`
`
`Paper No. ___
`Date Filed: March 8, 2017
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`
`_______________________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`_______________________
`
`MYLAN PHARMACEUTICALS INC.,
`Petitioners,
`
`
`
`v.
`
`JANSSEN ONCOLOGY, INC.
`Patent Owner.
`
`_______________________
`
`Case IPR2016-01332
`Patent 8,822,438 B2
`
`_______________________
`
`PATENT OWNER’S RESPONSE
`
`JANSSEN EXHIBIT 2154
`Wockhardt v. Janssen IPR2016-01582
`
`
`
`
`
`Table of Contents
`
`I.
`
`INTRODUCTION ............................................................................................. 1
`
`II. CLAIM CONSTRUCTION ISSUES .............................................................. 5
`
`III. THE PRIOR ART RELIED UPON BY MYLAN ......................................... 7
`
`A. O’Donnell (Ex. 1003) ............................................................................. 7
`
`B.
`
`Barrie (Ex. 1005) .................................................................................... 8
`
`C. Gerber (Ex. 1004) ................................................................................... 9
`
`IV. THE RECORD REFUTES THE SCIENTIFIC BASIS OF MYLAN’S
`HINDSIGHT-DRIVEN OBVIOUSNESS CHALLENGE .........................10
`
`A. Mylan Incorrectly Contends that Abiraterone Acetate Causes the
`Same Side Effects as Ketoconazole ....................................................13
`
`1. Abiraterone Acetate and Ketoconazole Cause Very Different
`Effects on Steroid Biosynthesis ....................................................13
`
`2. Mylan Incorrectly Contends that Ketoconazole and Abiraterone
`Acetate Cause the Same Side Effects ...........................................17
`
`(a)
`
`There Is No Evidence that Ketoconazole Causes
`Mineralocorticoid Excess ...................................................18
`
`(b) There Was No Evidence in 2006 that Abiraterone Acetate
`Would Cause Mineralocorticoid Excess ............................19
`
`B. O’Donnell Establishes No Need for Glucocorticoid Replacement
`with Abiraterone Acetate ....................................................................20
`
`1. O’Donnell Reports No Side Effects of Abiraterone Acetate
`Warranting Glucocorticoid Replacement .....................................20
`
`2. O’Donnell’s Synacthen Test Results Did Not Establish a Need for
`Glucocorticoid Replacement with Abiraterone Acetate ...............24
`
`C. Ketoconazole with Prednisone Was Not Known to be “Safe and
`Effective” for Prostate Cancer in 2006 ..............................................27
`
`1. Gerber Did Not Establish that Ketoconazole with Prednisone Was
`Safe and Effective for Prostate Cancer .........................................27
`
`2. Other Prior Art Taught that Ketoconazole with Prednisone Was
`Not a Safe and Effective Treatment of Prostate Cancer ...............29
`
`
`
`D.
`
`Prednisone’s Severe Side-Effects Would Have Dissuaded Skilled
`Persons from Using It without a Clear Clinical Benefit...................30
`
`1. Use of Glucocorticoids Was Discouraged Because of Adverse
`Side Effects, Particularly in Prostate Cancer Patients ..................30
`
`2. The Prior Art Taught that Prednisone Could Fuel the Prostate
`Cancer ...........................................................................................32
`
`3.
`
`If Symptoms of Mineralocorticoid Excess Occurred, a Skilled
`Person Would Have Addressed Them with Other Available Drugs
`.......................................................................................................34
`
`In 2006, Prednisone Was Not Known to Have “Anti-Cancer
`Effects” ..................................................................................................35
`
`A Skilled Person Would Not Have Had a Reasonable Expectation
`of Success in Achieving the ’438 Patented Invention .......................37
`
`1. The Prior Art Provided No Basis for a Skilled Person to Expect
`Prednisone Would Provide Anti-Prostate Cancer Effects ............37
`
`2. The Unpredictability of Drug Combination Therapy For Prostate
`Cancer Precludes Obviousness .....................................................40
`
`E.
`
`F.
`
`G. Mylan’s Obviousness Grounds Rely on Hindsight ...........................41
`
`H. Mylan Has Failed to Meet Its Burden of Demonstrating
`Obviousness of the Claimed Invention ..............................................43
`
`1. O’Donnell and Gerber Do Not Provide a Reason to Use
`Prednisone with Abiraterone Acetate to Treat Prostate Cancer ...44
`
`2. Barrie and Gerber Do Not Provide a Reason to Use Prednisone
`with Abiraterone Acetate to Treat Prostate Cancer ......................46
`
`3. The Prior Art Taught Away from Using Prednisone in Prostate
`Cancer Patients .............................................................................47
`
`V. OBJECTIVE INDICIA OF NONOBVIOUSNESS CONFIRM THE
`PATENTABILITY OF THE CLAIMS ........................................................48
`
`A.
`
`The Claimed Methods Show Unexpected Results ............................48
`
`1. Multiple Clinical Studies Demonstrate Unexpected Anti-Cancer
`Effects of Glucocorticoid/ Abiraterone Acetate Combination
`Treatment ......................................................................................49
`
`2. Mylan’s Criticisms of the Invention’s Unexpected Results Are
`Unwarranted..................................................................................54
`
`
`
`iii
`
`
`
`B.
`
`C.
`
`D.
`
`Skepticism and the Failure of Others ................................................55
`
`The Claimed Invention Met a Long-Felt Need .................................57
`
`The Claimed Invention Has Significant Commercial Success ........58
`
`1. A Nexus Exists Between the Claimed Invention and ZYTIGA®’s
`Commercial Success .....................................................................59
`
`E. Mylan’s Blocking Patent Argument is Flawed .................................62
`
`VI. CONCLUSION ................................................................................................64
`
`
`
`
`
`iv
`
`
`
`IPR2016-01332
`Patent 8,822,438
`
`I.
`
`INTRODUCTION
`
`This proceeding involves a breakthrough in the treatment of prostate cancer
`
`patients with an advanced stage of the disease known as metastatic castration-
`
`resistant prostate cancer (“mCRPC”). Before the invention of U.S. Patent No.
`
`8,822,438 (“the ’438 patent”), these patients faced a dismal prognosis, with few
`
`meaningful treatment options. The invention – a two time FDA priority approved
`
`method of administering abiraterone acetate with prednisone that corresponds
`
`directly to the claims of the ’438 patent – changed this picture dramatically; prostate
`
`cancer patients treated with this therapy enjoy a striking increase in patient survival
`
`that could not have been predicted.
`
`Despite this unpredictable result, and the striking commercial success of this
`
`new and effective treatment, Mylan Pharmaceuticals Inc. (“Mylan”) contends it
`
`would have been obvious to the skilled person to co-administer prednisone with
`
`abiraterone acetate, advancing in its petition the same hindsight-infused theory of
`
`obviousness advanced previously by Amerigen Pharmaceuticals, Ltd. (See
`
`IPR2016-00286) (“Amerigen IPR”).1 (See Mylan ID at 2-3). Under that theory, a
`
`skilled person would have co-administered abiraterone with prednisone based on
`
`experiences with another “CYP17 inhibitor,” ketoconazole, which Mylan contends
`
`was a safe and effective method for treating prostate cancer before 2006.
`
`
`1
`IPR2016-0317 was joined with IPR2016-00286.
`
`1
`
`
`
`IPR2016-01332
`Patent 8,822,438
`
`
`Specifically, Mylan contends that, because a skilled person would have
`
`believed ketoconazole administration “resulted in adverse side effects including high
`
`blood pressure, hypokalemia and swelling associated with adrenocorticotropic
`
`hormone (“ACTH”) drive and mineralocorticoid excess,” it was routinely co-
`
`administered with glucocorticoids. Then Mylan asserts a skilled person would have
`
`expected those side effects to occur with abiraterone acetate because both it and
`
`ketoconazole are “CYP17 inhibitors.” Thus, according to Mylan, a skilled person
`
`would have administered prednisone with abiraterone acetate to address its “safety
`
`and tolerability.” (Pet. at 27-28); IPR2016-00286 (Paper 1 at 27).
`
`The record before the Panel now establishes that each of these assertions in
`
`Mylan’s petition is simply false.
`
`First, contrary to Mylan’s assertion, ketoconazole does not cause
`
`mineralocorticoid excess – it inhibits the production of mineralocorticoids (among
`
`other steroids) due to its non-specific suppression of various adrenal steroid synthesis
`
`pathways in the body. As Amerigen’s expert Dr. Serels admitted – whose opinion
`
`Mylan’s expert Dr. Garnick simply parrots – “mineralocorticoid excess would not
`
`occur with ketoconazole.” (Ex. 2151 (PO Resp) at 3-5, 13, 16-17; Ex. 2122 (Serels)
`
`at ¶8, 10 (emphasis added); Ex. 2037 (Serels) at 210:4-12; 211:1-20; Ex. 2126
`
`(Garnick) at 33:15-34:11, 36:7-16; Ex. 2127 (Serels) at 9:24-10:16). Ketoconazole
`
`with prednisone also was not seen as “safe and effective” for treating prostate cancer
`
`
`
`2
`
`
`
`IPR2016-01332
`Patent 8,822,438
`
`before 2006, and even today it has never been approved for prostate cancer
`
`treatment. Moreover, as O’Donnell reports, ketoconazole was often administered
`
`without side effects necessitating glucocorticoid replacement.
`
`Second, a skilled person would not have equated the side effects of
`
`ketoconazole with those of abiraterone acetate because these drugs do not have the
`
`same effects on the steroid biosynthesis pathways. Unlike ketoconazole, abiraterone
`
`acetate is a selective inhibitor of the enzymes in the pathways that lead to production
`
`of testosterone, and a skilled person would have understood it does not inhibit
`
`glucocorticoids to the same degree as ketoconazole. (Ex. 1020 (Harris) at 544; Fig.
`
`1, Fig. 2). In fact, the primary reference relied on by Mylan for its obviousness
`
`grounds expressly states that abiraterone acetate was “very well-tolerated” – directly
`
`refuting Mylan’s “common side effect” theory. (Ex. 1003 at 2322).
`
`Consequently, a skilled person would have seen no justification for co-
`
`administering another drug with abiraterone acetate, either to address
`
`“mineralocorticoid excess” or for any other “safety and tolerability” reason. (Id.; Ex.
`
`2028 ¶¶96, 109, 130, 147, 154, 155). And Mylan simply ignores the substantial
`
`evidence in this record establishing why a skilled person would not administer
`
`prednisone to prostate cancer patients absent some compelling clinical need, given its
`
`well-known and serious toxicities. (Ex. 2038 (Rettig) ¶109, 119; Ex. 2040 ¶¶67-68;
`
`Ex. 2126 (Garnick) at 87:19-89:2; Ex. 2125 (Dorin) at 30:19-31:4; 32:6-9).
`
`
`
`3
`
`
`
`IPR2016-01332
`Patent 8,822,438
`
`
`Mylan does advance a half-hearted argument that a skilled person would have
`
`been motivated to administer both agents because of “prednisone’s possible anti-
`
`cancer effects” (Pet. at 6, 40). But that theory was refuted by every oncology expert
`
`offering opinions in this and the Amerigen proceedings. As Mylan’s expert, Dr.
`
`Garnick, conceded, a skilled person in 2006 “would have believed that prednisone
`
`would not cause any measurable cancer-treating effect when combined with
`
`abiraterone acetate.” (Ex. 2126 (Garnick) at 140:7-15; 77:11-78:6; 78:19-79:19; Ex.
`
`1002 (Garnick) ¶89).
`
`In addition, the objective indicia of non-obviousness for the claimed invention
`
`is overwhelming. Mylan largely ignores the unexpected anti-cancer effects of
`
`prednisone when combined with abiraterone acetate and that the claimed invention
`
`met a long felt need. It then downplays the striking commercial success of the
`
`claimed treatment method, arguing there is no nexus between the claims and that
`
`success, and that the scale of that commercial success was not somehow sufficient.
`
`But as Amerigen’s expert acknowledged, the claims of the ’438 patent correspond
`
`directly to the approved indication for ZYTIGA® and its commercial success is
`
`directly dependent on the claimed co-administration with prednisone.
`
`At bottom, Mylan can point to nothing that even hints at the possibility that
`
`prednisone administered in combination with abiraterone acetate would have
`
`provided an unexpected anti-cancer benefit in treating prostate cancer. The record
`
`
`
`4
`
`
`
`IPR2016-01332
`Patent 8,822,438
`
`instead refutes each of Mylan’s contentions as to why a skilled person would have
`
`combined abiraterone acetate with prednisone to treat prostate cancer, and provides
`
`compelling evidence of objective indicia of non-obviousness of the claimed
`
`invention, including unexpected results, long-felt need, failure of others, and
`
`commercial success. The patentability of the contested claims should be affirmed.
`
`II. CLAIM CONSTRUCTION ISSUES
`
`Each of the claims of the ’438 Patent is directed to a “method for the treatment
`
`of a prostate cancer,” and each claim expressly requires administration of “a
`
`therapeutically effective amount of abiraterone acetate” and “a therapeutically
`
`effective amount of prednisone.” (Ex. 1001 at 16:15-20).
`
`The Panel construed the terms “treat,” “treating” and “treatment” to “include
`
`the eradication, removal, modification, management or control of a tumor or
`
`primary, regional, or metastatic cancer cells or tissue and the minimization or delay
`
`of the spread of cancer.” See Mylan ID at 5 (incorporating the Panel’s analysis from
`
`the Amerigen ID, including the construction of claim terms). Each of these required
`
`effects is directed to reducing the growth or spread of cancer itself. (Ex. 2028
`
`(Rettig) ¶77).
`
`Patent Owner understands the Panel’s construction to mean that administration
`
`of prednisone with abiraterone acetate, must at least cause an anti-cancer effect,
`
`regardless of whether it has any other non-anti-cancer effects. The Panel’s
`
`
`
`5
`
`
`
`IPR2016-01332
`Patent 8,822,438
`
`construction is thus consistent with the way that the district court construed this term
`
`in the co-pending litigation.
`
`Mylan nonetheless contends, as it did unsuccessfully in the district court, that
`
`the terms “treat,” “treating” and “treatment” under this construction do not require
`
`the administered prednisone to have any anti-cancer effect. (Pet. at 18, n.3).
`
`Mylan’s position cannot be reconciled with the specification or the claims. The only
`
`effects identified in the ’438 Patent disclosure from administration of prednisone are
`
`anti-cancer effects, and the claims expressly refer to the administration of prednisone
`
`causing those anti-cancer effects when co-administered with abiraterone acetate, as
`
`the district court found. (Ex. 2004; Ex. 2005 at 1).
`
`The Panel’s construction of “treat,” “treating” and “treatment” is consistent
`
`with the disclosure of the ’438 patent and should be maintained.
`
`Nonetheless, even under the modified claim construction urged by Mylan, the
`
`claims would not have been obvious. While Mylan argues a skilled person would
`
`have co-administered prednisone simply to improve the safety and tolerability of
`
`abiraterone acetate, the evidence proves the contrary – it shows that there was no
`
`reason for a skilled person to believe that abiraterone acetate would cause side effects
`
`warranting additional interventions, and that a skilled person would have considered
`
`other options if they occurred. See §IV.D.3. “Safety and tolerability” thus would
`
`have provided no reason to co-administer prednisone with abiraterone acetate at the
`
`
`
`6
`
`
`
`IPR2016-01332
`Patent 8,822,438
`
`time of the invention.
`
`III. THE PRIOR ART RELIED UPON BY MYLAN
`
`A. O’Donnell (Ex. 1003)
`
`O’Donnell published in 2004, and describes a series of three Phase 1 safety
`
`studies in which abiraterone acetate was tested in humans. (Ex. 1003 at 2318; Ex.
`
`2040 ¶¶19-20). O’Donnell reports that “[t]his is the first report of the effects of a
`
`specific 17α-hydroxylase/C17,20-lyase inhibitor in humans.” (Id.) (emphasis added);
`
`(Ex. 2038 ¶¶92-95). The studies were performed in patients with stable disease to
`
`evaluate the safety of abiraterone acetate and to determine the dose of abiraterone
`
`acetate that would result in maximum testosterone suppression in castrate and non-
`
`castrate men with prostate cancer. (Id.; Ex. 2126 (Garnick) at 150:11-14). As safety
`
`studies, the clinical trials reported in O’Donnell were not designed to examine, and
`
`did not establish, clinical efficacy of abiraterone acetate. (Ex. 2038 ¶95). Prostate
`
`cancer treatment was not an endpoint, and O’Donnell did not measure PSA or assess
`
`survival benefits of abiraterone acetate. (Ex. 2126 (Garnick) at 145:3-9; 150:11-14;
`
`Ex. 2124 (Ratain) at 72:9-21; Ex. 2037 at 178:11-18). O’Donnell reports that
`
`patients in the studies were not allowed to take concomitant steroids. (Id. at 2319;
`
`Ex. 2038 ¶191).
`
`O’Donnell reports that “abiraterone acetate was very well tolerated and no
`
`serious adverse events attributable to treatment were recorded.” (Id. at 2322; Ex.
`
`
`
`7
`
`
`
`IPR2016-01332
`Patent 8,822,438
`
`2040 ¶21). It also reports no hematologic or biochemical effects and no alteration in
`
`resting heart rate or blood pressure. (Id.). O’Donnell also reports that abiraterone
`
`acetate had “no effect on 17α-OH-progesterone production [the precursor to
`
`cortisol]” and “no significant effect on cortisol levels in these patients” (Id. at 2322-
`
`23; Ex. 2040 ¶22-25). O’Donnell concludes by proposing that the next study of
`
`abiraterone acetate should take place in castrate patients (i.e., concomitant GnRH
`
`dosing), notably omitting any mention of co-administered glucocorticoids. (Id. at
`
`2324; Ex. 2038 ¶¶54, 127-128).
`
`B.
`
`Barrie (Ex. 1005)
`
`Barrie (U.S. Patent 5,604,213) issued in 1997 and describes a novel class of
`
`compounds that are “powerful hydroxylase/lyase inhibitors.” (Ex. 1005 at 1:38-39).
`
`Abiraterone acetate is one of many compounds disclosed in the patent. (Id. at 27-30,
`
`Ex. 2040 ¶40). Barrie states that the compounds may be useful for treating prostate
`
`cancer and breast cancer. (Id. at Abstract; 10:47-56; Ex. 2038 ¶¶89-90).
`
`Barrie describes results of in vivo testing in male mice of abiraterone acetate,
`
`ketoconazole and one other compound. (Id. at 25:14-20; Ex. 2040 ¶¶40-42). Blood
`
`samples were tested for testosterone levels. (Id. at 25:26-28). In addition, adrenal
`
`glands, prostates, seminal vesicles, testes and kidneys of the mice were removed,
`
`weighed, and compared. (Id. at 25:28-31). In describing the test results, Barrie
`
`reports that whereas “[k]etoconazole caused an increase in adrenal weight at the two
`
`
`
`8
`
`
`
`IPR2016-01332
`Patent 8,822,438
`
`highest doses,” abiraterone acetate “had no significant effect, suggesting that
`
`abiraterone acetate did not inhibit corticosterone biosynthesis” (Id. at 25:45-48; Ex.
`
`2040 ¶¶43-45; Ex. 2038 ¶¶91, 103-105). Corticosterone is the major glucocorticoid
`
`in mice, and plays a similar role to cortisol in humans. (Ex. 2040 ¶¶11-12, 46; Ex.
`
`2038 ¶105). Barrie also disclosed in vitro data showing that abiraterone acetate may
`
`inhibit C17,20-lyase to a greater degree than 17α-hydroxylase suggesting that it would
`
`allow continued production of cortisol. (Ex. 2040 ¶41; Ex. 2038 ¶¶119-120).
`
`C. Gerber (Ex. 1004)
`
`Gerber describes a retrospective chart review of serum prostate specific
`
`antigen (“PSA”) levels for 15 patients with hormone refractory (i.e., castration
`
`resistant) metastatic prostate cancer who received ketoconazole and prednisone. (Ex.
`
`2038 ¶¶80-88; Ex. 2048 Table 2). It explains that ketoconazole was originally
`
`developed as an antifungal agent but was found to be a potent and non-selective
`
`inhibitor of adrenocortical and gonadal steroid synthesis. (Ex. 1004 at 1179).
`
`Gerber explains that prednisone was provided as “glucocorticoid replacement
`
`therapy” to counteract a known clinical side effect of ketoconazole, which inhibits
`
`production of all testicular and adrenal steroids. (Id. at 1179; Ex. 2038 ¶129).
`
`Gerber portrays any decline in PSA, regardless of degree or duration, as a
`
`“response.” (Ex. 1004 at 1178). By 2006, this type of PSA data was recognized as
`
`not being clinically meaningful. (Ex. 2038 ¶¶62-68; Ex. 2057 (Bubley) at Abstract,
`
`
`
`9
`
`
`
`IPR2016-01332
`Patent 8,822,438
`
`3461-62, 3464; Ex. 2047 (Stamey) Abstract). While Gerber reported that 80% of
`
`patients experienced some decline in PSA, the PSA declines were short-lived in 75%
`
`of these patients, which led the authors to conclude that they were “unlikely [to have]
`
`significant impact on survival” and “probably do not reflect significant disease
`
`regression.”2 (Ex. 1004 at 1178).
`
`IV. THE RECORD REFUTES THE SCIENTIFIC BASIS OF MYLAN’S
`HINDSIGHT-DRIVEN OBVIOUSNESS CHALLENGE
`
`Mylan’s obviousness challenge, like Amerigen’s before it, is a hindsight
`
`driven theory that rests on a series of demonstrably incorrect scientific assumptions
`
`and ignores evidence that contradicts its central premise – that a skilled person would
`
`have combined prednisone with abiraterone acetate to address its side effects when
`
`treating a prostate cancer patient.
`
`Mylan claims that because ketoconazole inhibited cortisol production, it
`
`would cause “mineralocorticoid excess” in a patient, which, in turn, would cause
`
`“adverse effects including hypertension, hypokalemia (decrease in circulating
`
`potassium levels), and fluid retention.” (Pet. at 27). To address that
`
`“mineralocorticoid excess” problem, Mylan contends it was “general knowledge” to
`
`
`2
`Gerber reports two patients had longer term 50% PSA declines, but does not
`
`provide any radiographic evidence or survival data to confirm these patients
`
`experienced a clinical benefit. (Ex. 2038 ¶182; Ex. 2046 (Therasse)).
`
`
`
`10
`
`
`
`IPR2016-01332
`Patent 8,822,438
`
`“administer a glucocorticoid, such as prednisone or hydrocortisone, to a patient … to
`
`reduce ACTH drive, and consequently, reduce mineralocorticoid excess.” (Id.) And
`
`then, to connect these supposed experiences to abiraterone acetate, Mylan portrays
`
`the two drugs simply as “CYP17 inhibitors.” (Id.).
`
`But every facet of Mylan’s obviousness theory is scientifically incorrect.
`
`First, at the time of the invention, a skilled person would not have assumed
`
`that side effects caused by ketoconazole would also be caused by abiraterone acetate
`
`because each drug causes very different effects on the steroid biosynthesis pathways
`
`in the body. (Ex. 2038 ¶¶97-108; Ex. 1003 at 2318; Ex. 1004 at 1177). Unlike
`
`ketoconazole, abiraterone acetate is a selective CYP17 inhibitor that does not inhibit
`
`synthesis of all adrenal steroids and thus does not cause the same types of side
`
`effects. (Id. at ¶106; Ex. 1003 at 2318 Fig. 1). And in fact, this is what O’Donnell
`
`reports from testing in human patients – abiraterone acetate was well tolerated by
`
`patients and did not cause the side effects Mylan alleges it would. (Ex. 1003 at
`
`Abstract, 2322; Ex. 2038 ¶107).
`
`Second, at the time of the invention, ketoconazole was known to not cause
`
`mineralocorticoid excess, but to diminish mineralocorticoid levels because it is a
`
`non-selective inhibitor of all steroid production. (Ex. 2038 ¶¶106, 129, 152; Ex.
`
`2066 (Mantero) at Abstract, 82; Ex. 1003 at 2318; Ex. 1020 at 544; Ex. 1004 at
`
`1177; Ex. 2090 (Tucker) at 2413-14; Ex. 2018 (Jubelirer) at 90). Mylan also
`
`
`
`11
`
`
`
`IPR2016-01332
`Patent 8,822,438
`
`provides no evidence that, in 2006, abiraterone acetate was known to cause side
`
`effects linked to mineralocorticoid excess. (Ex. 2038 ¶¶143-152).
`
`Third, even Mylan’s foundational assumption that ketoconazole in
`
`combination with prednisone was known in 2006 to “be a safe and effective
`
`treatment in human patients with hormone-refractory prostate cancer” is incorrect.
`
`(Pet. at 40). In 2006, it was recognized that ketoconazole with or without a
`
`glucocorticoid did not provide a survival benefit. (Ex. 2038 ¶¶168-170). Even today,
`
`ketoconazole, alone or with a glucocorticoid, has never been approved for the
`
`treatment of prostate cancer. (Id.; see also (Ex. 2037 (Serels) ¶¶175:20-176:8; Ex.
`
`2126 (Garnick) at 57:16-58:13).
`
`Fourth, Mylan ignores that in 2006 a skilled person would not administer
`
`prednisone to a prostate cancer patient without a demonstrated clinical need to do so,
`
`given the guidance in O’Donnell and that person’s knowledge that prednisone can
`
`cause undesirable side effects including exacerbation of the patient’s cancer. (Ex.
`
`2038 ¶¶49, 124-126, 131-142).
`
`Finally, the evidence uniformly contradicts Mylan’s unsupported assertion that
`
`prednisone was known in 2006 to have “possible anticancer effects.” (Pet. at 40, 41)
`
`(emphasis added). That evidence includes Mylan’s own expert, Dr. Garnick, who
`
`admitted that “a person of ordinary skill in the art in 2006 would not have believed
`
`that prednisone provided a therapeutically useful anti-cancer or anti-tumor effect to
`
`
`
`12
`
`
`
`IPR2016-01332
`Patent 8,822,438
`
`patients with prostate cancer” and that person would “have believed that prednisone
`
`would have had no anti-cancer effect when combined with abiraterone acetate.” (Ex.
`
`2126 (Garnick) at 140:7-15; 77:11-78:6; 78:19-79:19; Ex. 1002 ¶89, 90).
`
`Amerigen’s and Patent Owner’s experts agreed. (Ex. 2120 (Serels) ¶74; Ex. 2124
`
`(Ratain) at 90:13-16).
`
`Because the evidence contradicts the factual bases of Mylan’s obviousness
`
`challenge set forth in the Petition, Mylan cannot meet its burden in this proceeding.
`
`A. Mylan Incorrectly Contends that Abiraterone Acetate Causes the
`Same Side Effects as Ketoconazole
`
`Mylan’s obviousness theory is anchored on its assertion that, because
`
`abiraterone acetate and ketoconazole are both “CYP17 inhibitors,” they will cause
`
`the same side effects. (Pet. at 27-28). Mylan cites the necessity of treating those
`
`ketoconazole side effects as the reason why a skilled person would have been
`
`motivated to provide glucocorticoid replacement therapy to a patient treated with
`
`abiraterone acetate. (Id.). Mylan is incorrect on both points.
`
`1.
`
`Abiraterone Acetate and Ketoconazole Cause Very
`Different Effects on Steroid Biosynthesis
`
`Before 2006, ketoconazole and abiraterone acetate were known to not cause
`
`the same types of effects on the steroid synthesis pathways in the body. (Ex. 2126
`
`(Garnick) 29:5-17; 8:5-21; Ex. 2037 (Serels) at 59:20-24) Figures 1 and 2 below
`
`illustrate the different effects of each compound. (Ex. 2038 (Rettig) ¶¶25-39, Figs. 4
`
`
`
`13
`
`
`
`IPR2016-01332
`Patent 8,822,438
`
`and 5).
`
`Ketoconazole has far-reaching and profoundly inhibitory effects on the
`
`synthesis of all steroids. It is considered a non-selective steroid synthesis inhibitor.
`
`(Ex. 2040 ¶37; Ex. 2126 (Garnick) at 57:9-14; Ex. 2037 (Serels) at 74:13-75:14). As
`
`
`
`
`
`14
`
`
`
`IPR2016-01332
`Patent 8,822,438
`
`O’Donnell explains, “[k]etoconazole is relatively unselective, inhibiting both
`
`cholesterol side chain cleavage and 11β-hydroxylase.” (Ex. 1003 at 2318 (emphasis
`
`added); Ex. 2126 (Garnick) at 150:18-151:15; Ex. 2037 (Serels) at 80:9:12-124:23-
`
`125:7). “Cholesterol side chain cleavage” refers to the very first step in steroid
`
`synthesis, during which cholesterol is converted to pregnenolone, the building block
`
`for all steroid hormones. (Ex. 2038 ¶¶19-20; Ex. 2040 ¶44; Ex. 2126 (Garnick) at
`
`7:23-8:10; Ex. 2058 (Seifter) at 545-46, 551; Ex. 2086 (Princip. Endo.) at 705).
`
`A skilled person would have understood that inhibition of this first step by
`
`ketoconazole would suppress production of all other steroids, including not only
`
`testosterone, but also mineralocorticoids and glucocorticoids. (Ex. 2038 ¶¶145, 99-
`
`100; Ex. 2040 ¶36; Ex. 2065 at 1065; Ex. 2090 at 2413-14; Ex. 2066 at Abstract,
`
`82). Other prior art, including art cited by Mylan, explains that for this reason
`
`“replacement doses of [glucocorticoids] such as hydrocortisone may be required.”
`
`(Ex. 2018 at Table 3; Ex. 2090 at 2413-14; Ex. 1020 at 544; Ex. 1004 at 1177; Ex.
`
`1080 at 30).
`
`The skilled person would also appreciate that ketoconazole directly inhibits
`
`production of corticosterone and cortisol. (Supra Fig. 2; Ex. 2038 ¶¶97-100, Fig. 5).
`
`This is because, as O’Donnell notes, ketoconazole inhibits the action of 11β-
`
`hydroxylase. (Ex. 1003 at 2318).
`
`Abiraterone acetate, by contrast, is a selective CYP17 inhibitor, meaning it
`
`
`
`15
`
`
`
`IPR2016-01332
`Patent 8,822,438
`
`targets only CYP17. (Ex. 1003 at 2317 (abiraterone acetate is “a specific 17α-
`
`hydroxylase/C17,20-lyase inhibitor in humans.”) (emphasis added)). As illustrated in
`
`Figure 1 above, abiraterone acetate does not inhibit cholesterol side chain cleavage
`
`(an enzyme also referred to as “desmolase”) or 11β-hydroxylase. Therefore,
`
`abiraterone acetate permits production of numerous steroids whose production is
`
`inhibited by ketoconazole, including corticosterone and cortisol. (Ex. 2038 ¶¶97,
`
`103; Ex. 1005 at 25:45-49)
`
`Indeed, unlike previous clinical experience with ketoconazole, O’Donnell
`
`reports “there was no significant effect on cortisol levels” in patients treated with
`
`abiraterone acetate. (Ex. 1003 at 2222-23). Moreover, as seen in Fig. 1 (above),
`
`abiraterone acetate does not inhibit corticosterone which has glucocorticoid activity,
`
`and thus can compensate for cortisol, thereby preventing clinical symptoms. (Ex.
`
`2038 (Rettig) ¶¶25, 97; Ex. 2086 (Princip. Endocr. Ch. 72) at 705, 707; Ex. 2088
`
`(Princip Endocr. Ch. 78) at 752). As O’Donnell reports, abiraterone acetate was
`
`“well tolerated,” and no hypertension was seen. (Ex. 1003 at 2322).
`
`Barrie (Ex. 1005) similarly identifies important mechanistic differences
`
`between ketoconazole and abiraterone acetate in inhibiting androgen synthesis. It
`
`notes that administration of ketoconazole to mice “caused an increase in adrenal
`
`weight,” signifying an accumulation of cholesterol, the precursor to all steroids. (Ex.
`
`1005 at 25:45-46). As Dr. Auchus explains, a skilled person would understand this
`
`
`
`16
`
`
`
`IPR2016-01332
`Patent 8,822,438
`
`cholesterol backlog to result from ketoconazole’s inhibition of the first step of steroid
`
`synthesis, which prevents cholesterol from being converted into downstream
`
`steroids. (Ex. 2040 ¶44-45). Barrie reports, by contrast, that abiraterone acetate
`
`resulted in “no significant effect” in adrenal weight in in vivo testing in mice, which
`
`“suggests that [abiraterone acetate] did not inhibit corticosterone biosynthesis.” (Ex.
`
`1005 at 25:46-48). Corticosterone is the primary glucocorticoid in mice, and the lack
`
`of an effect on its production was particularly notable because corticosterone has
`
`glucocorticoid activity in humans that can compensate for reduced cortisol levels.
`
`(Ex. 2040 ¶¶44-47; Ex. 2038 ¶25; Ex. 2086 (Princip Endocr.) at 707; Ex. 2088
`
`(Princip. Endocr.) at 752).
`
`Dr. Garnick’s declaration omits any mention of these fundamental differences
`
`in the effects of ketoconazole and abiraterone acetate on the various steroid synthesis
`
`pathways, particularly those that might implicate a need for glucocorticoid
`
`replacement therapy. Remarkably, he makes no mention of ketoconazole’s ability to
`
`inhibit the production of all steroids. Instead, in his declaration, he repeatedly, and
`
`misleadingly, refers to ketoconazole as simply a “CYP17 inhibitor.” (Ex. 1002 ¶46,
`
`55, 58; Ex. 2037 (Serels) at 9:24-10:6; Ex. 2122 (Serels) ¶8; Ex. 2124 (Ratain) at
`
`48:22-49:11). His conclusory and unsupported opinions are entitled to no weight.
`
`2. Mylan Incorrectly Contends that Ketoconazole and
`Abiraterone Acetate Cause the Same Side Effects
`
`A central assumption of Mylan’s obviousness theory is that ketoconazole
`
`
`
`17
`
`
`
`IPR2016-01332
`Patent 8,822,438
`
`causes mineralocorticoid excess and its associated symptoms, such as hypertension,
`
`hypokalemia, or fluid retention, thus requiring glucocorticoid replacement. This, it
`
`contends, would have led a skilled person in 2006 to believe the same side effects
`
`would occur with abiraterone acetate (because ketoconazole and abiraterone are both
`
`“CYP17 inhibitors”), thereby requiring glucocorticoid replacement as well. (Pet. at
`
`57, 58). But Mylan’s and Ame