`
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`_____________________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`_____________________
`
`
`TEVA PHARMACEUTICALS USA, INC.
`Petitioner
`
`v.
`
`CORCEPT THERAPEUTICS, INC.
`Patent Owner
`_____________________
`
`PGR2019-00048
`Patent 10,195,214 B2
`_____________________
`
`SECOND DECLARATION OF DR. DAVID J. GREENBLATT, M.D.
`
`
`
`
`
`
`
`
`
`TEVA1067
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`
`

`

`Patent 10,195,214 B2
` TEVA1067 – Second Declaration of Dr. David J. Greenblatt, M.D.
`
`I, David J. Greenblatt, M.D., hereby declare as follows:
`
`1.
`
`I am over the age of eighteen (18) and competent to make this
`
`declaration.
`
`2.
`
`I have been retained as an expert witness on behalf of Teva
`
`Pharmaceuticals USA, Inc. for the above-captioned post-grant review (PGR). I am
`
`being compensated for my time in connection with this PGR at my standard
`
`consulting rate, which is $300 per hour, or $3000/day for work requiring out-of-
`
`state travel.
`
`3.
`
`I understand that this Declaration accompanies a reply in support of
`
`Teva’s petition for PGR involving U.S. Patent No. 10,195,214 (“the ’214 patent”)
`
`(TEVA1001). I also submitted a Declaration accompanying Teva’s petition in May
`
`2019. In that Declaration, I discussed my background and qualifications to offer
`
`expert opinions in this matter. My CV (TEVA1003) provides further information
`
`about my background and qualifications. In short, I am an expert in clinical
`
`pharmacology, pharmacokinetics, drug metabolism, and drug interactions—
`
`particularly those involving CYP3A inhibitors—and have been since prior to
`
`March 1, 2017.
`
`4.
`
`I understand that Corcept has proposed a definition for a person of
`
`ordinary skill in the art (“POSA”) that differs from my proposed definition in that,
`
`in Corcept’s view, the POSA or team making up the POSA also requires a medical
`
`- 1 -
`
`

`

`Patent 10,195,214 B2
`
` TEVA1067 – Second Declaration of Dr. David J. Greenblatt, M.D.
`
`professional with experience treating Cushing’s syndrome. I disagree with
`
`Corcept’s proposed definition and maintain that the definition set forth in my
`
`initial declaration is correct. TEVA1002, ¶18. However, the opinions set forth in
`
`my initial declaration and in this second declaration would be the same regardless
`
`of which definition is applied. I note that even under Corcept’s definition of a
`
`POSA I would still be an integral part of the “team” making up the POSA in view
`
`of my extensive knowledge and experience with clinical studies of drug-drug
`
`interactions involving CYP3A inhibitors.
`
`5.
`
`Counsel for Teva asked me to review and respond to certain opinions
`
`set forth in the declaration of Dr. F. Peter Guengerich (EX2056). I disagree with
`
`several of those opinions for reasons explained below. My failure to comment on a
`
`specific opinion of Dr. Guengerich’s should not be interpreted as an agreement
`
`with that opinion.
`
`6.
`
`In formulating my opinions in this case, I have considered all the
`
`references and documents cited in this declaration and my first declaration,
`
`including those listed below.
`
`Exhibit #
`1001
`
`1003
`1004
`
`Description
`Belanoff, J.K., “Concomitant Administration Of Glucocorticoid
`Receptor Modulators And CYP3A Inhibitors,” U.S. Patent No.
`10,195,214 B2 (filed June 19, 2017; issued February 5, 2019)
`Curriculum Vitae for David J. Greenblatt. M.D.
`Korlym Label (2012)
`
`
`
`- 2 -
`
`

`

`Patent 10,195,214 B2
`
` TEVA1067 – Second Declaration of Dr. David J. Greenblatt, M.D.
`
`
`Lee et al., Office of Clinical Pharmacology Review NDA 20687
`(Addendum, KorlymTM, Mifepristone) (2012)
`FDA Approval Letter for Korlym (mifepristone) tablets, NDA
`20217, dated February 17, 2012
`Tsunoda, S.M., et al., “Differentiation of intestinal and hepatic
`cytochrome P450 3A activity with use of midazolam as an in vivo
`probe: Effect of ketoconazole,” Clin. Pharmacol. Ther. 66(5): 461-
`471 (1999)
`Sitruk-Ware, R. and Spitz, I.M., “Pharmacological properties of
`mifepristone: toxicology and safety in animal and human studies,”
`Contraception 68: 409–420 (2003)
`Jang, G.R., et al., “Identification of CYP3A4 as the Principal
`Enzyme Catalyzing Mifepristone (RU 486) Oxidation in Human
`Liver Microsomes,” Biochem. Pharmacol. 52: 753-761 (1996)
`Greenblatt, D., “In Vitro Prediction of Clinical Drug Interactions
`With CYP3A Substrates: We Are Not There Yet,” Clin. Pharm.
`Ther. 95(2): 133-135 (2014)
`Greenblatt, D.J., et al., “Mechanism of cytochrome P450-3A
`inhibition by ketoconazole,” J. Pharm. Pharmacol. 63: 214–221
`(2011)
`Greenblatt, D.J. and von Moltke, L.L., “Clinical Studies of Drug-
`Drug Interactions: Design and Interpretation,” in Enzyme- and
`Transporter-Based Drug-Drug Interactions: Progress and Future
`Challenges. Pang, K.S. et al., ed., pp. 625-649, New York,
`Springer: (2010)
`Greenblatt, D.J., et al., “The CYP3 Family” in Cytochromes P450:
`Role in the Metabolism and Toxicity of Drugs and other
`Xenobiotics. Ionnides, C., ed., pp. 354-383, Royal Society of
`Chemistry: (2008)
`Ohno, Y., et al., “General Framework for the Quantitative
`Prediction of CYP3A4-Mediated Oral Drug Interactions Based on
`the AUC Increase by Coadministration of Standard Drugs,” Clin.
`Pharmacokinet. 46(8): 681-696 (2007)
`Nguyen, D. and Minze, S., “Effects of Ketoconazole on the
`Pharmacokinetics of Mifepristone, a Competitive Glucocorticoid
`
`1005
`
`1006
`
`1007
`
`1015
`
`1022
`
`1023
`
`1024
`
`1025
`
`1026
`
`1027
`
`1034
`
`
`
`- 3 -
`
`

`

`Patent 10,195,214 B2
`
` TEVA1067 – Second Declaration of Dr. David J. Greenblatt, M.D.
`
`
`1035
`1037
`
`1040
`
`1041
`
`1046
`
`1057
`
`1058
`
`1059
`
`1066
`
`1073
`
`1074
`
`1080
`
`Receptor Antagonist, in Healthy Men,” Adv. Ther. 34:2371–2385
`(2017)
`File History of U.S. Patent No. 10,195,214 B2
`Kaesar, B., et al., “Drug-Drug Interaction Study of Ketoconazole
`and Ritonavir-Boosted Saquinavir,” Antimicrobial Agents and
`Chemotherapy 53(2): 609–614 (2009)
`“A Guide to Drug Safety Terms,” FDA Consumer Health
`Information / U. S. Food and Drug Administration, (2012)
`downloaded from www.tinyurl.com/y6oao2sj
`“Guidance for Industry Drug Interaction Studies — Study Design,
`Data Analysis, and Implications for Dosing and Labeling,” U.S.
`Department of Health and Human Services, Food and Drug
`Administration, Center for Drug Evaluation and Research
`(CDER)., Center for Biologics Evaluation and Research (CBER)
`(2006)
`Greenblatt, D.J, et al., “Ketoconazole inhibition of triazolam and
`alprazolam clearance: Differential kinetic and dynamic
`consequences,” Clin. Pharmacol. Ther. 64(3):237-247 (1998)
`Greenblatt, D.J. and Koch-Weser, J., “Clinical Pharmacokinetics,”
` NEJM 293:702-705 (1975)
`Greenblatt, D.J. and Abourjaily, P.N., “Pharmacokinetics and
`Pharmacodynamics for Medical Students:A Proposed Course
`Outline,” J. Clin. Pharmacol. 56(10): 1180–1195 (2016)
`Friedman, H. and Greenblatt, D.J., “Rational Therapeutic Drug
`Monitoring,” JAMA 256(16): 2227–2233 (1986)
`Weber, J.M, et al., Other Review(s) NDA 202107 (KorlymTM,
`Mifepristone) (2012) [https://www.accessdata.fda.gov/drugsatfda_
`docs/nda/2012/202107Orig1s000OtherR.pdf]
`Banankhah, P.A., et al., “Ketoconazole-Associated Liver Injury in
`Drug-Drug Interaction Studies in Healthy Volunteers,” Journal of
`Clinical Pharmacology 56(10):1196–1202 (2016)
`Outeiro, N., et al., “No Increased Risk of Ketoconazole Toxicity in
`Drug-Drug Interaction Studies,” J. Clin. Pharmacol. 56(10):1203–
`1211 (2016)
`Locniskar, A., et al., “Interaction of diazepam with famotidine and
`cimetidine, two H2-receptor antagonsits,” J. Clin. Pharmacol.
`26(4):299–303 (1986)
`
`
`
`- 4 -
`
`

`

`Patent 10,195,214 B2
`
` TEVA1067 – Second Declaration of Dr. David J. Greenblatt, M.D.
`
`
`D. Greenblatt & J. Harmatz, “Ritonavir is the best alternative to
`ketoconazole as an index inhibitor of cytochrome P450-3A in
`drug-drug interaction studies,” Brit. J. Clin. Pharmacol.,
`80(3):342–50 (2015)
`E. Dunnigan et al., “Mifepristone (RU-486) in the treatment of
`Refractory Cushing’s Disease,” Endocrine Rev., Suppl. 1,
`31(3):S1201 (2010)
`Greenblatt, D.J, et al., “Kinetic and dynamic interaction study of
`zolpidem with ketoconazole, itraconazole, and fluconazole,”
`Clinical Pharmacol. Ther. 64(6):661–671 (1998)
`Declaration of F. Peter Guengerich, Ph.D.
`
`2023
`
`2036
`
`2054
`
`2056
`
`
`
`7.
`
`Dr. Guengerich states that he disagrees with my opinion that DDI
`
`studies were routine in the art prior to March 2017. EX2056, ¶75. Dr. Guengerich
`
`further states that, “[w]hen designing a DDI study, there are many different
`
`variables resulting in thousands of different options for a DDI study. For example,
`
`a POSA designing a DDI study would need to select, at least: study design (e.g.,
`
`randomized crossover, one-sequence crossover, parallel design), dosing regimen
`
`combinations (e.g., single dose/single dose, single dose/multiple dose, multiple
`
`dose/single dose, multiple dose/multiple dose), the strong CYP3A inhibitor, dose
`
`of mifepristone, dose of CYP3A inhibitor, whether the study is blinded or
`
`unblinded, and the patient population.” Id., ¶76.
`
`8.
`
`I strongly disagree with Dr. Guengerich’s suggestion that this list of
`
`variables renders the design and execution of clinical DDI studies complex. On the
`
`contrary, these studies are routine in my field and have been for decades.
`
`
`
`- 5 -
`
`

`

`Patent 10,195,214 B2
`
` TEVA1067 – Second Declaration of Dr. David J. Greenblatt, M.D.
`
`
`To be sure, there are choices one must make when designing a DDI
`
`9.
`
`study. But, so long as those choices are made in a scientifically responsible
`
`manner, they generally will not be expected to affect the outcome of the study or
`
`the reliability of the results.
`
`10. For example, with regard to Dr. Guengerich’s first variable—study
`
`design—those designing DDI studies generally prefer crossover designs because
`
`they allow subjects to serve as their own controls. In crossover designs, subjects
`
`participate (at least) twice, once in the control state (without perpetrator drug), and
`
`once with co-administration of the perpetrator drug. Crossover designs inherently
`
`control for inter-patient variability in the pharmacokinetic response to a given
`
`victim drug, so they tend to provide the strongest results from a statistical
`
`standpoint. I have used crossover designs in many of my own DDI studies. See
`
`TEVA1080; EX2054. When a crossover design is feasible, it will generally be
`
`used. Accordingly, Corcept’s decision to use a crossover design in its mifepristone-
`
`ketoconazole DDI study was an obvious and scientifically responsible choice.
`
`11. Sometimes, however, there may be reasons to avoid a crossover
`
`design—for example, if there are logistical constraints such that there is not time to
`
`perform a crossover study, or if the victim drug has some characteristic that makes
`
`the clinician hesitant to give subjects the victim drug twice. In those case, a parallel
`
`design can be used. One set of subjects takes victim without perpetrator, and a
`
`
`
`- 6 -
`
`

`

`Patent 10,195,214 B2
`
` TEVA1067 – Second Declaration of Dr. David J. Greenblatt, M.D.
`
`second group (completely different) takes victim plus perpetrator. Parallel-design
`
`studies generally need to use much larger sample sizes to control for inter-patient
`
`variability in the pharmacokinetic response to the victim drug. But, as long as they
`
`are designed correctly, they are still perfectly adequate when the circumstances call
`
`for them.
`
`12. The larger point is that these various study-design options have been
`
`known and applied in practice for decades. (For example, the results of the
`
`diazepam-famotidine-cimetidine crossover study cited above were published in
`
`1986, over thirty years before the priority date of the ’214 patent. TEVA1080.)
`
`Different designs may be preferable given the specific characteristics or objective
`
`of a given study. But choosing among them is not hard or complex for those with
`
`experience in this field. And, so long as the choice made is well-informed and
`
`scientifically responsible, the study will be expected to yield a reliable result.
`
`13. With respect to dosing, FDA Guidance explains that “[t]he
`
`inhibiting/inducing drugs and the substrates should be dosed so that the exposures
`
`of both drugs are relevant to their clinical use.” TEVA1041, 10. I agree with this
`
`recommendation and generally follow it in my own work.1
`
`
`1 I understand that Corcept has noted that FDA Guidance is a “draft”
`
`document. That is true, but many FDA Guidance documents are labeled “drafts,”
`
`
`
`- 7 -
`
`

`

`Patent 10,195,214 B2
`
` TEVA1067 – Second Declaration of Dr. David J. Greenblatt, M.D.
`
`
`14. Ketoconazole—the strong CYP3A inhibitor that the FDA selected for
`
`Corcept’s DDI study—is often dosed at 200 mg twice daily both in clinical
`
`practice and in clinical DDI studies. For example, we used this dosing regimen of
`
`ketoconazole in our ketoconazole-triazolam-alprazolam DDI study, the results of
`
`which were published in 1998. TEVA1046.
`
`15. Furthermore, as Dr. Guengerich acknowledges (citing one of my
`
`articles), ketoconazole is a widely used index strong CYP3A inhibitor precisely
`
`because its inhibition of CYP3A does not appear to be dose- or concentration-
`
`
`and all FDA Guidance documents state that they are non-binding. Those in the
`
`field nonetheless can and do consider FDA Guidance documents in designing
`
`clinical studies, including clinical DDI studies. In my experience, the FDA expects
`
`that those in the field will follow its recommendations unless there is a good reason
`
`not to do so. Accordingly, deviating from the FDA’s recommendations is generally
`
`unwise if the study at issue is going to be later used, for example, as evidence to
`
`support a request for FDA approval of a given drug or a given labeling change. See
`
`TEVA1041, 4 (“This guidance provides recommendations for sponsors of new
`
`drug applications (NDAs) and biologics license applications (BLAs) for
`
`therapeutic biologics who are performing in vitro and in vivo drug metabolism,
`
`drug transport, and drug-drug interaction studies.”).
`
`
`
`- 8 -
`
`

`

`Patent 10,195,214 B2
`
` TEVA1067 – Second Declaration of Dr. David J. Greenblatt, M.D.
`
`dependent. EX2056, ¶84 (citing TEVA1007, 467). Accordingly, “any dose of
`
`ketoconazole 200 mg or greater would result in the same CYP3A inhibition.” Id.
`
`So the dose of ketoconazole selected for a DDI study is generally not critical, so
`
`long as it is 200 mg or above. A POSA would have been aware of this well before
`
`March 2017.
`
`16. With regard to dosing of mifepristone, the 2012 Korlym Label permits
`
`co-administration of 300 mg mifepristone and strong CYP3A inhibitors. Based on
`
`that, a POSA would have expected that co-administration of 300 mg mifepristone
`
`and strong CYP3A inhibitors would be safe. Accordingly, Corcept’s choice to use
`
`the next-highest dose of mifepristone (600 mg) in its mifepristone-ketoconazole
`
`DDI study was a reasonable one—it allowed Corcept to test the safety of the dose
`
`that was immediately higher than the dose already expected to be safe.
`
`17.
`
`In any event, however, selection of the dose of mifepristone would not
`
`have been expected to affect the reliability of the results of the study. One could
`
`have obtained reliable data on the pharmacokinetic interaction of mifepristone and
`
`ketoconazole by using any approved dose of mifepristone (i.e., any dose between
`
`300 mg and 1200 mg).
`
`18. Dr. Guengerich states that Corcept’s DDI study was not routine
`
`“because the FDA required Corcept to run the DDI study with ketoconazole.”
`
`EX2056, ¶77. I disagree. While the FDA did later issue a recommendation against
`
`
`
`- 9 -
`
`

`

`Patent 10,195,214 B2
`
` TEVA1067 – Second Declaration of Dr. David J. Greenblatt, M.D.
`
`the use of ketoconazole in DDI studies, the general consensus in the field—as
`
`demonstrated in several articles published by myself and others—is that the FDA’s
`
`recommendation on this score was not supported by the evidence. TEVA1073,
`
`1196; TEVA1074, 1203. In fact, the article that Dr. Guengerich cites in this
`
`paragraph of his declaration—an article that I wrote—explained that there is
`
`“negligible evidence of a liver injury risk from ketoconazole used as an index
`
`CYP3A inhibitor in healthy volunteers.” EX2023, 343. The fact is that
`
`ketoconazole has been viewed as the gold-standard index strong CYP3A inhibitor
`
`for years, and it remains so today. Using ketoconazole for the mifepristone DDI
`
`study would have been an obvious choice even if the FDA had not explicitly told
`
`Corcept to use ketoconazole.
`
`19. With respect to the choice between blinded and unblinded, as FDA
`
`Guidance itself explains, DDI studies can usually be unblinded. TEVA1041, 10.
`
`That is particularly so with respect to studies aimed at evaluating pharmacokinetic
`
`effects because analysis of such endpoints is not subject to investigator bias. But—
`
`again—the choice between blinded and unblinded is not critical to the success of
`
`the study. Generally speaking, either design would work and would yield reliable
`
`results.
`
`20. Finally, with respect to patient population, clinical DDI studies are
`
`typically run with healthy subjects. I note that the FDA explicitly instructed
`
`
`
`- 10 -
`
`

`

`Patent 10,195,214 B2
`
` TEVA1067 – Second Declaration of Dr. David J. Greenblatt, M.D.
`
`Corcept to run the mifepristone-ketoconazole study with healthy patients.
`
`TEVA1066, 23. Again, however, the choice of patient population—within the
`
`bounds of scientifically responsible practice—will not be expected to affect the
`
`ultimate outcome or success of the study.
`
`21. The broader point is that the “variables” identified by Dr. Guengerich
`
`have been known and applied in practice for decades. There are no bewildering
`
`complexities, and no overwhelming and confusing array of options. The POSA
`
`would assess the characteristics of the victim and perpetrator drugs, and the overall
`
`objectives of the study, and choose the design that makes the most sense.
`
`22. Dr. Guengerich agrees with my opinion that it is not possible to
`
`quantify the extent of a given DDI in advance before running a clinical study.
`
`EX2056, ¶54. However, Dr. Guengerich states that he would have expected co-
`
`administration of 600 mg mifepristone and a strong CYP3A inhibitor to be
`
`“unsafe” based on five factors (identified in my first declaration) that affect the
`
`extent of a given CYP3A inhibitor-CYP3A substrate DDI. Id., ¶55.
`
`23.
`
`I disagree. These factors are merely qualitative guides to predicting
`
`whether a significant DDI will occur. They do not allow for any level of
`
`quantitative prediction about the extent of the DDI. Thus, a POSA looking at these
`
`factors would not be able to say, one way or another, whether co-administration of
`
`600 mg mifepristone and a strong CYP3A inhibitor was likely to be safe or unsafe.
`
`
`
`- 11 -
`
`

`

`Patent 10,195,214 B2
`
` TEVA1067 – Second Declaration of Dr. David J. Greenblatt, M.D.
`
`Instead, the POSA would have had to run a clinical DDI study to quantify the
`
`magnitude of the DDI. I have published articles making precisely this point. See,
`
`e.g., TEVA1023.
`
`24. Dr. Guengerich notes that strong CYP3A inhibitors increase AUC of
`
`sensitive CYP3A substrates by five-fold or more. EX2056, ¶23. To the extent Dr.
`
`Guengerich means to imply that this would have led a POSA to expect a five-fold
`
`or more increase in mifepristone AUC when co-administered with a strong CYP3A
`
`inhibitor, he is incorrect. Mifepristone is not a sensitive CYP3A substrate, see
`
`TEVA1034, 2377, nor would a POSA have expected mifepristone to be a sensitive
`
`CYP3A substrate prior to March 2017. It is not possible to predict whether a given
`
`CYP3A substrate will be “sensitive” or not prior to running a clinical DDI study.
`
`Moreover, as illustrated in a paper by Outeiro and colleagues, many (if not most)
`
`CYP3A substrates are not sensitive. TEVA1074, 1206. So even looking at the
`
`question from a purely probabilistic standpoint, there would have been no reason to
`
`expect that mifepristone would be a sensitive CYP3A substrate prior to running the
`
`clinical DDI study.
`
`25. Dr. Guengerich also states that the Dunnigan case report (EX2036)
`
`“would have indicated to a POSA that there is a clinically significant DDI”
`
`between mifepristone and ketoconazole. EX2056, ¶66. I have reviewed the
`
`Dunnigan case report, and I disagree with Dr. Guengerich’s opinion. Dunnigan
`
`
`
`- 12 -
`
`

`

`Patent 10,195,214 B2
`
` TEVA1067 – Second Declaration of Dr. David J. Greenblatt, M.D.
`
`provides incomplete data (from only three time points) on blood concentrations of
`
`mifepristone in a single patient who received ketoconazole and 600 mg
`
`mifepristone. Dunnigan is difficult, if not impossible, to interpret. First, Dunnigan
`
`does not indicate what the patient’s blood concentrations of mifepristone would
`
`have been in the absence of ketoconazole administration—data that would be
`
`necessary to form any hypothesis about the effect of ketoconazole. Second,
`
`Dunnigan states that 600 mg mifepristone was used but also contains data on
`
`mifepristone trough levels for doses of 300–1200 mg, the source of which is
`
`unclear. Finally, Dunnigan is a case report of data from a single patient. In general,
`
`case reports do not prove or disprove anything. Based on Dunnigan alone, I would
`
`not have been able to form an expectation one way or another about the qualitative
`
`or quantitative extent of the interaction between mifepristone and strong CYP3A
`
`inhibitors. I would have formed an expectation only based on more extensive
`
`clinical evidence, such as that provided by a clinical DDI study.
`
`26.
`
`I understand that Corcept has argued that a POSA would have
`
`expected a pharmacodynamic interaction between mifepristone and ketoconazole.
`
`Like pharmacokinetic interactions, however, pharmacodynamic interactions cannot
`
`be quantified in advance, prior to running a clinical DDI study. Clincal DDI studies
`
`catalogue adverse events and thereby allow the investigators to identify potentially
`
`harmful pharmacodynamic interactions.
`
`
`
`
`
`- 13 -
`
`

`

`Patent 10,195,214 B2
`
` TEVA1067 – Second Declaration of Dr. David J. Greenblatt, M.D.
`
`
`I hereby declare that all statements made herein of my own knowledge are
`
`true and that all statements made on information and belief are believed to be true,
`
`and further that these statements were made with the knowledge that willful false
`
`statements and the like so made are punishable by fine or imprisonment, or both,
`
`under Section 1001 of Title 18 of the United States Code.
`
`
`
`Respectfully submitted,
`
`
`
`____________________
`David J. Greenblatt, M.D.
`
`
`
`Date: June 4, 2020
`
`
`
`
`
`
`
`
`
`
`
`
`
`- 14 -
`
`