PGR2019-00002
`U.S. Patent 9,855,302 B2
`
`
`
`
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`___________________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`___________________
`
`GENOME & COMPANY,
`Petitioner
`
`v.
`
`THE UNIVERSITY OF CHICAGO,
`Patent Owner
`___________________
`
`Case No. PGR2019-00002
`U.S. Patent 9,855,302 B2
`___________________
`
`
`REPLY DECLARATION OF JONATHAN BRAUN, M.D., Ph.D., IN POST
`GRANT REVIEW OF U.S. PATENT NO. 9,855,302
`
`
`
`

`

`TABLE OF CONTENTS
`
`PGR2019-00002
`U.S. Patent 9,855,302 B2
`
`Page(s)
`
`I.
`
`II.
`
`INTRODUCTION ........................................................................................... 1
`
`APPLICABLE LEGAL PRINCIPLES ........................................................... 2
`
`III. EXTENSIVE HUMAN ONCOLOGY CLINICAL EXPERIENCE .............. 3
`
`IV. THE ‘302 PATENT DISCLOSURE ............................................................... 4
`
`A.
`
`V.
`
`The Claimed Functional Check Point Inhibitors ............................................. 4
`
`PROPERTIES OF BIFIDOBACTERIUM SPECIES ARE
`UNPREDICTABLE AND SOMETIMES STRAIN SPECIFIC ..................... 5
`
`A. Different Strains of Bifidobacterum Longum Drive Different Immune
`Responses .................................................................................................... 5
`
`B.
`
`C.
`
`D.
`
`E.
`
`F.
`
`G.
`
`H.
`
`I.
`
`O’Mahoney’s Data Is Sufficient To Draw Conclusions Regarding the the
`Properties of the Bifidbacterium Strains Described Therein ....................... 5
`
`Dong Described Immunostimulatory Bifidobacterium Longum and Its
`Methodology Is Sound ................................................................................. 8
`
`Lopez Describes Species And Strain Specific Effects Of Bifidobacterium
`And its Methodology is Sound .................................................................. 11
`
`Lee Describes Bifidobacterium Adolescentis Spm 2012 Exerting An Anti-
`Proliferative Effect Human Colon Cancer Cell Lines ............................... 14
`
`Reddy ......................................................................................................... 15
`
`Cancer Immunotherapy Is Vastly More Complex and Uncertain Than
`Simply Targeting Immune Checkpoints .................................................... 16
`
`Check Point Inhibitors Show Low Response Rates In A Limited Number
`Of Cancers And For Only A Subset Of Cancer Patients Having Those
`Cancers ...................................................................................................... 17
`
`Extensive and Undue Experimentation Is Required To Practice the Claims
`of the ‘302 Patent ....................................................................................... 20
`
`VI. The ‘302 Claims are Obvious ........................................................................ 23
`
`A.
`
`B.
`
`(Ground 2) Korman ‘401 in View of Singh and Dong Render Obvious
`Claims 1-9, 12-17, and 19-25, and 27-28 .................................................. 23
`
`Korman '401 in View of Kohwi Renders Obvious Claims 1-4, 7-9, 12-17,
`19-25, and 27-28 ........................................................................................ 25
`
`i
`
`

`

`1. Claims 5, 6, 9, 10, 11, 23, 24 and 26 Are Obvious ................................... 28
`
`C.
`
`Ground (9) Korman ‘401 in View of Mohania and Prakash ‘449 Renders
`Obvious Claims 1-9, 12-17, and 19-25, and 27-28 ................................... 29
`
`PGR2019-00002
`U.S. Patent 9,855,302 B2
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`

`

`I, Jonathan Braun, M.D., Ph.D., being of legal age, hereby declare affirm, and
`
`PGR2019-00002
`U.S. Patent 9,855,302 B2
`
`state the following:
`
`I.
`
`INTRODUCTION
`
`1.
`
`As I testified in my declaration signed October 2, 2018 (“Opening
`
`Declaration”), which I understand has been labeled as Exhibit 1002 in this
`
`proceeding, I have been retained on behalf of Genome & Company/Petitioner, to
`
`serve as an independent expert and provide expert opinions regarding U.S. Patent
`
`No. 9,855,302 (“the ‘302 patent”) (Ex. 1001).
`
`2.
`
`Relevant aspects of my educational background, career history, and
`
`other qualifications were provided in my Opening Declaration and will not be
`
`repeated here. (See Ex. 1002 at ¶¶ 6-14.)
`
`3.
`
`I provided testimony in my Opening Declaration regarding the
`
`technical subject matter of the ‘302 patent and the application of various references
`
`that are prior art to the ‘302 patent. In particular, I was asked to consider what a
`
`person of ordinary skill in the art would have understood from the ‘302 patent,
`
`whether the specification teaches a person of ordinary skill in the art how to make
`
`and use the claimed invention without undue experimentation, and whether the
`
`prior art discussed herein rendered the claimed invention obvious to a person of
`
`ordinary skill in the art.
`
`1
`
`

`

`PGR2019-00002
`U.S. Patent 9,855,302 B2
`
`4.
`
`In this Reply Declaration, I have been asked to consider Patent
`
`Owner’s Response (“POR”) and the Declaration of Dr. Mani (Ex. 2007) vis a vis
`
`my opinions regarding the enablement and validity of the claims of the ‘302 patent.
`
`5. My testimony in this declaration, like my Opening Declaration, is
`
`through the eyes of a person of ordinary skill in the art as defined in ¶¶ 38-40 of
`
`my Opening Declaration.
`
`6.
`
`I have also relied on the knowledge and experience I acquired from
`
`more than 33 years of practicing, researching, and teaching oncology, pathology,
`
`microbiology and immunology.
`
`7.
`
`I have personal knowledge of the facts and opinions set forth in this
`
`declaration, and, if called upon to do so, I would testify competently thereto. All
`
`of the opinions and conclusions found in this declaration are my own.
`
`8.
`
`I am being compensated at a rate of $550.00 per hour for my services.
`
`This compensation is in no way based on the content of my opinions or the
`
`outcome of this matter.
`
`
`
`II. APPLICABLE LEGAL PRINCIPLES
`
`9. My understanding of certain legal standards were previously stated in
`
`¶¶ 15-37 of my Opening Declaration and will not be repeated here.
`
`2
`
`

`

`PGR2019-00002
`U.S. Patent 9,855,302 B2
`III. EXTENSIVE HUMAN ONCOLOGY CLINICAL EXPERIENCE
`
`10.
`
`I have extensive human oncology clinical experience.
`
`11.
`
`In my 24 years as Chair of Pathology and Laboratory Medicine at
`
`UCLA, I have directed all surgical pathology and clinical laboratory services for
`
`UCLA Health, with responsibility for surgical pathology diagnosis for over
`
`200,000 patients, and 5.5 million laboratory tests.
`
`12. Over 65% of surgical pathology cases are pathology consultations to
`
`provide critical information for diagnosis or management of cancer patients, as are
`
`over 20% of clinical laboratory testing. In performing these clinical cancer
`
`services, I directed the activities of over 100 pathology-specialist clinicians and
`
`over 600 technical staff, as well their roles in the 12 concurrent biweekly cancer
`
`patient-management tumor boards where oncologists, radiation oncologists, and
`
`subspecialist pathologists radiologists together perform patient-level diagnosis and
`
`treatment-planning.
`
`13. As a member of senior leadership in UCLA Health, I participated in
`
`or led multiple line-management and strategy initiatives for our clinical cancer
`
`portfolio. Recent examples included the UCLA Health ambulatory cancer care
`
`strategy; the Urologic Oncology Institute and integrated neurocancer service line;
`
`regional proton-beam services; CAR-T cell therapy; and genome-wide NGS
`
`somatic and germline cancer clinical diagnostics).
`
`3
`
`

`

`PGR2019-00002
`U.S. Patent 9,855,302 B2
`
`IV. THE ‘302 PATENT DISCLOSURE
`
`A. The Claimed Functional Check Point Inhibitors
`
`14.
`
`In my Opening Declaration, I addressed the meaning of the claimed
`
`check point inhibitor (“CPI”). I opined that the ‘302 specification defined CPI
`
`functionally as “a protein or polypeptide that binds an immune checkpoint”
`
`including “antibodies or antigen binding fragments that bind to and inhibit an
`
`immune checkpoint protein” as well as “interfering nucleic acids” which inhibit an
`
`immune checkpoint protein. See Ex. 1002, ¶49, see also ¶¶ 47, 48, 50. I also
`
`declared that the functionally claimed CPI comprised a “limitess number of
`
`possibilities.” See Ex. 1002, ¶ 41.
`
`15. The ‘302 specification does not provide any guidance for making
`
`fragments or interfering nucleic acid molecules that successfully function as CPIs.
`
`16. Dr. Mani at his deposition (Ex. 1042, 19:7-29:6) references Exhibits
`
`2009, 2011-2014, and 2016 and states that some of those references may deal with
`
`fragments of antibodies that function as checkpoint inhihibitors. Ex. 1042 at 20:7-
`
`21:15.
`
`17. None of Exhibits 2009, 2011-2014, and 2016 describe a fragment of
`
`an antibody that functions as checkpoint inhibitor.
`
`18. Exs. 2009 and 2011 both describe anti-KIR antibodies. Ex. 2012
`
`describes anti-PD-1 and anti-LAG-3 antibodies. Ex. 2013 describes an LAG
`
`4
`
`

`

`PGR2019-00002
`U.S. Patent 9,855,302 B2
`
`fusion protein. Ex. 2014 describes an anti-VISTA antibody. And Ex. 2016
`
`describes an anti-TIM-3 antibody.
`
`19.
`
`I am not aware of any antibody fragments or interfering nucleic acids
`
`that have shown efficacy as a CPI in treating cancer.
`
`V.
`
`PROPERTIES OF BIFIDOBACTERIUM SPECIES ARE
`UNPREDICTABLE AND SOMETIMES STRAIN SPECIFIC
`
`A. Different Strains of Bifidobacterum Longum Drive Different
`Immune Responses
`
`20. PO represented that Bifidobacterium longum strain 1714 described in
`
`O’Mahoney (Ex. 1017) was immunosuppressive. PO stated:
`
`O'Mahoney et al. teach that the disclosed Bifidobacterium
`
`strain suppresses such immune responses.
`
`Ex. 1014, 123-132, 132.
`
`21.
`
`In a subsequent response, PO reiterated that “O’Mahoney describes
`
`suppression of immune response.” Ex. 1014, 57-65, 64.
`
`22. The ‘302 patent describes Bifidobacterium longum as
`
`immunostimulatory. Ex. 1001, 38:36-58; 39:12-40:54.
`
`B. O’Mahoney’s Data Is Sufficient To Draw Conclusions Regarding
`the Properties of the Bifidbacterium Strains Described Therein
`
`23. PO alleges that O’Mahony does not provide sufficient data or
`
`statistical analysis to justify the conclusions Petitioner ascribes to it. POR, 24.
`
`These allegations do not hold up to scrutiny.
`
`5
`
`

`

`PGR2019-00002
`U.S. Patent 9,855,302 B2
`24. PO states that the O’Mahony does not report any comparative in vivo
`
`testing of Bifidobacterium strains. POR, 20-21. This is untrue. O’Mahony [087,
`
`088] highlighted their own mouse previously published in vivo comparative testing
`
`of Bifidobacterium, ref. 8 [0151].
`
`25. PO alleges a series of deficits in O’Mahony, by mistakenly
`
`demanding that each example must test at once all the immunologic properties of
`
`the system. POR, 20-24. This willfully ignores the actual and reasonable design
`
`of examples, which cumulatively presents an orderly, sequential assessment of
`
`these properties. And, it presents them in distinct tissue settings, which exemplifies
`
`the immunologic “cascade” concept that O’Mahony highlights at [0088] and which
`
`is endorsed by the PO at POR, 22.
`
`26. Thus, PO alleges due to example 5 that “O’Mahony reports on just a
`
`handful of cytokine measurements”, and does not present IL-10. POR, 21.
`
`However, the series of examples actually present data for 6 cytokines (and the
`
`critical signaling intermediate, NFkB), most in both in vitro and in vivo settings.
`
`As I discuss below for Lopez, these cytokines span the key molecules in dendritic
`
`cell-T cell interaction, and O’Mahoney evaluates them in both tissue culture and
`
`animal contexts.
`
`
`
`
`
`6
`
`

`

`PGR2019-00002
`U.S. Patent 9,855,302 B2
`
`27.
`
`
`
`Cytokine
`
`in vitro
`
`in vivo
`
`IL-6
`
`IL-10
`
`IL-12
`
`Table 5
`
`
`
`Fig 4
`
`Fig 8
`
`Table 6
`
`Fig 7
`
`IFN-alpha
`
`Table 6
`
`
`
`IFN-gamma
`
`Table 6
`
`Fig 6
`
`TNF-a
`
`Fig 5
`
`Fig. 8, 16
`
`NF-kB
`
`
`
`Fig 9, 10
`
`
`
`28. PO criticizes O’Mahony for incomplete description of donor numbers
`
`and absence of statistics (POR, 20) and lack of positive and negative controls
`
`(POR, 22). However, such information and statistics is included in Fig. 4, 5, 6, 7,
`
`8, 9, 10, and 16.
`
`29. PO criticizes O’Mahony “for too many gaps… to reach any
`
`conclusions about differences between the 1714 and 35624 strains” (POR, 22).
`
`However, the central point of the O’Mahony patent is demonstration that certain
`
`Bifidobacterial species (notably 1714) can be strongly immunosuppressive.
`
`30. PO alleges that a POSITA requires an explanation for why “a
`
`difference in an effect on peripheral blood cells in vitro as corresponding to a
`
`7
`
`

`

`PGR2019-00002
`U.S. Patent 9,855,302 B2
`
`difference in an effect on cancer in vivo.” POR, 24. I note here, and detail in a
`
`subsequent section, that dendritic cell IL-12 and its induction of Th1 IFNgamma is
`
`well-understood to be a crucial component of anti-tumor immunity by stimulation
`
`of the immune system. The existence of Bifidobacterium sp. that are deficient in
`
`inducing immunostimulatory dendritic cells (due to relative low IL-12 production
`
`and high IL-10) is thus obvious to a POSITA as a significant unpredictability in the
`
`efficacy of a given species of Bifidobacterium on a particular cancer.
`
`C. Dong Described Immunostimulatory Bifidobacterium Longum and
`Its Methodology Is Sound
`
`31. PO alleges that a POSITA would have not have concluded from Dong
`
`that the B. longum described therein is immunostimulatory. In support thereof, PO
`
`alleges that Dong’s methodology is flawed. POR, 51. Patent Owner also alleges
`
`that Dong shows B. Longum’s upregulation of IL-10, a cytokine known to be anti-
`
`inflammatory (POR, 58).
`
`32. PO is incorrect. Dong describes Bifidobacterium longum as
`
`immunostimulatory. See Ex. 1002, ¶¶ 113, 170-74.
`
`33.
`
`I also note that the inventors of the ‘302 Patent, Drs. Gajewski, Sivan,
`
`and Corrales, expressly referenced Dong as describing immunostimulatory
`
`bifidobacterium longum. The inventors state: “[s]timulatory interactions between
`
`bifidobacteria [longum] and the host immune system, including those associated
`
`with interferon-(cid:1) (IFN(cid:1)) have been described previously. (13-16). We thus
`
`8
`
`

`

`PGR2019-00002
`U.S. Patent 9,855,302 B2
`
`hypothesized that members of this genus could represent a major component of the
`
`beneficial antitumor immune effects observed in JAX mice”. Ex. 2005, 1085. In
`
`the preceding equation from the inventors, reference number 15 is Dong.
`
`34. Contrary to PO’s assertions, Dong’s methodology is not flawed.
`
`35. Dong was published in “Early Human Development,” a peer-reviewed
`
`publication. Accordingly, the methodology for generating the data referenced
`
`therein passed the peer review process, and thus, was not judged to be flawed by an
`
`independent review process
`
`36. Further, PO incorrectly states that “…there was no way to know the
`
`baseline microbiota of Dong’s SPF rats.” (POR p. 54). In fact, it was already
`
`well-known that rodents bred together in a single facility have a shared baseline
`
`microbiome composition (Vanessa K. Ridaura, et al., Gut Microbiota from Twins
`
`Discordant for Obesity Modulate Metabolism in Mice, Science, 341:1079,
`
`1241214 (2013) (Ex. 1045). The rats studied by Dong were bred and housed in a
`
`single facility, so a POSITA would understand that all rats used in the study had a
`
`shared baseline micrbiome composition prior to the experimental intervention. In
`
`this regard, the PO misrepresented my response (Ex. 2033, 66:19–23), which
`
`actually concerned the microbiome composition after the experimental
`
`intervention.
`
`9
`
`

`

`PGR2019-00002
`U.S. Patent 9,855,302 B2
`37. PO alleges that Dong’s experimental design prevents attribution of the
`
`observations to B. longum. POR, 55. However, PO affirms that the BS group
`
`simply administered B. longum, and that Dong provides a direct comparison of the
`
`substantial differences in immunologic state of the BS group to a control which
`
`had not received this administration. A POSITA readily understands that any such
`
`differences between these two groups are attributable to B. longum administration.
`
`38. The PO allegation expands by questioning whether the effect of B.
`
`longum was induced by the organism alone. POR, 55. However, this question
`
`does not refute the effect of B. longum. A POSITA understands that, whether B.
`
`longum acted alone or in part via its ecologic effect on other microbiome members,
`
`administration of this organism elicited the observed host response.
`
`39. PO alleges that the RT-PCR cytokine analysis was unreliable, because
`
`Dong did not specify technical controls used in the RNA preparation. POR, 56.
`
`However, this is invalid for two reasons. First, technical controls for RNA
`
`isolation are so routinely used for such experimentation that they are rarely
`
`documented in publications, and indeed unwelcome by editors as superfluous. So,
`
`a POSITA would assume that Dong was using a routine protocol that included
`
`such technical quality checks. Second, contamination and degradation, if present,
`
`adds variability between individual samples regardless of their experimental group.
`
`Accordingly, they are detectable in the variance (e.g., standard error) of each
`
`10
`
`

`

`PGR2019-00002
`U.S. Patent 9,855,302 B2
`
`group. In contrast to the PO allegation , they would not render any observations
`
`unreliable, unless the variances undermined statistical significance for the observed
`
`findings. Since Dong reported variances and drew conclusions only on statistically
`
`significant findings a POSITA would understand that the PO allegation is specious.
`
`40. With respect to PO’s argument that B. longum induced upregulation
`
`of IL-10 (POR, 57), a cytokine known to be anti-inflammatory, Dong showed that
`
`Bifidobacterium longum induced maturation in dendritic cells characterized by
`
`increased expression of the immunostimulatory CD86 surface protein and IL-12
`
`cytokine. Dong also demonstrated their immunostimulatory function by capacity to
`
`induce T cell IFN-γ, the hallmark of Th1 activation, which in turn is the hallmark
`
`of immunostimulatory host response in cancer. Dunn et al., The Three Ez of
`
`Cancer Immunoediting, Annu. Rev. Immunol. 2004, 22:329-360. Ex. 1046.
`
`41. Significantly, the ‘302 patent also pointed to elevated levels of T-cell
`
`induced IFN-γ as evidence of Bifidobacterium’s immunostimulatory function to
`
`activate dendritc cells. Ex. 1001, 39:43-40:54.
`
`42. Accordingly, like the inventors of the ‘302 Patent, a POSITA would
`
`have concluded that Dong described B. Longum as being immunostimulatory.
`
`D. Lopez Describes Species And Strain Specific Effects Of
`Bifidobacterium And its Methodology is Sound
`
`43. PO’s expert attacks Lopez, asserting that its methodology is
`
`unreliable, and therefore, does not “provide[] credible evidence.” POR, 19, 24-36.
`
`11
`
`

`

`PGR2019-00002
`U.S. Patent 9,855,302 B2
`
`However, a subsequently published, peer reviewed journal cited Lopez as
`
`describing strain specific effects of Bifidobacterium that correlate to different in
`
`vivo effects on the immune system, i.e., Th1 and Th2 responses
`
`It is known that different probiotic bacteria present different effects upon the
`
`immune system [7, [Lopez]],.. Some strains promote Th1 responses,
`
`characterized by the production of IFN(cid:1) and TNF(cid:2), whereas other strains
`
`induce anti-inflammatory cytokines generating a Th2 profile [Lopez]
`
`Sanchez, Borja Sánchez et al., The Effects of Bifidobacterium breve on Immune
`
`Mediators and Proteome of HT29 Cells Monolayers., BioMed Research
`
`International, 2015 (Article ID 479140):1-6 (2015). (Ex. 1044).
`
`44. The above quote by Sanchez also shows, contrary to PO’s assertions
`
`(POR, 32), that the data in Lopez demonstrates how the immune system would be
`
`effected, in vivo, i.e, by generating a Th1 or Th2 response.
`
`45. PO also alleges that Lopez does not provide sufficient data or
`
`statistical analysis to justify the conclusions Petitioner ascribes to it.
`
`46. PO alleges that Lopez's data is skewed and cannot be analyzed
`
`quantitatively.
`
`47. This conclusion is clearly false. Cytokine and maturation marker data
`
`is not normally-distributed (affirmed by both Lopez and the PO and termed
`
`“skewed” per the POR). However, non-parametric statistics (e.g., Wilcoxon-Mann-
`
`12
`
`

`

`PGR2019-00002
`U.S. Patent 9,855,302 B2
`
`Whitney or Kendall W methods) are well-established and widely used to
`
`quantitatively analyze such data and these were the methods employed by Lopez.
`
`48. The PO further alleges that critical conclusions about Bifidobacterium
`
`differences cannot be quantitatively made with non-parametric statistical analysis,
`
`because such analysis does not permit direct donor comparisons. But as even the
`
`PO agrees, the ranking method of such analysis permits quantitative comparison of
`
`groups. This is how Lopez presented and interpreted the study findings. For
`
`example, see Lopez p. 160: “No relevant differences were observed in the
`
`production of IL-8 (p = 0.207, Kendall test) and IL-6 (p = 0.192), but,
`
`interestingly, independently of the donor, the variety of bifidobacterial strains
`
`differed substantially in their capacities to induce IL-1β (p = 0.041), IL-10 (p =
`
`0.001), TNFα (p = 0.008) and IL-12 (p = 0.008).”
`
`49. PO also alleges that Figure 2 does not provide sufficient information
`
`to determine whether there were any differences in the cytokine profiles and that
`
`there is insufficient information to conclude the differences in cytokine profiles
`
`amounted to biological differences. In support thereof, PO also alleges that Lopez
`
`analyzed only 6 cytokines. PO is wrong.
`
`50. The literature demonstrates that among more than 45 cytokines
`
`discussed in major reviews by 2015, only 9 were known to reflect dendritic cell
`
`immunstimulation. Patrick Blanco, et al., Dendritic cells and cytokines in human
`
`13
`
`

`

`PGR2019-00002
`U.S. Patent 9,855,302 B2
`
`inflammatory and autoimmune diseases, Cytokine & Growth Factor Reviews, 19
`
`(Issue 1):41-52 (2008)(Ex. 1047. Lopez reported on 6 of these 9 cytokines, and
`
`these 6 were the most widely studied for this biologic role.
`
`51. Moreover, one of these cytokines, interferon-gamma, demonstrates a
`
`biologic difference reported by Lopez that is crucial to anti-cancer immunity. It
`
`was already widely understood that interferon-gamma is specifically the product of
`
`T cells induced by IL-12 of immunostimulatory dendritic cells into the Th1 state.),
`
`Exs. 1046 and 1047. And, Th1 and their interferon-gamma production are crucial
`
`to anti-cancer immunity. Ex. 1046. It was obvious to a POSITA that the findngs of
`
`Lopez (differential IL-12 and IFN-gamma induction between Bifidobacterium
`
`species) was a biologic difference important to anti-cancer immunity. Thus,
`
`contrary to the opinion of PO’s expert, that the data in Lopez shows how the
`
`immune system would be effected, in vivo.
`
`E.
`
`Lee Describes Bifidobacterium Adolescentis Spm 2012 Exerting
`An Anti-Proliferative Effect On Human Colon Cancer Cell Lines
`
`52. PO asserts that a POSITA would not have concluded that the anti-
`
`proliferative effect of an N-butanol extract of Bifidobacterium Longum on human
`
`colon cancer cell line was due to Bifidobacterium Longum. In particular, PO
`
`alleges that Lee did not characterize the extract or run control experiments to
`
`confirm what the material was in the extract responsible for the antiproliferative
`
`effect.
`
`14
`
`

`

`PGR2019-00002
`U.S. Patent 9,855,302 B2
`53. Patent Owner is incorrect. Lee assayed three different extracts of
`
`Bifidobacterium adolescentis SPM 2012, only one of which – N-butanol - showed
`
`inhibition of proliferation of three colon cancer cell lines. A POSITA, therefore,
`
`would have assumed that the inhibition of proliferation of the three colon cancer
`
`cell was not an artifact.
`
`54. Further, while Lee does not state what component(s) of
`
`Bifidobacterium adolescentis SPM 2012 were extracted by the N-butanol and
`
`responsible for the anti-cancer activity, a POSITA would have understood the
`
`experiments in Lee to show that the Bifidobacterium adolescentis SPM 2012 was
`
`responsible for the anti cancer activity. Indeed, Lee stated that Bifidobacterium
`
`adolescentis SPM 2012 was chosen for the assays because it exhibited the highest
`
`efficacy of three Bifidobacterium adolescentis strains for inhibiting the growth of 3
`
`different colon cancer cell lines. The N-butanol experiment was merely to “further
`
`characterize the functional substances of B. adolescentis SPM0212.” P. 3.
`
`F. Reddy
`
`55. Subsequently published, peer reviewed articles described Reddy as
`
`showing Bifidobacteria exerting anti tumor activity.
`
`56. Reza Aghebati Maleki et al., Effects of some natural
`
`immunomodulatory compounds in combination with thalidomide on survival rate
`
`and tumor size in fibrosarcoma-bearing mice, Advanced Pharmaceutical Bulletin,
`
`15
`
`

`

`PGR2019-00002
`U.S. Patent 9,855,302 B2
`
`4, Suppl 1:465-470 (2014) (Ex. 1048, 466) states that Reddy describes
`
`“Bifidobacteria show inhibitory effects on colon … cancer[]”. Blanda Di Luccia et
`
`al., Lactobacillus gasseri SF1183 Affects Intestinal Epithelial Cell Survival and
`
`Growth, PLoS ONE, 8(7):e69102 (2013) (Ex. 1049, 1) states that Reddy describes
`
`“in vivo studies have shown the inhibitory activity of [Bifidobacteria Longum] on
`
`…colon tumors in animal models”
`
`G. Cancer Immunotherapy Is Vastly More Complex and Uncertain
`Than Simply Targeting Immune Checkpoints
`
`57. PO asserts that all cancer cells express and present foreign antigens on
`
`their surfaces, and therefore, all cancer cells are recognizable, at least in theory, by
`
`the immune system. POR, 4-9
`
`58.
`
`Immune checkpoints, however, suppress the functioning of the
`
`immune system, thereby allowing the tumor cells to evade the immune system.
`
`Thus, according to PO, by administering a checkpoint inhibitor to inhibit the
`
`function of the immune checkpoint, the functioning of the immune system will be
`
`restored and, in theory, all cancer cells, regardless of cancer type, will be
`
`recognized by the immune system and destroyed. POR, 8-9.
`
`59. PO’s theory ignores that fact that there are many immune mechanisms
`
`by which tumor cells evade the immune system.. Immune checkpoints are only one
`
`of those immune mechanisms.
`
`16
`
`

`

`PGR2019-00002
`U.S. Patent 9,855,302 B2
`60. For example, tumor cells can secrete immunosuppressive factors such
`
`as TGF-ß, indoleamine-pyrrole-2,3-dioxygenase (1DO), tryptophan-2,3-
`
`dioxygenase (TDO), galectin 3, or natural killer cell decoys. Ex. 2028, 308, 309.
`
`Tumor cells can also produce immunosuppressive metabolites such as lactic acid,
`
`adenosine, and prostaglandins. Id. They can also decrease their antigenicity by
`
`downregulating or extinguishing HLA class I molecules. Id.
`
`61. Non-tumor cells can also contribute to immunosuppressive
`
`mechanisms. These include “regulatory T [Treg] cells that could be attracted to
`
`some tumors by chemokines, and myeloid-derived suppressor cells and their
`
`contact-dependent immunosuppression, which involves nitrogen oxide (NO) and
`
`reactive oxygen species (ROS). Ex. 2028, 309.
`
`62. Another mechanism by which the tumor cells evade the immune
`
`system is the inability of the immune cells to penetrate the tumor. Ex. 2028, 310.
`
`63. PO’s expert admitted he did not account for any of these known
`
`alternative possibilities of immunosuppressive mechanisms when providing his
`
`enablement analysis. Ex. 1042: 47:21-59:3.
`
`H. Check Point Inhibitors Show Low Response Rates In A Limited
`Number Of Cancers And For Only A Subset Of Cancer Patients
`Having Those Cancers
`
`64. PO asserts that “cancers from all manner of tissue types and all
`
`degrees of mutational burden have been shown to respond to CPI therapy,” and
`
`17
`
`

`

`PGR2019-00002
`U.S. Patent 9,855,302 B2
`
`therefore, it “would not have been necessary to test all cancers for CPI efficacy.”
`
`See POR, 9-13, 6. PO alleges that a sample of clinical trial reports, whose results
`
`are tabulated in a demonstrative Table at page 10 of the POR, showed responses in
`
`cancers arising in a wide variety of tissue types that are collectively representative
`
`of all cancers. POR, 11. Further, PO alleges those reports showed responses in
`
`cancers throughout the mutational landscape, as shown in annotated Figures at p.
`
`11 and 12 of the POR. Id.
`
`65. Patent Owner grossly overstates its case. Those clinical trial reports
`
`provide data for a total of 3 different checkpoint inhibitors: CTLA-4, PD-1, and
`
`PD-L1; and for 16 types of cancer. None of the CPI’s are antibody fragments or
`
`interfering nucleic acids. Nor do sixteen cancers do not correspond to all “cancers
`
`from all manner of tissue types” as alleged by PO. Furthermore, the clinical trial
`
`reports do not provide data for cancers throughout the mutational landscape.
`
`Nearly 50% (13 of 29) of the cancers listed on the annotated Figure on page 11 of
`
`the POR were not shown as being responsive to checkpoint inhibitors, and the low
`
`response rates for each type of CPI are even more limited in terms of the cancer
`
`types: CTLA-4 (4/29); PD-1 (12/29); and PD-L1 (2/29). See, e.g., claims 13, 14,
`
`16-18, and 27-29. And approximately 25% percent of the cancers listed on a
`
`smaller sample size of cancers (3 of 12) were not shown as being responsive to
`
`checkpoint inhibitors.
`
`18
`
`

`

`PGR2019-00002
`U.S. Patent 9,855,302 B2
`66. Patent Owner also overstates the value of the “response rates” of those
`
`clinical trials. Most of the clinical trials were Phase I trials that tested for safety,
`
`not efficacy, and thus, did not include controls. Furthermore, many of the sample
`
`sizes were extremely small and thus could not rule out the natural variability of
`
`cancer progression. Accordingly, those clinical trial responses are preliminary
`
`findings at best.
`
`67. PO alleges Ex. 2037 shows a response rate of 2/3 for chronic
`
`lymphocytic leukemia administered a PD-1 antibody. Ex. 2037 actually shows that
`
`two of the patients exhibited stable disease. However, one patient had Stage A
`
`CLL and the other had Stage C CLL. Further, the patients were reported as having
`
`“advanced” CLL, not progressive CLL. (Dr. Mani was therefore wrong when he
`
`testified that the patients had progressive CLL, and therefore, acted as its own
`
`control. (Ex. 104*, 62:10-15 (“when a patient comes in with progressive disease,
`
`the patient is their own control”). Accordingly, given the extremely small sample
`
`size and the lack of a control, the reported response rates are not reliable.
`
`68. Ex. 2038 is a short poster presentation reporting on the “safety/
`
`tolerability” of a PD-1 antibody administered alone or in combination with an
`
`CTLA-4 antibody to glioblastoma patients (no control). Each arm comprised 10
`
`patients, with six of the ten patients showing an overall survival of 6 months. The
`
`trial did not include a control.
`
`19
`
`

`

`PGR2019-00002
`U.S. Patent 9,855,302 B2
`69. Ex. 2039 in its Abstract describes administering an CTLA-4 antibody
`
`to ten glioblastoma patients (no control). The overall survival (OS) for the 6 best
`
`patients was 5.1 months, which compares to “currently available treatments for
`
`recurrent glioblastoma [that] are inadequate [having] a mean survival [of]
`
`approximately six months.”
`
`70. Ex. 2040 reports on administering a CTLA-4 antibody to treat patients
`
`having advanced pancreatic ductal adenocarcinoma. Two out of fifteen patients
`
`were reported as showing stable disease at 7 and 22 weeks. The authors conclude
`
`that “there is insufficient data to conclude that ipilimab alone or the combination of
`
`ipilimab + GVAX leads to better clinical outcomes.” P. 388
`
`71. Ex. 2043 reports administering a PD-1 antibody to treat patients with
`
`platinum resistant ovarian cancer (no controls). Three of the thirteen patients
`
`exhibited partial responses. The authors concluded that the PD-1 antibody has
`
`“encouraging clinical efficacy.”
`
`I.
`
`Extensive and Undue Experimentation Is Required To Practice
`the Claims of the ‘302 Patent
`
`72. PO asserts that cancer treatment typically involves trial and error
`
`before “settling on a treatment that provides a satisfactory benefit,” and therefore,
`
`such trial and error testing is routine. POR at 16-19.
`
`73. PO’s assertions are groundless. It is undisputed the ‘302 specificatio