`571-272-7822
`
`
`
`
`
`Paper 68
`Entered: April 16, 2021
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`____________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`____________
`
`MYLAN INSTITUTIONAL LLC and PFIZER INC.,
`Petitioner,
`
`v.
`
`NOVO NORDISK A/S,
`Patent Owner.
`____________
`
`IPR2020-00324
`Patent 8,114,833 B2
`____________
`
`Record of Oral Hearing
`Held: March 26, 2021
`____________
`
`
`
`
`Before ERICA A. FRANKLIN, JOHN G. NEW, and
`SUSAN L.C. MITCHELL, Administrative Patent Judges.
`
`
`
`
`
`
`
`
`
`
`
`
`IPR2020-00324
`Patent 8,114,833 B2
`
`APPEARANCES:
`
`
`ON BEHALF OF PETITIONER MYLAN INSTITUTIONAL LLC:
`
`BRANDON M. WHITE, ESQ.
`Perkins Coie LLP
`700 13th Street NW, Suite 800
`Washington, DC 20005-3960
`(202) 654-6206
`bmwhite@perkinscoie.com
`
`
` ON BEHALF OF PETITIONER PFIZER INC.:
`
`MICHAEL JOHNSON, ESQ.
`Willkie Farr & Gallagher LLP
`787 Seventh Avenue
`New York, NY 10019-6099
`(212) 728-8137
`mjohnson1@willkie.com
`
`
`ON BEHALF OF PATENT OWNER:
`
`JEFFREY OELKE, ESQ.
`Fenwick & West LLP
`902 Broadway
`New York, NY 10010-6035
`(212) 430-2747
`joelke@fenwick.com
`
`
`The above-entitled matter came on for hearing on Friday, March 26,
`2021, commencing at 10:00 a.m. EDT, via Videoconference.
`
`
`
`
`
`
`
`
`
`2
`
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`
`IPR2020-00324
`Patent 8,114,833 B2
`
`
`
`P-R-O-C-E-E-D-I-N-G-S
`
`10:00 a.m.
`JUDGE FRANKLIN: Good morning, I'm Judge Franklin and with
`me are Judges New and Mitchell. We welcome you to this hearing for IPR
`2020-00324. IPR 2020-01252 has been joined with this proceeding, and
`this hearing relates to U.S. Patent 8,114,833.
`We will begin with Counsel for the parties introducing yourselves
`and when doing so, please spell your name for the court reporter, beginning
`with Counsel for Petitioners?
`MR. WHITE: This is Brandon White from Perkins Coie for
`Petitioner Mylan Institutional, B-R-A-N-D-O-N, W-H-I-T-E.
`JUDGE FRANKLIN: And is there anyone else for Petitioner?
`MR. JOHNSON: Good morning, Your Honor, this is Michael
`Johnson from Willkie Farr & Gallagher on behalf of the Petitioner Pfizer
`Incorporated and that's M-I-C-H-A-E-L, J-O-H-N-S-O-N.
`JUDGE FRANKLIN: Thank you, and who do we have for Patent
`Owner?
`MR. OELKE: Good morning, Your Honor, this is this is Jeff Oelke
`from Fenwick and West on behalf of the Patent Owner and my name is
`spelled J-E-F-F and the last name is O-E-L-K-E.
`JUDGE FRANKLIN: And do we have anyone else for Patent
`Owner?
`MR. OELKE: I don't believe so, Your Honor.
`
`
`
`3
`
`
`
`IPR2020-00324
`Patent 8,114,833 B2
`
`JUDGE FRANKLIN: All right, thank you. Consistent with the
`hearing order, each party has a total of 45 minutes for its presentation.
`Petitioners may reserve a portion of their time to respond to arguments
`presented by Patent Owner, and Patent Owner has also been authorized to
`reserve a portion of its time for rebuttal.
`Please be mindful that a court reporter is transcribing this hearing and
`there is no shared display for the demonstrative exhibits. So, when
`referring to a particular demonstrative exhibit or slide, clearly identify it by
`number so that we may all follow along.
`We are in receipt of Petitioner's objections to three of Patent Owner's
`demonstrative slides and we will reserve ruling on them. However, Patent
`Owner, you should be prepared to address any inquiries from this hearing
`regarding whether the slides noted in the objections contain new arguments.
`And lastly, please remember to mute yourself when you are not
`presenting and we will also mute ourselves until we have a question or
`comment for you. That way, hopefully we may avoid unnecessary
`feedback during this remote hearing. If during the hearing you encounter
`any technical difficulties, please let us know.
`And now, Petitioner, you may proceed after indicating any time you
`would like to reserve for rebuttal. Mr. White?
`
`
`MR. WHITE: Your Honors, I believe Mr. Johnson is going to be
`making the argument today.
`JUDGE FRANKLIN: Okay, Mr. Johnson, how much time would
`you like to reserve?
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
`
`
`4
`
`
`
`IPR2020-00324
`Patent 8,114,833 B2
`
`MR. JOHNSON: Your Honor, I would like to reserve 15 minutes.
`JUDGE FRANKLIN: Okay, when you're ready you may begin.
`MR. JOHNSON: Good morning, Your Honors and may it please the
`Board, my name, as I said, is Mike Johnson, I'm from Willkie Farr &
`Gallagher on behalf of the Petitioner Pfizer Incorporated.
`The crux of this matter comes down to the fundamental notion that
`companies can't patent that which they've already disclosed to the public, but
`that is exactly what Novo is trying to do here today with the '833 patent.
`Novo's scientists filed the application that would become the Flink
`reference in 2001 and 2002. And in that application they disclosed
`formulations that match up with the '833 patent and disclose all the
`elements. Not only that, but Novo specifically amended the Flink
`application in order to specifically contain references and claims to the
`propylene glycol-containing formulation.
`To illustrate this, if you direct your attention to Slide 17,Slide 17
`discloses the relevant text from the first four priority applications of the
`Flink reference, Exhibits 1108, 1009, 1110, and 1111, they're the -- the
`polyhydric alcohols mentioned only include mannitol, sorbitol, or glycerol.
`However, in January of 2002 Novo specifically amended this
`application to include reference to propylene glycol, or PG, which will be
`the focus of the argument today. It was this conscious decision by Novo to
`describe and claim propylene glycol as an isotonicity agent in Flink that
`forms the basis of Grounds 1 and 2 of the validity challenges today.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`
`
`
`5
`
`
`
`IPR2020-00324
`Patent 8,114,833 B2
`
`If you could turn to Slide 7, what we have here is a comparison
`between Claim 1 and Claim 14 of Flink, and you will recall that Claim 14 of
`Flink, as described by the Petitioners, provides a roadmap, a roadmap that
`clearly shows that the Flink reference discloses each and every element in a
`specific recitation and combination, as required for a reference to be
`anticipatory.
`I don't think there's any doubt if you look at the blue boxes that both
`Flink and the claims of the '833 are directed to formulations containing a
`GLP1 agonist. In the red, again, Claim 1, we see a pH of about 7 to 10, we
`see the exact same range in the Flink reference in the Claim 1 that's
`incorporated into Claim 14.
`Turning now to the green, Flink discloses a buffer and we'll see the
`buffer was an important part of the Flink reference, it was this buffer that
`contributed to the stability of the Flink formulations. And Flink specifically
`discloses disodium phosphate buffers as required by Claim 1. A person of
`skill in the art would understand that this disclosure of the disodium
`phosphate buffer would include the dihydrate form.
`And the forms of a buffer once they're in solution don't really matter
`anyway. And finally, there's the orange that kind of appear a little yellow
`on my screen, a box related to the isotonicity agents. Again, when you look
`at this disclosure of what's disclosed in Flink, you'll see that PG is one of the
`few exemplified isotonicity agents.
`And again, this wasn't by accident, this was a conscious choice that
`Novo made in order to secure early patent coverage for the PG formulation
`6
`
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
`
`
`IPR2020-00324
`Patent 8,114,833 B2
`
`in the Flink reference. And indeed, the U.S. counterpart to Flink has been
`listed in the orange book and asserted against other Defendants who are
`looking to make a liraglutide formulation.
`The fundamental issue confronted by the Board today is, is PG
`specifically disclosed by the Flink reference, and does this specific recitation
`of a small list of chemical compound of excipients isotonicity agents contain
`an anticipatory disclosure?
`Petitioners believe that it does. So, with that, I'd like to turn our
`attention to ground one, which of course is anticipation over the Flink
`reference. I don't know if we need to but for the record, Flink is obviously
`Exhibit 1004 and we'll be focusing first on the anticipation.
`Slide 8, Dr. Tessier, Patent Owner's expert confirmed that when you
`look at a claim, a person of skill in the art would not only look at what's in
`the claim, but they will look at what's in the specification and review the list
`of agents that are used. And I think when you do that, respectfully, with the
`Flink reference, Claim 14, which provides the overarching roadmap, it
`provides all the elements arranged and combined in exactly the fashion.
`Patent Owner makes a big deal about picking and choosing and other
`sorts of arguments to defeat the anticipation argument, but there isn't picking
`and choosing. Claim 14 provides the roadmap, Claim 14 specifically tells
`you to use a buffer in combination with an isotonicity agent, and then the
`rest of the specification, as we'll turn to focus on now, describes exactly
`what those agents are.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`
`
`
`7
`
`
`
`IPR2020-00324
`Patent 8,114,833 B2
`
`And for the record, the Dr. Tessier testimony is Exhibit 1077 from his
`deposition transcript at Page 17, Line 8 to Page 18, Line 14.
`So, focusing first on the buffer, and if I could draw your attention to
`Slide 9? Flink specifically discloses and exemplifies disodium phosphate
`buffers in its list of buffers that can be used as part of the invention. A
`person of ordinary skill in the art looking to interpret Claim 14 would look
`to a spec to understand what specific elements are claimed.
`And not only is disodium hydrogen phosphate recited in the longer list
`of potential buffers in Flink, but it is specifically called out as one of the
`preferred buffers. There are only four preferred buffers in Flink and
`disodium hydrogen phosphate is one of them, and this is at Flink on Page 18,
`Lines 27 to 35. Now, the Flink uses the phrase disodium hydrogen
`phosphate.
`I don't think there's any dispute here that disodium hydrogen
`phosphate is the same as disodium phosphate buffers. A person of ordinary
`skill in the art would understand they're just different nomenclature for the
`same thing, and indeed this was confirmed by Dr. Tessier again in his
`deposition transcript, Exhibit 1077 at Pages 77, Lines 22 through 78, Page
`78, Line 11.
`Patent Owners make much that the specific hydration state of the
`buffer is not disclosed here but the hydration states for buffers aren't
`particularly important. A buffer works by being in solution and once a
`buffer is in solution, what hydration state is started in is of no moment. Dr.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`
`
`
`8
`
`
`
`IPR2020-00324
`Patent 8,114,833 B2
`
`Tessier confirmed this understanding as well, that once they're in solution
`the hydration state doesn't matter.
`Again, that's at Exhibit 1077 Page 78, Line 17 to Page 80, Line 1.
`So, here there's a clear disclosure and just for the record, Slide 10 also
`demonstrates the specific hydrate that was used consistently in Example 7.
`Again, this is an excerpt from Flink at 45 to 46 that shows the specific buffer
`being disodium hydrogen phosphate, which again is disodium phosphate and
`the dihydrate.
`The preferred embodiments of Flink was four buffers, one of which is
`the Claim disodium phosphate buffer. The hydration state is simply a
`matter of convenience and design choice by a person of ordinary skill in the
`art.
`
`But in any event, when looking further into the specification at
`example 7, a person of ordinary skill in the art would understand that the
`dihydrate is specifically described as an embodiment of the invention of
`Claim 14. Therefore, a person of ordinary skill in the art would understand
`the Claim 14 discloses the disodium phosphate buffer.
`Turning our attention now to the element that I think is really the crux
`of the dispute between the parties, propylene glycol. Claim 14 requires
`isotonicity agent, I don't think there's any dispute about that if we back to
`Slide 7 and we look at it. Claim 14 also requires that the isotonicity agent is
`present at a specific concentration.
`So, what we want to do now is look at what does Flink teach you
`about the particular isotonicity agents that can be used? And if I could
`9
`
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
`
`
`IPR2020-00324
`Patent 8,114,833 B2
`
`direct your attention to Slide 16, what we have here is a call-out from the
`Flink reference juxtaposed at the top with the language we've been look at in
`Claim 14 that provides the roadmap, and then the specific disclosure of the
`type of isotonicity agent that can be used with Flink to accomplish the
`formulation that is claimed in Claim 14.
`And if you look, there's been much made, there are 18 specifically
`identified chemical compounds here that can all function as an isotonicity
`agent. There are all common excipients that are routinely used in
`pharmaceutical formulations, there's no mystery about them. And
`specifically, one group of agents is the polyhydric alcohols and if you notice,
`the first example of a polyhydric alcohol mentioned is propylene glycol.
`Recall we looked earlier about how this wasn't by accident, this was
`specifically included by Novo in January of 2002 to claim PG-containing
`formulations. Indeed, if you look at Exhibit 2051, an exhibit relied upon by
`Patent Owners, at Pages 3 to 4 we see that one of the inventors of Flink, Ms.
`Larsen, actually suggested to the inventor of the '833 the use of PG because
`of the liquid at room temperature.
`Again, the specific disclosure of all these isotonicity agents are
`contemplated in 14 and, therefore, Claim 14 provides the roadmap that
`combines all of these elements. We'll talk about the sheer number and
`some of the Patent Owner's arguments in the meantime. But again, this is a
`good time I think to turn to the concentration and the concentration ranges,
`essentially, overlap one another.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`
`
`
`10
`
`
`
`IPR2020-00324
`Patent 8,114,833 B2
`
`Flink discloses a concentration of 1 to 100, the 833 patent discloses 1
`to 50. As the Board noted in its institution decision of Genentech v.
`Hospira, which I'd like to call the Hospira decision. We argued
`overlapping ranges render the claim obvious when there's no unexpected
`results. Here the ranges are overlapping.
`Claim 14 specifically provides it and therefore, Claim 14 provides the
`roadmap to make all of these claim formulations in the exact same manner
`that the '833 patent does.
`Novo raises several arguments in response to this. They claim that a
`person of ordinary skill in the art wouldn't know it was stable, wouldn't
`know it was efficacious, wouldn't know that it's free from side effects. First
`of all, we'll explain more in the obvious context why all those are wrong, but
`more fundamentally, anticipation doesn't require that.
`There's no requirement of showing of efficacy, the claims don't
`require that the formulation is actually isotonic, they just require the
`inclusion -- Claim 14 of Flink just requires the inclusion of an isotonicity
`agent and a buffer. And these are all standard excipients, standard
`isotonicity agents and buffers that are used according to their natural
`function. They're used in many, many other pharmaceutical formulations.
`JUDGE NEW: Mr. Johnson, excuse me for a moment. I'd just like
`to step back briefly. What does our case law say about overlapping ranges
`in cases of anticipation? You mentioned obviousness in Hospira but what
`about anticipation?
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`
`
`
`11
`
`
`
`IPR2020-00324
`Patent 8,114,833 B2
`
`MR. JOHNSON: Hospira's anticipation as well, Your Honor. If
`there's an overlapping range for a disclosure of a single point within a range,
`I believe the case law establishes that there is anticipation. Here there is no
`particular criticality to the range and in terms of anticipation, the disclosure
`of an overlapping range anticipates.
`And I think that was the holding of Genentech Hospira as well.
`JUDGE NEW: Thank you.
`MR. JOHNSON: The other case law issue that Novo raises is this
`idea of the sheer number of volume of combinations and they say there's 216
`combinations when you multiply the 18 times the 12 buffers. We think this
`is the wrong analysis. First of all, we don't think there's a hard and fast rule
`in any of the cases as to an absolute number that constitutes anticipation.
`But at a minimum, Flink discloses 4 preferred buffers but a maximum,
`that should be the starting point, not the 12. And the isotonicity agents that
`are disclosed, there's 18 but if you recall, Claim 5 specifically requires that a
`pH of 7.4 that mannitol or sodium chloride are excluded. So, mannitol's out
`and the focus, Novo makes much of this focus of Flink on mannitol and
`glycerol.
`But really, if you look at the disclosure, mannitol and glycerol are the
`polyhydric alcohols, they're the one of five disclosed. They're admitted by
`Novo the focus of Flink so arguably I think you could look at this as saying
`there's 5 times 420, really, when you drill down on this. But either way, I
`think the Wrigley case is illustrative.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`
`
`
`12
`
`
`
`IPR2020-00324
`Patent 8,114,833 B2
`
`There you had a disclosure of an optional excipient, one that wasn't
`even required and it needed to be combined with 1 of 23 different flavoring
`agents. Again, here the list is smaller, it's stated that both our required --
`our disclosure in Flink is actually, I believe, narrower than the disclosure in
`Wrigley that was held to be anticipatory.
`The sheer volume is not an issue, it only takes one disclosed
`embodiment, one disclosed species of a prior art reference to anticipate a
`broad claim. And that's exactly what we have here, Claim 14 provides the
`roadmap that gets you to the formulation of Flink with a buffer that
`preferably can be disodium phosphate and an isotonicity agent which can be
`propylene glycol. Therefore, we respectfully submit there is anticipation.
`Briefly, Patent Owner's argue that there's a lack of enablement. This
`argument seems to be specious at best. These are standard routine
`formulations that a person of ordinary skill in the art, which in this case is a
`pretty high bar, Ph.D. formulator would be able to combine and make a
`formulation.
`Patent Owner's enablement argument really goes to this issue of you
`wouldn't know if it would be stable and you wouldn't know all of its
`properties, but that's not the standard for anticipation. The standard for
`anticipation, as described in In re Gleave, is simply could a person of
`ordinary skill in the art make this formulation?
`And there's really no dispute that a person of ordinary skill in the art
`could make a formulation that contains propylene glycol as a buffer,
`particularly a commonly known buffer. And just briefly touching on Slide
`13
`
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
`
`
`IPR2020-00324
`Patent 8,114,833 B2
`
`66, the '833 patent itself acknowledges that these formulations can be made
`by conventional techniques and this is at Column 12, Line 64 to Column 13,
`Lines 29. There's simply no issue with enablement.
`Moving on to the second round, which is similar, Claims 1 to 15 are
`not only anticipated but are obvious over the Flink reference, again the
`starting point here is Claim 14. And we start with the same disclosure that
`Flink discloses everything. The one shortfall, to the extent there is any, is
`the claims don't specifically recite the specific buffer and the specific
`isotonicity agent.
`We've covered how the spec describes those and a person of ordinary
`skill in the art would understand the claim in anticipation. But we'd also
`just direct your attention to Slide 34 and this is an excerpt from Figure 7 of
`Flink that's at Page 45 of Exhibit 1004. And again, you see formulations
`here in comparing it to Claim 1 of the '833, again you see formulations here
`that meet all of the elements except for the substitution of mannitol for
`propylene glycol.
`And we've talked at length about in the briefing why a person of
`ordinary skill in the art would be motivated to switch out propylene glycol
`from mannitol. Flink describes them both as isotonicity agents of the same
`variety, of the same type, the polyol, the polyhydric alcohol. A person of
`ordinary skill in the art would also understand that PG is a commonly used
`excipient in parenteral formulation.
`Slide 39 is an indication of this, this is an excerpt from Exhibit 1127,
`the Merck Index, and Exhibit 1028, the Japanese Pharmaceutical Excipients
`14
`
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
`
`
`IPR2020-00324
`Patent 8,114,833 B2
`
`Directory both describe propylene glycol as an isotonicity agent. Similarly,
`Slide 40 discloses 6 examples from other patents about the use of propylene
`glycol as an isotonicity agent.
`It was a common agent, it was known to be a liquid, which we'll talk
`about, and the exact meaning of this and it simply would be routine to
`substitute propylene glycol for mannitol because of the disclosure of Flink
`itself coupled with the knowledge of the person of ordinary skill in the art.
`Propylene glycol was known to be safe, it was known to be used in
`Patent
`formulations at a much higher concentration than its used here.
`Owners raise the argument that it was not known as an isotonicity agent. I
`think that classically, they point to some treatises that talked about the role
`of co-solvents and the like, again that's belied by Flink itself as well as
`numerous other references and patents describing Flink as an isotonicity
`agent. Similarly, they say PG was never approved in a product as an
`isotonicity agent but that's because in the United States, typically, only the
`primary functions of an excipient are included in the labeling.
`The fact that even Dr. Tessier acknowledges every solute in solution
`contributes to this isotonicity, PG is an isotonicity agent. Patent Owner
`raised concerns with hemolysis and membrane permeability as arguments
`that would teach away from the use of propylene glycol.
`Again, this is picking and choosing from the Hammarlund article that
`simply does not support the proposition. And Hammarlund, for example,
`has shown on Slide 48, it's Exhibit 2033 and this is an excerpt from Page 6,
`everything has the potential for hemolysis. Propylene glycol was routinely
`15
`
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
`
`
`IPR2020-00324
`Patent 8,114,833 B2
`
`used in parenteral formulations and a person of ordinary skill in the art
`would know it's safe.
`Patent Owner also takes issue with the membrane permeability of this
`to suggest it could cause cell rupture but, again, membrane permeability is
`exactly what you see with glycerol, one of the preferred agents in Flink, and
`it doesn't dissuade anyone from using that.
`JUDGE FRANKLIN: Mr. Johnson, you have just under five minutes
`left and I don't want to interrupt you going through Patent Owner's
`arguments there but I would like you to touch on briefly the motivation and
`reasonable expectation of success in selecting both the DPD buffer in
`combination with PG?
`MR. JOHNSON: Sure, I think the motivation to combine the
`disodium phosphate buffer comes from Flink itself but also from the fact
`that, like Remington establishes, the phosphate buffer is one of the most
`commonly used buffers. It is a standard off-the-shelf buffer, the dihydrate
`form does not matter as both sides' experts have acknowledged.
`And propylene glycol, the motivation to select it is found in Flink
`itself as a disclosed isotonicity agent. There is nothing special about the
`way these buffers work together or how they perform that would require you
`to say there's a motivation to put this and this. Claim 14 provides the
`roadmap that requires a buffer and an isotonicity agent. Disodium
`phosphate is one of the most common buffers, there's clear motivation to use
`that.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`
`
`
`16
`
`
`
`IPR2020-00324
`Patent 8,114,833 B2
`
`Propylene glycol was known as an isotonicity agent, propylene glycol
`was known that it could be a substitute for glycerol. And because of that, a
`person of ordinary skill in the art would be motivated to use propylene
`glycol in the formulations described by Claim 14. Separately, and also to
`touch briefly on Ground 3, that motivation would also come from Betz.
`There's been some question about whether Betz is prior art, I think the
`Dynamic Drinkware case clearly establishes that a Petitioner is allowed to
`rebut an attempt to annotate a reference by either establishing an earlier
`priority date, which is what we've done here, or questioning the conception
`of reduction of practice, which we've also done here.
`All the information about the conception of reduction of practice is
`outlying in our papers and in the motions to exclude. I will not touch on
`that but I think the Dynamic Drinkware specifically rejects the argument
`made by Patent Owner here that Petitioner waived the right to rely on the
`earlier priority date.
`And then if you turn to Slide 60, Flink addressed the same issue, they
`addressed the issue of improving the stability of peptide and protein-
`containing formulations. The Patent Owner criticized Betz because it
`relates to HGH but the '833 patent actually covers both HGH and GLP
`agonist in the specification.
`This is without merit. A person of ordinary skill in the art would
`look to other exemplary formulations and Betz actually discloses that
`propylene glycol and mannitol can be switched out. And I'm looking at
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`
`
`
`17
`
`
`
`IPR2020-00324
`Patent 8,114,833 B2
`
`Slide 63, which is an excerpt from Betz, which is Exhibit 1006, and this is
`the Chart 7 that we've talked about.
`And really, the key point to this chart is not whether one is better, one
`of these formulations, the results show that all these formulations were
`stable and the results show the only difference between these eight
`formulations is that mannitol decreases and propylene glycol increases in 1
`through 7.
`8 also includes glycol. But leaving that aside, these formulations
`show that propylene glycol can provide the same stability and isotonicity
`benefit as propylene glycol. Therefore, a person of ordinary skill in the art
`would be motivated by Flink and Betz to replace mannitol with propylene
`glycol.
`JUDGE FRANKLIN: Your time has expired but you may go into
`your rebuttal time if you would like.
`MR. JOHNSON: Okay. Briefly, unless the Board has any more
`questions on obviousness, I'd just like to turn very briefly to unexpected
`results in Slide 76. I think it's clear that there's a lot of issues with the
`unexpected result, as presented in the briefing and in the motion to exclude.
`The differences between these candidates are simply differences in degrees.
`The glycerol, xylitol, sorbitol, propylene glycol were all found to have
`suitable properties as replacement candidates for mannitol. The differences
`here are very minor and are differences in degree and not in type.
`And finally, with regards to unexpected results in Slide 78, the
`unexpected results evidence is not commensurate in scope with this. The
`18
`
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
`
`
`IPR2020-00324
`Patent 8,114,833 B2
`
`PG concentration ranges are varied, the pH isn't on point, the manufacturing
`scales, and there's also a question about whether or not liraglutide was used
`in those.
`So, with that, unless there are questions I will cede the rest of my time
`to rebuttal.
`JUDGE FRANKLIN: Thank you, let me check if we have questions.
`JUDGE NEW: No questions from Judge New.
`JUDGE MITCHELL: No questions from me.
`JUDGE FRANKLIN: All right, thank you. One moment. All
`right, and for Patent Owner, Mr. Oelke?
`MR. OELKE: Thank you, Your Honor.
`JUDGE FRANKLIN: How much time you would like to reserve for
`rebuttal?
`MR. OELKE: Five minutes, Your Honor.
`JUDGE FRANKLIN: Thank you, you may begin when you're ready.
`MR. OELKE: Thank you. May it please the Board that '833 patent
`concerns propylene glycol in formulations for GOP1 compounds. And I
`think it's important to understand what led to that intent. In the early 2000s,
`GOP1 compounds were just beginning to be developed, there were none that
`had been approved yet.
`And Novo Nordisk was a leader and a pioneer in this field and it had a
`promising GOP1 compound that it wanted to get into clinical trials. And
`they knew it had to be in a formulation that could be administered to these
`patients, these diabetes patients, and that would be through subcutaneous
`19
`
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
`
`
`IPR2020-00324
`Patent 8,114,833 B2
`
`administration. But there was an issue that they had to confront
`immediately, that is, that the formulators at Novo Nordisk had to confront.
`And if you turn to Slide 3 of Patent Owner's slides, you can see GOP1
`was known to be difficult to formulate. That's because liraglutide in GOP1
`compounds tend to fibrillate, that means fibrils of the compound tended to
`agglomerate together and they led to an unclear solution as they continued to
`agglomerate.
`So, in solution they would become unclear, leading to an unstable
`formulation, basically a formulation that couldn't be used. And you can see
`here on Slide 3, this was acknowledged in the art, GOP1 is a very difficult
`molecule to handle in solution. That's Exhibit 2011. And the Flink
`reference itself talks about this in the background, that's on the right-hand
`side of this slide.
`While much attention has been focused on pharmacological properties
`of GOP1 compounds. Hitherto, little is known about their physical,
`chemical, and solution structural properties. So, the inventors started out,
`the formulators started out, with a series of work to try to figure out how to
`solve this fibrillation issue with their candidate compound.
`And that's the work that's described in the Flink reference. You see
`here on Slide 4 they did a series of experiments and that's what's described in
`Flink. So, the early formulation work was to try to come up with a
`formulation that they could put into clinical trials but they had this problem
`with fibrillation. It's a physical stability problem so they did a series of
`experiments, and those are set out in the examples of the Flink reference.
`20
`
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
`
`
`IPR2020-00324
`Patent 8,114,833 B2
`
`Mostly, they focused on controlling the pH. And if you look at
`Examples 1 through 6, what you'll see is there's a trend. As the pH was
`raised into a more basic area, it got more physically stable formulations.
`And so their conclusion, you can see this on Slide 4, is that another aspect of
`the invention is the formulation has a pH of 7.5 to 10.
`And they did try other things to experiment with to deal with physical
`stability, they varied the temperature, they varied the concentration of the
`active ingredient. But they did not try experiments with what they
`considered background excipients at that time, so that included the buffer,
`the preservative and the isotonic agent.
`They used the same isotonic agent in those earlier experiments that
`they've used all along for other products before they were dealing with the
`liraglutide, and that was mannitol and glycerol, really two standard isotonic