UNITED STATES PATENT AND TRADEMARK OFFICE
`_______________________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`_______________________
`
`TEVA PHARMACEUTICALS USA, INC.,
`Petitioner
`
`v.
`
`CORCEPT THERAPEUTICS, INC.,
`Patent Owner
`_______________________
`
`Case PGR2019-00048
`U.S. Patent No. 10,195,214
`_______________________
`
`DECLARATION OF F. PETER GUENGERICH, Ph.D.
`
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`Corcept Ex. 2056, Page 1
`
`

`

`V.
`VI.
`
`C.
`
`I.
`II.
`III.
`IV.
`
`TABLE OF CONTENTS
`INTRODUCTION ...........................................................................................1
`QUALIFICATIONS AND MATERIALS RELIED UPON...........................1
`SUMMARY OF OPINIONS...........................................................................6
`TECHNICAL BACKGROUND .....................................................................6
`A.
`Drug-Drug Interactions .........................................................................6
`B. Mifepristone ..........................................................................................8
`1.
`Mifepristone is a CYP3A Substrate............................................8
`2.
`Mifepristone has Complicated Pharmacokinetics ....................10
`LEGAL STANDARDS .................................................................................11
`THE CLAIMS OF THE ’214 PATENT WOULD NOT HAVE BEEN
`OBVIOUS......................................................................................................13
`A.
`Dr. Greenblatt’s Definition of a POSA...............................................13
`B.
`Dr. Greenblatt’s Has Not Established That A POSA Would Have A
`Reasonable Expectation That The Claimed Methods Would Be Safe
`and Effective........................................................................................14
`If Anything a POSA Would Have Expected There to be a Significant
`Drug-Drug Interaction.........................................................................21
`1.
`A POSA Would Have Expected Mifepristone to Act As a
`Sensitive CYP3A Substrate That Would Be Significantly
`Affected By a Strong CYP3A Inhibitor....................................21
`The Only Available Clinical Data Indicated a Clinically
`Significant DDI.........................................................................26
`A POSA Would Expect The Ratio Of Ketoconazole’s
`Concentration To Its Inhibition Constant To Be Greater Than
`10, Indicating a High Probability of a Clinically Significant
`Interaction with Mifepristone ...................................................29
`Corcept’s DDI Study Was Not Routine..............................................32
`D.
`The Claimed Inventions Demonstrate Unexpected Results................33
`E.
`VII. CONCLUSION..............................................................................................36
`
`2.
`
`3.
`
`i
`
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`Corcept Ex. 2056, Page 2
`
`

`

`I, F. Peter Guengerich, hereby declare and state as follows:
`
`I submit this declaration on behalf of Corcept Therapeutics, Inc. (“Corcept”
`
`or “Patent Owner”), the owner of U.S. Patent No. 10,195,214 (“the ʼ214 Patent”),
`
`in connection with the Petition for Post-Grant Review filed by Teva
`
`Pharmaceuticals USA, Inc. (“Teva” or “Petitioner”).
`
`I.
`
`INTRODUCTION
`
`1.
`
`I have been asked to review and respond to certain of the opinions set
`
`forth in the Declaration of Dr. David J. Greenblatt, M.D., submitted on behalf of
`
`Petitioner.
`
`II. QUALIFICATIONS AND MATERIALS RELIED UPON
`
`2.
`
`I am the Tadashi Inagami Professor of Biochemistry in the
`
`Department of Biochemistry at the Vanderbilt University School of Medicine.
`
`3.
`
`I received a B.S. in Agricultural Science from the University of
`
`Illinois, Urbana in 1970. I then obtained my Ph.D. in Biochemistry from
`
`Vanderbilt University in 1973 under the guidance of Professor H.P. Broquist.
`
`Following that, I was a Postdoctoral Scholar in the laboratory of Professor M.J.
`
`Coon in the Department of Biological Chemistry at the University of Michigan
`
`Medical School.
`
`4.
`
`Subsequent to my postdoctoral work, in 1975 I started as an Assistant
`
`Professor of Biochemistry at the Vanderbilt University School of Medicine. In
`
`1
`
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`Corcept Ex. 2056, Page 3
`
`

`

`1980 I was named a tenured Associate Professor of Biochemistry at the Vanderbilt
`
`University School of Medicine, and in 1983 I became a (tenured) Professor of
`
`Biochemistry at the Vanderbilt University School of Medicine. Since that time, I
`
`have held several positions at Vanderbilt University School of Medicine,
`
`including: Director, Center in Molecular Toxicology (1981-2011), Harry Pearson
`
`Broquist Professor of Biochemistry (2007-2012), Interim Chairman, Department of
`
`Biochemistry (2010-2012), Stanford Moore Professor of Biochemistry (2013), and
`
`my current position as the Tadashi Inagami Professor of Biochemistry (2013-
`
`present).
`
`5.
`
`I have decades of experience studying and educating others about
`
`drug-drug interactions (“DDI”), including DDIs involving CYP3A inhibitors.
`
`Throughout my time at Vanderbilt, I have taught courses on multiple aspects of
`
`drug-drug interactions and pharmacokinetics to medical students, graduate
`
`students, and post-graduates. I am currently teaching Enzyme Kinetics and
`
`Mechanisms. I am also currently teaching Drug Metabolism & Safety, part of a
`
`Master’s degree class for postgraduate physicians. I teach pharmacokinetics and
`
`toxicokinetics in my Biochemical Toxicology class. In addition, I have created an
`
`online course for the Pharmacology Department dealing with pharmacokinetics of
`
`drug-drug interactions and called Enzyme Kinetics for Drug Discovery &
`
`Development.
`
`2
`
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`Corcept Ex. 2056, Page 4
`
`

`

`6.
`
`I have extensive experience in the fields of biochemistry and
`
`medicinal chemistry with an emphasis on mechanisms of activation and
`
`detoxication of drugs, chemical carcinogens, steroids, and toxicants and
`
`characterization of enzymes involved in these processes.
`
`7.
`
`During my career I have received numerous honors and distinguished
`
`lectureships, which are summarized in my curriculum vitae, which is attached as
`
`Appendix A.
`
`8.
`
`I have published more than 700 original peer-reviewed scientific
`
`articles and more than 270 invited reviews and chapters during my career.
`
`9.
`
`I currently serve on the Editorial boards for Chemistry and
`
`Biodiversity, Critical Reviews in Toxicology, and Drug Metabolism and
`
`Disposition. I was previously on the editorial advisory board for Nature Reviews
`
`in Drug Discovery. I also served as an Associate Editor of both the journals
`
`Molecular Pharmacology and Chemical Research in Toxicology, and since 2013, I
`
`have served as Deputy Editor of The Journal of Biological Chemistry.
`
`10.
`
`I am a member of the American Chemical Society, including the
`
`Divisions of Biological Chemistry, Medicinal Chemistry, and Chemical
`
`Toxicology. I served as Chair of the latter Division from 2007-2008 and have held
`
`several other offices in the Division of Chemical Toxicology. In 2009, I was
`
`named as an American Chemical Society Fellow, in the inaugural class for that
`
`3
`
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`Corcept Ex. 2056, Page 5
`
`

`

`program. I was a member of the 2019 inaugural class of Fellows of the American
`
`Society for Pharmacology and Experimental Therapeutics, and I received the B.B.
`
`Brodie Award for Achievement in Drug Metabolism from that society in 1992.
`
`Since 1997, I have chaired the International Advisory Committee for the biennial
`
`International Conferences on Cytochrome P450.
`
`11.
`
`I have served as a consultant for a number of companies including
`
`Merck, Sandoz (now Novartis), Schering-Plough (now part of Merck), Oxford
`
`Biomedical Research, Otsuka Pharmaceuticals, Orchid Biosciences, Targacept,
`
`Aventis, Pfizer, OSI, Novartis, Johnson & Johnson, Teva, Salix, Vertex, Biogen
`
`Idec, Barbeau Pharma/ProCor, BioPure, Resverologix, Medivation, GeminX,
`
`Angion Biomedica, Transcept, Purdue Pharma LP, and ZioPharm Oncology.
`
`Since 2015, I have been a member of the Expert Panel of the Flavor and Extract
`
`Manufacturers’ Association.
`
`12.
`
`I have been working on the important hepatic cytochrome P450
`
`proteins since 1973. Under my direction, my laboratory developed the technology
`
`for purifying several P450 proteins from human liver, including those now known
`
`as CYP 1A1, 1A2, 2C8, 2C9, 2D6, 2E1, and 3A4. My laboratory also developed
`
`useful methods for the heterologous expression of these and other human
`
`cytochrome P450 enzymes in bacteria and their purification and use in assays of
`
`the metabolism of drugs and other chemicals.
`
`4
`
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`Corcept Ex. 2056, Page 6
`
`

`

`13.
`
`I have testified several times as an expert witness since 1995,
`
`including for the following companies: Marion Merrill Dow, Pfizer, Merck, Glaxo-
`
`GSK, NCR, Amgen, King, Vanda, and Incyte.
`
`14. A copy of my curriculum vitae, including a list of publications I have
`
`authored, is attached to this declaration as Appendix A. A list of all other cases
`
`during the previous four years in which I testified as an expert at trial or by
`
`deposition is attached hereto as Appendix B.
`
`15. An identification of the material I have reviewed and/or relied upon is
`
`set forth in Appendix C thereto.
`
`16.
`
`I have further drawn on my experience in pharmacokinetics and
`
`pharmacodynamics. Each of the statements below reflects my opinion.
`
`17.
`
`I expressly reserve the right to supplement the opinions expressed
`
`herein, as well as the bases for my opinions, in response to additional expert
`
`declarations submitted by Teva, or any additional discovery or other information
`
`provided in this matter.
`
`18.
`
`I am being compensated for my time at my usual rate of $650 per
`
`hour. My compensation does not depend in any way on the substance of my
`
`testimony or the outcome of this or any other case.
`
`5
`
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`Corcept Ex. 2056, Page 7
`
`

`

`III.
`
`SUMMARY OF OPINIONS
`
`19.
`
`I have reviewed the ’214 patent (Ex. 1001), the Greenblatt Declaration
`
`(Ex. 1002), and other materials cited herein. I have been asked by counsel to use
`
`March 1, 2017 as the date of invention for the claims in the ʼ214 patent. In my
`
`opinion, Dr. Greenblatt has failed to establish that the claims of the ’214 patent
`
`would have been obvious over the references that he cites and/or general
`
`knowledge in the field, as of March 1, 2017.
`
`20.
`
`It is my opinion that as of the date of invention, a person of ordinary
`
`skill in the art (“POSA”) would not have had a reasonable expectation of
`
`successfully performing the methods claimed in the ’214 patent. In particular, a
`
`POSA would not have reasonably expected that 600 mg of mifepristone could be
`
`safely and effectively administered to patients with Cushing’s syndrome
`
`concomitantly with a strong CYP3A inhibitor.
`
`IV. TECHNICAL BACKGROUND
`
`A.
`
`Drug-Drug Interactions
`
`21. When two or more medications are administered at the same time,
`
`there is a risk that the drugs will interact with one another and alter the effects of
`
`one or both drugs on the patient. A DDI can take many forms. One example is a
`
`pharmacokinetic DDI, in which one drug (the inhibitor) inhibits the metabolic
`
`pathway or enzyme responsible for metabolizing a second drug (the substrate).
`
`6
`
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`Corcept Ex. 2056, Page 8
`
`

`

`Pharmacokinetic DDIs can cause a substantial increase in the blood and tissue
`
`concentrations of substrates and/or their metabolites, because the pathway
`
`responsible for metabolizing the substrate is inhibited. These increased
`
`concentrations can cause harmful side effects.
`
`22.
`
`The cytochrome P450 (CYP) family of enzymes is the major group of
`
`enzymes involved in drug metabolism. CYP3A is one member of the CYP family.
`
`Substances that are metabolized by CYP3A are called CYP3A substrates.
`
`23. Drugs metabolized by CYP3A enzymes have been shown to be
`
`susceptible to serious pharmacokinetic DDIs. The FDA has identified numerous
`
`CYP3A inhibitors and classified them as strong, moderate, and weak. Ex. 2046 at
`
`8-9. The FDA categorizes drugs that increase the “area-under-the-curve” (AUC)
`
`(i.e., exposure) of sensitive CYP3A index substrates ≥5-fold as strong CYP3A
`
`inhibitors. Id. Ketoconazole, itraconazole, and clarithromycin have been
`
`identified as strong CYP3A inhibitors. Id. In fact, ketoconazole, along with
`
`ritonavir, are the strongest inhibitors of CYP3A. Strong CYP3A inhibitors have
`
`been shown to significantly increase exposure to CYP3A substrates. For example,
`
`co-administration of ketoconazole and oral midazolam increased the AUC of
`
`midazolam concentration 16-fold (Ex. 1007 at 465) and co-administration of
`
`ketoconazole and triazolam increased triazolam exposure nearly 14-fold. Ex. 1026
`
`at 367.
`
`7
`
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`Corcept Ex. 2056, Page 9
`
`

`

`24. Co-administration of CYP3A inhibitors and substrates without
`
`accounting for pharmacokinetic DDIs has resulted in serious side effects, including
`
`death. For example, the CYP3A substrate terfenadine, which was approved by the
`
`FDA in 1985 and marketed under the name Seldane®, was found to cause
`
`potentially fatal heart problems, including cardiac arrhythmias due to QT
`
`prolongation, when given in combination with the CYP3A inhibitors ketoconazole
`
`and erythromycin. The combination of terfenadine and strong CYP3A inhibitors
`
`has been blamed for numerous deaths. Similarly, grapefruit juice (a moderate
`
`CYP3A inhibitor) has been shown to cause serious adverse effects when co-
`
`administered with certain CYP3A substrates and the package inserts and
`
`instructions for many of these drugs now carry warnings. Ex. 2064.
`
`B. Mifepristone
`
`25. Mifepristone (or “RU 486”) was first synthesized in 1981 in France.
`
`It was discovered to be an effective progesterone receptor antagonist. Many of the
`
`early mifepristone studies were focused on this mechanism, including the use of
`
`mifepristone as an abortifacient. In addition to its antiprogestin effect, it was also
`
`discovered that mifepristone acts as an antiglucocorticoid.
`
`1. Mifepristone is a CYP3A Substrate
`
`26.
`
`Following oral administration, mifepristone has an approximately
`
`70% absorption rate from the gut. Ex. 2050 at 113. Mifepristone also “undergoes
`
`8
`
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`Corcept Ex. 2056, Page 10
`
`

`

`[a] first-pass effect in the liver and gets partially metabolized and eventually its
`
`bioavailability is reduced to 40% in human beings.” Id. Mifepristone is
`
`metabolized into three active metabolites: mono-demethylated, hydroxylated, and
`
`di-demethylated. See, e.g., Ex. 1022 at FIG. 1; Ex. 1004 at 13. The metabolism of
`
`mifepristone into its three active metabolites is depicted in the following scheme:
`
`Ex. 1022 at Fig. 1.
`
`27.
`
`In 1996, Jang determined that mifepristone is metabolized by CYP3A.
`
`Id. at Abstract (“[D]ata clearly support CYP3A4 as the enzyme primarily
`
`responsible for mifepristone demethylation and hydroxylation in human liver
`
`microsomes.”). In fact, as Dr. Greenblatt states in his declaration, metabolism of
`
`mifepristone “is determined almost exclusively by the action of CYP3A.” Ex.
`
`1002 at ¶ 27. Jang also disclosed that “[c]hemical and immuno-inhibition of
`
`CYP3A4 resulted in significant inhibition of mifepristone metabolism, which was
`9
`
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`Corcept Ex. 2056, Page 11
`
`

`

`further confirmed through correlation analyses.” Ex. 1022 at 758 (emphasis
`
`added). In addition, Jang determined that CYP3A is responsible for the
`
`metabolism of the mono-demethylated metabolite of mifepristone. Id.
`
`2. Mifepristone has Complicated Pharmacokinetics
`
`28. Mifepristone exhibits complicated pharmacokinetics. First,
`
`pharmacokinetic studies done before the time of invention uniformly reported that
`
`mifepristone’s pharmacokinetic profile was non-linear. Following administration,
`
`mifepristone binds primarily with α1-acid glycoprotein (AAG). But at doses in the
`therapeutic range, AAG is rapidly saturated. Ex. 1004 at 13, Ex. 2013 at
`
`423. Upon AAG saturation, mifepristone levels in the blood do not increase
`
`proportionally with an increase in dose. Ex. 2013 at Abstract; see also Ex. 1004 at
`
`13 (“Exposure to mifepristone is substantially less than dose proportional.”); Ex.
`
`2060 at 80 (“Interestingly, within the dose range of 100-800 mg either the peak or
`
`steady state serum levels of RU 486 did not correlate with the dose ingested.”).
`
`29.
`
`Second, mifepristone exhibits highly variable inter-patient
`
`pharmacokinetics. For example, mifepristone blood levels vary significantly
`
`between patients administered the same dose. Ex. 2061 at 137 (disclosing a “very
`
`marked between-subject variation in the plasma concentrations after the same dose
`
`of RU 486.”). In addition, mifepristone exhibits variable inter-patient half-life.
`
`The Korlym label reports that the mean half-life of mifepristone following multiple
`
`10
`
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`Corcept Ex. 2056, Page 12
`
`

`

`doses is 85 hours with a very high standard deviation of 61 hours. Ex. 1004 at 13;
`
`see also Ex. 1005 at 26; Ex. 2050 at 115.
`
`30.
`
`Third, mifepristone’s metabolites “are believed to be
`
`pharmacologically active.” Ex. 1005 at 21. The Korlym label discloses that
`
`mifepristone’s “three active metabolites have greater affinity for the glucocorticoid
`
`receptor … 61%, 48%, and 45% … than either dexamethasone (23%) or cortisol
`
`(9%).” Ex. 1004 at 12. In addition, mifepristone’s active metabolites “are
`
`produced to [a] substantial extent.” Ex. 1005 at 21. Lee (Ex. 1005) discloses that
`
`the AUC of the mono-demethylated metabolite exceeds the AUC of mifepristone
`
`following a 600 mg dose of mifepristone. Id. The other two metabolites also
`
`achieve significant levels. Id. at 21-22.
`
`V.
`
`LEGAL STANDARDS
`
`31.
`
`I am not an attorney and, consequently, I offer no opinion on the law
`
`itself. My understanding of the pertinent law is outlined in this section and is the
`
`result of explanations provided by counsel. I have applied this understanding in my
`
`analysis.
`
`32.
`
`33.
`
`I understand that patents are presumed valid.
`
`I understand that a patent claim is invalid if it is obvious in view of
`
`the prior art. I further understand that the frame of reference for determining
`
`11
`
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`Corcept Ex. 2056, Page 13
`
`

`

`whether a patent is obvious is from the perspective of a POSA at the time the
`
`invention was made (in this case, the priority date of March 1, 2017).
`
`34.
`
`In analyzing obviousness in light of the prior art, I understand that it is
`
`important to understand the scope of the claims, the level of skill in the relevant
`
`art, the scope and content of the prior art, the differences between the prior art and
`
`the claimed invention, and any objective indicia of non-obviousness (also called
`
`secondary considerations of non-obviousness).
`
`35.
`
`I understand that a patent claim is obvious over multiple, combined
`
`references only if the prior art as a whole taught or suggested the invention, if there
`
`was a motivation or reason to combine the teachings of the prior art references to
`
`achieve the claimed invention, and if a POSA would have had a reasonable
`
`expectation of success in doing so.
`
`36.
`
`I also understand that the use of hindsight must be avoided in
`
`determining whether an invention would have been obvious because the
`
`obviousness of an invention is evaluated from the perspective of a POSA at the
`
`time the invention was conceived. In other words, it is improper to use the
`
`invention itself as a blueprint to assemble the prior art into the claimed inventions.
`
`37. As stated above, I understand that the obviousness inquiry also takes
`
`into account objective indicia of non-obviousness, or secondary considerations of
`
`non-obviousness. I understand that commonly recognized objective indicia of non-
`
`12
`
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`Corcept Ex. 2056, Page 14
`
`

`

`obviousness include: the failure of others to solve the problem addressed by the
`
`invention, the unexpected or surprising results of the invention, the existence of a
`
`long-felt but unmet need for the invention, the commercial success of the
`
`invention, and the copying of the invention by others.
`
`VI. THE CLAIMS OF THE ’214 PATENT
`WOULD NOT HAVE BEEN OBVIOUS
`
`38. Dr. Greenblatt opines that “claims 1-13 of the ’214 patent would have
`
`been obvious to a POSA over the combination of the Korlym Label and Lee” and
`
`“over the combination of the Korlym Label, Lee, and FDA Guidance.” Ex. 1002
`
`at ¶¶ 57, 117. I disagree. Having reviewed the relevant art, in my opinion, the
`
`claims of the ’214 patent would not have been obvious to a POSA.
`
`A.
`
`Dr. Greenblatt’s Definition of a POSA
`
`39. Dr. Greenblatt opines that “a POSA in the field of the ’214 patent
`
`would have had an M.D., a Pharm. D., and/or a Ph.D. in pharmacology or a related
`
`discipline, with at least four years of experience in treating patients with
`
`mifepristone and/or CYP3A inhibitors, or, alternatively, studying drug-drug
`
`interactions involving CYP3A inhibitors.” Id. at ¶ 18. He further states that the
`
`“POSA may have worked as part of a multidisciplinary team and drawn upon not
`
`only his or her own skills, but also taken advantage of certain specialized skills of
`
`others on the team.” Id. I have been asked to provide my opinions in this case
`
`from the perspective of Dr. Greenblatt’s POSA. I do not offer any opinion on the
`
`13
`
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`Corcept Ex. 2056, Page 15
`
`

`

`propriety of Dr. Greenblatt’s definition. I note that I am a POSA under his
`
`standard and have been for some time.
`
`B.
`
`Dr. Greenblatt’s Has Not Established That A POSA Would Have
`A Reasonable Expectation That The Claimed Methods Would Be
`Safe and Effective
`
`40.
`
`I understand that in order for the claims to be obvious, a POSA must
`
`have had a reasonable expectation of success, i.e., a reasonable expectation that
`
`600 mg of mifepristone daily in combination with a strong CYP3A inhibitor would
`
`be safe and effective in patients with Cushing’s syndrome. Nothing in Dr.
`
`Greenblatt’s declaration establishes that a POSA would have had a reasonable
`
`expectation of success.
`
`41. Dr. Greenblatt admits that the Korlym label and Lee “do not disclose
`
`the specific dose adjustment from 900 or 1200 mg mifepristone to 600 mg
`
`mifepristone when mifepristone is co-administered with a strong CYP3A
`
`inhibitor.” Ex. 1002 at ¶ 68. Nevertheless, Dr. Greenblatt asserts that because “the
`
`label expressly permits doses of up to 1200 mg per day,” it was “reasonably likely
`
`that 600 mg [of mifepristone] would be well tolerated and therapeutically effective
`
`when co-administered with a strong CYP3A inhibitor.” Id. at ¶ 69. Dr.
`
`Greenblatt’s opinion is inconsistent with other portions of his own declaration and
`
`with the understanding of a POSA at the time of the invention.
`
`14
`
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`Corcept Ex. 2056, Page 16
`
`

`

`42. Dr. Greenblatt repeatedly opines that a POSA would not have any
`
`expectation about the effect of a strong CYP3A inhibitor on mifepristone without a
`
`clinical study. For example, Dr. Greenblatt states that “the quantitative extent to
`
`which clearance and exposure [of a substrate] are modified by the inhibitor”
`
`“cannot be predicted.” Id. at ¶ 31. Dr. Greenblatt also admits, “[t]he state of the
`
`art does not allow us to predict these values a priori, or even based on in vitro
`
`studies” and “actual quantitative modifications can only be determined by an actual
`
`clinical study.” Id.
`
`43.
`
`In addition, under Dr. Greenblatt’s analysis, a POSA could only have
`
`an expectation about the safety of 600 mg of mifepristone in combination with a
`
`strong CYP3A inhibitor if clinical data were already available. For example, Dr.
`
`Greenblatt states that “drug interactions involving CYP3A inhibitors and CYP3A
`
`substrates are not quantifiable in advance, and the only way to determine the
`
`precise extent of the interaction for a given substrate is to perform a clinical study.”
`
`Id. at ¶ 37. According to Dr. Greenblatt, “[t]he bottom line is that one simply does
`
`not know the precise extent or clinical significance of a specific DDI until the
`
`interaction is clinically tested.” Id. at ¶ 33.
`
`44. Dr. Greenblatt, however, does not cite any clinical study that would
`
`have allowed a POSA to have a reasonable expectation that 600 mg of
`
`mifepristone in combination with a strong CYP3A inhibitor would be safe and
`
`15
`
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`Corcept Ex. 2056, Page 17
`
`

`

`effective in patients with Cushing’s syndrome. In fact, Dr. Greenblatt makes clear
`
`that his opinions are not based upon clinical data. See, e.g., id. at ¶ 37 (“Corcept,
`
`however, had not performed such a study on mifepristone, so the extent of the
`
`interaction and its clinical significance (or lack thereof) was unknown.”); id. at ¶
`
`129 (“[A]s is evident from the materials in the Drug Approval Package for
`
`Korlym, there was no clinical experience regarding the interaction between
`
`mifepristone and strong CYP3A inhibitors.” (emphasis in original)). Accordingly,
`
`Dr. Greenblatt’s own logic leads to the conclusion that a POSA would not have
`
`had a reasonable expectation of success.
`
`45.
`
`The statements in Lee (Ex. 1005) that Dr. Greenblatt relies upon also
`
`do not establish that a POSA would have had a reasonable expectation of success.
`
`For example, Dr. Greenblatt relies on Lee’s disclosure that the results of the
`
`recommended DDI studies “will help provide more therapeutic options available to
`
`Cushing’s patients and appropriate labeling of mifepristone when co-administered
`
`with CYP3A inhibitors.” See, e.g., Ex. 1002 at 61 (quoting Ex. 1005). Dr.
`
`Greenblatt, however, fails to put this quote in the correct context.
`
`46.
`
`First, Dr. Greenblatt omits that the authors of Lee, reviewers from the
`
`Office of Clinical Pharmacology at the FDA, clearly recommended
`
`contraindicating Korlym and strong CYP3A inhibitors and avoiding co-
`
`administration of mifepristone and moderate CYP3A inhibitors. See Ex. 1005 at
`
`16
`
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`Corcept Ex. 2056, Page 18
`
`

`

`37-38. This is confirmed by the Cross Discipline Team Leader Review. See Ex.
`
`2055 at 8 (“[T]he clinical pharmacology reviewer makes the following
`
`recommendations…. Concomitant use of strong inhibitors of CYP3A is
`
`contraindicated.”). Thus, when the authors of Lee referred to “more therapeutic
`
`options” they did so with the expectation that co-administration of mifepristone
`
`and strong or moderate CYP3A inhibitors would not be allowed in the Korlym
`
`label. Indeed, the 2012 Korlym Label did not exist at the time Lee was written.
`
`Thus, “more therapeutic options” referred to amending the label to allow any co-
`
`administration of mifepristone and a CYP3A inhibitor, not amending the label to
`
`allow co-administration of mifepristone at doses higher than the 300 mg limit that
`
`was ultimately included.
`
`47.
`
`Second, the relevant portion of Lee states:
`
`A drug-drug interaction study with ketoconazole is
`recommended as a Post Marketing Requirement (PMR).
`The goal of this study is to get a quantitative estimate of
`the change in exposure of mifepristone following co-
`administration with ketoconazole. Based on the results of
`this study, the effect of moderate CYP3A inhibitors on
`mifepristone pharmacokinetics may need to be addressed.
`This will help provide more therapeutic options available
`to Cushing’s patients and appropriate
`labeling of
`mifepristone when co-administered with CYP3A
`inhibitors.
`
`Ex. 1005 at 5. Similarly, Lee also states:
`
`DDI studies to evaluate the inhibitory effects of strong
`CYP3A4 inhibitor (ketoconazole is recommended) is
`17
`
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`Corcept Ex. 2056, Page 19
`
`

`

`recommended. Based on the results further evaluation to
`determine the effect of a moderate CYP3A4 inhibitor on
`mifepristone pharmacokinetics may be needed.
`Id. at 38. In other words, Lee recommended that Corcept run a DDI study with
`
`ketoconazole and, depending on the results, run an additional study with
`
`mifepristone and a moderate CYP3A inhibitor. A POSA would have understood
`
`that the only reason to run a DDI study with a moderate CYP3A inhibitor after
`
`running a DDI study with ketoconazole is if the first study showed that co-
`
`administration with ketoconazole causes a clinically significant increase in
`
`mifepristone exposure. Thus, Lee would not have indicated an expectation that a
`
`DDI study with mifepristone and ketoconazole would show that co-administration
`
`of more than 300 mg mifepristone and a strong CYP3A inhibitor was safe. Rather,
`
`a POSA would have understood that Lee expressly contemplates that the results of
`
`the ketoconazole DDI study would require contraindication and a second study
`
`would be necessary to see if co-administration with even moderate CYP3A
`
`inhibitors was safe.
`
`48. Dr. Greenblatt also fails to establish that a POSA would have had a
`
`reasonable expectation of success, because he ignored multiple properties of
`
`mifepristone that a POSA would have relied on in predicting a potential interaction
`
`between mifepristone and a strong CYP3A inhibitor. Mifepristone’s non-linear
`
`pharmacokinetics, inter-patient variability, CYP3A inhibition, and active
`
`18
`
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`Corcept Ex. 2056, Page 20
`
`

`

`metabolites would have prevented a POSA from having a reasonable expectation
`
`that there would not be a clinically significant interaction between mifepristone
`
`and a strong CYP3A inhibitor.
`
`49. As explained above, mifepristone’s non-linear pharmacokinetics
`
`result in blood levels that do not increase proportionally with dose. See supra at ¶
`
`28. In fact, the Korlym label discloses that “[e]xposure to mifepristone is
`
`substantially less than dose proportional.” Ex. 1004 at 13. Dr. Greenblatt,
`
`however, seemingly failed to appreciate that mifepristone demonstrates non-linear
`
`pharmacokinetics, stating that “[e]xposure—measured either by Cmax or AUC—is
`directly proportional to the drug dosing rate (for example, in milligrams per day),
`
`and inversely proportional to the drug’s clearance.” Ex. 1002 at ¶ 25 (emphasis
`
`added).
`
`50.
`
`In addition, Dr. Greenblatt ignores the fact that mifepristone exhibits
`
`highly variable inter-patient pharmacokinetics. See supra at ¶ 29. A POSA would
`
`have been concerned about any increase in mifepristone exposure in patients who
`
`exhibited higher levels of mifepristone even without a strong CYP3A inhibitor.
`
`51. Dr. Greenblatt also fails to mention that mifepristone and its active
`
`metabolites are CYP3A inhibitors themselves. Ex. 1005 at 9-10. Ketoconazole,
`
`like mifepristone, is a CYP3A substrate. Ex. 2019 at 6. A POSA would have been
`
`concerned about the potential for a clinically significant DDI with ketoconazole
`
`19
`
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`Corcept Ex. 2056, Page 21
`
`

`

`because mifepristone’s inhibitory properties were likely to affect both mifepristone
`
`and ketoconazole levels. Mifepristone inhibits CYP3A’s metabolism of
`
`ketoconazole, resulting in higher ketoconazole levels. A POSA would have
`
`expected that higher ketoconazole concentrations would in turn lead to increased
`
`inhibition of mifepristone metabolism. The resulting uncertainty regarding
`
`mifepristone’s inhibitory effect presents an additional complicating factor that
`
`would have prevented a POSA from having a reasonable expec