IN THE UNITED STATES DISTRICT COURT
`FOR THE DISTRICT OF DELAWARE
`
`C.A. No.: 1:20-cv-00755-RGA
`
`HIGHLY CONFIDENTIAL
`
`)))))))))))
`
`UNITED THERAPEUTICS
`CORPORATION,
`
`Plaintiff,
`
`v.
`LIQUIDIA TECHNOLOGIES, INC.,
`Defendant.
`
`SUPPLEMENTAL DECLARATION OF ROBERT R. RUFFOLO, PH.D.
`I, Robert R. Ruffolo, Ph.D., declare as follows:
`1.
`I have been retained by counsel for United Therapeutics Corporation (“UTC”) as
`an expert consultant to UTC, Plaintiff in the above-captioned litigation.
`2.
`I submitted a declaration (“Opening Declaration”) in support of UTC’s Opening
`Claim Construction Brief with Respect to U.S. Patent Nos. 9,593,066 and 9,604,901, which was
`dated February 5, 2021.
`3.
`In my Opening Declaration, I set forth my opinion that a POSA would
`understand the plain and ordinary meaning of the claim term “contacting the solution
`comprising treprostinil from step (b) with a base to form a salt of treprostinil” that appears in
`claims 1 and 8 of the ’901 patent, and that this term is not given any special meaning or
`definition. See Opening Declaration, ¶¶ 94-96.
`4.
`I have reviewed Liquidia’s Answering Claim Construction Brief, in which
`Liquidia contends that (i) statements by UTC in Inter Partes Review (“IPR”) proceedings
`before the Patent Trial and Appeal Board (“PTAB”) require a departure from plain and ordinary
`meaning according to its proposed construction: “contacting the solution comprising treprostinil
`from step (b) with a base to form a salt of treprostinil, wherein the salt is formed without
`isolation of treprostinil after alkylation and hydrolysis”; and (ii) that it “provided UTC with
`
`1
`
`IPR2020-00770
`United Therapeutics EX2033
`Page 1 of 13
`
`

`

`documentary evidence that treprostinil in Liquidia’s LIQ861 is isolated prior to salt formation
`and cannot infringe.” See Liquidia’s Answering Claim Construction Brief (“Resp. Br.”) at 21-
`24 (citing to Exs.1 7-9, 28) (emphasis omitted).
`5.
`As explained below, I do not agree that a POSA would understand the statements
`on which Liquidia relies to require a departure from plain and ordinary meaning according to
`Liquidia’s proposed construction. Furthermore, Liquidia’s proposed construction would
`exclude a preferred embodiment, and would thus be inconsistent with a POSA’s understanding
`of the ’901 patent specification, prosecution history and IPR proceedings.
`6.
`This declaration is intended to supplement (not replace) my Opening
`Declaration; my relevant background and qualifications may be found in my Opening
`Declaration (¶¶ 3-29) and accompanying exhibits thereto.
`I.
`SUMMARY OF OPINIONS
`7.
`It is my opinion that the following terms from the ’066 and ’901 patents should
`have the following constructions:
`Claim Term
`Patent(s) and
`Claims
`’066 patent, claims
`1 and 8
`’066 patent, claims
`6 and 8; ’901
`patent, claim 6
`’066 patent, claims
`6 and 8; ’901
`patent, claim 6
`
`“a process”
`
`“ambient
`temperature”
`
`“stored” / “storing”
`/ “storage”
`
`“pharmaceutical
`batch”
`
`’901 patent, claims
`1-4, 6, and 8
`
`Construction
`
`plain and ordinary meaning
`
`plain and ordinary meaning
`
`“require that the stored material possesses
`stability sufficient to allow manufacture and
`which maintains integrity for a sufficient
`period of time to be useful for the preparation
`of a pharmaceutical composition or a
`pharmaceutical product”
`“a specific quantity of treprostinil (or its salt)
`that is intended to have uniform character and
`quality, within specified limits, and is
`produced according to a single manufacturing
`order during the same cycle of manufacture,
`
`1 “Ex.” refers to the exhibits attached to the Declaration of Douglas W. Cheek in Support of
`Defendant’s Answering Claim Construction brief unless otherwise noted.
`
`2
`
`IPR2020-00770
`United Therapeutics EX2033
`Page 2 of 13
`
`

`

`Claim Term
`
`Patent(s) and
`Claims
`
`’901 patent, claims
`1 and 8
`
`“contacting the
`solution comprising
`treprostinil from
`step (b) with a base
`to form a salt of
`treprostinil”
`
`Construction
`
`wherein the uniform character and quality is
`such that it still contains impurities resulting
`from the method by which it is produced”
`plain and ordinary meaning
`
`II.
`
`LEGAL STANDARDS
`8.
`As set forth in my Opening Declaration, I understand that the meaning of claim
`terms is a legal issue for the court to decide. I also understand that the court determines the
`meaning of the claims based on the perspective of a person of ordinary skill in the art (“POSA”)
`as of the filing date of the patent application. I also understand that the court relies on the
`claims, patent specification and prosecution history of the patent to determine the meaning of
`the claim terms. I also understand that the court can consider expert testimony regarding how a
`POSA would understand the claim terms in view of the claims, patent specification and
`prosecution history. In this declaration, I offer my opinions about how a POSA would have
`understood the meaning of the claim terms in view of the claims, patent specification and
`prosecution history, as well as the IPR proceedings cited by Liquidia.
`9.
`I understand that Liquidia has argued that “UTC’s statements amount to a clear
`and unmistakable acknowledgement that the claims of the ’901 patent do not permit isolation of
`treprostinil prior to salt formation.” Resp. Br. at 21. I understand from counsel that claim scope
`is not disavowed unless the patentee clearly and unambiguously surrenders it. I understand that
`the standard for finding disavowal has been described as “exacting” and that courts consider a
`POSA’s understanding after reading the claims, specification and prosecution history.
`
`3
`
`IPR2020-00770
`United Therapeutics EX2033
`Page 3 of 13
`
`

`

`III.
`
`POSA’S UNDERSTANDING OF THE CLAIM TERM
`A.
`“contacting the solution comprising treprostinil from step (b) with a base to
`form a salt of treprostinil”
`The claim term “contacting the solution comprising treprostinil from step (b)
`10.
`with a base to form a salt of treprostinil” appears in claims 1 and 8 of the ’901 patent. The
`claims state:
`
`Claim 1: “A pharmaceutical batch consisting of treprostinil or a salt thereof and
`impurities resulting from (a) alkylating a benzindene triol, (b) hydrolyzing the product of
`step (a) to form a solution comprising treprostinil, (c) contacting the solution comprising
`treprostinil from step (b) with a base to form a salt of treprostinil, (d) isolating the salt
`of treprostinil, and (e) optionally reacting the salt of treprostinil with an acid to form
`treprostinil, and wherein the pharmaceutical batch contains at least 2.9 g of treprostinil or
`its salt.” Ex. 2 at Claim 1 (emphasis added).
`
`Claim 8: “A method of preparing a pharmaceutical batch as claimed in claim 1,
`comprising (a) alkylating a benzindene triol, (b) hydrolyzing the product of step (a) to
`form a solution comprising treprostinil, (c) contacting the solution comprising
`treprostinil from step (b) with a base to form a salt of treprostinil, (d) isolating the salt
`of treprostinil, and (e) optionally reacting the salt of treprostinil with an acid to form
`treprostinil.” Id. at Claim 8 (emphasis added).
`11.
`As discussed in my Opening Declaration (¶¶ 95-96), a POSA would understand
`this term according to its plain and ordinary meaning.
`12.
`In its Responsive Brief, Liquidia cited to statements made in the context of an
`IPR concerning the ’901 patent. These underlying IPR documents were attached to Liquidia’s
`Brief as exhibit 7 (the Patent Owner’s Preliminary Response), exhibit 8 (Rehearing Request),
`exhibit 9 (Patent Owner’s Response), and exhibit 28 (Rehearing Request Denial). Liquidia
`relies on a handful of statements from these documents, which were not expert opinions, to
`contend that a POSA would understand the ’901 patent claims to require an additional limitation
`
`4
`
`IPR2020-00770
`United Therapeutics EX2033
`Page 4 of 13
`
`

`

`that is not recited in the claims. In my opinion, Liquidia’s assertions are not consistent with how
`a POSA would understand the cited statements in their complete context, and in view of the
`patent claims, specification, prosecution history, and the training and knowledge of a POSA.
`13.
`As an initial matter, Liquidia submitted no opinion from any expert in support of
`its Answering Brief in order to support its proposed construction.
`14.
`Liquidia further ignores that a POSA reviewing documents filed in connection
`with the ’901 proceedings would consider the declarations of Prof. Pinal, UTC’s expert. Prof.
`Pinal’s assessment is unambiguous. He states:
`It is noteworthy that in Dr. Winkler’s [Liquidia’s expert in the IPR
`proceedings] analysis, opposite actions, such as isolating vs. not isolating
`treprostinil, operate in the same direction. I note that this isolation
`limitation Dr. Winkler seems to try to be addressing is actually a limitation
`from the ’066 patent—not isolating the treprostinil before contacting it
`with a base is not an explicit limitation of claim 1 of the ’901 patent.
`
`Ex 10, ¶157 (emphasis added). Liquidia’s interpretations of attorney statements in the ’901 IPR
`proceedings are not consistent with Prof. Pinal’s unambiguous opinion concerning the ’901
`patent’s claim scope, and they are not consistent with how a POSA would interpret those
`statements in their full context and in light of Prof. Pinal’s opinion and the ’901 patent
`specification and claims.
`15.
`I disagree with a number of statements and assertions in Liquidia’s Responsive
`Brief. For example, Liquidia cites to UTC’s Patent Owner’s Response as purportedly
`supporting its proposed construction that would require “contacting the solution comprising
`treprostinil from step (b) with a base to form a salt of treprostinil, wherein the salt is formed
`without isolation of treprostinil after alkylation and hydrolysis.” The language on which
`Liquidia relies states:
`Claim 1 requires “contacting the solution comprising treprostinil from step
`(b) with a base to form a salt of treprostinil.” The claim’s preamble
`requires the pharmaceutical batch be one “consisting of” what results from
`the recited steps. Together, this language means treprostinil is not isolated
`from the solution formed in step (b) before forming a salt in step (c). Just
`as in the ’901 patent’s Example 3 (reactor charged with treprostinil
`
`5
`
`IPR2020-00770
`United Therapeutics EX2033
`Page 5 of 13
`
`

`

`solution “35-40 L from the previous step”), claim 1 requires the solution
`in which treprostinil is formed be used directly in the next salt-forming
`step without isolating treprostinil in between.
`
`Ex. 9 at 11. A POSA would not understand these statements to constitute an unambiguous
`
`disclaimer that would support the requirements of Liquidia’s proposed construction. As
`
`explained in greater detail below, a POSA would recognize that this statement does not
`
`accurately summarize the experimental detail it cites to, including the ’901 patent’s Example 3
`
`(“Conversion of Treprostinil to Treprostinil Diethanolamine Salt (1:1)”2), and the “‘35-40 L
`
`from the previous step’” described in Example 2 (“Hydrolysis of Benzindene Nitrile”)3. The
`
`reasons a POSA would not understand this statement to be an unambiguous disclaimer or
`
`disavowal of claim scope include:
`
` As an initial matter, nowhere in the ’901 patent specification or claims does it state that
`
`isolation of treprostinil before the salt formation step must not, or cannot, be performed.
`
`To the contrary, the patent indicates that “the treprostinil salts can be synthesized from
`
`the solution of treprostinil without isolation” (Ex. 2 at 17:8-10 (emphasis added)), which
`
`clearly and unambiguously indicates that isolation is not required prior to the salt
`
`formation step, and may be performed if desired.
`
` The statement Liquidia relies on does not cite to any expert declaration.
`
` A POSA would recognize that the statement is incorrect based on what the patent
`
`specification describes in Example 2. The statement indicates that the solution in which
`
`treprostinil is formed is used in the salt formation step. As indicated in Example 2,
`
`2 Ex. 2 at 12:19-13:49.
`3 Ex. 2 at 11:1-12:17.
`
`6
`
`IPR2020-00770
`United Therapeutics EX2033
`Page 6 of 13
`
`

`

`however, treprostinil is formed in a basic (alkaline) aqueous solution containing an
`
`alcohol (methanol), and this solution is not carried forward to the salt formation step.
`
` The actual solution that was carried forward to the salt formation step in Example 2 is an
`
`organic phase solution (and not an aqueous phase solution) containing treprostinil, and
`
`this occurs after the treprostinil that was formed in the aqueous phase described above is
`
`transferred to an organic phase. As is clear from Example 2, the solution in which
`
`treprostinil is formed, which is the basic (alkaline) aqueous phase, is first acidified to
`
`protonate treprostinil, and this unionized form of treprostinil is then extracted into ethyl
`
`acetate (an organic solvent), and it is this treprostinil in the organic phase that is what is
`
`carried forward to the salt formation step, and not the solution in which treprostinil was
`
`formed, which was in the aqueous phase. It is this organic phase containing treprostinil,
`
`that follows the phase transition from the aqueous phase, that represents the “35-40 L
`
`from the previous step” that was used “in [the] next step”, which is the salt formation step
`
`described in Example 3. Accordingly, a POSA would recognize that the unsupported
`
`statement on which Liquidia relies could not unambiguously alter the scope of the ’901
`
`patent claims as Liquidia proposes.
`
` Furthermore, treprostinil itself has already been isolated and separated from many
`
`impurities (although not all impurities) through the many purification steps that occur in
`
`Examples 2 and 3. Simply because treprostinil is still in a solution when used in the salt
`
`formation step does not mean that treprostinil has not been isolated (as discussed in detail
`
`below). As such, a POSA would understand that the passage in the Patent Owner’s
`
`Response upon which Liquidia relies is incorrect to the degree it suggests that Examples
`
`2 and 3 describe synthesizing treprostinil without isolating it prior to salt formation.
`
`7
`
`IPR2020-00770
`United Therapeutics EX2033
`Page 7 of 13
`
`

`

` As discussed below, it is common for compounds to be isolated (i.e., separated from
`
`many impurities and other contaminants resulting from previous chemical reactions) and
`
`retained in a solution state in commercial manufacturing processes for practical reasons,
`
`such as facilitating material transfer.
`
` Nowhere in the patent specification or claims is it required that treprostinil be isolated,
`
`either in a solid form or in solution, prior to being utilized in the salt formation step.
`
`For all of the reasons delineated above, Liquidia’s limitation in the claim term
`16.
`that isolation of treprostinil is not permitted prior to the salt formation step is unsupported. A
`POSA reviewing the ’901 patent specification and claims, as well as in view of Dr. Pinal’s
`expert declaration, would understand that the ’901 patent does not require that the solution in
`which treprostinil is formed be used directly in the next salt-forming step without isolating
`treprostinil in between.
`17.
`And to support this position further, a POSA would also understand that the cited
`experimental procedures in the ’901 patent specification disclose, among other things:
`
` Synthesizing treprostinil through hydrolysis in an aqueous phase containing potassium
`hydroxide (KOH). Ex. 2 at 11:39-45.
`
` The formation of treprostinil is monitored via thin-layer chromatography (“TLC”), and
`following the reaction’s completion, the reaction mixture is “quenched” by the addition
`of hydrochloric acid. Id. at 11:45-53.
`
` A number of purification steps are performed on the quenched reaction mixture prior to
`the salt formation step in Example 2, which include i) in vacuo removal and discarding
`of solvent, ii) extraction with the organic solvent, ethyl acetate, “to remove impurities
`soluble in ethyl acetate”, iii) washing of combined organic layers with water, iv)
`washing with aqueous sodium bicarbonate (NaHCO3), v) washing with saturated salt
`(NaCl) solution, vi) filtering, vii) drying over anhydrous sodium sulfate (Na2SO4), viii)
`treatment with activated carbon, ix) further filtration through a Celite® 545 pad, x)
`
`8
`
`IPR2020-00770
`United Therapeutics EX2033
`Page 8 of 13
`
`

`

`washing with ethyl acetate “until no compound was seen on TLC of the washings”, and
`xi) concentration by evaporation. Id. at 11:53-12:13.
`
`
`
`
`
`In addition to Examples 2 and 3 of the specification referred to by Liquidia, a number of
`additional purification steps were also performed after formation of the benzindene
`nitrile in Example 1 (“Alkylation of Benzindene Triol”4) prior to those purification steps
`described above in Example 2, including i) filtering the reaction mixture with/without
`Celite pad, ii) washing with acetone and iii) concentrating in vacuo.
`
`It is clear that these many purification steps result in isolation of treprostinil (from many
`impurities and contaminants) prior to the salt formation step, regardless of whether the
`isolated treprostinil is in an aqueous phase, organic phase or solid phase.
`
` Nowhere in these Examples in the specification of the ’901 patent or its claims is the
`elimination of any of the purification steps in the synthesis of treprostinil following the
`alkylation step required, and indeed, many purification steps are utilized in this process.
`
` Likewise, there is no requirement for eliminating the isolation of treprostinil following
`the alkylation step that appears anywhere in the Examples in the specification of the
`’901 patent or its claims.
`18.
`A POSA would also understand that these multiple purification steps described
`in the patent specification result in the significant removal of impurities and contaminants after
`formation of the nitrile and the preparation of treprostinil, and prior to the salt recrystallization
`steps. The ’901 specification states: “[t]he impurities carried over from intermediate steps (i.e.,
`alkylation of triol and hydrolysis of benzindene nitrile) are removed during the carbon treatment
`and the salt formation step.” Id. at 17:1-4. Notably, a POSA would understand that these
`experimental details do not support Liquidia’s assertion that treprostinil cannot be isolated prior
`to the salt formation step, and the Patent Owner’s Response’s direct reference to the
`experimental details of the ’901 specification would provide context for the Patent Owner’s
`
`4 Ex. 2 at 10:12-67.
`
`9
`
`IPR2020-00770
`United Therapeutics EX2033
`Page 9 of 13
`
`

`

`statement that does not support Liquidia’s contention that those statements unambiguously
`disavow claim scope according to Liquidia’s proposed construction. Indeed, the experimental
`details of the patent specification underscore that isolation, or lack thereof, are permissive in
`view of the statement that “the treprostinil salts can be synthesized from the solution of
`treprostinil without isolation.” Id. at 17:8-10. Accordingly, there is no requirement in the patent
`specification or claims that the synthesis of treprostinil salts from treprostinil must be performed
`without isolation of treprostinil. Although it can be desirable in a commercial-scale
`manufacturing process to eliminate steps for efficiency, the ’901 patent specification and claims
`do not require that any of the purification or isolation steps be eliminated.
`19.
`The multiple purification steps described above in Examples 1 and 2 also result
`in a change in the solvent system that contains treprostinil, from an aqueous solvent system in
`which treprostinil was formed, to an organic solvent system containing ethyl acetate after
`elimination of water, prior to the salt formation step described in Example 3. In particular,
`according to the experimental detail in Example 2 of the ’901 patent specification, treprostinil is
`synthesized in a methanolic potassium hydroxide aqueous solution (i.e., a basic aqueous
`solution containing an alcohol), and that the solvent is subsequently changed to an organic
`solvent (ethyl acetate) through a series of acidification, extraction and drying (i.e., elimination
`of water) steps, among others. Claim 1 of the ’901 patent, which refers to “(a) alkylating a
`benzindene triol” and “(b) hydrolyzing the product of step (a) to form a solution comprising
`treprostinil” does not specify a specific solvent system. Claim 1 of the ’901 patent does not
`refer to or restrict what can be done with treprostinil with respect to isolating, drying, subjecting
`to carbon treatment, or otherwise purifying, nor does the patent claim restrict or limit how the
`resulting treprostinil is subsequently processed. And importantly, the claim does not prohibit
`treprostinil from being isolated at this point if desired or necessary. A POSA would further
`understand that claim 1’s reference to “contacting the solution comprising treprostinil from step
`(b)” does not restrict that step only to the solvent in which treprostinil is formed, dried,
`subjected to carbon treatment, or otherwise purified. A POSA would realize that this step in
`
`10
`
`IPR2020-00770
`United Therapeutics EX2033
`Page 10 of 13
`
`

`

`claim 1 could be used to prepare salts of treprostinil without isolation if desired, which would
`significantly increase the efficiency of large commercial-scale production of treprostinil salts,
`but that including an isolation step could be performed if desired, and the addition of such a step
`is not prohibited by the patent specification or claims.
`20.
`Liquidia contends that the experimental details of the ’901 patent specification
`somehow support its proposed construction. A POSA would recognize, however, that
`Liquidia’s assertion does not make sense in light of the fact that its proposed construction
`excludes a preferred embodiment in Example 2 of the ’901 patent.
`21.
`At times, Liquidia appears to suggest that the key attribute of “isolation” that its
`proposed construction would exclude is the removal of solvent from treprostinil to form a solid.
`See Resp. Br. at 24. A POSA would not agree with Liquidia’s attorney argument, and the
`presence or absence of solvent is not relevant to how a POSA would understand the term
`“isolation” in the context of commercial pharmaceutical manufacturing, the specification, or the
`claims. Rather, a POSA would understand that a compound can be purified to isolation, even
`while it is maintained in a solvent, which is often done for practical considerations (e.g.,
`material transfer) in manufacturing plants. Likewise, a compound that has not been isolated to a
`high level of purity can have all of its solvent removed such that the compound exists as a dry
`powder in the solid state, but yet still have present many contaminating materials such that it is
`not isolated at all. The physical state of a compound, be it in a solvent solution or in a solid
`state, is not a reflection of the degree to which that compound has been isolated from impurities
`and contaminants carried through in the various reaction mixtures. Accordingly, Liquidia’s
`proposed construction of the step “(c) contacting the solution comprising treprostinil from step
`(b) with a base to form a salt of treprostinil” inserts the term, “isolation”, that is not recited in
`that claim limitation, and a POSA would not understand this additional limitation to be required
`by the claims. The insertion by Liquidia into the claim construction for a requirement that the
`isolation of treprostinil prior to salt formation be eliminated is inconsistent with how a POSA
`would understand that term based on the ’901 patent specification and claims.
`
`11
`
`IPR2020-00770
`United Therapeutics EX2033
`Page 11 of 13
`
`

`

`A POSA also would not understand the cherry-picked statements that Liquidia
`22.
`alleges indicate that the ’901 process “requires” that the isolation step be eliminated to
`accurately describe the ’901 invention, especially when the claim language is understood in
`light of the specification, claims and prosecution history. Furthermore, a POSA would
`understand the claim language in the context of commercial pharmaceutical manufacturing
`where it is common to re-purify specific batches of an API as needed. See, e.g., Opening
`Declaration, ¶56. A POSA would understand that the claims permit, and most certainly do not
`prevent, re-purification of the solution comprising treprostinil as needed as part of its
`manufacturing process.
`23.
`In sum, a POSA’s understanding of the meaning of the claim term “(c)
`contacting the solution comprising treprostinil from step (b) with a base to form a salt of
`treprostinil” would be governed by his or her reading of the patent itself combined with their
`knowledge of the art, and especially as the art relates to commercial pharmaceutical
`manufacturing to which the ’901 patent is directed. A POSA would not understand the cited
`statements from the IPR to change the clear scope of the claims, and would understand this
`claim term to have its plain and ordinary meaning.
`
`12
`
`IPR2020-00770
`United Therapeutics EX2033
`Page 12 of 13
`
`

`

`
`
`United Therapeuticsa
`
`I declare under penalty ofperjury under the lawof the state ofCalifornia thatthe
`
`foregoing is truc and correct.
`
`Dated: April 2, 2021
`
`Robert R. Ruffolo,
`
`Ph.D.
`
`IPR2020-00770
`
`IPR2020-00770
`United Therapeutics EX2033
`Page 13 of 13
`
`