Supernus Pharm. v. Torrent Pharm.

Full title: SUPERNUS PHARMACEUTICALS, INC., Plaintiff, v. TORRENT PHARMACEUTICALS…

Court: United States District Court, D. New Jersey

Date published: Jan 30, 2024

Civil Action 21-06964 (GC) (DEA) (D.N.J. Jan. 30, 2024)

Opinion

Civil Action 21-06964 (GC) (DEA)

01-30-2024

SUPERNUS PHARMACEUTICALS, INC., Plaintiff, v. TORRENT PHARMACEUTICALS LTD., et al., Defendants.

Charles M. Lizza William C. Baton Sarah A. Sullivan SAUL EWING LLP Edgar H. Haug Nicholas F. Giove Richard F. Kurz Andrew S. Wasson Jason A. Kanter Anna N. Lukacher Jessica M. Stookey Chinmay P. Bagwe HAUG PARTNERS LLP Counsel for Plaintiff Supernus Pharmaceuticals, Inc Rebekah R. Conroy STONE CONROY LLC Neal Seth (pro hac vice) Wesley Weeks (pro hac vice) Lawrence Sung (pro hac vice) Corey Weinstein (pro hac vice) WILEY REIN, LLP Counsel for Defendants Torrent Pharmaceuticals Ltd. and Torrent Pharma Inc.

CASTNER, District Judge

Charles M. Lizza William C. Baton Sarah A. Sullivan SAUL EWING LLP Edgar H. Haug Nicholas F. Giove Richard F. Kurz Andrew S. Wasson Jason A. Kanter Anna N. Lukacher Jessica M. Stookey Chinmay P. Bagwe HAUG PARTNERS LLP Counsel for Plaintiff Supernus Pharmaceuticals, Inc

Rebekah R. Conroy STONE CONROY LLC Neal Seth (pro hac vice) Wesley Weeks (pro hac vice) Lawrence Sung (pro hac vice) Corey Weinstein (pro hac vice) WILEY REIN, LLP Counsel for Defendants Torrent Pharmaceuticals Ltd. and Torrent Pharma Inc.

OPINION FILED UNDER TEMPORARY SEAL

CASTNER, District Judge

This matter comes before the Court upon a Hatch-Waxman patent-infringement action brought by Plaintiff Supernus Pharmaceuticals, Inc. (Supernus) against Defendants Torrent Pharmaceuticals Ltd. and Torrent Pharma Inc. (Torrent), stemming from Torrent's Abbreviated New Drug Application seeking approval to market a generic version of Supernus's drug product. Supernus alleges that Torrent's ANDA Products infringe claims of three patents in suit: claim 14 of U.S. Patent No. 8,992,989 ('989 patent); claim 14 of U.S. Patent No. 9,549,940 ('940 patent); and claims 13 and 23 of U.S. Patent No. 9,622,983 ('983 patent).¹

1 Collectively, the “Asserted Claims” of the “Patents-in-Suit.” The '989 patent's trial exhibit is PTX058; the '940 patent's, PTX059; the '983 patent's, PTX061.

The Court held a bench trial on Supernus's infringement claims and Torrent's invalidity defenses. In this Opinion, the Court resolves disputes over infringement and validity of the Patents-in-Suit, including reserved decisions on motions in limine and trial objections and motions. The Court discusses the procedural history and trial record, and then makes findings of fact and conclusions of law in accordance with Federal Rule of Civil Procedure (“Rule”) 52(a). The findings of fact are based on the Court's observations at trial, credibility judgments, and review of all admissible record evidence.

For the reasons set forth below, and other good cause shown, the Court finds that Torrent's ANDA Products INFRINGE on the Patents-in-Suit and that the asserted patents are VALID.²

2 When Supernus rested its case-in-chief, Torrent moved for judgment as a matter of law on infringement. (Tr. 420:10-19.) Rule 52(c) states:

If a party has been fully heard on an issue during a nonjury trial and the court finds against the party on that issue, the court may enter judgment against the party on a claim or defense that, under the controlling law, can be maintained or defeated only with a favorable finding on that issue. The court may, however, decline to render any judgment until the close of the evidence. A judgment on partial findings must be supported by findings of fact and conclusions of law as required by Rule 52(a).

The Court reserved on Torrent's motion. For the reasons set forth in this Opinion, Torrent's motion is

I. FACTS ESTABLISHED

The following facts are stipulated, not disputed, or established by undisputed record evidence in the Final Pretrial Order, the Joint Proposed Findings of Fact, or both³:

3 The Final Pretrial Order is at ECF No. 133, and the Joint Proposed Findings of Fact (PFF) is at ECF No. 169. For several proposed facts, Torrent disputes their relevance but not their veracity. (See PFF ¶¶ 22-24, 26-27 (Def.) (asserting that facts are “not relevant to the litigation and therefore . . . not relevant to the proposed findings of fact”).) The Court recites facts that it finds are both relevant under Federal Rule of Evidence 401 and established through undisputed, credible, and admissible evidence.

A. The Parties

1. Plaintiff Supernus is a corporation organized and existing under the laws of Delaware, having its principal place of business at 9715 Key West Avenue, Rockville, Maryland 20850. (PFF ¶ 1 (citing ECF No. 133 ¶ 29).)

2. Defendant Torrent Pharmaceuticals Ltd. is a corporation operating and existing under the laws of India, with its principal place of business at Torrent House, Off. Ashram Road, Ahmedabad, Gujarat, 380009, India. (PFF ¶ 2 (citing ECF No. 133 ¶ 30).)

3. Defendant Torrent Pharma Inc. is a Delaware limited liability company having its principal place of business at 150 Allen Road, Suite 102, Basking Ridge, N.J. 07920. (PFF ¶ 3 (citing ECF No. 133 ¶ 31).)

4. Defendant Torrent Pharmaceuticals Inc. is a subsidiary of Torrent Pharmaceuticals Ltd. (PFF ¶ 4 (citing ECF No. 133 ¶ 32).)

B. Background

5. This is a civil action for patent infringement that is brought by Supernus against Torrent arising under the patent laws of the United States, Title 35, United States Code.⁴(PFF ¶ 5 (citing ECF No. 133 ¶ 21).)

4 This Court thus has subject-matter jurisdiction over all claims and counterclaims in this action under 28 U.S.C. §§ 1331 and 1338. The parties do not dispute this Court's personal jurisdiction over them. (ECF No. 133 ¶ 2.)

6. Supernus asserts that Torrent infringes claims of three patents - United States Patent Nos. 8,992,989 ('989 patent); 9,549,940 ('940 patent); and 9,622,983 ('983 patent) (collectively, the “Patents-in-Suit”). (PFF ¶ 6 (citing ECF No. 133 ¶ 23).)

7. This case involves the drug product Trokendi XR^®, which is sold by Supernus and is indicated for treatment of epilepsy: initial monotherapy for the treatment of partial-onset or primary generalized tonic-clonic seizures in patients 6 years of age and older; adjunctive therapy for the treatment of partial-onset, primary generalized tonic-clonic seizures, or seizures associated with Lennox-Gastaut syndrome (LGS) in patients 6 years of age and older; and preventive treatment of migraine in patients 12 years of age and older. (PFF ¶ 7 (citing ECF No. 133 ¶ 22).)

8. Supernus owns the FDA approved New Drug Application (NDA) No. 201635 for Trokendi XR^®, topiramate extended-release capsules, 25 mg, 50 mg, 100 mg, and 200 mg. (PFF ¶ 8 (citing ECF No. 133 ¶¶ 28, 33).)

9. Pursuant to 21 U.S.C. §§ 355(b)(1) and 355(c)(2), FDA's publication titled “Approved Drug Products with Therapeutic Equivalence Evaluations” (commonly known as the “Orange Book”) lists ten (10) patents as covering Supernus's Trokendi XR^®. The Orange Book lists the Patents-in-Suit and also the following patents, which are no longer asserted against Torrent: United States Patent Nos. 8,298,576 ('576 patent), 8,298,580 ('580 patent), 8,663,683 ('683 patent), 8,877,248 ('248 patent), 8,889,191 ('191 patent), 9,555,004 ('004 patent), and 10,314,790 ('790 patent). (PFF ¶ 9 (citing ECF No. 133 ¶ 34).)

10. The inventors assigned ownership of all Orange Book Patents to Supernus. (PFF ¶ 10 (citing ECF No. 133 ¶ 36).)

11. Torrent filed Abbreviated New Drug Application No. 215918 (Torrent's ANDA) with the FDA seeking approval to engage in commercial marketing of a generic version of Trokendi XR^® (Torrent's ANDA Products). (PFF ¶ 11 (citing 133 ¶ 55).)

12. On or about June 15, 2021, Torrent sent a letter pursuant to 21 U.S.C. § 355(j)(2)(B) and 21 C.F.R. § 314.95 regarding the Torrent ANDA Products and the '576, '580, '683, '248, '191, '989, '940, '004, '983, and '790 patents (the “June 15 Notice Letter”) to Supernus. (PFF ¶ 12 (citing ECF No. 133 ¶ 56).)

13. Torrent's ANDA references Trokendi XR^® (topiramate extended-release capsules) as the reference listed drug, or “RLD.” (PFF ¶ 13 (citing ECF No. 133 ¶ 57).)

14. The Torrent ANDA Products are extended release formulations of topiramate. (PFF ¶ 14 (citing ECF No. 133 ¶ 59).)

15. The Torrent ANDA Products are topiramate extended-release capsules, 25 mg, 50 mg, 100 mg, and 200 mg. (PFF ¶ 15 (citing ECF No. 133 ¶ 58).)

16. Torrent has asserted affirmative defenses stating that the Patents-in-Suit are invalid and not infringed, directly or indirectly. (PFF ¶ 17 (citing ECF No. 133 ¶ 24).)

17. Torrent Pharmaceuticals Ltd. also brought counterclaims for a declaratory judgment of non-infringement with respect to the Patents-in-Suit. (PFF ¶ 18 (citing ECF No. 133 ¶ 25).)

18. Supernus asserted affirmative defenses stating that, among other things, the Patents-in-Suit are valid and enforceable, and infringed by Torrent's ANDA Products. (PFF ¶ 19 (citing ECF No. 133 ¶ 26).)

C. The Accused Product

19. The composition of Torrent's ANDA Products is disclosed in the Quality Overall Summary (QOS) that Torrent submitted to the FDA. (PFF ¶ 20 (citing PTX106.3-5; Patel Tr. 65:12-66:2, 75:11-17, 75:20-22; Tr. 240:22-242:24, 232:2-237:24.).)⁵

5 Yogesh Patel, Assistant General Manager at Torrent, oversaw the formulation and development of Torrent's ANDA Products. (PFF ¶ 133 (citing Patel Tr. 18:4-19, 19:4-20:4).) In lieu of live testimony, excerpts of Mr. Patel's video deposition were played at trial. (See ECF No. 166.)

20. Torrent's ANDA Products comprise gelatin capsules that are filled with beads where each bead comprises (i) sugar sphere coated with a seal coating, (ii) a topiramate drug coating layer coated on top of each sugar sphere, and (iii) a release controlling polymer coating layer coated on top of the topiramate-coated beads. (PFF ¶ 21 (citing Patel Tr. 21:18-24, 22:5-15; Tr. 240:22-242:24, 232:2-237:24; PTX106.3-5).)

21. At the core of the Torrent ANDA Products beads is an inert sugar sphere. (PFF ¶ 22 (Pl.) (citing PTX106.3; Patel Tr. 66:3-12; Tr. 240:22-241:8).)

22. A seal coating suspension is coated on top of the sugar sphere. (PFF ¶ 23 (Pl.) (citing PTX106.3; Patel Tr. 66:13-16; Tr. 240:22-241:18).)

23. A drug coating suspension is coated on top of the seal coating suspension layer. The drug coating suspension layer contains the active pharmaceutical ingredient (API), topiramate, along with various excipients. The beads coated with the drug coating suspension are also referred to as “drug coated pellets.” (PFF ¶ 24 (Pl.) (citing PTX106.3; Patel Tr. 67:2-5, 67:19-22; Tr. 240:17-242:2).)

24. Torrent's ANDA Products comprise polyvinyl acetate dispersion as a release controlling polymer. (PFF ¶ 25; PTX106.3-4.)

25. The common polymer coated beads are filled into capsules, and each capsule may contain hundreds of beads. (PFF ¶ 26 (Pl.) (citing PTX106.4-5; Patel Tr. 21:18-24; Tr. 240:17-241:5; Tr. 61:19-62:2).)

26. Torrent developed a formulation “to mimic the physiochemical characteristics of RLD [i.e., Supernus's Trokendi XR^® product]. No other formulations were explored.” (PFF ¶ 27 (Pl.) (citing Patel. Tr. 302:8-17; PTX106.29).)

27. Torrent represented to the FDA that its ANDA Product is bioequivalent to Supernus's Trokendi XR^®. (PFF ¶ 28 (citing Patel Tr. 55:3-7, 251:1-11).)

D. The Patents-in-Suit and the Asserted Claims

28. Supernus is asserting claim 14 of the '989 patent, claim 14 of the '940 patent, and claims 13 and 23 of the '983 patent (the “Asserted Claims”) against Torrent in this action. (PFF ¶ 29 (citing ECF No. 133 ¶ 49).)

29. The '989 patent is titled “Sustained-Release Formulations of Topiramate.” (PFF ¶ 30 (citing ECF No. 133 ¶ 37).)

30. The '989 patent issued on March 31, 2015. (PFF ¶ 31 (citing ECF No. 133 ¶ 38).)

31. The '989 patent claims priority to U.S. Provisional Patent Application No. 60/859,502, which has a priority date of November 17, 2006. (PFF ¶ 32 (citing ECF No. 133 ¶ 39).)

32. The '989 patent issued from U.S. Patent Application No. 14/499,462. (PFF ¶ 33 (citing ECF No. 133 ¶ 40).)

33. Claim 14 of the '989 patent reads:

14. A sustained release formulation of topiramate comprising topiramate as an active ingredient, which is released immediately and continuously upon administration from the formulation, the formulation comprising:

(a) an extended release (XR) topiramate-containing component, comprising a coating material selected from the group consisting of cellulosic polymers and acrylic polymers, and, optionally,

(b) an immediate release (IR) topiramate-containing component comprising:

(i) a complexing agent selected from the group consisting of hydroxypropyl-beta-cyclodextrin, beta-cyclodextrin, gamma-cyclodextrin, alpha-cyclodextrin, cyclodextrin, and cyclodextrin derivative, and/or

(ii) an enhancing agent selected from the group consisting of Vitamin E TPGS, glutamic acid, glycine, sorbitol, mannose, amylose, maltose, mannitol, lactose, sucrose, glucose, xylitose, dextrins, glycerol-polyethylene glycol oxystearate, polyethylene glycol-32 glyceryl palmitostearate, sodium lauryl sulfate, polyoxyethylene sorbitan monooleate, benzyl alcohol, sorbitan monolaurate, polyethylene-polypropylene glycol, polyethylene glycol-3350, polyvinylpyrrolidone-K25, oleic acid, glyceryl monooleate, sodium benzoate, cetyl alcohol, sucrose stearate, crospovidone, sodium starch glycolate, croscarmellose sodium, carboxymethylcellulose, starch, pregelatinized starch, hydroxypropylmethylcellulose (HPMC), substituted hydroxypropylcellulose, microcrystalline cellulose sodium bicarbonate, calcium citrate, sodium docusate, menthol, and combinations thereof,

wherein the XR component exhibits a maximum plasma concentration of topiramate in vivo at 16 or more hours after a single initial dose.

[(PFF ¶ 34 (citing ECF No. 133 ¶ 50).)]

34. The '940 patent is titled “Sustained-Release Formulations of Topiramate.” (PFF ¶ 35 (citing ECF No. 133 ¶ 41).)

35. The '940 patent issued on January 24, 2017. (PFF ¶ 36 (citing ECF No. 133 ¶ 42).)

36. The '940 patent claims priority to U.S. Provisional Patent Application No. 60/859,502, which has a priority date of November 17, 2006. (PFF ¶ 37 (citing ECF No. 133 ¶ 43).)

37. The '940 patent issued from U.S. Patent Application No. 15/259,841. (PFF ¶ 38 (citing ECF No. 133 ¶ 44).)

38. Claim 14 of the '940 patent reads:

14. A sustained release formulation of topiramate comprising between 0.5% to 85%, by weight of the formulation, topiramate, which is released immediately and continuously upon administration from the formulation, the formulation comprising:

(a) an extended release (XR) topiramate-containing component, comprising a coating material selected from the group consisting of cellulosic polymers and acrylic polymers, and, optionally,

(b) an immediate release (IR) topiramate-containing component comprising:

(i) a complexing agent selected from the group consisting of hydroxypropyl-beta-cyclodextrin, beta-cyclodextrin, gamma-cyclodextrin, alpha-cyclodextrin, cyclodextrin, and cyclodextrin derivative, and/or

(ii) an enhancing agent selected from the group consisting of Vitamin E TPGS, glutamic acid, glycine, sorbitol, mannose, amylose, maltose, mannitol, lactose, sucrose, glucose, xylitose, dextrins, glycerol-polyethylene glycol oxystearate, polyethylene glycol-32 glyceryl palmitostearate, sodium lauryl sulfate, polyoxyethylene sorbitan monooleate, benzyl alcohol, sorbitan monolaurate, polyethylene-polypropylene glycol, polyethylene glycol-3350, polyvinylpyrrolidone-K25, oleic acid, glyceryl monooleate, sodium benzoate, cetyl alcohol, sucrose stearate,

crospovidone, sodium starch glycolate, croscarmellose sodium, carboxymethylcellulose, starch, pregelatinized starch, hydroxypropylmethylcellulose (HPMC), substituted hydroxypropylcellulose, microcrystalline cellulose sodium bicarbonate, calcium citrate, sodium docusate, menthol, and combinations thereof,

wherein the XR component exhibits a maximum plasma concentration of topiramate in vivo at 16 or more hours after a single initial dose.

[(PFF ¶ 39 (citing ECF No. 133 ¶ 51).)]

39. The '983 patent is titled “Sustained-Release Formulations of Topiramate.” (PFF ¶ 40 (citing ECF No. 133 ¶ 45).)

40. The '983 patent issued on April 18, 2017. (PFF ¶ 41 (citing ECF No. 133 ¶ 46).)

41. The '983 patent claims priority to U.S. Provisional Patent Application No. 60/859,502, which has a priority date of November 17, 2006. (PFF ¶ 42 (citing ECF No. 133 ¶ 47).)

42. The '983 patent issued from U.S. Patent Application No. 15/259,856. (PFF ¶ 43 (citing ECF No. 133 ¶ 48).)

43. Claim 13 of the '983 patent reads:

13. A sustained release formulation of topiramate comprising topiramate in the form of micronized particles ranging from 1 μm to 1000 μm in size, which topiramate is released immediately and continuously upon administration from the formulation, the formulation comprising:

(a) an extended release (XR) topiramate-containing component, comprising a coating material selected from the group consisting of cellulosic polymers and acrylic polymers, and, optionally,

(b) an immediate release (IR) topiramate-containing component comprising:

(i) a complexing agent selected from the group consisting of hydroxypropyl-beta-cyclodextrin, beta-cyclodextrin, gamma-cyclodextrin, alpha-cyclodextrin, cyclodextrin, and cyclodextrin derivative, and/or

(ii) an enhancing agent selected from the group consisting of Vitamin E TPGS, glutamic acid, glycine, sorbitol, mannose, amylose, maltose, mannitol, lactose, sucrose, glucose, xylitose, dextrins, glycerol-polyethylene glycol oxystearate, polyethylene glycol-32 glyceryl palmitostearate, sodium lauryl sulfate, polyoxyethylene sorbitan monooleate, benzyl alcohol, sorbitan monolaurate, polyethylene-polypropylene glycol, polyethylene glycol-3350, polyvinylpyrrolidone-K25, oleic acid, glyceryl monooleate, sodium benzoate, cetyl alcohol, sucrose stearate, crospovidone, sodium starch glycolate, croscarmellose sodium, carboxymethylcellulose, starch, pregelatinized starch, hydroxypropylmethylcellulose (HPMC), substituted hydroxypropylcellulose, microcrystalline cellulose sodium bicarbonate, calcium citrate, sodium docusate, menthol, and combinations thereof,

wherein the XR component exhibits a maximum plasma concentration of topiramate in vivo at 16 or more hours after a single initial dose.

[(PFF ¶ 44 (citing ECF No. 133 ¶ 52).)]

44. Claim 23 of the '983 patent, which depends from claim 13, reads:

23. The formulation according to claim 13 reads: wherein the XR component further comprises a pore former selected from the group consisting of glucose, fructose, mannitol, mannose, galactose, sorbitol, pullulan, dextran, hydroxyalkylcelluloses, carboxyalkylcelluloses, hydroxypropylmethylcellulose, cellulose ethers, acrylic resins, polyvinylpyrrolidone, cross-linked polyvinylpyrrolidone, polyethylene oxide, carbomer, diols, polyols, polyhydric alcohols, polyalkylene glycols, polyethylene glycols, polypropylene glycols or block polymers thereof, polyglycols, poly(a-w)alkylenediols, alkali metal salts, alkaline earth metal salts, and combinations thereof.

[(PFF ¶ 45 (citing ECF No. 133 ¶ 53).)]

E. Claim Construction

45. For purposes of this litigation, the Parties have agreed upon, and the Court has adopted, the following definition of a person of ordinary skill in the art (POSA):

[S]omeone in the 2006 time frame with at least a Bachelor of Science degree in Pharmaceutical Sciences or a related field, approximately three to five years of experience in a drug delivery technology or related field, and working knowledge regarding pharmacokinetics (or a person of commensurate education and experience).

[(PFF ¶ 57 (citing ECF No. 88 at 9; ECF No. 133 ¶ 60).)]

46. As per the Court's December 2, 2022 Markman Order, the following constructions apply to the Asserted Claims:

Patent Claim Term	Court's Construction
‘‘immediately and continuously” ('989 patent claim 14; '940 patent claim 14; '983 patent claim 13)	Requires no construction-plain and ordinary meaning “Without delay and without interruption"
“extended release (XR) component” or “extended release component” “extended release (XR) topiramatecontaining component(s)” or "extended release topiramate-containing components)” ('989 patent claim 14; '940 patent claim 14; '983 patent claims 13 and 23)	“A component that releases topiramate in a continuous manner and is adjusted in such a way that 80% of the active ingredient is released in vitro in the predetermined period of time”
“comprising a coating material selected from the group consisting of cellulosic polymers and acrylic polymers” ('989 patent claim 14; '940 patent claim 14; '983 patent claim 13)	Requir es no construction-plain and ordinary meaning

[(PFF ¶ 58 (citing ECF No. 89 at 3-4; ECF No. 88 at 33; ECF No. 133 ¶ 61).)]

F. Witnesses Presented at Trial

47. Supernus's expert witness Mansoor A. Khan, R.Ph., Ph.D., was admitted as an expert in the field of pharmaceutical formulations and drug delivery - specifically, the development, evaluation, and characterization of sustained release formulations and the ingredients in such formulations. (PFF ¶ 59 (citing Tr. 221:20-24, 231:22-24; PTX119).)

48. Supernus's expert witness David R. Taft, Ph.D., was admitted as an expert in pharmacokinetics and pharmacodynamics. (PFF ¶ 71 (citing Tr. 162:8-17).)

49. Supernus's fact witness, Padmanabh Bhatt, Ph.D., is one of the inventors of the Patents-in-Suit. (PFF ¶ 124 (citing Tr. 56:8-17; PTX058.2; PTX059.2; PTX061.2).). Dr. Bhatt is a founding executive of Supernus. (PFF ¶ 126 (citing Tr. 52:22-24).) When Supernus was founded in December 2005, Dr. Bhatt was the head of pharmaceutical sciences and was involved in the development of Trokendi XR^®. (PFF ¶ 127 (citing Tr. 52:22-24, 55:4-19).) At present, Dr. Bhatt is the Chief Scientific Officer and a senior vice president of intellectual property at Supernus. (PFF ¶ 128 (citing Tr. 53:13-15).) As a senior vice president of intellectual property at Supernus, Dr. Bhatt is responsible for managing outside counsel for intellectual property related matters both in the U.S. and worldwide. (PFF ¶ 130 (citing Tr. 54:13-25).) Dr. Bhatt is not an attorney. (PFF ¶ 131 (citing Tr. 54:11-12).) Dr. Bhatt earned an undergraduate degree and a master's degree in pharmacy from the University of Bombay. He also earned a Master of Science degree and a Ph.D. in pharmaceutical chemistry from the University of Kansas. (PFF ¶ 132 (citing Tr. 52:6-16).)

50. Torrent's expert witness Linda Felton, Ph.D., was admitted as an expert in drug formulation and modified drug delivery systems. (PFF ¶ 82 (citing Tr. 553:16-17).)

51. Torrent's expert witness David P. Rotella, Ph.D., was admitted as an expert in structural chemistry and organic chemistry. (PFF ¶ 96 (citing Tr. 449:22-23).).

52. Torrent's fact witness, Yogesh Patel, is Assistant General Manager at Torrent, and oversaw the formulation and development of the Torrent ANDA Products. (PFF ¶ 133 (citing Patel Tr. 18:4-19, 19:4-20:4).) Mr. Patel has a master's degree in pharmaceutics. (PFF ¶ 134 (citing Patel Tr. 11:18-23).) Mr. Patel is Torrent's designated Rule 30(b)(6) witness. (PFF ¶ 135 (citing Patel Tr. 23:4-16, 24:13-19, 26:7-27:16, 28:16-29:15, 31:10-32:10; PTX067).)

II. BACKGROUND

A. Procedural History

Supernus filed separate infringement actions against Torrent and Ajanta Pharma Limited and Ajanta Pharma USA Inc. (ECF No. 1; ECF No. 43.)⁶ The two actions were later consolidated. (ECF No. 43.) Torrent counterclaimed for declarative judgments of noninfringement and asserted an affirmative defense that Supernus's patents were invalid.⁷ Torrent's basis, according to its noninfringement contentions,⁸ was that “Torrent's ANDA Products do not comprise any claimed controlling coating material including cellulosic and Acrylic polymer in composition” in the Asserted Claims of the Patents-in-Suit. (ECF No. 127-1 at 7-13.⁹)

6 Civil Action No. 21-14268 (Torrent); Civil Action No. 21-6964 (Ajanta).

7 (Defs.' Answer to Complaint, Affirmative Defenses, & Counterclaim, Supernus Pharmaceuticals, Inc. v. Torrent Pharmaceuticals Ltd., et al., Civil No. 21-14268 (D.N.J. Sept. 29, 2021), ECF No. 11.)

8 This District's local patent rules required Torrent to serve “Non-Infringement Contentions” describing its basis for opposing Supernus's infringement assertion. L. Pat. R. 3.6(e). The contentions needed to “include a chart identifying each claim at issue in the case and each limitation of each claim at issue” and “specifically identify[ing] for each claim which claim limitation(s) is/(are) literally absent from” Torrent's ANDA. Id. Torrent's noninfringement contentions also covered other patents that Supernus no longer asserts.

9 Page numbers for record cites (i.e., “ECF Nos.”) refer to the page numbers stamped by the Court's e-filing system and not the internal pagination of the parties.

At the Markman stage, the Court held a hearing and construed certain of the patents' claim terms - namely, the definition of a person of ordinary skill in the art, or “POSA”; “immediately and continuously”; “extended release (XR) component”; and “comprising a coating material selected from the group consisting of cellulosic polymers and acrylic polymers.” (ECF Nos. 88, 89.)¹⁰

10 A transcript of the Markman hearing is at ECF No. 87.

Supernus and the Ajanta defendants eventually dismissed their claims against each other. (ECF No. 104.) Shortly after, Supernus and Torrent stipulated to reducing their claims and defenses: Supernus agreed to pursue only the Asserted Claims; Torrent agreed to withdraw reports of two experts; and both parties agreed to withdraw their requests to strike certain portions of each other's expert reports. (ECF No. 106.)

In early July 2023, the Court entered the Final Pretrial Order. (ECF No. 133.) The Final Pretrial Order detailed, among other things, the legal issues, stipulated facts, contested facts, and potential fact and expert witnesses. (Id.)

Leading up to trial, the Court addressed three motions in limine - one from Supernus and two from Torrent. (ECF Nos. 126, 127, 128, 129.) The Court partially granted Supernus's motion, precluding Torrent “from eliciting from its expert testimony about the absence of evidence or Supernus's alleged failure to prove that Torrent's ANDA products contain an extended release (XR) topiramate-containing component.” (ECF No. 151 at 7.) The Court permitted Torrent, however, to cross-examine Supernus's expert “about the absence of such evidence or Supernus's satisfaction of its burden of proof,” reserving on deciding whether that cross-examination testimony would ultimately be admissible. (Id. at 8.)

As to Torrent's motions, the Court reserved on deciding (1) whether to preclude certain testimony of Supernus's expert Mansoor A. Khan, and (2) whether to preclude Supernus from asserting infringement under the doctrine of equivalents at trial. (Id.) The latter issue, the Court ruled, was “tantamount to a request for summary judgment of noninfringement” and should thus be resolved after trial. (Id. at 7.)

The Court then held a four-day bench trial on Supernus's infringement claims and Torrent's invalidity defenses. Following trial, the parties submitted Joint Proposed Findings of Fact along with their respective post-trial briefs. (ECF Nos. 169, 170, 171.) After reviewing the papers, the Court heard the parties' closing arguments.

B. Summary of the Arguments

Supernus argues that Torrent's ANDA Products infringe every limitation of the Asserted Claims, literally and under the doctrine of equivalents. (PFF ¶¶ 136-139; ECF No. 170.) Although, to prevail, Supernus must prove infringement of every claim limitation, the chief infringement dispute is whether Torrent's ANDA Products meet the claim limitation requiring “an extended release (XR) topiramate-containing component, comprising a coating material selected from the group consisting of cellulosic polymers and acrylic polymers.”

Torrent's literal-infringement defense is twofold. First, for its ANDA Products' coating material, Torrent uses polyvinyl acetate, which Torrent contends is not an acrylic polymer.¹¹Second, Torrent asserts that Supernus has not shown that Torrent's ANDA Products contain an extended-release component within the Court's construction of that term. As to the doctrine of equivalents, Torrent argues that the prosecution history estops Supernus from recapturing what it disclaimed in a narrowing amendment. But even if Supernus is not so estopped, Torrent still contests that its ANDA Products are equivalent. In addition, Torrent asserts that the Asserted Claims are invalid because they are not enabled and they lack written description.

11 Supernus does not assert that polyvinyl acetate is, or is equivalent to, a cellulosic polymer.

III. ISSUES TO BE DECIDED

The Court must resolve three main issues: first, whether Supernus has proven by a preponderance of the evidence that Torrent's ANDA Products infringe the Asserted Claims of the Patents-in-Suit, either literally or under the doctrine of equivalents; second, whether Torrent has proven by clear and convincing evidence that the Asserted Claims of the Patents-in-Suit are not enabled; and third, whether Torrent has proven by clear and convincing evidence that the Asserted Claims of the Patents-in-Suit lack adequate written-description support. (ECF No. 133 ¶¶ 15, 441-484.)

If there is no literal infringement, the Court must decide whether the doctrine of prosecution history estoppel bars Supernus from asserting that Torrent's use of polyvinyl acetate infringes the acrylic polymers claim term under the doctrine of equivalents. (Id. ¶ 442.) To get there, the Court must first decide whether Supernus's claim amendment during the prosecution of an earlier patent was narrowing such that a presumption of estoppel applies. And if one does apply, Supernus must have rebutted the presumption by a preponderance of the evidence.¹²

12 The Final Pretrial Order included two more issues: (1) whether Supernus is entitled to recover its costs and expenses in this action; and (2) whether Torrent has demonstrated that under the disclosure-dedication doctrine, the mention of polyvinyl alcohol in the specification would prevent a finding that polyvinyl acetate infringes the acrylic polymers claim term under the doctrine of equivalents. (ECF No. 133 ¶¶ 443, 451-452.) But Supernus's trial briefs did not mention costs-expenses recovery, and Torrent's trial briefs did not mention the disclosure-dedication doctrine. (ECF Nos. 139 & 170 (Supernus), 140 & 171 (Torrent).) As a result, the Court need not address either issue in this Opinion.

IV. DISCUSSION

The Court will discuss the relevant legal standards and how they apply to key parts of the trial record.

A. Literal Infringement

Patent infringement occurs when a person “without authority makes, uses, offers to sell, or sells any patented invention, within the United States . . . during the term of the patent.” 35 U.S.C. § 271(a). “To prove infringement, a patentee” - here, Supernus - “must supply sufficient evidence to prove that the accused product or process contains, either literally or under the doctrine of equivalents, every limitation of the properly construed claim.” Eli Lilly & Co. v. Hospira, Inc., 933 F.3d 1320, 1328 (Fed. Cir. 2019) (quoting Seal-Flex, Inc. v. Athletic Track & Court Const., 172 F.3d 836, 842 (Fed. Cir. 1999)). “If any claim limitation is absent from the accused [product], there is no literal infringement as a matter of law.” Bayer AG v. Elan Pharm. Rsch. Corp., 212 F.3d 1241, 1247 (Fed. Cir. 2000).

Supernus must demonstrate infringement by a preponderance of the evidence. Eli Lilly, 933 F.3d at 1328 (citation omitted). A preponderance of the evidence means “the greater weight of evidence, evidence which is more convincing than the evidence which is offered in opposition to it.” Hale v. Dep't of Transp., F.A.A., 772 F.2d 882, 885 (Fed. Cir. 1985); see In re Winship, 397 U.S. 358, 371-72 (1970) (Harlan, J., concurring) (defining a preponderance of evidence as “the existence of a fact is more probable than its nonexistence”). “A patentee may prove infringement by any method of analysis that is probative of the fact of infringement,” and “circumstantial evidence may be sufficient.” Martek Biosciences Corp. v. Nutrinova, Inc., 579 F.3d 1363, 1372 (Fed. Cir. 2009) (internal citations omitted).

1. Asserted Claim Limitations Not in Dispute

Though Torrent challenges only two elements of the Asserted Claims, Supernus still must show that Torrent's ANDA Products contain every limitation of the Asserted Claims. Therefore, the Court will first consider Supernus's proof of each limitation not in dispute.¹³ The undisputed claim limitations are as follows:

13 The Court's discussion of undisputed elements will be expeditious. See LG Elecs., Inc. v. Conversant Wireless Licensing S.A.R.L., 759 Fed.Appx. 917, 925 (Fed. Cir. 2019) (“The [Patent Trial and Appeal Board] is ‘not required to address undisputed matters' or arguments about limitations with which it was never presented.” (quoting In re NuVasive, Inc., 841 F.3d 966, 974 (Fed. Cir. 2016))); Abraxis Bioscience, Inc. v. Mayne Pharma (USA) Inc., 467 F.3d 1370, 1380 n.8 (Fed. Cir. 2006) (deeming arguments waived where a party “conceded that the dispute with respect to the doctrine of equivalents was limited to the ‘way' prong of the analysis”).

. The sustained release formulation of topiramate limitation: claim 14 of the '989 patent, claim 14 of the '940 patent, and claim 13 of the '983 patent each recite the limitation “a sustained release formulation of topiramate.” (PFF ¶ 140.)

. The topiramate as an active ingredient limitation: claim 14 of the '989 patent recites the limitation “comprising topiramate as an active ingredient.” (PFF ¶ 147.)

. The released immediately and continuously limitation: claim 14 of the '989 patent, claim 14 of the '940 patent, and claim 13 of the '983 patent each recite the limitation “[topiramate is] released immediately and continuously upon administration from the formulation.” (PFF ¶ 149.)

. The optional immediate release topiramate-containing component limitation: claim 14 of the '989 patent, claim 14 of the '940 patent, and claim 13 of the '983 patent each recite the limitation: “optionally, an immediate release (IR) topiramate-containing component . . . .” (PFF ¶ 255.)

. The maximum plasma concentration (‘Tmax ') of topiramate in vivo at 16 or more hours limitation: claim 14 of the '989 patent, claim 14 of the '940 patent, and claim 13 of the '983 patent each recite the limitation: “wherein the XR component exhibits a maximum plasma concentration of topiramate in vivo at 16 or more hours after a single initial dose.” (PFF ¶ 257.)¹⁴

. The 0.5% to 85% by weight of the formulation, topiramate limitation: claim 14 of the '940 patent recites the limitation: “comprising between 0.5% to 85%, by weight of the formulation, topiramate.” (PFF ¶ 297.)

. The topiramate in the form of micronized particles limitation: claim 13 of the '983 patent recites the limitation: “topiramate in the form of micronized particles ranging from 1 um to 1000 um in size.” (PFF ¶ 299.)

. The pore former limitation: claim 23 of the '983 patent, which depends from claim 13, recites the limitation: “The formulation according to claim 13, wherein the XR component further comprises a pore former selected from the group consisting of . . . polyvinylpyrrolidone . . . .” (PFF ¶¶ 303-304.)

14 Tmax is the time at which the maximum level of the drug within a dosage form has been released from the dosage form. (See Tr. 84:2-8 (Bhatt) (defining Tmax as “the time it takes for the plasma concentration to reach its highest point” recorded “in units of hours or minutes”); Tr. 169:17-22 (Taft) (defining Tmax as the “time point during the experiment [that] you see the peak plasma level,” or Cmax).)

To prove infringement of these claim limitations, Supernus offered testimony of two expert witnesses, who explained how Torrent's ANDA submissions to the FDA supported their opinions; highlighted testimony of Torrent's Rule 30(b)(6) witness admitting that Torrent's ANDA Products contain these claim limitations; and cited, for some limitations, the parties' stipulated facts. (See generally PFF ¶¶ 136-155, 255-308; ECF No. 170 at 13-15.) The experts' testimony is well-reasoned and credible, and the Torrent witness's admissions are straightforward. In addition, Supernus's proposed findings of fact for these claim limitations - although not necessarily stipulated - are not disputed. (See PFF ¶¶ 136-155, 255-308 (Def) (stating throughout that “Torrent is not disputing this element of the claims”).) And Torrent did not offer evidence to the contrary at trial. All told, the Court finds that Supernus has established by a preponderance of the evidence that Torrent's ANDA Products contain the claim limitations not in dispute.

2. Asserted Claim Limitations in Dispute

Now to the disputed limitations: “an extended release (XR) topiramate-containing component, comprising a coating material selected from the group consisting of cellulosic polymers and acrylic polymers.” (ECF No. 171 at 7; ECF No. 88 at 29.)

a. The extended release (XR) topiramate-containing component limitation

The Court starts with the XR component element: “an extended release (XR) topiramate-containing component.” ('989 patent col. 21 ll. 5-6; '940 patent col. 21 ll. 64-65; '983 patent col. 21 ll. 37-38.)¹⁵

15 The claims and specification of the Patents-in-Suit use extended release and XR interchangeably. The Court does the same here.

Supernus's motion in limine. Before trial, the Court ruled that Torrent could not offer rebuttal evidence of Supernus's failure to prove that Torrent's ANDA Products contain an XR component. (ECF No. 151 at 7.) At trial, Supernus restated its objection to Torrent's cross-examining Supernus's experts about the XR component limitation. (Tr. 409:10-22.) The Court will now resolve the reserved issue: whether the testimony that Torrent elicited on cross-examination about the XR component is admissible. The Court rules that it is.

Supernus argues that a party cannot challenge claim limitations that the party did not identify in its noninfringement contentions. The problem with Supernus's proposition is that it would shift the ultimate burden of proving infringement of every claim limitation from the patentee to the accused infringer. See Int'l Rectifier Corp. v. IXYS Corp., 361 F.3d 1363, 1369 (Fed. Cir. 2004) (“Comparison of the claims to the accused device requires a factual determination that every claim limitation or its equivalent is found in the accused device.”). The United States Supreme Court has rejected such a burden shift. See Medtronic, Inc. v. Mirowski Fam. Ventures, LLC, 571 U.S. 191, 203 (2014) (reversing a “change in the ordinary rule imposing the burden of proving infringement upon the patentee” in declaratory judgment actions).¹⁶ The burden of proof stays with the patentee even for limitations that the accused infringer did not identify in noninfringement contentions. See Medtronic Inc. v. Bos. Sci. Corp., 558 Fed.Appx. 998, 1000 (Fed. Cir. 2014) (“This court agrees with the district court that Medtronic's noninfringement contentions based on certain elements alleged to be missing from its devices do not relieve [the patentee's infringement expert] of the requirement to opine on the presence of structure meeting every claim limitation . . . .”).¹⁷The Court will not relieve Supernus of its burden here.

16 For a collection of authorities “establish[ing] that the burden of proving infringement generally rests upon the patentee,” see Medtronic, 571 U.S. at 198-99.

17 In analogous contexts, courts have refused to apply local rules in ways that would modify pleading standards under national rules. See Tyco Fire Prod. LP v. Victaulic Co., 777 F.Supp.2d 893, 904 (E.D. Pa. 2011) (“A local rule must be consistent with-but not duplicate-[the national] federal statutes and rules . . . .” (quoting Fed.R.Civ.P. 83(a)(1)) (alteration in Tyco)); TSMC Tech., Inc. v. Zond, LLC, 2014 WL 7498398, at *7 n.12 (D. Del. Jan. 8, 2014) (“Regardless of what internal procedures a court deploys for patent cases, such procedures could not modify the pleading requirements set forth in the Federal Rules of Civil Procedure, nor allow for disposing with such requirements.”).

To be sure, proper stipulations help expedite litigation, see SRI Int'l v. Matsushita Elec. Corp. of Am., 775 F.2d 1107, 1131 (Fed. Cir. 1985) (“[T]rial judges . . . discourage unnecessary pleadings and encourage stipulation of undisputed matters.”), and a party's decision to dispute overwhelmingly established facts, just because it can, would be in poor judgment. But here, there was little prejudice or surprise to Supernus that its expert's opinions must sustain scrutiny. Besides, Torrent's examination on this limitation was relatively brief. (See Tr. 388:2-393:9.) The Court therefore rules that in these circumstances, the testimony that Torrent elicited on cross-examination about the XR component is admissible for Torrent's argument that Supernus has not met its burden of proof for this element. That part of Supernus's motion is denied.

Torrent's motion in limine. The Court denies Torrent's motion in limine to preclude Dr. Khan's opinion about this limitation's predetermined period of time. (ECF No. 128.)

Torrent's motion highlights a discrepancy between the Court's claim construction and the Asserted Claims. At the Markman stage, the Court construed extended release component to mean, “as stated expressly in the specification, a component that ‘releases topiramate in a continuous manner and is adjusted in such a way that 80% of the active ingredient is released in vitro in the predetermined period of time.'” (ECF No. 88 at 25; '989 patent col. 6 ll. 30-33 (emphasis added).)¹⁸ The predetermined period of time, the Court continued, “refers to the period of time necessary to achieve the ‘predetermined release profile' for the particular ‘sustained release' formulation.” (ECF No. 88 at 25.) The Court reasoned that “[t]his construction is consistent with the sentence following the definition for ‘XR component,' which states, ‘[b]y way of example, and by no means limiting the scope of the invention, the period of time may be not more than 24 hours, not more than 16 hours, not more than 12 hours, not more than 8 hours, or not more than 4 hours, depending on desired attributes of the final product.” (Id.; '989 patent col. 6 ll. 33-38.)

18 At the Markman stage, the Court referred to the specification of the '580 patent. (ECF No. 88 at 3 n.2.) That patent is no longer in suit. So now, references to the specification mean the '989 patent's specification, which the parties agree is identical to the '580 patent's specification. (Tr. 49:3-50:9.)

The discrepancy is that unlike other, no-longer-asserted claims, the Asserted Claims do not expressly recite predetermined periods of time. For example, claim 14 of the unasserted '248 patent recites the following as its final clause: “wherein the XR component releases topiramate in a continuous manner and such that greater than or equal to about 80% of the topiramate is released in vitro in less than or equal to about 12 hours.” ('248 patent col. 22 ll. 12-15, ECF No. 1-4 at 21 (emphasis added).) That claim language recites a predetermined period of “about 12 hours.” Compare that with the final clause of the Asserted Claims, which concerns not the T80%¹⁹ but the Tmax: “wherein the XR component exhibits a maximum plasma concentration of topiramate in vivo at 16 or more hours after a single initial dose.” ('989 patent col. 21 ll. 33-35; '940 patent col. 22 ll. 25-27; '983 patent col. 21 ll. 65-67.)

19 T80% is “[t]he time at which [80%] of the drug within a dosage form has been released from the dosage form . . ., where [80%] is the percent of drug that has been released.” (See '989 patent col. 3 ll. 61-64 (“The time at which a specified percentage of the drug within a dosage form has been released from the dosage form is referred to as the ‘T.sub.x' value, where ‘x' is the percent of drug that has been released.”).)

Based on this, Supernus contends that the period of time for each “particular” sustained release formulation depends on the context of each individual claim. (ECF No. 170 at 16.) Since the Asserted Claims do not recite a predetermined period of time in which 80% of the active ingredient is released, Dr. Khan sought guidance in the specification's definitions of immediate release formulation and sustained release. (Tr. 250:10-251:3 (Khan).) Immediate release formulation is defined as “a formulation that releases greater than or equal to about 80% of the pharmaceutical agent in less than or equal to about 1 hour.” ('989 patent col. 3 ll. 33-35 (emphasis added).) Sustained release is defined as “release of a pharmaceutical agent in a continuous manner over a prolonged period of time.” ('989 patent col. 3 ll. 44-46 (emphasis added).) And prolonged period of time is defined as “a continuous period of time of greater than about 1 hour, preferably, greater than about 4 hours, more preferably, greater than about 8 hours, more preferably greater than about 12 hours, more preferably still, greater than about 16 hours up to more than about 24 hours.” ('989 patent col. 3 ll. 47-52.) So Dr. Khan selected “about 1 hour” as “a default in the definition” of prolonged period of time. (Tr. 251:2-3 (Khan).)

Dr. Khan then reviewed the in-vitro dissolution studies in Torrent's ANDA. (PTX109.39.) A dissolution profile shows that Torrent's ANDA Products release 80% of the drug between about eight to 10 hours. (PTX109.39; Tr. 252:9-253:7 (Khan).) Other data in Torrent's ANDA show that its ANDA Products release on average 11% of the drug in one hour and 80% in eight hours. (PTX109.38-39.) Given this data, Dr. Khan opined that Torrent's ANDA Products infringe the XR component limitation. (Tr. 249:24-25 (Khan).) Supernus adds that the parties stipulated that Torrent's ANDA Products are “extended release formulations of topiramate.” (ECF No. 170 at 18 (citing ECF No. 133 ¶ 59, which states, “The Torrent ANDA Products are extended release formulations of topiramate”).) In Supernus's Request for Admission No. 4, however, Torrent denied that its “ANDA Products include an extended-release component.” (ECF No. 170-1 at 6.)

Torrent does not deny that its ANDA Products are extended release formulations. (PFF ¶¶ 171-172.) Rather, Torrent argues that Supernus did not meet the burden of proving that the ANDA Products contain an XR component “as construed by the Court.” (PFF ¶¶ 171-172 (Def.) (citing ECF No. 88 at 33); Tr. 750:25-751:1).) Torrent attacks Dr. Khan's selection of about one hour as the so-called default prolonged period of time, for three reasons. (ECF No. 171 at 20.)

First, Dr. Khan selected a predetermined period of time based on the different term prolonged period of time. (ECF No. 171 at 21.) Torrent says the prolonged period of time concerns exceeding a period of time for releasing topiramate, whereas the predetermined period of time concerns releasing topiramate in vitro within a period of time. (ECF No. 171 at 21.)²⁰Torrent asserts that Dr. Khan simply applied the same definition and analysis for both the XR component and sustained release claim elements. (ECF No. 171 at 21.)

20 (Compare '989 patent col. 3 ll. 47-52 (defining prolonged period of time as “a continuous period of time of greater than about 1 hour, preferably, greater than about 4 hours, more preferably, greater than about 8 hours, more preferably, greater than about 12 hours, more preferably still, greater than about 16 hours up to more than about 24 hours”), with '989 patent col. 6 ll. 30-39 (describing predetermined period of time, “[b]y way of example, and by no means limiting the scope of the invention,” as “not more than 24 hours, not more than 16 hours, not more than 12 hours, not more than 8 hours, or not more than 4 hours, depending on desired attributes of the final product”) (emphasis added).)

Second, the Court construed the predetermined period of time as referring to “the period of time necessary to achieve the ‘predetermined release profile' for the particular ‘sustained release' formulation” (ECF No. 171 at 21 (quoting ECF No. 88 at 25) (emphasis added by Torrent)), yet Dr. Khan did not consider “the specifics of Torrent's ANDA formulation in determining the appropriate ‘predetermined period of time' to use in his analysis” (ECF No. 171 at 21-22). Dr. Khan merely selected the lowest period of time described in the prolonged period of time definition regardless of the other, more preferable time periods listed there - “preferably, greater than about 4 hours, more preferably, greater than about 8 hours, more preferably greater than about 12 hours, more preferably still, greater than about 16 hours up to more than about 24 hours.” (ECF No. 171 at 22 (quoting '989 patent col. 3 ll. 47-52).)

Third, Dr. Khan's analysis is contrary to the Court's Markman Opinion, which rejected a proposed construction “seeking to construe ‘extended release component' as the inverse of ‘immediate release component.'” (ECF No. 171 at 22 (quoting ECF No. 88 at 26).)

The Court finds that several facts dictate that more likely than not, Torrent's ANDA Products meet the XR component limitation. Although not dispositive, Torrent's ANDA submission describes its ANDA Products as “extended-release” products. (PTX106.2; PTX161.4.) Indeed, the invention's novelty is the sustained release formulation that requires a continuous release of topiramate over a prolonged period of time beyond about one hour. Torrent omitted this element from its noninfringement contentions, and Torrent did not seek to amend its noninfringement contentions even after the Court issued its Markman Opinion. In consequence, Dr. Khan's opinion on this element is unrebutted.

The Court does not rely solely on Dr. Khan's opinion. For one thing, the in-vitro dissolution data in Torrent's ANDA show that its ANDA Products release 80% of topiramate within eight to 10 hours, which far exceeds Dr. Khan's “about 1 hour” selection. For another, the ANDA Products' in-vitro release rate appears to be consistent with the specification's conditions for a bead population whose Tmax in vivo is at least 16 hours. Torrent's ANDA Products' in-vitro dissolution T80% of eight hours, for instance, meets the in-vitro dissolution “conditions” for the XR2 bead population in Example 6 of the Patents-in-Suit. (See '989 patent Example 6.) According to Example 6, a bead population that exhibits an in-vitro dissolution T80% at five to eight hours also exhibits a Tmax “for a single initial dose in-vivo” at “>=16 h” - that is, “a maximum plasma concentration . . . in vivo at 16 or more hours after a single initial does.” (Id.)

The Court therefore concludes that Supernus has proven infringement of this limitation by a preponderance of the evidence.

b. The acrylic polymers limitation

Next, the acrylic polymers limitation: “comprising a coating material selected from the group consisting of cellulosic polymers and acrylic polymers.” (PFF ¶ 174.) At the Markman stage, the Court ruled that this limitation needed no further construction - “the plain meaning of this phrase governs the meaning of the disputed term.” (PFF ¶ 178; ECF No. 88 at 30.) But that ruling concerned the entire limitation; the parties did not seek construction of the term acrylic polymers. Since then, the parties have disagreed over the plain meaning of acrylic polymers. The issue is thus whether the polyvinyl acetate dispersion that Torrent uses as the coating material in its ANDA Products falls within the plain and ordinary meaning of acrylic polymers.²¹ Supernus says that it does; Torrent says that it does not.

21 The parties agree that Torrent's ANDA Products' coating material comprises polyvinyl acetate. (PFF ¶ 25; see also ECF No. 171 at 11 (“For infringement, the only question before the Court is ‘whether polyvinyl acetate is or is not an acrylic polymer.'”).) They also agree that “a POSA viewing the ‘acrylic polymers' claim term would agree that ‘no construction [is] necessary because a POSA knows what the polymer is' because a ‘POSA uses [them] all the time.'” (ECF No. 171 at 11 (quoting PFF ¶ 248 (Pl.)).)

When courts say that a term requires no construction, they mean that further elaboration is unnecessary because the term is common parlance and its meaning is clear. Summit 6, LLC v. Samsung Elecs. Co., 802 F.3d 1283, 1291 (Fed. Cir. 2015); see Schumer v. Lab'y Computer Sys., Inc., 308 F.3d 1304, 1312 (Fed. Cir. 2002) (“These are not technical terms or art, and do not require elaborate interpretation.” (quoting Brown v. 3M, 265 F.3d 1349, 1352 (Fed. Cir. 2001))). But here, the term acrylic polymers is more technical, and the parties clash over its plain and ordinary meaning. Indeed, both parties called experts to present their competing knowledge of the term's “special meaning and usage in the field” of the invention. See Multiform Desiccants, Inc. v. Medzam, Ltd., 133 F.3d 1473, 1477 (Fed. Cir. 1998). It follows that the Court must also measure the experts' testimony against the written record. See Medrad, Inc. v. MRI Devices Corp., 401 F.3d 1313, 1319 (Fed. Cir. 2005) (noting that “the ordinary meaning of the term” is viewed not “in a vacuum” but “in the context of the written description and the prosecution history” (quoting DeMarini Sports, Inc. v. Worth, Inc., 239 F.3d 1314, 1324 (Fed. Cir. 2001))); Phillips v. AWH Corp., 415 F.3d 1303, 1318 (Fed. Cir. 2005) (“[A] court should discount any expert testimony ‘that is clearly at odds with the claim construction mandated by the claims themselves, the written description, and the prosecution history.'” (quoting Key Pharms. v. Hercon Lab'ys Corp., 161 F.3d 709, 716 (Fed. Cir. 1998))); see also Mosaic Brands, Inc. v. Ridge Wallet LLC, 55 F.4th 1354, 1362 (Fed. Cir. 2022) (noting critically that a party arguing the plain and ordinary meaning of a claim term “offers no intrinsic evidence for this position, only extrinsic dictionary definitions and attorney argument”).

This is all to say, the Court cannot resolve the issue without engaging in further claim construction. See Teva Pharms. USA, Inc. v. Sandoz, Inc., 574 U.S. 318, 331 (2015) (noting that consulting extrinsic evidence to understand “the background science or the meaning of a term in the relevant art during the relevant time period” may necessitate Markman-type factfinding “where those subsidiary facts are in dispute”); see also O2 Micro Int'l Ltd. v. Beyond Innovation Tech. Co., 521 F.3d 1351, 1360 (Fed. Cir. 2008) (“The purpose of claim construction is to ‘determin[e] the meaning and scope of the patent claims asserted to be infringed.'” (quoting Markman v. Westview Instruments, Inc., 52 F.3d 967, 976 (Fed. Cir. 1995), aff'd, 517 U.S. 370 (1996))). For clarity, the Court will discuss the type and priority of evidence that it will consider.²²

22 Other courts have found it necessary to provide additional claim construction as part of infringement rulings following bench trials. See, e.g., NobelBiz, Inc. v. Glob. Connect, L.L.C., 701 Fed.Appx. 994, 997 (Fed. Cir. 2017) (“The district court had the responsibility to determine the scope of the asserted claims, and ‘[a] determination that a claim term “needs no construction” or has the “plain and ordinary meaning” may be inadequate when a term has more than one “ordinary” meaning or when reliance on a term's “ordinary” meaning does not resolve the parties' dispute.'” (quoting O2 Micro Int'l, 521 F.3d at 1361)); Hoganas AB v. Dresser Indus., Inc., 9 F.3d 948, 950 n.7 (Fed. Cir. 1993) (noting with approval the trial court's interpretation of a term “after a full bench trial”); Apex Inc. v. Raritan Computer, Inc., 325 F.3d 1364, 1369 (Fed. Cir. 2003) (reviewing an infringement ruling after a bench trial where the district court first construed disputed limitations and then found no infringement); Renishaw PLC v. Marposs Societa' per Azioni, 158 F.3d 1243, 1248, 1253 (Fed. Cir. 1998) (affirming district court's construction following a bench trial); Azurity Pharms., Inc. v. Alkem Lab'ys Ltd., - F.Supp.3d -, 2023 WL 3254117 (D. Del. May 4, 2023) (engaging in claim construction following a bench trial).

“[T]he ordinary and customary meaning^[23] of a claim term is the meaning that the term would have to a person of ordinary skill in the art in question at the time of the invention, i.e., as of the effective filing date of the patent application.” Phillips, 415 F.3d at 1313. To ascertain that meaning, the Court first considers the intrinsic evidence: the claims, specification, and prosecution history. Markman, 52 F.3d at 979 (quoting Unique Concepts, Inc. v. Brown, 939 F.2d 1558, 1561 (Fed. Cir. 1991))).^[24]

[23] The Federal Circuit Court of Appeals has used the phrases plain and ordinary meaning and ordinary and customary meaning interchangeably. See, e.g., Laitram, LLC v. Ashworth Bros., 2023 WL 3449148, at *3 (Fed. Cir. May 15, 2023); Ruckus Wireless, Inc. v. Innovative Wireless Sols., LLC, 824 F.3d 999, 1002-03 (Fed. Cir. 2016); Thorner v. Sony Computer Ent. Am. LLC, 669 F.3d 1362, 1365 (Fed. Cir. 2012).

[24] See also Phillips, 415 F.3d at 1313 (“[POSA] is deemed to read the claim term not only in the context of the particular claim in which the disputed term appears, but in the context of the entire patent, including the specification.”); Blazer v. Best Bee Bros. LLC, 2022 WL 16954848, at *4 (Fed. Cir. Nov. 16, 2022) (“The specification is especially significant when a claim term possesses ‘a range of possible ordinary meanings in context' or otherwise features ‘genuine uncertainties.'” (quoting Straight Path IP Grp., Inc. v. Sipnet EU S.R.O., 806 F.3d 1356, 1361 (Fed. Cir. 2015))).

If the intrinsic record is not telling enough, the Court must consider extrinsic evidence: “expert and inventor testimony, dictionaries, and learned treatises.” Id. at 980; see Teva Pharms., 574 U.S. at 331 (“In some cases, . . . the district court will need to look beyond the patent's intrinsic evidence and to consult extrinsic evidence in order to understand, for example, the background science or the meaning of a term in the relevant art during the relevant time period.”).^[25] An expert may testify as to “how those skilled in the art would interpret the claims.” Markman, 52 F.3d at 979 (citation omitted).^[26]

[25] See also Actelion Pharms. LTD v. Mylan Pharms. Inc., 85 F.4th 1167, 1173-74 (Fed. Cir. 2023) (noting that district courts must consult extrinsic evidence when the intrinsic evidence is not telling enough for claim construction); Vitronics Corp. v. Conceptronic, Inc., 90 F.3d 1576, 1583 (Fed. Cir. 1996) (“Extrinsic evidence may also be considered, if needed to assist in determining the meaning or scope of technical terms in the claims.” (quoting Pall Corp. v. Micron Separations, Inc., 66 F.3d 1211, 1216 (Fed. Cir. 1995))).

[26] Though “expert testimony can be useful to . . . establish that a particular term in the patent or the prior art has a particular meaning in the pertinent field,” of no use are “conclusory, unsupported assertions by experts as to the definition of a claim term” or “any expert testimony that is clearly at odds with the claim construction mandated by the claims themselves, the written description, and the prosecution history, in other words, with the written record of the patent.” Phillips, 415 F.3d at 1318 (internal citations and quotation marks omitted).

To be sure, a dispute among experts testifying about a claim term's meaning is factual for the Court to resolve. Teva Pharms., 574 U.S. at 332. Indeed, “‘[e]xperts may be examined to explain terms of art, and the state of the art, at any given time,' but they cannot be used to prove ‘the proper or legal construction of any instrument of writing.'” Id. (quoting Winans v. New York & E.R. Co., 62 U.S. 88, 100-01 (1858)). After resolving the factual dispute, the Court turns to the legal issue: it must “interpret the patent claim in light of the facts as [the Court] has found them.” Id.

The intrinsic evidence. The Court begins with the intrinsic record, starting with the claims. The Asserted Claims require a “sustained release formulation.” (See, e.g., '989 patent col. 21 l. 1.) This requirement indicates that the claimed acrylic polymers must be suitable for sustained release formulations. Other than that, the Asserted Claims shed little light on the meaning of acrylic polymers, so the Court turns to the specification for guidance. Genentech, Inc. v. Wellcome Found. Ltd., 29 F.3d 1555, 1563-64 (Fed. Cir. 1994).

The specification tells us more. It says “[t]he coating material is preferably selected from a group comprising cellulosic polymers, such as ethylcellulose, methylcellulose, hydroxypropyl cellulose, hydroxypropylmethyl cellulose, cellulose acetate, and cellulose acetate phthalate; polyvinyl alcohol; acrylic polymers such as polyacrylates, polymethacrylates and copolymers thereof, and other water-based or solvent-based coating materials.” ('989 patent col. 7 ll. 4-13.) From this the Court gathers a few things. The use of such as indicates that the claimed acrylic polymers are not limited to the examples listed. Indeed, the specification later describes an exemplary copolymer not expressly included in that list: the XR8 formulation in Table 1, “Acrylic polymers (Eudragit^® RL30D/RS30D),” which Dr. Felton described as an “ammonio methacrylate copolymer.” ('989 patent Table 1; Tr. 556:14-17 (Felton).) The specification likewise uses such as to list non-exhaustive examples of claimed cellulosic polymers. It does not do the same for polyvinyl alcohol, which is in the singular form unlike cellulosic polymers and acrylic polymers.

In addition, the specification uses semicolons to offset cellulosic polymers, polyvinyl alcohol, and acrylic polymers. (ECF No. 171 at 9.) This punctuation indicates that for this specification, these polymers are “separate and distinct” from one another. See In re Affinity Labs of Texas, LLC, 856 F.3d 902, 907 (Fed. Cir. 2017) (finding that semicolons offsetting each of five steps of a method claim “strongly indicates that each step is separate and distinct”); In re Pelz, 379 Fed.Appx. 975, 978 (Fed. Cir. 2010) (noting that using “semicolons to separate the particulars for that list . . . indicates that the three positions are exemplary subsets”).

But the Patents-in-Suit neither define these scientific terms nor disclaim a certain meaning. As a result, a POSA's understanding of the terms' plain and ordinary meaning is still unclear. The Court thus turns to the prosecution history.

“The prosecution history, in particular, may be critical in interpreting disputed claim terms, and even where prosecution history statements do not rise to the level of unmistakable disavowal, they do inform the claim construction.” Univ. of Massachusetts v. L'Oreal S.A., 36 F.4th 1374, 1379 (Fed. Cir. 2022) (quoting Personalized Media Commc'ns, LLC v. Apple Inc., 952 F.3d 1336, 1340 (Fed. Cir. 2020)) (internal quotation marks omitted); see Canopy Growth Corp. v. GW Pharma Ltd., 2023 WL 3048243, at *5 (Fed. Cir. Apr. 24, 2023) (opting not to decide whether amendments amounted to disavowal or disclaimer where the prosecution history showed that the patent “discloses three non-overlapping embodiments while claiming only one of them” (citing L'Oreal S.A., 36 F.4th at 1379)).

Relevant here is the applicants' August 2012 amendment during the prosecution of U.S. Patent Application No. 12/926,931 ('931 application). (ECF No. 170 at 33; PTX233.1.) An earlier version of subsection (a) of claim 104 in the '931 application read: “(a) at least two different extended release topiramate-containing components, wherein each component comprises a release controlling coating specific for its component . . . .” (PTX233.210.) The examiner rejected this claim for inadequately describing the claimed release controlling coating. (ECF No. 170 at 33; PTX233.230.) The applicants responded that they disagreed but would amend “solely in the interest of compact prosecution.” (PTX233.249.) Their amendment added that the release controlling coating must “comprise[] a coating material selected from the group consisting of cellulosic polymers and acrylic polymers.” (PTX233.242; ECF No. 170 at 34.) The applicants remarked that the amended claims “now recite species of the noted genera.” (PTX233.249.) The specification also discloses cellulosic polymers, polyvinyl alcohol, and acrylic polymers, though this amendment added only cellulosic polymers and acrylic polymers. (Compare '989 patent col. 7 ll. 4-13 (specification), with PTX233.242 (amendment).)

Still, nothing in the prosecution history disclaims a plain meaning of acrylic polymers or prescribes it a special definition.

All intrinsic evidence considered, including a reading of the entire Patents-in-Suit, the Court would benefit from considering extrinsic evidence. The Court, in its discretion, turns to the experts who testified at trial about how a POSA would understand the plain and ordinary meaning of acrylic polymers and whether that meaning covers polyvinyl acetate. See Aventis Pharma S.A. v. Hospira, Inc., 675 F.3d 1324, 1331 (Fed. Cir. 2012) (noting the district court's “sound discretion” to use extrinsic evidence to learn about “the field of the invention” and what a POSA “would understand claim terms to mean” (quoting Phillips, 415 F.3d at 1319)).

The infringement experts. Supernus's infringement expert Mansoor A. Khan, R.Ph., Ph.D., was admitted as an expert in the field of pharmaceutical formulations and drug delivery - specifically, the development, evaluation, and characterization of sustained release formulations and the ingredients in such formulations. (PFF ¶¶ 59, 69; Tr. 231:22-24.)

Torrent called two experts for its infringement rebuttal - David P. Rotella, Ph.D., an expert in structural chemistry and organic chemistry; and Linda Felton, Ph.D., an expert in drug formulation and modified drug delivery systems. (PFF ¶¶ 82-95, 96.) Both were admitted as experts. But in post-trial submissions, Supernus challenged Dr. Rotella's qualification as a POSA. The Court will now address that challenge.

During claim construction, the Court adopted an undisputed definition of POSA:

[S]omeone in the 2006 time frame with at least a Bachelor of Science degree in Pharmaceutical Sciences or a related field, approximately three to five years of experience in a drug delivery technology or related field, and working knowledge regarding pharmacokinetics (or a person of commensurate education and experience).

[(PFF ¶ 57; ECF No. 88 at 9 (emphasis added).)]

This definition requires “at least a Bachelor of Science degree in Pharmaceutical Sciences or a related field”; Dr. Rotella received a bachelor's degree in pharmacy in 1981 and an advanced degree in medicinal chemistry in 1985. (Tr. 447:23-448:2; DTX025.) The definition also requires “experience in a drug delivery technology or related field”; Dr. Rotella has experience in medicinal chemistry and drug discovery, which he testified are fields related to “a drug delivery technology” field. (Tr. 447:25-448:2; see Tr. 446:8-11 (Rotella) (testifying that organic chemistry is “essentially the same thing” as drug-delivery technology).) And Dr. Rotella testified that he has a working knowledge of pharmacokinetics. (Tr. 446:12-22, 447:24-25.)

On these credentials, Dr. Rotella was admitted as an expert in the fields of structural chemistry and organic chemistry. (Tr. 448:6-449:23.) In qualifying Dr. Rotella, Torrent made clear that he would offer his opinions from a POSA's perspective. (Tr. 447:17-448:5.) Specifically, Torrent asked Dr. Rotella if he was familiar with the Court's construction of the term POSA, if he considered himself to be a POSA, and if he applied a POSA standard in forming his opinions - to all of which Dr. Rotella responded affirmatively. (Tr. 447:17-448:5.)

At that time, Supernus objected, but not to Dr. Rotella's opining as a POSA:

MR. SETH (ATTORNEY FOR TORRENT). Okay. Your Honor, at this time we offer Dr. Rotella as an expert witness qualified to testify as an expert in the fields of structural chemistry, organic chemistry, and to testify about non-infringement.

THE COURT. Any objection by Supernus?

MR. KURZ (ATTORNEY FOR SUPERNUS). Your Honor, Supernus doesn't object to the extent that they're offering Dr. Rotella as an expert in medicinal chemistry or organic chemistry or perhaps structural chemistry. I'm not quite sure what counsel is getting at with the rest of what he asked for.

THE COURT. At this point in the trial there should be no objections to the parties offering an expert opinion as what their witness is going to provide expert testimony on, so I'm not sure of the disconnect here.

MR. KURZ. Anyway, Your Honor, Supernus has no objection to them offering Dr. Rotella as an expert in medicinal chemistry or organic chemistry.

THE COURT. Is there -- I'm not sure I'm following. Is there a disconnect here, Mr. Seth?

MR. SETH. If I'm understanding correctly, there's no objection; is that right?

MR. KURZ. For medicinal chemistry and organic chemistry. I'm not quite sure. He said something else, I thought.

THE COURT. Yeah, let's go back. What are you offering Dr. Rotella as an expert in?

MR. SETH. I think I said structural chemistry, organic chemistry, and to testify about non-infringement.

THE COURT. So offering as an expert in structural chemistry, organic chemistry, and offering an opinion with respect to infringement.

MR. SETH. Non-infringement.

THE COURT. Non-infringement, correct, right. Any objection with respect to Dr. Rotella as an expert in that area?

MR. KURZ. With respect to offering testimony in organic chemistry and structural chemistry, there's no problem. I'm not quite sure what it is to be an expert in non-infringement, Your Honor. But as far as offering that to prove or to testify about non-infringement, we have no objection to that.

THE COURT. Okay. So the Court accepts Dr. Rotella as an expert in structural chemistry and organic chemistry.

MR. SETH. Okay.

MR. KURZ. Thank you, Your Honor.

[(Tr. 448:6-449:25.)] Nor did Supernus previously move in limine to preclude Dr. Rotella's opinions as a POSA. Yet in its post-trial brief, Supernus argued that “Dr. Rotella's testimony from the perspective of a POSA is legally improper and substantively incorrect,” because Dr. Rotella is a medicinal chemist without the requisite experience in drug-delivery technology. (ECF No. 170 at 22-23 (emphasis omitted).)

The Court sees two reasons not to disqualify Dr. Rotella as a POSA. First, Supernus had plenty of notice of Dr. Rotella's background and qualifications, and the Final Pretrial Order described how he would rebut Dr. Khan's opinion. (ECF No. 133 at 83.) That Supernus waited until post-trial briefing to object to Dr. Rotella's status as a POSA is enough for the Court to find that Supernus waived its objection. See Kreepy Krauly U.S.A., Inc. v. Sta-Rite Indus., Inc., 152 F.3d 949 (Fed. Cir. 1998) (ruling that by failing to object when the court qualified a witness as an expert, the party “waived any objection to the court's decision to qualify the expert”).

Second, since Supernus waived its challenge to Dr. Rotella's experience as being “a related field,” the Court views Supernus's arguments against Dr. Rotella's POSA status as truly credibility attacks on his qualifications. For instance, Supernus argues that medicinal chemistry (Dr. Rotella's field) is not a field related to drug delivery. (ECF No. 170 at 23-24.) Supernus reasons that unlike the pharmaceutical science of drug delivery, medicinal chemistry studies biologically active substances (here, topiramate), not inactive ingredients or excipients (here, polymer coatings). (PFF ¶ 110; ECF No. 170 at 23; see Tr. 496:24-25 (Rotella) (“Medicinal chemistry is the study of the chemistry of biologically active substances.”).) Supernus also cites Dr. Felton's testimony that the POSA definition “doesn't include a medicinal chemist,” who “is somebody who would synthesize a chemical entity, and that's not what these patents are about.” (Tr. 584:7-18.)^[27]

[27] Dr. Felton's opinion that the POSA definition “doesn't include a medicinal chemist” might carry more weight if Dr. Rotella were only a medicinal chemist. But as mentioned, Dr. Rotella's bachelor's degree in pharmacy and professional experience, which Supernus delinquently challenged, meet the broad POSA definition requiring a “Bachelor of Science degree in Pharmaceutical Sciences” and experience in a field related to drug delivery technology. (See also Tr. 750:7-10 (Torrent arguing that Dr. Felton was correct that being a medicinal chemist is not by itself enough to be a POSA but that Dr. Rotella has the requisite experience).)

Supernus also asserts that Dr. Rotella admittedly has never designed a sustained-release formulation, is not involved in conducting formulation research, and has never developed a drug with FDA approval. (ECF No. 170 at 24 (citing Tr. 497:8-18, 499:10-13, 498:5-7 (Rotella)).) And Supernus contests that Dr. Rotella's experience working with formulators “for an unspecified period of time” qualifies him as a POSA. (ECF No. 170 at 24.) These attacks may raise credibility issues, but at this stage, the Court finds that Dr. Rotella's testimony as a POSA should not be stricken from the trial record.^[28]

[28] The Court also agrees with Torrent that Supernus opened the door to testimony about chemistry when Dr. Khan included an empirical formula as part of his definition of acrylic polymers and consulted a chemistry webbook in forming his definition. (Tr. 747:13-19.) The Court discusses Dr. Khan's definition next.

The Court turns to the parties' dispute over a POSA's understanding of acrylic polymers.

Dr. Khan's definition of acrylic polymers. Supernus's expert Dr. Khan first testified that a POSA would agree that the claim term acrylic polymers needs no construction, “because a POSA knows what the polymer is,” a “POSA knows what function it plays,” a “POSA uses [them] all the time.” (Tr. 257:6-15.) But because the parties dispute the term's meaning, Dr. Khan sought “to come up with something so that everybody understands.” (Tr. 257:16-24; see Tr. 255:10-11 (Khan) (“I was trying to get at the intent of the developer here, the formulator here.”).) The challenge for Dr. Khan, he said, was that acrylic polymers represent a broad group, so broad that he knew of no single definition encompassing all of them. (See Tr. 257:19-21 (Khan) (“There's no single definition that I'm aware of that covers all [the] acrylic polymers.”); Tr. 262:3-4 (Khan) (“[B]ut I haven't seen a single definition that covers all the acrylic polymers.”).)^[29]

[29] (See also Tr. 256:2-8 (Khan) (“So I was trying to look at what the acrylic polymer means. It may mean different things to different people. It's a pretty broad term. Some people may use a very narrow definition of that. Some people may use a broad definition of that. In the excipient world we use broad, so many different types of acrylic polymers in the pharmaceutical development.”).)

So, trying to conceptualize what a POSA would understand acrylic polymers to mean here, Dr. Khan fashioned his own “operative definition”: “A polymer containing a monomer from the series of olefin acids with the general formula [CnH2n-2O2] or their derivatives.” (Tr. 258:1-4 (cleaned up); PFF ¶ 181.) At closing arguments, Supernus explained that “to be helpful, to have some words that Your Honor could utilize to do your infringement analysis, Dr. Khan put some meat on the ruling that is plain and ordinary meaning[,] and this is what he arrived at.” (Tr. 723:1-4.) To get this definition, Dr. Khan drew from a POSA's “general understanding” based on everyday usage of acrylic polymers. He also consulted Torrent's ANDA submission, the specification, and several extrinsic sources.

Starting with the intrinsic record, Dr. Khan testified that a POSA reading the patents would understand that the plain and ordinary meaning of acrylic polymers must encompass at least the polymers exemplified in the specification: here,^[30] “polymers such as polyacrylates, polymethacrylates and copolymers thereof” and the “Acrylic polymers (Eudragit^®RL30D/RS30D).”^[31](ECF No. 170 at 11; PFF ¶ 55 (Pl.) (citing Tr. 254:6-255:11, 258:8-259:4, 263:2-264:5 (Khan)); PFF ¶¶ 187-189, 251 (Pl.).) Eudragit^® RL 30 D and Eudragit^® RS 30 D are the tradenames of two polymethyl methacrylates listed in the Handbook of Pharmaceutical Excipients as pharmaceutically acceptable acrylic polymers. (PFF ¶ 184 (Pl.) (citing Tr. 272:6-14, 312:6-21 (Khan); PTX113.3; PTX115.5); Tr. 556:12-17, 628:18-21 (Felton).) They are used for extended-release applications. (PFF ¶ 185 (citing Tr. 627:18-21 (Felton)).)

[30] Mindful of the rule that “[e]xperts may be examined to explain terms of art, and the state of the art, at any given time, but they cannot be used to prove the proper or legal construction of any instrument of writing,” Teva Pharms., 574 U.S. at 332 (internal quotations omitted), the Court considers expert testimony only to find how a POSA would understand the plain and ordinary meaning of acrylic polymers in a similar context.

[31] The Federal Circuit “has cautioned against interpreting a claim term in a way that excludes disclosed embodiments, when that term has multiple ordinary meanings consistent with the intrinsic record.” Helmsderfer v. Bobrick Washroom Equip., Inc., 527 F.3d 1379, 1383 (Fed. Cir. 2008) (citing Verizon Servs. Corp. v. Vonage Holdings Corp., 503 F.3d 1295, 1305 (Fed. Cir. 2007)); see Abbott Lab'ys v. Andrx Pharms., Inc., 473 F.3d 1196, 1211 (Fed. Cir. 2007) (rejecting a definition of the term pharmaceutically acceptable polymer because the “definition would not cover some of the very polymers listed” in the specification).

But other grades of Eudragit^® polymers, Dr. Khan explained, would not be suitable acrylic polymers for the Patents-in-Suit. (PFF ¶ 186 (Pl.) (citing Tr. 375:2-376:23 (Khan)).) For example, Eudragit L^®, an enteric polymer, is delayed release and thus does not release “immediately and continuously” as the claims require. For another, Eudragit E^®, an organic polymer, is not a dispersion and is used in organic solutions. (PFF ¶ 186 (Pl.).) The difference in available acrylic polymers indicates that acrylic polymers, as used here, must include not all acrylic polymers - only ones suitable for extended-release applications.

In fashioning his definition, Dr. Khan also reviewed Torrent's ANDA submission and supporting data. Specifically, Dr. Khan analyzed what ingredient Torrent used, whether it fit into his definition, and whether it had the same properties and “chemistry related to the functioning of this polymer.” (Tr. 256:20-23 (Khan).) Dr. Khan clarified that he focused not on the “pharmacological action” but on “the use as an excipient.” (Tr. 256:23-257:1.) “Based on all that,” Dr. Khan testified, “I knew that it's the same polymer, acrylic polymer.” (Tr. 257:2-3.)

Dr. Khan also reviewed “some of the scientific literature” - “for example, the National Institute of Standards and Technology,” or “NIST,” “chemistry workbook.^[32] (Tr. 256:15-17.) The problem, according to Dr. Khan, was that “a lot of chemistry handbooks do not define things,” and instead “have structures.” (Tr. 260:3-8.) This touches a critical theory of Supernus's case: when classifying a polymer, a pharmaceutical scientist focuses on the polymer's function, not its structure. (PFF ¶ 55 (Pl.) (citing Tr. 345:17-346:1, 355:56-20 (Khan)).) It is where pharmaceutical scientist and chemists diverge, according to Dr. Khan:

[32] The parties describe the NIST as “a Federal Government organization that sets standards.” (PFF ¶ 229.)

That's where the difference between chemist and pharmaceutical scientist come. A chemist would look at it as an ester because . . . yeah, as a structure you would see it as an ester. But a pharmaceutical scientist will not see that as a difference because the function doesn't change.

[(Tr. 345:17-21.)]

Dr. Khan ultimately landed on the Webster's New World Dictionary of the American Language, Second College Edition,^[33] which offers two definitions of the word acrylic. The first definition is “designating or of a colorless, pungent acid, CH2:CHCOOH, obtained by the oxidation of acrolein.” (PTX116.3.) Dr. Khan testified that this first definition is too narrow, as it does not even cover polymethacrylates, one of the acrylic polymers listed in the specification. (PFF ¶¶ 203-204 (Pl.) (citing Tr. 260:25-263:13 (Khan); PTX116.3; '989 patent col. 7 ll. 6-13)).)

[33] Dr. Khan added that because this dictionary was copyrighted in 1986, it would have “a definition . . . that I thought was closest to what's there in the patent,” whose priority date is in 2006. (Tr. 259:14-19.) Webster's was copyrighted in 1986 and was therefore a reference that was available to a POSA before the 2006 priority date. (PFF ¶ 200 (Pl.).)

Dr. Khan then turned to the second definition: “designating or of a series of olefin acids with the general formula CnH2n-2O2.” (PTX116.3.) He felt that this definition was better but still not broad enough to cover all acrylic polymers in the specification - namely, “Acrylic polymers (Eudragit^® RL30D/RS30D)” in Table 1, which are derivatives of “a polymer containing a monomer from the series of olefin acids with the general formula CnH2n-2O2.” (PFF ¶ 55 (Pl.) (citing Tr. 257:4-259:4, 261:25-264:5, 325:7-19 (Khan); PTX116.3; '989 patent Table 1).) And so, Dr. Khan added to the second definition the words or their derivatives.^[34]

[34] Dr. Khan also testified that he “had to include the word[s] ‘or their derivatives' to encompass the acrylic polymers listed in the specification, . . . because those polymers are not acids.” (Tr. 347:16-20.)

Dr. Khan's definition also includes the phrase “a polymer containing a monomer.” A polymer is made of many monomers bonded together through a chemical reaction called polymerization. (PFF ¶¶ 193-194 (citing Tr. 265:3-266:25 (Khan), 467:6-9 (Rotella)).) But not all monomers can form polymers. (See Tr. 539:22-540:11 (Rotella) (“[S]ome of these molecules may not be able to form polymers.”).) So to meet Dr. Khan's definition, a compound must be a polymer, which means that it must have monomers that can form a polymer. (PFF ¶¶ 194, 196-197 (Pl.).)

Thus, for a polymer to meet Dr. Khan's definition of acrylic polymers, the monomer must meet “the general formula CnH2n-2O2” or be a derivative of such a compound. (PFF ¶ 214 (Pl.) (citing 257:4-258:6 (Khan)).) The general formula CnH2n-2O2 indicates the ratio of carbon atoms (C) to hydrogen atoms (H) to oxygen atoms (O). (PFF ¶ 215 (Pl.) (citing Tr. 267:23-268:16 (Khan)).) The monomer must also be “a monomer from the series of olefin acids.” (PFF ¶ 216 (Pl.) (citing 267:4-15, 268:16-17 (Khan)).) For a monomer to be an olefin acid, it must have “an olefin component and an acid component.” (PFF ¶ 217 (Pl.) (citing Tr. 267:4-15, 268:16-17 (Khan)).) An olefin is a double-bonded structure. (PFF ¶ 218 (Pl.) (citing Tr. 267:11-15 (Khan)).)

According to Dr. Khan, polyvinyl acetate fits within his acrylic polymers definition. The monomer that is used to make polyvinyl acetate is vinyl acetate. (PFF ¶ 195.) Vinyl acetate satisfies the general formula in Dr. Khan's definition: CnH2n-2O2. (PFF ¶ 226 (Pl.) (citing Tr. 267:4-268:25 (Khan)).) Vinyl acetate's chemical formula is C4H6O2 - that is, four carbon (n = 4), six hydrogen (2n-2 = 6), and two oxygen atoms. (ECF No. 170 at 27 (citing PTX117.1; PTX113.3 at Heading 4; Tr. 268:2-16 (Khan)).) Vinyl acetate is also “a monomer from the series of olefin acids.” (Id.) Dr. Khan testified that an olefin acid has “an olefin component and an acid component.” (Tr. 267:4-15, 268:16-14.) The following table illustrates vinyl acetate's structure in two ways - on the left as a line drawing, and on the right showing all of the atoms, with arrows included to identify each component:

(Image Omitted) [(ECF No. 170 at 27.)]

Dr. Khan testified that the term olefin refers to the double-bonded structure, which is also known as a vinyl group. (Tr. 267:11-15; see Tr. 270:6 (Khan) (“The vinyl is olefin.”).)^[35] Dr. Khan identified the olefin group in vinyl acetate as the “CH, double-bond CH2,” or “CH=CH2,” group. (PTX117; ECF No. 170 at 28 (citing Tr. 267:19-20, 270:18-20, 271:12-13 (Khan)).) For the acid component, Dr. Khan identified the “CH3CO” portion of the structure. (PTX117; ECF No. 170 at 28 (citing Tr. 267:20-21, 270:17-18, 271:10-11 (Khan)).) In fact, according to the NIST WebBook, vinyl acetate is also known as “acetic acid vinyl ester,”^[36] whose structure includes olefin (the vinyl group) and acid (acetic acid) components. (ECF No. 170 at 28 (citing PTX117; Tr. 269:11-270:4 (Khan)); PFF ¶¶ 230-231 (Pl.)).)

[35] (See PFF ¶ 227 (Def.) (“The terms ‘olefin' and ‘vinyl' both refer to a carbon-carbon double bond (C=C), also known as a functional group, found in molecules.” (citing Tr. 460:22-24, 469:16-19 (Rotella)).)

[36] Dr. Khan testified that “an ester is a derivative of an acid,” given that “acid and alcohol react to make an ester.” (Tr. 346:13-16.) Dr. Khan also testified that vinyl acetate is an ester that forms from acetic acid, thus making vinyl acetate a derivative of acetic acid. (Tr. 346:21-347:3.) But Dr. Khan testified that “acetic acid is different from acrylic acid,” because acrylic acid has acid and olefin components, whereas “acetic acid doesn't have the olefin.” (Tr. 347:4-9.) “That's why vinyl acetate is” acrylic “but acetic acid is not.” (Tr. 347:9-10.)

Thus, if the Court accepts Dr. Khan's definition as the plain and ordinary meaning of acrylic polymers, Torrent's use of polyvinyl acetate in its ANDA Products literally infringes the Patents-in-Suit.

Torrent's response. Torrent rejects what it calls Dr. Khan's “Frankenstein” definition, claiming that he invented a self-serving definition by modifying a definition in a general-purpose dictionary that Supernus gave him. Torrent argues that Dr. Khan's definition is erroneous for several reasons. For starters, Webster's is not the type of technical resource that a POSA would ever consult. In fact, a POSA reading the specification would not feel the need to find a so-called special definition of acrylic polymers. This is because, as Dr. Khan conceded, a POSA would consult a dictionary only for unfamiliar terms. (Tr. 335:24-336:4 (Khan).) Yet, as Dr. Khan opined, a POSA would be familiar with the term acrylic polymers. (Tr. 336:5-8 (Khan).) So instead of being based on a general-purpose dictionary definition, a POSA's understanding of the plain and ordinary meaning of acrylic polymers is reflected in the myriad of technical literature in the field. And none of the technical literature in evidence refers to or classifies polyvinyl acetate as an acrylic polymer. Indeed, Dr. Khan has never seen vinyl acetate described by name as an acrylic polymer. (ECF No. 171 at 11 (citing Tr. 312:22-25, 314:22-24 (Khan)).)

Torrent attacks Dr. Khan's motive in crafting his definition. Torrent first asserts that Dr. Khan approached his POSA task with the wrong lens: he testified that he “was trying to get at the intent of the developer here, the formulator here,” and “the intent of this author of that patent.” (ECF No. 171 at 14 (citing Tr. 255:10-11, 322:17-19 (Khan)).)^[37] Indeed, “the focus in construing disputed terms in claim language is not the subjective intent of the parties to the patent contract when they used a particular term”; it “is on the objective test of what one of ordinary skill in the art at the time of the invention would have understood the term to mean.” Markman, 52 F.3d at 986. The exception to this rule is the parties' subjective intent “as documented in the prosecution history.” Id. at 985.

[37] (See also Tr. 313:11-19 (Khan) (“I would say that vinyl acetate has vinyl background. Vinyl acetate has vinyl background. Methacrylates and acrylates also have vinyl background, as a polymer, when I look at the background, both have vinyl structures, so for me it is an acrylic polymer, but in terms of somebody specifically using the word, I don't know.”).)

Torrent also asserts that Dr. Khan added the words or their derivatives to the Webster's definition “after a long discussion and conversation” with Supernus's counsel to give him flexibility to apply that definition to other polymers outside the scope of the Webster's definition. (ECF No. 171 at 14 (quoting Tr. 262:2-10, 347:16-20 (Khan)).)^[38]

[38] Torrent also asserts that Dr. Khan's trial testimony conflicts with his deposition testimony, where Torrent says Dr. Khan agreed that “absent the addition of the words ‘or their derivatives,' the Webster's definition does not cover vinyl acetate.” (ECF No. 171 at 14 (quoting Tr. 343:12-344:4 (Khan)); Tr. 339:11-340:3 (Khan).) At his deposition, when asked if he agreed that the monomer “vinyl acetate does not fit within either definition of acrylic provided in Webster's,” Dr. Khan responded as follows: “Yeah, I thought I indicated that just the definition acrylic has nothing to do with acrylic polymers, just the definition of acrylic. If you're asking for vinyl, vinyl acetate, not polyvinyl, vinyl acetate, vinyl acetate is not an acid.” (Tr. 343:12-21.) Dr. Khan's response makes sense, as Supernus submits, because “[v]inyl acetate is a monomer, and thus would not be described as a polymer.” (ECF No. 170 at 28.) At trial, Torrent confronted Dr. Khan with this deposition testimony. (Tr. 336:21-344:17.) Dr. Khan then clarified that he responded “Yeah” not to Torrent's question but to counsel's “asked and answered” objection, insisting that he “repeatedly said it does fall into it, and that's an acrylic polymer.” (Tr. 344:6-17.) On redirect, Supernus read into the trial record excerpts of Dr. Khan's deposition testimony that “polyvinyl acetate does fall into” Webster's second definition of acrylic. (Tr. 394:2-396:9.) Torrent objected to Supernus's reading, and the Court reserved on its ruling. (Tr. 396:11-14.) The Court now overrules Torrent's objection. Under Federal Rule of Evidence 106, “[i]f a party introduces all or part of a writing or recorded statement, an adverse party may require the introduction, at that time, of any other part-or any other writing or recorded statement-that in fairness ought to be considered at the same time.” Torrent's use of a small part of Dr. Khan's deposition transcript created confusion about whether Dr. Khan believes that polyvinyl acetate, as opposed to the monomer vinyl acetate, fit Webster's second definition of acrylic before Dr. Khan added the phrase or their derivatives. In fairness, Supernus was entitled to repair the record by reading other excerpts of Dr. Khan's deposition transcript giving context to the allegedly conflicting testimony.

The technical literature. For coating materials, according to Torrent's expert Dr. Felton, formulators “categorize the polymers in three different categories: Cellulosic polymers, acrylic polymers, and vinyl polymers.” (ECF No. 171 at 8 (citing Tr. 554:7-10 (Felton)).) For its part, Torrent relies on excerpts from a variety of technical literature that “uniformly refer to polyvinyl acetate as a vinyl polymer, not an acrylic polymer.” (Id. at 11.)

First is a table from Aqueous Polymeric Coatings for Pharmaceutical Dosage Forms, which Dr. Felton co-edited. (Id. at 8 (citing Tr. 554:9-10 (Felton); DTX019 at TORTOP00127212).) Table 1.2, titled Polymers Used for Aqueous-Based Coating as Solution or Dispersion, lists three “[t]ype[s]” of polymers: “Cellulose derivatives”; “Acrylic resin, poly(meth) acrylates”; and “Polyvinyl derivatives.” (DTX019 at TORTOP00127212.) Polyvinyl acetate is in the polyvinyl derivates category and not the acrylic category. (ECF No. 171 at 8.) Dr. Felton testified that this table is consistent with how a POSA would classify polymers. (Tr. 557:17-24.) But Dr. Felton also testified that she does not consider the three categories to be “definitional.” (Tr. 578:22-579:4.) Dr. Khan, for his part, testified that the table includes “[o]nly a small subset of acrylic polymer” and “exclude[s] a lot of acrylic polymers.” (Tr. 302:4-12.)

Second is chemical manufacturer BASF's technical information bulletin for Kollicoat SR 30 D, the exact brand of polyvinyl acetate that Torrent uses in its ANDA Products. (DTX020; see Tr. 493:19-555:25 (Rotella).) The bulletin refers to Kollicoat SR 30 D as a vinyl polymer. (ECF No. 171 at 9 (citing Tr. 294:12-296:7 (Khan); DTX020).) Dr. Rotella noted that the bulletin “connotes directly the structure of the polymer” and includes “no mention . . . of anything being related to an acrylic polymer” or acrylate derivatives. (Tr. 494:6-25; see DTX020 at TORTOP001126491 (depicting the structural formula).) Dr. Khan agreed that BASF did not describe polyvinyl acetate as an acrylic polymer, but he insisted that it “do[es]n't necessarily have to” describe it that way. (Tr. 295:19-24.)

Third is the Handbook of Pharmaceutical Excipients, a source that provides descriptions and properties of pharmaceutical excipients. (PTX113.3; PFF ¶ 232.)^[39] The parties agree that the Handbook is a “reliable source that is widely used by formulators.” (PFF ¶ 232 (citing Tr. 271:15-272:3 (Khan)); Tr. 298:25-299:3 (Khan).) The Handbook describes polyvinyl acetate dispersion as a vinyl polymer. (ECF No. 171 at 9 (citing Tr. 296:9-21, 297:6-10 (Khan); PTX113.3).) Confronted with this description, Dr. Khan testified that, like BASF's bulletin, the Handbook need not describe the polymer as within the “broad group” that is acrylic polymers, adding that the Handbook does not say that polyvinyl acetate is not an acrylic polymer. (Tr. 296:22-297:2.) And, as Dr. Khan noted, the Handbook cites as a specific reference BASF's technical information bulletin for Kollicoat SR 30 D. (Tr. 297:15-298:9; PTX113.4.) So, it makes sense that the Handbook and BASF's bulletin describe polyvinyl acetate similarly.

[39] Dr. Khan relied on the Handbook during his direct examination. (See Tr. 271:19-25 (Khan) (“This is a handbook that we routinely use, I think every formulator uses.”).)

Fourth is Remington: The Science and Practice of Pharmacy. (DTX207 at 932.) Remington notes that “[a]lternatives to the cellulose ethers are acrylic copolymers (eg, methacrylate and methyl methacrylate copolymers) and vinyl polymers (eg, polyvinyl alcohol).” (ECF No. 171 at 10 (citing DTX207 at 932).) Dr. Khan testified that this excerpt of Remington, like Table 1.2 in Aqueous Polymeric Coatings, is “just one description by one author,” is “not all inclusive,” and “doesn't encompass a lot of broad acrylic polymers that I used.” (Tr. 307:15-22.)

Fifth is the NIST Chemistry WebBook entry for Acetic acid ethenyl ester, which Dr. Khan referenced during direct examination. (PTX117; Tr. 308:6-15 (Khan).) The entry lists “[o]ther names” for the compound acetic acid ethenyl ester, including “acetic acid vinyl ester” and “vinyl acetate.” (PTX117.) Dr. Khan testified that although the NIST WebBook does not identify the compound as “acrylic,” it describes the compound in a way such that a pharmaceutical scientist would consider the compound acrylic. (Tr. 309:13-23.)^[40] Dr. Khan explained that he would not expect a chemistry book like the NIST Chemistry WebBook to use the specific name acrylate when describing vinyl compounds, because chemists view compounds more narrowly than do pharmaceutical scientists. (Tr. 310:6-16.) Still, Dr. Khan conceded that he did not know any source that uses the word acrylic as part of the chemical name for vinyl acetate; only “[t]he use effect.” (Tr. 310:17-20, 312:22-313:3.) This includes his own publications. (Tr. 312:22-313:3 (Khan).)

[40] Torrent attacks Dr. Khan's testimony, “For me it's an acrylic polymer,” as conveying his personal opinion rather than a POSA's understanding. (ECF No. 171 at 10 (quoting Tr. 308:6-20).) The Court disagrees. Dr. Khan made clear that he opined from a POSA's perspective.

Sixth is Polymer Science: A Textbook for Engineers and Technologists - specifically, section 2.5 of Chapter 2 titled Classification of Polymers. (DTX027 at TORTOP00126557.) In the subsection titled Acrylic Polymers, it notes that “[i]n this class, the polymers are derived from acrylic acid CH2=CH-CO-OH and methacrylic acid CH2=C(CH3)-CO-OH.” (DTX027 at TORTOP00126561; see Tr. 489:1-8 (Rotella).) In a separate subsection titled Polyvinyl Esters, it notes that “[p]olyvinyl acetate (PVA) belongs to this class, widely used in the form of aqueous emulsions for the manufacture of paints.” (DTX027 at TORTOP00126562; see Tr. 489:9-12 (Rotella).) Torrent's expert Dr. Rotella noted that Polymer Science's method of classifying polymers matches the way in which Remington and Aqueous Polymeric Coatings classify polymers: by the polymers' structure. (Tr. 489:13-17.) He added that he was “unaware of any textbook that illustrates that polyvinyl acetate is an example of an acrylic polymer.” (Tr. 489:18-19.)

Last is Hawley's Condensed Chemical Dictionary. (DTX033.) In Hawley's, the definition of acrylic polymers refers readers to acrylic resins, which “is another word for acrylic polymers,” according to Dr. Rotella. (Tr. 491:21-25; DTX033.) The definition also notes that acrylic resins include “acrylonitrile.” (Tr. 491:21-492:6.) Dr. Rotella testified that acrylonitrile “does not fit the empirical formula” of Dr. Khan's definition of acrylic polymer. (Tr. 492:3-6.) In addition, Hawley's defines acrylic acid using a structure. (Tr. 490:24-491:20; DTX033.) Dr. Rotella acknowledged, however, that Hawley's also refers to acrylonitrile as “vinyl cyanide,” though he added that “nobody calls it that” (PFF ¶ 238 (Pl.); Tr. 508:15-509:23).

Dr. Rotella's rebuttal opinion. Torrent's expert Dr. Rotella disagrees with Dr. Khan's use of “the general formula CnH2n-2O2” and the functional groups “olefin acids” to define acrylic polymers. Dr. Rotella opined that chemical compounds are defined by their structure, and that a POSA would not define a compound's structure based on its empirical formula or “functional groups that might be present in [the] structure.” (Tr. 464:16-24 (Rotella).) An empirical formula describes “the number and type of atoms that exist in a molecule” - for example, in Dr. Khan's definition, “how many carbon atoms there are, how many hydrogen atoms there are, how many oxygen atoms there are.” (Tr. 453:19-22, 456:2-18 (Rotella).) But according to Dr. Rotella, an empirical formula is “imprecise” and “confers no information about structure.” (Tr. 453:19-22, 456:2-18.) In fact, an empirical formula does not determine or describe chemical structure; “[t]he bonding of atoms to other atoms is what describes it.” (Tr. 537:24-538:8 (Rotella).) “The point here,” Dr. Rotella said, “is that structure is unique and empirical formula is not. . . . And, therefore, using empirical formula to help define structure is inappropriate, incomplete, and insufficient.” (Tr. 538:7-11; see Tr. 465:2 (Rotella) (“Structure is unique; empirical formula is not.”); Tr. 540:13-19 (Rotella) (“Empirical formula, in my opinion, as a person of skill in the art, is irrelevant. It simply is being employed here to define something. And that definition is, as I said, inappropriate, incomplete, and imprecise.”).)

Dr. Rotella opined that “vinyl acetate is not an acrylic derivative[,] because the structures are different.” (Tr. 471:22-23.) Indeed, Dr. Rotella testified, “vinyl polymers such as polyvinyl acetate polymers are characterized and classified differently” from acrylic polymers, “because they have different chemical structures.” (Tr. 484:15-18.) He testified that the two terms and their descriptions in Hawley's are “mutually exclusive.” (Tr. 493:7-10.) Acrylic polymers and vinyl polymers are “chemically and structurally distinct.” (Tr. 506:19-23.)

Dr. Rotella testified that molecules could have the same empirical formula but have different chemical structures.^[41] (Tr. 458; see Tr. 460:7-8 (Rotella) (“These two molecules have the same empirical formula but a different chemical structure.”).) To illustrate, Dr. Rotella testified that acrylic acid, for example, has an olefin, which is “also sometimes called a vinyl group.” (Tr. 460:21-24.) He then compared acrylic acid with another, unidentified molecule that has the same empirical formula, including the olefin, but has a different chemical structure. (Tr. 461:4-10.) Dr. Rotella's point was this: “[E]mpirical formula is not a determinant of structure. Structure determines structure, and how atoms are bonded to each other determines structure. Empirical formulas are not unique; structure is, and structure can be different.” (Tr. 460:9-14.)

[41] Dr. Rotella taught some chemistry fundamentals. “[M]olecules are made up of atoms. . . . And those atoms are connected to each other by bonds.” (Tr. 458:24-459:3.) “A polymer . . . is a large molecule with a collection of monomeric units. And those monomeric units can be the same or they can be different.” (Tr. 467:3-5.) “Polymerization is simply a general term for the chemical reaction by which monomers are connected to each other to make a polymer, and that's a chemical reaction that is, in many cases, very well established. It can be carried out by machines and it can be carried out by technicians according to recipes.” (Tr. 467:7-12.) A “carbonyl functional group” is written as “C=O.” (Tr. 468:10-15.) “[Y]ou need the elements of that . . . olefinic bond to make new bonds with other units in the structure,” and so when polymerization occurs, the olefin “disappears.” (Tr. 469:18-25 (cleaned up).)

Dr. Rotella then compared the polymerization of an acrylic derivative with that of a vinyl acetate. (Tr. 467.) He explained that in the demonstratives, R “represents a variable group in the structure” - “[i]t can be changed to other atoms.” (Tr. 470:8-11.) When R is hydrogen (H), the molecule is acrylic acid; when R is a “CH3 group,” the molecule is a “methyl group.” (Tr. 470:11-15.) Though the molecules are different, both “are acrylic derivatives because they are derived from acrylic acid or . . . structurally associated with acrylic acid.” (Tr. 470:16-20.) But “regardless of what this R is,” Dr. Rotella testified, a POSA would never “confuse an acrylic derivative . . . with vinyl acetate, because the structures are different” and “[t]hose differences in structure in the monomers are relayed faithfully into the polymers.” (Tr. 471:24-472:3.)

Dr. Rotella then discussed examples of acrylic monomers that derive from acrylic acid. The examples were acrylic esters, which are “a family of molecules”; methyl methacrylate; acrylamides, which are also “a family of molecules”; and acrylonitrile. (Tr. 475:9-476:25.) He testified that although acrylamides and acrylonitrile “contain a vinyl group” - that is, an olefin - and “are used in polymers,” these two molecules' formulas do not fit the empirical formula in Dr. Khan's acrylic polymers definition, because the molecules have a nitrogen atom and no oxygen atom. (Tr. 477:1-18, 478:1-4.) This illustrates, according to Dr. Rotella, “the imprecise, imperfect, and incomplete nature of using an empirical formula as a means for defining the term ‘acrylic polymers.'” (Tr. 477:15-18.)

Dr. Rotella also testified that not all acrylic polymers have the same structure. (Tr. 477:21-23.) And “depending on the specifics of structure,” Dr. Rotella continued, “these molecules can have distinct empirical formulas.” (Tr. 477:24-478:1.)

Having discussed the “acrylic polymer family,” Dr. Rotella shifted to describing vinyl acetate monomer and polymer units. (Tr. 478:5-9.) He explained that the vinyl acetate monomer's structure includes an olefin, which is where the “monomer undergoes polymerization . . . to create the polyvinyl acetate polymer.” (Tr. 478:21-24.) He testified that “this monomer and this polymer are structurally distinct from the acrylic monomers and acrylic polymers.” (Tr. 479:11-13.)

Dr. Rotella also described the structural differences between an acrylic derivative monomer and a vinyl acetate monomer. Starting with acrylic derivatives, Dr. Rotella testified that “by definition of acrylic derivative,” its structure must have an olefin unit (i.e., two carbon atoms double-bonded together, or “C=C”), with one of the double-bonded carbon atoms bonded to another carbon atom (“C-C”), with that other carbon atom bonded to either an oxygen or a nitrogen atom. (Tr. 480:13-481:13.) He testified that the “one tiny exception to this” is “acrylonitrile,” whose second of the single-bonded carbon atoms is triple-bonded to a nitrogen atom. (Tr. 481:14-18.)

Turning to vinyl acetate, Dr. Rotella testified that vinyl acetate's structure has the same olefinic unit, and that “a vinyl polymer is defined, first off, by this olefinic unit,” and “then there is another atom,” which can be an oxygen or a carbon atom. (Tr. 482:7-24.) “The acrylic derivative,” by contrast, “is defined by that olefin functional group, a carbon and then either an oxygen or a carbon.” (Tr. 482:24-483:1.) Dr. Rotella also testified, referencing a demonstrative exhibit, that “where these molecules are similar is in the olefin functional group . . . . [I]t's identical in both.” (Tr. 480:6-7.)

Asked to respond to “Dr. Khan's position that these two molecules” - the acrylic derivative and vinyl acetate - “are nearly identical or don't have widely differing structures,” Dr. Rotella testified as follows:

Well, you know, depending on your definition of widely different or not widely different you might say, well, okay, that's a possibility, but they are different. It doesn't matter if there's one atom or several atoms that are different. The fact that the basic units in the two structures are different makes these molecules different. And it prevents a person of skill in the art from confusing them. They are mutually exclusive.

[(Tr. 483:7-18.)]

Dr. Rotella then compared the chemical structures of polymethyl acrylate and polyvinyl acetate polymers that Dr. Khan had discussed:

(Image Omitted) [(PTX195 (polymethyl acrylate) (left); PTX196 (polyvinyl acetate) (right).)]

Dr. Rotella contended that the depiction of the polymethyl acrylate structure “could have been cherrypicked because it happens to have the same empirical formula as polyvinyl acetate.” (Tr. 484:1-5.) By “cherrypicked,” Dr. Rotella meant that both structures depicted the empirical formula of Dr. Khan's definition, though “a different acrylic polymer could have been picked” as the comparison. (Tr. 484:21-485:24.) But still, Dr. Rotella testified, “these two structures are different,” and despite their similarities, “a [POSA] would not confuse them[,] because they are structurally distinct.” (Tr. 484:8-12, 485:17-21.)

Dr. Rotella was then asked why these differences are important. He responded that “the differences in structure lead to differences in clarification and description of polymers,” adding that “polymers are often described in terms of structural classes, as was done in the patent, for example.” (Tr. 486:10-20.) Dr. Rotella testified that “the patent uses the terminology ‘cellulosic polymers' and ‘acrylic polymers.' And both of those terms refer to a wide family of structural units that are structurally distinct from each other.” (Tr. 486:21-24.) Dr. Rotella further testified that the term vinyl polymers “is a distinct structural class of molecules that have distinct chemical structures from either of the two claimed in the patents-in-suit here. And those structures are different. And those vinyl polymers are not -- are specifically not mentioned and not claimed in the claims in the patents-in-suit.” (Tr. 486:25-487:5.)

On cross-examination, some concerns were raised about the reliability of Dr. Rotella's opinions. As to his experience, Dr. Rotella has collaborated with people “who have designed [sustained-release] formulations,” but he has never designed one himself or developed a drug that received FDA approval. (Tr. 497:8-498:7.) In his work from 2006 and earlier, he had not used the Handbook of Pharmaceutical Excipients, because he was not involved with carrying out formulation research. (Tr. 498:24-499:13.) He has no personal experience developing a product using polyvinyl acetate or polymethacrylate. (Tr. 499:14-21.) And he has no “direct experience with” or “personal knowledge about whether poly methacrylate” or polyvinyl acetate “can be used as a coating material for release-controlling coatings and in an extended-release formulation of a drug,” which Dr. Rotella testified is “outside the scope of my expertise.” (Tr. 501:2-17.)

Another concern stems from two errors that Dr. Rotella made in his expert report. First, Dr. Rotella acknowledged that just two weeks before trial, he submitted an errata sheet to his expert report, revising the name “hydroxypropyl methyl cellulose” to “methyl cellulose,” an error that he chalked up to a typo. (Tr. 502:13-18.) But when confronted with the error, Dr. Rotella testified that his errata sheet was incorrect - “I made a mistake in correcting a mistake.” (Tr. 505: 2-506:17.) Second, in the report's sentence “A direct comparison of the structure of polyvinyl acetate and polyvinyl acrylate polymers are shown below,” Dr. Rotella erroneously wrote “polyvinyl acrylate polymers” instead of “acrylate polymers.” (Tr. 520:17-521:8.)

Perhaps most important, Dr. Rotella conceded that there are circumstances where “acrylic compounds” and “vinyl compounds” are not mutually exclusive. (Tr. 517:12-15.) The concession came from a series of questions about the book Polymer Chemistry, on which Dr. Rotella relied in his expert report. (PTX192; Tr. 512:20-22.) Figure 1.4 of Polymer Chemistry shows the “polymerization of an olefinically unsaturated compound.”

(Image Omitted) [(PTX192.17.)] Dr. Rotella first acknowledged that both the structure of the unit depicted in Figure 1.4 and the structures of the polyacrylonitrile, polyvinyl acetate, polymethyl acrylate units depicted, respectively, in PTX198 share a “similar functional group”: the olefin, which “is necessary for the polymerization reaction.” (Tr. 515:23-516:13.)

(Image Omitted) [(PTX198 (from left to right: polyacrylonitrile, polyvinyl acetate, polymethyl acrylate); see also Tr. 510:4-511:14 (Rotella) (identifying the structures).)]

Supernus then confronted Dr. Rotella with this excerpt from Polymer Chemistry: “Vinyl compounds can often be polymerized by a chain-growth mechanism. Here, the double bond is converted into two single bonds (see Fig. 1.4).” (PTX192.16; Tr. 516:20-24.) Dr. Rotella testified that the book uses the term vinyl to refer to the olefinic unit that is common among the structures of the polymers depicted in PTX198. (Tr. 516:25-517:6.) Still, Dr. Rotella testified, a POSA “recognizes that there is variability in that R and that the polymers derived therefrom can be characterized and should be characterized differently because they are structurally distinct, depending on what you're comparing.” (Tr. 517:7-11.) But ultimately, Dr. Rotella testified that “acrylic compounds contain a vinyl unit. They contain an olefinic unit.” (Tr. 517:12-15.) Dr. Rotella nevertheless contested that the existence of a vinyl unit dictates how a POSA characterizes polymers. (Tr. 517:17-19.)^[42]

[42] Torrent offered a cleaned-up version of this testimony in its proposed findings of fact: “References may use bond nomenclature [to] describe a molecule, but again, this nomenclature does not mean that anything with a vinyl bond, for instance, is automatically a vinyl polymer.” (PFF ¶ 243 (Def.) (citing Tr. 517:2-6 (Rotella)).) At closing arguments, Supernus argued that “every acrylic polymer has a vinyl backbone. . . . So the acrylic polymers are vinyl polymers” (Tr. 737:16-18); Torrent countered that “just because a molecule has a vinyl bond doesn't make it a vinyl polymer” (Tr. 748:19-20).

In the same light, Supernus confronted Dr. Rotella with a 2004 article by D. Jones titled Pharmaceutical Applications of Polymers for Drug Delivery - excerpts of which were read into the record.^[43] (Tr. 521:17-19, 522:23-24.) In section 1.2 titled Examples of Pharmaceutical Polymers, it states, “As described previously, there are a wide range of polymers that are used as components of pharmaceutical formulations.” (Tr. 521:23-522:22.) Underneath section 1.2 is subheading 1.2.1 titled vinyl polymers. (Tr. 523:2-5.) And underneath that section is subheading 1.2.1.1 titled poly methacrylates. (Tr. 523:6-8.) Based on this limited review of the article, Dr. Rotella surmised that the author “characterize[d] polymers based on the existence of a vinyl group, as we've seen in others.” (Tr. 523:16-18.) He reasoned that “the characterization here goes from the so-called vinyl polymers, that is, polymers that contain an olefin, to cellulose polymers, and it breaks those down structurally as well.” (Tr. 523:18-21.) “And so,” Dr. Rotella concluded, “I think what he's referring to here is -- if I look at this in context -- is the functional unit in a molecule that undergoes the polymerization reaction, not in terms of structural characterization which he does differently.” (Tr. 523:22-25.)

[43] Under Federal Rule of Evidence 803(18)(B), a “statement contained in a treatise, periodical, or pamphlet” is not hearsay if “(A) the statement is called to the attention of an expert witness on cross-examination or relied on by the expert on direct examination; and (B) the publication is established as a reliable authority by the expert's admission or testimony, by another expert's testimony, or by judicial notice.” The rule permits admissible statements to “be read into evidence but not received as an exhibit.” Fed.R.Evid. 803(18). Supernus used the D. Jones article on cross-examination. Although Dr. Rotella had never seen the article before, Torrent did not object to Supernus's reading it into the record.

Also underneath the vinyl polymers subsection is subheading 1.2.1.2 titled polyvinyl alcohol. (Tr. 524:13-21.) Dr. Rotella testified that polyvinyl alcohol is “structurally distinct from polyvinyl acetate because . . . the polymer contains a different structure.” (Tr. 524:22-525:1.)

The next line of questions fielded a well-made objection. While cross-examining Dr. Rotella, Supernus tried to run a Google search of “popular types of acrylic polymers.” Supernus anticipated that the search would return a link to a chemical manufacturer's website showing that “at least some people who are in the chemical arts would understand that polyvinyl acetate is an acrylic polymer.” (Tr. 526:17-527:2.) Supernus's idea was to contradict Dr. Rotella's opinion that a POSA “would not confuse and not mistake polyvinyl acetate for an acrylate polymer because they are structurally distinct.” (Tr. 526:9-15.) Torrent argued that this was not a proper basis for impeachment and challenged the search results' authenticity and evidentiary value. (Tr. 527:7-11.) Supernus responded that the search would rebut Dr. Rotella's opinion that “no one or no POSA would consider polyvinyl acetate as being a kind of acrylic polymer.” (Tr. 528:17-25.)

Impeachment by contradiction is a permissible theory of impeachment. Primus v. Target Corp., 532 Fed.Appx. 314, 315 (3d Cir. 2013) (citing United States v. Gilmore, 553 F.3d 266, 271 (3d Cir. 2009); United States v. Kincaid-Chauncey, 556 F.3d 923, 932 n.8 (9th Cir. 2009)). But here, the Internet marketing materials, which Supernus offers for their substantive truth, are hearsay. For the Court to admit them, they must meet a hearsay exception. See Pernix Ireland Pain Dac v. Alvogen Malta Operations Ltd., 316 F.Supp.3d 816, 826 (D. Del. 2018) (“An effort to impeach a witness with an out-of-court statement made by another person necessarily relies on the truth of the out-of-court statement and, consequently, requires a hearsay exception for admission.” (quoting 30B Charles Alan Wright, et al., Fed. Prac. & Proc. Evid. § 6728 (2018 ed.))).^[44] Supernus cited no hearsay exception. More important, as Torrent correctly asserted, “people on the internet [are] not a [POSA].” (Tr. 529:4-5.) As a result, the Internet materials do not in fact contradict Dr. Rotella's testimony about a POSA's understanding. See Primus, 532 Fed.Appx. at 315 (“Absent an actual contradiction, the evidence could not impeach by contradiction.”). The Court therefore finds that the Internet materials are unreliable and inadmissible, and it will not consider the related testimony in its decision.

[44] C.f. Huzinec v. Six Flags Great Adventure, LLC, 2023 WL 1433633, at *3 (3d Cir. Feb. 1, 2023) (ruling that YouTube videos offered to prove that Six Flags was on notice of a danger, not the truth of the matter asserted, were not hearsay); Express Homebuyers USA, LLC v. WBH Mktg. Inc., 323 F.Supp.3d 784, 789 n.3 (E.D. Va. 2018) (finding that results of the Internet search “we buy houses” were not hearsay where they were offered to show not that “companies advertising home-buying services in fact buy houses” but that “many third parties currently use and have used the ‘we buy houses' phrase”).

Finally, when confronted with the claim limitations at issue, Dr. Rotella agreed that “for any structure to fit within the claim language, it would need to be able to form a polymer.” (Tr. 539:17-20 (referencing DTX170 ¶ 45).) He acknowledged that some of the molecules that he used to demonstrate the notion that molecules can have the same empirical formula but different structures “may not be able to form polymers.” (Tr. 539:22-540:11; see Tr. 466:3-25.)^[45]

[45] At trial, Supernus objected to Torrent's use of DDX107, which depicted chemical structures that were not disclosed in Dr. Rotella's report. (See Tr. 421-435.) Because neither DDX107 nor Dr. Rotella's related direct-examination testimony meaningfully influenced the Court's decision, the Court need not resolve the objection.

The Court's conclusions. The Court finds that Supernus has not shown by a preponderance of the evidence that a POSA would understand the plain and ordinary meaning of acrylic polymers to include polyvinyl acetate.

Supernus contends that the term acrylic polymers needs no construction, as Dr. Khan testified, “because a POSA knows what the polymer is,” “knows what function it plays,” and “uses [them] all the time.” According to Supernus, pharmaceutical scientists focus on a polymer's function, not on its structure. (PFF ¶ 55 (Pl).)^[46] As Dr. Khan sees it, if a pharmaceutical scientist knows that a polymer functions like an acrylic polymer, the scientist need not consult a definition to know that it is an acrylic polymer. By Dr. Khan's analogy, people who have seen an elephant know what an elephant is; his definition was purposed to “describe an elephant to people who haven't seen” one before. (Tr. 322:2-12.)

[46] (See Tr. 314:10-12 (Khan) (“So within acrylic, there is vinyl set of polymers that are separated out from acrylic, but they work the same way. They work as acrylic polymer.”).)

The problem for Supernus is that a purely function-based description - such as “a release controlling coating” (PTX233.210) - is something the PTO examiner rejected during the patent prosecution:

Defining . . . release coatings in terms of functional features or elements does not suffice in the absence of a disclosure of structure features or elements of the . . . release coatings which would have the stated function.

Applicants are describing what . . . release coatings do rather than what they are. Describing a compound by its functions [(release coatings)] will not substitute for written description of the structure of the compound. Describing the function of a compound fails to distinguish the compound from other molecules or agents that can perform the same functions.

[(PTX233.230.)]

Hoping to solve that problem, Dr. Khan crafted a definition that he believed both covered the acrylic polymers in the Patents-in-Suit and included a structure feature: the olefin acid. (Tr. 267:8-268:22 (Khan).) Indeed, Dr. Rotella also referred to an olefin as a “functional group” and to an acid as “a different functional group.” (Tr. 455:16-19, 480:5-7, 482:22-25, 516:6-13 (Rotella).)

But the Court is not satisfied that a POSA would more likely than not read the specification and arrive at Dr. Khan's exact definition. The Court expects that if a pharmaceutical scientist viewed the olefin acid as the structure feature that, coupled with the empirical formula, defined a polymer as acrylic, vinyl, or both, that view might oft appear in technical literature. Yet Dr. Khan's definition is not in any of the technical literature in evidence. Even in Webster's, the source from which Dr. Khan drew his definition, the acrylic definition needed Dr. Khan's modification to cover the claimed acrylic polymers.^[47] And, as Torrent established through cross-examination, Dr. Khan is “not aware of any references that discuss or describe vinyl acetate as an acrylic polymer.” (Tr. 314:22-24.)

[47] The Court rejects Torrent's argument that general-use dictionaries cannot be used to define scientific terms. The Federal Circuit has cited similar dictionaries for definitions of chemical compounds. See, e.g., Ultimax Cement Mfg. Corp. v. CTS Cement Mfg. Corp., 587 F.3d 1339, 1348 (Fed. Cir. 2009) (noting the definition of anhydrite from Webster's Third New International Dictionary of the English Language Unabridged (1986)). General-use dictionaries are unhelpful when a claim term has a different meaning in the art than it has in common, lay usage. See Vanderlande Indus. Nederland BV v. I.T.C., 366 F.3d 1311, 1321 (Fed. Cir. 2004) (“[A] general-usage dictionary cannot overcome credible art-specific evidence of the meaning or lack of meaning of a claim term.”). Torrent has not argued that acrylic polymers means something different in the drug-delivery context than in lay usage.

On the other hand, the record supports that no all-encompassing definition of acrylic polymers may exist. That explains the, as Torrent put it, “Frankenstein” nature of Dr. Khan's definition. But record evidence also shows that acrylic polymers and vinyl polymers are not always mutually exclusive. For instance, Hawley's refers to acrylonitrile, an acrylic polymer, as “vinyl cyanide,” showing that acrylic polymers can be described as vinyl polymers. For another, Dr. Rotella acknowledged that the 2004 D. Jones article characterizes polymers with vinyl groups as vinyl polymers. The article supports Dr. Khan's opinion that a pharmaceutical scientist prioritizes a polymer's function and thus focuses on its functional group - that is, the olefin, also called the vinyl group.

In addition, Dr. Rotella offered nothing but his testimony to support his “definitions” of vinyl polymers and acrylic polymers, which were vague and overly sensitive to structure:

DR. ROTELLA. And a vinyl polymer is defined, first off, by this olefinic unit. And then there is another atom here in this example. It happens to be oxygen. But in other examples it can be a carbon atom, but that carbon atom then is bonded in different ways to other carbon atoms present in this structure. But in this specific example, so vinyl acetate is defined by the oxygen atom and the olefin functional group here. The acrylic derivative is defined by that olefin functional group, a carbon and then either an oxygen or a carbon. And those are different. And that's why the term “acrylic” means something. That's why the term “vinyl” means something. Those terms confer structure. And the word “acrylic” confers the structure as I have just outlined it for you and illustrated it for you.

[(Tr. 482:17-483:5.)]

The Court does not find that a POSA would view structural differences as narrowly as Dr. Rotella views them. Dr. Rotella's testimony highlights his sensitivity to even slight differences between structures. He testified that polyvinyl alcohol and polyvinyl acetate are “structurally distinct” because they contain different structures. He also testified that “[i]t doesn't matter if there's one atom or several atoms that are different. The fact that the basic units in the two structures are different makes these molecules different. And it prevents a person of skill in the art from confusing them. They are mutually exclusive.” (Tr. 483:7-18.)

In addition, as knowledgeable as Dr. Rotella is about organic chemistry and drug discovery, he lacks experience in drug formulation and delivery. Yet Dr. Felton, an expert in the field of drug formulation and modified drug delivery systems (Tr. 553:10-17 (Felton)), did not provide testimony corroborating Dr. Rotella's narrow view of classifying polymers. On the topic, Torrent established only her opinion that Aqueous Polymeric Coatings's depicting cellulosic, acrylic, and vinyl polymers separately is consistent with how formulators classify polymers. (Tr. 557:17-21.) Even so, Dr. Felton testified that the depiction was not definitional. Beyond that, her infringement testimony concerned only the doctrine of equivalents.

In sum, the Court finds that there may be more than one way to classify a polymer, and that there are times when acrylic polymers and vinyl polymers may overlap. That said, Supernus has not proven by a preponderance of the evidence that a POSA would apply Dr. Khan's definition to conclude that polyvinyl acetate falls within the plain and ordinary meaning of acrylic polymers. The Court therefore rules that Torrent's ANDA Products do not literally infringe the Asserted Claims.

B. Doctrine of Equivalents

Next, the Court decides whether Supernus proved by a preponderance of the evidence that Torrent's ANDA Products infringe the acrylic polymers limitation under the doctrine of equivalents. “Under [the doctrine of equivalents], a product or process that does not literally infringe upon the express terms of a patent claim may nonetheless be found to infringe if there is ‘equivalence' between the elements of the accused product or process and the claimed elements of the patented invention.” Kyocera Senco Indus. Tools Inc. v. Int'l Trade Comm'n, 22 F.4th 1369, 1375 (Fed. Cir. 2022) (quoting Warner-Jenkinson Co. v. Hilton Davis Chem. Co., 520 U.S. 17, 21 (1997)).

But first, Supernus must prove by a preponderance of the evidence that its prosecution history does not estop it from asserting infringement by equivalence.

1. Prosecution History Estoppel

“Prosecution history estoppel applies as part of an infringement analysis to prevent a patentee from using the doctrine of equivalents to recapture subject matter surrendered from the literal scope of a claim during prosecution.” Pharma Tech Sols., Inc. v. LifeScan, Inc., 942 F.3d 1372, 1380 (Fed. Cir. 2019) (quoting Trading Techs. Int'l, Inc. v. Open E Cry, LLC, 728 F.3d 1309, 1322 (Fed. Cir. 2013)). “Prosecution history estoppel can occur in two ways: ‘either (1) by making a narrowing amendment to the claim (“amendment-based estoppel”) or (2) by surrendering claim scope through argument to the patent examiner (“argument-based estoppel”).'” Id. (quoting Conoco, Inc. v. Energy & Env't Int'l, L.C., 460 F.3d 1349, 1363 (Fed. Cir. 2006)).

An amendment is narrowing if it is necessary to satisfy a requirement of the Patent Act and secure the patent. Festo Corp. v. Shoketsu Kinzoku Kogyo Kabushiki Co., 535 U.S. 722, 736-37 (2002). Indeed, “[a] patentee who narrows a claim as a condition for obtaining a patent disavows [its] claim to the broader subject matter, whether the amendment was made to avoid the prior art or to comply with § 112.” Id. at 737. In contrast, “[i]f a § 112 amendment is truly cosmetic, then it would not narrow the patent's scope or raise an estoppel.” Id. at 736-37. A “patentee's decision to narrow [its] claims through amendment may be presumed to be a general disclaimer of the territory between the original claim and the amended claim.” Pharma Tech Sols., 942 F.3d at 1380 (quoting Festo, 535 U.S. at 725). This is often called the “Festo presumption.”

Torrent invokes amendment-based estoppel. It argues that during the prosecution of the '580 patent,^[48] the applicants made a narrowing amendment surrendering polyvinyl acetate as a release controlling polymer. (ECF No. 171 at 24-27; see PTX233.1 ('931 application).) The Court must therefore analyze the prosecution history.

[48] The '580 patent was prosecuted under U.S. Patent Application No. 12/926,931 (“the '931 application”). (PTX233.1.) In the Markman Opinion, the Court found that “[t]he parties also agree that, because the patents in suit are part of the same patent ‘family,' the ‘prosecution history of related patents is relevant to all patents in that family.'” (ECF No. 88 at 8 (citations omitted).) Thus, the '580 patent prosecution history is relevant to the Patents-in-Suit.

An earlier version of subsection (a) of claim 104 in the '931 application read “(a) at least two different extended release topiramate-containing components, wherein each component comprises a release controlling coating specific for its component. . . .” (PTX233.210.)

In an August 8, 2012 Office Action, the examiner rejected claim 104 of the '931 application for “failing to comply with the written description requirement” under § 112. (PTX233.229.) The examiner reasoned as follows:

The disclosure of a species may provide an adequate written description of a genus when the species disclosed is representative of the genus. The present claims encompass any and all enhancing agents, complexing agents and release coatings that can exist. There is substantial variability among the species of enhancing agents, complexing agents and release coatings encompassed within the scope of the claims because the enhancing agents, complexing agents and release coatings have widely differing structures and

corresponding biological activities. Defining the enhancing agents, complexing agents and release coatings in terms of functional features or elements does not suffice in the absence of a disclosure of structure features or elements of the enhancing agents complexing agents and release coatings which would have the stated function.

Applicants are describing what the enhancing agents, complexing agents and release coatings do rather than what they are. Describing a compound by its functions (enhancing agents, complexing agents and release coatings) will not substitute for written description of the structure of the compound. Describing the function of a compound fails to distinguish the compound from other molecules or agents that can perform the same functions.

A description of a genus may be achieved by means of a recitation of a representative number of species falling within the scope of the genus or of a recitation of structural features common to the members of the genus, which features constitute a substantial portion of the genus. Regents of the University of California v Eli Lilly Co., 119 F.3d 1559, 1569, 43 U.S.P.Q.2d 1398, 1406 (Fed Cir 1997). Consequently, the Examiner notes that the claimed invention which is drawn to a genus of enhancing agents, complexing agents and release coatings may be adequately described if there is a (1) sufficient description of a representative number of species, or (2) by disclosure of relevant, identifying characteristics sufficient to describe the claimed invention in such full, clear, concise and exact terms that a skilled artisan would recognize applicant was in possession of the claimed invention. Here, the specification discloses only a few common structural features shared by the members of the claimed genus. Since the claimed genus encompasses compounds yet to be discovered, the disclosed structural feature does not constitute a substantial portion of the claimed genus.

Therefore, the disclosure of several examples of enhancing agents, complexing agents and release coatings and others listed in the specification, does not provide an adequate description of the claimed genus of enhancing agents, complexing agents and release coatings because the claims are broader than the disclosure.

[(PTX233.230-231.)]

In response, the applicants amended subsection (a) of claim 104, adding this underlined clause: “(a) at least two different extended release topiramate-containing components, wherein each component comprises a release controlling coating specific for its component and comprising a coating material selected from the group consisting of cellulosic polymers and acrylic polymers. . . .” (PTX233.242.) In their remarks, the applicants first acknowledged that claim 104 “stand[s] rejected for failing to comply with the written description requirement” - “[i]n particular, . . . that the specification fails to adequately describe . . . ‘release controlling coating'.” The applicant then stated that “[a]lthough Applicants respectfully disagree, solely in the interest of compact prosecution, Applicants have amended the claims to now recite species of the noted genera.” (PTX233.249.)

Supernus does not deny that the amendment was narrowing for patentability reasons. Indeed, the examiner rejected the original claim for written-description shortcomings. The responsive amendment was purposed to address the examiner's concerns about compliance with § 112. And given the examiner's note that the original claim encompassed “any and all . . . release coatings that can exist,” including “compounds yet to be discovered” (PTX233.230-231), the original claim covered polyvinyl acetate. C.f. Intendis GMBH v. Glenmark Pharms. Inc., USA, 822 F.3d 1355, 1365-66 (Fed. Cir. 2016) (holding that where the claims as originally written could not have included the accused equivalent, the amendment was made for clarification, not patentability). For these reasons, the Court finds that there is a presumption of prosecution history estoppel.^[49]

[49] In support of estoppel, Torrent asserts that “Supernus chose to use the closed transition phrase ‘consisting of' in its amendment.” (ECF No. 171 at 26.) Because the Court finds that the Festo presumption applies regardless of this argument, the Court need not analyze it.

2. Tangential Relation

The issue, then, is whether Supernus has rebutted the presumption. There are three ways to overcome the presumption: “(1) the rationale underlying the amendment bears no more than a tangential relation to the equivalent in question; (2) the equivalent was unforeseeable at the time of the application; or (3) there was some other reason suggesting that the patentee could not reasonably be expected to have described the equivalent.” Bio-Rad Lab'ys, Inc. v. 10X Genomics Inc., 967 F.3d 1353, 1364 (Fed. Cir. 2020) (citing Festo, 535 U.S. at 740-41). The standard to rebut the presumption is a preponderance of the evidence. Integrated Tech. Corp. v. Rudolph Techs., Inc., 734 F.3d 1352, 1359 (Fed. Cir. 2013).

Supernus invokes only the “tangential relation” exception: that “the rationale underlying the amendment . . . bore no more than a tangential relationship to the alleged equivalent (i.e., polyvinyl acetate).” (ECF No. 170 at 33.) “The tangential relation inquiry ‘focuses on the patentee's objectively apparent reason for the narrowing amendment,' which ‘should be discernible from the prosecution history record.'” Pharma Tech Sols., 942 F.3d at 1380 (quoting Integrated Tech., 734 F.3d at 1358). Courts considering the tangential relation exception ask “whether the reason for the narrowing amendment was peripheral, or not directly relevant, to the alleged equivalent.” Ajinomoto Co. v. Int'l Trade Comm'n, 932 F.3d 1342, 1354 (Fed. Cir. 2019) (quoting Festo Corp. v. Shoketsu Kinzoku Kogyo Kabushiki Co., 344 F.3d 1359, 1369 (Fed. Cir. 2003)).^[50]

[50] Tangential means “touching lightly or in the most tenuous way.” Eli Lilly, 933 F.3d at 1331 (quoting Webster's Third New International Dictionary (2002)).

Supernus argues that “the objectively apparent reason for the amendment was to alleviate the Patent Office's explicitly stated concern that the claims, as then written, encompassed coatings that ‘have widely differing structures and corresponding biological activities.'” (ECF No. 170 at 34.) Polyvinyl acetate, Supernus says, “has the same biological activity and a structure that is nearly identical to the structure of the exemplary acrylic polymers listed in the patent specification.” (Id.)

Torrent counters that “the clear purpose of Supernus's amendment was to overcome a § 112 rejection by claiming specific polymers (cellulosic and acrylic polymers) and surrendering other polymers (vinyl polymers).” (ECF No. 171 at 28.) Torrent asserts that Supernus puts all of its weight behind one sentence from the prosecution history: that “[t]here is substantial variability among the species of enhancing agents, complexing agents and release coatings encompassed within the scope of the claims because the enhancing agents, complexing agents and release coatings have widely differing structures and corresponding biological activities.” (Id.)

The Court must resolve whether the objectively apparent reason for Supernus's amendment narrowing the release coatings to cellulosic polymers and acrylic polymers was more than tangentially related to polyvinyl acetate.

“[W]hether the patentee has established a merely tangential reason for a narrowing amendment is [a question of law] for the court to determine from the prosecution history record without the introduction of additional evidence, except, when necessary, testimony from those skilled in the art as to the interpretation of that record.” Festo, 344 F.3d at 1370. “[T]he reason for an amendment, where the tangential exception is invoked, cannot be determined without reference to the context in which it was made.” Eli Lilly, 933 F.3d at 1332. The Federal Circuit has described this analysis as “case-specific,” cautioning that “analogies to other cases [are] less helpful than a direct consideration of the specific record . . . and what it shows about the reason for amendment and the relation of that reason to the asserted equivalent.” Id. at 1332 n.5. Indeed, prosecution history estoppel has not applied “in similar circumstances, where the prosecution record differed.” Id. at 1333; see Intervet Inc. v. Merial Ltd., 617 F.3d 1282, 1291 (Fed. Cir. 2010) (“[T]here is no hard-and-fast test for what is and what is not a tangential relation.”).

That said, it is helpful to understand how the Federal Circuit has addressed the tangential exception in comparable circumstances.

In Eli Lilly & Co. v. Hospira, Inc., the patent owner narrowed claims during prosecution to recite “pemetrexed disodium” instead of “an antifolate.” 933 F.3d at 1325-26, 1330-31. The accused equivalent was pemetrexed ditromethamine, which is functionally identical to pemetrexed disodium. Id. at 1327. The Federal Circuit concluded that “[t]he reason for Lilly's amendment . . . was to narrow original claim 2 to avoid Arsenyan, which only discloses treatments using methotrexate, a different antifolate,” and “not to cede other, functionally identical pemetrexed, salts.” Id. at 1331. Thus, claiming the functionally equivalent pemetrexed salts was only tangential to overcoming prior art disclosing an antifolate other than pemetrexed.

There is more to take from the Eli Lilly decision. First, the Federal Circuit does not “demand perfection from patent prosecutors.” Id. at 1332. Even “inartful” amendments may be saved depending on the context. See Id. (“The prosecution record implies that Lilly's amendment, inartful though it might have been, was prudential in nature and did not need or intend to cede other pemetrexed salts.”).

Second, a patentee invoking the tangential exception need not “prove that it could not have drafted a claim that literally encompassed” the accused equivalent. Id. at 1332. The patentee must show that the accused equivalent was “‘peripheral, or not directly relevant,' to its amendment.” Id. (quoting Festo, 344 F.3d at 1369). And “the reason for an amendment . . . cannot be determined without reference to the context in which it was made.” Id.

Finally, the Federal Circuit rejected the “bright-line rule” that “where the reason for the amendment and the equivalent in question both relate to the same claim element, the tangential exception does not apply.” Id. at 1333. The unanimous panel made clear that “an amendment that narrows an existing claim element” often “evinces an intention to relinquish that claim scope” - but that “this consideration is not dispositive.” Id. (citing Honeywell Int'l, Inc. v. Hamilton Sundstrand Corp., 523 F.3d 1304, 1315-16 (Fed. Cir. 2008), as an example of where the intention to relinquish was so evinced)); see Honeywell, 523 F.3d at 1322 (Newman, J., dissenting) (“It does not suffice to say that by narrowing a claim by adding an element, ipso facto the relation to the accused equivalent element is not tangential. This criterion relates to why an amendment was made; it does not become irrebuttable simply when the accused equivalent concerns the same element that was added by amendment.”). Even though the narrowing amendment and the accused equivalent both related to the particular type of salt in the invention, the Federal Circuit held that “the particular type of salt to which pemetrexed is complexed relates only tenuously to the reason for the narrowing amendment, which was to avoid” the prior art. Eli Lilly, 933 F.3d at 1331.

By contrast, in Amgen Inc. v. Amneal Pharmaceuticals LLC, the Federal Circuit rejected use of the tangential exception. 945 F.3d 1368 (Fed. Cir. 2020). There, the accused product used “pregelatinized starch” as a binder, yet the asserted claim did not list pregelatinized starch in its Markush group reciting binders. Id. at 1380. In the prosecution history, the patent owner had revised the claim's binder limitations to be in Markush group format to overcome prior art references that taught the use of pregelatinized starch as a binder. Id. at 1382. Thus, the Federal Circuit concluded that the amendment - made to avoid prior art that contains the accused equivalent - was more than tangential to the accused equivalent. Id. Because the prior art references taught the use of the alleged equivalent for the claimed function, the tangential exception to prosecution history estoppel could not apply. Id.

In Bio-Rad Laboratories, Inc. v. 10X Genomics Inc., the inventors amending claims specified that their invention uses “non-fluorinated microchannels” that are chemically distinct from “fluorinated surfactant” carrier fluid so that “the carrier fluid and the microchannels in the claimed invention would not react with each other.” 967 F.3d at 1365. “In amending the claims,” the Federal Circuit noted, “the patentees sought to distinguish the claimed invention from [the prior art], which disclosed fluorinated microchannel wall coatings that would react with the carrier fluid.” Id. The accused equivalent was microchannels containing a nominal amount of fluorine that is “not chemically distinct from the carrier fluid” but still “cannot react with the carrier fluid.” Id. The Federal Circuit concluded that “the inventors surrendered microchannels coated with fluorine ‘for a purpose-not those containing de minimis amounts of fluorine that have no effect on how the microchannel functions in the system.'” Id. Thus, the narrowing amendment had only a tangential relation to the equivalent - “negligibly fluorinated microchannels, or, put differently, microchannels with non-fluorinated properties.” Id. at 1366.

In reaching that decision, the Federal Circuit in Bio-Rad measured its facts against those in Amgen and Eli Lilly. In Amgen, the amendment was more than tangential, because it was “made to avoid prior art that contains the equivalent in question”; in Bio-Rad, by contrast, the prior art “did not teach the use of the alleged equivalent-negligibly fluorinated microchannels or those with no fluorinated properties.” Bio-Rad, 967 F.3d at 1366 (citing Amgen, 945 F.3d at 1382). And in Eli Lilly, the amendment was tangential, in part because it was made to avoid prior art, which did not disclose the accused equivalent; likewise, in Bio-Rad, the prior art disclosed fluorinated microchannels and not the accused equivalent “microchannels with no fluorinated properties.” Bio-Rad, 967 F.3d at 1366 (citing Eli Lilly, 933 F.3d at 1325-27, 1331).

In this case, no one disputes that the applicants' remarks show that the amendment was made to address the examiner's concerns. (See ECF No. 170 at 35 (“[T]he applicants amended claim 104 of the '931 Application to address the Examiner's concerns.”); ECF No. 171 at 29 (“Supernus specifically told the USPTO that the amendment was made ‘in the interest of compact prosecution' to address the Examiner's concerns.” (quoting PTX233.249)) (cleaned up); Tr. 563:12-16 (Felton) (“It says that the applicants have amended the claims in a manner that ought to address the examiner's concerns.”).) Torrent argues that the Court, in discerning the rationale for the amendment, should focus not on the examiner's reasons for its rejection, but on the applicants' remarks. (ECF No. 171 at 28-29.) Based on those remarks, Torrent asserts, the objectively apparent reason for the amendment could not have been to respond to the examiner's widely differing structures and corresponding biological activities comment, because the applicants did not specifically identify that comment in their response. In essence, Torrent proposes that the Court focus on what the applicants did not say rather than what they responded to.

The Court cannot do so. Again, the Court's inquiry “focuses on the patentee's objectively apparent reason for the narrowing amendment” based on “the prosecution history record.” Pharma Tech Sols., 942 F.3d at 1380 (quoting Integrated Tech., 734 F.3d at 1358). This focus does not make the tangential exception unavailable just because the applicants did not neatly recite a particular sentence from the examiner's opinion, despite what Torrent would prefer. Indeed, divorcing the examiner's reasons from the applicants' response would disregard “the context in which [the amendment] was made.” Eli Lilly, 933 F.3d at 1332; accord Malvern Panalytical Inc. v. TA Instruments-Waters LLC, 2023 WL 7171484, at *8 (Fed. Cir. Nov. 1, 2023) (considering how “a reasonable reader” could interpret the prosecution history (citing Ecolab, Inc. v. FMC Corp., 569 F.3d 1335, 1343 (Fed. Cir. 2009))).

In context, the prosecution history shows that the applicants amended the claim to address the examiner's concerns that “the scope of the claims” encompassed “any and all . . . release coatings that can exist” or have “yet to be discovered,” including coatings with “widely differing structures and corresponding biological activities.” (PTX233.230-231.) The examiner explained that by “describing the function of a compound” - here, release control - the original claims “fail[ed] to distinguish the compound from other molecules or agents that can perform the same functions.” (PTX233.230.) The examiner advised that the applicants had to disclose “structure features or elements of the . . . release coatings which would have the stated function.” (PTX233.230.) The examiner noted that the specification disclosed “several examples of . . . release coatings,” but that disclosure did “not provide an adequate description of the claimed genus of . . . release coatings because the claims” - which then lacked the species of genus disclosed in the specification - “are broader than the disclosure.” (PTX233.231.) In response, the applicants amended the claim to include the “species of the noted genera” that were disclosed in the specification. (PTX233.249.)

Dr. Felton testified that the objectively apparent reason was to address the examiner's concerns that the language “was just too broad.” (Tr. 563:15-22.) But Dr. Felton's opinion ignores the context of the examiner's rejection. The spirit of the rejection is captured by the examiner's position that the claim as originally written included “widely differing structures and corresponding biological activities.” Hence why Dr. Khan testified that a POSA would interpret the objectively apparent reason for the amendment as “to obtain a . . . claims scope that includes polymers that are not widely differing, in structures that are not widely different in their biological properties.” (Tr. 286:7-18 (emphasis added).)

The Court here, like the Federal Circuit in Eli Lilly, finds it unlikely that a competitor would have been justified in assuming that if it used a release coating polymer “with the same biological activity” and “a structure that is nearly identical” to the structure of the specific acrylic polymers exemplified in the patent specification, it would not infringe the Patents-in-Suit. See Eli Lilly, 933 F.3d at 1332 (“[I]t is unlikely that a competitor would have been ‘justified in assuming that if he [made an equivalent pemetrexed salt], he would not infringe [the '209 patent].'” (quoting Kinzenbaw v. Deere & Co., 741 F.2d 383, 389 (Fed. Cir. 1984)) (alterations in Eli Lilly)). The Court therefore concludes that Supernus's proposed objectively apparent reason for the amendment - “to obtain a . . . claims scope that includes polymers that are not widely differing, in structures that are not widely different in their biological properties” - is more likely than not the true reason.

Torrent's arguments to the contrary are unpersuasive. Torrent argues that “[i]f Supernus's amendment to recite the specific cellulosic and acrylic polymer categories was based on some unrecited structure, it would have ‘fail[ed] to distinguish the compound from other molecules or agents that can perform the same functions' and thus failed to resolve the examiner's concerns.” (ECF No. 171 at 30 (quoting PTX233.230).) In other words, including acrylic polymers does not adequately convey a distinct structure. But as the Court found earlier, a POSA reading the specification would understand that the plain and ordinary meaning of acrylic polymers included the exemplary acrylic polymers described in the specification. In fact, opposing literal infringement, Torrent argued that the term acrylic confers structure. (See Tr. 483:1-5 (Rotella) (“And that's why the term ‘acrylic' means something. That's why the term ‘vinyl' means something. Those terms confer structure. And the word ‘acrylic' confers the structure as I have just outlined it for you and illustrated it for you.”).) See Eli Lilly, 933 F.3d at 1332 (“It does not follow . . . that [an] amended claim becomes so perfect in its description that no one could devise an equivalent.” (quoting Festo, 535 U.S. at 738) (alterations in Eli Lilly)).

Torrent also argues that “Supernus clearly had the language to describe vinyl polymers, including polyvinyl acetate, and the ‘the original application once embraced' those polymers, yet Supernus ultimately chose not to claim them.” (ECF No. 171 at 30 (citing PTX233.242; quoting Festo, 535 U.S. at 734); see Tr. 559:10-12 (Felton) (testifying that the term “release-controlling coating” captures “all three categories of polymers”).) Torrent is correct that the original claim covered, in the examiner's words, “any and all . . . release coatings that can exist,” which would include polyvinyl acetate. (PTX233.230.) But that fact is not dispositive. In similar cases where an original claim theoretically encompassed the accused equivalent, the Federal Circuit opted not to apply “[s]uch a draconian preclusion . . . beyond a fair interpretation of what was surrendered.” Intervet, 617 F.3d at 1292; see, e.g., Eli Lilly, 933 F.3d at 1330-31 (holding that a narrowing amendment was only tangential to the accused equivalent even though the equivalent was within “the territory between the original claim and the amended claim”).^[51] The objectively apparent reason for the amendment here was not to disclaim polyvinyl acetate or some attribute of polyvinyl acetate; it was to disclaim release coatings that have widely differing structures and corresponding biological activities. In fact, the amendment was “peripheral, or not directly relevant,” to polyvinyl acetate. Eli Lilly, 933 F.3d at 1332 (quoting Festo, 344 F.3d at 1369).

[51] “After all, the tangential exception only exists because applicants over-narrow their claims during prosecution.” Eli Lilly, 933 F.3d at 1332.

Torrent cites three cases that are compelling but ultimately inapposite. (ECF No. 171 at 30-31.) First, in Duramed Pharmaceuticals, Inc. v. Paddock Laboratories, Inc., at the examiner's suggestion, Duramed amended its independent claim from tablets “coated with a moisture barrier coating” to tablets with a “moisture barrier coating compris[ing] ethylcellulose.” 715 F.Supp.2d 552, 556 (S.D.N.Y. 2010), aff'd, 644 F.3d 1376 (Fed. Cir. 2011). The district court ruled that “[b]ecause Duramed's amendment concerned the type of [moisture barrier coating] to be used, and the alleged equivalent [polyvinyl alcohol] is another type of moisture barrier coating, Duramed's narrowing amendment was not tangential to the alleged equivalent.” Id. at 565. In other words, the tangential exception did not apply, because the amendment and the accused equivalent both related to the moisture barrier coating.

The Federal Circuit held that Duramed did not meet the unforeseeability exception because the prior art application's disclosure of the accused equivalent moisture barrier coating established that the accused equivalent was known in the field of pharmaceutical compositions as of the time of Duramed's narrowing amendment. Duramed Pharms., Inc. v. Paddock Lab'ys, Inc., 644 F.3d 1376, 1381 (Fed. Cir. 2011). But the tangential exception was not discussed on appeal. Had it been, the Federal Circuit may have rejected the above-mentioned “bright-line rule” that it later rejected in Eli Lilly: “where the reason for the amendment and the equivalent in question both relate to the same claim element” - in Duramed, the moisture barrier coating - “the tangential exception does not apply.” Eli Lilly, 933 F.3d at 1333. Indeed, “such a bright-line rule is both contrary to the equitable nature of prosecution history estoppel . . . and inconsistent with the equitable spirit that animates the doctrine of equivalents.” Id. (internal citations omitted).^[52]

[52] Torrent notes that in Duramed, Supernus's counsel's firm opposed applying the tangential exception. (ECF No. 171 at 31.) That counsel has been on both sides of this issue matters less given that each case depends on its unique prosecution history. See Eli Lilly, 933 F.3d at 1333 (collecting cases showing that “prosecution history estoppel d[id] not apply in similar circumstances” “where the prosecution record differed”).

This Court has the benefit of Eli Lilly's guidance. Here, the applicants amended the claim to better describe the release controlling coating. That may be - indeed it is - enough to create a presumption of estoppel. See Cross Med. Prod., Inc. v. Medtronic Sofamor Danek, Inc., 480 F.3d 1335, 1341 (Fed. Cir. 2007) (“[I]f a § 112 amendment is necessary and narrows the patent's scope-even if only for the purpose of better description-estoppel may apply.” (quoting Honeywell Int'l Inc. v. Hamilton Sundstrand Corp., 370 F.3d 1131, 1142 (Fed. Cir. 2004))).^[53] But following Eli Lilly, that alone does not bar Supernus from invoking the tangential exception to accuse an equivalent release controlling coating (here, polyvinyl acetate) of infringement. See Eli Lilly, 933 F.3d at 1333-34 (concluding that the consideration “that an amendment that narrows an existing claim element evinces an intention to relinquish that claim scope” “is not dispositive because the rest of the prosecution history, and the '209 patent itself, show that it is implausible that the reason for Lilly's amendment was to surrender other pemetrexed salts”).^[54]

[53] Again, “these circumstances create a presumption of estoppel” that “the patentee may still rebut.” Cross Med. Prod., 480 F.3d at 1341 (citation omitted).

[54] Compare Pharma Tech Sols., 942 F.3d at 1381, 1383-84 (rejecting the tangential exception where the inventors' remarks “distinguish[ing] the prior art based on the newly added sequential ‘converting' and ‘comparing' limitations” surrendered the accused equivalent's systems, which did not have a “comparing” limitation, because the “comparing” limitation “was integral to the inventors' . . . amendment” and “necessary to overcome the prior art”).

Second, in Cross Medical Products, Inc. v. Medtronic Sofamor Danek, Inc., the applicant told the examiner that to overcome § 112 rejections, it added a “thread depth limitation” to capture the way that the “stabilizer aspect of the invention operated”: with “threads extend[ing] toward the channel to a depth below the top of the stabilizer.” 480 F.3d at 1343. The accused equivalent, however, did “not include threads extending ‘to a depth below the top of the stabilizer'” and thus did “not capture this aspect of the invention.” Id. Because the accused equivalent “relate[d] to the amendment as shown even by the applicant's own statements,” the tangential exception did not apply. Id.

Critical to the Cross Medical Products decision was that an invention either had the threads or did not, and the applicant made clear that the inclusion of threads is what distinguished the invention from prior art. Here, the amendment was not so polarizing. Patentability did not depend on, nor was it influenced by, polyvinyl acetate's absence from the claim. And nothing in the prosecution history indicates that the applicants amended the claim to distinguish their formulation from release controlling coatings in prior art. Rather, the amendment addressed the need for “a disclosure of structure features or elements of the . . . release coatings which would have the stated function.” (PTX233.230.)

Third, and finally, Torrent cites International Rectifier Corp. v. IXYS Corp. ^[55] for the proposition that “because the amendment added specific structure to overcome a Section 112 rejection and the asserted equivalent related to the specific structure added by amendment, the reason could not be deemed tangential.” (ECF No. 171 at 29-30.) The Court disagrees with Torrent's framing of that case's holding. In International Rectifier, the amendment adding the limiting term adjoining more than tangentially related to the specific structure because it “recited precisely the structure it disclosed,” which did not include the opposite non-adjoining structure. 515 F.3d at 1359; see Lite-Netics, LLC v. Nu Tsai Cap. LLC, 60 F.4th 1335, 1346 (Fed. Cir. 2023) (“While we have recognized that literal failure to meet a claim limitation does not necessarily constitute a ‘specific exclusion,' we have found ‘specific exclusion' where the patentee seeks to encompass a structural feature that is the opposite of, or inconsistent with, the recited limitation.” (quoting Augme Techs., Inc. v. Yahoo! Inc., 755 F.3d 1326, 1335 (Fed. Cir. 2014)) (internal citations omitted)); see, e.g., Cross Med. Prod., 480 F.3d at 1343 (finding that an invention whose distinguishing feature was its threads could not be equivalent to an invention that had no threads); Integrated Tech., 734 F.3d at 1355, 1359 (holding that because the patentee relied on physical contact to overcome prior art, it could not “prove by a preponderance of the evidence that . . . the ‘objectively apparent reason for the narrowing amendment' was only tangentially related to the [no-touch] equivalent” (quoting Festo, 344 F.3d at 1369)).

[55] 515 F.3d 1353, 1359 (Fed. Cir. 2008).

In contrast, here the Court has already found that vinyl polymers and acrylic polymers are not mutually exclusive and can overlap. (See § IV.A.2.b above.) Recall, as Dr. Rotella testified, both vinyl polymers and acrylic polymers are at least somewhat defined by their “identical” vinyl groups. (See Tr. 480:6-7 (Rotella) (“[W]here these molecules are similar is in the olefin functional group here. You see it's identical in both.”); Tr. 482:17-483:5 (Rotella) (defining vinyl polymer, “first off, by this olefinic unit,” and acrylic polymer “by that olefin functional group”).) Thus, unlike the adjoining-not adjoining components in International Rectifier, the threads-no threads components in Cross Medical Products, or the touch-no touch components in Integrated Technology, here vinyl polymers and acrylic polymers are not necessarily “the opposite of, or inconsistent with,” each other.

The Court's finding that vinyl polymers and acrylic polymers are not mutually exclusive undermines Torrent's argument that where there are only three categories of polymers (cellulosic, acrylic, and vinyl), the omission of one of them is disclaiming. Even so, the Court finds that the Patents-in-Suit do not disclaim all vinyl polymers. As mentioned, the specification's use of semicolons to offset cellulosic polymers, polyvinyl alcohol, and acrylic polymers indicates that these polymers are separate from one another. Dr. Khan agreed with this reading. (Tr. 318:22-319:3.) He also agreed that polyvinyl alcohol is a vinyl polymer. (Tr. 317:5-13, 319:4-5.) Using that testimony, Torrent contends that the Patents-in-Suit distinguish between three categories of polymers - cellulosic polymers; vinyl polymers, with polyvinyl alcohol as the representative example; and acrylic polymers. If Torrent is correct, the omission of polyvinyl alcohol in the Asserted Claims may prevent Supernus from claiming any vinyl polymer.

But that is neither what Dr. Khan said nor what the specification says. Dr. Khan was careful not to concede that the list represents “distinct categories” of polymers. (See Tr. 319:6-14.) Dr. Khan's carefulness makes sense. The specification's word choice makes clear that when it intends to identify a polymer category, it states a polymer category. Cellulosic polymers and acrylic polymers are categories of polymers. Polyvinyl alcohol is not; it is a specific vinyl polymer. In fact, Dr. Rotella testified that polyvinyl alcohol and polyvinyl acetate are different polymers. (Tr. 524:22-525:1.) So construing “polyvinyl alcohol” as representing all vinyl polymers would not only alter the plain text and ignore that the specification treats categories of polymers and specific polymers as different, but also ignore the differences between polyvinyl alcohol and other vinyl polymers. (See Tr. 524:22-525:1 (Rotella) (testifying that polyvinyl alcohol and polyvinyl acetate are “structurally distinct”).) For this specification, polyvinyl alcohol is not a cellulosic polymer or an acrylic polymer. That's all.

The Court therefore finds that Supernus qualifies for the tangential exception and has rebutted the Festo presumption. As a result, the Court need not address Supernus's argument that “[b]ecause the Asserted Claims do not recite the claim term that was amended in the '931 Application, there is no prosecution history estoppel.” (ECF No. 170 at 36 n.14.) Nor must the Court address Supernus's preemptive strike against the disclosure-dedication doctrine (id. at 36-37), as Torrent did not assert that doctrine in defense.

3. Equivalence Tests

Having found that Supernus rebutted the presumption of prosecution history estoppel, the Court turns to the issue whether Torrent's ANDA Products infringe under the doctrine of equivalents.

There are two tests for evaluating equivalence. The first is the “function-way-result,” or “FWR,” test: “whether the accused product performs ‘substantially the same function in substantially the same way to obtain the same result.'” Mylan Institutional LLC v. Aurobindo Pharma Ltd., 857 F.3d 858, 866 (Fed. Cir. 2017) (quoting Graver Tank & Mfg. Co. v. Linde Air Prod. Co., 339 U.S. 605, 608 (1950)). The second is the “insubstantial difference” test: “whether the accused product or process is substantially different from what is patented.” Id. (citing Graver Tank, 339 U.S. at 608).

The Federal Circuit has said that for “non-mechanical” cases involving the “chemical arts,” the “[in]substantial differences test may be more suitable than FWR for determining equivalence.” Id. at 869. For example, “[h]ow a particular component of a composition, or substituent of a compound, functions in a human or animal body, or in what way, may not be known or even knowable (although, as technology evolves, that may change).” Id. at 867. In that scenario, the patentee could not prove that each limitation satisfies the FWR test. Thus, courts must decide which test is “better suited to the particular facts of th[e] case.” Id. at 867-70. a. Function-Way-Result Test

The Court starts with FWR.

Function. Supernus alleges that both polyvinyl acetate and the acrylic polymers listed in the specification “function as release controlling polymers as part of a release controlling coating.” (ECF No. 170 at 29 (citing Tr. 276:1-24 (Khan)).) In support, Supernus first notes how the '989 patent describes the coating: “In one embodiment, the extended release (XR) component is contained in at least one population of beads coated with a coating that modifies and controls the release of topiramate from the beads (release controlling coating).” (PTX058.13 col. 6 ll. 39-42.) This description tells us, as Dr. Khan said, that the function of the coating material is “the release controlling coating polymer.” (Tr. 276:17-24.) Supernus then compares the '989 patent specification to Torrent's ANDA, which says the “Pharmaceutical Function” of “Polyvinyl acetate dispersion” is “Release controlling polymer.” (PTX106.4.) Dr. Khan testified that the function described in the '989 patent specification and the function disclosed in Torrent's ANDA are “[i]dentical, virtually identical.” (Tr. 277:1-13.)

Supernus also asserts that Torrent's own witnesses confirm Dr. Khan's opinion. Dr. Felton “admitted that the function of polyvinyl acetate and ‘acrylic polymers' that are expressly listed in the patent specification (e.g., polymethacrylates) ‘is to control the release of drug.'” (PFF ¶ 326 (Pl.).) The admission came when Supernus read this excerpt from Dr. Felton's edition of Aqueous Polymeric Coatings for Pharmaceutical Dosage Forms: “The water-insoluble polymers (e.g., ethylcellulose, polyvinyl acetate, neutral or quaternary poly(meth)acrylates) can be applied for controlled drug delivery, especially for prolonged and sustained release.” (Tr. 579:6-17 (quoting DTX019 at TORTOP00127213).) Dr. Felton agreed that these polymers are used “to control the release of drug.” (Tr. 579:18-21.) And Yogesh Patel, who oversaw the formulation and development of Torrent's ANDA Products (PFF ¶ 133), testified that Torrent uses polyvinyl acetate as a release-controlling polymer in its ANDA product (Patel Tr. 72:2-12, 75:24-76:5, 77:6-78:14, 169:25-170:8; PTX106.29).

Torrent makes two counterarguments. Torrent first counters that it is not enough that the release controlling coating controls the release of the active ingredient; the coating must do so in such a way that achieves an 80% drug release at a predetermined period of time. (PFF ¶¶ 231-327 (Def.) (citing Tr. 575:14-22 (Felton)).)^[56] In support, Torrent cites three specification excerpts. First, “[t]he release controlling coating is specific for every population of beads and determines the rate of release of topiramate from the given bead population.” (ECF No. 171 at 32-33 (quoting '989 patent col. 6 ll. 42-45).) Second, “wherein the active ingredient is released from the formulation at a sustained rate along a pre-determined release profile.” ('989 patent col. 2 ll. 23-25.) Third, “there exists a mathematical relationship between the release controlling coating level among the cured beads and the 80% in vitro release time. Pre-determined target profiles can therefore be achieved by the interpolation or extrapolation of the relationship curve.” ('989 patent col. 7 ll. 28-33.)

[56] Dr. Felton testified that based on “the Court's construction and the context of the patent,” she understood the function of the claimed acrylic polymers to be “controlling the release of the active ingredient, the topiramate, . . . in such a way as to achieve a T80, 80 percent drug release at a predetermined period of time.” (Tr. 575:10-19.)

Based on these excerpts, Torrent asserts that “[a]s the only required element of the extended release components, the recited cellulosic and acrylic polymers['] function is to allow the active ingredient to be ‘released from the formulation at a sustained rate along a pre-determined release profile.'” (ECF No. 171 at 32-33 (quoting '989 patent col. 2 ll. 23-25) (emphasis added by Torrent).) In other words, “the function of the claimed acrylic polymers is to control the release of the active ingredient such that 80 percent of the drug [has been] release[d] at the predetermined period of time.” (PFF ¶¶ 231-327 (Def.) (citing Tr. 573:14-22 (Felton)).)

Torrent argues that Dr. Khan ignored this pre-determined release profile aspect of the function. (ECF No. 171 at 33.) Instead, Dr. Khan “relied on his personal experience with polymer coatings.” (Id.) In doing so, Torrent asserts, Dr. Khan disregarded that the FWR test “focuses on an examination of the claim and the explanation of it found in the written description of the patent.” AquaTex Indus., Inc. v. Techniche Sols., 479 F.3d 1320, 1326 (Fed. Cir. 2007) (internal citation and quotation marks omitted).

Torrent next counters that Dr. Khan admitted that many acrylic polymers do not function as he described. (ECF No. 171 at 33 (citing Tr. 373:18-21, 376:10-377:6 (Khan)).) This is problematic, according to Torrent, because although the Asserted Claims recite only acrylic polymers, Dr. Khan admittedly excluded “hundreds, or maybe thousands, of acrylic polymers.” (Id. (citing Tr. 376:10-377:6 (Khan)).)

Supernus contests that “the claimed acrylic polymers must alone function to achieve ‘80 percent drug release at a predetermined period of time.'” (ECF No. 170 at 30 (quoting Tr. 575:14-19 (Felton)).) Supernus argues that Dr. Felton conflated the function of the acrylic polymers with the Court's construction of the extended release (XR) topiramate-containing component. (Id.) Recall, once more, the Court construed extended release (XR) topiramate-containing component to mean “[a] component that releases topiramate in a continuous manner and is adjusted in such a way that 80% of the active ingredient is released in vitro in the predetermined period of time.” (Id. (quoting ECF No. 88 at 33).) Supernus asserts that “[n]either the Court's Markman Order nor the patent specification requires that the ‘acrylic polymers'-by themselves-function to achieve 80% drug release in a predetermined period of time.” (Id.) In fact, Supernus adds, “the specification discloses that the release rate of topiramate is controlled by a combination of factors, including, for example, the ‘type and concentration of a pore former' and the ‘process parameters.'” (Id. (quoting '989 patent col. 7 ll. 13-24).)

The Court agrees with Supernus. At trial, Torrent argued that “there's a relationship between the polymer thickness and the T80 time, the time when the 80 percent releases,” and that a POSA “can use that mathematical relationship and determine how to coat their beads and achieve” a “release time . . . of 80 percent of the T80 within the predetermined period of time.” (Tr. 369:13-370:3 (rearranged for clarity).) In so arguing, Torrent referenced this part of the specification: “For example, with all things being equal, there exists a mathematical relationship between the release controlling coating level among the cured beads and the 80% in vitro release time. Pre-determined target profiles can therefore be achieved by the interpolation or extrapolation of the relationship curve.” ('989 patent col. 7 ll. 28-33.)

But as Supernus submits, the acrylic polymers are only part of the relationship. The patent specification discloses that several factors affect the rate of release of topiramate:

The release-controlling coating is population-specific in the sense that the rate of release of topiramate from every bead population is controlled by at least one parameter of the release controlling coating, such as the nature of the coating, coating level, type and concentration of a pore former, process parameters and combinations thereof. Thus, changing a parameter, such as a pore former concentration, or the conditions of the curing, . . . allows to change the release of topiramate from any given bead population and to selectively adjust the formulation to the pre-determined release profile.

[('989 patent col. 7 ll. 13-24.)]

In addition, although Dr. Khan admittedly excluded many acrylic polymers that function differently, he testified that he excluded ones that “are not used for sustained release.” (Tr. 376:20-21.) Indeed, the Asserted Claims require a “sustained release formulation.” (See, e.g., '989 patent col. 21 l. 1.)

Thus, the Court finds that the specification coupled with the experts' and Mr. Patel's testimony establish that both polyvinyl acetate and the acrylic polymers listed in the Patents-in-Suit function as release-controlling polymers.

Way. Supernus next alleges that “[p]olyvinyl acetate functions in the same way as the polymers listed in the specification-by creating a barrier that controls the entry of aqueous fluid into the coated pellets.” (ECF No. 170 at 30 (citing Tr. 277:14-280:2, 280:20-281:11, 279:7-280:2 (Khan); PTX113.3).) According to the Handbook of Pharmaceutical Excipients, Eudragit^® RL and RS, tradenames of two polymethacrylates, which are exemplary acrylic polymers recited in the Patents-in-Suit, “are used to form water-insoluble film coats for sustained-release products.” (PTX115.5.) Consistent with that text, Dr. Khan testified that the acrylic polymers listed in the patent work by forming “a water insoluble film coating on the bead” such that “the drug diffuses out” of the water-insoluble film coating. (PFF ¶ 329 (Pl.) (citing Tr. 278:6-279:6 (Khan); PTX115.5).)

Here, too, Supernus asserts that Torrent's witnesses confirm Dr. Khan's opinion. Supernus says “Dr. Felton admitted that the way both polymethacrylates and polyvinyl acetate function in sustained release formulations is by acting as a ‘diffusional barrier[].'” (ECF No. 170 at 31; Tr. 580:2-12 (Felton).) After asking Dr. Felton about the polymers' function, Supernus confronted her with more text from Aqueous Polymeric Coatings: “The water-insoluble polymers (e.g., ethylcellulose, polyvinyl acetate, neutral or quaternary poly(meth)acrylates) . . . are insoluble in the entire gastrointestinal tract and act as a diffusion barrier as long as the coating remains intact. The API is released by a diffusion-controlled mechanism through the polymer film itself (often after swelling) or through pores.” (DTX019 at TORTOP00127213 (emphasis added).) Dr. Felton agreed that this text explained the way in which those polymers work, adding that “[t]hey are diffusional barriers, and that's why they're used as coating materials to control the release of drug.” (Tr. 580:2-12.)

Torrent advances a different way theory. As it did for the polymers' function, Torrent argues that Supernus must show the way in which a POSA achieves the pre-determined release profiles. (ECF No. 171 at 33-34.) Torrent asserts that “[t]he patent provides no support to show that the acrylic polymers listed in the patent achieve the results claimed in the patents.” (PFF ¶¶ 330-332.) Though not dispositive, Torrent cites no expert testimony in support. Torrent only attacks Dr. Khan's position that “it's not very important” “whether Torrent's ANDA Products obtained . . . any pre-determined target release profile based on modifying the coating thickness in the way described in the patent.” (Tr. 367:11-20.)

It is true, as mentioned, that the FWR test focuses on “an examination of the claim and the explanation of it found in the written description of the patent,” as well as, “[i]n some cases, the patent's prosecution history.” AquaTex Indus., 479 F.3d at 1326, 1328 (quoting Vehicular Techs. Corp. v. Titan Wheel Int'l, Inc., 141 F.3d 1084, 1090 (Fed. Cir. 1998)). “But this, of course, does not mean that discussion of the equivalence of the function, way, or result between a claimed invention and an accused product is irrelevant when the claims and specification of a patent are silent on the subject.” Stumbo v. Eastman Outdoors, Inc., 508 F.3d 1358, 1364 (Fed. Cir. 2007). “When the claims and specification of a patent are silent as to the result of a claim limitation, . . . we should turn to the ordinarily skilled artisan.” Id.; see Intendis, 822 F.3d at 1362 (“We have never held that a patent must spell out a claim element's function, way, and result in order for the doctrine of equivalents to apply as to that element.”). In Intendis, the Federal Circuit rejected that “a determination of the claimed element's function is limited to a review of the intrinsic record.” 822 F.3d at 1362. “The relevant inquiry,” the Federal Circuit added, “is what the claim element's function in the claimed composition is to one of skill in the art, and a fact finder may rely on extrinsic evidence in making this factual determination.” Id. (citing Zenith Lab'ys, Inc. v. Bristol-Myers Squibb Co., 19 F.3d 1418, 1425 (Fed. Cir. 1994)).

In its literal-infringement rebuttal, Torrent argued that the textbook Aqueous Polymeric Coatings - on which Supernus relies for the way prong - was “reflective of, and consistent with, how a POSA would view the categories of polymers available for use as a release control coating in sustained release formulations.” (ECF No. 171 at 8 (citing Tr. 557:17-24 (Felton)).) The literature Aqueous Polymeric Coatings is reliable here too. Thus, the Court may use it to determine how a POSA understands the way in which the release-controlling polymers function.

Considering Dr. Khan's and Dr. Felton's testimony and the scientific literature, along with the lack of support for Torrent's argument, the Court finds that both the polyvinyl acetate in Torrent's ANDA Products and the acrylic polymers listed in the specification function in the same way: by creating a barrier that controls the entry of aqueous fluid into the coated pellets.

Result. Supernus last alleges that the polyvinyl acetate in Torrent's ANDA Products achieves the same result as the polymers listed in the specification - controlled release of topiramate. (ECF No. 170 at 31; see Tr. 280:3-11 (Khan) (testifying that the result is “you control the release of the drug” topiramate).) Once again, Supernus points to Dr. Felton's testimony. After covering the function and way prongs, Supernus confronted Dr. Felton with one more excerpt from Aqueous Polymeric Coatings: “The release rate of the drug can mainly be influenced by the type of polymer, the thickness of the film, and the addition of excipients, such as pore formers.” (Tr. 580:17-19 (Felton) (quoting DTX019 at TORTOP00127213).) Asked if she agreed that “this is the result of how those three polymers” - that is, “ethylcellulose, polyvinyl acetates, neutral or quaternary poly(meth)acrylates” - “are used, being that you can adjust the release rate of the drug,” Dr. Felton testified as follows:

I think that's something that a person of ordinary skill would understand, that the release rate of the drug can be influenced by the type of polymer that's used, the thickness, which is how much coating is added to the substrate, and the addition of other excipients, including pore formers.

[(Tr. 580:22-581:4.)]

Satisfied with the result prong, the Court concludes that Supernus has provided “particularized testimony and linking argument as to the” function, way, and result of the claimed acrylic polymers and the polyvinyl acetate in Torrent's ANDA products. AquaTex Indus., 479 F.3d at 1328 (quoting Texas Instruments Inc. v. Cypress Semiconductor Corp., 90 F.3d 1558, 1567 (Fed. Cir. 1996)).

b. Insubstantial Differences Test

The insubstantial-differences test asks “whether the substitute element plays a role substantially different from the claimed element.” Warner-Jenkinson, 520 U.S. at 40; see, e.g., Mylan Institutional, 857 F.3d at 869 (“But the court failed to consider whether the key reagent in the process, manganese dioxide, was substantially different from the claimed reagent, silver oxide, and hence whether the substitution for, and omission of, silver oxide left the accused infringer outside of the bounds of the claims.”). Here, the issue is whether polyvinyl acetate in Torrent's ANDA Products is substantially different from the claimed acrylic polymers.

Supernus says it is not. Dr. Khan testified that from a pharmaceutical scientist's perspective, the polymers in the Patents-in-Suit and the polyvinyl acetate dispersion in Torrent's ANDA Products are insubstantially different. (Tr. 281:14-24.) To illustrate, Dr. Khan compared polymethacrylate, one of the polymers listed in the patent, to Torrent's polyvinyl acetate:

(Image Omitted) [(PTX195 (left);^[57] PTX196 (right).)]

[57] The parties have referred to polymethacrylate, poly acrylate, and polymethyl acrylate interchangeably. (See ECF No. 170 at 32 (describing PTX195 as polymethacrylate and polymethyl acrylate); Tr. 402:12-17 (Khan) (accepting counsel's statement “poly acrylate, i.e., poly methacrylate”).)

Dr. Khan testified that the two polymers have “very, very close” structures - “[t]hey do not have widely differing structures” - specifically, the same vinyl structure. (Tr. 281:19-21, 286:19-288:1.) He also testified that they have the same biological properties, that they both release drug in the entire gastrointestinal tract, and that they are both used for the same purpose. (Tr. 286:19-288:1.)

Perhaps most telling, Dr. Rotella thrice refused to testify that the two polymers have “widely differing” structures. Asked on direct examination to respond to “Dr. Khan's position that these two molecules” - comparing an acrylic derivative monomer with vinyl acetate - “are nearly identical or don't have widely differing structures,” Dr. Rotella testified as follows: “Well, you know, depending on your definition of widely different or not widely different, you might say, well, okay, that's a possibility, but they are different.” (Tr. 483:7-13.)

On cross-examination, Dr. Rotella avoided two more opportunities to testify unequivocally that the structures of polyacrylonitrile, polyvinyl acetate, and polymethyl acrylate are widely differing:

MR. KURZ (ATTORNEY FOR SUPERNUS). And just so I'm clear on your testimony, because I remember your counsel brought up whether these are widely differing structures or at least two of the structures on the screen are widely differing. And is it your testimony that these three have widely differing structures?

DR. ROTELLA. I would say that these three structures are different. And I would also repeat what I said previously that the one on the left and the one on the far right are representatives of acrylic polymers. The structures are clearly different from each other. And as I have stated previously, the polymer derived from acrylonitrile and the polymer derived from methyl acrylate are representative examples of acrylic polymers. The polymer in the middle derived from vinyl acetate is not, because of those structural differences.

Q. Okay. And just so the record is clear, is it your opinion that these are widely differing structures?

A. I said that they are different structures.

[(Tr. 511:24-512:15 (referencing PTX198).)]

During his examination, Dr. Rotella implored the Court to “[b]elieve your eyes, because they're right.” (Tr. 460:6-7, 471:12-13, 480:6-7.) The Court finds that these structures are not widely differing, and Dr. Rotella's reluctance to testify otherwise further supports that the structures' differences are insubstantial. Because the structural differences between polyvinyl acetate and the claimed acrylic polymers are insubstantial, and because the polymers pass the FWR test, the Court concludes that Torrent's use of polyvinyl acetate in its ANDA Products infringes the Asserted Claims under the doctrine of equivalents.

C. Invalidity

A patent is presumed valid. 35 U.S.C. § 282(a). The party asserting invalidity of a patent or claim bears the burden of establishing its invalidity. Id. Rebutting the validity presumption requires clear and convincing evidence of invalidity. Sciele Pharma Inc. v. Lupin Ltd., 684 F.3d 1253, 1260 (Fed. Cir. 2012) (citing 35 U.S.C. § 282; Microsoft Corp. v. i4i Ltd. P'ship, 564 U.S. 91, 95 (2011)). “Clear and convincing evidence places in the fact finder ‘an abiding conviction that the truth of [the] factual contentions are highly probable.'” Procter & Gamble Co. v. Teva Pharms. USA, Inc., 566 F.3d 989, 994 (Fed. Cir. 2009) (quoting Colorado v. New Mexico, 467 U.S. 310, 316 (1984)). This is a “high burden of proof created by the necessary deference to the PTO.” Sciele, 684 F.3d at 1260. To be sure, the burden does not require “extremely clear and convincing evidence” or “crystal clear and convincing evidence.” Id.

1. Enablement

A patent is invalid if it cannot enable a POSA to “make and use” the claimed invention. 35 U.S.C. § 112(a). To win on a lack of enablement theory, “a challenger must show by clear and convincing evidence that a person of ordinary skill in the art would not be able to practice the claimed invention without ‘undue experimentation.'” Alcon Rsch. Ltd. v. Barr Lab'ys, Inc., 745 F.3d 1180, 1188 (Fed. Cir. 2014) (citing In re Wands, 858 F.2d 731, 736-37 (Fed. Cir. 1988)); ALZA Corp. v. Andrx Pharms., LLC, 603 F.3d 935, 940 (Fed. Cir. 2010) (“To be enabling, the specification of a patent must teach those skilled in the art how to make and use the full scope of the claimed invention without undue experimentation.” (citations omitted)).

“[A]n enablement determination is made retrospectively, i.e., by looking back to the filing date of the patent application and determining whether undue experimentation would have been required to make and use the claimed invention at that time.” Enzo Biochem, Inc. v. Calgene, Inc., 188 F.3d 1362, 1371-72 (Fed. Cir. 1999). To determine if experimentation is “undue,” courts consider the so-called Wands factors:

(1) the quantity of experimentation necessary, (2) the amount of direction or guidance presented, (3) the presence or absence of working examples, (4) the nature of the invention, (5) the state of the prior art, (6) the relative skill of those in the art, (7) the predictability or unpredictability of the art, and (8) the breadth of the claims.

[Cephalon, Inc. v. Watson Pharms., Inc., 707 F.3d 1330, 1336 (Fed. Cir. 2013) (quoting Wands, 858 F.2d at 737)^[58]]

[58] “[I]t is not necessary that a court review all the Wands factors to find a disclosure enabling. They are illustrative, not mandatory.” Streck, Inc. v. Rsch. & Diagnostic Sys., Inc., 665 F.3d 1269, 1288 (Fed. Cir. 2012) (quoting Amgen, Inc. v. Chugai Pharm. Co., 927 F.2d 1200, 1213 (Fed. Cir. 1991)).

A reasonable amount of routine experimentation is permissive. ALZA, 603 F.3d at 940.

Torrent submits that for the Patents-in-Suit “to be enabled for the full scope of the claims, a POSA must be able to make and use formulations of topiramate that achieve the appropriate ‘predetermined release rate' for” the “incredibly broad variety of conditions” that the claims-covered formulations intend to treat. (ECF No. 171 at 36-37.) Yet for claims so broad and an invention so complex, Torrent argues, the Patents-in-Suit teach too little. Torrent first argues that the patents claim a “near infinite number of formulations” yet give examples of only seven formulations. (Id. at 37; PFF ¶ 467 (Def.).) Next, the patents do not disclose the exact dissolution methodology that Supernus used. (ECF No. 171 at 38.) Finally, the patents lack “any connection between the formulations of the two critical example[s], Examples 1 and 6.” (Id. (citing Tr. 598:13-17 (Felton)).)

The Court will discuss each of Torrent's challenges and the Wands factors it implicates.

a. Seven examples for infinite formulations

Torrent's first challenge implicates the breadth of the claims and amount of direction presented factors. Torrent argues that the Patents-in-Suit do not provide enough information to teach what they claim. For its part, Torrent first relies on remarks from the prosecution of the '475 application that led to a patent not in suit. The applicants remarked that “the number of possible populations of beads available to the pharmacologist is near infinite.” (DTX015 at 29.) Confronted with these remarks at trial, Supernus's witness Padmanabh Bhatt, Ph.D., a founding executive of Supernus and one inventor of the Patents-in-Suit, testified that the remarks meant that “we can make literally hundreds and thousands of formulations,” and “[i]t can take years and years to do that.” (PFF ¶¶ 124, 126-128; Tr. 134:3-9.) Dr. Bhatt testified that they “could have done hundreds, if not thousands, of examples for table 1,” though they disclosed only seven. (Tr. 126:15-21, 135:6-14.)^[59] Thus, Torrent argues that there is a “disconnect between what the asserted patents claim (near infinite number of formulations) and what the asserted patents actually teach (7 formulations).” (ECF No. 171 at 37.)

[59] In its invalidity discussion, the Court cites Dr. Bhatt's factual, nonexpert testimony only as to his personal knowledge of “the claimed invention and its development.” Verizon Servs. Corp. v. Cox Fibernet Virginia, Inc., 602 F.3d 1325, 1340 (Fed. Cir. 2010) (citing Voice Techs. Grp., Inc. v. VMC Sys., Inc., 164 F.3d 605, 615 (Fed. Cir. 1999)).

In further support, Torrent cites testimony of Dr. Felton, who felt that “the breadth of the claims is quite large,” also highlighting the “near infinite” remark. (Tr. 608:11-14; DTX015 at 29.) Dr. Felton testified that there are “numerous different types of polymers that are acrylic polymers, and there's also a number of cellulosic polymers,” and that there are “a lot of different polymers that could be used to coat the beads in order to get the desired release profiles of those beads.” (Tr. 609:16-22.) “And within each of the polymers,” she added, “there's different weight gains, there's different plasticizers and other excipients that might be added to the coatings to get them to adhere to the dosage form and function like they're supposed to.” (Tr. 609:23-610:1.)

Supernus calls Torrent's “near infinite” argument misleading and taken out of context. (ECF No. 170 at 46.) The Court agrees. The remark concerned claims in an application for a familial patent that is indisputably not asserted or even “directly related in this case.” (ECF No. 170 at 46 (citing PFF ¶¶ 396, 467); Tr. 620:23-621:2 (Felton).) That application is not useful here, as its claims differ from those at issue in a meaningful way. The '475 application remark relates to claims that recite specific combination-ratios of multi-bead formulations (ECF No. 170 at 37 (citing PFF ¶¶ 396, 467; Tr. 621:13-16, 621:25-622:7, 622:21-23, 623:16-24 (Felton); DTX015 at 13, 23-24, 29)), as opposed to single-bead formulations, which the Patents-in-Suit describe (PFF ¶¶ 396, 410-411, 445, 448-449, 457-458, 462, 493-494, 501-502, 506, 512 (Pl.)). In fact, the claims' only single-bead formulation, designated as Formulation 7, had been crossed out. (DTX015 at 13, 23-24; Tr. 621:17-62324.)

By contrast, Formulation 7 in Example 6 of the '989 patent discloses a sustained-release formulation that contains only one population of extended-release beads. ('989 patent Table 5.) Example 6 also reports the in-vivo pharmacokinetic testing of three single-bead populations, designated as XR1, XR2, and XR3, with the data for each bead's population reported in Figure 3. (ECF No. 170 at 41; '989 patent Fig. 3, Example 6.) The Patents-in-Suit's description of single-bead formulations makes the unrelated application and its remarks less meaningful here.

In addition, the remark was part of a disagreement over the claims' obviousness, not its enablement or written description. In making the remark, the applicants stated that they were responding to the examiner's allegation that “the release-profile is considered a result-effective variable which is obvious to optimize . . . with a combination of independent pellet populations which release topiramate at different layers.” (DTX015 at 29, SUPTXR0002277.) The applicants then remarked in relevant part as follows:

[T]he Examiner has not explained what would have guided the ordinary artisan to specifically select the bead populations as identified in items 1-3 of claims 34 and 92. Indeed, . . . there is no teaching in the art that would have led the ordinary artisan to the populations of beads per se identified by the present inventors.

[(DTX015 at 29.)]

In this context, the “near infinite” remark addressed allegations of obviousness - it did not relate to the claims' scope.

For these reasons, the Court does not give weight to the “near infinite” remark. This challenge does not support a non-enablement finding.

b. Dissolution test methodology

Torrent's next challenge implicates the amount of direction presented, relative skill of a POSA, and predictability of the art factors. There is no dispute that the Patents-in-Suit do not disclose the inventors' exact dissolution methodology. The issue is whether they needed to disclose it for enablement.

As a rule, “a patent need not teach, and preferably omits, what is well known in the art.” Streck, 665 F.3d at 1288 (quoting Hybritech Inc. v. Monoclonal Antibodies, Inc., 802 F.2d 1367, 1384 (Fed. Cir. 1986)). That rule, however, is “merely a rule of supplementation, not a substitute for a basic enabling disclosure.” ALZA, 603 F.3d at 940-41 (quoting Auto. Techs. Int'l, Inc. v. BMW of N. Am., Inc., 501 F.3d 1274, 1283 (Fed. Cir. 2007)). Indeed, “the specification, not the knowledge of those skilled in the art, ‘must supply the novel aspects of an invention' to satisfy the enablement requirement.” Streck, 665 F.3d at 1289 (quoting Auto. Techs., 501 F.3d at 1283).

Torrent argues that the patents needed to disclose the exact dissolution methodology. Dr. Felton testified that if a POSA does not use the “exact same methodology” as the one used in the patent, the POSA's T80% results will be “meaningless in reference to this patent,” because “it would be like comparing apples and oranges.” (Tr. 602:20-603:5.) Without that information, Dr. Felton continued, a POSA “would have to conduct the equivalent of a clinical trial,” which “is a huge undertaking, and certainly more than routine experimentation.” (Tr. 610:18-21.)

Dr. Felton testified that she found only “one small[,] generalized statement about the dissolution methodology” in the specification:

The release rates referred to herein are determined by placing a dosage form to be tested in a medium^[60] in an appropriate dissolution bath. Aliquots of the medium, collected at pre-set intervals, are then injected into a chromatographic system fitted with an appropriate detector to quantify the amounts of drug released during the testing intervals.”

[(Tr. 601:12-602:7; '989 patent col. 3 ll. 65-67, col. 4 ll. 1-3.)]

[60] The medium is “the liquid that we're going to put the product in.” (Tr. 602:2-3 (Felton).)

Dr. Felton testified that this statement does not say “what dissolution methodology was used” or “what type of equipment to use.” (Tr. 602:8-11.) Although “[t]here's a number of different standardized pieces of equipment that are used for dissolution testing,” Dr. Felton added, “that's not disclosed here, so I don't know which is used.” (Tr. 602:11-13.) “So while the person of ordinary skill is familiar with the dissolution test and could recreate his or her own test,” Dr. Felton concluded, “if they're not using the exact same methodology, they're not going to get an equivalent T80 to be able to compare.” (Tr. 602:20-24.)

Dr. Bhatt agreed that the Patents-in-Suit disclose “no specific information about the parameters or the equipment that's used in order to generate data.” (Tr. 118:8-11.) But Dr. Bhatt insisted that they did not need to include specific information; it is enough to say that the methodology “was a dissolution study.” (Tr. 110:23-24.) This is because the standard dissolution methodologies are publicly available. The United States Pharmacopeia (USP) publishes guidance on the dissolution apparatuses and media to use for a given purpose. Dr. Bhatt described USP as “a standard reference book for conducting testing of pharmaceutical products.” (Tr. 110:16-18.) “For solid dosage forms,” Dr. Bhatt testified, the guidance “detail[s] USP dissolution bath number two apparatus as a standard dissolution testing in vitro testing method.” (Tr. 109:11-16.) The FDA also publishes guidance on how to test extended-release products. (Tr. 110:19-20.) So pharmaceutical scientists doing formulation development, including testing formulations in vitro, know to visit USP and FDA guidance. (Tr. 109:5-110:24.)

Echoing Dr. Bhatt, Dr. Khan testified that for dissolution methodology, “[e]verybody knows [to] go to USP.” (Tr. 634:20-21.) In fact, there are only seven approved USP methods, and “most of them use USP apparatus one or apparatus two, because others are used for other purposes, not for tablets and capsules.” (Tr. 634:11-635:5.) Even his “simple laboratory” has “full dissolution equipment.” (Tr. 634:16-18.) Dr. Khan also testified that the “FDA routinely publishes . . . product-specific guidance,” including a “topiramate dissolution method.” (Tr. 635:10-14.) And if a method is not available, “sponsors submit a controlled correspondence to FDA,” requesting “a method for dissolution.” (Tr. 635:14-22.)

For reasons discussed after further factfinding below, the Court finds that Torrent did not meet its burden to prove that the Patents-in-Suit needed to describe the inventors' exact dissolution methodology.

c. Connection between Examples 1 and 6

Torrent's final challenge implicates the quantity of experimentation, amount of direction presented, working examples, relative skill of a POSA, and predictability of the art factors. Torrent argues that there is no connection between Example 1's and Example 6's bead formulations. (ECF No. 171 at 38 (citing Tr. 598:13-17 (Felton)).) Example 1, Table 1, lists three XR1 beads (XR1a, XR1b, XR1c), four XR2 beads (XR2a, XR2b, XR2c, XR2d), and one bead each for XR3 to XR8. ('989 patent Table 1; Tr. 597:22-598:1 (Felton).) Dr. Felton testified that she “could not understand how the compositions in table 1 related to” Example 6, which lists only three beads, designated as XR1, XR2, and XR3, and does not state their compositions. (Tr. 598:8-12, 606:22-607:3, 610:15-18; '989 patent col. 18 l. 26.) So Dr. Felton concluded that “the examples in table 1 do not correspond at all to examples in example 6.” (Tr. 598:13-14.) She opined that without that connecting information, a POSA “would have to test the product in vivo. They would have to create beads, test them in vivo. It would be like an entire drug development project, a huge undertaking.” (Tr. 603:23-604:4.)

No one disputes that none of the 13 exemplary beads in Table 1 directly corresponds to any of the three beads designated as XR1, XR2, and XR3 in Example 6. Rather, Supernus argues that “[b]ased on the teachings in Example 1 and Figures 1 and 2, a POSA would have understood how to make XR formulations that match the parameters listed in Example 6 to achieve Tmax . . . in vivo at 16 or more hours after a single initial dose.” (PFF ¶ 444 (Pl.).)

The Court analyzes the specification to determine whether a POSA would draw a connection between Examples 1 and 6.

Along with describing several embodiments, the specification describes a multi-step method of preparing bead formulations:

The current invention additionally encompasses a method of preparing formulations of topiramate, comprising an extended release component, and an optional immediate release component, wherein topiramate is released from the formulation at the sustained rate along the pre-determined release profile. The method comprises the following steps:

1. determining the desired release profile;

2. determining specific amounts of the extended release component and the immediate release component necessary to produce the pre-determined release profile; and

3. incorporating the specified amounts of the components into the formulation.

[('989 patent col. 11 ll. 25-36.)]

Example 1, titled Extended Release Beads Preparation, teaches how to prepare the beads contained in the product capsule. ('989 patent col. 14 l. 15; Tr. 61:21-62:23 (Bhatt).) Relevant here, the disclosure teaches the “Coating of the Core with a Release Controlling Coating.” ('989 patent col. 14 l. 33.) To that end, Table 1, titled Composition and process parameters for the extended Release, exemplifies 13 different beads that Supernus developed and designated as follows: XR1a, XR1b, XR1c, XR2a, XR2b, XR2c, XR2d, XR3, XR4, XR5, XR6, XR7, and XR8. (See '989 patent Table 1, col. 14 ll. 40-43 (“The release controlling coating dispersion is sprayed onto the bed to evenly coat the core to a desired coating level as exemplified in Table 1.”); Tr. 63:7-11, 64:19-23 (Bhatt).)

(Table Omitted) [('989 patent Table 1.)]

In each bead's column, the top row discloses the name of the release-controlling polymer, or “RC coating material,” sprayed onto that bead. All but one bead uses ethylcellulose. ('989 patent Table 1; Tr. 598:3-7 (Felton).) The one bead, XR8, uses an acrylic-polymer combination of Eudragit^® RL 30 D and RS 30 D. ('989 patent Table 1; Tr. 598:3-5 (Felton).) Some beads also use a “pore former,” which helps speed up the drug's release.^[61] (Tr. 66:15-67:15 (Bhatt).) If the bead uses a pore former, the second and third rows disclose the pore former's name and the ratio of the RC coating material to the pore former. ('989 patent Table 1; Tr. 63:23-64:3 (Bhatt).)

[61] For examples of suitable pore formers, see '989 patent col. 7 ll. 48-67.

The fourth row shows the bead's RC coating level, which discloses how much RC coating material to use relative to the bead's total weight. (Tr. 74:4-9 (Bhatt).) The fifth row, Product temperature during coating, discloses the temperature of air flowing through the equipment as the beads are coated. (Tr. 74:14-24 (Bhatt); see '989 patent col. 14 ll. 31-33 (“Inlet airflow rate and product temperature are adjusted to keep the batch from spray-drying the coating material or over-wetting the spheres.”).) The sixth and seventh rows, Over-coat material and Over-coat coating level, respectively, disclose the “nonfunctional coating” that rounds and smoothens the bead's surface “so that the beads can then be easily filled into a capsule.” (Tr. 74:25-75:12 (Bhatt).)^[62]

[62] For more information on overcoats, see '989 patent col. 8 ll. 4-13.

The bottom row, Curing method, discloses the equipment used for curing. (Tr. 75:17-78:3 (Bhatt).) More details on the curing process are provided in Example 5. ('989 patent col. 17 ll. 45-67, col. 18 ll. 1-10; Tr. 78:5-19 (Bhatt).) Example 5 discloses, for instance, “assisting solvents” that “dramatically reduce[] the curing time necessary to produce a desired release profile and a level of stability” from “up to two weeks” to “several hours.” ('989 patent col. 11 ll. 58-67; see '989 patent col. 13 ll. 1-6 (“In most instances, less than 4 hours of solvent-assisted curing resulted in more complete curing of the extended release dosage forms than 2 weeks of heat-only oven curing of the same dosage forms.”).)

Dr. Bhatt testified that Table 1 does not describe every detail, as some details are known in the art. (Tr. 65:4-8.) Dr. Felton did not take issue with Example 1 and Table 1; she understood them just fine. (See Tr. 607:5-8 (Felton) (“So, in example 1, we saw that the inventors disclosed how to make compositions. But the person of ordinary skill knew how to make and coat beads. So that's not what's novel.”).) Her issue remained that the 13 beads in Table 1 “don't relate at all to what's shown in example 6.” (Tr. 607:12-14 (Felton).)

Example 6, titled Sustained Release Formulations of Topiramate, “gives a plan of study for the human subjects” to see how formulations “perform in the body in vivo.” ('989 patent Example 6; Tr. 79:10-11, 80:8-9 (Bhatt).) Example 6 describes the inventors' “6-way crossover study,” where volunteers underwent six stages of testing, receiving a different bead each stage so that every volunteer receives every bead over the course of the study. ('989 patent col. 18 l. 23; Tr. 79:10-80:2 (Bhatt); PFF ¶ 443 (Pl.).) The study included administration of “three extended release compositions, designated here” - in Example 6 - “as XR1, XR2 and XR3,” each representing a single-bead formulation. ('989 patent col. 18 ll. 24-30; Tr. 80:22-25 (Bhatt).) The study also included administration of “two immediate release compositions ((Topamax®, Ortho-McNeil Neurologics, Inc.) (25 mg BID), and an immediate release bead formulation) and an enhanced immediate release (IR) bead composition.” ('989 patent col. 18 ll. 26-30.) Data from the Example 6 in-vivo study are shown in Figures 3 and 4. (See '989 patent col. 18 ll. 30-33 (referencing Figures 3 and 4).)

Figure 3 shows the “single dose topiramate plasma concentration profiles for XR1, XR2 and XR3.” ('989 patent Fig. 3, col. 4 ll. 53-54, col. 18 ll. 30-31; Tr. 82:19-22 (Bhatt).) Each bead has its own data-curve. The X axis is time in hours over 72 hours starting from the participant's swallowing the capsule, and the Y axis is the plasma concentration of topiramate at different times over 72 hours. (Tr. 83:3-20 (Bhatt); Tr. 167:24-168:1 (Taft).) The apex of each bead's curve reflects that bead's Tmax. (Tr. 84:2-3 (Bhatt).) For XR2 and XR3 each, the Tmax is about 25 hours, which is “obviously greater than 16 hours,” as the Asserted Claims require. (Tr. 85:1-5 (Bhatt); '989 patent col. 21 ll. 33-35.)

(Image Omitted) [('989 patent Fig. 3.)]

Figure 4 also shows topiramate plasma concentration profiles, this time for the IR formulations - two of Supernus's IR beads and, as the control formulation, Topamax. ('989 patent Fig. 4, col. 4 ll. 55-56, col. 18 ll. 31-33; Tr. 85:6-25 (Bhatt).)

(Image Omitted) [('989 patent Fig. 4.)]

Circling back, Example 6 further states,

The extended release populations XR1, XR2 and XR3, and an immediate release (IR) population are selected in such a way as to be defined by at least one of the three following sets of conditions:

1. for the steady state, for XR1, 1.70CmaxIR>=CmaxXR1>=1.30CmaxIR for XR2, 0.40CmaxIR>=CmaxXR2>=0.20CmaxIR for XR3, 0.25CmaxIR>=CmaxXR3>=0.05CmaxIR

2. for in-vitro dissolution, for XR1, 1.5 h<=T80%<=4 h for XR2, 5 h<=T80%<=8 h for XR3, 8 h<T80%<=10 h

3. for a single initial dose in-vivo, for XR1, 4 h<=Tmax<=8.5 h for XR2, Tmax>=16 h for XR3, Tmax>=16 h.

[('989 patent col. 18 ll. 36-55.)]

Dr. Felton understood these conditions. She confirmed that “for in-vitro dissolution,” “the desire is that the XR1 needs to have a T80 of between 1.5 and four” hours; the XR2 bead, “a T80 of between five and eight hours”; and the XR3 bead, “greater than eight hours or equal to or less than ten hours.” (Tr. 617:11-618:8; PFF ¶ 454 (Pl.).) Though emphasizing that the inventors' exact dissolution methodology is critical, Dr. Felton agreed that regardless of the methodology, the conditions dictate that “the XR2 dissolution for T80 is slower than the XR1 dissolution” and faster than the XR3 dissolution. (Tr. 619:7-620:2.) But still, Dr. Felton's issue was that she could find “no data in here that . . . say what the XR1 is or that that XR1 had that dissolution data that met this criteria.” (Tr. 617:21-23.)

Both sides agree that a POSA would conduct in-vitro dissolution testing before in-vivo testing. (See Tr. 600:11-16 (Felton) (“I don't want to go to a clinical test until I've tested something in the lab.”); Tr. 79:4-9 (Bhatt) (testifying that the process requires testing formulations in vitro to identify formulations for human clinical study).) They also agree that in-vitro dissolution testing is relatively simple. (See Tr. 600:1-5 (Felton) (testifying that the “in-vitro dissolution” test “just means testing the drug and measuring the release of the drug in a beaker,” which involves “a lot of parameters” but is “the simplest test”); Tr. 620:11-15 (Felton) (agreeing that “develop[ing] a dissolution test, coat[ing] beads with different weight gains, conduct[ing] the dissolution test, and analyz[ing] the data” “would be normal, everyday skills that a POSA could do”); Tr. 634:11-636:8 (Khan) (testifying that in-vitro dissolution testing is “very standard” and “very routine”).) Therefore, the Court finds that the specification provides enough direction for a POSA to understand that to carry out Example 6, the POSA must first identify beads that meet the T80% conditions “for in-vitro dissolution.”

Figures 1 and 2 of the Patents-in-Suit show the results of in-vitro dissolution testing. ('989 patent Fig. 1, Fig. 2; see '989 patent col. 4 ll. 47-52 (describing Figures 1 and 2 as showing “the time of release of 80% of the drug vs. % wt. gain of release controlling coating for” specific “coated extended release beads”); '989 patent col. 7 ll. 28-45 (describing Figures 1 and 2 as “in-vitro dissolution test[s]”).)

Figure 1 graphs the results of in-vitro experiments calculating how long it took for 80% of topiramate to release from three different sets of beads, each coated with Surelease^® ethylcellulose in varying thickness, into the dissolution media. ('989 patent Fig. 1; Tr. 68:22-12 (Bhatt).) In simpler terms, the test measured how fast topiramate released from the beads and dissolved in surrounding liquid, depending on how much Surelease^® RC coating was on the beads. When plotted in log scale, which is used to scale down data that increase exponentially, the datapoints formed a nearly straight line sloping upward. (Tr. 69:16-19, 71:19-72:2 (Bhatt).)

(Image Omitted) [('989 patent Fig. 1.)]

Figure 2 shows the results of other in-vitro experiments like Figure 1's but with a few differences. ('989 patent Fig. 2.) First, the beads were coated with ethylcellulose along with a pore former. (Tr. 72:6-9 (Bhatt); see Tr. 661:3-5 (Khan) (“That's the difference between figure 1 and Figure 2. One of them has pore former and the other one doesn't have a pore former.”).) Second, the Y axis reflects time in hours, but it is not converted into a log scale. (Tr. 72:11-15 (Bhatt); see Tr. 661:5-6 (Khan) (“So one is log relationship. The other one is non-log relationship.”).) Finally, Figure 1 plots three datapoints, whereas Figure 2 plots four datapoints. (Tr. 72:22-23 (Bhatt).) But as in Figure 1, Figure 2's datapoints form a nearly straight line sloping upward. (Tr. 72:16-21 (Bhatt).) Dr. Felton understood that Figure 2 “is a summary of the dissolution data” for “four different release-controlling coating weight gains,” estimating that “the individual dissolution data was probably plotted, and the T80, the time for 80 percent release, was then calculated.” (Tr. 615:20-616:8.)

(Image Omitted) [('989 patent Fig. 2.)]

Both figures have equations representing the slope formula of the datapoints. The closer R2is to 1.0, the straighter the line. (Tr. 70:9-16 (Bhatt).) The nearly straight lines in both figures reflect the linear co-relationship between RC coating thickness and release time: the thicker the RC coating, the longer it takes to release 80% of the drug. (Tr. 69:19-21, 70:9-16 (Bhatt); see Tr. 616:9-14 (Felton) (“The higher the amount of polymer that was applied, the longer it took to get to the T80.”).) Indeed, the specification refers to Figures 1 and 2 as exemplifying that, “with all things being equal, there exists a mathematical relationship between the release controlling coating level among the cured beads and the 80% in vitro release time.” ('989 patent col. 7 ll. 28-47.) Critical in these experiments, the only parameter varying between the beads was the amount of RC coating. (Tr. 70:22-71:15 (Bhatt); Tr. 653:21-654:5 (Khan).) Thus, Supernus says, a POSA can use the linear equation to find the T80% of any amount of coating along the line. (Tr. 73:3-14 (Bhatt).)

The parties disagree about the usefulness of the figures' exemplifying the linear co-relationship. The disagreement manifested during a late-trial objection, which the Court will now resolve.

On the invalidity issue, the parties agreed to present evidence in the following order: Torrent calls Dr. Felton in its case-in-chief, Supernus calls Dr. Khan in rebuttal, and Torrent recalls Dr. Felton in reply. (ECF No. 133 ¶ 506.) During Torrent's case-in-chief, when Dr. Felton opined on the disconnect between Examples 1 and 6, she specifically mentioned the “mathematical models” in Figures 1 and 2 as follows:

MR. SUNG (ATTORNEY FOR TORRENT). And in addition to the lack of that information, was there anything else that you found in your analysis?

DR. FELTON. Well, there's other -- there's some other things related to the mathematical models that are used in figures 1 and 2. And the patent makes clear that those mathematical models which would allow a person of ordinary skill to predict how much coating I would need to apply to achieve a specific T80, those mathematical relationships don't hold for all of the compositions. So a person of ordinary skill would have to recreate, again, recreate these types of mathematical models. But again, without the dissolution methodology that was used, finding a T80 using these mathematical models is meaningless.

[(Tr. 603:06-18.)]

At that time, Dr. Felton did not elaborate on her opinion that “those mathematical relationships don't hold for all of the compositions.” In Supernus's rebuttal, Dr. Khan explained step by step how a POSA would use the patent to arrive at the formulations described in the Asserted Claims, once mentioning that “all the information is there very easily.” (See generally Tr. 637:7-646:3.) In reply, Torrent presented - for the first time - a demonstrative exhibit showing Dr. Felton's failed attempt at using the mathematical equation in Figure 1 to predict the T80% of five exemplary beads from Table 1. (Tr. 679:4-681:5.) The demonstrative was in Dr. Felton's binder of direct-examination exhibits, though it was not referenced in Torrent's case-in-chief.

Supernus objected, arguing that Dr. Felton's calculations went to the heart of Torrent's case-in-chief. (Tr. 681:6-682:13.) Torrent countered that it “opted not to do this on direct,” but that Dr. Felton's calculations rebutted Dr. Khan's opinion “that someone could easily follow from example 1 to example 6, and that it was simple for a POSA to be able to go through the entire patent.” (Tr. 681:13-16, 682:14-22, 683:12-17, 688:3-21.) Torrent told the Court that Dr. Khan's rebuttal testimony was “the first time that he ha[d] provided information from his perspective,” and that “prior to that point during the trial proceedings, there was no information that Dr. Khan was going to say that a person of ordinary skill in the art could easily walk from example 1 to example 6.” (Tr. 683:21-684:1.) Supernus responded that Torrent's representation was verifiably inaccurate based on Dr. Khan's expert report and deposition, Torrent's preparation and disclosure of the demonstrative in advance, and Supernus's pretrial disclosure of “the exhibits and everything we would be using on direct with Dr. Khan.” (Tr. 684:8-16.)

For expediency, the Court let Torrent finish questioning Dr. Felton about her calculations and let Supernus recall Dr. Khan in another rebuttal, reserving on both Supernus's objection to Torrent's reply-examination and Torrent's subsequent objection to the Court's modifying the order of evidence presentation. (Tr. 700:9-701:21.)

The spirit of the federal rules on rebuttal testimony applies to Torrent's reply-examination. The “purpose of rebuttal testimony is to explain, repel, counteract or disprove the evidence of the adverse party.” Vay v. Huston, 725 Fed.Appx. 112, 116 (3d Cir. 2018) (quoting United States v. Chrzanowski, 502 F.2d 573, 576 (3d Cir. 1974)). “At trial, rebuttal evidence is limited ‘to that which is precisely directed to rebutting new matter or new theories presented by the defendant's case-in-chief.'” Crowley v. Chait, 322 F.Supp.2d 530, 550 (D.N.J. 2004) (quoting Step-Saver Data Sys., Inc. v. Wyse Tech., 752 F.Supp. 181, 193 (E.D. Pa. 1990), aff'd in part, rev'd in part on other grounds, 939 F.2d 91 (3d Cir. 1991)). It is not “an opportunity for the correction of any oversights in the [party-proponent]'s case in chief.” Smithkline Beecham Corp. v. Ranbaxy Lab'ys, LTD., Civ. No. 03-2158, 2006 WL 8459047, at *4 (D.N.J. Feb. 23, 2006) (quoting Crowley, 322 F.Supp.2d at 551). “[E]vidence which properly belongs in the case-in-chief but is first introduced in rebuttal may be rejected, so as to avoid prejudice to the [party-opponent] and to ensure the orderly presentation of proof.” Vay, 725 Fed.Appx. at 117 (quoting Emerick v. U.S. Suzuki Motor Corp., 750 F.2d 19, 22 (3d Cir. 1984)).

The Court finds that Dr. Felton's calculations exceeded the purpose of Torrent's reply. The notion that Torrent could not anticipate, or that “there was no information” suggesting, what Dr. Khan would testify as to the connection between Examples 1 and 6 is undermined by the record. Dr. Khan's opinion was no surprise. Before trial, Torrent was on notice that Dr. Khan would testify that the specification (1) “teaches a POSA how to make and use the claimed inventions,” citing Example 1, and (2) “provides clear guidance on how to create sustained release formulations containing one or more XR components that deliver topiramate with the claimed pharmacokinetic parameters,” citing Example 6. (ECF No. 139 at 25-26.) That Dr. Khan used the word easily changes nothing.

What's more, Dr. Felton's calculations are not purposed merely to rebut Dr. Khan's testimony. They are the entire basis for her opinion that “those mathematical relationships don't hold for all of the compositions” - which Torrent elicited in its case-in-chief. The “punchline,” Torrent argued, was that “based on the calculations from Dr. Felton, only one example of the 13 [beads] that are in table 1 . . . even fits example 6. And as a result of that, even the examples that [Supernus] points to as supporting their written description or enablement begins to collapse.” (Tr. 688:6-14.) In other words, Dr. Felton's calculations are the proof that undermines the usefulness of the Patents-in-Suit's examples and thus their enablement and written description. It was the “sum and substance of the evidence” that Torrent presented in its case-in-chief on invalidity. Step-Saver, 752 F.Supp. at 193.

Contrary to Dr. Felton's suggestion in her reply testimony, the equations in Figures 1 and 2 are not meant to apply universally to all 13 beads in Table 1. Indeed, changing an in-vitro dissolution parameter - namely, the RC coating material or pore former - would make “a similar graph, but the equation, the slope of the line will be different.” (Tr. 71:10-15 (Bhatt); see Tr. 659:18-660:3 (Khan) (testifying that changing the polymer will change the slope and intercept value but not the relationship).) In Supernus's surrebuttal, Dr. Khan testified that Dr. Felton had “taken a bead population that probably is belonging to that log population” in Figure 1 and “try[ed] to fit [it] into a non-log. So obviously those numbers will look very, very high.” (Tr. 706:15-19.) Dr. Felton's “fundamental problem,” as Dr. Khan put it, was that she tried to apply “the equation and the data” for one bead population to all 13 exemplary beads. (Tr. 706:20-25.) But in fact, each of the 13 exemplary beads requires its own set of in-vitro dissolution experiments. (See Tr. 704:25-705:4 (Khan) (“Each one of the columns [in Table 1] that we have, . . . we have to perform an experiment, that different coating thicknesses. Each one of them [is] different. So you can perform experiment for column one, get that, column two.”).

Thus, the takeaway from Figures 1 and 2 is not their particular datapoints or slope formula. Nor is it that a POSA could plug data from other experiments into the figures' equations to calculate other formulations' release times. It is that they exemplify (1) how to analyze in-vitro dissolution data to find beads' T80%, and (2) that the RC coating thickness or the presence of pore formers correlates with release time. So although many variables affect release time, the Patents-in-Suit teach that to modulate release time, a POSA need only modulate the amount of RC coating or use of a pore former, leaving all other parameters equal.

Based on the Patents-in-Suit's direction and evidence of a POSA's skillset, the Court draws the connection between Examples 1 and 6 as follows. Example 1 teaches how to prepare the beads. Figures 1 and 2 show that a POSA can modulate thickness of an RC coating material or a pore former to achieve a T80% that meets the conditions “for in-vitro dissolution” in Example 6. Upon completing in-vitro dissolution experiments and preparing single-bead XR formulations that satisfy those in-vitro conditions in Example 6, the POSA designates each formulation as the XR composition whose T80% range matches the formulation's T80%: XR1 (1.5 h<=T80%<=4 h), XR2 (5 h<=T80%<=8 h), or XR3 (8 h<T80%<=10 h) in Example 6. After that, Dr. Khan testified, the POSA can simply rely on the bead-designations' Tmax data “for a single initial dose in-vivo” provided in Example 6. (See Tr. 642:23-643:10 (Khan) (“[I]f I coat the beads, get the dissolution between five to ten hours, I know I get the Tmax value because the patent already shows me that thing.”).)

With these findings, the Court revisits Torrent's claim that the specification must describe the inventors' exact dissolution methodology for in-vitro testing. Torrent argues that “without specific information about the dissolution parameters and method that Supernus used to generate the in-vitro data in Figures 1 and 2, a POSA could not rely on that information to determine which formulations described in Example 1 might be acceptable under Example 6.” (ECF No. 171 at 37 (citing Tr. 602:25-603:5 (Felton)).) Torrent attacks Supernus's contention that “a POSA could look up the appropriate dissolution method,” flagging that neither Dr. Bhatt nor Dr. Khan “testified as to the actual method that should be used.” (Id. at 37-38 (citing PFF ¶ 54 (Pl.)).) Torrent argues that without the inventors' exact methodology, “the data in Figures 1 and 2 would be meaningless,” and a POSA could not follow or replicate the examples. (Id. at 38 (citing Tr. 602:25-603:5 (Felton)).)

It was incumbent upon Torrent, who bears the burden to prove invalidity, to make clear why the data would be “meaningless.” The Court is not satisfied that Torrent has done so. Torrent repeatedly cites Dr. Felton's refrain that using another dissolution methodology would produce “meaningless” data and “be like comparing apples and oranges.” For this assertion, Torrent offers only Dr. Felton's calculations showing that the equations in Figures 1 and 2 do not apply universally to all polymers - an undisputed fact that misses the point of the figures. Torrent presented nothing to show, for instance, in what ways the differences among dissolution methodologies make the results meaningless or untranslatable from Example 1 to Example 6, or that a POSA does not have access to the apparatuses that USP or FDA recommends for this type of in-vitro dissolution testing. And again, that different dissolution methods “have different parameters” (Tr. 602:14-19 (Felton)) matters less given the specification's showing that, with those parameters being equal, a POSA can modulate release time by modulating only the thickness of the RC coating or a pore former (see Tr. 655:16-22 (Khan) (asking “why would I want to know all the variables?” if modulating coating thickness alone enables a POSA to achieve the desired T80%)).

Therefore, Torrent's dissolution methodology and connection between Examples 1 and 6 challenges, which implicate the amount of direction presented, relative skill of a POSA, and predictability of the art factors, do not support a non-enablement finding.

Torrent likens this case to Amgen Inc. v. Sanofi.^[63] In Amgen, the United States Supreme Court rejected Amgen's “functional” claim to “an entire class of things” - there, antibodies - “defined by their function” rather than their “physical characteristics or chemical properties.” 598 U.S. at 602, 610, 613.^[64] Specifically, Amgen did not claim “any particular antibody described by amino acid sequence,” which is how a particular antibody is commonly identified; it claimed “‘the entire genus' of antibodies” that performed two particular functions. Id. at 600, 602 (citation and some internal quotation marks omitted). The challenge for Amgen, the Supreme Court said, was that its described methods for making other antibodies forced a scientist “to engage in ‘painstaking experimentation' to see what works.” Id. at 614 (citation omitted).

[63] 598 U.S. 594 (2023).

[64] Supernus distinguishes the functional claim in Amgen from the Asserted Claims here. (ECF No. 170 at 43.) Because the Court ultimately rejects Amgen's applicability for other reasons, it need not address the asserted distinction.

But the Supreme Court qualified its holding in two ways. First, a specification need not always “describe with particularity how to make and use every single embodiment within a claimed class.” Id. at 610-11. “For instance,” the Supreme Court said, “it may suffice to give an example (or a few examples) if the specification also discloses ‘some general quality . . . running through' the class that gives it ‘a peculiar fitness for the particular purpose.'” Id. (quoting Consol. Elec. Light Co v. McKeesport Light Co., 159 U.S. 465, 475 (1895)). Second, “a specification may call for a reasonable amount of experimentation to make and use a patented invention,” such as when “some small difference in the proportions” of the claimed ingredient are required. Id. at 611-12 (citing, e.g., Wood v. Underhill, 46 U.S. 1 (1847); Mins. Separation v. Hyde, 242 U.S. 261 (1916)).

This case is different from Amgen in two meaningful ways. First, unlike Amgen's claim, the Asserted Claims do not claim an “entire genus” of release-controlling coatings regardless of physical characteristics or chemical properties. They claim sustained release formulations of topiramate comprising an XR component with cellulosic or acrylic polymers. (See, e.g., '989 patent col. 21 ll. 1-8.) Recall, again, Dr. Rotella's testimony that “the term ‘acrylic' means something” - it “confer[s] structure.” (Tr. 483:1-5 (Rotella).) The XR component must also exhibit “a maximum plasma concentration of topiramate in vivo at 16 or more hours after a single dose.” (See, e.g., '989 patent col. 21 ll. 33-35.) In short, the Asserted Claims require more than Amgen's claim.

Second, the Patents-in-Suit do not require “painstaking experimentation to see what works” as Amgen's patent required. See Amgen, 598 U.S. at 614 (citations and quotation marks omitted). Torrent argues that a POSA cannot “predict the rate of release for a given formulation without running [his or her] own experiments.” (ECF No. 171 at 40.) Indeed, Dr. Khan testified that for in-vitro dissolution testing, “if I'm changing the polymer, I have to perform at least those three, four experiments to have that” curve. (Tr. 660:7-9.) But Dr. Khan added, “Once I have that curve, then I can predict anything within that. But I have to make my own curve.” (Tr. 660:9-11.) Thus, once a POSA performs the first in-vitro dissolution test, gathers the data, and creates the slope formula - the parties' experts agree that these steps are routine^[65] - the curve will tell the POSA how much RC coating will achieve a certain T80%. Using that information, the POSA can adjust the RC coating precisely as needed to achieve a desired release rate.

[65] (See Tr. 638:4-6 (Khan) (“[T]he coating process is extremely well known, and I think Dr. Felton also admitted that it's very easy, any -- it's a routine formulator will do it.”); Tr. 620:11-15 (Felton) (agreeing that “develop[ing] a dissolution test, coat[ing] beads with different weight gains, conduct[ing] the dissolution test, and analyz[ing] the data” “would be normal, everyday skills that a POSA could do”).)

Explaining the simplicity of this process, Dr. Khan testified as follows:

MR. WEEKS (ATTORNEY FOR TORRENT). And you would agree that you cannot predict what figures 1 and 2 would look like if you switched from using ethylcellulose to one of the Eudragit^®acrylic polymers, for example, correct?

DR. KHAN. I am not worried at all. I provide the coating, I do the dissolution, and I calculate the time for 80 percent dissolution. If that dissolution, if it is releasing faster, I provide more coating so that my T[8]0 time increases. And if I'm not satisfied with that, I provide even more coating. So in one experiment, I keep on increasing the coating until I get the dissolution time, so I'm not at all worried about other polymers not giving the data. Actually, I got the teaching from it, how to perform and how to coat. I will use that and I get my own data.

. . .

Q. The addition of a pore former significantly change the relationship between polymer weight gain and T80 dissolution time. You could not have predicted that in advance without running experiments, right?

A. So when I run experiment, I will run without pore former. If that dissolution is too slow, then I will add pore former. And once I add pore former, I know now in my experiments I need pore former. Then I start the process, right. So I will increase the coating weight gain, perform dissolution. And this time in my coating material, I have added a pore former excipient therein. But I perform the same experiment. Now I add a pore former.

[(Tr. 659:5-660:23.)]

And, once more, after making beads that satisfy the “in-vitro dissolution” conditions in Example 6, a POSA can rely on the “single initial dose in-vivo” data that correspond with the beads' in-vitro release time. (See '989 patent col. 18 ll. 46-55; see also Tr. 653:21-654:25 (Khan) (“What this patent did, even though there are so many variables there, they said, . . . focus on coating thickness. And that thickness will modulate their dissolution, and dissolution will get to your PK.”).)

Even if Torrent had shown that a POSA could not rely on the in-vivo data in Example 6, the Court is not convinced that performing an in-vivo study would be undue experimentation. Dr. Felton testified that in-vivo clinical studies are beyond a POSA's skillset, reasoning that even she has “never conducted a clinical trial” despite her qualifications exceeding the POSA construction. (Tr. 613:5-20.) This is true even if a POSA hired a provider to perform the in-vivo study, because, as Dr. Felton understands “based on the legal description” that she received from counsel, all that matters is whether the POSA has the skillset. (Tr. 613:21-614:12.) Yet Dr. Khan countered that clinical studies are not as burdensome as Dr. Felton describes them. He testified that contrary to Dr. Felton's testimony, “[y]ou don't need FDA permission” to conduct single dose in-vivo studies; you only need permission from your institutional review board, or “IRB.” (Tr. 632:18-633:15.) He also testified that his graduate students routinely conduct these studies. (Tr. 633:17-634:10.) Dr. Taft likewise testified that “this type of clinical bioavailability test is a routine test that can be executed by a person of ordinary skill.” (Tr. 169:23-25.) Faced with only these competing, equally credible opinions, the Court cannot rule that Torrent met the clear-and-convincing standard.

The Court finds that the Patents-in-Suit require some trial and error but not the type of “random trial-and-error discovery” that gave the Supreme Court pause. See Amgen, 598 U.S. at 615; see also Wyeth & Cordis Corp. v. Abbott Lab'ys, 720 F.3d 1380, 1386 (Fed. Cir. 2013) (“Even ‘a considerable amount of experimentation is permissible,' as long as it is ‘merely routine' or the specification ‘provides a reasonable amount of guidance' regarding the direction of experimentation.” (quoting Johns Hopkins Univ. v. CellPro, Inc., 152 F.3d 1342, 1360-61 (Fed.Cir. 1998))).^[66] Therefore, the fact that a POSA must perform in-vitro dissolution testing, which implicates the quantity of experimentation, relative skill of a POSA, and predictability of the art factors, does not favor a non-enablement finding.

[66] That Torrent took seven years to develop its ANDA Products (PFF ¶ 53, 569 (Def.)) is not alone dispositive. See Amgen, 598 U.S. at 615 (“[E]nablement is not measured against the cumulative time and effort it takes to make every embodiment within a claim.”). Though Torrent presented evidence as to what could delay a POSA, Torrent never explained what actually took it so long to develop its ANDA Products.

Having found none of Torrent's challenges persuasive, the Court rules that Torrent has not shown by clear and convincing evidence that the Patents-in-Suit are not enabling.

D. Written Description

Under § 112(a), a patent's “description must clearly allow persons of ordinary skill in the art to recognize that the inventor invented what is claimed.” Ariad Pharms., Inc. v. Eli Lilly & Co., 598 F.3d 1336, 1351 (Fed. Cir. 2010) (internal citation, quotation marks, and brackets omitted). “To satisfy the written description requirement, a patent's specification must ‘reasonably convey[ ] to those skilled in the art that the inventor had possession of the claimed subject matter as of the filing date.'” Novartis Pharms. Corp. v. Accord Healthcare, Inc., 38 F.4th 1013, 1016 (Fed. Cir. 2022) (quoting Ariad Pharms., 598 F.3d at 1351). “[W]hether a patent complies with the written description requirement will necessarily vary depending on the context. Specifically, the level of detail required . . . varies depending on the nature and scope of the claims and on the complexity and predictability of the relevant technology.” Immunex Corp. v. Sandoz Inc., 395 F.Supp.3d 366, 380 (D.N.J. 2019), aff'd, 964 F.3d 1049 (Fed. Cir. 2020) (quoting Ariad Pharms., 598 F.3d at 1351). The party challenging a patent's written description must do so through clear and convincing evidence. Invitrogen Corp. v. Clontech Lab'ys, Inc., 429 F.3d 1052, 1072 (Fed. Cir. 2005).

Much of the Court's enablement discussion also applies to Torrent's written-description challenge. (See ECF No. 171 at 42 (“The same deficiencies in the asserted patent specifications described above regarding enablement, compel a similar result regarding inadequate written description.”).) For Figures 1 and 2, for instance, Torrent asserts its failed dissolution methodology challenge. (See PFF ¶¶ 425-442 (Def.).) The Court will discuss only new arguments.

Torrent argues that the Patents-in-Suit do not show that the in-vitro mathematical relationship would hold for any formulation other than the RC coating materials used in Figures 1 and 2. (ECF No. 171 at 44.) Yet Torrent does not contest that the patents claim that the relationship would hold for each exemplary bead described in Table 1. Compare Regents of the Univ. of California v. Eli Lilly & Co., 119 F.3d 1559, 1567-69 (Fed. Cir. 1997) (rejecting a description of “a process for obtaining human insulin-encoding cDNA” that lacked any “sequence information indicating which nucleotides constitute human cDNA” or “further information in the patent pertaining to that cDNA's relevant structural or physical characteristics”), Bos. Sci. Corp. v. Johnson & Johnson, 647 F.3d 1353, 1364 (Fed. Cir. 2011) (rejecting a description that included “no examples of macrocyclic lactone analogs of rapamycin” (the claimed genus) and generally “no guidance on how to properly determine whether a compound is a macrocyclic lactone analog of rapamycin”), with Ajinomoto, 932 F.3d at 1361 (accepting a description that expressly provided four examples of “more potent promoters” where a POSA could “make relatively predictable changes to the native promoter to arrive at a more potent promoter”).

Torrent also argues that “there is no support in the specification showing that a single-component formulation could be used to achieve a target predetermined release profile.” (ECF No. 171 at 45.) Not so. In Figure 2, which shows results of in-vitro dissolution testing with single-bead formulations, each of the four plotted data points meet the XR1, XR2 or XR3 conditions set forth in Example 6: (i) about a 2% release coating had about a 2.8 hour T80% time; (ii) about a 3% release coating had about a 3.2 hour T80% time; (iii) about a 5.5% release coating had about a 5.5 hour T80% time; and (iv) about a 9.5% release coating had about a 9.5 hour T80% time.^[67] ('989 patent Fig. 2, Example 6.) Compare these release rates with Example 6's conditions “for in-vitro dissolution”: for XR1, 1.5 h<=T80%<=4 h; for XR2, 5 h<=T80%<=8 h; for XR3, 8 h<T80%<=10 h. And in Figure 3, which shows results of in-vivo studies involving single-bead formulations, the XR2 and XR3 beads each exhibit a Tmax of about 24 hours, satisfying the claimed Tmax of 16 or more hours after a single initial dose. ('989 patent Fig. 3, Example 6; Tr. 85:1-5 (Bhatt).)

[67] The Court overrules Torrent's objection that Supernus's noting these datapoints violates FRE 702's requirements. (See PFF ¶ 423 (Def.).) The data are in the specification, which is part of the intrinsic record that the Court can review without the help of expert testimony.

Finally, contrary to Torrent's assertion, a patent's disclosing a handful of examples for far-reaching claims is not in itself improper. (See ECF No. 171 at 45.) See Ariad Pharms., 598 F.3d at 1351 (“Nor do we set out any bright-line rules governing, for example, the number of species that must be disclosed to describe a genus claim, as this number necessarily changes with each invention, and it changes with progress in a field.”). “A written description of an invention involving a chemical genus, like a description of a chemical species, requires a precise definition, such as by structure, formula, [or] chemical name of the claimed subject matter sufficient to distinguish it from other materials.” Idenix Pharms. LLC v. Gilead Scis. Inc., 941 F.3d 1149, 1164 (Fed. Cir. 2019) (quoting Bos. Sci., 647 F.3d at 1363) (internal quotation marks omitted) (alterations in Bos. Sci.). Here, the Patents-in-Suit satisfy this mandate. Recall, once more, Dr. Rotella's testimony that “the term ‘acrylic' means something” - it “confer[s] structure.” (Tr. 483:1-5 (Rotella).)

As none of Torrent's written-description arguments is availing, the Court rules that Torrent has not shown by clear and convincing evidence that the Patents-in-Suit lack adequate written description.

V. CONCLUSIONS OF LAW

For these reasons, the Court rules that Torrent's AND A Products INFRINGE on the Asserted Claims of the Patents-in-Suit, and that the Patents-in-Suit are VALID. An appropriate Order follows.

DENIED.