.

.New claim 36 is for the product of the process of claims 32-35.

.The insertion of new claims 37-42 relating to the use of the compounds of the invention, which claims are not relevant to the present decision.

The Evidence

In support of the opposition the opponent lodged 2 declarations and accompanying exhibits.  One was from Professor Stephen Angyal who holds the degrees of Doctor of Philosophy, Master of Science and Doctor of Science, and who is the author of some 160 scientific publications on subjects concerned with organic chemistry.  He states at page 2 that "The nomenclature of organic chemistry has always been one of my special interests".  Prof. Angyal makes the following points in his declaration:

.The terms "n-propyl" and "isopropyl" are not synonymous.

Prof Angyal declares:

"The scope of the respective terms is unquestionably different and does not overlap" (paragraph 12).

.In the general formula of the compounds of the invention at page 4 of the original Merck specification (Exhibit SJA1) R3 was defined as n-propyl or sec-butyl.  Prof. Angyl declares at paragraph 8:

"... As a skilled chemist, I would have been in no doubt that the isopropyl compounds were excluded from the general formula as originally filed; as indeed were compounds substituted with any other alkyl group apart from n-propyl or sec-butyl.  I say unhesitatingly that any competent chemist would be of the same view as expressed by me in this paragraph."

.In the accepted Merck specification, application no 527532 (Exhibit SJA2), R3 was defined as isopropyl or sec-butyl.  Prof Angyl declares at paragraph 9.1: 

"... The scope of the general formula now includes those compounds which can structurally be considered to be related to other isopropyl compounds, including isopropyl alcohol, but does not include within its scope the compounds structurally related to n-propyl alcohol.  As a result, claims 1 and 32 now embrace the isopropyl compounds, which were not included in the claims before."

.The original Merck specification discloses only n-propyl or sec-butyl for the 25-position substituent.  Prof Angyal declares at paragraph 21.1:

"To sum up, my reading of the original specification presented here as Exhibit SJA1 is that it discloses only the possibility of the 25 substituent being

n-propyl or sec-butyl.  I think that any organic chemist or worker in the field would also hold this view."

and at paragraph 21.2:

"... The overall meaning of the specification of Exhibit SJA1 to me is that the C-076 starting

materials have n-propyl of sec-butyl groups in the 25 position and that these groups would be unchanged by any of the transformations discussed in the specification, so that the n-propyl and sec-butyl groups would also be present in the new products."

.The compound labelled B-41D of Sankyo application no 61668/80 (Exhibit SJA5) falls within claim 1 of the accepted Merck application.  Prof Angyal declares at paragraph 24:

"... In fact, the compound B-41D is identical with the compound of claim 1 of Exhibit SJA2 when R1 is hydrogen, R2 is hydrogen, R3 is isopropyl and R4 represents hydrogen.  Thus, there is no doubt that the compound B-41D is one of the various compounds covered by claim 1 on page 66 of the specification presented here as Exhibit SJA2."

The opponent's second declaration was from Dr Martin Horner, a registered patent attorney who holds the degrees of Bachelor of Arts, Master of Arts and Doctor of Philosophy all in the field of chemistry.  Dr Horner states in paragraph 3:

"I agree with the view expressed by Professor Angyal that there is no basis whatsoever in the specification of Merck application 42389/78 as originally lodged at the Australian Patent Office for there being an isopropyl substituent at the 25 position in either the C-076 starting materials or the new compounds disclosed therein." 

It is not clear to me on what basis he is qualified to "agree".  Dr Horner then makes the unqualified statement in paragraph 6: 

"It is therefore apparent to me on the evidence the accepted Merck application 527,532 includes new subject matter not present in the specification as originally lodged at the Australian Patent Office, as regards the possibility of an isopropyl substituent at the 25 position in both the C-076 starting materials and in the new compounds which are disclosed and claimed." 

Again I am unclear as to the basis for his conclusion.  The remainder of Dr Horner's declaration contains assertions as to the priority dates of various claims of the accepted Merck application based on this conclusion (paragraphs 10-25), and consequent assertions that the Sankyo compound B-41D prior publishes certain of those accepted claims (paragraphs 26-42).
         The evidence-in-answer comprises 3 declarations and accompanying exhibits.  2 of these declarations were made by 2 of the inventors of the Merck application, namely Dr Michael Fisher and Dr John Chabala.  The other declaration was made by Dr Georg Albers-Schonberg, one of the inventors of Australian patent no 513641 which discloses and claims C-076 compounds.
         Dr Chabala is an employee of Merck & Co, Inc holding the position of Senior Director, Medicinal Chemical Research since July 1987.  He holds a Bachelor of Science degree in Organic Chemistry and a degree of Doctor of Philosophy in Organic Chemistry.  In his declaration and accompanying exhibits Dr Chabala describes how he duplicated some of the preparations disclosed in the Merck specification.  He declares in paragraph 8:

"... I proceeded to duplicate Preparations 6 and 8 and Examples 15 to 18;"

In paragraphs 9-23 Dr Chabala describes the preparations and his conclusions.  In some of these preparations mixtures of the major (a) component and the minor (b) component were obtained and isolated, while other preparations describe the subsequent isolation of the minor (b) component from the mixture of (a) and (b).  Dr Chabala then describes how the actual structure of the (b) component was determined from analytical data and comparisons with published data.  From the statements in his declaration and the attached exhibits Dr Chabala makes the following points in paragraph 23:

"The foregoing and the attached Exhibits demonstrate unequivocally that C-076 compounds exist as a group of homologous pairs of compounds in unequal amounts and that

the minor (b) component is prepared simultaneously and in conjunction with the major (a) component of the C-076 compounds, that the structure of the minor components is iso-propyl at the 25-position and the structure of the major component is sec-butyl at the 25-position; and that

the structure of such groups does not change during the course of the reactions described in Australian Application No. 527,532 (42389/78)."

Dr Albers-Schonberg is a research chemist employed by the Merck, Sharp & Dohme Research Laboratories.  He holds the degrees of Diploma from the Swiss Federal Institute of Technology and a Doctor of Philosophy in Organic Chemistry.  Dr Albers-Schonberg describes how the structure of the C-076 compounds was determined.  He makes the following points:

."The avermectin compounds (originally referred to as C-076 compounds) were first discovered in the fermentation broth of Streptomyces avermitilis.  The broth was first made in April of 1975 and realized to be highly active when the broth or crude extracts thereof had antiparasitic activity.  The structure of the active compound or compounds in the broth was totally unknown at that time.  Then, starting in January of 1976, samples of the fermentation materials (believed to be 4 compunds) were sent for structural investigation." (paragraph 7)

.

"In January 1976, the first small samples of the anthelmintic fermentation products C-076, later to be called avermectins, were received from our isolation group for spectroscopic analysis.  Further fractionation of these still complex samples by thin layer chromatography and bioassays identified four chromatographically apparently uniform components (A1, A2, B1 and B2) as of interest.  Mass spectral analysis, however, showed that each of these consisted of a major "a"-component (90-95%) and a minor "b"-component (5-10%).  Hence, the notations A1a, A1b, A2a, A2b, etc.  Elemental composition of all eight components were assigned and isomeric pairs (B2a/A2b and B1a/A1b) clearly distinguished by their mass spectral fragmentations.  Also, the glycosidic nature of the compounds was recognized at this time." (paragraph 8).

."These four samples A1, A2, B1 and B2 were used for 13C NMR analysis and further mass spectrometry.  At the time we had only a 100 MHz instrument available for 1H NMR, and these spectra were of little help (see below) because of insufficient resolution.  The 13C spectra, on the other hand, were highly informative by producing complete lists of the carbon atoms of the major components A1a, A2a, B1a and B2a; but they gave no information on the minor components which we had not yet been able to separate from the major components and which were present in too small quantities in the mixed samples to be visible in these spectra." (paragraph 10).

."Based on the mass spectral fragmentations and 13C NMR spectra, we composed schematic structures showing all functional groups, numbers of rings and unsaturations as a visual aid to scan the natural products literature for possibly related structures.  This search led to the then just published milbemycins, whose structures after some changes could account for all our data ..." (paragraph 11).

."We could conclude at this time that the "b"-avermectins carried a saturated three-carbon substituent at C25, either n-propyl or isopropyl.  In the absence of relevant data (beyond the mass spectral data which proved the close "a"/"b" analogy but said nothing about the detailed nature of the C25 substituent) we hypothesized that the "b"-avermectins could contain an n-propyl group.  This would permit a uniform biogenetic scheme for the milbemycins and avermectins and also the more conventional macrolide antibiotics which had been shown to be constructed by the producing microorganisms from acetate and propionate precursors. ..." (paragraph 13).

."While the 13C propionate experiments were in progress, our Synthetic Chemistry Department succeeded to separate the avermectins B1a and B1b by high-pressure liquid chromatography.  We had also in the meantime acquired a new 300 MHz NMR spectrometer (approximately some time before the middle of 1977).  300 MHz 1H NMR spectra of pure 22,23-dihydro B1 compound unambiguously proved the absence of a straight chain carbon substituent and the presence of an isopropyl group at C25)." (paragraph 15).

From the statements in his declaration and the attached exhibit Dr Albers-Schonberg makes the following summary in paragraph 16:

"... We have, to data, never seen data from the b-avermectins which correctly or incorrectly could be interpreted in terms of an n-propyl substituent.  All actual data, when they finally became available, supported or proved the isopropyl structure.  Consideration of n-propyl structures was a scientific hypothesis in the absence of data, to be proven or disproven by further experimentation.  For this reason, prior to the resolution of the problem our original patent application specified only "butyl" and "propyl" substituents.  The n-propyl structure was included in patent applications filed while the above biogenic hypothesis was being considered and before the isopropyl structure was observed on NMR."

Dr Fisher is a chemist employed in the laboratories of Merck, Sharp & Dohme Research Laboratories.  He holds a degree of Ph.D. in Chemistry.  He has been involved in synthetic chemical research on C-076 compounds and has filed several patent applications on synthetic variations of those compounds.
         Dr Fisher declares in paragraph 8 that he is in "complete accord" with the observations made by Dr Chabala, in particular that:

.In relation to the corresponding "a" and "b" compounds,

"... No differences in chemical reactivity were observed and indeed, no differences would be expected." (paragraph 9).

."... none of the series of reactions carried out on the B1a and B1b compounds had any effect on the 25-position alkyl substituent.  While a wide variety of chemical reactions were carried out on the C-076 molecule ... none of such reactions had the slightest effect on the structure of the 25-position alkyl.  This is evident from the analytical data presented by Dr Chabala which showed that the various characterizing data produced by the 25-position groups did not change as each chemical reaction was carried out.  Again, from a knowledge of the chemical reactions being carried out, I would expect this result to occur.  None of the chemical reactions are of the type which I, as a skilled chemist, would assume would react with or otherwise alter an alkyl group." (paragraph 10).

."... from the data presented in the Chabala Statutory Declaration with respect to the structure of the 25-position substituent, it is clear that the data are completely consistent with an "a" component structure of sec-butyl and a "b" component structure of isopropyl ..." (paragraph 11).

Dr Fisher then makes the following points:

."Concerning the Albers-Schonberg Statutory Declaration, I have read the same and agree that the structure of the 25-position group was and is isopropyl ... No data have ever been seen by me that proved an n-propyl structure for the

C-076 compounds and no 25- n-propyl compounds have ever been observed as part of the C-076/avermectin synthetic research program." (paragraph 12).

."... There was never any intention to claim any compounds in the accepted specification that were not described in the original specification. ...  In the case of both the original and accepted specifications, the intent was to claim the same compounds, i.e. the compounds inherently produced following the description of the patent application ..." (paragraph 15).

."... the structure of the "b" compounds which were actually prepared using the procedures described always was isopropyl ..." (paragraph 15).

."... Due to the paucity of data which was originally obtained early during the synthetic program into C-076 compounds, and due to a hypothesized biogenic scheme for the production of C-076 compounds (see the Albers-Schonberg Statutory Declaration) the structure for the 25-position group "b" component was initially proposed as

n-propyl.To be as complete as possible this proposed structure was included in the original patent application.  When additional data were obtained, it was realized that the actual structure was isopropyl rather than n-propyl. ..." (paragraph 15).

."... following the description of either the original or the accepted specification for the preparation of the above-named C-076 compound, the compound of exactly the same structure as B-41D would have been produced.  This is so irrespective of the nature of the written structural formula. ..." (paragraph 16).

."... There can be no new matter when exactly the same material is obtained by following either the original specification or the granted specification." (paragraph 22).

."... Anyone skilled in the art of macrocyclic natural products would realize that when very closely related products are isolated together, it is very common to describe those compounds together in those cases where the difference does not involve a reactive group..." (paragraph 26).

Submissions
         Mr McCormack's submissions on behalf of the opponent were based on the contention that the Merck application as lodged was fatally flawed in not containing sufficient information to put the invention in relation to the "b" compounds into practice.  He submitted that the amendments to the specification made under s.49(4) of the 1952 Act to change the substituent at the 25-position from n-propyl to isopropyl involved new matter.  The priority date of claims to this "new matter" should therefore be determined under s.159A of the 1952 Act.  In Mr McCormack's view the earliest possible priority date for certain claims of the Merck application would be 18 December 1981, the date when the amendments to change n-propyl to isopropyl were first lodged.  However he submitted that the priority date could be November 1982, the date of lodgement of the fourth statement of proposed amendmentss, if s.159A(5) were to operate.
         On behalf of the applicant Dr Bennett submitted the Merck specification as lodged correctly disclosed the "b" series of compounds both in terms of the starting materials and the empirical formula but that there was a scientific mistake as to the structural formula.  According to Dr Bennett there was no added subject matter in identifying correctly the original material in the accepted Merck application and hence s.159A is not relevant for determining priority dates.  She emphasised that the point to be decided in the present opposition was not whether the amendments to the Merck application should have been allowed but was concerned only with the question of priority dates.
         Dr Bennett considered the present situation was similar to the facts in Farbwerke Hoechst Aktiengesellschaft vormals Meister Lucius & Bruning v Commissioner of Patents (1971) 45 ALJR 235. In that case a general formula representing new products was subsequently found to be incorrect and the patentee sought to amend the specification to indicate that the formula was only a "presumed" formula. Gibbs J expressed no view as to whether an amendment by the substitution of the true formula would have been permissible and Dr Bennett drew the inference that the judge would have allowed an amendment to correct the formula. The substance of the invention in Farbwerke was, according to Dr Bennett, a process, and the structure of the compounds was irrelevant in the context of that process. There were no compounds of the stated formula and similarly in Merck's case the n-propyl compounds did not exist. Dr Bennett relied on Gibbs J's statement at page 239:

"On ordinary principles of construction, if the correct part of the description defines the subject matter with certainty, the incorrect part may be rejected to enable effect to be given to the document as a whole."

to support her submission that the amendments to change n-propyl to isopropyl were allowable and did not involve new subject matter.  Mr McCormack relied on a different passage from the Farbwerke decision where Gibbs J stated at page 240:

"if ... the formula applied to some of the compounds produced by the process, it could not be rejected as a mere erroneous addition to the description, but would have to be given effect as limiting the class of compounds to which the description applied."

to support the view that the amendments to the Merck application were directed to new matter and therefore take the later priority date, since the original formula correctly applied to some of the compounds.
         In discussing the formula the High Court distinguished the Farbwerke case from Chevron Research Co's Patent [1970] RPC 580 where the new claims were wider but included the old claims. Dr Bennett considered the Chevron decision was clearly distinguishable from the present case since Merck have substituted isopropyl for n-propyl and was not trying to add matter.  However Dr Bennett submitted the principles in Chevron support allowance of Merck's amendments.  In response Mr McCormack submitted Merck had not discharged the heavy onus referred to in Chevron of putting forward the full evidence to support their amendment.
         Mr McCormack submitted the facts in the present case were mirrored in Zambon S.p.A.'s Patent [1971] RPC 95 where the patentee had a choice of two alternatives for the position of a substituent and opted for the incorrect structure. Similarly he suggested Merck made an error of judgement in opting for n-propyl rather than isopropyl and this meant the application was flawed from the beginning. In response Dr Bennett submitted the Zambon decision stands by itself, with the judge deciding the case on its facts.  Zambon deliberately elected to claim one formula when they had good reason to doubt the correctness of that formula and therefore, as a matter of discretion, they were not allowed to amend.
         To support his contention concerning Merck's alleged error of judgement Mr McCormack referred to paragraph 15 of the Albers-Schonberg declaration which, in his interpretation, indicates definitive results as to the structure of the 25-position were obtained in 1977.  This was before the Australian application was lodged and yet Merck opted to claim the incorrect substituent.  Dr Bennett pointed out, according to the evidence, definitive results were only obtained late in 1978 (paragraph 17 of the Albers-Schonberg declaration), and there is no direct evidence as to the time frame in the latter half of 1978 that suggests any recklessness on the part of the applicant.  The evidence gives a full explanation of why n-propyl was originally believed to be the correct substituent and how the isopropyl structure was determined.  Dr Bennett emphasised this evidence stands uncontradicted.  Amendments were subsequently sought to the application as lodged to correct the structural formula and Dr Bennett referred to the declarations of Fisher and Albers-Schonberg as evidence that there has never been any data to show that the C-25 substituent was
n-propyl.

Mr McCormack submitted as there was no character-isation of a specific "b" compound in the specification as lodged, one must question whether there was any disclosure of a "b" compound at all.  All the examples in the Merck specification were directed to "a" compounds and references to "b" compounds were difficult to find.  There was merely some information on page 23 of the specification to the effect that the "a" and "b" compounds occur together.  Mr McCormack pointed out between publication and acceptance of the Merck application a considerable body of information was inserted into the specification relating to the "b" compounds, including a number of additional preparations and claims.  In his view these amendments disclosed new subject matter in relation to the "b" compounds.
         In response Dr Bennett submitted the original specification disclosed the "b" compounds by the structural formula, by claim 2 which was solely directed to "b" compounds and by the discussion of the ratio of "a" to "b" at page 23.  In her view the description at page 23 meant when the specification referred to "a" compounds it meant "a" and "b" compounds.  The evidence also made it clear the "b" compounds were made.  Dr Bennett suggested it was not enough to look at the quantity of additional matter inserted into the specification but the content must also be considered.  The new preparations describe the separation of the "a" and "b" compounds and provide more information consistent with Merck's attitude of full disclosure.
Decision
         The issue to be resolved is to determine the priority dates of the claims of the accepted Merck application.  Mr McCormack
put to me the accepted claims involved "new matter" and therefore the priority dates should be determined under s.159A.
Section 159A(1) of the Patents Act 1952 reads as follows:

(1)Where -

(a)an application and complete specification have been accepted; and

(b)in an action or other proceeding before the Commissioner or prescribed court, the Commissioner or the prescribed court finds that a claim of the specification claims matter (in this sub-section

referred to as "the new matter") in substance disclosed in the specification as a result of amendment of the specification,

the Commissioner or the prescribed court shall treat as the priority date of that claim the date on which the new matter was disclosed in a statement of proposed amendments of the complete specification lodged under section 49, section 52A or section 52D or in a request made under Part VIII seeking leave to make amendments of the complete specification.

I must therefore determine whether the accepted claims which refer to the substituent at the 25-position as iso-propyl claim matter in substance disclosed in the specification as a result of the amendment from n-propyl which was the original definition of the 25-substituent.
         From the declarations lodged in the evidence in answer it is apparent to me the substituent at the 25-position in the
C-076 compounds always was iso-propyl.  Dr Chabala's evidence shows how he duplicated some of the preparations of the Merck application and was able to determine the 25-substituent remained unchanged during the course of the reactions.  This evidence is supported in the declarations of Dr Fisher and Dr Albers-Schonberg who both depose they have never seen data proving an n-propyl structure for the C-076 compounds.  The evidence also shows how the biogenic hypothesis led to the misassignment of the 25-substituent as
n-propyl and how the subsequent nmr data established the iso-propyl structure.  While the exact time at which the correct structure was determined is not clear to me from the evidence, I cannot agree with Mr McCormack's contention that definitive results were obtained before Merck lodged its Australian application.  All I can conclude from the declaration of Dr Albers-Schonberg is definitive results as to the correct structure were obtained "late in 1978" (paragraph 17) and, in the absence of other evidence, I cannot conclude Merck were careless in nominating n-propyl originally as the 25-substituent.
         Even though I am satisfied the correct substituent at the 25-position of the compounds of the Merck application is iso-propyl, and I accept the explanation of how the original substituent was incorrectly identified, it does not necessarily follow that the accepted claims do not contain new matter.  The evidence from Prof Angyal, which was accepted by the applicant, clearly establishes
n-propyl and iso-propyl are different radicals.  Prof Angyal also gives a detailed analysis of the compounds included by the general formula defining the compounds of the invention in both the original and the accepted Merck applications.  As a skilled chemist and an expert in the nomenclature of organic chemistry Prof Angyal draws the conclusion in paragraph 8 I have quoted previously that isopropyl compounds were excluded from the general formula of the compounds in the original Merck application.  I also note his conclusion in paragraph 9.1 (supra) concerning the scope of the general formula of the accepted compounds.
         From this evidence it is clear to me the compounds of the invention when defined in terms of the general formula in the original Merck application differed from the compounds defined by the general formula of the accepted application due to the substituent at the 25-position.
         However it was argued before me that this change in the definition of the compounds of the invention did not involve new matter.  Dr Bennett referred me to the decision in the Farbwerke case (supra) and I would agree with her the facts are similar to the present case.  However the question of amending the formula was not argued before Gibbs J and I cannot speculate as to whether he would have allowed such an amendment.  Gibbs J did allow the deletion of the reference to the incorrect formula in the process claims on the basis the description of the process was correct and the deletion would be to allow effect to be given to the document as a whole.  However, as Mr McCormack pointed out, in Farbwerke the formula did not apply to any of the compounds and Gibbs J distinguished this situation from that where, as in the present case, the formula correctly applied to some compounds.  Claim 1 in Farbwerke was of a similar form to claim 1 of the Merck application, being to a compound defined in terms of a general formula.  Concerning the interpretation of claim 1 Gibbs J said in the passage bridging pages 238 and 239:

"it was submitted that claim 1, properly interpreted in the context provided by the specification as a whole, is a claim to a compound produced by one of the processes described in the specification.  I find it impossible to construe claim 1 in that way.  The words of the claim show clearly that it is a claim to a compound described only by reference to a formula ..."

Much of Merck's evidence and submissions go to the issue of the correctness of the processes described, and that the compounds are defined in terms of those processes and hence the amendment to the formula does not involve new matter.  I cannot construe the claims to the compounds per se as being limited to compounds produced by a particular process.  Following Gibbs J in Farbwerke I am bound to construe those claims as claims to compounds described only by reference to the formula given.  Since the formulae in the original and accepted applications differ in the radical at the 25-position, I am led to the conclusion the claims to compounds per se defined in terms of the formula are directed to different subject matter in the original and accepted applications.
         I would agree with Dr Bennett the decision in Chevron (supra) can be distinguished on the facts since Merck have substituted one compound for another rather than adding extra compounds.  The decision in Chevron is based on obvious mistake which I do not believe to be at issue in the present case.
         Mr McCormack relied heavily on the decision in Zambon (supra) in his submissions.  While I would agree the facts are similar in relation to substitution of one formula for another, as I have said above I do not believe Merck were reckless in proposing n-propyl originally.  In Zambon the patentees had doubts as to the correctness of the formula well before publication of the specification and as matter of discretion, having regard to the way in which they had dealt with the matter, were not allowed to amend.
         In relation to the question of disclosure of the "b" compounds in the original specification, I would agree with Dr Bennett there was some disclosure of those compounds.  However the point I am presently considering concerns the structural formula of those compounds.  Throughout the original specification the "b" compounds are characterised in terms of an n-propyl substituent at the 25-position.  I can find no disclosure in the original specification pointing to an iso-propyl substituent at the 25-position.  Although, as I have previously said, the processes described in the original specification inevitably led to compounds with an iso-propyl substituent at the 25-position, the compounds per se are defined in terms of a structural formula which nominates n-propyl as the 25-substituent.  This definition is in my view clear and unambiguous both at page 4, where the compounds of the invention are stated to have the given structural formula, and in claim 1.  The compounds are defined not in terms of a process but by a structural formula.  I therefore conclude that, in the accepted specification, claims to compounds defined in terms of a structural formula in which the 25-substituent is iso-propyl involve new matter and should be accorded a later priority date in accordance with s.159A.

         The statement of amendments to change the substituent at the 25-position from n-propyl to iso-propyl was lodged at the Patent Office on 18 December 1981.  The earliest possible priority date for claims to compounds defined in terms of an iso-propyl substituent is therefore 18 December 1981.  Although Mr McCormack put to me s.159A(5) could operate to give the later priority date as the date of lodgement of the fourth statement of proposed amendments, I do not believe anything turns on this issue in the present case and so I will not consider the matter further.
         Both parties agreed that the compound B-41D in application No 61668/80 by Sankyo Co Ltd is identical to the compound of claim 1 of the accepted Merck application in which R1 is hydrogen, R2 is hydrogen, R3 is iso-propyl, R4 is hydrogen and the broken line is a single bond.  The Sankyo application was published on 5 March 1981 and therefore I must consider whether it constitutes a prior publication in respect of one compound of the Merck application.
         I have found claims to compounds defined in terms of a structural formula in which the 25-substituent is iso-propyl, have the earliest possible priority date of 18 December 1981.  Claim 1, when it defines R3 as iso-propyl, takes this priority date.  Since this claim includes the Sankyo compound B-41D it is prior published with respect to that one compound.  Similarly, claim 2 includes the Sankyo compound and is also prior published.  Claims 3-19 are directed to compounds having a sec-butyl substituent at the 25-position and therefore I do not need to consider further these claims.  Claim 32 is the only other claim which refers to a structural formula, and since this claim is directed to a process which is not common to the Sankyo application I do not need to consider it further.
         I must now consider the remainder of the claims which do not refer to a structural formula.  Claim 20 is an added claim to a compound by name, which compound, if it falls within the structural formula of accepted claim 1, is identical to the Sankyo B-41D compound.  Dr Bennett put to me this claim should be interpreted in the light of the starting materials and the definition of the R3 radical at the 25-position is irrelevant.  Dr Horner deposes the definition of the compound relies on the definition of the 25-position substituent, which definition was changed from the original to the accepted specification.  There is no direct evidence before me as to how claim 20 should be interpreted.  However I am inclined to agree with Dr Horner's interpretation that the claim relies on the definition of the 25-substituent.  The claim is a per se claim to a compound however it may be prepared.  To fall within the definition of the compounds of the invention at page 4 and in claim 1, the compound of claim 20 must fall within the structural formula of those compounds.  It must therefore have an iso-propyl substituent at the 25-position.  As I have found there was no specific disclosure of iso-propyl in the original specification I conclude claim 20 is also directed to new matter, and as such has the earliest priority date of 18 December 1981.  I also believe to conclude otherwise would be inconsistent with my finding in relation to claim 1.  I therefore consider claim 20 is also prior published by the Sankyo application.
         Concerning the remainder of the claims which may include the B-41D Sankyo compound, there is no evidence before me as to how those claims should be interpreted.  Claims directed to a process, such as claim 33, clearly do not involve new matter since the process remained unchanged in the accepted specification.  Claims to a "compound obtainable" by a specified process were agreed by both parties to be product claims.  Dr Bennett submitted the claims depended on the starting materials which remained unchanged from the original to the accepted application.  Mr McCormack submitted the wording of these claims should be limited to the compounds obtained by the process since the term "obtainable" implied a de facto claim to the product itself.  I am inclined to agree with Mr McCormack the term "obtainable" does not limit the product to one which must be prepared by the particular process.  Therefore, for the same reasons as given above for claim 20, claims 21, 22, 28 and 31 which include the compound identical with the Sankyo B-41D compound are prior published.
Conclusion
         I have found the opposition succeeds on the ground of anticipation.  Since this concerns only one of the many compounds included in the Merck specification, I allow the applicant 60 days from the date of this decision to propose amendments with a view to removing the anticipated subject matter.

I award costs as following the event, in favour of the opponent.

(JANET WERNER)
   Supervising Examiner