Filed on behalf of: Corcept Therapeutics, Inc.
`
`
`
`
`Entered: May 20, 2019
`
`
`
`
`
`
`
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`_______________________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`_______________________
`
`NEPTUNE GENERICS, LLC,
`Petitioner,
`
`v.
`
`
`CORCEPT THERAPEUTICS, INC.,
`Patent Owner.
`_______________________
`Case IPR2018-01494
`U.S. Patent No. 8,921,348
`_______________________
`
`PATENT OWNER’S RESPONSE
`
`
`
`
`

`

`IPR2018-01494 (USP 8,921,348)
`
`
`
`Patent Owner’s Response
`
`TABLE OF CONTENTS
`
`I.
`II.
`
`Page
`INTRODUCTION ........................................................................................... 1
`BACKGROUND ............................................................................................. 6
`A. Mifepristone and Its Unusual Pharmacologic Properties ...................... 6
`B.
`The ’348 Patent Invention ................................................................... 12
`C.
`Neptune’s Validity Challenge ............................................................. 14
`III. CLAIM CONSTRUCTION .......................................................................... 17
`IV. PETITIONER’S PRIOR ART REFERENCES ............................................ 18
`A.
`Belanoff & Murphy Mental Disorder References ............................... 18
`1.
`Belanoff ’953 ............................................................................ 19
`2.
`Belanoff ’848 ............................................................................ 21
`3.
`Belanoff 2002 ............................................................................ 23
`4.
`Chu & Belanoff ......................................................................... 25
`5. Murphy ...................................................................................... 26
`Sitruk-Ware Contraceptive Reference ................................................ 27
`B.
`PETITIONER’S GROUNDS FOR UNPATENTABILITY ......................... 30
`V.
`VI. PETITIONER HAS FAILED TO CARRY ITS BURDEN OF
`PROVING, BY A PREPONDERANCE OF THE EVIDENCE, THAT
`THE ’348 PATENT CLAIMS WOULD HAVE BEEN OBVIOUS ............ 31
`A. Ground 1 .............................................................................................. 31
`1.
`Belanoff ’848 Would Not Have Motivated a POSA to
`Measure and Monitor Mifepristone Serum Levels In
`Evaluating Efficacy ................................................................... 32
`
`i
`
`

`

`IPR2018-01494 (USP 8,921,348)
`
`
`
`Patent Owner’s Response
`
`2.
`
`3.
`
`B.
`
`Belanoff ’848 Would Not Have Motivated a POSA to
`Target the Key 1300 ng/mL Level, and a POSA Would
`Have Had No Reasonable Expectation of Success in
`Achieving the Key 1300 ng/mL Level Based on the
`Belanoff ’848 Dosages .............................................................. 35
`Based on Belanoff ’848 and Knowledge in the Art, a
`POSA Would Have Been Surprised to Learn that a
`Threshold Blood Level of 1300 ng/mL Could Be Used to
`Adjust Dosing ........................................................................... 36
`Ground 2 .............................................................................................. 38
`1.
`Nothing in the References Overcomes the Teaching Away
`in the Field From Assessing Efficacy and Adjusting Dose
`Based on Measuring Drug Serum Levels ................................. 40
`Sitruk-Ware Does Not Establish a Correlation Between
`Mifepristone Dose and Serum Level of 1300 ng/mL ............... 42
`a)
`Cmax is Not a Mean Serum Level .................................... 42
`b)
`Petitioner’s Expert Admitted He Failed to Disclose
`that Sitruk-Ware Reports Data from the Inaccurate
`RIA Method .................................................................... 44
`Sitruk-Ware’s Antiprogestin, Single-Dose, Cmax,
`Disclosure Would Not Motivate a POSA to Arrive
`at the Key 1300 ng/mL Level for Efficacy ..................... 46
`The Federal Circuit’s Decision in Cyclobenzaprine
`Establishes that the Lack of Disclosure of a PK-PD
`Relationship in Petitioner’s Cited Prior Art Renders the
`’348 Patent Invention Non-Obvious ......................................... 50
`A POSA Would Have Had No Motivation to Combine The
`Contraceptive Reference (Sitruk-Ware) With The Mental
`Disorder References (Belanoff & Murphy) .............................. 53
`Ground 3 .............................................................................................. 55
`C.
`D. Ground 4 .............................................................................................. 55
`
`2.
`
`3.
`
`4.
`
`c)
`
`ii
`
`

`

`IPR2018-01494 (USP 8,921,348)
`
`
`
`Patent Owner’s Response
`
`E.
`Ground 5 .............................................................................................. 56
`Ground 6 .............................................................................................. 57
`F.
`G. Dr. Heikinheimo’s Failure Confirms that a POSA Would Have
`Had No Reasonable Expectation of Success ....................................... 57
`The Patent Office Already Determined That the ’348 Invention
`Was Non-Obvious Because the Correlation Between the Serum
`Level and Efficacy Was Unknown...................................................... 59
`VII. CONCLUSION .............................................................................................. 61
`
`
`H.
`
`iii
`
`

`

`IPR2018-01494 (USP 8,921,348)
`
`
`
`Patent Owner’s Response
`
`TABLE OF AUTHORITIES
`
`CASES
`
`Page(s)
`
`Cuozzo Speed Techs., LLC v. Lee,
`136 S. Ct. 2131 (2016) ...................................................................................... 17
`In re Cyclobenzaprine Hydrochloride Extended-Release Capsule Patent Litig.,
`676 F.3d 1063 (Fed. Cir. 2012) ............................................................ 51, 52, 53
`
`Endo Pharms. Inc. v. Actavis LLC,
`No. 2018-1052, 2019 WL 1967605 (Fed. Cir. May 3, 2019) ..................... 38, 50
`Leo Pharm. Prods., Ltd. v. Rea,
`726 F.3d 1346 (Fed. Cir. 2013) .................................................................. 58, 59
`Valeo N. Am., Inc. v. Schaeffler Techs. AG & Co. KG,
`No. IPR2016-00502, 2017 WL 2664384 (P.T.A.B. June 20, 2017) .......... 38, 50
`REGULATIONS
`37 C.F.R. § 42.100(b) ............................................................................................ 17
`
`
`iv
`
`

`

`IPR2018-01494 (USP 8,921,348)
`
`
`
`Patent Owner’s Response
`
`EXHIBIT LIST
`Description
`Ex.
`2001 Corcept Therapeutics, KORLYM® (mifepristone),
`http://www.corcept.com/korlym.html (last visited Nov. 19, 2018)
`
`2002 Burford Capital, Learn About Burford’s Litigation & Arbitration Finance
`Solutions, The Leading Global Finance Firm Focused on Law,
`http://www.burfordcapital.com/about/ (last visited on Nov. 19, 2018)
`
`2003 U.S. Patent No. 8,598,149 (“the ’149 Patent”)
`
`2004 MedlinePlus Medical Encyclopedia, Therapeutic Drug Levels,
`http://www.nlm.nih.gov/medlineplus/ency/article/003430.htm (last visited on
`Nov. 19, 2018) (“Medical Encyclopedia”)
`
`2005 N.N. Sarkar, Mifepristone: Bioavailability, Pharmacokinetics and Use-
`Effectiveness, 101 EUROPEAN JOURNAL OF OBSTETRICS AND GYNECOLOGY
`AND REPRODUCTIVE BIOLOGY, 113-120 (2002) (“Sarkar”)
`
`2006 Oskari Heikinheimo, et al., Quantitation of RU 486 in Human Plasma by
`HPLC and RIA After Column Chromatography, 34 CONTRACEPTION No. 6,
`613-624 (Dec. 1986) (“Heikinheimo 1986”)
`
`2007 Oskari Heikinheimo, et al., Antiprogesterone RU 486—a Drug for Non-
`Surgical Abortion, 22 ANNALS OF MEDICINE, 75-84 (1990) (“Heikinheimo
`1990”)
`
`2008
`
`[Corrected] Declaration Of Michelle L. Ernst In Support Of Patent Owner’s
`Motion For Pro Hac Vice Admission Under 37 C.F.R. § 42.10(C)
`
`2009 Transcript of deposition of Mikko Oskari Heikinheimo, Ph.D., taken May 2,
`2019 in Neptune Generics, LLC v. Corcept Therapeutics, Inc., IPR No. 2018-
`01494 (“Heikinheimo Deposition”)
`
`2010 Unit conversion table, marked as Exhibit 1 to the Heikinheimo Deposition
`(“Heikinheimo Deposition Exhibit 1”)
`
`v
`
`

`

`IPR2018-01494 (USP 8,921,348)
`
`
`
`Patent Owner’s Response
`
`Ex.
`Description
`2011 Ann Robbins & Irving M. Spitz, Mifepristone: Clinical Pharmacology, 39
`CLIN. OBSTET. GYNECOL. 436 (1996) marked as Exhibit 2 to the Heikinheimo
`Deposition (“Robbins-Spitz”)
`
`2012
`
`Irving M. Spitz & C.W. Bardin, Clinical Pharmacology of RU 486—An
`Antiprogrestin and Antiglucocorticoid, 48 CONTRACEPTION 403 (1993)
`marked as Exhibit 3 to the Heikinheimo Deposition (“Spitz-Bardin”)
`
`2013 Donna Shoupe et al., Effects of the Antiprogesterone RU 486 in Normal
`Women, 157 AM. J. OBSTET. GYNECOL. 1415 (1987) marked as Exhibit 4 to
`the Heikinheimo Deposition (“Shoupe”)
`
`2014 Declaration of Hartmut Derendorf, Ph.D. in Support of Patent Owner’s
`Response (“Derendorf”)
`
`2015 Curriculum Vitae of Hartmut Derendorf, Ph.D.
`
`2016 Declaration of Dr. Ned H. Kalin, M.D. in Support of Patent Owner’s
`Response (“Kalin”)
`
`2017 Curriculum Vitae of Ned H. Kalin, M.D.
`
`
`
`
`
`vi
`
`

`

`IPR2018-01494 (USP 8,921,348)
`
`
`
`Patent Owner’s Response
`
`I.
`
`INTRODUCTION
`Mifepristone (most widely known as “RU-486” or “The Abortion Pill”) has
`
`been approved in Europe since 1988 and the United States since 2000 to terminate
`
`early pregnancy. Mifepristone binds to two receptors in human tissue–the
`
`progesterone receptor (“PR”) and the glucocorticoid receptor (“GR”). By binding
`
`to PR, a single dose of mifepristone terminates pregnancy. At the priority date of
`
`United States Patent No. 8,921,348 (“the ’348 Patent”), mifepristone’s activity at PR
`
`following administration of a single dose (the approved dosing regimen for
`
`termination of pregnancy) had been widely studied. By contrast, the ’348 Patent
`
`focused on an entirely different and much less thoroughly investigated activity of
`
`mifepristone–the wide range of physiologic responses and potential therapeutic
`
`benefits triggered by mifepristone’s binding at GR when administered for many
`
`days.
`
`The ’348 Patent discloses the surprising discovery that achieving a
`
`mifepristone serum level of 1300 ng/mL after seven days or more of dosing was
`
`critical to treating patients effectively. The ’348 Patent’s discovery of this threshold
`
`serum level ran counter to the prevailing wisdom with respect to the treatment of
`
`chronic disorders, such as mental disorders—namely, that there was no correlation
`
`between a medication’s serum levels and its efficacy. Furthermore, the ’348 Patent
`
`identified this critically important serum level despite the fact that (1) no
`
`1
`
`

`

`IPR2018-01494 (USP 8,921,348)
`
`
`
`Patent Owner’s Response
`
`pharmacokinetic-pharmacodynamic relationship was known in the field for
`
`mifepristone, especially with respect to its activity at GR, and (2) in spite of
`
`mifepristone’s non-linear PK profile and complex metabolism, which made
`
`identification of the relationship between dose, serum concentration and efficacy
`
`extraordinarily complex and unpredictable.
`
`The mifepristone art since the 1980s, including art authored by Petitioner’s
`
`expert, Dr. Heikinheimo, largely focused on the drug’s maximum blood level
`
`(known as “Cmax”) and activity at PR following administration of a single dose. In
`
`addition, there was universal agreement that the pharmacokinetic profile of
`
`mifepristone was non-linear, with no dose/concentration-response and wide
`
`variability (exceeding 800%) between patients. In other words, there was no way to
`
`predict how much of a mifepristone dose would make its way to a patient’s blood
`
`where it can have its clinically intended effect.
`
`The claimed method has nothing to do with the Cmax levels or PR activity
`
`because neither have any bearing on the disorders treated by the ’348 method.
`
`Instead, the clinical benefits of the claimed method derive from activity at GR and
`
`serum levels after at least seven days of dosing. In 2007, a POSA knew that plasma
`
`concentrations after seven days of dosing (i.e., at “steady state”) were very different
`
`from Cmax serum levels. A POSA further understood that one specific Cmax value did
`
`not inform what (steady state) serum level would be achieved after multiple days of
`
`2
`
`

`

`IPR2018-01494 (USP 8,921,348)
`
`
`
`Patent Owner’s Response
`
`dosing. Therefore, a POSA would not rely on Sitruk-Ware’s single dose Cmax since
`
`the claims expressly require multiple doses over a period of time–at least 7 days–
`
`and then adjusting dose to achieve the 1300 ng/mL level. The ’348 Patent
`
`surprisingly discovered that a serum concentration of 1300 ng/mL after at a least
`
`week of daily dosing was an important threshold for determining whether to adjust
`
`a patient’s dose and allowed earlier, simpler and more objective assessment of the
`
`efficacy of treatment.
`
`The Board instituted trial based on a combination of three references about a
`
`preliminary clinical trial (Belanoff 2002), a case report on a single patient (Chu and
`
`Belanoff), and a contraception review article purporting to describe peak blood
`
`serum levels of mifepristone in a small number of patients following a single dose
`
`administration (Sitruk-Ware). Under the Board’s analysis, a POSA would have
`
`correlated the dose administered in the clinical trial with the peak blood serum levels
`
`disclosed
`
`in Sitruk-Ware and derived
`
`the claimed method by
`
`routine
`
`experimentation. As discussed in detail below, the analysis cannot stand under a
`
`more complete record.
`
`By way of example, first, the Board misapprehended the data in Sitruk-Ware,
`
`describing the blood serum level of drug as “mean serum level” when it was actually
`
`a mean Cmax. The Cmax, a single (maximum) point occurring less than two hours
`
`after administering a single dose of drug would tell a POSA nothing about blood
`
`3
`
`

`

`IPR2018-01494 (USP 8,921,348)
`
`
`
`Patent Owner’s Response
`
`levels after seven daily doses when the relevant parameter is steady state or trough
`
`level of drug. Indeed, other Heikinheimo literature reported steady state blood levels
`
`below the critical 1300 ng/mL level (i.e., 1100 ng/mL) and that blood levels after
`
`multiple doses (at the relevant Cmin parameter) decreased between days 3 and 4.
`
`Second, Dr. Heikinheimo admitted in his deposition that critical data in
`
`Sitruk-Ware was derived using radioimmunoassay (RIA) methodology that greatly
`
`overestimates the amount of mifepristone in blood serum. Dr. Heikinheimo did not
`
`disclose this fact in his original declaration although he admitted it would be
`
`important for the Board to understand. Third, the Sitruk-Ware reference itself
`
`acknowledges that, due to mifepristone’s non-linear pharmacokinetics, blood levels
`
`of the drug are essentially identical, regardless of dose in the relevant range, which
`
`would have taught away and discouraged a POSA from seeking to correlate blood
`
`levels with efficacy or to determine the need for dose adjustment. Indeed, Dr.
`
`Heikinheimo’s own publications reported no correlation between dose or blood
`
`serum level and clinical effect.
`
`Nonetheless, Petitioner alleges that the ’348 Patent invention is obvious. It
`
`does so in the face of the fact that its own expert had been experimenting with
`
`mifepristone since at least the 1980s, authoring over 60 articles on the subject, and
`
`yet never discovered that a threshold for mifepristone efficacy existed, and never
`
`arrived at the 1300 ng/mL clinical threshold and resulting method claimed in the
`
`4
`
`

`

`IPR2018-01494 (USP 8,921,348)
`
`
`
`Patent Owner’s Response
`
`’348 Patent. Thus, Petitioner needs to resort to hindsight and ignore important claim
`
`limitations to support its arguments. Petitioner cobbles together various references
`
`to suggest that a few patients in a few small trials would have received doses
`
`resulting in a blood level of mifepristone over 1300 ng/mL, but Petitioner’s
`
`argument ignores that those references do not teach a skilled person that 1300 ng/mL
`
`is a threshold for efficacy, much less suggest that patient blood levels should be
`
`compared to that threshold and treatment adjusted accordingly. It is telling that, over
`
`this long period of intensive study of mifepristone, neither Petitioner’s expert nor
`
`anyone else suggested the existence of any threshold level of mifepristone, nor
`
`identified the threshold discovered by the ’348 Patent. Only by relying on the
`
`discovery in the ’348 Patent teaching the utility of the 1300 ng/mL level as a
`
`threshold does Petitioner purportedly find it in the art.
`
`It does so while entirely disregarding that a POSA would have had no
`
`motivation or reasonable expectation of success to perform the claimed method
`
`(evidenced by Dr. Heikinheimo’s failure) due to the non-linear PK profile which
`
`presented layers of complication. Petitioner also disregards the clear Federal Circuit
`
`law expressly recognizing that the discovery of a method of achieving a key serum
`
`level or PK parameter for efficacy is inventive where there is no known PK/PD
`
`relationship in the art, as is the case here. Even ignoring these flaws in its analysis,
`
`Petitioner fails to show a reasonable motivation to combine disparate references
`
`5
`
`

`

`IPR2018-01494 (USP 8,921,348)
`
`
`
`Patent Owner’s Response
`
`from entirely different clinical applications—contraceptives and mental disease—
`
`—e.g., contraceptives requiring only acute mifepristone administration to block the
`
`action of the hormone progesterone, and chronic diseases requiring repeated, long-
`
`term mifepristone administration to block the action of the hormone cortisol.
`
`Finally, Petitioner fails to address that the invention claimed in the ’348 Patent was
`
`completely surprising to a POSA as set forth by Patent Owner’s expert. Mifepristone
`
`is an unusual drug
`
`that has an extremely unusual pharmacokinetic/
`
`pharmacodynamics profile and relationship. Petitioner’s hindsight-driven attempt
`
`to divine a correlation out of unrelated data sets involving small numbers of patients
`
`must fail in view of the overwhelming evidence of non-obviousness.
`
`For these reasons, Petitioner has not carried its burden of proving that the
`
`challenged claims in each ground are unpatentable. Patent Owner therefore
`
`respectfully requests that the Board confirm the patentability of the claims.
`
`II. BACKGROUND
`The ’348 Patent reports that patients should be monitored for blood levels of
`
`mifepristone, that those levels should be compared against a 1300 ng/mL threshold
`
`after seven or more days of dosing, and that the dose should be adjusted accordingly
`
`despite many obstacles in the art.
`
`A. Mifepristone and Its Unusual Pharmacologic Properties
`First, mifepristone exhibits a non-linear pharmacokinetic profile within a
`
`6
`
`

`

`IPR2018-01494 (USP 8,921,348)
`
`
`
`Patent Owner’s Response
`
`wide range of clinically relevant doses (i.e., greater than about 50 mg or 100 mg and
`
`up to at least 800 mg, as stated by Petitioner’s expert and submissions). Derendorf
`
`¶¶62-68 (Ex. 2014); Ex. 2009, 32. The blood level of drug in a patient does not
`
`linearly increase as the dose is increased. As explained by Patent Owner’s expert,
`
`Dr. Derendorf, while drugs with a linear PK profile show a direct relationship or
`
`correlation between dose and serum concentration, drugs with non-linear PK profiles
`
`do not demonstrate a proportional dose-drug concentration relationship. Derendorf
`
`¶¶46-55. In other words, for a linear drug, one could increase the dose eight-fold
`
`and expect the serum level to also increase eight-fold. By contrast, for mifepristone
`
`doses of 100 mg to 800 mg, one could increase the dose by 8-fold and expect to see
`
`no significant change in blood level. Derendorf ¶64. The lack of a similar,
`
`proportional relationship for drugs with a non-linear profile, complicates the
`
`understanding of the resultant PK profile or any particular parameter based on a
`
`given dose. Derendorf ¶55.
`
`Non-linear drugs therefore necessitate numerous studies, including complex
`
`statistical modeling of the PK profile, in order to (1) understand how dose
`
`corresponds with blood drug levels, and (2) then understand how dose and drug
`
`blood levels relate to efficacy. Derendorf ¶55; Ex. 2009, 44, 47-49. This latter
`
`component is the pharmacodynamics aspect, which is separate and yet interrelated
`
`to pharmacokinetics. Derendorf ¶¶56-58.
`
`7
`
`

`

`IPR2018-01494 (USP 8,921,348)
`
`
`
`Patent Owner’s Response
`
`As of the 2007 priority date of the ’348 Patent, mifepristone’s non-linearity
`
`was reported extensively. Derendorf ¶¶62-68. Petitioner’s own expert reported that
`
`the plasma concentration did not correlate with dosing. Rather, plasma
`
`concentration did not differ significantly following administration of between 100
`
`mg and 800 mg of the drug. Ex. 2006, 9 (“Interestingly, even though there was an
`
`eight-fold difference in the oral dose, the mean plasma RU 486 concentration did
`
`not differ significantly at any point measured between the 800 mg and 100 mg group;
`
`except at 2 hours (p<0.05).”); Ex. 1012, 5 (“The plasma concentrations of RU 486
`
`did not differ significantly when the single oral dose of RU 486 was increased from
`
`100 to 800 mg.”); Ex. 2007, 7 (“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.”); see also Ex. 1011, 2; Ex. 1014, Ex. 1013, 1, 4.
`
`In addition to the lack of correlation between dose and serum concentration,
`
`the scientific literature reported that there was no correlation between dose and
`
`clinical effect. See Ex. 1013, 1 (“No clear dose-response correlation with clinical
`
`performance has been found.”); Ex. 2007, 8 (“The similar serum levels of RU 486
`
`measured following various doses seems to explain the observation that there is no
`
`correlation between the regimen of RU 486 and the clinical outcome.”); Ex. 2007, 7
`
`(“After the first clinical studies with RU 486 it soon became apparent that the clinical
`
`outcome was not correlated with doses of RU 486 administered.”); see also
`
`8
`
`

`

`IPR2018-01494 (USP 8,921,348)
`
`
`
`Patent Owner’s Response
`
`Derendorf ¶¶62-68.
`
`Consequently, workers in the field could not predict the effect of a dosage
`
`change or expect that an increase in dose would yield higher blood levels, and that
`
`unpredictability extended to the clinical effect of the drug as well, with no
`
`meaningful correlation between changes in dose and clinical effect. Derendorf ¶¶62-
`
`68. While some of these observations were focused on the anti-progestin effects of
`
`mifepristone, the anti-glucocorticoid effect was similarly unpredictable and lacked
`
`explanation in ordinary pharmacokinetic terms. Id. Indeed, Petitioner’s expert
`
`described as an “enigma” the observed effect that mifepristone seemed to lack
`
`systemic glucocorticoid effect at doses below 400 mg. Ex. 1012, 6 (“[I]t still remains
`
`an enigma why systemic antiglucocorticoidal effects are virtually never seen at RU
`
`486 doses below 400 mg.”); see also Ex. 2006, 11 (“This justifies further studies on
`
`the effect of various oral doses on metabolism of RU 486.”); Ex. 2009, 111-112.
`
`This was particularly puzzling as a POSA understood from the literature that the
`
`blood level resulting from 200 mg and 600 mg doses was the same, but yet, the 600
`
`mg dose produced an antiglucocorticoidal effect, and the 200 mg dose did not. Thus,
`
`a POSA would not have had any expectation that blood level correlated with
`
`efficacy.
`
`Second, in addition to its non-linear PK, the identification of the efficacious
`
`serum level was complicated by mifepristone’s complex metabolism. As explained
`
`9
`
`

`

`IPR2018-01494 (USP 8,921,348)
`
`
`
`Patent Owner’s Response
`
`by Dr. Derendorf, and various of the prior art references, mifepristone exhibits rapid
`
`metabolism and certain of its metabolites (e.g., the monodemethylated metabolite)
`
`show higher serum levels than the drug itself after administration. Derendorf ¶¶69-
`
`74. Thus, this further adds to the complication in determining any dose-serum/dose-
`
`response correlation. Derendorf ¶¶69-74. This is particularly so because the prior
`
`art, including that authored by Dr. Heikinheimo, demonstrated that not all analytical
`
`methods were sensitive enough to separate mifepristone from its metabolites in
`
`measuring and reporting serum levels. Derendorf ¶¶73-74; Ex. 2009, 88, 33-34.
`
`RIA was not capable of separating mifepristone from its metabolites, and therefore,
`
`serum levels reported by RIA overestimated the amount of mifepristone in the blood
`
`because it also reported the levels of the metabolites. Id. Dr. Heikinheimo
`
`confirmed during his deposition that the key data he relies upon in Sitruk-Ware was
`
`generated by the inaccurate RIA method and presents overestimated serum levels
`
`(that also include the metabolites). Ex. 2009, 88-89, 33-34; see also Derendorf
`
`¶¶110-115, 160-165.
`
`Dr. Heikinheimo also testified and published data using the “more specific”
`
`chromatography assay which showed “values [that] were lower” than the RIA assay.
`
`Ex. 2013, 2-3. Dr. Heikinheimo explained that “[p]resumably, the values after
`
`chromatography represent true RU 486 concentrations.” Ex. 2013, 3. Thus, the
`
`inaccurate and unreliable reporting in the prior art, including in Sitruk-Ware,
`
`10
`
`

`

`IPR2018-01494 (USP 8,921,348)
`
`
`
`Patent Owner’s Response
`
`complicated the understanding of those skilled in the art, such as Dr. Belanoff, in
`
`attempting to understand the mifepristone serum levels that would result from a
`
`given dose. Derendorf ¶¶ 73-74, 112-115. A POSA would have thus known that
`
`the Sitruk-Ware reference was unreliable, and therefore, not instructive.
`
`Third, the majority of the research and development with the mifepristone
`
`compound since its discovery in the 1980s was focused on its antiprogestin
`
`properties for various contraceptive applications. Dr. Heikinheimo confirmed that
`
`his 35 years of experience with mifepristone was largely focused on its antiprogestin
`
`applications, and explained that he does not have any experience treating the
`
`disorders described in the ’348 Patent. Ex. 2009, 20-22. Thus, there was less
`
`development and understanding
`
`in
`
`the art with respect
`
`to
`
`the drug’s
`
`antiglucocorticoid properties, and relevant applications, such as the mental disorders
`
`described in the ’348 Patent, including for Cushing’s syndrome.
`
`The antiprogestin literature also failed to provide information relevant to
`
`understanding the PK/PD relationship for antiglucocorticoid applications, including
`
`because the mechanism of action for these two pathologies are very different. Kalin
`
`¶¶52-54 (Ex. 2016). For example, before the priority date, mifepristone was
`
`typically administered and effective through a single dose for contraceptive
`
`purposes—and therefore the available studies were largely limited to single-dose
`
`studies. Kalin ¶¶52-53. Further, any dose-serum, or dose-response information
`
`11
`
`

`

`IPR2018-01494 (USP 8,921,348)
`
`
`
`Patent Owner’s Response
`
`from these studies is largely inapplicable to the PK/PD relationship that needed to
`
`be understood for identifying the efficacious dose and serum level for chronic
`
`treatments—typically characterized by multiple, continuous dosing as opposed to
`
`acute treatments such as for contraception. Kalin ¶54.
`
`B.
`The ’348 Patent Invention
`Despite these impediments in the art, the ’348 Patent reports results from
`
`hundreds of patients in clinical trials, and the elusive correlation between serum level
`
`and efficacy. Specifically, the ’348 Patent reports that, on average, patients having
`
`a blood serum concentration of mifepristone below 1300 ng/mL showed effectively
`
`no benefit over that observed for placebo, while patients having blood levels above
`
`that level showed significant clinical improvement. Ex. 1001, 2:14-20, 2:34-40,
`
`Figs. 2, 5. This discovery paved the way for the claimed treatment method, in which
`
`patients are managed during treatment, not by inefficient and subjective patient
`
`written surveys and interviews, but by objective testing blood levels. Kalin ¶¶42-
`
`51. The ’348 method provides an ability to optimize dose, whereas the prior art
`
`methodology merely provided a mechanism for determining whether a patient’s
`
`symptoms had improved. Id.
`
`The ’348 Patent describes large-scale clinical trials involving hundreds of
`
`patients—far more than the small numbers of patients reported in the prior art
`
`studies. Compare Ex. 1001, Fig. 2 (reporting over 400 patients) with, e.g., Ex. 1023
`
`12
`
`

`

`IPR2018-01494 (USP 8,921,348)
`
`
`
`Patent Owner’s Response
`
`(1 patient); Ex. 1006 (4 patients); Ex. 1008 (5 patients); Ex. 1007 (30 patients); cf.
`
`Ex. 2009, 52, 73, 76 (testifying studies with four-five patients are small). These
`
`large-scale clinical trials showed that “[f]or the same dose of mifepristone, the blood
`
`serum levels can differ by as much as 800% from one patient to another.” Ex. 1001,
`
`1:32-34; see also id. at 2:54-55, 6:8-11. Despite this variability, the data in the patent
`
`demonstrated that the key threshold for clinical efficacy was a mifepristone serum
`
`level of 1300 ng/mL after seven days of daily dosing. See Ex. 1001, 1:40-49, 2:62-
`
`3:5, 10:41-50. Derendorf ¶118; Kalin ¶¶47-51.
`
`The discovery described in the ’348 Patent followed decades of investigation
`
`into mifepristone pharmacology, including that of Petitioner’s expert, that had failed
`
`to identify this crucial threshold. This was the first reliable correlation between
`
`mifepristone blood levels and antiglucocorticoid effect, or efficacy. Compare with
`
`Ex. 1006, Ex. 1007, Ex. 1008, Ex. 1010, Ex. 1023, Ex. 1024 (expressly considering
`
`the use of mifepristone to treat mental disorders but not disclosing any correlation
`
`between mifepristone serum level and antiglucocorticoid efficacy); see also Ex.
`
`1013, 1; Ex. 2007, 7-8; Kalin ¶¶21-51.
`
`Based on this discovery, the ’348 Patent designed an efficient and reliable
`
`treatment protocol—i.e., a method of optimizing mifepristone levels in the blood—
`
`involving comparing against the 1300 ng/mL threshold. The ’348 Patent describes
`
`a three-step process. First, a patient is treated with seven or more daily doses of
`
`13
`
`

`

`IPR2018-01494 (USP 8,921,348)
`
`
`
`Patent Owner’s Response
`
`mifepristone over a period of seven or more days. Ex. 1001, 16:29-30. Second, the
`
`serum levels of a patient are tested to determine whether the blood levels of
`
`mifepristone are greater than 1300 ng/mL. Ex. 1001, 16:31-33. Third, the daily
`
`dose is adjusted to achieve mifepristone blood levels greater than 1300 ng/mL. Ex.
`
`1001, 16:34-35.
`
`The patented method is an enormous advancement over the prior art which
`
`described monitoring the efficacy of mifepristone treatment by mere observation of
`
`clinical symptoms; the ’348 invention provides more precise and objective measures
`
`for evaluating mifepristone treatment. Kalin ¶¶48-51. The ’348 Patent instead
`
`describes a reliable and efficient blood test that can be used by physicians to ensure
`
`efficacy across patients despite the 800% inter-patient variability in dose response.
`
`Id.
`
`C. Neptune’s Validity Challenge
`Neptune’s Petition presents the same arguments already considered during
`
`prosecution and is supported by a minimal 13-page declaration from its expert, Dr.
`
`Heikinheimo. During Dr. Heikinheimo’s deposition, it was revealed that Neptune’s
`
`counsel failed to take even the basic steps to prepare a proper declaration from Dr.
`
`Heikinheimo. For example, Neptune’s counsel did not instruct Dr. Heikinheimo on
`
`the patent law obviousness standard as is evidenced by the complete absence of a
`
`14
`
`

`

`IPR2018-01494 (USP 8,921,348)
`
`
`
`Patent Owner’s Response
`
`legal standard section in Dr. Heikinheimo’s declaration and Dr. Heikinheimo’s
`
`admissions during the deposition. See generally Ex. 1004; see also Ex. 2009, 41.
`
`Dr. Heikinheimo explained during his deposition that Neptune’s counsel did
`
`not instruct him in the law and that he did not conduct his analysis under the patent
`
`law obviousness standard. See, e.g., Ex. 2009, 40 (objection omitted) (Q. Did you
`
`analyze whether the claims of the ’348 Patent would be obvious? A. So if I
`
`understand you correctly, to me these claims were other self-evident. Q. And did
`
`you do an analysis of whether the claims were obvious under the patent law? A. I’m
`
`a scientist, not an expert on U.S. patent law.”); id. at 41 (Q. And – and so did you
`
`have an understanding when you were analyzing the claim