`571-272-7822
`
` Paper 32
`Entered: June 29, 2021
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`____________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`____________
`
`SYNGENTA CROP PROTECTION AG,
`Petitioner,
`
`v.
`
`FMC CORPORATION,
`Patent Owner.
`____________
`
`PGR2020-00028
`Patent 10,294,202 B2
`____________
`
`Record of Oral Hearing
`Held: June 16, 2021
`____________
`
`Before SUSAN L. C. MITCHELL, ZHENYU YANG, and
`CYNTHIA M. HARDMAN, Administrative Patent Judges.
`
`
`
`
`
`PGR2020-00028
`Patent 10,294,202 B2
`
`APPEARANCES:
`
`ON BEHALF OF THE APPELLANT:
`
`
`MICHAEL E. ANDERSON, ESQUIRE
`TONI-JUNELL HERBERT, ESQUIRE
`DAVID KLECYNGIER, ESQUIRE
`FABIAN KOENIGBAUER, ESQUIRE
`Baker Hostetler LLP
`1050 Connecticut Avenue, N.W.
`Washington, D.C. 20036
`
`
`
`
`ON BEHALF OF PATENT OWNER:
`
`
`KATHRYN GREY, ESQUIRE
`SUSAN MORRISON, ESQUIRE
`DOROTHY P. WHELAN, ESQUIRE
`Fish & Richardson P.C.
`222 Delaware Avenue
`17th Floor, P.O. Box 1114
`Wilmington, Delaware 19801
`
`EDWARD L. BRANT
`ROBERTE M.D. MAKOWSKI
`FMC Corporation
`2929 Walnut Street
`Philadelphia, Pennsylvania 19104
`
`
`
`
`The above-entitled matter came on for hearing on Wednesday, June
`16, 2021, commencing at 1:00 p.m. EDT, by Webex.
`
`
`
`
`2
`
`
`
`PGR2020-00028
`Patent 10,294,202 B2
`
`
`P R O C E E D I N G S
`- - - - -
`JUDGE HARDMAN: Good afternoon, everyone. I am Judge
`Hardman, and with me are Judges Mitchell and Yang. This is the final oral
`hearing in PGR2020-00028. The Petitioner is Syngenta Crop Protection
`AG. The Patent Owner is FMC Corporation. We have a court reporter
`present on the line, and the Board will issue a transcript for this hearing,
`which will be made part of the record in due course. Let’s begin by taking
`appearances. Counsel for Petitioner, please introduce yourself and your
`team and identify who will be speaking today.
`MR. ANDERSON: Good afternoon, Your Honor. This is Michael
`Anderson, from Baker Hostetler. I’m here on behalf of Petitioner Syngenta
`Crop Protection AG. And with me today are my colleagues, Toni-Junell
`Herbert and Fabian Koenigbauer. I will be speaking during the presentation.
`JUDGE HARDMAN: Thank you, Mr. Anderson. And, Counsel for
`Patent Owner, please introduce yourself and your team and let us know who
`will be speaking.
`MS. MORRISON: Good afternoon, Your Honor. This is Susan
`Morrison, from Fish and Richardson, on behalf of Patent Owner FMC
`Corporation. With me today are my colleagues, Dorothy Whelan and
`Deanna Reichel. And I will be speaking on behalf of Patent Owner today.
`JUDGE HARDMAN: Thank you, Counsel. And thank you all for
`your flexibility, in participating in this all-video hearing. And as a reminder,
`if you encounter technical difficulties during the hearing, please immediately
`speak up or contact the team member who provided you with the connection
`information. And if you come to a good-faith belief that the pace of this
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`
`
`3
`
`
`
`PGR2020-00028
`Patent 10,294,202 B2
`
`hearing is preventing you from adequately representing your client, please
`also let us know. We’ll consider some expansion of the allotted time. Now,
`we understand that Petitioner has filed two Motions to Exclude, and we do
`not expect to rule on those Motions today, but the parties are welcome, but
`not obligated, to argue those Motions, during your allotted time. And we
`will address them in our final written decision.
`Now, each side has 60 minutes to present its case, and we will do our
`best to keep track of time, but we also suggest that you and your colleagues
`do the same. Petitioner will present its arguments first. And, Mr. Anderson,
`would you like to reserve any time for rebuttal?
`MR. ANDERSON: Yes, Your Honor. I’d like to reserve 10 minutes,
`if I could?
`JUDGE HARDMAN: And, Counsel for Patent Owner, you may have
`the last word today, if you wish. Would you like to reserve any of your
`time?
`
`MS. MORRISON: Yes, I’d like to reserve 10 minutes, as well.
`Thank you.
`JUDGE HARDMAN: Okay. As a reminder, please identify yourself
`each time you speak and mute your microphone when not speaking. The
`panel has access to the entire record, including your demonstratives. And
`please refer to each demonstrative paper or exhibit by the appropriate
`designation, and just give us a second or two to call up the paper or exhibit
`that you’re referring to. And our expectation is that, unless absolutely
`necessary, counsel for the opposing party will not interrupt the other party’s
`presentation. So, with that, we are ready to begin. If you bear with me, I
`will set the clock for 50 minutes. And, Mr. Anderson, you may begin.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`
`
`4
`
`
`
`PGR2020-00028
`Patent 10,294,202 B2
`
`
`MR. ANDERSON: Thank you, Your Honor. Again, this is Michael
`Anderson. Good afternoon. May it please the Board, I am here speaking
`today on behalf of the Petitioner, Syngenta Crop Protection AG. We have
`prepared some demonstrative slides of -- can that be -- if the Board is
`willing, I can put those on the screen, as a share. I will do that now. So,
`those should be on the screen. Please let me know if they are not.
`JUDGE HARDMAN: We do see them. Thank you.
`MR. ANDERSON: Okay. So, today, we have organized the -- our
`argument into four pieces. The first piece is going to be the lack of
`enablement argument. Then, we’ll move to the insufficiency of the written
`description. Then, we’ll move to the Petitioner’s arguments, related to Dr.
`Hunt’s patents, which are our first Motion to Exclude. And, finally, we will
`briefly touch on the Motion to Exclude certain opinions from the Patent
`Owner’s expert, Dr. Dayan.
`We’ll begin with the lack of enablement arguments, which was
`ground 1 to the petition, and that pertains to claims 1 through 3, 9 through
`13, and 21 through 30. Now, Section 112 of the Patent Act makes it clear
`that the specification of the patent must make -- it must enable a person of
`ordinary skill in the art to make and use the claimed invention. Well, that
`general principle is undisputed here. The parties have spent a lot of time
`debating how this requirement is actually applied in practice. And the focus
`of my discussion today is going to be on the use aspect of the make and use
`prong of Section 112.
`I think the first issue that’s really the subject of dispute is exactly what
`must be enabled. We think it’s important to note that it’s not sufficient
`under the law to merely enable a few embodiments. The patent must enable
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`
`
`5
`
`
`
`PGR2020-00028
`Patent 10,294,202 B2
`
`a person of ordinary skill in the art to make and use the full scope of the
`claims without undue experimentation. And as we see in the Trustees v.
`Boston University case, the Federal Circuit has held that, even where you
`have no dispute that five out of six of the embodiments were enabled, the
`patent was actually still found to be invalid for lack of enablement. As the
`Trustees’ court explained, if the Patent Owner wants to exclude others, its
`patent needs to teach the public how to make and use the full scope of that
`invention. That’s part of the quid pro quo of the entire patent bargain.
`And as we see in the MagSil Corp., as the MagSil Corp. further
`explained, enablement prevents Patentees from claiming more than they
`actually invented. So, a Patentee, who seeks broad claim language, puts
`itself at peril of losing any claim that it cannot fully enable, across the full
`scope of the patent. So, essentially, the scope of the patent and the
`specification must be less than or equal to the scope of the enablement,
`excuse me, the scope of the claims must be less than or equal to the scope of
`the enablement.
`Now, in 2021, moving to the next slide, in Amgen v. Sanofi, the
`Federal Circuit, again, reiterated that the specifications disclosure must be at
`least commensurate for the scope of the claims. Now, the court said you
`don’t need to describe how to make and use every possible variant, but there
`must be reasonable enablement of the entire scope of the range claim. And
`the court specifically noted that it’s appropriate to look out -- look at the
`amount of effort to obtain embodiments outside of the scope of the disclosed
`examples and guidelines. And the Sanofi court concluded that, in that
`particular case, given the breadth of the claims and the unpredictability in
`the art, no reasonable factfinder could conclude that there was an adequate
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`
`
`6
`
`
`
`PGR2020-00028
`Patent 10,294,202 B2
`
`guidance beyond a very narrow working examples of the patent. That’s
`exactly the same case that we have here.
`We also believe that the Idenix case, which is also noted on this slide,
`is very similar, as well. There, the court noted that a specification that
`requires a person of ordinary skill in the art to engage in iterative trial and
`error process is not enabling. The Idenix court specifically said it’s not
`sufficient to merely identify a starting point for future research. In citing to
`the Wyeth & Cordis Labs case, which is also in the record here, that
`involves screening tens of thousands of compounds for activity. The Idenix
`court stated that, notwithstanding that screening compounds for
`effectiveness might be routine, the fact that you must screen many, many
`thousands, which must first be synthesized, is -- was undue. The Idenix, like
`the Wyeth case, specifically noted that where practicing the full scope would
`have required excessive experimentation, even if you were using routine
`procedures, a patent is invalid for lack of enablement.
`I’m moving to --
`JUDGE HARDMAN: Counsel, can I just interrupt you for a second?
`MR. ANDERSON: Yes, Your Honor.
`JUDGE HARDMAN: Regarding routine procedures, Patent Owner
`has identified modeling techniques, such as QSAR, Q-S-A-R. Do you agree
`that that is a routine technique that’s used in this field?
`MR. ANDERSON: I would agree that it is a technique that is used in
`the field. It is not necessarily the predominant, but it is a technique that has
`been used. We would certainly disagree, regarding the scope and the utility
`of modeling and what it can be used to do. So, for example, we believe that
`there’s a very big difference between a global QSAR, which would model
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`
`
`7
`
`
`
`PGR2020-00028
`Patent 10,294,202 B2
`
`the entire scope of a -- of the claim here, all 10 to the 32nd compounds,
`versus a local QSAR, which would create a model that is nearly covering the
`small scope of the compounds identified. So, it is potentially a routine, or it
`is something that can be done, but it is certainly not a cure all for what is
`missing in this particular case.
`JUDGE HARDMAN: Thank you.
`MR. ANDERSON: Sure. Moving to the next slide, pardon me, I
`think -- there we go. Now, FMC has argued that, in this particular case, it’s
`not like Sanofi or Idenix because those cases involve functional claim
`language. But the Federal Circuit’s precedent, including the In re Surrey
`case and the In re Fouche cases, that we’ve cited here, make it clear that the
`requirement that the use be enabled applies equally to claims that are
`directed to compounds or structures. And In re Surrey, in particular, I think,
`is very analogous to the examined case here. It involves compound claims
`that were described in the specification as having the potential
`anticonvulsant properties.
`And the Federal Circuit specifically noted that, manifestly, a
`disclosure which does not adequately establish compounds as useful for an
`assertive purpose does not adequately describe how to use those compounds
`either. It went on to explain that the possible existence of compounds,
`falling within the scope of the claims but not having the utilities set forth in
`the Appellant’s specification, would properly be considered in connection
`with a rejection on undue breadth. So.
`JUDGE HARDMAN: Counsel, if I could interrupt you there. I’m
`sorry. I believe Patent Owner made an argument that there’s a different
`enablement standard for the compound claims, such as -- well, I won’t
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`
`
`8
`
`
`
`PGR2020-00028
`Patent 10,294,202 B2
`
`specify the particular claims because I don’t want to get them wrong. But
`there are challenged claims that are directed to just compounds and other
`challenged claims that also have recitation regarding uses in herbicide or
`herbicidal activity. And I believe Patent Owner argued that there’s really
`two different standards there, and that for the compound claims, if this
`specification discloses a practical utility, which I think this specification here
`does, that you really have to sort of take that at face value and that, alone,
`satisfies the enablement requirement. Did Petitioner respond to that
`particular argument?
`MR. ANDERSON: Well, I think that particular argument is actually
`undermined directly by -- I mean, I think the position that we have taken is
`that there is no double-standard. It is the same standard. And Surrey, for
`example, it specifically addresses that issue. Surrey says an applicant is not
`entitled to a claim for a large group of compounds, merely on the basis of
`showing that a selected few are useful and a general suggestion of similar
`utility and others. That’s precisely the case here. I don’t believe there’s two
`standards. There may be a slight difference in the amount of
`experimentation that is required, when you have purely functional language.
`But I think it’s a question of degree, not a question of a different standard. I
`think the Surrey case just clearly comes out and says it is not sufficient to
`simply state a general utility.
`And that’s similar to the In re Fouche case, as well, where, you know,
`you had a case where there was a compound claim using Markush groups,
`which are actually much smaller Markush groups than at issue here. And
`the court specifically addressed a rejection based on lack of enablement,
`regarding how to use the claimed compounds. And it said, when Applicant
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`
`
`9
`
`
`
`PGR2020-00028
`Patent 10,294,202 B2
`
`undertakes to define its invention by the recitation of a Markush group, he
`must enable one skilled in the art to make and use at least one composition,
`employing each member of the group. So, both of these cases were cited in
`our reply brief, as evidence that the standard is essentially the same, across
`all of the types of compounds. I would also note that Surrey specifically
`noted that, where a claim is directed to a group of compounds, there should
`be reasonable assurances that all or substantially all of them are useful. And
`in Surrey, as in here, the court specifically noted that’s particularly true
`where the claims are directed to a more or less artificial selection of
`compounds, which is essentially what we have here. Does that answer Your
`Honor’s question?
`JUDGE HARDMAN: Yes, thank you.
`MR. ANDERSON: If I could move to the next slide. The court is
`familiar with the Wands factors. So, I’m not going to go into those. But
`they have become the standard measure for assessing enablement, and
`whether the level of experimentation required to enable a particular claim is
`undue. I’m actually going to begin with the eighth factor here, which is the
`breadth of the claims because, in our view, the amount of guidance and the
`presence of examples all need to be viewed in the context of the breadth of
`the claims.
`So, we’ll start, of course, where all the claims -- the challenged claims
`relate to claim 1. They all depend on claim 1. And claim 1, here, spans
`more than three columns of the patent, from lines 285 to 287. It concludes
`compounds having a particular correlated known core, and there are 10
`possible positions that may be substituted off of that core. You see Q1, Q2,
`R1 through R6, and then Y1 and Y2. Our enablement argument, in
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`
`
`10
`
`
`
`PGR2020-00028
`Patent 10,294,202 B2
`
`particular, focuses on four of those groups, the R6, R1, Q1, and Q2, which
`we’ve circled in the slide, slide number 7.
`As we discussed in our briefing, and as Dr. Hunt detailed in his
`declaration, the large number of positions and the large number of
`substituents that are available at each position, and in particular at Q1, Q2,
`R1, and R6, mean that, even under a very conservative estimate, there are at
`least 10 to the 32nd compounds, 4x10 to the 32nd compounds, covered by
`claim 1. And the only uses described in the patent for any of these
`compounds is as an herbicide or as a plant growth regulator. And to the
`extent this is a reference to a plant growth regulator, the only data actually
`relates to herbicidal activity, uses herbicide. So, even if there’s a reference
`to a plant growth regulator, there’s actually no data supporting that.
`So, looking just first at the Q1 position, we can see that the claimed
`description of just one position occupies almost an entire column of space.
`On the left hand of this slide, Q1, itself, is defined, and it starts with a
`phenyl, which is where most of the testing in the 202 patent focus is, but it
`also includes five to six member heterocyclic ring systems, which can have
`up to four heteroatoms or an eight to 10 membered heteroaromatic ring
`system, which, again, can have up to four heteroaromatic atoms. And Dr.
`Hunt explained that there are over 1,000 possible heterocyclic ring systems
`that would fall in the scope of Q1, alone, without even considering R7, and
`R10, and R9.
`So, Q2, and, of course, R7 and R9, in the context of Q1, each are very
`large groups, and there may be up to five possible independently substituted
`group, outside of the ring. Depending on whether it’s a carbon or a nitrogen
`atom, you’ll either take from R7 or R9. The Q2 position is of a similar
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`
`
`11
`
`
`
`PGR2020-00028
`Patent 10,294,202 B2
`
`scope to Q1. It starts with a phenyl or napthalenal ring system. But it also
`includes five to six member hetero -- fully unsaturated heteroaromatic rings,
`in eight to 10 membered heteroaromatic bicyclic ring systems. And, again,
`each ring or ring system can have up to five independently selected
`substituents, from R10 and R11, which are actually the same as the R7 and
`R9 groups, shown up there respectively.
`So, the summary, you -- both the Q1 and Q2 groups, alone, each
`include thousands of substituents, and the vast majority of these substituents
`are heterocyclic rings or ring systems. And that thousand is without even
`considering the possible substitutions at each of the five positions off of
`those ring systems. Now, there may be up to five independent substitutions
`on each. As we said, R7 and R9 are the possible substitutions off of carbon
`and nitrogen atoms, for Q1, and that R10 and R11 are the possible
`substitutions off of carbon and nitrogen in the Q2 position. So, R7 and R10,
`they include 44 groups of substituents. R9 and R11 include 24 groups of
`substituents. So, this leads to a huge number of possible variations at the Q1
`and the Q2 position.
`Moving to the next slide, as Dr. Hunt explained, even if you only look
`at the phenyl group, within Q1 and Q2, so, you’re only looking at the one
`phenyl compound, and you’re not considering all the possible heterocycles,
`there would still be at least 167 million possibilities, at both of the Q1 and
`Q2 positions. Again, that’s without even considering possible heterocycles.
`JUDGE HARDMAN: Counsel, beyond just the sheer numbers here
`of choices, can you talk about the different properties that appear within
`these substituents, within the claim group, and how the properties differ
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`
`
`12
`
`
`
`PGR2020-00028
`Patent 10,294,202 B2
`
`across the full scope of the claim group, with respect to how that would
`impact activity?
`MR. ANDERSON: Sure. So, Dr. Hunt actually addressed this a little
`bit in his declaration, and we think, also, that the data, itself, goes to this
`issue, as well. So, when you look at the Q1 position -- let me see here, and
`I’ll just fast forward, if Your Honor doesn’t mind. As we said, there are
`thousands of possible positions and heterocycles. And Dr. Hunt explained
`that only 36 of those tested compounds had a Q1 that was not a phenyl, and
`so, when we start seeing how the heterocycles are actually inputted there, we
`can see that there’s a tremendous variation in -- or there’s a significant
`variation in the activity.
`So, there are only 36 compounds that have a non-phenyl or a
`heterocyclic Q1. Four of those, the pyrazoles, have no activity, at all. Three
`other pyrazoles have very little activity, even at the highest screening level
`configurations. And so, we also didn’t test the vast bulk of the R7 and R9
`configurations, which were tested on the heterocyclic groups. Now, as Dr.
`Hunt explained, in this particular context, the fact that you have inactivity
`and diminished activity exhibited by many of the compounds with
`heterocyclic, the Q1 groups, shows that activity is actually unpredictable.
`You have at least some of the compounds with heterocyclic rings,
`particularly the five membered heterocyclic rings, having little or no
`activity, and then, also, the fact that some Q1s with pyrazole showed no
`activity, and some showed very little activity, depending on what was
`substituted, indicates that there may be sensitivity to the nature and location
`of substituent groups.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`
`
`13
`
`
`
`PGR2020-00028
`Patent 10,294,202 B2
`
`
`And as Dr. Hunt explained in his declaration, you know, even very
`small substitutions can have large impacts on activity. And this is actually
`being borne out by the data, itself. So, just in Q1, we have a small number
`of groups that are being tested, that are not phenyls. And we can already see
`the level of inactivity rising and the unpredictability of that activity. Does
`that answer Your Honor’s question?
`JUDGE HARDMAN: Yes, thank you, Counsel.
`MR. ANDERSON: So, that is the positions of Q1 and Q2. As we
`said, there will be 167 million, looking -- this is at slide 10, 167 million
`possibilities at either position, without even considering the number of
`heterocycles. Then, we move to R1 and R6. So, R1 has 43 groups, R2 has -
`- R6 has 42 groups. And there’s a tremendous amount of possible variations
`in the R1 and the R6 position. So, both include the G1 subgroup, which is
`separately defined and may include any one of several different ring
`systems, each of which may be optionally substituted and independently
`substituted with up to five members from R13. And the R13 group includes
`39 groups it has at least 188 different substituents in it. So, you have, again,
`a tremendous amount of variation. And as Dr. Hunt explained in his
`declaration, even if you ignore the variation in the 43 R1 groups and the 42
`R6 groups, and you’re looking just at G1, you have at least 16 and a half
`billion possible substituents, at both R1 and R6. So, the bottom line is that
`your claim scope is vast. You have at least 4x10 to the 32 compounds, over
`167 million substituents at Q1 and Q2, and then 16 and a half billion
`substituents at both R1 and R6.
`Just to briefly touch on -- so, that’s essentially the Wands factor 1,
`which is the scope of the claims. Looking at the Wands factors 4, 5, and 7,
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`
`
`14
`
`
`
`PGR2020-00028
`Patent 10,294,202 B2
`
`those relate to the nature of the invention, the state of the prior art, and the
`level of predictability in the art. Here, the invention relates to herbicides and
`to agrochemistry, and there’s no dispute that agrochemistry is an
`unpredictable art. And moving to the next slide, that’s because, as Dr. Hunt
`explained, this lack of predictability in agrochemicals and in herbicides,
`more specifically, is really driven, in large part, by the fact that you can have
`these small chains, in what appear to be very minor substitutions, can have
`very significant effects on biological substitution or activity.
`So, you know, for example, in 142, he explained what might appear to
`be a minor substitution of fluorine or chlorine instead of a hydrogen can
`have a dramatic effect on activity by altering phenomena, such as
`lipophilicity, metabolism, plant uptake, and translocation. Now, notably,
`these opinions, regarding the unpredictability in the agrochemical field, were
`not rebutted by FMC or Dr. Dayan. And as we mentioned, the conclusion is
`actually supported by the biological data, in the 202 patent, itself, which I’ll
`address in just a moment.
`Before I do that, I just wanted to touch briefly on the level of skill in
`the art. Here, the level of skill in the art is relatively high. It’s at least a
`bachelor’s degree, with several years of experience. However, as the court
`noted in Idenix, the level of skill in the art is not sufficient to enable the
`patents in a case like this. In Idenix, the court found that, although the level
`of skill in the art is high, the patent doesn’t provide enough meaningful
`guidance for working examples across the full scope of the claim to allow a
`person of ordinary skill in the art to determine which nucleosides in that case
`would be effective or not, without extensive screening. That’s essentially
`the same here. You have a vast breadth of experimentation that would be
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`
`
`15
`
`
`
`PGR2020-00028
`Patent 10,294,202 B2
`
`required in an unpredictable field to try to bridge the huge gap between the
`disclosures of the specification and the breadth of the claims.
`JUDGE HARDMAN: Isn’t though, Counsel, one major difference
`between Idenix and this, the facts here, the agrochemical field is a mature
`field, which I think the Fed Circuit found was not the case, in the Idenix
`case. So, doesn’t the maturity of the field and the fact that the skilled artisan
`has a lot of tools at their disposal sort of counter the unpredictability and,
`you know, the expectation here, the skilled artisan is working in an
`unpredictable field, but does have the expectation of the unpredictability and
`the tools to sort of address that?
`MR. ANDERSON: Actually, Your Honor, I’m not sure I would agree
`with that, and I think if you -- in particular, when we start talking about the
`literature that we cited for the QSARs, for example, they make it clear that
`predictability is still actually very limited, even though this is the mature
`field. So, as we had referenced in our briefing, when you look at the, I think
`it’s exhibits 218, and 217, and 221, when they start talking about QSARs,
`they make it clear, the references, and these are references that Dr. Dayan,
`himself, cited, that the predictability in the field is essentially limited to
`compounds of a similar nature to those tested. So, I might be able to create a
`model, you know, the art is mature enough to create, yes, potentially, a
`model. But that model is not actually going to allow me to predict beyond
`the scope of those limited compounds that I’ve tested. So, I have a chemical
`space and my predictions are limited to the chemical space that I’ve used to
`create the model.
`Similarly, the art comes out and specifically says, you know, you
`can’t do a model without validation, and validation requires testing across,
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`
`
`16
`
`
`
`PGR2020-00028
`Patent 10,294,202 B2
`
`essentially, the entire scope of compounds that you want to predict. So, not
`only do you need, in order to create a model or to really, effectively,
`understand and predict activity, not only do you need, essentially, activity
`across the full -- or testing across the full space, but then to actually test your
`predictions, you need to have a vast amount of testing that’s, again,
`representative of the chemical space, to make sure that what you’re testing is
`actually accurate, what your predictions are, are actually accurate.
`So, I -- while it is a mature field, in some regards, I think that the
`maturity still has limits, and you, in theory, could do modeling, even in
`pharmaceuticals, but the problem you’d run into, again, is you’re limited by
`your starting materials. So, I think, in a way, I --
`JUDGE HARDMAN: I guess it just -- sorry, go, go ahead.
`MR. ANDERSON: What’s that? No, no, I was going to say, so,
`while you have a mature or potentially mature field or some maturity in the
`field, that maturity is actually still limited by the unpredictability. The tools,
`themselves, are still limited.
`JUDGE HARDMAN: Help me understand your point a little bit more
`on that the modeling -- that you can’t extrapolate beyond the tested
`compounds because isn’t the point of modeling to extrapolate beyond what
`you’ve put into the model to tell you something about what’s not in the
`model? So, I’m having a little bit of trouble understanding your point there.
`MR. ANDERSON: Well, I mean, to some extent, Your Honor, I
`mean, this is actually not necessarily just our position. This is actually
`taking directly out of some of the exhibits, and I have cited here what’s
`actually slide 34. These are exhibits that Dr. Dayan, himself, cited. And
`you can have a model, and you’ll be able to predict a chemical space around
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`
`
`17
`
`
`
`PGR2020-00028
`Patent 10,294,202 B2
`
`which you have similar compounds, and so, you might be able to extend a
`little bit beyond the compounds you’ve tested, and you might be able to
`extrapolate beyond it. It’s not just limited to the very specific compounds,
`but it’s compounds within, essentially, the same classes that you’ve already
`tested, and the same scope, the same sterics, the same -- for example, here, it
`talks about the same log P, the same physical chemical parameters, the same
`sterics, the same electronics, that you have positions, you know, electron
`donors in the same position. You have electron withdrawing or polar
`molecules in the same position. When you have all of those things, then you
`have ability to predict with things that have similar properties.
`In this case, however, we have such a limited scope, compared to the
`broad scope of the patent, that you just cannot predict it. So, we’re not
`saying that you would be limited to a model, if I have 350 compounds, that
`that compound is only going to be limited to, you know, maybe 100
`compounds, additional compounds, but it’s going to be limited in space to
`things that are very similar. And that’s what the art, itself, is telling us. I
`mean, this is Dr. Dayan’s own recommendations, in his own art, are saying
`that you cannot extrapolate beyond the area of the training set.
`JUDGE HARDMAN: Thank you, Counsel.
`MR. ANDERSON: So, you have some -- you can, you know,
`extrapolate and move a little bit, which, you know, maybe as -- I think Dr.
`Dayan used the example of, well, you might be able to go to the boundaries
`of your training set, but you really can’t go beyond the boundaries of your
`training set. And that’s the problem here, is that the training set is so
`limited, compared to the vast scope of the claims.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`
`
`18
`
`
`
`PGR2020-00028
`Patent 10,294,202 B2
`
`
`So, as we were saying, the level of skill in the art is high, but whereas
`here, you have a vast amount of -- or a vast scope. The huge -- the level of
`skill in the art i