UNITED STATES PATENT AND TRADEMARK OFFICE
`
`
`
`
`
`
`
`
`
`
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`
`
`
`
`
`SANDOZ INC.,
`APOTEX INC., APOTEX CORP.,
`EMCURE PHARMACEUTICALS LTD.,
`HERITAGE PHARMA LABS INC.,
`HERITAGE PHARMACEUTICALS INC.,
`GLENMARK PHARMACEUTICALS, INC., USA,
`GLENMARK HOLDING SA,
`GLENMARK PHARMACEUTICALS, LTD., MYLAN
`LABORATORIES LIMITED, TEVA PHARMACEUTICALS USA, INC.,
`FRESENIUS KABI USA, LLC, and WOCKHARDT BIO AG,
`
`Petitioners
`
`v .
`
`ELI LILLY AND COMPANY,
`
`Patent Owner.
`
`Case IPR2016-003181
`U.S. Patent 7,772,209
`
`
`
`
`
`
`
`
`PETITIONER SANDOZ INC.’S REPLY TO PATENT OWNER’S
`RESPONSE TO ITS PETITON FOR INTER PARTES REVIEW
`
`
`1 Cases IPR2016-01429, IPR2016-01393, and IPR2016-01340 have been joined
`
`with the instant proceeding.
`
`

`
`
`
`I.
`
`II.
`
`TABLE OF CONTENTS
`
`INTRODUCTION ........................................................................................... 1
`
`LILLY CANNOT DISCLAIM ITS OWN PRIOR ART STUDIES
`SHOWING FOLIC ACID PRETREATMENT WORKED ............................ 2
`
`A.
`
`Lilly’s Own Worzalla, Hammond, and ’974 Patent Show Folic
`Acid Ameliorated Pemetrexed Toxicity And Preserved Efficacy ........ 2
`
`1. Worzalla Expressly Teaches Folic Acid Pretreatment
`Reduces Pemetrexed Toxicity And Improves Efficacy .............. 3
`
`2.
`
`3.
`
`Hammond Teaches Only Benefits From Vitamin
`Pretreatment With Pemetrexed ................................................... 5
`
`The ’974 Patent Teaches Folic Acid Pretreatment Reduced
`Toxicity “Without Adversely Affecting Therapeutic
`Efficacy” ..................................................................................... 8
`
`B.
`
`Near Contemporaneous FDA Correspondence Confirms The
`Prior Art Taught Vitamin Pretreatment Prior To June 1999 ................. 8
`
`III. A POSA WOULD HAVE BEEN MOTIVATED TO ADD VITAMIN
`B12 PRETREATMENT BASED ON NIYIKIZA AND CALVERT ............ 10
`
`A. Niyikiza And Calvert’s Link Between Pemetrexed Toxicity And
`Homocysteine Provided The POSA Reason To Pretreat With
`Vitamin B12 .......................................................................................... 10
`
`B.
`
`EP005 And Nutritional Literature Provide Additional Reason To
`Add Vitamin B12 To The Pretreatment Regimen ................................ 13
`
`IV. A POSA WOULD HAVE BEEN MOTIVATED IN JUNE 1999 TO
`REDUCE PEMETREXED’S TOXICITY .................................................... 15
`
`V.
`
`THE POSA HAD A REASONABLE EXPECTATION OF SUCCESS
`IN USING VITAMIN PRETREATMENT TO REDUCE TOXICITY
`WITHOUT COMPROMISING EFFICACY ................................................ 17
`
`A.
`
`B.
`
`The Prior Art Indicated Folic Acid Pretreatment Would Be
`Successful ............................................................................................ 17
`
`A POSA Likewise Would Have Had No Concerns About
`Vitamin B12 Pretreatment .................................................................... 19
`
`VI. THE BOARD SHOULD DISCOUNT TESTIMONY OF LILLY’S
`UNQUALIFIED AND BIASED EXPERTS................................................. 22
`
`i
`
`

`
`VII. LIMITATIONS REQUIRING STANDARD DOSAGES AND
`SCHEDULES CANNOT SAVE THE ’209 PATENT CLAIMS ................. 23
`
`A.
`
`B.
`
`C.
`
`350-600 µg Dosages Of Folic Acid Were Standard Prior To June
`1999 ..................................................................................................... 23
`
`Intramuscular Injections of 1000 µg Vitamin B12 Administered
`According To The Claimed Schedule Were Standard ........................ 24
`
`The ’974 Patent Discloses the Exact Pretreatment Schedules
`Required By The Claims ..................................................................... 25
`
`VIII. PURPORTED SKEPTICISM ........................................................................ 25
`
`IX. CONCLUSION .............................................................................................. 27
`
`
`
`ii
`
`

`
`TABLE OF AUTHORITIES
`
`CASES
`
`Arkie Lures, Inc. v. Gene Larew Tackle, Inc.,
`119 F.3d 953 (Fed. Cir. 1997) ..............................................................................26
`
`Astrazeneca LP v. Breath Ltd.,
`No. 08-1512, 2014 WL 2526909 (D.N.J. June 4, 2014) ......................................26
`
`Aventis Pharma Deutschland v. Lupin, Ltd.,
`499 F.3d 1293 (Fed. Cir. 2007) ............................................................................12
`
`Extreme Networks, Inc. v. Enterasys Networks, Inc.,
`395 F. App’x 709 (Fed. Cir. 2010) .......................................................................22
`
`In re Fulton,
`391 F.3d 1195 (Fed. Cir. 2004) ..................................................................... 16, 25
`
`In re Mouttet,
`686 F.3d 1322 (Fed. Cir. 2012) ............................................................................16
`
`Nilssen v. Osram Sylvania, Inc.,
`504 F.3d 1223 (Fed. Cir. 2007) ............................................................................23
`
`Ormco Corp. v. Align Tech., Inc.,
`463 F.3d 1299 (Fed. Cir. 2006) ............................................................................12
`
`Osram Sylvania v. Am. Induction Techs.,
`701 F.3d 698 (Fed. Cir. 2012) ..............................................................................23
`
`PharmaStem Therapeutics, Inc. v. ViaCell, Inc.,
`491 F.3d 1342 (Fed. Cir. 2007) ........................................................................3, 22
`
`Spectralytics, Inc. v. Cordis Corp.,
`649 F.3d 1336 (Fed. Cir. 2011) .............................................................................. 7
`
`
`
`
`
`
`iii
`
`

`
`I.
`
`INTRODUCTION
`
`It is undisputed that prior to June 1999, Niyikiza I linked pemetrexed’s
`
`toxicity to elevated pretreatment homocysteine levels. It was also well-known that
`
`combining folic acid and vitamin B12 provided a synergistic reduction in
`
`pretreatment homocysteine levels. The ’209 patent’s claimed combination of a
`
`known folic acid pretreatment regimen to reduce pemetrexed’s toxicity (Worzalla
`
`or Hammond I) with a known folic acid and vitamin B12 regimen suitable for
`
`reducing pretreatment homocysteine levels (Calvert and/or EP005) did nothing
`
`more than yield a predictable result: reduction of the pretreatment homocysteine
`
`levels linked to pemetrexed’s toxicity in cancer patients. Indeed, Lilly’s expert,
`
`Dr. Bruce Chabner, admitted that it was known prior to June 1999 that folic acid
`
`and vitamin B12 pretreatment would decrease pemetrexed’s toxicity. Ex. 1074,
`
`88:12-18, 89:14-22.
`
`Faced with a compelling case of obviousness, Lilly attacks its own peer-
`
`reviewed prior art, which is the same art on which the Board instituted this
`
`proceeding. These publications disclosed and advocated the use of vitamin
`
`pretreatment with pemetrexed. Lilly now reinterprets the data in those publications
`
`contrary to their express conclusions based largely on cherry-picked, and often out-
`
`dated statements concerning other antifolates. Notwithstanding Lilly’s attempt to
`
`1
`
`

`
`spin them otherwise, the prior art’s express teachings render the claimed vitamin
`
`pretreatment regimen with pemetrexed obvious.
`
`II. LILLY CANNOT DISCLAIM ITS OWN PRIOR ART STUDIES
`SHOWING FOLIC ACID PRETREATMENT WORKED
`
`A. Lilly’s Own Worzalla, Hammond, and ’974 Patent Show Folic
`Acid Ameliorated Pemetrexed Toxicity And Preserved Efficacy
`
`Worzalla, Hammond, and the ’974 patent expressly conclude that folic acid
`
`pretreatment preserved pemetrexed’s efficacy while reducing its toxicity:
`
`Worzalla:
`
`“Folic acid supplementation was demonstrated to preserve
`
`antitumor activity of [pemetrexed] while reducing toxicity.”
`
`Ex. 1013, p.3235.
`
`Hammond II: “These results indicate that folic acid supplementation
`
`appears to permit [pemetrexed] dose escalation.” Ex. 1014,
`
`p.866; Ex. 1015, p.129.
`
`’974 patent:
`
`“We have now discovered that the toxic effects of
`
`lometrexol and related GAR-transformylase inhibitors
`
`[which includes pemetrexed] and other antifolate agents
`
`which bind to folate binding protein (FBP) . . . can be
`
`significantly reduced by the presence of FBP binding agent,
`
`without adversely affecting therapeutic efficacy.” Ex. 1005,
`
`1:47-53; Ex. 1001, l:58-61.
`
`In seeking approval of pemetrexed, Lilly represented to the FDA that the data in
`
`Hammond and Worzalla support the above findings. Ex. 1076, ELAP00013767, -
`
`2
`
`

`
`82; Ex. 1077, ELAP00199798, -814. Lilly now disclaims the results of the above-
`
`mentioned studies even though each was performed by or in collaboration with
`
`Lilly. Exs. 1005, 1013-15. Lilly and its experts cannot rewrite Lilly’s own prior
`
`art to save the ’209 patent from obviousness. See PharmaStem Therapeutics, Inc.
`
`v. ViaCell, Inc., 491 F.3d 1342, 1361-63 (Fed. Cir. 2007).
`
`1. Worzalla Expressly Teaches Folic Acid Pretreatment
`Reduces Pemetrexed Toxicity And Improves Efficacy
`
`There is nothing “conclusory” about Worzalla. PO Resp. 32. Worzalla
`
`explains exactly why a POSA would expect folic acid pretreatment to reduce
`
`pemetrexed’s toxicity while preserving its efficacy:
`
`LFD animals with high levels of folate supplementation demonstrated
`
`decreased lethality to [pemetrexed] compared to conventional diet
`
`animals, suggesting that folate intake can be manipulated to achieve
`
`greater therapeutic effects. Oral folic acid dramatically decreased the
`
`toxicity of [pemetrexed] and preserved antitumor activity (albeit at
`
`higher dose levels) in these mice (Figure 2).
`
`Ex. 1013, p.3238 (emphasis added).
`
`Worzalla’s Figure 2 supports Worzalla’s conclusions. Figure 2 shows a
`
`much broader window of doses with 100% tumor inhibition and no lethality for the
`
`supplemented mice compared to mice who received no folate (LFD, no
`
`supplementation) and relatively low folate (standard diet). Id.; Ex. 1075, ¶¶56-57;
`
`Ex. 1067. Worzalla reported that folic acid pretreatment worked better with
`
`3
`
`

`
`pemetrexed than it did in earlier lometrexol studies, which resulted in reduced
`
`toxicity, but no efficacy “even at low doses” of folic acid. Ex. 1013, p.3238. In
`
`arguing that Worzalla shows that “folic acid pretreatment reduces pemetrexed
`
`efficacy,” Lilly thus improperly focuses on only one half of the equation (efficacy),
`
`and ignores Worzalla’s showing of a beneficial interplay between toxicity and
`
`efficacy resulting from folic acid pretreatment. PO Resp. 31.
`
`The Board correctly found at institution that a POSA would understand that
`
`animal studies like Worzalla “have applicability to humans.” Paper 14, Institution,
`
`p.17; Ex. 1074, 120:23-121:4; Ex. 1075, ¶11. Indeed, many have cited Worzalla to
`
`support the use of folic acid pretreatment in humans both pre- and post-June 1999.
`
`Ex. 1014; Ex. 1015; Ex. 1047, p.39; Ex. 1078, p.21; Ex. 1079, p.205; Ex. 1080,
`
`p.168; Ex. 1081, p.1078; Ex. 1075, ¶54; Ex. 1102, p.55; Ex. 1103, p.38; Ex. 1104,
`
`p.1016.
`
`Finally, Lilly’s critique of Worzalla as involving a “special type of tumor
`
`engineered to be especially sensitive to antifolates,” i.e., L5178Y/TK-/HX-
`
`lymphoma (PO Resp. 30), ignores that the contrast between the particularly
`
`sensitive tumor cells versus the relatively less sensitive healthy cells is precisely
`
`why Worzalla’s model works with folic acid pretreatment. Ex. 1075, ¶55.
`
`4
`
`

`
`2. Hammond Teaches Only Benefits From Vitamin
`Pretreatment With Pemetrexed
`
`Try as they might to support their claim that Hammond “taught away” from
`
`folic acid pretreatment (PO Resp. 28-29), Lilly and Dr. Chabner cannot change
`
`what Hammond actually states. The Hammond abstracts uniformly applaud the
`
`benefits of folic acid pretreatment as “perm[mitting] [pemetrexed] dose escalation
`
`by ameliorating toxicity.” Exs. 1015 & 1014; Ex. 1083, p.371 (Lilly article
`
`reporting Hammond showed “folic acid ameliorates toxicities permitting dose
`
`escalations”). Hammond reports that the “study was initiated to determine if [folic
`
`acid] supplementation permits significant dose-escalation above the recommended
`
`phase II dose of MTA alone.” Exs. 1015 & 1014. Thus, Hammond’s objective
`
`was to reduce toxicities to permit higher doses that would improve efficacy.
`
`Ex. 1075, ¶35.
`
`Lilly and Dr. Chabner’s teaching away argument is predicated on turning
`
`Hammond into something it’s not: a study on efficacy. Id. ¶ 41. In Hammond’s
`
`Phase I clinical study, patients with a variety of cancers who had failed to respond
`
`to previous therapies were given pemetrexed with folic acid pretreatment to
`
`determine a maximum tolerated dose (“MTD”) of pemetrexed. Id. ¶ 37; Ex. 2126,
`
`161:4-9, 161:23-162:6. As Dr. Chabner acknowledged, variables unrelated to the
`
`drug under investigation can impact efficacy in such Phase I trials, e.g., “cross
`
`resistance” and the “patient’s tolerance to drugs” due to the prior chemotherapy.
`
`5
`
`

`
`Ex. 1074, 52:16-53:22; Ex. 1075, ¶¶39, 44. Thus, the partial response reported in
`
`Hammond II would be viewed as encouraging, but could not be used to quantify
`
`efficacy. Ex. 1075, ¶¶35, 37.
`
`Nonetheless Lilly improperly draws efficacy and toxicity comparisons
`
`between Hammond and other studies reported in, for example, Rinaldi and
`
`Rusthoven. Lilly’s comparisons are flawed. Ex. 1075, ¶¶36-49. First, Dr.
`
`Chabner acknowledged that a POSA “would understand that the Hammond trials
`
`were not statistically designed to permit quantitative comparisons of the results in
`
`the Hammond trials to another Phase I trial.” Ex. 1074, 78:20-79:12; Ex. 1118,
`
`p.3-4; Ex. 1075, ¶39; Ex. 1095, p.3-4. Second, the lack of major anti-tumor
`
`responses certainly does not “point[ ] away from pretreating patients with folic
`
`acid,” as Lilly argues, since a POSA would recognize that the Phase I study
`
`participants showed some antitumor responses, despite their already “dismal
`
`prognosis.” PO Resp. 27; Ex. 2126, 162:5; Ex. 1075, ¶37. Third, Hammond omits
`
`individual patient data, including number of pemetrexed courses (Ex. 1074, 77:1-
`
`78:4), the type(s) of cancer, (id. 80:7-19), and the amount and type of the
`
`previously-failed therapy (id. 84:3-85:4). These omissions preclude comparisons
`
`to other treatment regimens. Ex. 1075, ¶44. Fourth, Lilly predicates its
`
`comparison on Hammond being a so-called “Phase IB” study, even though the
`
`“Phase IB” format was not well-recognized until well after June 1999. Hammond
`
`6
`
`

`
`neither purports to be nor qualifies as Phase IB. See Ex. 1074, 21:22-28:11;
`
`Ex. 1075, ¶43; Ex. 1063; Compare Ex. 1082-0011 (Lilly website 2/6/2006) with
`
`Ex. 1082-0015 (Lilly website 4/28/2006); Ex. 1063; Ex. 1098.
`
`
`
`Lilly rationalizes its improper comparisons between Phase I studies with
`
`Laohavinij, which Lilly claims made the “very type of comparison for another
`
`antifolate (lometrexol)[.]” PO Resp. 26 (citing Ex. 2031). Lilly omits that
`
`Laohavinij’s brief comparison did not draw any conclusions about efficacy, but
`
`instead postulated that differences could be because “optimal therapeutic
`
`conditions have not been defined,” e.g., MTD. Ex. 2031, p.333; Ex. 1074, 99:24-
`
`100:5. Still, Laohavinij encouraged folic acid pretreatment, concluding
`
`“lometrexol toxicity can be modulated by folic acid supplementation in patients”
`
`and suggesting the reported folic acid study would “facilitate the future
`
`development.” Ex. 2031, pp.333-34; Ex. 1074, 99:24-100:5.
`
`Contrary to Lilly’s assertions, Hammond expresses no concerns about
`
`kidney toxicity with folic acid pretreatment. And with good reason; there is no
`
`basis in Hammond or any other pre-June 1999 study for Lilly’s litigation-inspired
`
`concern about pemetrexed’s impact on the kidneys. Ex. 1075, ¶¶50-53, 58-59;
`
`Ex. 1100, p.2331; PO Resp. 26-29.
`
`Finally, Spectralytics, Inc. v. Cordis Corp., 649 F.3d 1336, 1343 (Fed. Cir.
`
`2011) is inapt. PO Resp. 29. Unlike the Spectralyics prior art, which taught away
`
`7
`
`

`
`from an invention that “embrac[ed] vibrations,” Hammond expressly teaches folic
`
`acid pretreatment with pemetrexed. 649 F.3d at 1343; Ex. 1015.
`
`3.
`
`The ’974 Patent Teaches Folic Acid Pretreatment
`Reduced Toxicity “Without Adversely Affecting
`Therapeutic Efficacy”
`
`Lilly’s ’974 patent discloses that antifolate toxicity for “lometrexol and
`
`related GAR-transformylase inhibitors . . . can be significantly reduced by the
`
`presence of a FBP binding agent [folic acid], without adversely affecting
`
`therapeutic efficacy.” Ex. 1005, 1:47-53. Pemetrexed inhibits several enzymes,
`
`including GAR-transformylase. Ex. 1004, ¶35; Ex. 1075, ¶¶ 60-64. Lilly’s
`
`complaints that the patent has “no data” and “precious little detail” are belied by
`
`three columns of animal data (Ex. 1005, 6:57-8:47), its report of “pilot studies in
`
`humans” (id. 8:49-58), and Lilly’s decision to list the ’974 patent in the FDA’s
`
`Orange Book as covering pemetrexed (Ex. 1025, p.145). The ’974 patent’s
`
`disclosures were not suddenly nullified when the patent expired in 2011 and no
`
`longer protected Lilly’s monopoly on pemetrexed. See Ex. 1025, p.145.
`
`B. Near Contemporaneous FDA Correspondence Confirms The
`Prior Art Taught Vitamin Pretreatment Prior To June 1999
`
`Lilly’s and Dr. Chabner’s interpretations of Worzalla, Hammond, and
`
`Laohavinij in this IPR directly contradict Lilly’s prior characterizations of those
`
`references near or prior to June 30, 2000 (the ’209 patent’s earliest priority date).
`
`In February 2000, Lilly told FDA that Worzalla supported Lilly’s folic
`
`8
`
`

`
`acid/vitamin B12 pretreatment proposal because Worzalla showed it was “possible
`
`to decrease toxicity to healthy tissue while maintaining antitumor effect[.]”
`
`Ex. 1084, ELAP0013767. Lilly claimed Worzalla’s data “support[ed] the
`
`hypothesis that folic acid supplementation can protect healthy tissue from the toxic
`
`effects of [pemetrexed] with retention of antitumor activity.” Id., ELAP0013769;
`
`Ex. 1077, ELAP00199798.
`
`With respect to Hammond, in December 1999, Lilly told FDA that
`
`Hammond “suggest[ed] that folic acid supplementation in this study permits dose
`
`escalation by ameliorating toxicity since heavily and minimally pretreated patients
`
`tolerate [pemetrexed] at doses of 700 and 925 mg/m2 respectively.” Ex. 1077,
`
`ELAP00199798, -809.
`
`Lilly cited Laohavinij to FDA as support for adding a vitamin pretreatment
`
`regimen to its study protocol. Ex. 1077, ELAP00199793, -97 to -98, -814.
`
`Lilly also told FDA that it was known in 1999 that folic acid
`
`supplementation with pemetrexed would be beneficial, noting that it opted to
`
`“proceed with vitamin supplementation (folic acid together with vitamin B12)”
`
`because “external consultants were in unanimous agreement that the addition of
`
`folic acid should lower plasma homocysteine levels and thus offer protection to
`
`patients.” Ex. 1085, ELAP00013458, -59. Yet, Lilly now claims that the very
`
`same prior art it relied on to seek FDA approval taught just the opposite.
`
`9
`
`

`
`III. A POSA WOULD HAVE BEEN MOTIVATED TO ADD VITAMIN
`B12 PRETREATMENT BASED ON NIYIKIZA AND CALVERT
`
`A. Niyikiza And Calvert’s Link Between Pemetrexed Toxicity And
`Homocysteine Provided The POSA Reason To Pretreat With
`Vitamin B12
`
`Lilly does not dispute the core teachings of Niyikiza and Calvert, which
`
`Sandoz’s petition identified as the basis for adding vitamin B12 to the folic acid
`
`pretreatment regimen of Worzalla or Hammond. Paper 2, pp.9-10. As it must,
`
`Lilly acknowledges that “Niykiza I identifies a correlation between certain baseline
`
`homocysteine levels and pemetrexed toxicity” and that “Calvert I further teaches
`
`that elevated homocysteine can be caused by both folic acid and vitamin B12
`
`deficiencies.” PO Resp. 37; Ex. 1074, 150:12-16, 161:1-164:3. Thus, a POSA
`
`would have been motivated to add vitamin B12 to the known folic acid pretreatment
`
`regimen of Worzalla and Hammond I and in doing so, the POSA would have had a
`
`reasonable expectation that both the patient’s homocysteine level would be
`
`reduced and pemetrexed-related toxicities would be reduced. None of Lilly’s
`
`criticisms undercut these core teachings.
`
`10
`
`

`
`Lilly’s distinction between the correlation of homocysteine levels and
`
`pemetrexed toxicity and causation of this toxicity deliberately misses the point.2
`
`PO Resp. 38; Ex. 1075, ¶68. Lilly’s expert, Dr. Zeisel, put it well when he
`
`acknowledged that Niyikiza “suggests that there might be something about having
`
`a higher level than 10 [µM homocysteine] that was related to why people were
`
`getting toxicity from pemetrexed, and that’s what a POSA would have taken out of
`
`that set of studies [Niyikiza].” Ex. 1086, 29:7-12; Ex. 1006. Calvert indicates the
`
`elevated homocysteine levels observed by Niyikiza signaled a functional folate
`
`deficiency, which the POSA would have understood was a cause of pemetrexed
`
`toxicity. Ex. 1075, ¶¶74-75. Regardless, whether the elevated pretreatment
`
`homocysteine levels or the functional folate deficiency ultimately cause non-tumor
`
`cells to be more susceptible to toxicity is irrelevant because the patent’s claims do
`
`not require that the toxicity be caused by a particular physiological basis. Ex. 1001;
`
`Ex. 1075, ¶68; Ex. 1091, ¶41. By June 1999, it was well known that pretreating
`
`with folic acid and vitamin B12 (low levels of which are common causes of
`
`elevated homocysteine) would reduce pretreatment homocysteine levels and the
`
`
`2 Dr. Schiff did not “admit[ ]” that homocysteine does not cause pemetrexed
`
`toxicity (PO Resp. 38), but instead explained at his deposition that “[o]ne can’t
`
`infer that one way or another” from Niyikiza. Ex. 2126, 116:17-120:14.
`
`11
`
`

`
`POSA would have appreciated that the resulting reduced homocysteine correlated
`
`with reduced pemetrexed toxicity.
`
`Niyikiza and Calvert identified a problem to be solved: elevated baseline
`
`homocysteine levels. The ’209 patent’s pretreatment regiment is obvious because
`
`the nature of that problem suggested the invention, i.e., adding folic acid and
`
`vitamin B12 pretreatment to reduce baseline homocysteine. See Ormco Corp. v.
`
`Align Tech., Inc., 463 F.3d 1299, 1307-08 (Fed. Cir. 2006); Ex. 1075, ¶¶ 68, 74.
`
`Lilly thus demands more than what obviousness requires when it complains that
`
`Niyikiza “does not make any recommendations” on how to lower homocysteine
`
`and that Calvert “makes no suggestion as to how the POSA should use the
`
`information [from Niyikiza].” PO Resp. 38-39. Obviousness does not require that
`
`the invention be spelled out in the prior art. See Aventis Pharma Deutschland v.
`
`Lupin, Ltd., 499 F.3d 1293, 1301 (Fed. Cir. 2007); Ex. 2126, 167:20-21.
`
`Lilly also mischaracterizes Niyikiza II in arguing that “Niyikiza II analyzed
`
`whether patients’ MMA levels were correlated with pemetrexed toxicity, and
`
`found they were not.” PO Resp. 40. Not so. Niyikiza II merely notes that no such
`
`correlation “was seen.” Ex. 1016 (emphasis added). As Dr. Schiff explained, and
`
`Drs. Chabner and Zeisel admitted, a POSA would understand only that Niyikiza
`
`did not find the correlation with MMA, and would not exclude the possibility that
`
`it exists. Ex. 1075, ¶¶ 65-73; Ex. 1074 153:7-13, 153:24-154:1; Ex. 1086, 116:19-
`
`12
`
`

`
`20; Ex. 1091, ¶42. Because MMA and homocysteine were known to be highly
`
`correlated to one another, no conclusions can be drawn based on lack of data
`
`showing a correlation. Ex. 1004, ¶80; Ex. 2126, 126:22-130:15; Ex. 1074, 157:23-
`
`158:2. Indeed, Lilly itself explained that the “B12 deficiency marker,
`
`methylmalonic acid, was highly correlated with homocysteine and was therefore
`
`removed from the initial multivariate analysis conducted in 1998 to eliminate
`
`issues of collinearity.” Ex. 1088, ELAP001930.
`
`Lilly mischaracterizes Sandoz’s motivation for adding vitamin B12
`
`pretreatment as the possibility of a correlation between vitamin B12 and toxicity.
`
`PO Resp. 41. Rather, vitamin B12 pretreatment would have been obvious in 1999
`
`because Niyikiza I taught elevated baseline homocysteine levels were linked to
`
`toxicity and the POSA would have known folic acid and vitamin B12 together
`
`reduce homocysteine (e.g., Calvert, EP005). Paper 2, p.10; Ex. 1033. The point
`
`Lilly latches onto was a preemptory rebuttal, which simply pointed out that
`
`Niyikiza II would not deter the POSA from this otherwise obvious vitamin
`
`combination.
`
`B.
`
`EP005 And Nutritional Literature Provide Additional Reason To
`Add Vitamin B12 To The Pretreatment Regimen
`
`Lilly does not dispute that EP005 states that another antifolate
`
`(methotrexate) increases homocysteine and “tells [the POSA] that you can [control
`
`blood homocysteine levels] with a combination of vitamin B12 and folic acid.” Ex.
`
`13
`
`

`
`1033, p.4, 11; Ex. 1074, 185:24-186:7; Ex. 1075, ¶77. Instead, Lilly attempts to
`
`discount EP005’s teachings as “irrelevant” because pemetrexed does not itself
`
`increase homocysteine like methotrexate. PO Resp. 43. But EP005 broadly
`
`teaches administering folic acid and vitamin B12 to lower homocysteine regardless
`
`of cause. See Ex. 1033, p.4; Ex. 1075, ¶¶77, 79-80.
`
`Lilly ignores Niyikiza when it complains that EP005 “provides no reason to
`
`administer folic acid and vitamin B12 pretreatment in the context of pemetrexed
`
`chemotherapy.” PO Resp. 43. Niyikiza teaches that elevated baseline
`
`homocysteine predicts pemetrexed toxicity. Ex. 1016; Ex. 1075, ¶78. EP005
`
`provides powerful motivation to pretreat with a combination of folic acid and
`
`vitamin B12 to produce a synergistic lowering of homocysteine “which go[es]
`
`substantially beyond what might be expected from a simple additive effect of the
`
`action of these drugs.” Ex. 1033, p.11:20-25.
`
`Lilly contradicts itself by criticizing EP005, Ubbink I, Brattstrom,
`
`Bronstrup, and Ubbink II as “irrelevant” to cancer treatment (PO Resp. 43, 48),
`
`while at the same time relying on general nutritional concepts relating to the
`
`methyl trap not tied to cancer. E.g., PO Resp. 6-7, 37. Further, Lilly relied on
`
`some of these same “nutritional” references before the FDA when seeking
`
`approval to pretreat cancer patients with vitamins before administering
`
`pemetrexed. Ex. 1076, ELAP00013782. The ’209 patent’s background also
`
`14
`
`

`
`identifies so-called general nutritional references as relevant and there is no dispute
`
`that a POSA would know of general nutritional concepts. Ex. 1001, 2:16-24; PO
`
`Resp. 14; Ex. 1004, ¶13.
`
`IV. A POSA WOULD HAVE BEEN MOTIVATED IN JUNE 1999 TO
`REDUCE PEMETREXED’S TOXICITY
`
`The POSA would have sought to improve pemetrexed’s therapeutic index
`
`via vitamin pretreatment notwithstanding Lilly’s out-of-context quotes
`
`characterizing pemetrexed’s toxicities’ as “manageable,” “tolerable,” and
`
`“reversible.” PO Resp. 1, 10, 11, 21, 28 (citing Ex. 1045, p.107; Ex. 1047, p.39;
`
`Ex. 2030, p.82; Ex. 1052, p.1198). While the references laud pemetrexed as a
`
`promising antifolate pre-1999, they dispel Lilly’s rosy picture of effectively
`
`managing pemetrexed toxicity. The prior art reports a high incidence of Grade III
`
`and IV toxicities associated with pemetrexed, which Dr. Chabner acknowledged
`
`would be “dangerous.” Ex. 1074, 223:14-224:5; Ex. 1075, ¶¶12-20. One study
`
`reported a 72% incidence of Grade III and IV neutropenia with pemetrexed, more
`
`than double the 30% incidence that Dr. Chabner testified was the maximum
`
`threshold. Ex. 1074, 224:23-225:3 (“You wouldn’t want a 30 percent incidence”),
`
`254:10-13; Ex. 1070. Indeed, Lilly’s own Rusthoven article suggests folic acid
`
`pretreatment to reduce so-called “tolerable” pemetrexed-related toxicity. E.g.,
`
`Ex. 1052, p.1198.
`
`15
`
`

`
`Lilly’s suggestion that a POSA would look to impractical alternatives3
`
`highlights why folic acid and vitamin B12 emerge from the prior art as the lead
`
`candidates to address pemetrexed toxicity. PO Resp. 23, 41-42. As explained by
`
`Dr. Schiff, these “alternatives” do not address the full breadth of pemetrexed’s
`
`toxicities, were unknown for chemotherapy, required cumbersome administration,
`
`or would have negatively impacted efficacy. Ex. 1075, ¶¶21-33; Ex. 1072, p.14;
`
`Ex. 2033, p.2805; Ex.1086, 38:12-39:9; Ex. 1097; Ex. 1091, ¶¶32-35; Ex. 1074,
`
`261:1-262:6, 267:16-25; 276:24-277:25, 278:4-14. And even if viable, these
`
`alternatives would not have taught away from using folic acid and vitamin B12
`
`pretreatment to reduce pemetrexed toxicity. See, e.g., In re Fulton, 391 F.3d 1195,
`
`1201 (Fed. Cir. 2004); In re Mouttet, 686 F.3d 1322, 1334 (Fed. Cir. 2012) (“[J]ust
`
`because better alternatives exist in the prior art does not mean that an inferior
`
`combination is inapt for obviousness purposes.”).
`
`
`3 Lilly points to granulocyte colony stimulating factor (“GCSF”), betaine,
`leucovorin, and dose reductions rescue as alternatives.
`
`16
`
`

`
`V. THE POSA HAD A REASONABLE EXPECTATION OF SUCCESS
`IN USING VITAMIN PRETREATMENT TO REDUCE TOXICITY
`WITHOUT COMPROMISING EFFICACY
`
`A. The Prior Art Indicated Folic Acid Pretreatment Would Be
`Successful
`
`Worzalla, Hammond, and the ’974 patent’s plain statements about the
`
`benefits of folic acid pretreatment would have provided a POSA with a reasonable
`
`expectation of success in using folic acid pretreatment with pemetrexed. That
`
`expectation is not undercut by Lilly’s purported efficacy concerns, which stem
`
`almost entirely from Lilly’s flawed reading of a couple of sentences in an article by
`
`Laohavanij that concerns another antifolate, lometrexol. E.g., Ex. 2120, ¶¶66-68,
`
`96, 100, 166, 197, 200, 221; PO Resp. 9, 20-22, 26, 33 (relying on Ex. 2031,
`
`p.333).
`
`First, as noted previously, Laohavinij found that folic acid pretreatment was
`
`promising with the antifolate in question (lometrexol). See supra p.7; Ex. 1075,
`
`¶¶45-46.
`
`Second, the objectives of Laohavinij’s phase I study did not include
`
`efficacy. Instead, Laohavinij referenced earlier studies by Grindey and Alati on
`
`efficacy, both of which unequivocally reported “[r]eversal of the toxicity but not
`
`the antitumor activity of Lometrexol by folic acid” and thus encouraged folic acid
`
`pretreatment. Ex. 2031, p.326 (refs. 10-11); Ex. 1011; Ex. 1090, p.2331 (ref.16).
`
`17
`
`

`
`Third, Dr. Chabner admitted that his reading of Laohavinij as discouraging
`
`folic acid pretreatment was tainted by his incorrect belief that Lilly abandoned
`
`lometrexol due to Laohavinij. Ex. 1074, 100:18-101:13, 102:11-105:13; Ex. 1012,
`
`p.277. In fact, clinical trials combining lometrexol and folic acid continued well
`
`after Laohavinij – until at least 2002. Ex.1075, ¶48; Ex. 1099, NIH.
`
`Fourth, Laohavinij relies on a 1948 Farber paper that concerned a different
`
`antifolate (aminopterin) that works primarily by inhibiting dihydrofolate reductase
`
`(“DHFR”), which is not the primary target of pemetrexed. Ex. 1009, pp.787, 792;
`
`Ex. 1091, ¶17; Ex. 1075, ¶¶95-96. In June 1999, the POSA would understand that
`
`this 50-year old paper concerning childhood acute leukemia has limited application
`
`for a multi-targeted antifolate like pemetrexed, particularly given more recent
`
`studies like Worzalla and Hammond evidencing that folic acid pretreatment
`
`reduced pemetrexed’s toxicity while preserving efficacy. See Ex. 1074, 179:5-10;
`
`Ex. 1075, ¶¶34, 86, 95-96. The same holds true for Laohavinij itself: the POSA
`
`would look to Worzalla and Hammond’s more recent teachings about the benefits
`
`of folic acid pretreatment with pemetrexed over Laohavinij’s earlier lometrexol
`
`study.
`
`Fifth, Lilly attempts to explain its incorrect reading of Laohavinij by
`
`arguing that folic acid has the same effect on both cancer and healthy cells, making
`
`folic acid a purported “antidote” to pemetrexed. PO Resp. 2, 5, 19; Ex. 2120,
`
`18
`
`

`
`¶178; Ex. 2118, ¶49-50. This is untrue. Cancer cells’ rapid division and high rates
`
`of folate catabolism make them more susceptible to antifolates than normal cells.
`
`Ex. 1012, p.270; Ex. 1004, ¶29; Ex. 1091, ¶¶18-24.¶
`
`B. A POSA Likewise Would Have Had No Concerns About
`Vitamin B12 Pretreatment
`
`Lilly’s purported concern that vitamin B12 would “feed the tumor” by
`
`releasing large amounts of folate from the “methyl trap” relies on the same flawed
`
`premise as its argument that a POSA would be concerned that folic acid would
`
`reduce pemetrexed’s efficacy. Ex. 1086, 141:3-16; Ex. 1075, ¶97; PO Resp. 35-
`
`37. To the extent any such concern ever existed, it was debunked for pemetrexe