On behalf of: Par Pharmaceutical, Inc.
`
`Entered: August 3, 2017
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`_______________________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`_______________________
`
`PAR PHARMACEUTICAL, INC.,
`Petitioner
`
`v.
`
`NOVARTIS AG
`Patent Owner
`_______________________
`
`Case IPR2016-01479
`U.S. Patent No. 9,006,224
`_______________________
`
`Before LORA M. GREEN, CHRISTOPHER L. CRUMBLEY, and
`ROBERT A. POLLOCK, Administrative Patent Judges.
`
`SUPPLEMENTAL DECLARATION OF MARK J. RATAIN, M.D. IN
`SUPPORT OF PETITIONER’S REPLY IN THE INTER PARTES REVIEW
`OF U.S. PATENT NO. 9,006,224
`
`Ex. 1119-0001
`
`

`

`
`
`CONTENTS
`
`I.
`
`II.
`
`INTRODUCTION AND QUALIFICATIONS ............................................... 1
`
`UNDERSTANDING OF THE GOVERNING LAW ..................................... 1
`
`III. The Person of Ordinary Skill in the Art of the ’224 Patent ............................. 2
`
`IV. Scope of Declaration and Summary of Opinions ............................................ 7
`
`V.
`
`The ’224 Patent Would Have Been Obvious to a POSA in November
`2005 ............................................................................................................... 11
`
`A.
`
`B.
`C.
`
`PNETs That Failed Cytotoxic Chemotherapy Would Not Be
`Expected to Be “More Resistant” to Molecularly Targeted
`Therapies ............................................................................................. 11
`Preclinical Models for NETs ............................................................... 19
`Oberg & Eriksson Would Not Be Considered More Relevant by
`a POSA than Oberg 2004 .................................................................... 22
`D. Dr. Kulke Incorrectly Characterizes the Status of mTOR
`Inhibitors as Anticancer Agents in November 2005 ........................... 23
`A POSA Would Have Conceptualized mTOR Inhibitors as a
`Class of Similar Agents ....................................................................... 26
`A POSA Would Have Known PTEN Had Been Implicated in
`NET ..................................................................................................... 29
`G. A POSA Would Have Had a Reasonable Expectation that an
`mTOR Inhibitor, Like Everolimus, Would Have Antitumor
`Activity in Advanced PNETs .............................................................. 31
`
`E.
`
`F.
`
`VI. There Are No Objective Indicia Indicating a POSA Would Have
`Found the Claims Non-Obvious .................................................................... 36
`
`A. Dr. Kulke’s Analysis Does Not Indicate a POSA Would Have
`Found the Claims of the ’224 to Exhibit Unexpected Results ............ 36
`The Claimed Methods Did Not Satisfy Any Long-Felt But
`Unmet Needs ....................................................................................... 45
`
`B.
`
`
`
`i
`
`Ex. 1119-0002
`
`

`

`
`
`I, Mark J. Ratain, M.D., resident of Chicago, Illinois, hereby declare as
`
`follows:
`
`I.
`
`INTRODUCTION AND QUALIFICATIONS
`
`1.
`
`I am the same Mark J. Ratain who submitted a declaration in support
`
`of Par’s petition for inter partes review for U.S. Patent No. 9,006,224 (Ex. 1001,
`
`“the ’224 patent”). My qualifications and experience are identified in that
`
`declaration. (Ex. 1003, Ratain Decl.; Ex. 1004, Ratain CV.)
`
`2.
`
`I have been asked to consider and respond to the opinions and
`
`analyses of Novartis’s expert, Dr. Matthew Kulke, submitted in support of
`
`Novartis’s Response to the ’224 Petition.
`
`3. My work in this matter is currently being billed at my standard rate of
`
`$800 per hour, with reimbursement for necessary and reasonable expenses. My
`
`compensation is not in any way contingent upon the outcome of the petition. I
`
`have no financial interest in the outcome of this proceeding or any related
`
`litigation.
`
`II. UNDERSTANDING OF THE GOVERNING LAW
`
`4. My understanding of the legal analysis for obviousness is outlined in
`
`my original declaration submitted in this proceeding. (Ex. 1003.) Below I outline
`
`additional legal standards of which I have been informed by counsel for Par that
`
`are relevant to my analysis in this supplemental declaration.
`
`1
`
`Ex. 1119-0003
`
`

`

`
`
`5.
`
`I am informed by counsel for Par that certain factors, sometimes
`
`known as “secondary considerations of non-obviousness,” must be considered, if
`
`present, when in an obviousness determination. These secondary considerations
`
`include: (i) satisfaction of a long-felt but unmet need, (ii) unexpected results, (iii)
`
`skepticism of others of the invention, (iv) teaching away from the invention, (v)
`
`commercial success, (vi) praise by others for the invention, and (vii) copying by
`
`other companies.
`
`6.
`
`I am informed by counsel for Par that “unexpected results” require
`
`that (1) there be a difference in properties between the claimed subject matter and
`
`the closest prior art and (2) this difference would have been unexpected to a POSA
`
`as of the date of the alleged invention. For the purposes of my analysis, I consider
`
`the disclosures of Oberg 2004 (Ex. 1027) and Duran (Ex. 1011) to be the closest
`
`prior art.
`
`III. THE PERSON OF ORDINARY SKILL IN THE ART OF THE ’224
`PATENT
`7.
`
`I agree with Dr. Kulke that the claims of the ’224 patent are directed
`
`to methods of treating PNET and are not limited to the treatment of patients in
`
`clinical trials. (Ex. 1070, Kulke Dep. 173:3-9.) I also agree with Dr. Kulke that
`
`patients with PNETs are treated by medical oncologists. (Ex. 1070, Kulke Dep.
`
`43:19-23.)
`
`2
`
`Ex. 1119-0004
`
`

`

`
`
`8.
`
`I have reviewed court submissions made by Novartis in the related
`
`litigation before the United States District Court for the District of Delaware
`
`involving the same ’224 patent. In that proceeding, Novartis, and Dr. Kulke,
`
`offered the opinion with respect to the ’224 patent that a POSA “would have had
`
`(1) a Ph.D. in biology, biochemistry, pharmaceutical sciences, molecular biology,
`
`cancer biology, or other biological sciences; and/or (2) a medical degree and
`
`experience conducting preclinical, clinical, and/or laboratory research relating to
`
`cancer of the neuroendocrine system, including PNETs.” (Ex. 1121, D.I. 68-1 at
`
`40 of 302, ¶ 147 (emphasis added).) Thus, before the District Court, Novartis and
`
`Dr. Kulke identify that a POSA would not necessarily have had “experience
`
`conducting preclinical, clinical, and/or laboratory research relating to cancer of the
`
`neuroendocrine system, including PNETs.” I agree with the opinion offered by Dr.
`
`Kulke in that proceeding that a POSA would not have necessarily had had such
`
`research experience, as this is consistent with Dr. Kulke’s opinions that the claims
`
`of the ’224 patent are not limited to a clinical trial setting and that patients with
`
`PNETs are treated by medical oncologists. (Ex. 1070, Kulke Dep. 43:19-23,
`
`173:3-9.)
`
`9.
`
`Novartis asserts that I do not qualify as a POSA under the definition it
`
`and Dr. Kulke proposed in this proceeding because I do not have “special expertise
`
`in PNET.” (Patent Owner Response (“POR”) 5-6.) Putting aside that Novartis’s
`
`3
`
`Ex. 1119-0005
`
`

`

`
`
`two proposed definitions are inconsistent, Novartis’s assertion is incorrect. My
`
`testimony at my deposition was that “I didn’t have any special expertise in PNET
`
`as distinguished from any other cancer.” (Ex. 2040 at 302:9-11 (emphasis
`
`added).) I am an expert in cancer drug development, early clinical trials, and
`
`preclinical analyses. (E.g., id. at 303:24-304:9.) Further, by November 2005, I
`
`had preclinical laboratory research experience relating to NETs, including in cell
`
`signaling pathways. (Id. 304:25-307:5.)
`
`10.
`
`In addition, by November 2005, I had treated patients with NETs,
`
`including PNETs, and had enrolled such patients in phase 1 clinical trials.
`
`11. Prior to November 2005, I was involved in the early clinical
`
`development of a different mTOR
`
`inhibitor, AP23573 (also known as
`
`ridaforolimus and deforolimus). (Ex. 1082, Desai 2004; Ex. 1083, Desai 2005.)
`
`As noted on the poster presentation, at least one neuroendocrine patient was
`
`enrolled in the study and treated with this rapamycin derivative. (Ex. 1108, Desai
`
`2005 poster at Phase I Trials – Patient Population.) Therefore, I have conducted
`
`clinical research related to NETs.
`
`12. Further, I have specifically published on the use of mTOR inhibitors
`
`in the treatment of NETs that included an analysis of multiple clinical studies of
`
`treatments for NETs. (Ex. 1106, O’Donnell & Ratain 2007.) Dr. Kulke
`
`acknowledges that expertise in NETs includes analysis of clinical data on the
`
`4
`
`Ex. 1119-0006
`
`

`

`
`
`treatment of NETs. (Ex. 1070, Kulke Dep. 51:8-18.) Thus, even under the
`
`arbitrary (and inconsistent) standard for NET expertise put forward by Novartis
`
`and Dr. Kulke in this proceeding, I certainly qualify as having research experience
`
`“relating to NETs.”
`
`13.
`
`I further note that Novartis submitted a declaration from Dr. Lebwohl,
`
`one of the co-inventors of the ’224 patent, as an “expert in the field of Oncology
`
`clinical development,
`
`including clinical
`
`studies of
`
`treating pancreatic
`
`neuroendocrine cancer,” during prosecution of the ’224 patent. (Ex. 1002 at page
`
`1005, 9/24/2013 Declaration Under 37 C.F.R. § 1.132 at page 1.) I have worked
`
`with Dr. Lebwohl and have known him for decades, as we attended both Harvard
`
`College (graduating the same year with the same major) and Yale Medical School
`
`together. To the extent that Novartis and Dr. Lebwohl have represented Dr.
`
`Lebwohl’s experience as reflecting an expert in PNETs, I would similarly qualify
`
`as an expert. I have searched PubMed for his publications to identify his
`
`experience in cancer drug development. (Ex. 1097, Lebwohl PubMed search
`
`results.) As of November 2005, I have been unable to identify a single paper that
`
`Dr. Lebwohl published regarding the treatment of PNETs or even NETs generally.
`
`His work focused on the investigation of new drugs in cancer, especially solid
`
`tumors and in particular breast cancer. (Id.) His first publication on NETs does
`
`not appear until 2010, well after the November 2005 filing date of the ’224 patent.
`
`5
`
`Ex. 1119-0007
`
`

`

`
`
`(Id. at Reference 20.) Furthermore, Dr. Lebwohl’s biography notes that he has
`
`“focused on the clinical development of new cancer drugs, first at Bristol Myers
`
`Squibb, where he lead [sic] the development of JM-216, an oral platinum, and
`
`ixabepilone, an epothilone, as well as other compounds. (Ex. 1079, Lebwohl
`
`Biography.) His biography also notes that “he was responsible for the
`
`development of Afinitor (RAD001) from phase 1 through five NDA/MAA
`
`approvals” and that “he also led the clinical development of PTK787, a VEGFR
`
`inhibitor, panobinostat, a deacetylase inhibitor, and PKC412, a FLT3 kinase
`
`inhibitor.” (Id.) Having been an investigator for both JM-216 and PTK787
`
`(vatalanib) with Dr. Lebwohl, we have interacted directly during his time both at
`
`Bristol Myers Squibb and Novartis. (Ex. 1081, DeMario 1999; Ex. 1111, Schilsky
`
`2008.)
`
`14. Dr. Lebwohl’s pharmaceutical industry experience “focused on the
`
`clinical development of new cancer drugs” parallels my academic experience. In
`
`this context of developing new cancer drugs, an oncologist is motivated to use
`
`specific drugs in new ways to treat one or more diseases. (E.g., Ex. 1004 at 11.)
`
`15. Thus, to the extent that Dr. Lebwohl has special expertise in clinical
`
`trial design for NETs, as he informed the Patent Office, I have similar expertise.
`
`16.
`
`I have also searched for information regarding the publications and
`
`experience of Dr. Peter Wayne Marks, the other co-inventor. (Ex. 1100, Marks
`
`6
`
`Ex. 1119-0008
`
`

`

`
`
`PubMed Search Results.) Dr. Marks is currently the Director, Center for Biologics
`
`Evaluation and Research for the U.S. Food and Drug Administration. (Ex. 1120,
`
`Marks FDA website.) According to the FDA website, Dr. Marks served as
`
`Clinical Director of Hematology at Brigham and Women’s Hospital in Boston,
`
`prior to working “for several years in the pharmaceutical industry on the clinical
`
`development of hematology and oncology products.” (Id.) After that, he moved to
`
`Yale University, and his responsibilities included leadership of the Adult Leukemia
`
`Service. (Id.) I have also searched PubMed for Dr. Marks’ publications and have
`
`not identified any publications on neuroendocrine cancers. (Ex. 1100, Marks
`
`PubMed Search Results.)
`
`IV. SCOPE OF DECLARATION AND SUMMARY OF OPINIONS
`I have been asked to respond to certain opinions offered by Dr. Kulke
`17.
`
`related to Novartis’s allegations that the claims of the ’224 patent are not obvious.
`
`18.
`
`I disagree with Dr. Kulke that a POSA would have considered NETs
`
`that failed to respond to cytotoxic chemotherapy “were more difficult to treat than
`
`chemotherapy-naïve patients” with molecularly targeted therapies, such as mTOR
`
`inhibitors or other growth factor signaling inhibitors. (POR 3, 13-15; Ex. 2041,
`
`Kulke Decl. ¶¶31, 58-65.) A POSA would not have considered the failure of
`
`PNETs to respond to cytotoxic chemotherapy agents to indicate that the tumor
`
`7
`
`Ex. 1119-0009
`
`

`

`
`
`would be less likely to respond to a molecularly targeted therapy, such as an
`
`mTOR inhibitor.
`
`19.
`
`I also disagree with Dr. Kulke that a POSA would not have
`
`considered the CA20948 cell line as a model for NETs. (Ex. 2041, Kulke Decl.
`
`¶¶131-152.) Dr. Kulke bases his opinion on the fact that CA20948 is a rat tumor
`
`cell line derived from an azaserine-induced acinar cell adenocarcinoma. (Id. at
`
`¶¶132-137.) But, in a paper he co-authored before his involvement in this
`
`litigation, Dr. Kulke identified a different rat tumor cell line derived from an
`
`azaserine-induced acinar cell adenocarcinoma as a “Neuroendocrine Tumor
`
`Model.” (Ex. 1089, Heidari 2013 at Title.) I agree with Dr. Kulke’s published
`
`opinion that such tumor cell lines can serve as useful models for NETs.
`
`20.
`
`I also disagree with Dr. Kulke that a POSA would have considered the
`
`disclosure of Oberg & Eriksson more relevant to developing a new treatment for
`
`advanced PNETs. (Ex. 2041, Kulke Decl. ¶118.) A POSA would have considered
`
`the teaching of the entire prior art, and in particular the art specifically directed to
`
`oncologists in developing new therapies for NETs.
`
`21.
`
`I also disagree with Dr. Kulke that as of November 2005, the role of
`
`mTOR inhibitors as anticancer agents was “uncertain.” (E.g., Ex. 2041, Kulke
`
`Decl. at §V.E, ¶¶71-78.) By that time, significant clinical data were available
`
`demonstrating the antitumor activity of mTOR inhibitors as a class, and the clinical
`
`8
`
`Ex. 1119-0010
`
`

`

`
`
`development of mTOR inhibitors, including everolimus and temsirolimus, had
`
`advanced to phase 2 and phase 3 studies. In light of the extensive studies and
`
`advanced clinical development efforts for mTOR inhibitors as a class, a POSA
`
`would not have considered this class to have an “uncertain” role as anticancer
`
`agents.
`
`22.
`
`I also disagree with Dr. Kulke that as of November 2005, a POSA
`
`would have considered rapamycin, everolimus, and temsirolimus to have any
`
`differences in pharmacological activities. (E.g., Ex. 2041, Kulke Decl. ¶¶153-155,
`
`221-223, 235-237.) Dr. Kulke does not cite any data indicating that these rapalogs
`
`had any such pharmacological differences. To the contrary, the prior art, Novartis,
`
`and the ’224 patent consistently identify all three compounds as a group with
`
`similar anticancer activity. (See §V.E, below.) Therefore, in my opinion, a POSA
`
`would have considered rapamycin and its derivatives to have similar biological
`
`activity.
`
`23.
`
`I also disagree with Dr. Kulke that PTEN and the mTOR pathway had
`
`not been implicated in the treatment of PNETs as of November 2005. (Ex. 2041,
`
`Kulke Decl. ¶¶85-99.) Oberg 2004 (Ex. 1027) and Duran (Ex. 1011) both
`
`specifically identify the use of mTOR inhibitors as treatments for these tumors. In
`
`addition, Duran identifies that the basis for their clinical study in NETs was based
`
`on the reported role of PTEN in NETs. (Ex. 1011, Duran at 3096Ab.) Therefore,
`
`9
`
`Ex. 1119-0011
`
`

`

`
`
`in my opinion, a POSA would have been well aware that PTEN and the mTOR
`
`pathway had been implicated in the treatment of PNETs.
`
`24.
`
`I also disagree with Dr. Kulke’s assertion that a POSA would not have
`
`had a reasonable expectation that everolimus would be therapeutically effective in
`
`advanced PNETs. Dr. Oberg, an undisputed expert in the treatment of NETs,
`
`identified rapamycin as a treatment for NETs and stated that “clinical trials . . . are
`
`planned.” (Ex. 1027, Oberg 2004 at 60.) A POSA would have credited the
`
`teaching of such a notable expert and have reasonably expected that mTOR
`
`inhibitors would be therapeutically effective. In addition, Duran identified that
`
`temsirolimus had antitumor activity in NETs. (Ex. 1011, Duran at 3096Ab.)
`
`Given the identity of mTOR inhibitors as a class, a POSA would have considered
`
`these data for temsirolimus to reasonably identify the antitumor activity of mTOR
`
`inhibitors in NETs. Therefore, in my opinion, the prior art would have provided a
`
`POSA a reasonable expectation that an mTOR inhibitor, like everolimus, would be
`
`therapeutically effective in advanced NETs, including PNETs.
`
`25. Finally, I disagree with Dr. Kulke that there is any evidence of
`
`unexpected results for everolimus in the treatment of advanced PNETs compared
`
`to the closest prior art. I further disagree that everolimus satisfied any long-felt but
`
`unmet need for treatments for advanced PNETs.
`
`10
`
`Ex. 1119-0012
`
`

`

`
`
`V. THE ’224 PATENT WOULD HAVE BEEN OBVIOUS TO A POSA IN
`NOVEMBER 2005
`26.
`
`I have reviewed Dr. Kulke’s declaration and disagree with the bases
`
`for several of the opinions he offers therein. Because I disagree with many of the
`
`opinions he offers and have identified several factual inaccuracies, Dr. Kulke’s
`
`opinions do not change my opinion that a POSA would have considered the claims
`
`of the ’224 patent to be obvious over the prior art in November 2005.
`
`A.
`
`PNETs That Failed Cytotoxic Chemotherapy Would Not Be
`Expected to Be “More Resistant” to Molecularly Targeted
`Therapies
`27. Dr. Kulke opines that a POSA would have expected PNETs that had
`
`failed cytotoxic chemotherapy would be “more resistant or aggressive disease[s].”
`
`(Ex. 2041, Kulke Decl. ¶¶58-65.) For the following reasons, I disagree that, as of
`
`November 2005, a POSA would have expected PNETs that had previously failed
`
`cytotoxic chemotherapy to be “more resistant” to a molecularly targeted therapy,
`
`such as an mTOR inhibitor.
`
`28.
`
`In support of his opinion, Dr. Kulke cites only four papers: Ex. 1023,
`
`Moertel; Ex. 2012, Delaunoit; Ex. 2011, Kouvaraki; and Ex. 2074, Ramanathan.
`
`(Ex. 2041, Kulke Decl. ¶¶ 58-65.) But, as Dr. Kulke acknowledged at his
`
`deposition, each of these papers identify only that patients who had previously
`
`failed cytotoxic chemotherapy had lower response rates to further cytotoxic
`
`11
`
`Ex. 1119-0013
`
`

`

`
`
`chemotherapy. (Ex. 1070, Kulke Dep. 103:19-106:25 (agreeing that all four papers
`
`relate only to cytotoxic chemotherapies).)
`
`29.
`
`I agree with Dr. Kulke that molecularly targeted therapies are not
`
`cytotoxic chemotherapies. (Ex. 1070, Kulke Dep. 15:16-25.) I also agree with Dr.
`
`Kulke that mTOR inhibitors are molecularly targeted therapies, thus distinct from
`
`cytotoxic chemotherapies. (Id. at 158:8-15.) Therefore, in my opinion a POSA
`
`would not expect a solid tumor, including a PNET, that failed cytotoxic
`
`chemotherapy to be resistant to or more difficult to treat with a molecularly
`
`targeted therapy, such as an mTOR inhibitor. And the fact that such tumors were
`
`more likely to be resistant to further cytotoxic chemotherapy is irrelevant to the
`
`claims of the ’224 patent.
`
`30. Notably, none of the four references on which Dr. Kulke relies
`
`includes any data demonstrating that treatment with any molecularly targeted
`
`therapy resulted in a different response rate depending on whether or not the tumor
`
`was chemotherapy-naïve.
`
`31. All patients with advanced PNETs
`
`treated with cytotoxic
`
`chemotherapy will fail and very few patients even have responses to cytotoxic
`
`chemotherapy, as Dr. Kulke published in the prior art. (See Ex. 1101, McCollum
`
`2004 at 485, concluding that cytotoxic chemotherapy “failed to demonstrate
`
`substantial antitumor activity in patients with advanced [PNETs]”) & id.
`
`12
`
`Ex. 1119-0014
`
`

`

`
`
`(describing cytotoxic chemotherapy as “palliative” and not curative).) A POSA
`
`would be familiar with the dogma that patients resistant to first-line cytotoxic
`
`chemotherapy are often resistant to multiple cytotoxic chemotherapy agents. (E.g.,
`
`Ex. 1092, Kaye 1995 at 6A-40S (“Clinical drug resistance is a major obstacle to
`
`successful [cytotoxic] chemotherapy for cancer. When it occurs, resistance to a
`
`wide range of agents is noted.”); Ex 1098, Malhotra 2003 at S3 (“The most
`
`common mechanism of drug resistance is related to altered gene expression. Cells
`
`in the G0 phase are generally resistant to all drugs active in the S phase.”).)
`
`32. A POSA would therefore be motivated to avoid additional cytotoxic
`
`chemotherapy in these chemotherapy-resistant patients and instead seek an
`
`anticancer agent with a completely different mechanism of action, such as a
`
`molecularly targeted agent. Indeed, Dr. Kulke offered precisely this opinion in the
`
`litigation related to Sutent (sunitinib), which is indicated for advanced PNETs. In
`
`that litigation, Dr. Kulke testified that when a PNET becomes refractory to a
`
`treatment, including a cytotoxic chemotherapy drug, “you have to look for another
`
`type of therapy.” (Ex. 1095, Kulke Sutent Testimony at 798:11-21 (emphasis
`
`added).) I agree with the opinion that Dr. Kulke offered in the Sutent litigation.
`
`33. Consistent with this opinion, the prior art included multiple examples
`
`of molecularly targeted therapies having antitumor activity after failure of
`
`cytotoxic chemotherapy. For example, the prior art reported that temsirolimus had
`
`13
`
`Ex. 1119-0015
`
`

`

`
`
`“substantial antitumor activity in relapsed MCL [mantle cell lymphoma],”
`
`indicating that an mTOR inhibitor had demonstrated activity in tumors that
`
`previously resisted cytotoxic chemotherapy. (Ex. 1118, Witzig Aug. 2005 at
`
`5347.) Other molecularly targeted therapies had been shown to have antitumor
`
`activity
`
`in patients who had progressed after
`
`treatment with cytotoxic
`
`chemotherapy. (E.g., Ex. 1078, Cohen 2003 at Table 3 (describing EGFR inhibitor
`
`having antitumor activity after failure of first-line cytotoxic chemotherapy); Ex.
`
`1116, Burris 2004 at 14 (“Clinical responses were observed at a variety of doses
`
`[of lapatinib, an EGFR/ErbB-2 receptor inhibitor] in these heavily pretreated
`
`patients with metastatic disease.”).)
`
`34. Additionally, Dr. Kulke indicates that streptozocin and chlorozotocin
`
`“ha[ve] different mechanisms of action,” in support of his opinion. (Ex. 2041,
`
`Kulke Decl. ¶61.) But a POSA would not have understood streptozocin and
`
`chlorozotocin to have different mechanisms. Chlorozotocin is an analog of
`
`streptozocin, and both are nitrosoureas. (Ex. 1091, Johnston 1975 (describing
`
`chlorozotocin as a synthetic analog of streptozocin).) A POSA would have
`
`expected any solid tumor that had been previously treated with a nitrosourea,
`
`including advanced PNETs, to fail to respond to another nitrosourea, as the prior
`
`art explicitly identified. (Ex. 1096, Kvols 1987 at 81 (identifying that “[n]o
`
`responses were observed among the four previously treated patients, suggesting a
`
`14
`
`Ex. 1119-0016
`
`

`

`
`
`similar mechanism of action and cross resistance [of chlorozotocin]
`
`to
`
`streptozocin.”).)
`
`
`
`Chlorozotocin
`
`Streptozotocin
`
`
`
`
`
`
`
`
`
`
`
`35. Next, Dr. Kulke relies on Delaunoit to indicate that a POSA would
`
`expect failure of prior cytotoxic chemotherapy to be “significantly associated with
`
`worse outcomes,” relying on the lower median survival times for patients
`
`previously treated with cytotoxic chemotherapy. (Ex. 2041, Kulke Decl. ¶62,
`
`citing Ex. 2012.) Delaunoit reports shorter overall survival times for PNETs
`
`previously treated with cytotoxic chemotherapy, but that at most that would
`
`suggest that these tumors were not responsive to cytotoxic chemotherapy, which
`
`would not have lead a POSA to expect that such patients would not benefit from a
`
`molecularly targeted agent, such as sunitinib or everolimus.
`
`36. For
`
`the same reason,
`
`the reports of failure of second-line
`
`chemotherapies in Kouvaraki and Ramanathan, cited by Dr. Kulke, would not have
`
`lead a POSA to expect that PNETs after failure of cytotoxic chemotherapy would
`
`15
`
`Ex. 1119-0017
`
`

`

`
`
`be more challenging to treat with a molecularly targeted therapy. (Ex. 2041, Kulke
`
`Decl. ¶¶63-64, citing Exs. 2011 & 2074.) In fact, Ramanathan’s report that
`
`dacarbazine has little activity in PNETs previously treated with streptozocin (Ex.
`
`2074 at 1141) is not surprising as dacarbazine and streptozocin are both alkylating
`
`agents and thus have similar mechanisms of action, as Dr. Kulke has stated in his
`
`publications. (Ex. 1076, Chan & Kulke 2014 at 375, (identifying “streptozocin and
`
`other alkylating agents” and citing references 60 (Ex. 1023, Moertel) and 61 (Ex.
`
`2011, Kouvaraki)).) As such, these references do not support Dr. Kulke’s opinion
`
`that a POSA would have expected that failure to respond to an alkylating agent
`
`would result in a tumor more resistant to a therapy with a completely different
`
`mechanism of action, such as an mTOR inhibitor.
`
`37. Compelling contemporaneous evidence shows that investigators (e.g.,
`
`medical oncologists) understood
`
`that previous
`
`treatment with cytotoxic
`
`chemotherapy was not likely to influence response to treatment with molecularly
`
`targeted therapies. As of November 2005, clinical trials investigating molecularly
`
`targeted therapies allowed patients with NETs, including PNETs, to enroll who had
`
`previously been treated with cytotoxic chemotherapy. (Ex. 1011, Duran at 3096Ab
`
`(allowing prior cytotoxic chemotherapy for protocol administering mTOR
`
`inhibitor); Ex. 2049, Carr (allowing prior cytotoxic chemotherapy for protocol
`
`administering a kinase inhibitor, imatinib, a molecularly targeted therapy); Ex.
`
`16
`
`Ex. 1119-0018
`
`

`

`
`
`2099, Kulke (allowing prior cytotoxic chemotherapy for protocol administering
`
`sunitinib, a vascular endothelial growth factor receptor inhibitor, a molecularly
`
`targeted therapy (Ex. 1070, Kulke Dep. 15:11-15)); Ex. 1112, Shah 2004 (allowing
`
`prior cytotoxic chemotherapy for a protocol administering bortezomib, a
`
`proteasome inhibitor, a molecularly targeted therapy); Ex. 1090, Hobday 2005
`
`(allowing prior cytotoxic chemotherapy for protocol administering gefitinib, an
`
`epidermal growth factor receptor inhibitor, a molecularly targeted therapy).) Such
`
`inclusion criteria indicate that these investigators studying NETs did not consider
`
`such previously treated tumors to be unlikely to benefit from molecularly targeted
`
`therapies. To the contrary, inclusion of these patients indicates that the
`
`investigators considered those patients to be just as likely to respond to
`
`molecularly targeted therapies as those who had not previously undergone
`
`cytotoxic chemotherapy, particularly in the absence of a preplanned analysis of the
`
`impact of prior cytotoxic chemotherapy.
`
`38.
`
`If such prior treatment had been considered to be an important
`
`consideration, scientific standards would have mandated that such patients be
`
`excluded from the clinical trial. Indeed, Dr. Kulke excluded NET patients with
`
`prior cytotoxic chemotherapy (dacarbazine) in his study of a combination
`
`treatment that included a different cytotoxic chemotherapy agent (temozolomide).
`
`This exclusion criterion demonstrates that the investigators, including Dr. Kulke,
`
`17
`
`Ex. 1119-0019
`
`

`

`
`
`believed that including such previously treated patients would affect the response
`
`to the treatment protocol that included a second chemotherapy agent with a similar
`
`mechanism. (Ex. 1068, Chan 2012 at 2964.) This contrasts with the Duran (Ex.
`
`1011), Carr (Ex. 2049), Kulke 2005 (Ex. 2099), Shah 2004 (Ex. 1112), and
`
`Hobday 2005 (Ex. 1090) studies referenced in paragraph 37 above, where the
`
`inclusion of such previously-treated patients demonstrates that those investigators
`
`believed previous treatment would not affect the outcome.
`
`39.
`
`In addition, if a POSA had expected differences in responses between
`
`prior treated tumors and chemotherapy-naïve tumors, a POSA would have
`
`expected that such differences would be reported in the analysis of clinical results.
`
`Dr. Kulke did not identify any analyses of clinical study results for the treatment of
`
`NETs or PNETs that indicated a difference in response between previously treated
`
`tumors and chemotherapy-naïve
`
`tumors when administered non-cytotoxic
`
`chemotherapies. Indeed, even Dr. Kulke’s own publications from clinical studies
`
`conducted as of November 2005 include no such analysis, corroborating my
`
`opinion that as of November 2005 a POSA would not have expected PNETs that
`
`had previously failed cytotoxic chemotherapy to be resistant to molecularly
`
`targeted therapies. (Ex. 2099, Kulke 2005 (reporting no difference in prior
`
`treatment); Ex. 1066, Kulke 2008 (reporting no difference in prior treatment for a
`
`study enrolling patients before November 2005); Ex. 1068, Chan 2012 (same).)
`
`18
`
`Ex. 1119-0020
`
`

`

`
`
`40. Therefore,
`
`the activities of medical oncologists
`
`(and other
`
`investigators) and the published clinical trial results, as of November 2005, support
`
`a conclusion that a POSA would not have expected previously treated PNETs to
`
`respond differently from chemotherapy-naïve PNETs to a molecularly targeted
`
`therapy, such as an mTOR inhibitor.
`
`41. The results of the RADIANT-3 study further corroborate my opinion
`
`that a POSA would not have expected prior cytotoxic chemotherapy to affect the
`
`antitumor activity of a molecularly targeted therapy. Dr. Kulke cites Ex. 2022,
`
`Yao 2011 to identify that the results of everolimus treatment in patients with and
`
`without prior cytotoxic chemotherapy “are comparable.” (Ex. 2041, Kulke Decl.
`
`¶262.) Notably, Yao 2011 does not include any discussion that such “comparable”
`
`results were unexpected or surprising. (Ex. 2022, Yao 2011 at Figure 1C.) If a
`
`POSA in November 2005 truly would have expected some difference between the
`
`activity in PNETs with and without previous cytotoxic chemotherapy, it would be
`
`appropriate for Yao, reporting the only direct comparison of previously treated and
`
`chemotherapy-naïve tumors, to comment that no such difference was observed in
`
`contrast to what would have been expected at the time.
`
`B.
`Preclinical Models for NETs
`42. Dr. Kulke has opined that a POSA would not have considered
`
`CA20948 a preclinical model for NETs. (E.g., Ex. 2041, Kulke Decl. ¶¶131-152.)
`
`19
`
`Ex. 1119-0021
`
`

`

`
`
`43. Dr. Kulke admits that NETs, including PNETs, are somatostatin
`
`receptor-positive tumors and that CA20948 is a model of somatostatin receptor-
`
`positive tumors. (Ex. 2041, Kulke Decl. ¶144; Ex. 1070, Kulke Dep. 149:23-
`
`150:2.)
`
`44. Dr. Kulke identifies three main reasons to support his opinion that a
`
`POSA would not have considered CA20948 as a model for NETs in humans:
`
`(1) CA20948 derives from a rat tumor; (2) CA20948 derives from an acinar tumor
`
`of the exocrine pancreas; and (3) CA20948 derives from an azaserine-induced
`
`tumor. (Ex. 2041, Kulke Decl. ¶¶132-136; Ex. 1070, Kulke Dep. 62:21-63:14).
`
`But Dr. Kulke’s opinion is contradicted by his own publication regarding another
`
`preclinical model for NETs. Dr. Kulke co-authored a paper in 2013 in which he
`
`identifies the rat tumor cell line AR42J as a “Neuroendocrine Tumor Model.” (Ex.
`
`1089, Heidari at Title (“Neuroendocrine Tumor Model”) & 6866 (describing use of
`
`AR42J “rat pancreatic carcinoma”).) The AR42J cell line is a rat tumor, derived
`
`from an acinar tumor of the exocrine pancreas that was induced by azaserine—
`
`precisely the same characteristics underlying Dr. Kulke’s opinions in this
`
`proceeding. (Ex. 1053, Weckbecker at 15.) Thus, in his own work predating this
`
`proceeding, Dr. Kulke apparen