`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`_________________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`_________________
`
`LUPIN LTD. and LUPIN PHARMACEUTICALS INC.
`
`Petitioners,
`
`v.
`
`HORIZON THERAPEUTICS, LLC
`
`Patent Owner.
`_________________
`
`IPR2017-00829
`U.S. Patent No. 9,095,559
`
`
`
`PETITIONERS’ REPLY
`
`
`
`
`Mail Stop PATENT BOARD
`Patent Trial and Appeal Board
`United States Patent and Trademark Office
`PO Box 1450
`Alexandria, Virginia 22313-1450
`
`
`
`

`

`
`
`TABLE OF CONTENTS
`
`Page
`
`1. Horizon Mischaracterizes Both the Claims and State of the Prior Art ............... 1
`(a) POSAs Used Ammonia Levels to Adjust Drug Dosage ............................... 2
`(b) A POSA Would Have Been Motivated to Adjust Drug Levels in Patients
`Near the ULN Because the Prior Art Taught POSAs to Maintain Stable
`Plasma Ammonia Levels .............................................................................. 3
`(c) The Prior Art Does Not Demonstrate Complacency When Plasma
`Ammonia Was At or Under the ULN ........................................................... 5
`(d) Claim 5 is Not Patentable ............................................................................ 10
`(e) POSAs Used Fasting Plasma Ammonia Levels .......................................... 11
`2. There Was Motivation to Combine the Cited Prior Art References ................. 12
`(a) Any Purported Differences Between the Nitrogen Scavenging Drugs in
`Simell and the ’859 Publication are Irrelevant to Simell’s Role in the
`Unpatentability Analysis ............................................................................. 12
`(b) LPI is a UCD ............................................................................................... 14
`(i) A POSA Would Have Combined the Cited References Even Though They
`Do Not All Address Dosing Based on Normal Fasting Plasma Ammonia
`Levels .......................................................................................................... 16
`3. Horizon’s Arguments Regarding POSA Level and Dr. Vaux’s Qualifications
`Should be Rejected ............................................................................................ 18
`(a) Horizon’s Proposed POSA Level is Artificially Stringent ......................... 18
`(b) Dr. Vaux Treats UCD Patients and is an Expert in the Field of the Claimed
`Subject Matter ............................................................................................. 21
`4. Lupin Has Shown a Reasonable Expectation of Achieving the Claimed Subject
`Matter, and Horizon’s Legally Flawed Argument to the Contrary Should be
`Rejected ............................................................................................................. 22
`5. Conclusion ......................................................................................................... 24
`
`

`

`
`
`1. Horizon Mischaracterizes Both the Claims and State of the Prior Art
`
`Patent Owner Horizon greatly overstates the scope of the claimed subject
`
`matter, mischaracterizing it as “an ingenious solution” to the “problem” of “how to
`
`treat a patient suffering from a nitrogen retention disorder,” while referencing
`
`“accumulation of potentially fatal levels of ammonia.” (POR at 1.) Methods of
`
`using glyceryl tri-[4-phenylbutyrate] to lower plasma ammonia levels were known
`
`long before the ’559 patent, and Horizon did not solve the problem of how to lower
`
`fatal ammonia levels. Indeed, the ’559 patent claims are not even related to
`
`treating acute hyperammonemia. They are instead narrowly directed to fine-tuning
`
`or optimizing drug dosage in patients who already have normal fasting plasma
`
`ammonia levels, based on comparing the patient’s fasting plasma ammonia levels
`
`to the ULN for plasma ammonia. The use of ammonia as a biomarker for
`
`evaluating drug dosages was well known; fine-tuning drug dosage was well within
`
`the ordinary skill of physicians; and the effect of carrying out the claimed methods
`
`would be to routinely maintain normal plasma ammonia levels—a goal that was
`
`fully reported in the prior art. Because the challenged claims are to nothing more
`
`than “the predictable use of prior art elements according to their established
`
`functions,” KSR Int’l Co. v. Teleflex Inc., 550 U.S. 398, 417 (2007), they should be
`
`cancelled as obvious.
`
`
`
`

`

`
`
`(a) POSAs Used Ammonia Levels to Adjust Drug Dosage
`
`In arguing that plasma ammonia levels “were not a basis upon which to increase
`
`the dosage of a drug” (POR at 46), Horizon directly contradicts the prior art. For
`
`example, the ’857 Publication—which shares an inventor with the ’559 patent—
`
`clearly teaches using ammonia levels to determine whether to adjust drug dosage.
`
`(See Petition at 22-23.)
`
`Horizon lists several purported drawbacks of using ammonia as a biomarker,
`
`none of which change the fact that POSAs used ammonia levels when adjusting
`
`drug dosage. And the claimed methods fail to address these purported drawbacks
`
`in any event. For example, citing Ex. 2015, Horizon contends that inherent
`
`difficulties with the interpretation of blood ammonia levels “undermined its
`
`usefulness as a diagnostic tool.” (POR at 12, 49.) This argument misses the mark.
`
`Ex. 2015 discusses the difficulty in diagnosing a specific UCD based on the
`
`specific numerical level of ammonia (Ex. 2015 at 75), but the ’559 claims do not
`
`cover using a specific numerical value of ammonia to diagnose diseases. Thus
`
`Horizon did not solve this purported problem.
`
`Nor have the ’559 patent claims solved any of the other purported problems
`
`with ammonia that Horizon raised, including diurnal fluctuation; artificially high
`
`levels if a patient is catabolic due to fasting, exercise, surgery, an infection, or
`
`pregnancy; and lack of correlation of ammonia levels with clinical status. (POR at
`
`
`
`2
`
`

`

`
`
`12, 48-50.) The prior art disclosed all of these issues with ammonia, yet POSAs
`
`still used ammonia levels to adjust drug dosage. (See, e.g., Ex. 1007 at [0232]
`
`(“[D]ose adjustment would be based on repeated measurement of urinary PAGN as
`
`well as assessment of dietary protein and plasma ammonia.”).)
`
`(b) A POSA Would Have Been Motivated to Adjust Drug Levels in
`Patients Near the ULN Because the Prior Art Taught POSAs to
`Maintain Stable Plasma Ammonia Levels
`
`In trying to undercut a motivation to combine, Horizon disputes that a goal
`
`of nitrogen scavenging therapy is to maintain a “stable” plasma ammonia level.
`
`(POR at 33-34.) Once again, Horizon ignores the express teachings of the prior
`
`art, which make clear that maintenance of plasma ammonia levels within normal
`
`limits, and below the ULN, is one of the objectives of drug therapy. See, e.g., Ex.
`
`1007 at [0046] (noting desire to “maintain a stable level of plasma ammonia”);
`
`[0182] (noting that HPN-100-treated subjects will typically “achieve and maintain
`
`normal plasma ammonia levels”); [0209] (discussing consistent reduction of
`
`ammonia to below about 40 µmol/L); Ex. 1020 at 3327 (noting that plasma
`
`ammonia “should be maintained within normal limits”). Horizon further asserts
`
`that due to the known variability in plasma ammonia levels, a stable level cannot
`
`be achieved. (POR at 34.) But that is exactly Petitioners’ point: because ammonia
`
`levels were known to vary, a POSA would have been motivated to keep the
`
`baseline ammonia levels low, such that despite transient fluctuations, e.g. in
`
`
`
`3
`
`

`

`
`
`response to meals, a patient would be stably maintained within normal limits—a
`
`goal that is clearly reflected in the prior art statements cited above.
`
`Horizon asserts that the prior art did not indicate what should be adjusted or
`
`changed in a patient with a normal plasma ammonia level, citing Procter &
`
`Gamble Co. v. Teva Pharm. USA, Inc., 566 F.3d 989, 996-97 (Fed. Circ. 2009).
`
`(POR at 35.) This case is inapposite. There, the claimed chemical compound was
`
`found non-obvious where the prior art gave no guidance as to the effect of
`
`modifications needed to transform a prior art compound into the claimed
`
`compound. Teva, 556 F.3d at 996-97. That is nothing like the case here, where
`
`factors that affect plasma ammonia were well-known, and thus how to lower and
`
`maintain plasma ammonia levels was well-known. For example, the ’859
`
`Publication taught that “[d]ietary protein intake or drug dosage or both could be
`
`adjusted to attain a normal or desired plasma ammonia level.” (Ex. 1007 at
`
`[0226].)
`
`This same disclosure undercuts Horizon’s contention that when optimizing
`
`therapy for a UCD patient who already has a normal plasma ammonia level,
`
`“clinicians wound not have increased drug dosage, but instead would have
`
`evaluated their diet or health.” (POR at 35.) Clearly the prior art taught evaluating
`
`not only diet, but ammonia and drug levels too. (Ex. 2007 at [0226], [0016]
`
`(“Dietary protein can be limited, but a healthy diet requires a significant amount of
`
`
`
`4
`
`

`

`
`
`protein, particularly for growing children thus in addition to controlling dietary
`
`protein intake, drugs that assist with elimination of nitrogen are used to reduce
`
`ammonia build-up (hyperammonemia).”).) Horizon’s citation of Ex. 2019 is
`
`unavailing. That reference, dated 2012, is not prior art because it published after
`
`the claimed September 30, 2011 priority date. (Ex. 2006 ¶2.) See 35 U.S.C. § 103
`
`(a patent may not be obtained if the claimed would have been obvious before the
`
`effective filing date of the claimed invention).
`
`(c) The Prior Art Does Not Demonstrate Complacency When Plasma
`Ammonia Was At or Under the ULN
`
`Horizon takes an unreasonably myopic view of the prior art, positing that it
`
`“consistently taught that there was no need to change the course of treatment for a
`
`UCD patient if that patient’s plasma ammonia level was within the normal range,”
`
`and contending that POSAs were focused only “on avoiding extremely high
`
`plasma ammonia levels.” (POR at 2; 4 (“none of the art suggests that there is any
`
`need to continue to tinker with what has already been fixed and considered
`
`normal”).) Horizon points to no express teaching supporting this view. No
`
`instruction in the prior art directs POSAs to focus only on “extremely high”
`
`ammonia levels, or to stop titrating drug dosage as soon as a patient’s plasma
`
`ammonia level hits the ULN.
`
`According to Horizon, the ’859 Publication teaches increasing dosage only
`
`when plasma ammonia levels are inadequate, which Horizon misleadingly equates
`
`
`
`5
`
`

`

`
`
`with “above normal.” (POR at 4, 40, 51-52.) In fact, the ’859 Publication’s
`
`disclosure is broader. Nowhere does it limit “inadequate” plasma ammonia levels
`
`to those “above normal.” Instead, it teaches POSAs to adjust dosage to achieve a
`
`“normal” or “desired” ammonia level: “Dietary protein intake or drug dosage or
`
`both could be adjusted to attain a normal or desired plasma ammonia level, e.g., a
`
`level below about 40 µmol/L.” (Ex. 1007 at [0226] (emphasis added).) It also
`
`makes clear that any value less than the ULN is appropriate: “a plasma ammonia
`
`level of less than about 40 µmol/L, or of not greater than 35 µmol/L would
`
`indicate that the treatment was effective.” (Ex. 1007 at [0074], [0085]) (emphasis
`
`added). Accordingly, the ’859 Publication teaches a range of acceptable and
`
`desirable plasma ammonia levels, and nowhere teaches away from continuing to
`
`treat patients who are already at or below the ULN. It also teaches that a benefit of
`
`glyceryl tri-[4-phenylbutyrate] is that it leads to average ammonia levels that are
`
`typically lower than those achieved with other nitrogen scavenging drugs. (Ex.
`
`1007 at [0209].) This demonstrates that POSAs were not merely content to reach
`
`the ULN, but in fact recognized that lower ammonia levels are better.
`
`Horizon points to no prior art evidencing complacency in the face of normal
`
`plasma ammonia levels. It cites the ’157 Publication (Ex. 2012), another
`
`publication by the same inventor of the ’559 patent, for a purported teaching to
`
`
`
`6
`
`

`

`
`
`adjust dosage only when ammonia is above the ULN.1 (POR at 42.) But the ’157
`
`Publication does not teach adjusting drug dosage only when PAGN levels correlate
`
`with plasma ammonia levels above the ULN, as Horizon contends. (Id.) Rather, it
`
`teaches that POSAs should customize treatment for individual patients, stating that
`
`even where PAGN output exceeds 10 g/day (a level that purportedly correlates to
`
`ammonia levels ≤ the ULN), POSAs may monitor ammonia to assess whether the
`
`dosage is generally sufficient. (Ex. 2012 at [0297].) And, like the ’859
`
`Publication, the ’157 Publication suggests adjusting drug dosage to attain “a
`
`normal or desired plasma ammonia level,” meaning that the disclosure is not
`
`limited to increasing dose only when ammonia is above the ULN. (Id. at [0299]
`
`(emphasis added).)
`
`Horizon also misreads Berry (Ex. 1016) as disclosing the increase of
`
`ammonia scavenging drugs only when the patient’s plasma ammonia levels are 3x
`
`the ULN. (POR at 43.) Berry provides guidelines in the case of intercurrent
`
`illness, when protein breakdown can cause rapid ammonia accumulation. (Ex.
`
`1016 at 4-5.) In such cases, Berry suggests that even where the patient’s “plasma
`
`ammonia level is <100 µmol/L,” POSAs should increase drug dosage by
`
`approximately 50% (id. at 4-5), as a prophylactic against ammonia accumulation.
`
`1 Horizon also points to Ex. 2019 (see POR at 42-43), but this reference is not prior
`
`art. See supra pg. 5.
`
`
`
`7
`
`

`

`
`
`And, because levels “<100 µmol/L” include normal levels, Berry actually instructs
`
`POSAs to increase drug dosage even for patients with normal ammonia levels—
`
`contrary to Horizon’s argument.
`
`Horizon also cited Barsotti (Ex. 1015) and Consensus (Ex. 2025) for the
`
`proposition that practitioners cared only about ammonia levels far above normal.
`
`(POR at 44.) Again, Horizon takes the references out of context. Barsotti
`
`recognizes ammonia levels in excess of 1 mmol/L as being associated with acute
`
`hyperammonemia (Ex. 1015 at 1), but nowhere suggests that only levels above this
`
`amount indicate abnormality in nitrogen homeostasis. Horizon contends that the
`
`Consensus shows that there was no agreement as to whether to use nitrogen
`
`scavenging drugs when plasma levels are 3x the ULN, but that statement relates to
`
`a threshold for when emergency treatment with IV therapy should be started. (Ex.
`
`2025 at 3.) Horizon ignored the more pertinent statement that such an elevated
`
`ammonia level “does require adjustment of therapy or better compliance with the
`
`recommended treatment regimen.” (Id.)
`
`And while Horizon cites two references indicating that above-normal plasma
`
`ammonia levels are acceptable (see POR at 44), other references clearly establish
`
`that the goal was to maintain plasma ammonia levels below the ULN. (See, e.g.,
`
`Ex. 1016 at 3 (“The goal of treatment is to maintain normal levels of plasma
`
`ammonia through the use of the low-protein diet and medication while allowing for
`
`
`
`8
`
`

`

`
`
`normal growth. . . . The biochemical targets for optimal UCD control are noted in
`
`Table VI. These include a plasma ammonia level of <40 µmol/L . . . .”); Ex. 1007
`
`at [0226], [0074], [0085].)
`
`Horizon also argues that because treating UCDs is so complicated, and
`
`because it was only recently that patients began surviving UCD diagnoses, that
`
`clinicians were content to achieve a normal plasma ammonia level and stop further
`
`optimization. (POR at 40.) Survival rates began improving following the
`
`introduction of nitrogen scavenging drugs in the 1970’s. (See Ex. 2017 at 1; Ex.
`
`1026 at 31:9-23.) This does not mean that clinicians would remain complacent and
`
`happy with mere survival. It is the natural inclination of POSAs to keep
`
`innovating, and the prior art as a whole shows a continued focus on optimizing
`
`care for UCD patients.
`
`Finally, in an attempt to suggest a hesitancy to increase drug dosage,
`
`Horizon posits that clinicians typically use the lowest possible dosage of any
`
`medication to minimize side effects. (POR at 44.) But Horizon only cites
`
`evidence of massive overdoses due to prescribing errors. (Id. at 44-45.) In fact,
`
`the prior art showed that at the dosage levels at issue here, toxicity was not a
`
`concern:
`
`. . . HPN-100 exhibits no indications of toxicity at equimolar doses
`when compared to the approved PBA dosage of 20 g/day and a dose
`2-3 times the equivalent of 20 grams of PBA is unlikely to produce
`PAA blood levels leading to AEs [adverse events]. Moreover,
`
`
`
`9
`
`

`

`
`
`tolerability of taking HPN-100 is much higher than for PBA . . . . In
`some patients or clinical settings, HPN-100 doses well above the
`approved PBA dosage are expected to be beneficial . . . .
`
`(Ex. 1007 at [0086]; see also [0203]-[0204]; Ex. 2009 at 4 (“[A]lternative pathway
`
`therapies have been found to be remarkably nontoxic in humans at the doses
`
`recommended to treat patients with urea cycle disorders.”).)
`
`Horizon’s efforts to liken this case to Leo Pharm. Prods. Ltd. v. Rea, 726
`
`F.3d 1346 (Fed. Cir. 2013), where the inventors were the first to recognize and
`
`then solve a problem, should be rejected. (POR at 38.) Here, Horizon fails to
`
`articulate exactly what problem was purportedly recognized and solved. In any
`
`event, as Dr. Vaux explained, POSAs were aware that even patients with normal
`
`ammonia levels could experience events that routinely take them outside of normal
`
`limits, which was contrary to the goal of maintaining stable levels under the ULN.
`
`(Ex. 1002 at ¶51; Petition at 24-25.) Accordingly, Horizon was by no means the
`
`first to recognize that the inherent variability of ammonia levels presented an
`
`obstacle to maintaining stable plasma ammonia levels, even in seemingly well-
`
`controlled patients.
`
`(d) Claim 5 is Not Patentable
`
`Horizon also contends that the prior art did not suggest targeting any specific
`
`plasma ammonia level below the ULN, including below half the ULN as recited in
`
`claim 5. (POR at 6, 58.) But as discussed, the prior art clearly taught POSAs to
`
`
`
`10
`
`

`

`
`
`administer drug to obtain normal ammonia levels, i.e. any value under the ULN,
`
`which includes values less than half the ULN, as recited in claim 5. (See, e.g., Ex.
`
`1007 at [0085], [0226]; Ex. 1016 at 3.)
`
`Because the prior art teaches a range that includes the claimed plasma
`
`ammonia level, the claimed value is prima facie obvious. See Iron Grip Barbell
`
`Co., Inc. v. USA Sports, Inc., 392 F.3d 1317, 1321-22 (Fed. Cir. 2004); In re
`
`Peterson, 315 F.3d 1325, 1329 (Fed. Cir. 2003). Although it had the burden to do
`
`so, Horizon has not submitted any evidence that the claimed ammonia value
`
`exhibits unexpected results over the prior art disclosure of ammonia values that are
`
`less than the ULN. See In re Peterson, 315 F.3d at 1330; Petition at 21-22. While
`
`Horizon contends that the claimed subject matter provides “assurance that a patient
`
`would have an optimally controlled ammonia level and would be unlikely to suffer
`
`from hyperammonemia” (POR at 3), it hasn’t supported this contention with any
`
`evidence, or shown how this purported result is different than what existed in the
`
`prior art.
`
`(e) POSAs Used Fasting Plasma Ammonia Levels
`
`Horizon’s contention that prior to the ’559 patent, clinicians “did not take
`
`into account whether a plasma ammonia level was taken in a fasted or fed state
`
`when making treatment decisions” (POR at 2), directly contradicts the prior art.
`
`Use of fasting plasma ammonia levels was routine (see Petition at, e.g., 15-16 and
`
`
`
`11
`
`

`

`
`
`n.2, 19-20, 25-26; see also Ex. 1005 at 2; Ex. 1006 at 17; Ex. 1015 at 1, Ex. 1010),
`
`and is even acknowledged in Horizon’s own exhibits. Ex. 2033 states: “[F]asting
`
`plasma levels of ammonium, branched chain amino acids, arginine, and serum
`
`plasma protein should be maintained within normal limits . . . .” (Ex. 2033 at 47
`
`(emphasis added).) Ex. 2024 states the same. (Ex. 2024 at 35, Table 8 (see *).)
`
`Ex. 2020 states that in cases of suspected hyperammonemia, “[e]xcept in
`
`emergency situations, sampling should be done in fasting state or at least 4-6 h
`
`after a meal,” citing Barsotti (Ex. 1015). (Ex. 2020 at 6 (emphasis added).)
`
`2.
`
`There Was Motivation to Combine the Cited Prior Art References
`
`In disputing motivation to combine, Horizon lobs several unfounded
`
`criticisms at the Simell reference. These criticisms do not undercut the utility a
`
`POSA would have gotten from Simell, which is the suggestion to use fasting
`
`plasma blood levels.
`
`(a) Any Purported Differences Between the Nitrogen Scavenging
`Drugs in Simell and the ’859 Publication are Irrelevant to Simell’s
`Role in the Unpatentability Analysis
`
`According to Horizon, differences in the metabolism, routes of
`
`administration, and indications between the sodium benzoate and phenylacetate
`
`used in Simell and the glyceryl tri-[4-phenylbutyrate] used in the ’859 Publication
`
`would have discouraged POSAs from combining these references. (POR at 5.)
`
`Petitioners and the Board, however, relied on Simell for use of a fasting plasma
`
`
`
`12
`
`

`

`
`
`ammonia level. (Petition at 25-26; Institution Decision at 8.) Horizon fails to
`
`establish why the purported differences between the drugs have any bearing on this
`
`issue.
`
`Horizon further contends that Petitioners have not identified anything
`
`“indicating that the dosing of sodium benzoate or phenylacetate is interchangeable
`
`with glyceryl tri-[4-phenylbutyrate].” (POR at 27.) Even if interchangeable
`
`dosing mattered to the way Simell is being used here (which it does not), Horizon
`
`ignores that glyceryl tri-[4-phenylbutyrate] is metabolized to phenylacetate. (Ex.
`
`1007 at [0021].) The prior art even provides straightforward methods of
`
`converting a patient to glyceryl tri-[4-phenylbutyrate] based on the patient’s
`
`current dosage of phenylacetate. (See, e.g., Ex. 1007 at [0116]-[0119]). Further,
`
`although the ’859 Publication is silent with respect to the route of administration of
`
`the phenylacetate, as Dr. Enns acknowledged, this drug has only ever been
`
`available for intravenous administration. (Ex. 2006 at ¶90; Ex. 1026 at 29:21-
`
`30:11.) Given the teachings of the prior art, any purported differences between
`
`these drugs would not have been an obstacle to combining Simell with the ’859
`
`Publication.
`
`Horizon’s contention that the drugs in Simell and the ’859 Publication have
`
`different indications (POR at 26) should also be rejected as based solely on post-
`
`art. RAVICTI was approved in 2013 (Paper 9 at 1), and its purported label (Ex.
`
`
`
`13
`
`

`

`
`
`2041) shows a 2016 date. This is long after the priority date, making any
`
`differences in the indications for the drugs in Simell and the ’859 Publication
`
`irrelevant to obviousness. See 35 U.S.C. § 103.
`
`(b) LPI is a UCD
`
`Simell shows the use of a fasting plasma ammonia level in a protocol that
`
`uses a nitrogen scavenging drug to reduce plasma ammonia level. (Ex. 1005 at 1-
`
`2.) Horizon contends that POSAs would not have considered Simell because the
`
`patients suffered from lysinuric protein intolerance (“LPI”), which Horizon
`
`contends is not a UCD. (POR at 27.) This is contrary to Simell itself, which
`
`unambiguously refers to LPI as a UCD: “We studied metabolic changes caused by
`
`these substances and their pharmacokinetics in a different urea cycle disorder,
`
`lysinuric protein intolerance (LPI) . . . .” (Ex. 1005 at 1 (Abstract) (emphasis
`
`added).)
`
`Simell is not alone in referring to LPI as a UCD or discussing LPI in the
`
`context of UCD treatments. At least eight of Horizon’s own exhibits do the same.
`
`For example:
`
`
`
`In Ex. 2022, Dr. Enns discusses the treatment of hyperammonemia,
`
`including for UCDs. He cited Simell for its teachings on the pharmacokinetics and
`
`adverse reactions of sodium phenylacetate/sodium benzoate. (Ex. 2022 at 6, 8.)
`
`Dr. Enns nowhere stated that Simell’s data was irrelevant other UCD patients.
`
`
`
`14
`
`

`

`
`
`
`
` Ex. 2009, titled “Alternative Pathway Therapy for Urea Cycle
`
`Disorders: Twenty Years Later,” and Ex. 2018, titled “Alternative Pathway
`
`Therapy for Urea Cycle Disorders,” both review the use of nitrogen scavenging
`
`drugs to lower ammonia, and discuss Simell’s pharmacokinetic data. (Ex. 2009 at
`
`2, 3; Ex. 2018 at 5.) Again, at no point did the authors state that Simell’s data is
`
`irrelevant to UCD patients or relevant only to LPI patients.
`
`Additional references likewise address LPI in connection with UCD
`
`treatment. (See, e.g., Ex. 2016 at 2, Table 1 (listing LPI in a table titled “Urea
`
`cycle and related disorders”); Ex. 2013 at 1-2 (discussing LPI in the context of
`
`“Urea Cycle and Related Disorders”); Ex. 2023 at 2 (stating that LPI “should also
`
`be included in urea cycle disorders”); Ex. 2015 at 3, Table 2 (listing LPI as a
`
`UCD); Ex. 2044 at 2, Table 1 (same).)
`
`Simell teaches the use of nitrogen scavenging drugs to reduce plasma
`
`ammonia levels—subject matter squarely related to the claims. This disclosure is
`
`relevant to UCD treatment, as exemplified by the many references that refer to LPI
`
`as a UCD, and even cite Simell itself. Accordingly, Horizon’s attacks on the
`
`relevance of Simell should be rejected. See Innovation Toys, LLC v. MGA Entm’t,
`
`Inc., 637 F.3d 1314, 1321 (Fed. Cir. 2011) (“A reference is reasonably pertinent if .
`
`. . it is one which, because of the matter with which it deals, logically would have
`
`commended itself to an inventor’s attention in considering his problem.”).
`
`
`
`15
`
`

`

`
`
`(i) A POSA Would Have Combined the Cited References
`Even Though They Do Not All Address Dosing Based
`on Normal Fasting Plasma Ammonia Levels
`
`Horizon disputes the motivation to combine because the references “do not
`
`address the dosing of a nitrogen scavenging drug based on normal fasting plasma
`
`ammonia levels.” (POR at 29.) Horizon improperly focuses on what each
`
`reference teaches separately, not on the teachings of the prior art as a whole. See
`
`In re Merck & Co., 800 F.2d 1091, 1097 (Fed. Cir. 1986); In re Keller, 642 F.2d
`
`413, 425 (C.C.P.A. 1981).
`
`Horizon’s attempt to distinguish Simell and Brusilow 1984 by arguing that
`
`these references disclose choosing a dosage based on body weight, not plasma
`
`ammonia (POR at 30), is misleading. While the prior art (and the ’559 claims) use
`
`ammonia values as a factor in considering whether to adjust the dosage of the
`
`nitrogen scavenging drug, neither posits a quantitative relationship between the
`
`value of the ammonia level and drug dosage. (See, e.g., Ex. 1026 at 63:20-65:3.)
`
`Accordingly, Horizon’s criticism misses the mark. Horizon also contends that
`
`Simell and Brusilow 1984 do not place significance on fasting plasma ammonia
`
`levels (POR at 30-31), but this needlessly picks apart individual prior art references
`
`rather than viewing the prior art as a whole. As discussed in the petition, POSAs
`
`were well aware of good reasons to use fasting plasma ammonia levels (see
`
`
`
`16
`
`

`

`
`
`Petition at, e.g., 15-16 and n.2, 19-20, 25-26), and would have been aware of those
`
`reasons whether or not these two references discuss why to use fasting levels.
`
`Horizon also argues that there is no motivation to combine because Simell
`
`and Brusilow 1984 do not address treatment for a patient who already has normal
`
`ammonia levels. (POR at 30, 31, 41.) This again overlooks the larger context. As
`
`discussed here and in the Petition, POSAs were well aware of good reasons to
`
`maintain patients at normal ammonia levels, and would have been aware of these
`
`reasons whether or not these two references disclose treatment of patients with
`
`normal levels.
`
`Horizon again argues that Blau is irrelevant because it is concerned only
`
`with diagnosis, not treatment, of UCDs. (POR at 32.) As Dr. Vaux testified,
`
`treatment and diagnosis are intimately intertwined. (Ex. 2034 at 23:20-24:14.)
`
`Moreover, Petitioners cite Blau for its disclosure of using fasting plasma ammonia
`
`levels—a fact that is well-supported in the prior art. (Petition at 25; see also supra
`
`Section 1(e); Ex. 2034 at 41:17-25.)
`
`Horizon’s arguments regarding lack of motivation to combine are not
`
`persuasive because they do not address Petitioners’ proposed combination.
`
`
`
`17
`
`

`

`
`
`3. Horizon’s Arguments Regarding POSA Level and Dr. Vaux’s
`Qualifications Should be Rejected
`
`(a) Horizon’s Proposed POSA Level is Artificially Stringent
`
`Horizon spends six pages of its reply criticizing Lupin’s proposed POSA
`
`level and proffering an alternative. Horizon’s emphasis on this issue is curious,
`
`because it never argues that the differences in proposals affects the outcome here.
`
`Horizon criticizes Lupin’s proposal because it does not detail the quantum of
`
`“specialized training” needed in the diagnosis or treatment of inherited metabolic
`
`disorders, such as UCDs. (POR at 17.) Horizon suggests that the training received
`
`during a general pediatrics residency is insufficient, but in fact, one of the core
`
`competencies for achieving board certification in pediatrics is specialized training
`
`in the diagnosis and acute and long-term management of UCDs. (Ex. 1025 at 26;
`
`Ex. 1026 at 18:25-23:8.) Moreover, as Dr. Vaux explained, “specialized” training
`
`includes direct experience with inherited metabolic diseases such as UCDs, or
`
`formal training such as classwork or clinical experience. (Ex. 2034 at 21:5-16.)
`
`Horizon also criticizes Lupin’s definition for referencing diagnosis or
`
`treatment, contending that a physician trained in only diagnosing UCDs would not
`
`understand treatment of UCD patients. (POR at 19.) But as Dr. Vaux testified,
`
`POSAs experienced in diagnosing UCDs would also be experienced in treating
`
`them, and vice versa. (Ex. 2034 at 23:11-24:14).
`
`
`
`18
`
`

`

`
`
`Finally, Horizon criticizes Lupin’s definition because it relates to any
`
`inherited metabolic disorder, not just UCDs. But it cannot be disputed that Lupin’s
`
`definition includes POSAs trained in UCDs (Petition at 8), rendering this criticism
`
`meaningless.
`
`The main distinction between the parties’ proposals is that Horizon requires
`
`two specific board certifications. (POR at 16.) Horizon never explains why these
`
`certifications are required, stating only that they demonstrate that Dr. Enns
`
`possesses “‘the ability to integrate clinical and genetic information and understand
`
`the uses, limitations, interpretations, and significance of specialized laboratory and
`
`clinical procedures.’’’ (Id.) But Horizon hasn’t explained what clinical and
`
`genetic information needs to be integrated to practice the claims or understand the
`
`prior art, hasn’t identified any “specialized laboratory and clinical procedures” (let
`
`alone any relating to UCDs), and hasn’t specified what uses, limitations, or
`
`interpretations of such procedures are pertinent. Nor has it shown that the inventor
`
`or authors of the cited prior art references have such dual board certifications.
`
`(See, e.g., Ex. 1026 at 15:7-11.)
`
`The cited references are representative of the level of ordinary skill in the
`
`art, and make clear that Horizon’s requirements are artificially stringent. See
`
`Okajima v. Bourdeau, 261 F.3d 1350, 1355 (Fed. Cir. 2001). For example, Ex.
`
`2041 states: “RAVICTI should be prescribed by a physician experienced in
`
`
`
`19
`
`

`

`
`
`management of UCDs.” (Ex 2041 at 2.) Dr. Vaux is such a physician. (Ex. 1002,
`
`¶1.) The label does not state that the drug should be administered or the dosage
`
`adjusted only by someone with the dual board certifications proposed by Horizon.
`
`Additionally, Horizon’s Exhibit 2040, titled “Current Strategies for the
`
`Management of Neonatal Urea Cycle Disorders,” acknowledges that the
`
`“pediatrician or primary care provider plays a crucial role in caring for a patient
`
`with a UCD.” (Ex 2040 at 9.) And Ex. 1007 plainly teaches “physicians” and
`
`“doctors” how to adjust doses of glyceryl tri-[4-phenylbutyrate]. (Ex. 1007 at
`
`[0223], [0224].) Neither reference specifies a need for dual board certifications.
`
`While it is true that hyperammomenic crises can be life-threatening events
`
`and must be treated by a multidisciplinary team (see, e.g., Ex