(cid:1)
`
`
`
`The Opponent:
`
`
`The Applicant:
`
`[emblem]
`State of Israel
`Commissioner of Patents, Designs and Trademarks
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Opposition to Patent Application
`No. 172563
`
`TEVA PHARMACEUTICAL INDUSTRIES LTD.
`By counsel S. Horowitz & Co.
`
`MERCK SHARP & DOHME CORP.
`By Counsel Liad Whatstein & Co.
`
`
`
`D E C I S I O N
`
`
`
`
`
`
`
`1.
`
`2.
`
`3.
`
`I have before me the opposition of Teva Pharmaceutical Industries Ltd. (hereinafter: “Teva” or the
`“Opponent”) to Patent Application no. 172563 (hereinafter: the “Patent Application” or the
`“Application”), which was filed by Merck & Co., Inc., USA (hereinafter: “Merck” or the
`“Applicant”).
`
`The Patent Application was filed on April 18, 2008, and it is entitled: “Phosphoric acid salt of
`dipeptidyl peptidase IV inhibitors”. The application claims priority on the basis of Provisional
`Application No. 60/482161 (hereinafter: the “Provisional Application”), which was filed in the
`United States on June 24, 2003 (hereinafter: the “Priority Date” or the “Effective Date”).
`
`The Application was published in Patents Journal 7/2008, and the Opponent notified of its
`opposition to it on October 6, 2008.
`
`Background
`
`4.
`
`The Application claims a dihydrogen phosphate salt as well as a monohydrate crystalline form of
`the same salt of the active ingredient known as 4-oxo-4- [3-(trifluoromethyl)-5,6-dihydro[1,2,4]
`triazolo[3,4-a]pyrazine-7(8H)-yl]-1-2(2,4,5-trifluorophenyl)butan-2-amine, which will hereinafter
`be referred to as “sitagliptin”. Sitagliptin is a compound that has an inhibitory activity on the
`enzyme dipeptidyl peptidase-IV (DPP-4), an activity which, according to the Patent Application,
`constituted a new approach in the treatment of type-2 diabetes.
`
`5.
`
`The application includes 26 claims, one of which is independent, as stated below:
`
`
`
`
`
`1 of 49
`
`IPR2020-01045, Teva Ex. 1021
`IPR2020-01060, DRL Ex. 1021
`IPR2020-01072, Sun Ex. 1021
`
`

`

`[emblem]
`State of Israel
`Commissioner of Patents, Designs and Trademarks
`(cid:1)
`
`
`
`Claim No. 1 claims the dihydrogen phosphate salt; Claims 2 and 3 claim the salt which is the
`subject of Claim 1, where the chiral center of the molecule has an (R) configuration and an (S)
`configuration, respectively; Claim 4 claims the salt which is the subject of Claim 1, when it is
`characterized as a monohydrate crystal; Claims 5 to 11 claim the salt which is the subject of Claim
`4, having different characteristics; Claims 12 to 17 claim a medicinal substance comprising various
`weight concentrations of the crystal that is the subject of Claim 4; Claim 18 claims a
`pharmaceutical composition comprising a therapeutically effective amount of the salt in Claims 1
`or 4, in combination with one or more pharmaceutically acceptable carriers; Claims 19 to 22 are
`process and use claims; Claims 23 to 26 are omnibus claims.
`
`6.
`
`This is the language of Claim 1:
`
`
`
`“a dihydrogenphosphate salt of 4-oxo-4-[3-(trifluoromethyl)-5,6-dihydro[1,2,4]triazolo[4,3-
`a]pyrazine-7(8H)-yl]-1-2(2,4,5-trifluorophenyl)butan-2-amine of structural formula l:
`
`(cid:4)(cid:2)
`
`
`
`(cid:5)(cid:2)
`
`The Parties' Evidence
`
`The Opponent submitted on its behalf: an expert opinion of Prof. Abu Serajuddin (hereinafter:
`“Serajuddin 1”), an expert opinion of Dr. Leonard J. Chyall (hereinafter: “Chyall”) and the
`affidavit of Mr. Darryl W. Hendricks.
`
`The Applicant submitted on its behalf: an expert opinion on behalf of Prof. Jerry L. Atwood
`(hereinafter: “Atwood”), an affidavit of Dr. Robert M. Wenslow and an affidavit of Mr. Robert Di
`Vincenzo.
`
`The Opponent submitted evidence in reply by way of an additional expert opinion of Prof.
`Serajuddin (hereinafter: “Serajuddin 2”), and an additional expert opinion of Dr. Chyall.
`
`
`
`
`
`2
`
`2 of 49
`
`IPR2020-01045, Teva Ex. 1021
`IPR2020-01060, DRL Ex. 1021
`IPR2020-01072, Sun Ex. 1021
`
`7.
`
`8.
`
`9.
`
`(cid:1)(cid:1)
`
`

`

`[emblem]
`State of Israel
`Commissioner of Patents, Designs and Trademarks
`(cid:1)
`
`10. After the main round of evidence, both parties filed additional evidence, on which I shall elaborate
`below. I note, however, that on behalf of the Applicant, these were the affidavits of Prof. Atwood
`(dated August 26, 2012; March 20, 2013; September 13, 2013), which will hereinafter be referred
`to as “Atwood 1”; “Atwood 2”; and “Atwood 3”, respectively. On behalf of the Opponent these
`were the affidavits of Dr. Chyall (dated January 24, 2013; and February 19, 2013), which will be
`hereinafter referred to as “Chyall 1” and “Chyall 2”, respectively.
`
`The Uncontested Facts
`
`11.
`
`International Publication WO 03/004498 (hereinafter: “Publication ‘498”) discloses a Markush
`formula which includes a large number of compounds and exemplifies 33 of them. Example 7 of
`Publication ‘498 describes the hydrochloride salt of sitagliptin.
`
`12. Publication ‘498 also includes a list of preferred acids for the formation of salts of the compounds
`(p. 10 of the Publication). Among these acids is also the phosphoric acid which forms the
`phosphate salt. The dihydrogen phosphate salt of sitagliptin, which is claimed in the Patent
`Application that is before me, will hereinafter be referred to as the “DHP salt”.
`
`13. The DHP salt is a salt consisting of one molecule of sitagliptin and one molecule of phosphoric
`acid in a ratio of 1:1, and it is formed by the transfer of the first of three protons of the phosphoric
`acid to sitagliptin.
`
`Novelty
`
`14. According to the Opponent, the Applicant itself proposed to prepare the DHP salt before the
`Priority Date as transpires from Publication ‘498, by indicating the phosphoric acid as one of eight
`preferred acids in that publication. According to the Opponent, the question of novelty should be
`examined in accordance with the infringement test. In other words, it is the Opponent's position
`that if performing the stated in Publication ‘498 constitutes an infringement of the Patent
`Application, then the invention claimed in the Patent Application lacks novelty. The Opponent
`further argues in this context that this test holds even when the previous publication does not
`explicitly exemplify the invention, but performing the instructions in the prior art inevitably leads
`to the invention.
`
`In the present case, the Opponent argues that an person of ordinary skill in the art who would react
`the active ingredient sitagliptin with phosphoric acid to form a salt as proposed in ‘498 would
`necessarily obtain the DHP salt as the only stable phosphate salt. Therefore, the Opponent holds
`that the DHP salt was disclosed in that publication in a manner that takes away novelty from the
`Patent Application. The Applicant, on the other hand, holds that such activity could result in
`additional products, such that it was not possible to expect the resulting product prior to performing
`experiments that test it.
`
`3
`
`3 of 49
`
`IPR2020-01045, Teva Ex. 1021
`IPR2020-01060, DRL Ex. 1021
`IPR2020-01072, Sun Ex. 1021
`
`15.
`
`(cid:1)(cid:1)
`
`

`

`[emblem]
`State of Israel
`Commissioner of Patents, Designs and Trademarks
`(cid:1)
`
`16.
`
`In other words, the question being asked here is whether it is sufficient that a compound and an
`acid that are included in a previous application as one of the options therein negate novelty from a
`patent application that claims that specific salt.
`
`17. Publication ‘498 teaches compounds that are included in a Markush formula and processes for their
`preparation. As is well known, the use of a Markush formula allows the patentee to obtain certain
`protection over the variety of chemical compounds that are included in the formula, as well as over
`the processes of their preparation, if these are patentable in accordance with the terms prescribed in
`the Law.
`
`18.
`
`In order to negate novelty from an invention, it must be shown that there is a single prior
`publication that fully includes the invention’s components in a manner that enables a person of
`ordinary skill in the art to carry out the invention (CA 345/87 Hughes Aircraft Company v. State of
`Israel, PD 44(4), 45, 102-105 (1990)) (hereinafter: “Hughes”). Therefore, when a prior publication
`expressly claims a substance, it will clearly negate novelty from a later application seeking
`protection over the same substance:
`
`“A specific disclosure of a substance invalidates a claim to the substance,
`regardless of question of advantage.”
`(Blanco White, 4th edition, pp. 4-110)
`
`
`19. However, when a prior publication does not describe a known compound, but rather the compound
`in its free base form, its manner of preparation and its preferred salts, the question arises whether it
`can negate novelty from a later application that includes one specific compound from among all the
`possible compounds that exist in it.
`
`20.
`
`In the decision regarding opposition to patent application 55660 (15.1.84), it was stated that: “For
`a chemist, a compound is not a known compound until it has actually been prepared and
`identified by one of its properties or characteristics.”
`
`21. This question was discussed again, albeit in obiter dictum, in CA 8802/06 Unipharm Ltd. v.
`SmithKline Beecham Plc (published in Nevo, 05-18-2011) (hereinafter: “SmithKline”). The court
`held that in order to recognize the patentability of a product which is a component of a group
`described in a prior patent, it must be shown that the product is new and has an inventive step, as
`follows:
`
`“Thus, for example, the novelty requirement defined in section 4 of the Patents
`Law requires examining whether a property discovered in a component or
`components from a group described in a prior patent was disclosed and
`known; the inventive step requirement set forth in Section 5 of the Patents
`
`IPR2020-01045, Teva Ex. 1021
`IPR2020-01060, DRL Ex. 1021
`IPR2020-01072, Sun Ex. 1021
`
`4
`
`4 of 49
`
`(cid:1)(cid:1)
`
`

`

`[emblem]
`State of Israel
`Commissioner of Patents, Designs and Trademarks
`(cid:1)
`
`Law requires examining whether a new property discovered in a selected
`component, for which a new patent is being sought, is a property that
`embodies a substantial advantage that may constitute an appropriate
`consideration for granting a monopoly to the inventor.”
`
`22.
`
`In other words, according to this ruling, in order to determine that a component was already
`described in a prior publication, such component need not be expressly exemplified; it is sufficient
`that it is described as part of a group for the question of novelty to arise in respect of it.
`
`23. This was adopted in the work directives of the Patents Authority (see Section 6.5 of Appendix F,
`Edition 4, Work Directive F/23.1), where it is stated:
`
`“If the publication that discloses the broad group does not include examples
`that describe the members of the narrow group, and the specification of the
`application under consideration does not describe a substantial and
`unexpected advantage of the entire claimed group, the Examiner must
`indicate a deficiency in regard to Sections 4 and 5 of the Law, based on the
`publication that describes the broad group to which the claimed group of
`members belongs.”
`
`24. Although this section of the Work Directives was incorporated into them after the time this
`Opposition was heard, it is based on the judgment of the Supreme Court in the SmithKline case,
`published in May 2011.
`
`25. On this question, the British approach is different, following the European Directive and the case
`law of the European Union. In the case of Dr Reddy’s Laboratories Ltd v Eli Lilly & Co [2009]
`EWCA 1362, it was determined that for a prior publication to negate the novelty of a compound, it
`must describe the compound individually (“individualized description”). Sometimes, selecting the
`same compound from among all the compounds that are included in the Markush formula is
`nothing more than finding a needle in a haystack. Therefore, the Markush formula will not preclude
`novelty from a single compound included in it:
`
`“The contention amounts to this: that every chemical class disclosure discloses
`each and every member of the class. It would, it seems, even apply if the
`formula had simply been written down without any suggested utility.
`
`I reject the contention for two reasons: firstly as a matter of a priori reasoning
`and secondly because it is inconsistent with settled EPO Board of Appeal case
`law.
`
`IPR2020-01045, Teva Ex. 1021
`IPR2020-01060, DRL Ex. 1021
`IPR2020-01072, Sun Ex. 1021
`
`5
`
`5 of 49
`
`(cid:1)(cid:1)
`
`

`

`[emblem]
`State of Israel
`Commissioner of Patents, Designs and Trademarks
`(cid:1)
`
`First then, the a priori considerations apart from case-law. An old question
`and answer runs as a follows: “Where does a wise man hide a leaf? In a
`forest.” It is, at least faintly, ridiculous to say that a particular leaf has been
`made available to you by telling you that it is in Sherwood Forest. Once
`identified, you can of course see it. But if not identified you know only the
`generality: that Sherwood Forest has millions of leaves.” (Dr Reddy’s
`Laboratories Ltd v Eli Lilly & Co [2009] EWCA 1362)
`
`It is noted that in that case, it was a Markush formula that included 1019 compounds, of which
`86,000 were preferred. What degree of detail amounts to individualized description has not been
`decided in the same matter:
`
`“It is not necessary here to go into what is sufficient to amount to an
`“individualized description.” “Obviously the question may partly be one of
`degree, but other considerations may come in too, for instance, the specificity
`of any indicated purpose for making the compounds.”
`
`The court reiterated this rule also in more recent decisions (GlaxoSmithKline UK Ltd v Wyeth
`Holdings LLC [2016] EWHC 1045 (Ch) (12 March 2016, paragraph 157).
`
`26. Literature in the United Kingdom also adopted this rule to a large extent. The scholars Richard
`Miller QC, Guy Burkill QC, Colin Briss QC and Douglas Campbell (in their book “Terrell on the
`Law of Patents”, 11-46 (17th ed.)), discussed the requirement for an individualized description that
`can preclude novelty:
`
`“So far as novelty was concerned, the court rejected the broad proposition
`that the disclosure of a generalized class necessarily amounted to disclosure of
`each and every member of it. Such rejection applied whether the earlier class
`was described by way of a general formula or by an itemized listing out of all
`the members. What was required for anticipation was an “individualized
`
`IPR2020-01045, Teva Ex. 1021
`IPR2020-01060, DRL Ex. 1021
`IPR2020-01072, Sun Ex. 1021
`
`6
`
`6 of 49
`
`(cid:1)(cid:1)
`
`

`

`[emblem]
`State of Israel
`Commissioner of Patents, Designs and Trademarks
`(cid:1)
`
`27.
`
`description” of the later claimed compound or class. Absent that, it could not
`be said that performing any part of the earlier generalised disclosure would
`inevitably produce a result within the later claim. This approach mirrored
`that of the EPO and Germany.” (Terrell on the law of Patents, paragraph 11-61)
`In our case, Publication ‘498 exemplifies the manner of preparation of 7 compounds, including the
`process of preparation of the hydrochloride salt of sitagliptin (Example 7). Further down, 26
`additional compounds are specified, but their preparation methods are not described. However, the
`Publication states that their method of preparation is similar to that described in respect of
`Examples 1 to 7. That is, a total of 33 compounds were exemplified in this publication. In Claim
`15, each of these 33 compounds and their pharmaceutically acceptable salts are claimed. The
`publication also teaches that the compounds described are basic and therefore salts can be prepared
`from them using pharmaceutically acceptable acids, the list of which is provided on page 10 of the
`Publication. At the end of the list, 8 acids were indicated as particularly preferred, one of which is
`the phosphoric acid.
`28. The Applicant claims that out of all these possible alternatives, Publication ‘498 does not teach or
`give preference to any one specific compound of the general formula. The Applicant also states that
`this publication does not include reference to additional properties such as side effects, toxicity,
`absorption, bioavailability, preferred form of administration, physical and chemical properties of
`the compounds, and more.
`In SmithKline, the court held that even though the free base, rosiglitazone, was exemplified in the
`prior publication, its salt, maleate, was not exemplified and therefore this salt constitutes a “new
`substance”:
`“the 228 Patent refers to salts that belong to the group of TZD components
`claimed therein (p. 4 of the 228 Patent) and it lists 31 examples of preparation
`of TZD components. Example 30 in the specification of the 228 Patent
`describes a process for the preparation of rosiglitazone(p. 38 of Patent 228),
`and in the claims chapter of the 228 Patent, rosiglitazone is claimed in Claim
`12 (pp. 41-42 of Patent 228). However, as the lower court ruled, and justly so,
`the rosiglitazone maleate salt is not mentioned in the 228 Patent. It is not
`superfluous to refer again in this context to the expert opinion of the
`Appellants' expert, Prof. Breuer, in which it is clearly and boldly stated that
`the 228 Patent does not expressly describes the rosiglitazone-maleate salt
`(Section 9 of the “first opinion” dated February 26, 2004. Reference should
`also be made to his testimony before the lower court, in which he
`acknowledged that the rosiglitazone maleate salt is ‘itself a new substance.’”
`(Para. 41 of the judgment) (emphases added)
`
`29.
`
`IPR2020-01045, Teva Ex. 1021
`IPR2020-01060, DRL Ex. 1021
`IPR2020-01072, Sun Ex. 1021
`
`7
`
`7 of 49
`
`(cid:1)(cid:1)
`
`

`

`[emblem]
`State of Israel
`Commissioner of Patents, Designs and Trademarks
`(cid:1)
`
`30.
`
`It seems that this statement of the Supreme Court does not contradict the general ruling of the court
`in the SmithKline case, according to which the disclosure of a group that was described previously
`also means the disclosure of its components. I will further note that although in our case the DHP
`salt was not described, yet the broad group (namely, the group of the 33 free bases and the 8
`preferred acids) was indeed described. Out of the 33 bases described, the preparation methods of
`only seven, including sitagliptin, were described. Thus the group from which it was recommended
`to select is in fact even narrower and includes only 56 options.
`
`31. The application of the rule set forth in the SmithKline judgment to our case requires examining
`“whether a property discovered in a component or components from a group that was
`described in a prior patent was manifest and known.” The Applicant’s contention is that a
`person skilled in the art would not have known that the DHP salt is stable, with suitable solubility
`and other properties that make it suitable for its designation as a medicinal preparation. In other
`words, the properties discovered in the DHP salt were not known to a person skilled in the art from
`Publication ‘498. I cannot accept this contention. Once sitagliptin was claimed as a free base and
`also its pharmaceutically acceptable salts were claimed (in Claim 15), the applicant disclosed its
`opinion that all these are beneficial and proper for the purposes that it asserted, otherwise why did
`the applicant claim them? Moreover, since the inventor recommended 8 particularly preferred acids
`that were suitable for coupling with the compounds in the form of the free base that were
`exemplified, it cannot be argued that their preference over other acids was not manifest and known.
`
`32. The Applicant's attempt to claim that nothing can be learned from Publication ‘498 is reflected in
`the testimony of Dr. Wenslow on its behalf (p. 136 line 28 to p. 137 line 18 of the minutes of the
`hearing dated February 2, 2015):
`
`“Adv. Band: So why is Merck telling the world about pharmaceutical salts in
`this case?
`Wenslow: It's a boilerplate of what possibly could happen down the road,
`Adv. Band: The claims are also boiler claims at Merck?
`Wenslow: In this patent, the claims a boilerplate. The DP claims are,
`Adv. Band: What do you mean by boilerplate claim?
`Wenslow: So, so the 33 compounds, that's what I'm talking about boilerplate
`claim, in a sense that there are 33 compounds and it's any one of those
`compounds could have the DPP4 activity, they didn't know at the time.
`
`Adv. Band: And so when, when they wrote in the claims, anyone of those
`compounds, including sitagliptin and pharmaceutical salts, pharmaceutically
`acceptable salts, that does not mean that to Merck or to the reader, that they
`actually made pharmaceutical salts. Maybe no salts at all.
`
`IPR2020-01045, Teva Ex. 1021
`IPR2020-01060, DRL Ex. 1021
`IPR2020-01072, Sun Ex. 1021
`
`8
`
`8 of 49
`
`(cid:1)(cid:1)
`
`

`

`[emblem]
`State of Israel
`Commissioner of Patents, Designs and Trademarks
`(cid:1)
`
`Wenslow: You don't have to make a salt, there is no reason to have to make
`salt,
`Adv. Band: Ok, but was it a possibility part of the teaching?
`Wenslow: I mean it's mentioned, so maybe it's a possibility,”
`
`33. Further on, Dr. Wenslow was asked what he, as a person skilled in the art, learned from a
`recommended list in another patent, and he replied as follows (p. 162 lines 7-13 of the same
`minutes):
`
`“Adv. Band: Again when you read this description in patent 798, what do you
`understand the message of the patentee, what is he trying to tell you? What's
`the difference
`in his mind between the whole
`list, which
`includes
`hydrochloride but also phosphate, and the shorter list which for in this case
`includes hydrochloride but does not include phosphate.
`
`Wenslow: I can't speak towards the inventor of this patent. All I can say is in
`salt research there is absolutely no way we would limit ourselves to any type of
`list besides those that a toxicologist would say are safe for human use.”
`
`34.
`
`It is clear that an approach according to which patents are not intended to teach anything to persons
`skilled in the art nullifies the content of the patent system, of which entire purpose is to disclose the
`invention to the public in return for granting a monopoly to the inventor. This is especially
`inconceivable when it is stated by a person who is registered as one of the inventors in the Patent
`Application that is before me. On the relationship between the disclosure of the invention and the
`receipt of the consideration for it, see LCA 6025/05 Merck & Co. Inc. v. Teva et al. (Published in
`Nevo, 05-19-2011):
`
`“The monopoly granted to the patent holder is conditional on the disclosure of
`all the components of the invention to the general public as part of the
`publication of the acceptance of the patent application. The purpose of this is
`first to enable the filing of oppositions to the patent application by those
`disputing the right to register the patent on the invention. Second, the full
`disclosure is intended to enable the public to utilize the details of the invention
`after the exclusivity period has ended. Thus, alongside the protection afforded
`by the law to the patent owner, by granting a monopoly on the invention
`
`for a specified period of time, the patent owner is obligated to share all the
`details of the invention with the general public (HCJ 280/60 Abic Chemical
`Laboratories Ltd. v.
`the Competent Authority for
`the Importation of
`Pharmaceutical Preparations, PD 15 1323, 1336-1337 (1961) (hereinafter:
`
`IPR2020-01045, Teva Ex. 1021
`IPR2020-01060, DRL Ex. 1021
`IPR2020-01072, Sun Ex. 1021
`
`9
`
`9 of 49
`
`(cid:1)(cid:1)
`
`

`

`[emblem]
`State of Israel
`Commissioner of Patents, Designs and Trademarks
`(cid:1)
`
`“Abic”); CA 665/84 Sanofi Ltd. v. Unipharm Ltd., PD 41(4) 729, 742-743 (1987)
`(hereinafter: “Sanofi”); CA 227/86 Schechter v. Avmetz Ltd., PD 44(2) 846,
`852-854 (1990) (hereinafter: “Schechter”). The proprietary right that may be
`acquired, if a patent application is approved, involves the cost of disclosing the
`details of the invention to the public at an early stage of the patent application
`process, and taking the risk that competitors will use that information to their
`advantage, as shall be explained below.”
`
`35. Moreover, an approach such as that of the Applicant denies in fact any possibility of relying on
`information described in patents both for the purpose of further research and as prior publications
`precluding novelty or inventive step.
`
`36.
`
`It follows from the foregoing that in view of the stated in Publication ‘498, the formation of a
`phosphate salt of sitagliptin was known before the Effective Date. The Applicant's claim is that
`there is no certainty as to which phosphate salt will form (monohydrogen or dihydrogen) and
`therefore this publication does not preclude novelty from the patent application. In contrast, the
`Opponent claims that according to the “inherent anticipation” rule, this salt is included in
`Publication ‘498, since it is the only [salt] that can be formed by reacting sitagliptin as a free base
`and phosphoric acid.
`
`The Inherent Anticipation Rule
`
`37. The Opponent's contention, then, is that the DHP salt was inherently described in Publication ‘498,
`and therefore this publication takes away novelty from the invention before me. This is because,
`according to the Opponent, this salt is the only stable salt that is expected to be obtained as the
`product of a reaction between sitagliptin and phosphoric acid, a reaction that was proposed in
`Publication ‘498.
`
`38. As was clarified in the Opponent's summations, its contention that the DHP salt is the only product
`possible in such a reaction was raised in response to the opposition division's decision in the
`opposition to the corresponding European patent. In that decision, it was ruled that once the
`existence of other phosphate salts of sitagliptin could not be ruled out beyond any reasonable doubt
`and no experiments were presented to support the position that there was only one option of
`forming a salt with phosphoric acid, the Opposition must be dismissed.
`
`39. From the point of view of patent law, the Opponent needed the “inherent anticipation” rule. Its
`meaning is that following the instructions in the prior publication will inevitably lead to executing
`the invention, and therefore the invention was inherently described in the prior publication. For a
`
`IPR2020-01045, Teva Ex. 1021
`IPR2020-01060, DRL Ex. 1021
`IPR2020-01072, Sun Ex. 1021
`
`
`
`10
`
`10 of 49
`
`
`
`(cid:1)(cid:1)
`
`

`

`[emblem]
`State of Israel
`Commissioner of Patents, Designs and Trademarks
`(cid:1)
`
`prior publication describing the method of preparation of an invention to inherently preclude
`novelty from a later application, the publication must describe the method of preparation so that its
`actual implementation will yield only one result – the result described in the later application (see
`Opposition to Patent Application No. 115314, Eli Lilly and Company v. Unipharm Ltd., paragraph
`44 (03-11-2004)).
`
`40. Note well: it is not sufficient that it is “most likely” that repeating the process described in the
`application will lead to the result described in the previous publication – it must be shown that the
`outcome reached is the only possible outcome, as follows:
`
`likelihood, does not suffice” (Inhale
`“Likelihood, even overwhelming
`Therapeutic Systems Inc. v Quadrant healthcare Plc. [2002] RPC 21)
`
`41.
`
`In addition, the party wishing to deny novelty from an application, by performing experiments that
`repeat the art stated in the invention in order to examine the result, must precisely follow the
`instructions of the publication and take care that the experiments are conducted in accordance with
`the knowledge at the relevant date and without hindsight:
`
`“In repeating any prior art proposal by experiment with a view to proving
`lack of novelty, the party seeking to do so must adhere faithfully to the
`teaching of the source document. Moreover, the interpretation of the cited
`document for the purpose of such experiment is deemed to be assessed as of
`the date of its publication, and the interpretation is also deemed to be the job
`of the skilled worker, whose knowledge must not be supplemented by later
`knowledge.” (Terrell, p. 11-42)
`
`42. As stated, the Opponent argues that performing the [instructions] in Publication ‘498 will
`inevitably lead to obtaining the DHP salt as the only stable salt. In order to prove or refute this
`contention, both the Applicant and the Opponent conducted experiments. Prof. Atwood testified on
`that matter on behalf of the Applicant, and the Opponent submitted an affidavit by Dr. Chyall. The
`theoretical background of these experiments was described in the expert opinions of Prof.
`Serajuddin and Prof. Atwood.
`
`43. Before describing these experiments, their theoretical background will be clarified. It is centered on
`a rule the existence of which was questioned – the “delta pKa” rule (hereinafter: also “ΔpKa”).
`
`11
`
`11 of 49
`
`IPR2020-01045, Teva Ex. 1021
`IPR2020-01060, DRL Ex. 1021
`IPR2020-01072, Sun Ex. 1021
`
`
`
`
`
`(cid:1)(cid:1)
`
`

`

`[emblem]
`State of Israel
`Commissioner of Patents, Designs and Trademarks
`(cid:1)
`
`44. Ka is a value expressing the strength of an acid, namely its tendency to transfer a proton (H+). This
`value is measured in the reaction between the acid and water. The more the acid reacts with the
`water, that is, the reaction is in the direction of the products, the higher this value is. Conversely,
`the more the reaction is directed towards the reactants, namely only some of the acid reacts with the
`water, the lower this value is. pKa is defined as -logKa. Consequently, the weaker the acid, the
`higher its pKa value (Appendix B to Prof. Atwood's affidavit).
`
`45. The Opponent contends that one can predict whether reacting a base and acid will form a stable
`salt, in accordance with the delta pKa rule. The Opponent holds that according to this rule, there
`should be a difference of at least two units between the pKa value of the acid and the pKa value of
`the base. In other words, the pKa value of the acid must be at least two units lower than the
`corresponding value of the base.
`
`46. By contrast, the Applicant argued that the rule is different and is reflected in the professional
`literature. According to the Applicant, it is sufficient that the difference is positive (on the part of
`the acid) for the reaction to take place and for salt to be formed (see Gould, Appendix G of Prof.
`Atwood's Opinion).
`
`“Clearly to form a salt the pKa of the conjugate acid has to be less than or
`equal to the pKa of the basic center of the drug.”
`
`47.
`
`In our case, the pKa value of the active ingredient sitagliptin is 7.7, while the phosphoric acid has
`three hydrogens and, accordingly, three pKa values: 2.12, 7.21, and 12.67 (Serajuddin 1, Section
`49, Appendices 10-11). The Opponent’s argument is that since the delta pKa applies only in
`relation to the first hydrogen of the phosphoric acid (a difference of 5.58 units), a transfer of the
`first proton to the sitagliptin is expected in a reaction between sitagliptin and phosphoric acid,
`formingonly the DHP salt. In other words, in the reaction between sitagliptin and phosphoric acid,
`only one stable salt can be formed.
`
`48. The Applicant believes that at least two possible salts are possible as a result of said reaction: a salt
`in which the acid transfers one proton and another salt in which the acid transfers two protons.
`Moreover, it claims that the salt can be formed in non-aqueous solutions, such as a mixture of water
`and methanol, where the pKa values of the acid and of the base vary.
`
`49.
`
`In addition, the Applicant claims that there is no certainty that it will be possible to precipitate or
`isolate a salt that is formed in solution, regardless of the pKa difference. According to the
`Applicant, there is a long way between preparing salt in solution and preparing a solid salt which
`comes out of solution, and this, it argues, is a research challenge in itself. In support of this, the
`
`IPR2020-01045, Teva Ex. 1021
`IPR2020-01060, DRL Ex. 1021
`IPR2020-01072, Sun Ex. 1021
`
`
`
`12
`
`12 of 49
`
`
`
`(cid:1)(cid:1)
`
`

`

`[emblem]
`State of Israel
`Commissioner of Patents, Designs and Trademarks
`(cid:1)
`
`Applicant referred to opposing examples where there was no success in isolating solids and
`extracting the salt from the solution despite differences of 2 units and more in the pKa value.
`
`50.
`
`It seems that the parties agree that the question of whether salt formed in solution can be
`precipitated or isolated differs