CORCEPT THERAPEUTICS, INC. v. TEVA PHARMACEUTICALS USA, INC.

IN THE UNITED STATES DISTRICT COURT

FOR THE DISTRICT OF NEW JERSEY

CAMDEN VICINAGE

CORCEPT THERAPEUTICS, INC.,

Plaintiff,

Civil Action No. 18-03632

v. (RMB) (LDW) (consolidated)

TEVA PHARMACEUTICALS OPINION

USA, INC.,

Defendant.

APPEARANCES:

Charles M. Lizza

William C. Baton

Sarah A. Sullivan

Alexander L. Callo

SAUL EWING LLP

One Riverfront Plaza, Suite 1520

Newark, New Jersey 07102-5426

F. Dominic Cerrito (pro hac vice)

Eric C. Stops (pro hac vice)

Evangeline Shih (pro hac vice)

Daniel C. Wiesner (pro hac vice)

John P. Galanek (pro hac vice)

Nicholas A. LoCastro

QUINN EMANUEL URQUHART & SULLIVAN, LLP

51 Madison Avenue, 22nd Floor

New York, New York 10010

On behalf of Plaintiff Corcept Therapeutics, Inc.

Liza M. Walsh

Christina I. Gannon

Jessica K. Formichella

Lauren R. Malakoff

WALSH PIZZI O’REILLY FALANGA LLP

Three Gateway Center

100 Mulberry Street, 15th Floor

Newark, New Jersey 07102

J.C. Rozendaal (pro hac vice)

Uma N. Everett (pro hac vice)

William H. Milliken (pro hac vice)

Anna G. Phillips (pro hac vice)

Brady P. Gleason (pro hac vice)

Brooke N. McLain (pro hac vice)

STERNE, KESSLER, GOLDSTEIN & FOX, P.L.L.C.

1101 K Street NW, 10th Floor

Washington, D.C. 20005

On behalf of Defendant Teva Pharmaceuticals USA, Inc.

RENÉE MARIE BUMB, Chief United States District Judge:

This is a “Hatch-Waxman Act” case for patent infringement brought by

Plaintiff Corcept Therapeutics, Inc. (“Corcept”) against Defendant Teva

Pharmaceuticals USA, Inc. (“Teva”) pursuant to 35 U.S.C. § 271(b) and 271(e)(2)(A).

Teva seeks to market and sell a generic version of Corcept’s 300 mg branded drug,

Korlym® (mifepristone), prior to the expiration of certain Corcept patents. Following

Teva’s submission of Abbreviated New Drug Application No. 211436 to the United

States Food and Drug Administration for commercial approval of its generic drug

product, Corcept initiated this suit seeking, inter alia, a declaration of patent

infringement and a permanent injunction restraining Teva from launching its generic

mifepristone product prior to the expiration of the Patents-in-Suit (defined below).

On February 27, 2023, the Court denied the parties’ cross-motions for summary

judgment, observing that it was critical to hear “expert testimony regarding [Teva’s]

label’s dosing regimens available when healthcare providers co-administer

mifepristone with strong CYP3A inhibitors.” [Docket No. 229.] This action

proceeded to a three-day bench trial on September 26, 2023. [Transcript of Bench

Trial, Docket Nos. 284, 285, 286 (“Tr.”).] The sole issue presented to the Court was

whether Teva induced infringement of claims 10–13 of United States Patent No.

10,195,214 (“the ’214 Patent”) and claims 1, 6, 7, and 9 of United States Patent No.

10,842,800 (“the ’800 Patent”) (the “Patents-in-Suit”). [See generally Def.’s Pretrial

Br., Docket No. 269; Pl.’s Pretrial Br., Docket No. 270.] Claim 10 of the ’214 Patent

and claims 1 and 6 of the ’800 Patent are the only independent claims asserted. The

Court heard evidence, including live testimony from the parties’ respective expert

witnesses,1 and thereafter directed the submission of post-trial briefing.2

Having considered the parties’ briefs, the evidence of record, and the applicable

law, the Court concludes that Corcept has not met its burden of proof: it has not

demonstrated that there is a likelihood of direct infringement of the asserted claims in

the future because it has not established that physicians are likely to coadminister

mifepristone with a strong CYP3A inhibitor at an infringing sequence and/or dosage

now or in the future. Indeed, Corcept failed to introduce credible record evidence that

anyone has ever previously infringed the asserted claims. See Genentech, Inc. v. Sandoz

Inc., 55 F.4th 1368, 1380 (Fed. Cir. 2022). Additionally, Corcept failed to demonstrate

that Teva possesses the specific intent to encourage physicians to infringe the asserted

claims because Teva’s proposed product label does not clearly “encourage,

recommend, or promote” infringement. See Takeda Pharms. U.S.A., Inc. v. West-Ward

Pharm. Corp., 785 F.3d 625, 631 (Fed. Cir. 2015). Accordingly, the Court will enter

1 The Court heard live testimony from Dr. Joseph Belanoff (co-founder, CEO,

and President of Corcept, and the inventor associated with the Patents-in-Suit); Dr. Ty

Carroll (Corcept’s expert witness); and Dr. Peter Snyder (Teva’s expert witness). The

Court also heard deposition testimony via video from Dr. Adrian Dobs, Mr. Gregg

DeRosa, Mr. Craig Jones, Mr. Sean Maduck, and Dr. Andreas Moraitis.

2 The parties submitted their briefs accordingly. [Def.’s Post-Trial Br., Docket

No. 288; Pl.’s Post-Trial Br., Docket No. 289; Def.’s Post-Trial Reply Br., Docket No.

294; Pl.’s Post-Trial Reply Br., Docket No. 295.]

judgment in Teva’s favor. This Opinion constitutes the Court’s findings of fact and

conclusions of law pursuant to Federal Rule of Civil Procedure 52(a).

I. BACKGROUND

A. The Parties, Standing, and Jurisdiction.

Corcept is a publicly traded corporation organized under the laws of the State

of Delaware, with a principal place of business in Menlo Park, California. [Final Pre-

Trial Order (“FPTO”), Tab 2, Stipulated Facts ¶ 2, Docket No. 273.] Teva is a

corporation organized under the laws of the State of Delaware, with a principal place

of business in Parsippany, New Jersey. [Id. ¶ 1.] (Teva’s global headquarters is located

in Tel Aviv, Israel.)

As the assignee of the Patents-in-Suit, [Tr. at 127:4–10 (Dr. Belanoff) (citing

PTX-001, PTX-003)], Corcept has standing to maintain this litigation. See Enzo APA

& Son, Inc. v. Geapag A.G., 134 F.3d 1090, 1093 (Fed. Cir. 1998) (“[A]n assignee is the

patentee and has standing to bring suit for infringement in its own name.”) (citing 35

U.S.C. § 100(d)).

Because this action arises under the patent laws of the United States, this Court

exercises subject matter jurisdiction pursuant to 28 U.S.C. §§ 1331, 1338(a), 2201, and

2202.

B. The Hatch-Waxman Act.

Under the Federal Food, Drug, and Cosmetic Act (“FDCA”), 21 U.S.C. § 301,

et seq., the United States Food and Drug Administration (“FDA”) must approve a new

drug prior to its introduction on the market. 21 U.S.C. § 355(a). To secure the FDA’s

approval, a drug applicant may file a New Drug Application (“NDA”) that includes

certain specified information, including examples of the proposed label for the drug

and any patents that claim the drug, or a method of using the drug, if an infringement

claim could reasonably be asserted. Id. § 355(b)(1)(A)(vi), (viii). “The FDA publishes

the names of approved drugs and their associated patent information in the Approved

Drug Products with Therapeutics Equivalence Evaluations list, commonly referred to as the

‘Orange Book.’ ” AstraZeneca LP v. Apotex, Inc., 633 F.3d 1042, 1045 (Fed. Cir. 2010).

Where a drug applicant seeks the FDA’s approval to introduce a generic version

of an approved drug, the drug applicant may file an Abbreviated New Drug

Application (“ANDA”). 21 U.S.C. § 355(j). An ANDA is a more streamlined

submission than an NDA, allowing the applicant to rely on the safety and efficacy

information of a drug listed in the Orange Book if the applicant can demonstrate

“bioequivalence.” See AstraZeneca, 633 F.3d at 1045; see also 21 U.S.C. § 355(j)(2)(A).

Congress permitted the submission of ANDAs, rather than full NDAs, under the

Hatch-Waxman Act,3 as part of a deliberate and carefully constructed scheme to

balance competing policy priorities in the pharmaceuticals industry. On the one hand,

3 The “Hatch-Waxman Act” is the more common referent for the Drug Price

Competition and Patent Term Restoration Act of 1984, Pub. L. No. 98-417, 98 Stat.

1585 (codified at 21 U.S.C. §§ 355, 360(c), 35 U.S.C. §§ 156, 271, 282), as amended

by the Medicare Prescription Drug Improvement and Modernization act of 2003, Pub.

L. No. 108-173, 117 Stat. 2066. Sometimes the law is referred to as the “Hatch

Waxman Amendments” to the FDCA. See, e.g., Andrx Pharms., Inc. v. Biovail Corp.,

276 F.3d 1368, 1370–71 (Fed. Cir. 2002).

Congress sought to encourage “pioneering research and development of new drugs,”

while on the other hand, “enabling competitors to bring low-cost, generic copies of

those drugs to market.” Andrx Pharms., Inc. v. Biovail Corp., 276 F.3d 1368, 1371 (Fed.

Cir. 2002); see also Eli Lilly & Co. v. Medtronic, Inc., 496 U.S. 661, 676 (1990) (explaining

that Congress authorized ANDAs, substantially shortening the time and effort needed

to obtain marketing approval, to eliminate the de facto extension of a patent’s term, and

to enable new drugs to be marketed more quickly and cheaply).

Under a safe harbor provision, it is not an act of patent infringement to research,

develop, test, and collect information for submission of a drug application to the FDA.

See 35 U.S.C. § 271(e)(1) (“It shall not be an act of infringement to make, use, offer to

sell, or sell within the United States or import into the United States a patented

invention . . . solely for uses reasonably related to the development and submission of

information under [applicable law].”); Merck KGaA v. Integra Lifesciences I, Ltd., 545

U.S. 193, 202 (2005); Glaxo, Inc. v. Novopharm, Ltd., 110 F.3d 1562, 1568 (Fed. Cir.

1997). Under § 271(e)(2)(A), however, it is an act of patent infringement for a generic

drug applicant to file an ANDA seeking approval of a generic drug product claimed

by a valid and unexpired patent. Glaxo Grp. Ltd. v. Apotex, Inc., 376 F.3d 1339, 1351

(Fed. Cir. 2004); Andrx Pharms., 276 F.3d at 1371. This provision, based on a

somewhat “artificial” act of infringement, has been considered a “jurisdictional-

conferring statute” invoked to establish a case or controversy in a declaratory judgment

action. Glaxo Grp., 376 F.3d at 1351 (citing Allergan, Inc. v. Alcon Labs., 324 F.3d 1322,

1330 (Fed. Cir. 2003)); see also Eli Lilly & Co., 496 U.S. at 676 (explaining that

§ 271(e)(2) is based on a “somewhat artificial[] act of infringement for a very limited

and technical purpose”).

As part of the approval process, an ANDA applicant must file one of four

certifications concerning every patent that claims the listed drug that is the subject of

the application, or a method of using the drug: (I) “that such patent information has

not been filed,” (II) “that such patent has expired,” (III) “the date on which such patent

will expire,” or (IV) “that such patent is invalid or will not be infringed by the

manufacture, use, or sale of the new drug for which the application is submitted.” 21

U.S.C. § 355(j)(2)(A)(vii)(I)–(IV). An applicant who files a “Paragraph IV”

certification must give notice to the holder of the patents at issue that an ANDA has

been filed, and the applicant must state the basis for its belief that the patents are not

valid or will not be infringed. Id. § 355(j)(2)(B)(i), (ii). The ANDA becomes effective

45 days thereafter unless the patentee commences suit. Id. § 355(c)(3)(C). Subject to

certain exceptions, if a patent infringement suit is commenced, then the ANDA will

not become effective until the expiration of an automatic 30-month stay. Id.; see also

Eli Lilly & Co., 496 U.S. at 677–78 (explaining certification process).

C. Cushing’s Syndrome and Treatment.

Relevant to the dispute before the Court is the rare and debilitating endocrine

disorder, Cushing’s syndrome. Endogenous Cushing’s syndrome is characterized by

excessive production of the adrenal hormone cortisol. [Tr. at 120:15–16 (Dr.

Belanoff); Tr. at 382:1–3 (Dr. Snyder); see also Tr. at 52:18–53:2 (Dr. Dobs)

(differentiating endogenous and exogenous Cushing’s syndrome, the latter signifying

a cause outside of the body).] Endogenous Cushing’s syndrome often results from the

presence of a tumor in the adrenal or pituitary glands that stimulates, either directly or

indirectly, the overproduction of cortisol. [Tr. at 120:16–19 (Dr. Belanoff); Tr. at

382:3–8 (Dr. Snyder).] As Dr. Snyder put it, excess cortisol is “really bad.” [Tr. at

382:12.] It can lead to hyperglycemia (i.e., high blood sugar), hypertension, obesity,

osteoporosis, depression, sexual dysfunction, menstrual disorders, and infertility,

among other problems and complications. [Tr. at 382:11–17 (Dr. Snyder); see also, e.g.,

Rosario Pivonello et al., The Treatment of Cushing’s Disease, Endocrine Rev., Aug. 2015,

at 385, 386 (describing manifestations of Cushing’s syndrome) (PTX-015.01).]

The first-line treatment for a patient suffering from Cushing’s syndrome is

surgery. [Tr. at 53:15–54:11 (Dr. Dobs); Tr. at 120:22–25 (Dr. Belanoff); Tr. at

382:21–24 (Dr. Snyder).] Excising the “offending tumor” is preferred for multiple

reasons, not least of which is that it is the most efficient option from a medical and

financial “point of view.” [Tr. at 53:21–54:1 (Dr. Dobs); see also Tr. at 382:23–24 (Dr.

Snyder).] A majority of those who suffer from hypercortisolism undergo surgery and

are cured. [Tr. at 236:12–18 (Dr. Carroll); see Tr. at 120:24–121:4 (Dr. Belanoff)

(explaining that surgery is only possible “about half the time”).] However, if surgical

intervention is not an option for the patient or surgery has already yielded an

unsuccessful result, physicians will consider medical therapies (i.e., medication). [Tr.

at 382:25–383:3 (Dr. Snyder); Tr. at 54:22–55:3 (Dr. Dobs); see also, e.g., Pivonello et

al., supra, at 396 (Fig. 2) (depicting schematic algorithm for treating Cushing’s disease,

prioritizing surgery followed by medical therapy) (PTX-015.12).]

Today, there are a few medications available to treat Cushing’s syndrome when

surgery is not an option. [See Tr. at 55:24–56:2 (Dr. Dobs).] One group of medications

that is often prescribed—called “CYP3A inhibitors”—impedes the synthesis of cortisol

in the adrenal glands, thereby “blocking” excessive cortisol at its source. This group

includes ketoconazole and metyrapone, among others. [Tr. at 383:6–12 (Dr. Snyder).]

CYP3A inhibitors are a broad therapeutic class of drugs that target multiple different

diseases, often addressing specific manifestations of Cushing’s syndrome. [See Tr. at

129:6–131:6 (Dr. Belanoff) (explaining that CYP3A inhibitors, such as ketoconazole,

itraconazole, nefazodone, ritonavir, nelfmavir, and clarithromycin, are often the “best

choice” for addressing a particular manifestation of Cushing’s syndrome).] However,

these medications also suppress the activity of CYP3A enzymes, which are the

enzymes that metabolize most drugs. [See Tr. at 127:19–24 (Dr. Belanoff).] As a

result, there is the potential for adverse drug-drug interactions. [See Tr. at 127–28 (Dr.

Belanoff); see also generally Lynette K. Nieman, Medical Therapy of Hypercortisolism

(Cushing’s Syndrome), UpToDate (last updated Aug. 29, 2022) [DTX-034]).]

Another medication is mifepristone, which “antagonizes the effects of cortisol,”

[Tr. at 121:5–6 (Dr. Belanoff)], by blocking cortisol from binding to a glucocorticoid

receptor, [Tr. at 383:9–12 (Dr. Snyder); see generally Pivonello et al., supra, at 455–58

(PTX-015.71–74)]. Unlike a CYP3A inhibitor that prevents production of cortisol,

mifepristone manages cortisol’s effects. [See, e.g., Tr. at 121:5–6 (Dr. Belanoff).] The

major benefit of mifepristone is that it acts quickly and can be used to stabilize a very

sick patient, while that patient waits for surgery, for instance. [Tr. at 385:6–13 (Dr.

Snyder); Tr. at 57:15–58–12 (Dr. Dobs).] The major drawback, however, is inherent

in mifepristone’s function: by blocking cortisol’s “action,” mifepristone does not

actually inhibit cortisol production. Cortisol levels can thus increase even though

mifepristone is being used. This can lead to dosing problems because a prescribing

physician cannot use cortisol levels to determine the right dose. [See Tr. at 385–86 (Dr.

Snyder).]

Mifepristone is rarely, if ever, prescribed in combination with a strong CYP3A

inhibitor, such as ketoconazole, to treat Cushing’s syndrome. [See, e.g., Tr. at 58:19–

25 (Dr. Dobs) (explaining generally that she has prescribed mifepristone with

ketoconazole only “about two to three times” where a patient was “extremely ill”); Tr.

at 410:9–22 (Dr. Snyder) (noting that he has never co-administered mifepristone with

a strong CYP3A inhibitor and “can’t think of any situation where it would be

necessary,” where the benefits would outweigh the risk).] As will be discussed herein,

co-administration can be risky because it can increase mifepristone levels in the blood

and cause resulting health consequences, as strong CYP3A inhibitors block CYP3A

enzymes from metabolizing mifepristone. [Tr. at 212:1–7 (Dr. Carroll); Tr. at 127–28,

131 (Dr. Belanoff).]

In this regard, Dr. Snyder testified that a newer drug, osilodrostat, is now being

used to treat Cushing’s syndrome. [Tr. at 384, 386, 393 (Dr. Snyder).] It is a powerful

drug that acts quickly by inhibiting the synthesis of cortisol. [Tr. at 386 (Dr. Snyder).]

While osilodrostat’s branded version (Isturisa®) is only indicated for the treatment of

Cushing’s disease, Dr. Snyder testified that it is now widely used off-label for the

treatment of Cushing’s syndrome. [Tr. at 474, 476 (Dr. Snyder).] Its principal

advantage in this context is that physicians can rely on cortisol levels to determine the

proper dose. [Tr. at 384, 386 (Dr. Snyder).]

D. Korlym® and Discovery of the Claimed Invention.

In the years after Corcept was founded (in 1998), Dr. Belanoff and the company

focused on developing a treatment for psychotic depression. [Tr. at 117–120 (Dr.

Belanoff).] Because of cortisol’s systemic bodily effect, Dr. Belanoff was particularly

interested in, and focused Corcept’s research and development activities on, cortisol

modulation. [Tr. at 120 (Dr. Belanoff).] Corcept became aware that researchers had

reported the successful treatment of Cushing’s syndrome with mifepristone—a drug

that had been around since 1985—but that there was not an FDA-approved

medication on the market. [Tr. at 120–21, 123, 157 (Dr. Belanoff).] In 2007, Corcept

filed an Investigational New Drug Application to evaluate a mifepristone product for

the treatment of Cushing’s syndrome, and it soon designed a clinical study that

produced “potent and positive” results. [Tr. at 122 (Dr. Belanoff).] Corcept named

its efficacy study “SEISMIC.” [Tr. at 121–22 (Dr. Belanoff).] These results supported

Corcept’s NDA No. 202107 for approval of its product. [Tr. at 122 (Dr. Belanoff).]

In 2012, the FDA approved Corcept’s mifepristone product, Korlym®, for the

following indication: “for the control of hyperglycemia secondary to hypercortisolism

in adult patients with endogenous Cushing’s syndrome who have type 2 diabetes

mellitus or glucose intolerance and have failed surgery or are not candidates for

surgery.” [FPTO, Tab 2, Stipulated Facts ¶¶ 31, 33; see Tr. at 123:9–12 (Dr. Belanoff);

JTX-015.] When the drug was approved, it qualified for orphan status, with an

estimated market of 5,000 patients. [Tr. at 123:15–18 (Dr. Belanoff).] More than ten

years on, Korlym® is today prescribed to about 1,500 patients. [Tr. at 123:17–18 (Dr.

Belanoff).]

In connection with the product’s initial approval, the FDA required Corcept to

conduct drug-drug interaction studies to determine the effect of coadministration of

mifepristone with a strong CYP3A inhibitor. [Tr. at 136–37 (Dr. Belanoff).] The

FDA’s directive is reflected as a post-marketing requirement in its initial approval

letter. [PTX-012.3 (approval letter).] The FDA required post-marketing studies to be

conducted because it recognized that ketoconazole (an exemplary CYP3A inhibitor)

was widely used off-label to treat Cushing’s syndrome and that increased exposure to

mifepristone might result from coadministration, potentially leading to severe

hypokalemia and adrenal insufficiency. [Tr. at 136–39 (Dr. Belanoff); see PTX-012.3–

4.] Dr. Belanoff agreed with the FDA’s directive to conduct “rigorous” and “robust”

studies to quantify the effects of coadministration on mifepristone blood levels, as he

was fearful that concomitant use could lead to severe effects, including death. [Tr. at

131:20–25, 137:13–139:4.] He also testified that there was an awareness at the time of

approval that physicians were already administering mifepristone with a strong

CYP3A inhibitor in very sick patients in isolated cases. [Tr. at 132:6–11, 138:14–20;

PTX-012.4.] According to Dr. Belanoff and Dr. Carroll, doctors desired to

coadminister the drugs but avoided doing so, unless medically necessary, due to safety

concerns. [Tr. at 143:13–144:12, 146:8–22 (Dr. Belanoff); Tr. at 211:8–11, 211:21–

212:7 (Dr. Carroll).]

Corcept’s initial 2012 Korylm® label was approved with the following warnings

regarding coadministration of mifepristone and strong CYP3A inhibitors:

• Warnings & Precautions – Use of Strong CYP3A Inhibitors:

Concomitant use can increase mifepristone plasma levels

significantly. Use only when necessary and limit mifepristone dose

to 300 mg (5.6). [JTX-015.1.]

• Drug Interactions – CYP3A inhibitors: Caution should be used when

Korlym is used with strong CYP3A inhibitors. Limit mifepristone

dose to 300 mg per day when used with strong CYP3A inhibitors.

(5.6, 7.2). [JTX-015.1.]

• 5.6 Use of Strong CYP3A Inhibitors – Korlym should be used with

extreme caution in patients taking ketoconazole and other strong

inhibitors of CYP3A, such as itraconazole, nefazodone, ritonavir,

nelfinavir, indinavir, atazanavir, amprenavir, fosamprenavir,

boceprevir, clarithromycin, conivaptan, lopinavir, nefazodone,

posaconazole, ritonavir, saquinavir, telaprevir, telithromycin, or

voriconazole, as these could substantially increase the concentration

of mifepristone in the blood. The benefit of concomitant use of these

agents should be carefully weighed against the potential risks.

Mifepristone should be used in combination with strong CYP3A

inhibitors only when necessary, and in such cases the dose should be

limited to 300 mg per day. [See Warnings & Precautions (5.4), Drug

Interactions (7.2), and Clinical Pharmacology (12.3).] [JTX-015.6.]

• 7.2 CYP3A Inhibitors – Medications that inhibit CYP3A could

increase plasma mifepristone concentrations and dose reduction of

Korlym may be required. [¶] Ketoconazole and other strong inhibitors

of CYP3A, such as itraconazole, nefazodone, ritonavir, nelfinavir,

indinavir, atazanavir, amprenavir and fosamprenavir, boceprevir,

clarithromycin, conivaptan, lopinavir, mibefradil, nefazodone,

posaconazole, ritonavir, saquinavir, telaprevir, telithromycin, or

voriconazole may increase exposure to mifepristone significantly.

The clinical impact of this interaction has not been studied. Therefore,

extreme caution should be used when these drugs are prescribed in

combination with Korlym. The benefit of concomitant use of these

agents should be carefully weighed against the potential risks. The

dose of Korlym should be limited to 300 mg and used only when

necessary. [See Warnings & Precautions (5.6)]. [JTX-015.9–10.]

[JTX-015 (Corcept’s 2012 Korlym® Label).] Dr. Belanoff testified that the 2012 label

represented a “forceful[]” warning to physicians not to coadminister mifepristone with

strong CYP3A inhibitors. [See Tr. at 145–48, 149–50 (Dr. Belanoff) (citing prior

declaration to Patent Office, characterizing 2012 product label as “forcefully” warning

that coadministration might be dangerous).]

In the years that followed, as required, Corcept conducted a series of clinical

trials under a variety of conditions, hypothesizing that mifepristone blood levels would

“skyrocket” if mifepristone and a strong CYP3A inhibitor were taken together. [Tr.

at 128, 131 (Dr. Belanoff).] Rather, Corcept discovered that mifepristone blood levels

only increased marginally and that a physician could prescribe up to 900 mg of

mifepristone with a strong CYP3A inhibitor if dosed appropriately. [Tr. at 128–31,

151–55 (Dr. Belanoff).] Based on this discovery, Corcept filed a patent application in

March 2017 that led to the issuance of the ’214 Patent and ’800 Patent. Corcept also

asked the FDA to approve modifications to its original product label. [Tr. at 151 (Dr.

Belanoff).] The current label, revised in November 2019, includes the following

language regarding coadministration (with modifications noted in-line):

• Warnings and Precautions – Use of Strong CYP3A Inhibitors:

Concomitant use can increase mifepristone plasma levels

[significantly]. Use only when necessary and limit mifepristone dose

to 900 mg (5.6). [JTX-016.1 (Corcept’s 2019 Korlym® Label)

(strikethrough and emphasis added).]

e Drug Interactions —CYP3A inhibitors: Caution should be used when

KORLYM 1s used with strong CYP3A inhibitors. Limit mifepristone

dose to 900 mg per day when used with strong CYP3A inhibitors (7.2).

[J TX-016.1 (emphasis added).]|

e 2.5 Concomitant Administration with CYP3A Inhibitors -—

Ketoconazole and other strong inhibitors of CYP3A, such as

itraconazole, nefazodone, ritonavir, nelfinavir, indinavir, atazanavir,

amprenavir and fosamprenavir, clarithromycin, conivaptan,

lopinavir/ritonavir, posaconazole, saquinavir, telithromycin, or

voriconazole may increase exposure to mifepristone. KORLYM

should be used in combination with strong CYP3A inhibitors only

when necessary. [See Warnings and Precautions (5.6), Drug Interactions

(7.2).

Administration of KORLYM to patients already being treated with strong

CYP3A inhibitors:

e Start at a dose of 300 mg. If clinically indicated, titrate to a

maximum of 900 mg.

Administration of strong CYP3A inhibitors to patients already being treated

with KORLYM:

e Adjust the dose of KORLYM according to Table 1.

Table 1. Dose adjustment of KORLYM when strong CYP3A inhibitor is added

Current dose of Adjustment to dose of KORLYM

600 mg Reduce dose to 300 mg. If clinically indicated,

toamasinan omg

titrate to a maximuim of 900 mg

[JTX-016.4 (new instruction regarding coadmuinistration). |

e 5.6 Use of Strong CYP3A Inhibitors - KORLYM should be used

with [extreme] caution in patients taking ketoconazole and other

strong inhibitors of CYP3A, such as itraconazole, nefazodone,

ritonavir, nelfinavir, indinavir, atazanavir, amprenavir,

fosamprenavir, clarithromycin, conivaptan, lopinavir/ritonavir,

posaconazole, saquinavir, telithromycin, or voriconazole, as these

could [substantially] increase the concentration of mifepristone in the

blood. The benefit of concomitant use of these agents should be

16

carefully weighed against the potential risks. KORLYM should be

used in combination with strong CYP3A inhibitors only when

necessary, and in such cases the dose should be limited to 900 mg per

day. [See Warnings & Precautions (5.4), Drug Interactions (7.2), and

Clinical Pharmacology (12.3)]. [JTX-016.6 (strikethroughs and

emphasis added).]

• 7.2 CYP3A Inhibitors – Ketoconazole and other strong inhibitors of

CYP3A, such as itraconazole, nefazodone, ritonavir, nelfinavir,

indinavir, atazanavir, amprenavir and fosamprenavir,

clarithromycin, conivaptan, lopinavir/ritonavir, posaconazole,

saquinavir, telithromycin, or voriconazole may increase exposure to

mifepristone [significantly]. [The clinical impact of this interaction

has not been studied.] [Therefore, extreme] Caution should be used

when strong CYP3A inhibitors are prescribed in combination with

KORLYM. The benefit of concomitant use of these agents should be

carefully weighed against the potential risks. The dose of KORLYM

should be limited to 900 mg, and strong inhibitors of CYP3A should

be used only when necessary. [See Dosage and Administration (2.4),

Warnings & Precautions (5.6), and Clinical Pharmacology (12.3)]. [JTX-

016.11 (strikethroughs and emphasis added).]

[JTX-016 (Corcept’s 2019 Korlym® Label); see also Tr. at 151–55 (Dr. Belanoff)

(describing changes to label).] The Clinical Pharmacology section of the 2019

Korlym® label (Section 12.3) now includes data from Corcept’s clinical trials

reflecting the drug-drug interaction effects of coadministering mifepristone with a

strong CYP3A inhibitor. [JTX-016.16–17.] Specifically, Table 3 lists the average

ratios of blood concentrations of mifepristone with and without coadministration of a

variety of drugs. [Tr. at 403–04 (Dr. Snyder) (summarizing purpose of, and

information contained in, table).] Table 3 is reproduced in full below:

Table 3. Summary Table of KORLYM Drug-Drug Interaction Effects

Geometric Mean Ratio

Dosing of Coadministered Dosing of (analyte ratio with/without

Mifepristone Drug Coadministered drug coadministration)

Drug

10 davs simvastatin 80 mg single dose acid

simvastatin 10.40 | 7.02

1200 mg once daily for 5 alprazolam 180 | 0.81

10 davs alprazolam 1 mg single dose 4-hydroxy-alprazolam

. 0.76 0.39

10 days

17 days days 13-17 Metabolite It 1.02 1.06

Metabolite 2+ 1.67 1.69

tf ns |

mifepristone 1.10 1.20

14 days 14 days Metabolite 2+ 1,23 1.19

Metabolite 3+ 0,97 0.94

300 mg once daily for ips - -

*No effect = 90% CI within range 0.80 — 1.25

+See Section 12.2 for the relative potencies of the three metabolites

' Simvastatin 40 mg dose used as reference for the comparison. Result could be representative of other

oral drugs with CYP3A metabolism and high first pass effect: cyclosporine, midazolam, triazolam,

pimozide, sildenafil, sirolimus, and tacrolimus

? Result could be representative of other oral drugs with CYP3A metabolism and low first pass effect.

Clinical significance of any interaction will depend on the therapeutic margin of the drug.

3 Result could be representative of other oral drugs with CYP2C8/C9 metabolism

4 Plasma digoxin concentration should be measured after | to 2 weeks of concomitant use and following

usual clinical practice at appropriate intervals thereafter.

Result could be representative of other mild inhibitors of CYP3A

[JTX-016.16-17.]

18

E. The Patents-in-Suit.

There are two remaining patents in this litigation: the ’214 Patent and the ’800

Patent. Both involve methods of controlling hyperglycemia in certain patients who

have Cushing’s syndrome. [See generally JTX.001 (214 Patent); JTX.003 (800

Patent).| The claims embodied in these patents are based on Corcept’s findings

following the post-marketing studies that the FDA required, as summarized above.

The only asserted independent claim of the ’214 Patent is claim 10, which

involves coadministering 600 mg of mifepristone with a strong CYP3A inhibitor, after

titrating down from 1200 mg or 900 mg of mifepristone:

10. A method of controlling hyperglycemia secondary to

hypercortisolism in a patient with endogenous Cushing's

syndrome who is taking an original once-daily dose of 1200

mg or 900 mg per day of mifepristone, comprising the steps

of:

reducing, the original once-daily dose to an adjusted

once-daily dose of 600 mg mifepristone,

administering the adjusted once-daily dose of 600 mg

mifepristone and a strong CYP3A inhibitor to the

patient,

wherein said strong CYP3A inhibitor is selected from the

group consisting of ketoconazole, itraconazole, nefa-

zodone, ritonavir, nelfmavir, indinavir, boceprevir,

clarithromycin, conivaptan, lopinavir, posaconazole,

saquinavir, telaprevir, cobicistat, troleandomycin, tip-

ranivir, paritaprevir and voriconazole.

[JTX-001.37 (214 Patent, 68:47-63) (Claim 10).] Claim 1 of the ’800 Patent is nearly

identical to claim 10 of the ’214 Patent, except that it requires concomitant

administration of 900 mg of mifepristone with a strong CYP3A inhibitor, after titrating

down from 1200 mg of mifepristone:

19

1. A method of controlling hyperglycemia secondary to

hypercortisolism in a patient with endogenous Cushing’s

syndrome, said patient taking an original once-daily dose of

1200 mg per day of mifepristone, the method comprising the

steps of:

reducing the original once-daily dose to an adjusted

once-daily dose of 900 milligrams (mg) per day of

mifepristone, and

administering the adjusted once-daily dose of 900 mg per

day of mifepristone and a strong CYP3A inhibitor to

the patient,

wherein said strong CYP3A inhibitor is selected from the

group consisting of ketoconazole, itraconazole, nefa-

zodone, ritonavir, nelfinavir, boceprevir, clarithromy-

cin, conivaptan, lopinavir, saquinavir, telaprevir, cabi-

cistat, troleandomycin, tipranivir, and paritaprevir.

[JTX-003.41 (’800 Patent, 67:30-68:2) (Claim 1).]

Independent claim 6 of the ’800 Patent is practiced when 900 mg of mifepristone

is administered to a patient who is already taking a strong CYP3A inhibitor:

6. A method of controlling hyperglycemia secondary to

hypercortisolism in a patient with endogenous Cushing’s

syndrome, said patient taking a strong CYP3A inhibitor

selected from ketoconazole, itraconazole, nefazodone, rito-

navir, nelfinavir, indinavir, boceprevir, clarithromycin,

conivaptan, lopinavir, posaconazole, saquinavir, telaprevir,

cobicistat, troleandomycin, tiprantivir, paritaprevir and vori-

conazole, the method comprising administering to the

patient a once-daily dose of mifepristone of 900 milligrams

(mg) per day.

[JTX-003.41 (’800 Patent, 68:15-24) (Claim 6).]

F. Teva’s ANDA and Generic Mifepristone Product.

Teva filed ANDA No. 211436 with the FDA in December 2017 seeking FDA

approval to market, make, use, offer to sell, sell, and/or import a generic version of

Korlym® prior to the expiration of the ’214 Patent and ’800 Patent. [FPTO, Tab 2,

Stipulated Facts § 37.]| Teva’s proposed package insert is identical in all material

respects to Corcept’s 2019 Korlym® label. [Compare JTX-016 (Corcept’s 2019 label),

with JTX-011 (Teva’s proposed product label)], and Teva’s generic product is

20

bioequivalent to Korlym®, [FPTO, Tab 2, Stipulated Facts ¶ 41]. On August 2, 2020,

the FDA granted Teva’s ANDA final approval. [Id. ¶ 40.] As of the date hereof, Teva

has not yet launched its product.

II. LEGAL STANDARDS

Patent law provides that whoever actively induces infringement “shall be liable

as an infringer.” 35 U.S.C. § 271(b). To prevail on a theory of induced infringement

in the Hatch-Waxman context, a plaintiff must prove two elements: (1) direct

infringement, i.e., “if defendant’s drug [were] ‘put on the market, it would infringe the

relevant patent’ ”; and (2) specific intent, i.e., that the defendant “possessed the specific

intent to encourage another’s infringement.” Genentech v. Sandoz Inc., 55 F.4th 1368,

1376 (Fed. Cir. 2022) (quoting Vanda Pharms. Inc. v. West-Ward Pharms. Int’l Ltd., 887

F.3d 1117, 1129–30 (Fed. Cir. 2018)).4

In a Hatch-Waxman Act case involving a method-of-treatment patent, the

“mere existence of direct infringement by physicians, while necessary to find liability

for induced infringement, is not sufficient for inducement.” Takeda Pharms. U.S.A.,

Inc. v. West-Ward Pharm. Corp., 785 F.3d 625, 631 (Fed. Cir. 2015). Rather, where it is

alleged that the ANDA applicant’s proposed label induces physicians to infringe, the

label must “encourage, recommend, or promote infringement.” Id. (citations omitted);

4 A patent is directly infringed when a person “without authority makes, uses,

offers to sell, or sells any patented invention, within the United States or imports into

the United States any patented invention during the term of the patent.” 35 U.S.C. §

271(a).

see also Vita-Mix Corp. v. Basic Holding, Inc., 581 F.3d 1317, 1329 n.2 (Fed. Cir. 2009)

(“The question is not . . . whether a user following the instructions may end up using

the device in an infringing way. Rather, it is whether [the] instructions teach an

infringing use of the device such that we are willing to infer from those instructions an

affirmative intent to infringe the patent.”). If the “proposed label instructs users to

perform the patented method[,] . . . the proposed label may provide evidence of [the

ANDA applicant’s] affirmative intent to induce infringement.” AstraZeneca LP v.

Apotex, Inc., 633 F.3d 1042, 1060 (Fed. Cir. 2010).

The plaintiff, as patentee, bears the burden of proving induced infringement by

a preponderance of the evidence. Warner-Lambert Co. v. Apotex Corp., 316 F.3d 1348,

1366 (Fed. Cir. 2003) (citing Bristol-Myers Squibb Co. v. Royce Lab’ys, Inc., 69 F.3d 1130,

1135 (Fed. Cir. 1995)); SmithKline Diagnostics, Inc. v. Helena Lab’ys Corp., 859 F.2d 878,

889 (Fed. Cir. 1988).

III. DISCUSSION

Before proceeding to its analysis, the Court summarizes the parties’ arguments.

Corcept argues that it has met its burden of establishing Teva’s liability for induced

infringement under 35 U.S.C. § 271(b). [See Pl.’s Post-Trial Br., Docket No. 289.]

First, it contends that the expert testimony elicited at trial uniformly demonstrates that

Teva’s label instructs physicians to perform the patented method when

coadministering mifepristone with a strong CYP3A inhibitor. [Id. at 2–5.] Corcept

next claims that physicians have previously administered mifepristone and a strong

CYP3A inhibitor concomitantly for seriously ill Cushing’s syndrome patients and that

they are likely to do so in an infringing way in the future. [Id. at 5–11.] To this end,

it characterizes Dr. Snyder’s testimony concerning osilodrostat as a “red herring.” [Id.

at 11–13.]

Second, Corcept argues that Teva’s specific intent to induce is clear from Teva’s

label. Claiming that HZNP Medicines v. Actavis Laboratories UT, Inc., 940 F.3d 680 (Fed.

Cir. 2019) is distinguishable, Corcept contends that Teva’s label does not merely

describe infringing activity in an “if/then” manner, but rather instructs physicians to

perform the patented method when prescribing Teva’s generic mifepristone product.

[Pl.’s Post-Trial Br. at 13–17.] It argues that Dr. Synder’s opinions regarding

encouragement are not credible and that the weight of the evidence supports a finding

of intent to induce. [Id. at 17–19.] Corcept also submits that Teva’s inducement

argument does not apply to claim 6 of the ’800 Patent because all a physician must do

to practice the claim is administer 900 mg of mifepristone to a patient already taking a

strong CYP3A inhibitor. [Id. at 19–20.] Finally, Corcept asserts that Teva’s intent to

induce is evidenced by its failure to seek approval of alternative labeling. [Id. at 20.]

In opposition, Teva argues that Corcept has failed to meet its burden of

demonstrating inducement. [Def.’s Post-Trial Br., Docket No. 288.] Teva contends

that there is no record evidence that anyone has ever practiced the patented method,

and that the weight of the evidence suggests that coadministration of mifepristone and

a strong CYP3A inhibitor in the future is not likely. [Id. at 2–9.] In particular, Teva

stresses that the introduction of osilodrostat, a recently approved drug that inhibits the

synthesis of cortisol, makes coadministration of mifepristone and strong CYP3A

inhibitors obsolete. [Id. at 4–5.] Moreover, Teva maintains that its label does not

actively induce performance of the patented methods because it discourages the

concomitant use of mifepristone with a strong CYP3A inhibitor. [Id. at 9–14.] Teva

contends that HZNP is dispositive of Corcept’s inducement argument because Teva’s

label, like the generic label in HZNP, is permissive without encouraging the infringing

activity. [Id. at 14–17.] Next, it devotes much of the remainder of its submission to

distinguishing the cases on which Corcept relies. [Id. at 17–20.] In closing, Teva

opposes Corcept’s “carve-out” argument regarding alternative labeling. [Id. at 20.]

The Court discusses each of these arguments, to the extent relevant, below.

A. Direct Infringement.

The asserted claims of the ’214 Patent and the ’800 Patent are quite similar.

Claim 10 of the ’214 Patent has the following limitations: (1) it applies to a patient

who is taking a once-daily dose of 1200 mg or 900 mg of mifepristone to treat

endogenous Cushing’s syndrome; (2) the patient’s dose is reduced to 600 mg of

mifepristone; and (3) the patient is simultaneously administered a strong CYP3A

inhibitor. [See JTX-001.37.] Claim 1 of the ’800 Patent has the following limitations:

(1) it applies to a patient who is taking a once-daily dose of 1200 mg of mifepristone

to treat endogenous Cushing’s syndrome; (2) the patient’s dose is reduced to 900 mg

of mifepristone; and (3) the patient is simultaneously administered a strong CYP3A

inhibitor. [See JTX-003.41.] Claim 6 of the ’800 Patent has the following limitations:

(1) it applies to a patient who is taking a strong CYP3A inhibitor; and (2) the patient

is simultaneously administered a once-daily dose of 900 mg of mifepristone. [See JTX-

003.41.]

As noted above, to prove direct infringement, Corcept must establish that, if

Teva’s mifepristone drug product were put on the market, it would infringe the

asserted claims of the ’214 Patent and the ’800 Patent. See Vanda, 887 F.3d at 1129–

30 (explaining that the inquiry is hypothetical and asks whether the generic drug maker

is likely to market an infringing product). A court may consider all relevant evidence

in this regard. See, e.g., Ferring B.V. v. Watson Lab’ys, Inc.-Fla., 764 F.3d 1401, 1409

(Fed. Cir. 2014) (“The infringement determination is . . . based on consideration of all

the relevant evidence[.]”); Glaxo, 110 F.3d at 1570 (noting that district court “properly

considered the ANDA itself, the materials submitted by Novopharm to the FDA, and

other pertinent evidence” in conducting its infringement analysis). Though “a

patentee does not need to prove an actual past instance of direct infringement” to

establish liability for inducement, Vanda, 887 F.3d at 1129, 1129–30, “past conduct is

relevant to what will happen in the future,” Genentech, 55 F.4th at 1379, 1379–80.

1. Evidence of Prior Infringement.

The Court begins with the lack of record evidence demonstrating that anyone

has ever practiced the claimed methods, including during the ten-year span since

Korlym® was approved. Teva contends that Genentech “dismantles Corcept’s

infringement case” for this reason. [Def.’s Post-Trial Br. at 8–9; see also Def.’s Pre-

Trial Br. 9–10.] Corcept asserts that Teva’s reliance on Genentech is misplaced because

it did not “change[] the law” to “require” the Court to consider past conduct in

determining whether infringement will likely occur in the future. [Pl.’s Post-Trial Br.

at 5–6; see also Pl.’s Pre-Trial Br. 27–30.] The Genentech court looked outside of the

generic manufacturer’s proposed product label to prior physician conduct only because

it found that the label did not recommend practice of the claimed methods, Corcept

argues. [Pl.’s Post-Trial Br. at 5; see also Pl.’s Post-Trial Reply Br. at 4.]

In Genentech, the Federal Circuit considered certain patented methods for

avoiding adverse drug interactions between pirfenidone, a medication to treat

idiopathic pulmonary fibrosis, and fluvoxamine, a strong CYP1A2 inhibitor. 55 F.4th

at 1371, 1374. The drug-drug interaction patents at issue involved administering

pirfenidone to a patient already taking fluvoxamine by either discontinuing

fluvoxamine or titrating the pirfenidone dose and continuing fluvoxamine. Id. at

1374–75. The generic manufacturer’s proposed label warned about the drug-drug

interactions, specifically stating that “concomitant administration” is “not

recommended” and “[u]se of fluvoxamine . . . should be discontinued prior to

administration of pirfenidone and avoided during pirfenidone treatment.” Id. at 1375

(emphasis added). Still, the label permitted coadministration of the two agents when

necessary and provided dosing instructions in such case. See id. (“In the event that

fluvoxamine or other strong CYP1A2 inhibitors are the only drug of choice, dosage

reductions are recommended. Monitor for adverse reactions and consider

discontinuation of pirfenidone as needed [see Dosage and Administration (2.4)].”)

(brackets and emphasis in original).

The district court determined that the patentee had not met its burden of

establishing direct infringement, and the Federal Circuit affirmed for two principal

reasons. Id. First, there was no record evidence that any physicians had ever practiced

the patented methods, and “past conduct is relevant to what will happen in the future.”

Id. at 1379. Second, the generic manufacturer had presented evidence from physicians

who testified that “they had never prescribed pirfenidone to an [idiopathic pulmonary

fibrosis] patient taking fluvoxamine” and that they would prescribe a noninfringing

alternative (i.e., nintedanib) if placed in that situation. Id. at 1380. “The court did not

clearly err by considering all relevant evidence, including Sandoz’s proposed label and

physician practice.” Id.

Here, Corcept’s attempt to discredit Genentech is unavailing. Teva is not

claiming that a “split-panel decision” changed the law, as Corcept says [see Pl.’s Post-

Trial Br. at 5], nor that this Court must credit the lack of evidence of past infringement.

Rather, Teva correctly contends that such evidence is relevant. There is nothing new

about the proposition for which the case is cited. See, e.g., Takeda, 785 F.3d at 634

(affirming district court finding of no inducement based on insufficient evidence that

healthcare provider had actually practiced claimed methods and patentee had not

otherwise met threshold element of direct infringement) (citing Ricoh Co. v. Quanta

Computer Inc., 550 F.3d 1325, 1341 (Fed. Cir. 2008)); accord Limelight Networks, Inc. v.

Akamai Techs. Inc., 572 U.S. 915, 921–22 (2014) (explaining that inducement liability

can arise “if, but only if, there is direct infringement”) (citation omitted) (cleaned up).

Moreover, the Federal Circuit in Genentech did not hold that evidence of physician

practice is only relevant if the product label fails to recommend an infringing use.

Corcept’s reading of Genentech would collapse the distinct elements of induced

infringement into one inquiry focused exclusively on the language of the product label.

That is not the law. As the trial court explained,

The presence of language that “encourages, recommends or promotes”

an infringing use on a proposed label, without any additional evidence

showing such an infringing use will in fact occur, especially where there

is evidence that an infringing use likely will not occur, is insufficient for

a finding of induced infringement.

Genentech, Inc. v. Sandoz, Inc., 592 F. Supp. 3d 355, 376 (D. Del.), aff’d, 55 F.4th 1368

(Fed. Cir. 2022).5

Just as the Genentech courts concluded, and mindful of Corcept’s burden, this

Court finds the lack of record evidence that any physician has ever practiced the

asserted claims to be persuasive—here, as a starting point to its analysis. Neither Dr.

Carroll (Corcept’s expert) nor Dr. Snyder (Teva’s expert) has ever prescribed

mifepristone with a strong CYP3A inhibitor to treat Cushing’s syndrome. [Tr. at

266:9–12, 266:18–23, 270:3–9, 273:17–19 (Dr. Carroll); Tr. at 410:9–22 (Dr. Snyder).]

Though Dr. Carroll testified regarding other physicians’ occasional coadministration

5 The trial court did not reach the question whether the patentee had

demonstrated specific intent to induce infringement because it concluded that the

patentee had failed to prove direct infringement.

of mifepristone with strong CYP3A inhibitors to treat especially ill patients, [see, e.g.,

Tr. at 236:22–238:2 (referencing Dr. Hamrahian)], it is critical to note that Dr. Carroll

did not provide evidence that any physician had ever prescribed the drugs together at

an infringing sequence and/or dosage, [see Tr. at 270:20–23, 285:2–19.] Dr. Snyder

also did not identify any instances of past practice of the asserted claims. [Tr. at

392:22–393:5.] The parties’ experts thus did not testify regarding prior direct

infringement.

Similarly, the other Corcept (as well as Teva) witnesses did not provide

testimony that they had ever practiced Corcept’s patented methods. Dr. Dobbs, for

instance, indicated that she had combined ketoconazole with mifepristone “two to

three times” in her career to take advantage of their “different mechanisms of action”

for “extremely ill” patients, [Tr. at 58:16–25], but she did not provide testimony

concerning the dosage or sequence employed. The other witnesses are no different.

[See, e.g., Tr. at 96:18–23, 97:23–98:15 (Dr. Moraitis indicating that he has never

treated a Cushing’s syndrome patient with a combination of mifepristone and

ketoconazole and is not aware of any who have).] While the record undisputedly

reveals that physicians have coadministered mifepristone with a strong CYP3A

inhibitor in rare cases, [see, e.g., Tr. at 244:25–245:14 (Dr. Carroll crediting case report

of patient who both ketoconazole and mifepristone for treatment of Cushing’s

syndrome)], Corcept did not present the Court with any specific evidence where a

physician practiced the patented methods. This is notwithstanding the fact that

Korlym® has been commercially available since April 2012. The absence of this

relevant information leads to the conclusion that future direct infringement is not

likely. 6 See Genentech, 55 F.4th at 1379–80; Takeda, 785 F.3d at 634.

2. Evidence of Future Infringement.

Next, the Court addresses whether Corcept has met its burden of proof that it is

more likely than not that direct infringement of the asserted claims will occur in the

future based on other record evidence. Crediting Dr. Snyder’s testimony, this Court

6 On November 2, 2023, the Court observed in a Letter Order that Corcept had

not pointed to record evidence of prior instances of coadministration at an infringing

sequence or dosage. [Docket No. 292, at 2.] While not required, the Court

emphasized that “the existence—or absence—of such prior infringement evidence”

was important to its analysis. [Id.] The Court reminded Corcept that it had invited

this evidence during and after trial. [Id.] In its Post-Trial Reply Brief, Corcept

interpreted the Court’s letter as an opportunity to reopen the trial record to include a

set of belated declarations from medical professionals attesting, in summary fashion,

that each has practiced one or more of the claimed methods in the past. [Pl.’s Post-

Trial Reply Br. at 5 & Exs. A–F, Docket No. 295.] This is not what the Court invited,

nor will the Court grant Corcept’s request at this stage. Courts have considerable

discretion in deciding whether to reopen the trial record to permit additional evidence.

Zenith Radio Corp. v. Hazeltine Research, Inc., 401 U.S. 321, 331 (1971). In making this

decision, courts consider several factors, including (1) the burden on the parties and

their witnesses, (2) any undue prejudice that may result to the nonmoving party, and

(3) the impact on judicial economy. See Rochez Bros., Inc. v. Rhoades, 527 F.2d 891, 894

n.6 (3d Cir. 1975). The trial record may be reopened when the failure to adduce

sufficient evidence is based on a reasonable misunderstanding or when it is to ensure

substantial justice. See id. at 894–95. Here, Corcept has had ample opportunity during

this litigation to introduce specific evidence that physicians have practiced the patented

methods. Indeed, courts routinely consider this evidence when assessing whether a

patentee has made a threshold showing that future direct infringement is likely. See,

e.g., Takeda, 785 F.3d at 634. It would be too prejudicial now to consider Corcept’s

hearsay declarations over Teva’s objection, and it would be inappropriate and

burdensome to delay resolution of this matter to further examine this issue. Corcept

has not persuaded the Court that its lack of prior infringement evidence stems from a

reasonable misunderstanding, or that reopening the trial record is necessary to ensure

substantial justice. Therefore, Corcept’s request must be denied.

finds that, while clinicians may exercise their medical judgment to coadminister

mifepristone with a strong CYP3A inhibitor in isolated instances “when necessary,”

it is highly unlikely that physicians will practice the claimed methods.

First, the record evidence demonstrates that physicians generally avoid

coadministering mifepristone and strong CYP3A inhibitors. They avoid doing so

because of the challenges of dosing mifepristone and the dangers associated with

coadministering the two substances.

The evidence established that mifepristone does not inhibit the production of

cortisol, but rather blocks its effects. [Tr. at 121:5–6 (Dr. Belanoff); Tr. at 383:9–12

(Dr. Snyder).] As a result, mifepristone can cause a patient’s cortisol blood

concentrations to increase. [Tr. at 384:24–385:5 (Dr. Snyder).] Physicians thus cannot

rely on circulating blood levels of ACTH and/or cortisol to reflect therapeutic efficacy

of mifepristone when treating Cushing’s syndrome patients. [See Tr. at 158:25–163:20

(Dr. Belanoff); see also James W. Chu et al., Successful Long-Term Treatment of Refractory

Cushing’s Disease with High-Dose Mifepristone (RU 486), 86 J. Clinical Endocrinology &

Metabolism 3568, 3570, 3572 (2001) (explaining that difficulty in monitoring

therapeutic efficacy of mifepristone stems from lack of objective measure of receptor

activity) [DTX-043.0004].] This is mifepristone’s principal disadvantage in this

context. [See Tr. at 385:16–20 (Dr. Snyder) (“[T]he major drawback is that because it

doesn’t block the production of cortisol, one cannot use cortisol to determine if one is

giving the right dose. So one doesn’t know if one’s giving the right dose or too much

or too little.”).]

Strong CYP3A inhibitors, such as ketoconazole, impede the synthesis of

cortisol in the adrenal glands, thereby “blocking” excessive cortisol at its source. [Tr.

at 383:6–12 (Dr. Snyder).] Because these medications suppress the activity of CYP3A

enzymes, they can raise mifepristone blood concentrations and increase the chances

of severe side effects when taken together. [Tr. at 127:19–24 (Dr. Belanoff); Tr. at

303:9–22 (Dr. Carroll); Tr. at 395:5–8 (Dr. Snyder).] Partly on this basis, Dr. Moraitis

reported that he strongly prefers monotherapy because of the possibility of adverse

drug-drug interactions. [Cf. Tr. at 96–97 (explaining that he avoids combining agents

because combination usually fails and requires more aggressive intervention).]

Given this context, and opining on Teva’s label, Dr. Snyder persuasively

testified that physicians should “be very afraid” of coadministration, that it could cause

a “problem.” [See Tr. at 394:11, 397:22.] He explained that he would not likely use

mifepristone today to treat Cushing’s syndrome because of the dosing problem

outlined above. [Tr. at 386:9–11.] And the problem is not without consequence: too

little mifepristone, and the patient continues to have Cushing’s syndrome, with all its

attending consequences; too much mifepristone, and the patient’s blood pressure could

drop, even causing death. [Tr. at 385:21–25, 386:1–8 (Dr. Snyder).] “I am not aware

of any authority in the area of Cushing’s syndrome who recommends combining

mifepristone with another medication.” [Tr. at 413:10–12 (Dr. Snyder) (emphasis

added).] The Court finds Dr. Snyder’s testimony credible. It is significant that Dr.

Lynette Nieman, a leading authority on the treatment of Cushing’s syndrome, did not

mention combining mifepristone with other drugs in a topic review of hers published

in UpToDate®, a medical periodical for which Dr. Snyder serves a co-editor-in-chief.

[See Tr. at 413:21–418:20 (discussing Lynette K. Nieman, Medical Therapy of

Hypercortisolism (Cushing’s Syndrome), UpToDate (last updated Aug. 29, 2022) [DTX-

034]).7] Dr. Snyder suggests that the absence of any such discussion in sections of the

topic review addressing combination therapy is telling, [see Tr. at 418:7–10], and this

Court agrees. Indeed, Corcept had no real response.

Second, and relatedly, the existence of at least one non-infringing alternative

treatment to mifepristone suggests that physicians are not likely to directly infringe the

asserted claims in the future. As Dr. Snyder testified, the recent introduction of

osilodrostat has “changed” the “landscape.” [Tr. at 386:13–17.] Osilodrostat is a

“powerful” and “effective” drug that “acts rapidly.” [Tr. at 386:19–22.] Unlike

mifepristone, osilodrostat blocks cortisol production, so physicians can use cortisol

blood concentrations to determine effective dosage. [Tr. at 384:9–12, 386:17–22,

484:10–20.] Dr. Snyder opined that physicians are unlikely to coadminister

mifepristone and strong CYP3A inhibitors in the future because osilodrostat is a better

drug to treat Cushing’s syndrome. [Tr. at 393:11–18; see also Tr. at 489:22–24 (“So it

is safer to give osilodrostat than to give mifepristone, in spite of the fact that the blood

7 Still, the Court notes that the article refers to the danger of combining

ketoconazole with CYP3A4 substrates: “Ketoconazole is a strong inhibitor of

CYP3A4. Coadministration of CYP3A4 substrates should be avoided if possible and

may result in increased plasma concentrations of these drugs, with increased or

prolonged therapeutic or adverse effects.” Nieman, supra, at 4 (Initial Therapy –

Ketoconazole) [DTX-034.0004].

concentration has changed more with osilodrostat.”] This is so even though, as

Corcept elicited on cross-examination, osilodrostat is only indicated for the treatment

of Cushing’s disease. [See Tr. at 474:2–15 (Dr. Snyder) (explaining that Isturisa®, or

osilodrostat, while only FDA-approved for treatment of Cushing’s disease, is widely

used off-label for Cushing’s syndrome); Tr. at 476:21–25 (Dr. Snyder) (“[I]t is, I’d say,

even desirable to use osilodrostat for any form of Cushing’s syndrome, and preferable

to use it compared to mifepristone. I think it’s medically preferrable.”).] In summary,

Dr. Snyder concluded that in his personal experience, the benefits of coadministering

mifepristone and ketoconazole never outweigh the risks, especially since the

introduction of osilodrostat. [Tr. at 410:9–14.] The Court finds this testimony credible

and persuasive.

Corcept’s arguments are not to the contrary. For instance, Corcept argues that

Teva presented no evidence that “even a single prescription” of Korlym® has been lost

to osilodrostat. [Pl.’s Post-Trial Br. at 11, 11–12.] Corcept points to Dr. Belanoff’s

testimony that since the FDA approved Isturisa® (osilodrostat) in 2020, Corcept has

actually experienced increased sales. [See Tr. at 123:23–124:3.] He attributes the

increase in sales to heightened visibility of hypercortisolism treatments and better

screening by physicians. [Tr. at 124:7–125:1.] In other words, Dr. Belanoff claims

that the introduction of a competing drug, with a different mechanism of action, has

had a positive impact for the company. [See Tr. at 124:21–125:1.] While the Court

has no reason to disbelieve that greater awareness about hypercortisolism has certain

salubrious effects, Dr. Belanoff’s explanation about increased sales was by no means

a causal analysis. And Corcept failed to prove that the increase is because of the

introduction of osilodrostat. Dr. Belanoff’s testimony left more questions in the

Court’s mind than answers. [See Tr. at 412:20–21 (Dr. Synder stating that he was

“puzzled” by the suggestion that osilodrostat has increased Corcept’s sales of

Korlym®).] Therefore, based on the evidence elicited at trial, the Court cannot

conclude that osilodrostat will not be preferred to Korlym® in the future, as Dr. Snyder

persuasively testified.

Third, Corcept has not met its burden of demonstrating that direct infringement

is likely to occur even if a physician consults Teva’s proposed label to coadminister

mifepristone and a strong CYP3A inhibitor. As will be discussed further below, Teva’s

label does not induce physicians to administer the drugs together. Rather, the label

cautions against coadministration, providing dosing instructions if a physician

independently determines that it is medically necessary to prescribe mifepristone and

a strong CYP3A inhibitor together. Corcept has maintained that physicians will

infringe the asserted claims by following Teva’s label because the label’s instructions

are not “optional.” [Pl.’s Post-Trial Br. at 9–11, 13–17.] But this position ignores that

a physician could follow the instructions on Teva’s label and not infringe the claims.

As Corcept’s own expert, Dr. Carroll, acknowledged, administering 300 mg of

mifepristone to a patient who is already taking a strong CYP3A inhibitor would not

infringe claim 6 of the ’800 Patent. [Tr. at 267:5–9, 268:5–12.] Nor would administering

600 mg of mifepristone. A physician infringes claim 6 only by administering 900 mg of

mifepristone to a patient who is already taking a strong CYP3A inhibitor. [JTX-

003.41.]

Similarly, a physician who administers a strong CYP3A inhibitor to a patient

who is already taking 300 mg or 600 mg of mifepristone would not infringe claim 10 of

the ’214 Patent or claim 1 of the ’800 Patent. [See Tr. at 261:14–264:8 (Dr. Carroll)

(acknowledging that these scenarios, and others, fall outside the claimed methods).]

Claim 10 is only practiced when a Cushing’s patient is already taking 1200 mg or 900

mg of mifepristone, and the physician lowers the patient’s dose to 600 mg of

mifepristone and administers a strong CYP3A inhibitor. [JTX-001.37.] Claim 1 is

only practiced when a Cushing’s patient is already taking 1200 mg of mifepristone,

and the physician lowers the patient’s dose to 900 mg and administers a strong CYP3A

inhibitor. [JTX-003.41.]

Corcept largely ignores these non-infringing possibilities. Instead, it contends

that infringement will occur because most patients require 900 mg or 1200 mg of

mifepristone to achieve an efficacious result. [Pl.’s Post-Trial Br. at 9–10.] The Court

does not find this argument to be compelling. First, the testimony elicited on this score

from Dr. Belanoff and Dr. Carroll focused on the optimal dose reported in the

SEISMIC study that supported the FDA’s initial approval of Korlym® in 2012. [See

Tr. at 147:4–6 (Dr. Belanoff); Tr. at 320:23–321:2 (Dr. Carroll).] Only 34 patients

participated in this study. [Tr. at 333:6–9 (Dr. Carroll).] Despite years of discovery,

Corcept did not introduce any evidence regarding the average dose of Korlym® that

patients receive in practice, or the number of patients who have taken 300 mg, 600 mg,

900 mg, or 1200 mg of Korlym®, or the number of patients who have taken Korlym®

together with ketoconazole. The logical conclusion is that Corcept does not know this

information. [See Tr. at 83:19–21, 81:19–82:3, 82:10–14 (Mr. Maduck, Corcept’s

Chief Commercial Officer).] As a result, Corcept has not persuaded this Court that

today most patients who receive a strong CYP3A inhibitor are already taking at least

900 mg of Korlym® or that patients who are already receiving a strong CYP3A

inhibitor will be prescribed at least 900 mg of Korlym®. Therefore, the Court

concludes that physicians are not likely to titrate Cushing’s syndrome patients’ dosage

of mifepristone in an infringing manner. Accordingly, Corcept has not demonstrated

that physicians, following Teva’s label, would necessarily infringe the asserted claims.

Corcept’s other arguments fail to persuade the Court that direct infringement is

likely. For instance, Corcept contends that physicians will infringe the asserted claims

because ketoconazole remains the most commonly used medication to treat

manifestations of Cushing’s syndrome. [Pl.’s Post-Trial Br. at 7.] Corcept also argues

that patients taking mifepristone may require ketoconazole if they develop an

opportunistic infection, such as aspergillosis. [Id. at 8.] Even crediting the testimony

on which Corcept’s argument is based, [see, e.g., Tr. at 229:18–234:16 (Dr. Carroll)

(opining that it might be “medically necessary” to coadminister mifepristone with a

strong CYP3A inhibitor where certain infections occur in a patient with Cushing’s

syndrome) (first citing J.K. Oosterhuis et al., Life-Threatening Pneumocystis Jiroveci

Pneumonia Following Treatment of Severe Cushing’s Syndrome, 65 Netherlands J. Med. 215

(2007) [PTX-016]); then citing Pivonello et al., supra), the Court does not find that

infringement will likely result. As the Court has already explained, Dr. Carroll’s

testimony is not supported by any credible instances in which a physician practiced

the claimed method, and a physician can faithfully follow Teva’s label and avoid

infringement. Given these reasons, and Dr. Snyder’s testimony, the Court must reject

Corcept’s conclusion that infringement will occur.

Therefore, for the above reasons, the Court finds that Corcept has not met its

burden of demonstrating that direct infringement of the ’214 Patent and the ’800 Patent

is likely. See Takeda, 785 F.3d at 635 (finding that court permissibly concluded direct

infringement was unlikely based on testimony from physicians that they try to avoid

coadministration of drugs and lack of evidence that claimed method would actually

be practiced).

B. Specific Intent.

Next, the Court addresses whether Teva’s label “instructs users to perform the

patented method.” AstraZeneca LP v. Apotex, Inc., 633 F.3d 1042, 1060 (Fed. Cir. 2010).

To answer the question, the Court asks whether the label “encourage[s],

recommend[s], or promote[s] infringement.” Takeda, 785 F.3d at 631. “[M]ere

knowledge of possible infringement by others does not amount to inducement; specific

intent and action to induce must be proven.” Warner-Lambert Co. v. Apotex Corp., 316

F.3d 1348, 1364 (Fed. Cir. 2003) (citation omitted). The Court reaches this question

for the sake of completeness, even though it has concluded that Corcept has not shown

a likelihood that Teva’s generic product would directly infringe the asserted claims if

marketed. See Limelight Networks, 572 U.S. at 921–22 (explaining that likelihood of

direct infringement is a threshold requirement for inducement liability).

Corcept contends that Teva’s intent to induce infringement can be inferred from

Teva’s product label. [See Pl.’s Post-Trial Br. at 13–17.] Corcept maintains that the

label instructs physicians to follow dosage adjustment steps that infringe the asserted

claims, and it submits that the appropriate test is not whether the label recommends

coadministration, but whether the label recommends that physicians follow the

claimed method once it has been decided to administer the drugs together. [Id.]

Relying on HZNP, 940 F.3d at 702, Teva counters that “an instruction how to do

something is not an instruction whether to do it in the first place.” [Def.’s Post-Trial

Br. at 10, 14–17.]

In HZNP, the Federal Circuit considered whether Horizon had met its burden

of showing that Actavis’s product label induced a use of Actavis’s proposed product

that would infringe Horizon’s method-of-use patents. 940 F.3d at 699. The product

at issue was a topical solution for the treatment of osteoarthritis pain of the knees, and

Horizon’s patents addressed its method of application. Id. As the court summarized,

the patented method required three steps: (1) applying the topical solution; (2) waiting

for the area to dry; and (3) applying sunscreen, insect repellant, or a second topical

medication. Id. at 702. Actavis’s proposed label, materially identical to Horizon’s

label for its branded product, recommended a specified dosage of the solution and

instructed users to apply the solution to the clean, dry skin of the knee twice a day. Id.

at 699. It also warned that users should wait until the treated area was completely dry

before covering with clothing, applying sunscreen, insect repellent, or other

substances. Id. at 699–700.

Horizon argued that Actavis’s label reflected its specific intent to induce

infringement because it tracked, nearly exactly, Horizon’s three-step, patented method

of applying the topical solution. Id. at 701. Horizon contended that if a user sought

to apply sunscreen, insect repellant, or another substance, then direct infringement

would occur, as the user would follow the patented method and wait until the user’s

skin was dry before applying the other substance to the treated area. Id.

The Federal Circuit rejected Horizon’s argument. It explained that Actavis’s

label, properly construed, only required step one, i.e., application of the topical solution.

Id. at 702. Post-product application of sunscreen or another substance was not

required by the label. Rather, the label’s warning operated in an “ ‘if/then’ manner:

if the user wants to cover the treated area with clothing or apply another substance

over it, then the patient should wait until the area is dry.” Id. This construction, the

Federal Circuit explained, does not “encourage, recommend, or promote”

infringement; the label “merely provided guidance to patients about what to do if the

pat[i]ent desired to have anything come into contact with the knee after application of

the medication.” Id. While a user may infringe Horizon’s method-of-use patents, the

court reaffirmed that evidence of direct infringement is not sufficient to establish that

a proposed label instructs the user to perform the patented method. Id.

HZNP is instructive here. Just as the product label in HZNP did not encourage,

promote, or recommend the specified steps of the patented method because nothing

required post-product application of sunscreen or another substance, 940 F.3d at 702,

Teva’s label does not encourage physicians to practice the asserted claims because the

label does not encourage coadministration of mifepristone and a strong CYP3A

inhibitor. The label only provides instructions how to coadminister the substances to

achieve an efficacious result if a physician determines that it is medically necessary. In

this context, the label only provides “guidance” for physicians; it does not teach an

infringing use “such that we are willing to infer from those instructions an affirmative

intent to infringe the patent.” Vita-Mix, 581 F.3d at 1329 n.2.

The Court thus rejects Corcept’s attempts to distinguish HZNP. For instance,

Corcept argued that a recent decision by another court in this District demonstrates

why HZNP is distinguishable. [See Pl.’s Post-Trial Br. at 16–17 (discussing Janssen

Pharms., Inc. v. Mylan Labs. Ltd., 2023 WL 3605733 (D.N.J. May 23, 2023) (Padin, J.).]

In Janssen, the court confronted a patented method for reinitiating treatment of

schizophrenia in patients who had missed their last dose of a three-month paliperidone

palmitate extended-release injectable medication within a four-to-nine-month

window. 2023 WL 3605733, *5–7. The patent specifications described a “reinitiation

dosing regimen.” Id. at *7. The generic label at issue instructed clinicians to apply the

regimen in the “inevitable” situation that a patient returns between four and nine

months after a missed dose. Id. at *15–16. Rejecting the generic manufacturer’s

argument that its label did not induce infringement because a warning discouraged

missing a dose, the court explained that the label does not “discourage or make

optional the practice of the Asserted Claims (or any claimed steps) in the inevitable

situation that doses are missed.” Id. at *17. The Janssen court also suggested that

HZNP supported its finding of induced infringement because there “one of the claimed

steps was optional.” Id. (citing HZNP, 940 F.3d at 702).

Here, Corcept has failed to demonstrate that a similarly “inevitable” situation

is present. Unlike the missed doses in Janssen that were likely to occur based on the

credible record evidence that more than 50 percent of patients miss a dose, see Janssen,

2023 WL 3605733, at *15, there is no external trigger here that is likely to convince a

physician to coadminister mifepristone and a strong CYP3A inhibitor. The record

evidence reveals that physicians are unlikely to coadminister the two substances, and

even less likely to do so in an infringing manner. Furthermore, contrary to Janssen, a

physician here can faithfully follow Teva’s label and not infringe the asserted claims.

There are non-infringing options available. Therefore, Corcept’s efforts to differentiate

HZNP are unconvincing.

Two additional points bear emphasizing. First, Teva’s label does not

affirmatively encourage the coadministration of mifepristone with a strong CYP3A

inhibitor. The label does not outline any benefits of coadministration or indicate the

circumstances in which the agents should be coadministered. [Tr. at 399:16–22 (Dr.

Snyder).] Rather, Teva’s label warns physicians against coadministration, advising the

exercise of caution if they choose to do so.8 [See, e.g., JTX-011.1 (WARNINGS AND

8 Dr. Belanoff acknowledged that Corcept’s 2012 label represented a “forceful[]”

warning against coadministration, that physicians were taught to avoid the use of

mifepristone with a strong CYP3A inhibitor. [JTX-005.682 (Dr. Belanoff Decl. to

PRECAUTIONS – “Use of Strong CYP3A Inhibitors: Concomitant use can increase

mifepristone plasma levels. Use only when necessary and limit mifepristone dose to

900 mg (5.6).”); JTX-011.1 (DRUG INTERACTIONS – “CYP3A inhibitors:

Caution should be used when mifepristone is used with strong CYP3A inhibitors.”),

JTX-011.4, § 2.5 & Tbl. 1 (“Mifepristone tablets should be used in combination with

strong CYP3A inhibitors only when necessary.”); JTX-011.5–6, § 5.6 (“The benefit of

concomitant use of these agents should be carefully weighed against the potential

risks.”); JTX-011.9, § 7.2 (“Caution should be used when strong CYP3A inhibitors are

prescribed in combination with mifepristone. The benefit of concomitant use of these

agents should be carefully weighed against the potential risks.”).] Dr. Snyder credibly

testified that physicians would read Teva’s label as a warning not to coadminister

mifepristone and a strong CYP3A inhibitor, unless medically necessary (i.e., unless the

Patent Office); Tr. at 169:24–170:12, 171:12–15 (Dr. Belanoff).] After additional study

of mifepristone, Corcept discovered that coadministration did not raise the blood

concentration of mifepristone as much as was previously believed. Accordingly,

Corcept’s label was revised in degree: the 2019 label recommended “caution” instead

of “extreme caution” and indicated that the allowable dose of mifepristone was 900

mg instead of 300 mg of mifepristone, among other relevant changes. [Compare JTX-

016.6, with JTX-015.6.] Though there are meaningful differences between the 2012

label and 2019 label based on this breakthrough, the current label still represents a

warning against coadministration. Therefore, the label’s history reinforces the Court’s

conclusion that Teva’s label, like Corcept’s 2019 label, does not encourage the

coadministration of mifepristone and strong CYP3A inhibitors. See United Therapeutics

Corp. v. Sandoz, Inc., 2014 WL 4259153, at *18 (D.N.J. Aug. 29, 2014) (“[T]here is a

rather significant difference between a warning and an instruction. A warning

provides information regarding a potential risk. It does not prescribe a course of

action.”).

benefit to the patient outweighs the risks). [See Tr. at 393:11–18, 394:18–395:8,

396:14–18, 397:7–22, 399:4–12.]

Second, in the unlikely event that a physician decides to coadminister

mifepristone with a strong CYP3A inhibitor, Teva’s label does not provide exclusively

infringing instructions. Said differently, Teva’s label permits non-infringing options

for coadministering the substances. Infringement thus cannot be assumed. See Warner-

Lambert Co., 316 F.3d at 1364–65 (“[T]he request to make and sell a drug labeled with

a permissible (non-infringing) use cannot reasonably be interpreted as an act of

infringement (induced or otherwise) with respect to a patent on an unapproved use, as

the ANDA does not induce anyone to perform the unapproved acts required to

infringe.”). For instance, as Corcept’s own expert acknowledged, a physician could

prescribe up to 600 mg of mifepristone to a patient who is already taking a strong

CYP3A inhibitor and not infringe any of the asserted claims. [See Tr. at 268:5–270:9

(Dr. Carroll) (describing the series of steps required over the course of months for a

physician to infringe claim 6 of the ’800 Patent); see also JTX-001.37 (’214 Patent, claim

10); JTX-003.41 (’800 Patent, claim 1); JTX-003.41 (’800 Patent, claim 6).] Similarly,

prescribing a strong CYP3A inhibitor and 300 mg or 600 mg of mifepristone to a

patient would not infringe the asserted claims either. [See Tr. at 271:18–272:17 (Dr.

Carroll) (acknowledging that § 2.5 of Teva’s label would not result in infringement if

strong CYP3A inhibitor added where patient already taking 300 mg or 600 mg of

mifepristone); see also JTX-001.37 (’214 Patent, claim 10); JTX-003.41 (’800 Patent,

claim 1); JTX-003.41 (’800 Patent, claim 6).] See also supra Section III.A.2 (pp. 35–36).

Accordingly, because Teva’s label cautions physicians against coadministering

mifepristone with a strong CYP3A inhibitor unless medically necessary and because

infringement of Corcept’s patented method is not guaranteed if a physician decides to

coadminister the substances, the Court will not infer Teva’s specific intent to induce

infringement of the asserted claims. See HZNP, 940 F.3d at 702; Vita-Mix, 581 F.3d at

1329 n.2; Takeda, 785 F.3d at 631, 634–35.

Finally, Corcept’s remaining arguments fail to persuade the Court to adopt a

contrary conclusion. For example, Corcept submits that Teva’s intent to induce is

evidenced by its failure to seek alternative labeling given its position that

coadministration is dangerous. “[H]ad Teva actually believed this, it could have

petitioned the FDA to allow Teva to revert back to the original, more restrictive, 2012

Korlym label.” [Pl.’s Post-Trial Br. at 20.] It cites this Court’s disposition of

AstraZeneca LP v. Apotex, Inc., 623 F. Supp. 2d 615, 618 (D.N.J. 2009) (Bumb, J.).

Corcept’s argument is unconvincing. In AstraZeneca, Apotex’s attempt to develop an

alternative, non-infringing label was at the heart of the Court’s analysis of specific

intent, in part because the evidence suggested Apotex was aware of an infringement

problem and persisted in distributing its generic drug nevertheless. See id. at 616, 618;

see also AstraZeneca, 633 F.3d at 1060–61.

Here, as the Court previously explained, no such infringement problem exists,

and Corcept has not met its burden of demonstrating a likelihood of direct

infringement if Teva’s mifepristone were placed on the market. Furthermore, evidence

that a label was materially altered can demonstrate a defendant’s lack of intent to

induce infringement. See, e.g., United Therapeutics, 2014 WL 4259153, at *17–21

(rejecting a patentee’s argument that carve out should be discounted because

information remaining in generic product’s label amounted to an “implicit”

instruction). But the reverse is not true. The Court is not aware of any precedent

requiring an ANDA applicant to modify its product label to avoid direct infringement

arguments. That would turn the legal test “on its head.” See Takeda, 785 F.3d at 632

n.4 (explaining that patentee needs to show that ANDA applicant took affirmative

steps to induce, “not affirmative steps to make sure others avoid infringement”).

IV. CONCLUSION

In sum, the Court concludes that Corcept has not met its burden of proving

induced infringement pursuant to 35 U.S.C. § 271(b). Corcept has not demonstrated

that direct infringement of the asserted claims is more likely than not if Teva’s generic

mifepristone product were placed on the market, and Corcept has not persuaded the

Court of Teva’s affirmative intent to induce infringement based on the language

contained in Teva’s product label. Because the credible record evidence discloses that

no one has ever practiced the patented method, that physicians seek to avoid

coadministering mifepristone and strong CYP3A inhibitors, that non-infringing

alternatives are available, and that Teva’s product label discourages physicians from

coadministering, among other reasons, the Court will enter judgment in Teva’s favor.

An accompanying Order shall issue separately. Fed. R. Civ. P. 58(a).

s/Renée Marie Bumb

RENÉE MARIE BUMB

Chief United States District Judge

DATED: December 29, 2023