`571-272-7822
`
`
`
`
`IPR2016-00558, Paper No. 38
`June 20, 2017
`
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`______________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`______________
`
`LUPIN LIMITED,
`Petitioner,
`
`V.
`
`VERTEX PHARMACEUTICALS INCORPORATED,
`Patent Owner.
`______________
`
`Case IPR2016-00558
`Patent 6,436,989 B1
`______________
`
`Record of Oral Hearing
`Held: Wednesday, April 5, 2017
`
`
`BEFORE LORA M. GREEN, SHERIDAN K SNEDDEN, and
`ROBERT A. POLLOCK, Administrative Patent Judges
`
`The above-entitled matter came on for hearing on Wednesday, April 5,
`2017, commencing at 9:31 a.m. at the U.S. Patent and Trademark Office,
`600 Dulany Street, Alexandria, Virginia.
`
`
`
`
`
`
`Case IPR2016-00558
`Patent 6,436,989 B1
`
`
` A
`
` P P E A R A N C E S
`
`
`
`ON BEHALF OF THE PETITIONER:
`
`
`STEPHEN AUTEN, ESQUIRE
`JANE S. BERMAN, ESQUIRE
`TAFT, STETTINIUS & HOLLISTER, LLP
`111 East Wacker Drive
`Suite 2800
`Chicago, Illinois 606061
`(312) 527-4000
`
`
`ON BEHALF OF THE PATENT OWNER:
`
`
`LISA M. FERRI, ESQUIRE
`SCOTT A. McMURRY, ESQUIRE
`BRIAN W. NOLAN, ESQUIRE
`MAYER BROWN, LLP
`1221 Avenue of the Americas
`New York, New York 10020
`(212) 506-2340
`
`
`
`
`
`
`
`
`
`2
`
`
`
`
`Case IPR2016-00558
`Patent 6,436,989 B1
`
`
`
` P R O C E E D I N G S
` JUDGE SNEDDEN: Good morning. I'm Judge
`Snedden, and I have with me on the panel Judge Green
`and Pollock, and we are here for the final -- the
`oral hearing in IPR201600558. I'd like to start with
`appearances. We'll start with Petitioner. Please
`stand, introduce yourself, and who you have with you
`today.
` MR. AUTEN: Morning, Your Honor. I'm Steve
`Auten on behalf of Petitioner Lupin Limited.
` JUDGE SNEDDEN: Thank you. And who did you
`bring with you today?
` MR. AUTEN: I'm sorry, I didn't hear that
`part. That's my colleague, Jane Berman.
` JUDGE SNEDDEN: Okay, welcome. And Patent
`Owner?
` MS. FERRI: Good morning, Your Honor. Lisa
`Ferri from Mayer Brown, and I'm here on behalf of the
`Patent Owner, Vertex Pharmaceuticals. I have with me
`Brian Nolan, also Mayer Brown, and Scott McMurry,
`also from Mayer Brown.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`
`3
`
`
`
`
`Case IPR2016-00558
`Patent 6,436,989 B1
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
` JUDGE SNEDDEN: Okay, welcome. Okay. Each
`party, 45 minutes. We'll begin with Petitioner. You
`can reserve part of your time for rebuttal. Would
`you be reserving time today?
` MR. AUTEN: I'd like to reserve four minutes
`for rebuttal, Your Honor.
` JUDGE SNEDDEN: Four minutes?
` MR. AUTEN: Correct.
` JUDGE SNEDDEN: Okay. All right. With that,
`we'll begin. When you're ready, please stand and
`we'll begin.
` MR. AUTEN: Sure thing. May it please the
`Board, I'm Stephen Auten on behalf of Petitioner
`Lupin Limited. And we're here to talk about
`fosamprenavir and the claims of the '989 patent and
`whether it was obvious to make a prodrug of the
`amprenavir based on a few key certain facts. And
`that is, one, it would have been obvious to use the
`second most common pharmaceutical salt that had been
`used for four decades prior to the relevant date, and
`it would have been obvious to use the number one,
`most common, prodrug moiety for improving solubility
`
`4
`
`
`
`
`Case IPR2016-00558
`Patent 6,436,989 B1
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`and thus bioavailability for a prodrug of that type
`when the prior art completely predicted that such a
`drug would have the resistance profile and
`pharmacokinetic profile of amprenavir.
` So you've seen a lot of content in the papers
`dedicated to Grobelny and the dog data and that's all
`a red herring argument. That's all trying to
`distract you because what a person of ordinary skill
`in the art would have expected would have been to
`have the pharmacokinetic profile of amprenavir,
`because that's the prodrug that's being made of that
`particular drug. It would not be expected to have a
`pharmacokinetic profile of a completely different
`drug in Grobelny's patent application.
` And even if you accept that argument as true,
`we'll show you why even Grobelny would have predicted
`a successful prodrug made of amprenavir. So turning
`to our deck, starting with slide 3 -- we just,
`obviously laid the foundation for amprenavir being
`the obvious target and being in the prior art as of
`1994.
` Slide 4 actually lays the foundation for what
`
`5
`
`
`
`
`Case IPR2016-00558
`Patent 6,436,989 B1
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`is key for your decision today; and that is knowing
`in the prior art that amprenavir had the extremely
`good pharmacokinetic profile. This was known to one
`of ordinary skill in the art at the relevant time.
`And so a POSITA making a prodrug of this drug would
`expect that very same, very good, excellent
`pharmacokinetic profile of amprenavir.
` And, in fact, Vertex's experts agree with this
`premise, as he testified that the original product
`had very tight pharmacokinetics with minimal
`variability.
` So that should end the Board's query as to
`variability, because we know that an amprenavir had a
`non-variable, very tight pharmacokinetic profile, so
`a prodrug of that very same drug would have those
`exact same properties. Amprenavir is an obvious
`target because the literature calls it a lead
`compound, and not a lead compound from a new chemical
`entity type perspective you saw in the papers,
`because fosamprenavir is not a new chemical entity
`under any definition. It's a salt and ester of a
`previously approved drug. And that previously
`
`6
`
`
`
`
`Case IPR2016-00558
`Patent 6,436,989 B1
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`approved drug, of course, was amprenavir.
` So this isn't a case where you create a new
`active moiety and for obvious analysis you have to
`follow that lead compound rubric that you've seen on
`aryl pyrazoles cases and the other assorted
`authorities cited by my opponent. This is not that
`type of case. This is a salt or ester case. And it
`was known at that time that there were three other
`HIV protease inhibiters, and amprenavir was about to
`become the fourth and the lead in that class.
` So that's why it would be an obvious choice to
`start with that compound. But, as we see in slide --
`excuse me, back up to slide 5 -- you know that the
`drug is especially effective in treating HIV -- so
`there's no question from an efficacy perspective that
`the drug works. The problem we get to is in the next
`slide, slide 7. And that's known that we've got a
`solubility problem. Okay. And when it's known when
`you have a solubility problem there's low
`bioavailability.
` And how are you going to fix that? Well, you
`fix that by fixing the drug's -- you fix it by
`
`7
`
`
`
`
`Case IPR2016-00558
`Patent 6,436,989 B1
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`enhancing the drug's solubility. So we see there at
`the bottom of slide 7 how it talks about -- if you've
`got a poorly soluble drug, you're going to have poor
`bioavailability. And that's the exact problem that
`we have here with amprenavir.
` And this excerpt is taken from the first
`reference in our combination, the Roy patent, the
`'679 patent, which discloses amprenavir, discloses
`oral formulations of amprenavir, and discloses that
`prior formulation techniques didn't work. So when my
`opponent says that some of the, you know -- POSITA
`would have started with different formulation
`techniques to fix the solubility problem, well, the
`prior art already teaches us that that didn't work.
`And we'll see that in a second.
` That's it for you there on slide 8 talking
`about how standard formulation techniques for
`amprenavir didn't work. And they tried this vitamin
`ETPGS and it's shown that that's a surfactant and
`that would have been one of the ways you try to
`improve solubility of a drug. But here that didn't
`work. So what's the next step? The next step for a
`
`8
`
`
`
`
`Case IPR2016-00558
`Patent 6,436,989 B1
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`POSITA could be a prodrug. And that's a technique
`that's been known since the 1950's for improving the
`solubility and thus the bioavailability a drug. But
`Dr. Sinko, Vertex's expert, admits that this is a
`technique well known -- and known of course as of
`1997, our relevant date, that you can use phosphate
`ester prodrugs to improve solubility.
` And now as we fit this within the KSR paradigm
`we've got a finite number of progroups that we can
`use. And we put the promoiety groups up there for
`your reference: The esters, carbonates, carbamates,
`and of course you see phosphates in that group. Now,
`how do we select phosphates from that group? Well,
`that comes directly from their own expert, Dr. Sinko,
`who published this textbook -- he was the editor of
`this textbook on pharmaceutical sciences here -- and
`you see it from the excerpt that we've quoted, the
`most frequently used linkages for these prodrugs --
`and there's the -- called the -- for lack of a better
`word, "Markush group" there, the grouping of prodrug
`moieties and you see at the bottom of the page that
`the most common ones used are esters.
`
`9
`
`
`
`
`Case IPR2016-00558
`Patent 6,436,989 B1
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
` And when would you use an ester? Well, you
`would use an ester if you're working on membrane
`permeability or lipophilicity. Well, that's not our
`problem. Our problem is solubility. Well, how do we
`find an answer? We go to the next page in that book
`and, contrary to esters, here's the disclosure in his
`book: Phosphate ester prodrugs are prepared to
`enhance the solubility.
` So a POSITA knows, from that list of prodrug
`moieties that were possible, we're now narrowing to
`it to one particular type of prodrug group; and
`that's phosphate esters. And importantly, from this
`disclosure here, there are two things to take away.
`Number one, they say that you use phosphate esters to
`improve the solubility, which we covered. And number
`two, they also grouped it together with parenteral
`formulations.
` So you've seen in the paper talking about the
`Chong and the Hostetler references, and are those
`relevant? Well, they're relevant because they show
`that you can make a phosphate ester prodrug of these
`type of drugs and improve the solubility. And
`
`10
`
`
`
`
`Case IPR2016-00558
`Patent 6,436,989 B1
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`solubility --
` JUDGE POLLOCK: Are you suggesting that one of
`ordinary skill in the arts seeking to improve
`solubility would ignore lipophilicity and absorption?
` MR. AUTEN: No, it's certainly an important
`parameter here. But I'm saying if the target
`parameter you're trying to prove is solubility you're
`going with phosphate as the prodrug. If you want to
`improve lipophilicity, then be improving
`permeability, you'll go with an ester.
` So importantly here, Dr. Sinko admits that
`this premise was generally known in 1997 at the
`relevant date. And, from this take away here, the
`last sentence in the highlighted box talking about
`how these drugs are readily converted by phosphates
`present at the intestinal brush border. That's
`important because one of ordinary skill knows that
`when you ingest these phosphate drugs you're going to
`get conversion. You're going to get conversion. And
`when you have that conversion then you've got a
`bioavailable drug. You have none of this variability
`that they talk about with the Grobelny dog data. So
`
`11
`
`
`
`
`Case IPR2016-00558
`Patent 6,436,989 B1
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`we recite Grobelny because you need to know where in
`the backbone to make that particular modification.
`But they know because of that conversion, because of
`the phosphate -- phosphatases in the system here,
`that's going to convert the drug.
` So we know that as a matter course without
`relying on what is the pharmacokinetic data of
`Grobelny's prodrug. We know it's going have the
`pharmacokinetic profile of amprenavir, which we know
`is a very good pharmacokinetic profile.
` JUDGE SNEDDEN: And do we have evidence that
`this conversion is not variable?
` MR. AUTEN: We do have evidence that this
`conversion is not variable.
` JUDGE SNEDDEN: Where is that?
` MR. AUTEN: Well, we have it in two places.
`We have it in the Paredes reference, which is talking
`about the fast conversion of fosamprenavir to
`amprenavir. But we also know it -- right here in the
`prior art from this reference, and from the testimony
`of Dr. Sinko, that because of these phosphate esters
`-- these phosphatases, there's rapid conversion upon
`
`12
`
`
`
`
`Case IPR2016-00558
`Patent 6,436,989 B1
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`ingestion. We know that if you make this type of
`prodrug, you're going to have that rapid conversion.
`It's going to happen as a matter of course.
` And when you do that, the POSITA expects the
`very same pharmacokinetic profile of the parent drug
`and the same resistance profile of the parent drug
`and all those other in vivo attributes of that parent
`drug, or that prodrug, as compared to a completely
`drug in Grobelny.
` So here's Grobelny, the roadmap -- we called
`it "universe of one." It is the only oral HIV
`protease inhibitor in the prior art. There are other
`HIV protease from the prior art. This is the only
`oral one and it shows a roadmap that -- for the
`selection of the prodrug moiety -- so now we take it
`from the phosphate ester group we're going to show
`now it's the number one choice to make. And that's
`the choice that the applicants from the '989 patent
`followed.
` So here's the Grobelny roadmap, slide 14,
`showing bits of the prior art -- and then Grobelny
`teaches the very same problem that we have in the Roy
`
`13
`
`
`
`
`Case IPR2016-00558
`Patent 6,436,989 B1
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`patent. It teaches that if you have low solubility,
`you're going to have low bioavailability. And how do
`you fix that? Well, turning to slide 16 we see that
`you fix that with a functional group, a prodrug group
`-- or promoiety group. And which one do you pick?
`It's no surprise we go to slide 17 the two top
`embodiments that they pick are the phosphate and the
`phosphite.
` And so we know those are the top choices --
`those are the exact ones they made on the '989
`patent. And so to hear my colleague talk -- and she
`talks on her slide five about the Chaturvedi
`declaration and the number of compounds they had to
`make and only 4 of 24 compounds worked. Well, guess
`which four worked? The phosphate and the phosphite.
`The very same compound that Grobelny tells you will
`work.
` So as we're laying the foundation for things
`two and three, we're focused on solubility rather
`than the pharmacokinetic parameters. And what we see
`here is is the customary location for making that
`change on the prodrug as we see in slide 18 and then
`
`14
`
`
`
`
`Case IPR2016-00558
`Patent 6,436,989 B1
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`that dovetails with slide 19 showing the prior art.
` Now, Chong here is another HIV protease
`inhibitor on the prior art, but it shows, when you
`have this backbone, where do you make the change? At
`the hydroxyl location on that backbone. The same
`place that they made it in Grobelny. And I might add
`when you go back as to the chart, where they do the
`declaration where they have their 24 compounds, all
`but two made the change at the central hydroxyl. So
`a POSITA knows where to make this change in order to
`make the soluble drug.
` So turning back to slide 20, we've got the
`examples from the Grobelny application on where
`they're going to make that change on the central
`hydroxyl for that particular compound. And on slide
`21 we show what is the impact of that change. The
`impact is to improve the solubility -- and not just
`by a little bit, but by tremendously -- by orders of
`magnitude you see here that we've got it now on the
`order of milligrams per milliliter. Whereas in the
`prior art -- and it's in Vertex's deck, you'll see
`it, that at the time the solubility of those drugs
`
`15
`
`
`
`
`Case IPR2016-00558
`Patent 6,436,989 B1
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`was in the micrograms per milliliter. So a
`difference of a thousand or three orders of magnitude
`difference.
` JUDGE SNEDDEN: Much of the argument so far is
`focused on solubility. And, I mean, you know, it
`seems to support solubility. But how do we know that
`when you make these modifications, we make this
`prodrug, that you're actually going to get something
`that's bioavailable that's going to treat HIV?
` MR. AUTEN: Excellent question. And we know
`that from what we excerpted from Dr. Sinko's book
`that when you have this -- we know the conversion is
`going to happen. And when you know when you have
`that conversion, as we're going to hear from
`Dr. Ogden in the second in the deck -- when you have
`that conversion, you expect the very same
`pharmacokinetic parameters that you have off the
`parent drug.
` JUDGE SNEDDEN: It's my understanding that
`Patent Owner's argument is that Grobelny -- they show
`us a drug and the pathway that you would choose, that
`when they actually go through this process,
`
`16
`
`
`
`
`Case IPR2016-00558
`Patent 6,436,989 B1
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`bioavailability is what's unpredictable --
`treatment -- though unpredictable -- and, in fact,
`this drug wasn't successful.
` MR. AUTEN: But that's why it's a red herring
`argument. Because they're saying to make a prodrug
`of amprenavir you would expect the PK data of
`Grobelny's drug and that's not true. If you're going
`to make a prodrug of amprenavir, you would expect the
`pharmacokinetics of amprenavir, which we know to be
`very tight and excellent and good.
` So they're trying to distract and make this
`red herring argument that if you're going to make a
`prodrug of amprenavir you'd get the PK profile of
`Grobelny. And, in fact, they suit the -- flip the
`argument to suit their needs. Because they say that
`you'd have the PK profile of Grobelny, but from the
`resistance profile perspective they compare that to
`amprenavir. So they're changing their argument to
`suit their needs.
` JUDGE SNEDDEN: Right. I believe what we're
`-- there's a difference between going through the
`predictability of making something soluble and
`
`17
`
`
`
`
`Case IPR2016-00558
`Patent 6,436,989 B1
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`evidence to show predictability of something that's
`going to be bioavailable and thus useful in treating
`HIV.
` MR. AUTEN: Right. I'm aware -- that's where
`we're going to get you right now. So we're seeing
`even if you buy their argument, I think we still win
`because I think it is predictable based on, most
`notably, figure 2 of the Grobelny patent application
`showing rapid conversion. And as Dr. Ogden
`testified, once you have that rapid conversion you
`expect the pharmacokinetics of bioavailability and
`all those excellent in vivo parameters of that parent
`drug, which, in this case, is amprenavir.
` So turning to claim 2, this is just a
`selection of -- claim two cites it's various
`phosphate salts of amprenavir. And then, of course,
`no surprise they picked the disodium which is the
`same phosphate salt disclosed in Grobelny. And we've
`put it up there on now slide 25 to show them
`juxtaposed to each other -- how it's the same
`modification for the same disodium salt -- to improve
`the solubility. And as the board noted in the
`
`18
`
`
`
`
`Case IPR2016-00558
`Patent 6,436,989 B1
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`institution decision for claims 2 and 3, there's no
`bioavailability requirement. So in here, all we have
`to show is the expectation of improving solubility.
`As I mentioned a little bit ago, that improvement was
`three orders of magnitude. So when you have that,
`then of course -- turning to slide 27 you see --
` JUDGE GREEN: Can we really tease the
`solubility of compound claims from the
`bioavailability of the method claims? Because the
`sole reason to make this is as a HIV protease an
`inhibitor; right?
` MR. AUTEN: I'm having a hard time hearing
`you, Your Honor.
` JUDGE GREEN: I'm sorry. Can we really tease
`apart the solubility argument as to the compound
`claims and the bioavailability argument as to the
`method claims? Because the sole reason to make the
`compound is to use it as a therapeutic; correct?
` MR. AUTEN: That's correct, Your Honor, that's
`correct. But, for the way they've drafted the claims
`there is no bioavailability requirement in claims 2
`and 3.
`
`19
`
`
`
`
`Case IPR2016-00558
`Patent 6,436,989 B1
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
` JUDGE GREEN: No, I understand that. But the
`reason that the art suggests for making this is
`bioavailability, so it does -- it still goes to the
`reasonable expectation of success provided by the
`prior art.
` MR. AUTEN: Right. I agree with that. Yes,
`absolutely. So -- but as mentioned, we do have the
`reasonableness expectation of success for solubility
`as to claim 2. And then for claim 3, we're just
`talking about the calcium salt -- which I mentioned
`at the opening, is the second most common salt over
`the last four decades. So now we're on slide 30, and
`claim 3 is obvious for two reasons. Number one,
`calcium salt is disclosed in Grobelny, as we
`highlight here on slide 30, and then slide 31 merely
`mentions it's only on solubility. Slide 32, that's
`when we talk about the Bighley reference, and you see
`in the right-hand column there the two most common
`salts over the last 40 years for the calcium salt and
`the sodium salt.
` So there was no -- there was nothing inventive
`about selecting the calcium salt. And that's why we
`
`20
`
`
`
`
`Case IPR2016-00558
`Patent 6,436,989 B1
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`think this case falls for more from the rubric of the
`findings of the Apotex case we gave you where the
`Federal Circuit found and endowed a salt patent
`because it was only routine testing to get 1 of 53
`pharmaceutically acceptable salts and that one was
`used less than one percent of the time.
` Here we have use that's an order of magnitude
`greater, so there's nothing inventive about selecting
`that salt that had been used for the last 40 years.
` So here's where the rubber meets the roads --
`we're talking about the pharmacokinetics. Claims 10
`to 12, we put them up there for you. They add a few
`different limitations that we'll need to cover that
`carry our burden and that's mainly talking about the
`carrier -- the adjuvants here and then the
`administration to a mammal and then we'll get to the
`expectation of success in the bioavailability -- just
`want to make sure we're covered here.
` Amprenavir is known to be effective. We have
`that and the prior art and the Roy reference as to
`the effective amount. That's disclosed in the Roy
`patent as well. And then Grobelny teaches carriers,
`
`21
`
`
`
`
`Case IPR2016-00558
`Patent 6,436,989 B1
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`adjuvants, and other excipients that'd be quite
`typical in the pharmaceutical sciences. So there's
`nothing inventive as those particular limitations in
`claim 10.
` This is a very busy slide but the only take
`away for claim 17 is to know that Grobelny taught
`administration to mammals because that's a limitation
`in the claim as well. And then claims 11 and 12
`teach these combinations and so we have on the slide
`here the language of the claims 11 and 12 next to the
`disclosure in Roy -- disclosed in these very
`combinations of the Markush group.
` So I don't think there's any dispute among the
`parties as to what's the scope of the disclosure of
`Roy from this perspective as to the combination. The
`dispute is going to be over the expectation of
`success, which is what I'll cover right now.
` So, again, before we go through this road on
`the argument of the expectation of success -- I state
`again this is only if you buy their argument that if
`you make a prodrug of amprenavir you'd expect the PK
`profile of a completely different drug, which we
`
`22
`
`
`
`
`Case IPR2016-00558
`Patent 6,436,989 B1
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`think is the wrong premise. You would expect the PK
`profile of amprenavir --
` JUDGE SNEDDEN: I don't think I understand
`their argument that way.
` MR. AUTEN: I'm sorry?
` JUDGE SNEDDEN: I'm not sure I understand
`their argument that way -- the way you just
`presented. The way I understand their argument is
`that what you have is a drug in Grobelny and that
`adding -- making a prodrug of that drug would then
`insert unpredictability. And that just speaks to
`unpredictability of the field of prodrugs. Not so
`much as -- am I misunderstanding that right?
` MR. AUTEN: No, I don't think you're
`misunderstanding, but I think the missing link in
`that argument is that Grobelny is silent as the
`pharmacokinetics of the parent drug. We don't have
`that comparison of parent to prodrug. We don't have
`that. What we do have, though, is what we know here
`instead is the PK profile of amprenavir. We knew of
`that to be very good. We know that to be very tight.
`We know that had minimal variability.
`
`23
`
`
`
`
`Case IPR2016-00558
`Patent 6,436,989 B1
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
` We don't know that about Grobelny's parent
`drug, which is why that may change the expectation as
`to Grobelny's prodrug. But for amprenavir's prodrug
`it's very different. The expectation is to have that
`very tight, minimal variable pharmacokinetic profile.
` JUDGE SNEDDEN: Wasn't the PK profile of
`fosamprenavir different from amprenavir?
` MR. AUTEN: It's slightly different, but we
`submit that's a distinction without difference.
`That's a difference of degree not in and of kind.
`And in fact, as the witness has testified, there's no
`difference of efficacy of those drugs. And we'll
`talk about the other side effect profiles and
`resistant profiles and the words that used in the art
`are superimposable, identical, comparable,
`equivalent...
` JUDGE SNEDDEN: Right. But it seems like the
`differences is that Patent Owner thinks matters are
`-- that change in the PK data is significant enough
`that it leads to some improvement in the adverse
`effect profile and also in the mutational resistance
`profile.
`
`24
`
`
`
`
`Case IPR2016-00558
`Patent 6,436,989 B1
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
` MR. AUTEN: But here's the thing, Your Honor,
`they think that -- you just said yourself they think
`that. It's not been proven. If you look at the Wood
`reference, they use conditional terms that there
`"might" be improved antiviral activity or there
`"might" be improved durability. That's never been
`proven. And as we know, conclusory statements that
`are unsupported are not evidence of secondary
`considerations.
` JUDGE SNEDDEN: I do recall some evidence
`submitted in Exhibit 2056, the Chapman paper for
`example on Ross -- is another one where there's a
`statement in there that says you don't get the --
`what is it -- the IV50?
` MR. AUTEN: Right. That's a mutation, right.
`The I50V.
` JUDGE SNEDDEN: That doesn't appear in
`patients treated with unboosted --
` MR. AUTEN: That's really commonly unknown.
`That's really commonly unknown. Because that's where
`the literature is unclear because we have the Paredes
`references that called the resistance profiles
`
`25
`
`
`
`
`Case IPR2016-00558
`Patent 6,436,989 B1
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`"superimposable" and then we have other references
`that say that fosamprenavir doesn't select for I50V.
` But as we argue in our papers, that's a
`distinction without difference. And that's because
`-- for a couple of reasons. Number one, both
`amprenavir and fosamprenavir exhibit high resistance.
`So it doesn't make it any better. Number two, every
`HIV protease inhibitor is going to exhibit a
`resistance. Number three, the I50V alone mutation
`exhibits only the low level resistance two and
`threefold which, as their expert testified, even a
`increase of that three to fourfold would be an
`acceptable level of resistance for an HIV protease
`inhibitor. And in any case you need more mutations,
`up to ten times -- and as Dr. Ogden stated in his
`book -- you'd have to have multiple mutations to have
`tenfold resistance for fosamprenavir. So the fact
`that it may or may not search for I50V, it's a
`distinction without difference.
` JUDGE SNEDDEN: Can we actually go through
`that evidence? So let's go through the evidence and
`discussion -- have a discussion about the that
`
`26
`
`
`
`
`Case IPR2016-00558
`Patent 6,436,989 B1
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`difference between the Cmin data, the PKPD profile of
`fosamprenavir, and amprenavir and why that difference
`doesn't matter.
` MR. AUTEN: Sure. So from the PK perspective,
`here's Wood talking about how the equivalent steady
`state plasma levels -- and how there's no unexpected
`results from that pharmacokinetic perspective. And
`it further talks in Wood, as I mentioned a moment
`ago, on how it -- it's not a proven premise, that it
`may have an impact on antiviral activity or
`durability. That's not been proven. And then we
`have the Paredes reference here talking about the PK
`profiles of fosamprenavir and amprenavir being
`identical after administration.
` So the slight differences that's noted in some
`of the references, from our perspective that's a
`difference in degree, not in kind, and it doesn't
`impart patentability of the claims. Because from an
`overall perspective, the industry and the art, they
`see them identical, superimposable, and equivalent.
` As to the resistance profile, you wanted to
`chat about that, so as I mentioned a moment ago,
`
`27
`
`
`
`
`Case IPR2016-00558
`Patent 6,436,989 B1
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`Paredes calls them superimposable; FDA would've
`reviewed it, found the resistance pattern to be
`expected and comparable between fosamprenavir and
`amprenavir. And that's after -- if we take the FDA
`paper -- I know there's the motion to strike pending
`before the Board -- that's where Vertex had a
`conversation with the FDA about what the profile
`should be, so that's a little bit difference than the
`other references we have of record that other folks
`reporting it in the art.
` This is where we have Vertex or GSK telling
`the FDA certain things, and this is the FDA's view of
`that same information, of GSK's, Vertex's
`information?
` JUDGE SNEDDEN: What does that tell us?
` MR. AUTEN: It tells us -- it's another data
`point that's expected and comparable. I think it's
`more relevant because they made that finding after
`having that conversation about the underlying data
`rather than just merely observing the data in a
`report issued by GSK or Vertex. This is an ongoing
`-- it'd be like looking at the file history of a
`
`28
`
`
`
`
`Case IPR2016-00558
`Patent 6,436,989 B1
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`