`
`Filed: September 13, 2016
`
`
`Filed on behalf of: AbbVie Biotechnology Ltd.
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`
`__________________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`__________________
`
`COHERUS BIOSCIENCES INC.,
`Petitioner,
`
`v.
`
`
`ABBVIE BIOTECHNOLOGY LTD.,
`Patent Owner.
`
`__________________
`
`
`Case IPR2016-00172
`Patent No. 8,889,135 B2
`
`__________________
`
`
`
`PATENT OWNER’S RESPONSE
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`

`

`TABLE OF CONTENTS
`
`Case IPR2016-00172
`
`Page(s)
`
`I.
`
`INTRODUCTION ........................................................................................... 1
`
`A. Dr. Baughman’s Half-Life Analysis Is Flawed..................................... 2
`
`B. A POSA Would Have Been Concerned About Lower Minimum
`Drug Concentrations.............................................................................. 4
`
`C. Half-Life Does Not Provide Sufficient Information For
`Designing A Dosing Regimen ............................................................... 6
`
`II.
`
`FACTUAL STATEMENT .............................................................................. 8
`
`A.
`
`The Prior Art ......................................................................................... 8
`
`1.
`
`2.
`
`Kempeni ....................................................................................11
`
`van de Putte ...............................................................................14
`
`B.
`
`The PK Data in Kempeni ....................................................................16
`
`The ’135 Patent ...................................................................................18
`C.
`D. HUMIRA® ...........................................................................................18
`INSTITUTION DECISION ..........................................................................19
`
`III.
`
`IV. THE CHALLENGED CLAIMS WOULD NOT HAVE BEEN
`OBVIOUS OVER VAN DE PUTTE AND KEMPENI ...............................20
`
`A.
`
`The Available Pharmacokinetic Data, Including Half-Life,
`Would Not Have Provided The Requisite Motivation And
`Expectation Of Success .......................................................................21
`
`1.
`
`Dr. Baughman’s calculation of serum drug levels is
`wrong.........................................................................................22
`
`i
`
`
`
`
`
`
`

`

`Case IPR2016-00172
`
`2.
`
`3.
`
`4.
`
`5.
`
`6.
`
`7.
`
`A POSA would have expected the steady-state trough
`levels of the claimed dosing regimen to be substantially
`lower than those of the 20mg weekly van de Putte
`regimen ......................................................................................26
`
`In her deposition, Dr. Baughman agreed that the Cmin of
`the claimed dosing regimen would have been expected to
`be lower than a 20mg weekly dose and that a POSA
`would have sought to design a dosing regimen to avoid
`that result ...................................................................................28
`
`The available PK and clinical data would have taught
`away from the claimed invention ..............................................31
`
`The lower Cmin of an every-other-week regimen would
`have raised concerns about anti-drug antibodies ......................39
`
`Doubling both the dose and interval between doses can
`abolish efficacy .........................................................................43
`
`Half-life does not provide sufficient information to
`design a dosing regimen ............................................................44
`
`B.
`
`C.
`
`“Biweekly” Dosing In Kempeni Would Not Have Provided The
`Requisite Motivation And Expectation Of Success ............................49
`
`Petitioner’s Claim That The Invention Represents “Routine
`Optimization” Of A Prior Art Therapy Does Not Satisfy Its
`Burden Of Proof ..................................................................................50
`
`D. Objective Indicia Support the Nonobviousness of the
`Challenged Claims ..............................................................................55
`
`1.
`
`2.
`
`3.
`
`There was a long-felt but unmet need for new RA
`therapies ....................................................................................55
`
`Despite low predicted trough levels, the claimed
`invention is unexpectedly effective ..........................................56
`
`The claimed invention was a commercial success as a
`result of its efficacious and safe dosing regimen ......................58
`
`
`
`ii
`
`
`

`

`Case IPR2016-00172
`
`E.
`
`The Claims Require a Therapeutically Meaningful Level of
`Efficacy................................................................................................61
`
`V.
`
`CONCLUSION ..............................................................................................63
`
`
`
`
`
`
`
`iii
`
`
`

`

`
`
`TABLE OF AUTHORITIES
`
`Case IPR2016-00172
`
` Page(s)
`
`Federal Cases
`Arendi S.A.R.L v. Apple Inc.,
`--F.3d--, 2016 WL 4205964 (Fed. Cir. 2016) ..................................................... 20
`
`Avanir Pharms., Inc. v. Actavis S. Atl, LLC,
`36 F. Supp. 3d 475 (D. Del. 2014) ...................................................................... 45
`
`Avanir Pharms. Inc. v. Par Pharm., Inc.,
`612 F. App’x 613 (Fed. Cir. 2015) ..................................................................... 45
`
`BioMarin Pharm. Inc. v. Genzyme Therapeutic Products Ltd.,
`IPR2013-00534, Paper 81 (Feb. 23, 2015) ................................................... 51, 52
`
`Continental Can Co. v. Monsanto Co.,
`948 F.2d 1264 (Fed. Cir. 1991) .......................................................................... 60
`
`In re Cyclobenzaprine Hydrochloride Extended Release Capsule
`Patent Litig.,
`676 F.3d 1063 (Fed. Cir. 2012) .................................................................... 31, 45
`
`Demaco Corp. v. F. Von Langsdorff Licensing, Ltd.,
`851 F.2d 1387 (Fed. Cir. 1988) .......................................................................... 59
`
`Dr. Reddy’s Labs., Ltd. v. Galderma Labs., Inc.,
`IPR2015-01782, Paper 10 (Feb. 16, 2016) ......................................................... 45
`
`Eli Lilly & Co. v. Zenith Goldline Pharms., Inc.,
`471 F.3d 1369 (Fed. Cir. 2006) .......................................................................... 38
`
`Galderma Labs., L.P. v. Tolmar, Inc.,
`737 F.3d 731 (Fed. Cir. 2013) ............................................................................ 60
`
`Innopharma Licensing, Inc. v. Senju Pharm. Co.,
`IPR2015-00903, Paper 82 (July 28, 2016) ................................................... 59, 60
`
`Leo Pharm. Prods., Ltd., v. Rea,
`726 F.3d 1346 (Fed. Cir. 2013) .......................................................................... 55
`iv
`
`
`
`
`

`

`Case IPR2016-00172
`
`In re Magnum Oil Tools, Int’l, Ltd.,
`--F.3d--, 2016 WL 3974202 (Fed. Cir. 2016) ................................................. 7, 30
`
`Merck & Co., v. Teva Pharm. USA, Inc.,
`395 F.3d 1364 (Fed. Cir. 2005) .......................................................................... 60
`
`Pfizer, Inc. v. Apotex, Inc.,
`480 F.3d 1348 (Fed. Cir. 2007) .......................................................................... 51
`
`Procter & Gamble Co. v. Teva Pharms. USA, Inc.,
`566 F.3d 989 (Fed. Cir. 2009) ............................................................................ 20
`
`Syntex (U.S.A.) LLC v. Apotex, Inc.,
`407 F.3d 1371 (Fed. Cir. 2005) .......................................................................... 20
`
`Tec Air, Inc. v. Denso Mfg. Michigan Inc.,
`192 F.3d 1353 (Fed. Cir. 1999) .......................................................................... 38
`
`WBIP, LLC v. Kohler Co.,
`--F.3d--, 2016 WL 3902668 (Fed. Cir. 2016) ..................................................... 59
`
`Yamanouchi Pharm. Co. v. Danbury Pharmacal, Inc.,
`231 F.3d 1339 (Fed. Cir. 2000) .................................................................... 19, 38
`
`Federal Statutes
`
`35 U.S.C. § 103 ........................................................................................................ 19
`
`
`
`
`
`
`
`v
`
`
`

`

`Case IPR2016-00172
`
`I.
`
`INTRODUCTION
`The Board instituted inter partes review of U.S. Patent No. 8,889,135 (“the
`
`’135 patent”) on the single ground that claims 1-5 would have been obvious based
`
`on a combination of van de Putte (Ex. 1004) and Kempeni (Ex. 1003). In its
`
`decision, the Board indicated that a trial was required to resolve “whether a skilled
`
`artisan would have relied upon half-life or dose stretching based on known half-life
`
`to establish a dosing regimen for D2E7.” Paper 9, 19. The evidence now before
`
`the Board definitely answers that question in Patent Owner’s favor: a skilled
`
`artisan would not have relied on D2E7’s reported half-life to establish the claimed
`
`dosing regimen. Indeed, serious safety and efficacy concerns would have steered a
`
`POSA away from the claimed regimen.
`
`Among the many problems with Petitioner’s argument is that, under
`
`undisputed pharmacokinetic (“PK”) principles, a person of ordinary skill
`
`(“POSA”) would have expected the minimum drug concentration between each
`
`dose (“Cmin”) in the claimed dosing regimen to be substantially less than the
`
`minimum drug concentration in the dosing regimens disclosed in van de Putte.
`
`Those lower troughs of drug concentration would have raised both efficacy and
`
`safety issues. The available clinical and PK data, including the half-life
`
`information on which Petitioner relies, would thus have discouraged a POSA from
`
`1
`
`

`

`Case IPR2016-00172
`
`stretching the 20mg weekly regimen of van de Putte to a 40mg every-other-week
`
`regimen.
`
`A. Dr. Baughman’s Half-Life Analysis Is Flawed
`As the Board recognized, Petitioner’s obviousness theory is primarily based
`
`on the declaration of Dr. Sharon Baughman and her claim “that a skilled artisan
`
`would have been led to biweekly dosing based on the known 11.6 to 13.7 day half-
`
`life of D2E7.” Paper 9, 19. The centerpiece of that theory is Dr. Baughman’s
`
`contention that a POSA would have expected the claimed 40mg every-other-week
`
`dose to be effective because the amount of drug in the blood two weeks after the
`
`40mg dose would allegedly be higher than the amount of drug in the blood one
`
`week after a 20mg dose. Ex. 1006 ¶¶67-68, 71.
`
`Dr. Baughman’s analysis is based on a table in her declaration, cited by the
`
`Board in it institution decision. Paper 9, 18 (citing Ex. 1006 ¶¶66-67, 72). In the
`
`table, Dr. Baughman purports to calculate and compare the amount of D2E7
`
`circulating in a patient’s blood one week and two weeks following an initial dose
`
`of the 20, 40, or 80mg weekly doses described by van de Putte. Ex. 1006 ¶67.
`
`Based on her calculations, she concludes that the Cmin of a 40mg every-other-week
`
`dose would be higher than the Cmin of a 20mg weekly dose. Id. ¶71.
`
`Dr. Baughman’s analysis is flawed in numerous ways. But the critical flaw
`
`is that Dr. Baughman approximated the amount of drug in the blood after just a
`
`2
`
`

`

`Case IPR2016-00172
`
`single dose, rather than examining the steady-state drug concentrations achieved
`
`after multiple doses.1 The claimed dosing regimen is not to a single dose; it is an
`
`every-other-week dosing regimen administered for a time “sufficient to treat”
`
`rheumatoid arthritis (“RA”). Likewise, the efficacy data in van de Putte on which
`
`Dr. Baughman relies was generated after three months of weekly injections, not a
`
`single dose.
`
`As explained in the declaration of Dr. Alexander Vinks, once the
`
`concentration of D2E7 in the body reaches a steady-state, a POSA applying
`
`fundamental PK principles would have expected that a multi-dose regimen
`
`delivering twice as much drug at twice the dosing interval would produce higher
`
`peaks and lower troughs of drug concentration in the body. A POSA thus would
`
`have expected that the Cmin of a 40mg every-other-week dose would be
`
`substantially lower for the average patient than the Cmin of the 20mg weekly
`
`regimen of van de Putte, not higher as Dr. Baughman suggests. While insufficient
`
`information was available
`
`in
`
`the prior art
`
`to perform a
`
`true PK/PD
`
`(pharmacodynamic) correlation for D2E7, Dr. Vinks further shows that the
`
`
`1 Steady-state concentrations are reached when the amount of drug eliminated
`
`from the body over each dosing interval is equal to the amount that was absorbed
`
`into the body after the previous dose.
`
`3
`
`

`

`Case IPR2016-00172
`
`existing PK information taught away from the claimed 40mg every-other-week
`
`dosing regimen because it indicated that the steady-state trough concentrations
`
`(Cmin) for the 40mg every-other-week dosing regimen would have been
`
`substantially lower than the trough concentrations for the 20mg prior art weekly
`
`regimen.
`
`Dr. Baughman’s analysis improperly ignored these effects. At her
`
`deposition, however, Dr. Baughman conceded that, at steady-state, lower Cmin
`
`levels would have been the expectation of a POSA based on “pharmacokinetic
`
`principles.” Ex. 2072, 90:20-91:3.
`
`B. A POSA Would Have Been Concerned About Lower Minimum
`Drug Concentrations
`
`Dr. Baughman admitted that a POSA would have wanted to design a dosing
`
`regimen in which the Cmin would be at or higher than the Cmin of a dosing regimen
`
`previously shown to be safe and efficacious. Id., 68:15-20. Dr. Baughman further
`
`acknowledged that many in industry believed that Cmin was the best parameter for
`
`determining the threshold of efficacy for a dosing regimen. Ex. 1006 ¶62. For
`
`D2E7, a POSA would have been particularly concerned about low Cmin values in
`
`view of prior art clinical trials involving dosing based on a patient’s weight, in
`
`which amounts that would have been expected to result in greater circulating drug
`
`concentrations than the claimed dosing regimen were shown to be insufficient.
`
`4
`
`

`

`Case IPR2016-00172
`
`Those results indicated that the claimed dosing regimen would also have been
`
`insufficient.
`
`Dr. Baughman also concedes that a POSA would have been concerned about
`
`under-dosing patients with D2E7 because of the fear that too little drug in the
`
`blood would increase the risk of anti-drug antibodies (“ADAs”). Ex. 1006 ¶71. In
`
`addition to safety concerns, ADAs were known to block a biologic’s efficacy by
`
`increasing the speed at which the biologic is removed from the body or by
`
`interfering with the biologic’s ability to bind to its target. Once an immune
`
`response to an anti-TNFα biologic is generated, any loss of efficacy is typically
`
`permanent and the patient may no longer respond to the drug at all.
`
`Dr. Baughman primarily relies on her flawed Cmin analysis to dismiss these
`
`concerns with respect to the claimed invention. She and Petitioner’s clinical expert,
`
`Dr. O’Dell, further contend that the D2E7 prior art was silent about the existence
`
`of ADAs. They infer from this silence that a POSA would have dismissed the
`
`concern that too low a dose of D2E7 would trigger an immune response. Id.; Ex.
`
`1007 ¶41. This argument is also refuted by the evidence.
`
`As an initial matter, the factual predicate of Dr. Baughman’s opinion is
`
`wrong. There are indications in the prior art of the possibility of ADAs to D2E7 as
`
`well as concerns about their effect on safety and efficacy. See §IV.A.5.
`
`5
`
`

`

`Case IPR2016-00172
`
`Further, no inference can be drawn from the absence of experimental proof
`
`showing ADAs in the D2E7 prior art. The prior art references consist of a handful
`
`of abstracts and reviews providing preliminary information on early, on-going,
`
`studies. None of the references discussed testing for ADAs. And it is not
`
`surprising that such testing was not reported. As explained in the declaration of
`
`Jeffrey Sailstad, the assays for detecting ADAs during this time period were
`
`inadequate, typically requiring a “wash-out” period, i.e., a period in which D2E7
`
`was not administered. Consequently, reliable ADA assays could not be performed
`
`during an ongoing clinical trial, such as that reported in van de Putte and the other
`
`references.
`
`C. Half-Life Does Not Provide Sufficient Information For Designing
`A Dosing Regimen
`
`More generally, the core premise of Petitioner’s theory is incorrect. Half-
`
`life is inadequate for designing a dosing regimen because it does not impart any
`
`information about drug concentration, the key determinant of safety and efficacy.
`
`Indeed, in the case of subcutaneous dosing, terminal half-life (what is reported in
`
`Kempeni) does not even tell a POSA how long the drug remains in the body, let
`
`alone at the site of action. It is a measure of drug elimination after the drug has
`
`reached the blood stream, which takes days following subcutaneous administration
`
`of D2E7.
`
`6
`
`

`

`Case IPR2016-00172
`
`Nor is there any logical reason why a POSA would have been led to a dosing
`
`interval that is the same as the terminal half-life of a drug. In the absence of
`
`additional data, a dosing interval that is the same as a drug’s terminal half-life
`
`ensures substantial, usually undesirable, fluctuations in drug levels. Moreover, the
`
`evidence shows that for therapeutic monoclonal antibodies, half-life is not a
`
`reliable predictor of dosing interval. Approved prior art therapeutic antibodies
`
`were dosed more frequently and less frequently than their terminal half-lives.
`
`In short, the evidence before the Board refutes the factual underpinnings of
`
`the obviousness theory on which the Petition is based. The known clinical and PK
`
`data, including half-life, would not have “led” a POSA to the claimed invention—
`
`they would have taught away. A POSA would not have been motivated to try a
`
`dosing regimen that was likely to be less efficacious—if it worked at all—than the
`
`known dosing regimens in van de Putte, particularly when that regimen carried an
`
`increased risk of ADAs and when experience with weight-based doses indicated
`
`that the claimed dosing regimen would be inadequate. And even if a POSA had
`
`tried the claimed invention, there would have been no reasonable expectation that
`
`it would be sufficient to treat RA. For these reasons alone, Petitioner has failed to
`
`carry its burden of demonstrating that the challenged claims are unpatentable. In
`
`re Magnum Oil Tools, Int’l, Ltd., --F.3d--, 2016 WL 3974202, at *10 (Fed. Cir.
`
`7
`
`

`

`Case IPR2016-00172
`
`2016) (“[T]he Board must base its decision on arguments that were advanced by a
`
`party, and to which the opposing party was given a chance to respond.”).
`
`Nonetheless, in addition to the declarations of Dr. Vinks and Mr. Sailstad,
`
`Patent Owner also submits the declarations of Drs. Allan Gibofsky, Jerry
`
`Hausman, and Brian Harvey. These declarations show that the development of an
`
`antibody-based dosing regimen in Phase II and III clinical studies was anything but
`
`“routine” and that the claimed invention—the first approved dosing regimen for
`
`the most successful pharmaceutical product in the world—produced unexpected
`
`results, satisfied a long-felt need where others had failed, and is directly
`
`responsible for the enormous success of the commercial embodiment of D2E7 in
`
`treating RA. This additional evidence confirms the patentability of the challenged
`
`claims.
`
`II.
`
`FACTUAL STATEMENT
`A. The Prior Art
`In June 2001, biologic agents designed to block TNFα activity were a new
`
`class of drugs that had shown promise for treating RA. Ex. 1003, 1; Ex. 2069 ¶53.
`
`At that time, there were two FDA-approved anti-TNFα biologics: ENBREL® (a
`
`TNFα receptor fusion protein) and REMICADE® (a chimeric monoclonal antibody
`
`that contains both murine and human sequences). Ex. 2065 ¶¶30-32; Ex. 2069
`
`¶54. D2E7 (HUMIRA®) is also a monoclonal antibody but was developed from
`
`8
`
`

`

`Case IPR2016-00172
`
`human genetic material. Ex. 2069 ¶77. It was the first such antibody to be
`
`approved by the FDA and the first antibody of any kind approved by FDA for
`
`subcutaneous administration.2 Ex. 2066 ¶11; Ex. 2069 ¶59; Ex. 2027, 10-12.
`
`Clinicians in 2001 were faced with a number of promising options for the
`
`treatment of RA, but there was significant confusion and unanswered questions
`
`about how to apply the available therapies. Ex. 2061, 1; Ex. 2065 ¶¶15, 38, 58-59;
`
`Ex. 2074, 104:3-106:6. With respect to anti-TNFα biologics in particular, there
`
`were significant safety concerns as of 2001 and “studies to elucidate the optimal
`
`degree of TNF inhibition that [was] safe and effective in each patient [were]
`
`crucial.” Ex. 2063, 2; see also Ex. 2068 ¶¶18, 35-37; Ex. 2074, 152:17-24,
`
`152:25-155:21; Ex. 2069 ¶53, 62.
`
`The prior art pertaining to D2E7 contained preliminary data from four Phase
`
`I clinical trials and one Phase II trial. See Ex. 2065 ¶¶42-50; Ex. 2069 ¶78, 91, 94.
`
`Limited information about these early trials was published in abbreviated form in
`
`review articles and conference abstracts, including van de Putte (Ex. 1004) and
`
`Kempeni (Ex. 1003). See also Exs. 1005; 1009; 1017; 1018; 1019; 1023; 1024;
`
`
`2 The use of monoclonal antibodies as therapeutic agents was in its infancy in
`
`2001. Only 11 such antibodies had been approved, most for acute rather than
`
`chronic conditions. Ex. 2066 ¶11.
`
`9
`
`

`

`Case IPR2016-00172
`
`2020. Collectively, the D2E7 prior art discussed a variety of dosing strategies
`
`involving different routes of administration, dosing schedules, dosing amounts, and
`
`response rates. Ex. 2065 ¶16.
`
`The clinical trials that led to the claimed invention did not constitute the
`
`exercise of “routine optimization” of a dosing regimen. Developing a clinical trial
`
`for an investigational new drug is a complex and unpredictable endeavor. Ex.
`
`2074, 124:11-21; Ex. 2066 ¶¶7-8, 13, 15-18. As explained by Petitioner’s
`
`rheumatology declarant Dr. O’Dell, “[c]onfounding factors include funding for
`
`clinical research, patient recruitment, which drugs to use and in what combination,
`
`duration of the study, accepted end points to measure efficacy, dosage of the
`
`medication, numbers and kinds of control groups needed, and the numbers of
`
`patients in each study needed to show real efficacy differences.” Ex. 2062, 2; Ex.
`
`2074, 111:14-113:10. Clinical trials of a biologic product—particularly an
`
`investigational new drug that has not yet been approved for human use—require an
`
`enormous investment of resources and face a high risk of failure. Ex. 2066 ¶¶8,
`
`17-18, 20; Ex. 2074, 106:7-19; see also id., 30:18-20 (a POSA “can’t assume that a
`
`dose or an interval is going to be effective or appropriate until you’ve actually
`
`studied it.”)
`
`Notably, biologics routinely fail to advance towards approval at even the
`
`later phases of clinical trials for any number of reasons, including because of the
`
`10
`
`

`

`Case IPR2016-00172
`
`failure of a drug’s dosing regimen. Ex. 2066 ¶9, 20-21. Indeed, poor dose
`
`selection was identified as the leading reason for delay and denial of FDA
`
`approval based on a review of NDAs submitted between 2000 and 2012. Id. ¶19;
`
`Ex. 2182, 4, 6. Thus, contrary to Petitioner’s unsupported statement cited in the
`
`Institution Decision, the art was not “advanced” as of June 2001 (Paper 9, 14)—it
`
`was still in its infancy.
`
`1. Kempeni
`Kempeni discloses several early “weight-based” D2E7 prior art trials. Ex.
`
`1003, 2-3; Ex. 2069 ¶78; Ex. 2065 ¶¶ 43-46, 75. The first Phase I study (DE001)
`
`examined 120 total patients, divided into 5 groups, who received placebo or a
`
`single intravenous dose of D2E7 based on weight (0.5 to 10mg/kg). Ex. 1003, 2;
`
`Ex. 2069 ¶79-80, 84. This study was followed by an open-label extension
`
`(DE003) in which patients continued to receive intravenous injections based on
`
`their body-weight. Ex. 1003, 2; Ex. 2069 ¶86. In the second Phase I study
`
`(DE004), just 24 patients received weekly subcutaneous weight-based doses of
`
`either placebo or 0.5mg/kg D2E7 for three months. Ex. 1003, 2-3; Ex. 2069 ¶¶87-
`
`88. The third Phase I study reported in Kempeni (DE010) involved a head-to-head
`
`comparison of a single, 1mg/kg weight-based dose of D2E7 administered either
`
`subcutaneously or intravenously. Ex. 1003, 3; Ex. 2069 ¶89.
`
`11
`
`

`

`Case IPR2016-00172
`
`In each of the trials reported in Kempeni, patients who received a weight-
`
`based dose of 0.5mg/kg had to be up-dosed to maintain their responder status.
`
`This need to up-dose indicated that a 0.5mg/kg dose is insufficient for treating RA
`
`across the patient population. See Ex. 2065 ¶¶44-45, 47; Ex. 2069 ¶¶154-155. The
`
`claimed 40mg dosing regimen—which Petitioner equates with the Kempeni dose
`
`by multiplying the 0.5mg/kg dose by an assumed average patient weight of 80kg—
`
`therefore also would have been insufficient to treat RA. See Pet. 27 (Table).
`
`Moreover, the prior art disclosed up-dosing in patients administered 0.5mg/kg
`
`intravenously. Ex. 2065 ¶¶44, 47. Compared to that method, subcutaneous
`
`administration decreases the bioavailability of the administered drug. Accordingly,
`
`the claimed 40mg dosing regimen would have been understood to result in lower
`
`concentrations of drug than those resulting from intravenous administration of a
`
`0.5mg/kg dose to an 80kg patient. Ex. 2069 ¶¶31-34, 72-74, 126-127; see also Ex.
`
`2017, 29; Ex. 2119, 19-20.3
`
`
`3 Reporting on a study lacking an intravenous arm (DE004), Kempeni states
`
`“preliminary data” shows that D2E7 blood levels after multiple SC doses were
`
`“comparable” to those obtained following intravenous administration. Ex. 1003, 3.
`
`No information was provided as to the PK parameters being considered, the
`
`concentration levels of D2E7, or the number of subcutaneous administrations of
`
`
`
`12
`
`

`

`Case IPR2016-00172
`
`There is nothing in Kempeni that teaches or suggests a fixed dose of D2E7
`
`or every-other-week dosing. All of the doses were weight-based, and the
`
`purportedly “biweekly” phase of the DE003 study actually involved a mean dosing
`
`interval of 2.5 weeks in a weight-based, intravenously administered regimen. Ex.
`
`1003, 2-3.
`
`Kempeni also reports substantial variability in the effect of the drug on
`
`patients
`
`(called pharmacodynamic
`
`(“PD”)
`
`responses)
`
`following
`
`single
`
`administration. Ex. 2069 ¶152. Specifically, in the three highest dose groups of
`
`the DE001 trial, “40-70% of patients achieved DAS and ACR20 response status at
`
`24 hours to 29 days” post-treatment, indicating significant patient-to-patient
`
`variability both with respect to whether the drug would work and how long it
`
`would take.4 Ex. 1003, 2
`
`
`D2E7 needed to achieve “comparable” concentration levels. Ex. 2069 ¶88.; Ex.
`
`2072, 79:6-22.
`
`4 DAS (Disease Activity Score) and ACR20 (American College of Rheumatology)
`
`refer to composite criteria used to measure the effectiveness of RA treatments. Ex.
`
`1003, 1-2; Ex. 2065 ¶¶40-41; Ex. 2112, 1.
`
`13
`
`

`

`Case IPR2016-00172
`
`van de Putte
`
`2.
`The van de Putte abstract is a conference abstract that reports preliminary
`
`data from the first Phase II trial of D2E7. Ex. 1004; Ex. 2065 ¶ 48; Ex. 2069 ¶91.
`
`This trial, called DE007, featured a three-month placebo-controlled study in which
`
`patients received a fixed dose of 20, 40, or 80mg D2E7 administered
`
`subcutaneously on a weekly schedule. Ex. 1004; Ex. 2069 ¶92. DE007 was the
`
`first fixed dose trial; all of the previous trials had used weight-based dosing. Ex.
`
`2065 ¶¶42, 61.
`
`A total of 283 patients were randomized equally into the four arms. Ex.
`
`1004. The study was not powered to provide statistically meaningful comparisons
`
`between doses, but only to determine the statistical significance of each of the
`
`doses compared to placebo. Ex. 2065 ¶64. The data showed that while the 40 and
`
`80mg doses were numerically superior to the 20mg dose, each dose was
`
`statistically superior to placebo. Id. ¶¶64-66; Ex. 2069 ¶93. Although the authors
`
`concluded that “20, 40 and 80 mg/week were nearly equally efficacious,” this
`
`statement was based on a comparison of each group to placebo, not to each other.
`
`Ex. 1004.
`
`Additional information about the DE007 trial exists in the prior art, although
`
`it is not addressed by Petitioner. After 3 months, patients receiving placebo were
`
`switched to a 40mg weekly dose. Ex. 2129; Ex. 2069 ¶94. Those patients showed
`
`14
`
`

`

`Case IPR2016-00172
`
`greater improvement in the same 3 clinical outcome measures (ACR20, SJC, CRP)
`
`after just 3 months of 40mg weekly dosing compared to patients who received
`
`20mg weekly for 6 months.5 Id.; Ex. 2065 ¶67.
`
`
`
`ACR20
`SJC (median)
`CRP (median)
`
`Percent Response or Improvement
`Placebo/40mg
`20mg
`3 mos./6 mos.
`3 mos./6 mos.
`10/59
`49/56
`16/56
`42/54
`01/67
`54/59
`
`
`The difference between 20mg and the higher weekly doses becomes even
`
`more evident at 12 months. For example, the ACR50 for the 20mg weekly group
`
`remains essentially the same between 3 and 12 months. Ex. 1024. The 40mg and
`
`80mg weekly groups, on the other hand, experience a 60% increase in percentage
`
`of patients achieving ACR50. Id. In fact, at 12 months, the percentage of patients
`
`receiving 40 and 80mg weekly doses was numerically superior for every clinical
`
`measure outcome compared to 20mg weekly. Id.; Ex. 2065 ¶68.
`
`
`
`
`ACR20
`ACR50
`TJC (median)
`
`20mg
`3 mos./12 mos.
`49/46
`24/25
`52/51
`
`Percent Response or Improvement
`40mg
`3 mos./12 mos.
`59/60
`27/43
`57/60
`
`80mg
`3 mos./12 mos.
`56/56
`19/31
`53/60
`
`
`5 SJC refers to Swollen Joint Count; CRP refers to C reactive protein, a biomarker
`
`of inflammation. Ex. 2065 ¶¶40, 48.
`
`15
`
`

`

`SJC (median)
`CRP (median)
`
`Other contemporaneous reports of the same clinical study do not even
`
`39/52
`53/55
`
`56/65
`61/64
`
`Case IPR2016-00172
`
`54/59
`64/60
`
`mention the efficacy of 20mg weekly dosing, indicating that the 40mg weekly
`
`regimen was preferred among the three regimens. Ex. 2020; Ex. 2065 ¶69.
`
`Moreover, in the full-length, peer-reviewed article reporting the data, the authors
`
`stated that “[i]n most measures of efficacy at week 12, adalimumab [D2E7] 40 mg
`
`was associated with better results than the other doses.” Ex. 2130, 9 (emphasis
`
`added); Ex. 2065 ¶70; Ex. 2069 ¶95.
`
`The PK Data in Kempeni
`
`B.
`Kempeni reports three PK parameters: (1) mean total serum clearance, (2)
`
`steady-state volume of distribution, and (3) an “estimated mean terminal half-life”
`
`of 11.6 to 13.7 days. Ex. 1003, 2; Ex. 2069 ¶80. These three metrics are reported
`
`as ranges in Kempeni, which does not report any patient-specific PK information.
`
`Ex. 2069 ¶133.
`
`Clearance refers to the rate at which a drug is eliminated from the body and
`
`is typically expressed as mL/min. Ex. 2069 ¶30. Volume of distribution refers to
`
`the theoretical volume over which the drug is distributed. It is typically expressed
`
`as L/kg, where kg is the weight of the patient. Id.
`
`Terminal half-life, a calculated value, refers to the time taken for the
`
`concentration of the drug in the blood to fall by 50% during the elimination phase
`
`16
`
`

`

`Case IPR2016-00172
`
`of a PK profile (the period when the rate of drug elimination due to excretion
`
`and/or metabolism predominates). Ex. 2069 ¶¶29, 109. Terminal half-life
`
`provides no information about the “absorption” phase (the period when the drug
`
`moves from the site of administration to the blood) or “distribution” phase (the
`
`period in which the drug is distributed to other areas in the body). Id. ¶¶25-26, 32,
`
`109; see also Ex. 2017, 14-19. Thus, in the case of a subcutaneously administered
`
`drug, the terminal half-life does not reflect how long the drug is in the body, nor
`
`does it provide any information about how long the drug is at the site of action. Id.
`
`¶¶25-26, 32, 34. Finally, half-life does not itself reveal any information about the
`
`concentration of drug in the blood, the PK parameter of primary importance for
`
`designing a dosing regimen. Id. ¶¶36, 107, 108, 110; Ex. 2119, 41.
`
`Kempeni does not report key exposure metrics such as Cmin, Cmax, or AUC.
`
`Ex. 1006 ¶62; Ex. 2069 ¶¶79-80, 86-89. Cmax and Cmin respectively refer to the
`
`peaks and troughs of a concentration-time curve that graphs exposure to a drug.
`
`Ex. 2069 ¶35, 39. AUC is the total area under a PK curve an