Respondents Brief — Teva Pharmaceuticals USA, Inc., Petitioner v. GlaxoSmithKline LLC, et al.

No. 22-37

In the

Supreme Court of the United States

TEVA PHARMACEUTICALS USA, INC.,

Petitioner,

v.

GLAXOSMITHKLINE LLC,

SMITHKLINE BEECHAM (CORK) LIMITED,

Respondents.

On Petition For a Writ of Certiorari to

The United States Court of Appeals

for the Federal Circuit

BRIEF IN OPPOSITION

JUANITA R. BROOKS

Counsel of Record

JONATHAN E. SINGER

Fish & Richardson P.C.

12860 El Camino Real

Suite 400

San Diego, CA 92130

(858) 678-5070

brooks@fr.com

MICHAEL J. KANE

ELIZABETH M. FLANAGAN

Fish & Richardson P.C.

60 South Sixth Street

Suite 3200

Minneapolis, MN 55402

(612) 335-5070

eflanagan@fr.com

Counsel for Respondents

i

QUESTION PRESENTED

When a generic drug is doubly indicated for a patented use, and there is strong record evidence of intent and inducing conduct, can the generic manufacturer evade liability for induced patent infringement

merely because it did not include on its label (i.e.

“carved out”) one of the two indications corresponding

to the patented use?

ii

RULES 24(B) AND 29.6 STATEMENT

All parties are identified in the caption of this

brief. Respondent GlaxoSmithKline LLC, is a forprofit Pennsylvania corporation.

Respondent

SmithKlineBeecham (Cork) Limited is a for-profit organization. GlaxoSmithKline plc owns 10% or more

of the stock in respondents.

iii

TABLE OF CONTENTS

Page

QUESTION PRESENTED.......................................... i

RULES 24(B) AND 29.6 STATEMENT .................... ii

INTRODUCTION ........................................................1

STATEMENT ..............................................................3

I. COREG®—A PIONEERING, LIFE-PROLONGING

TREATMENT FOR CONGESTIVE HEART

FAILURE ..................................................................3

A. The Coreg® Breakthrough ...............................3

B. GSK’s ’000 Reissue Patent Covering the

Use of Carvedilol to Reduce the Risk of

Mortality From Heart Failure .........................6

II. TEVA’S BLATANT ATTEMPT TO END-RUN

GSK’S PATENT RIGHTS ............................................7

A. GSK’s Proper Listing of the ’069 and

’000 Patents in the Orange Book and

FDA’s U-233 Use Code .....................................8

B. Teva’s Game-Playing With Its Generic

Label to Ensure It Obtained All Uses of

Coreg® for Heart Failure ............................... 13

III.THE PROCEEDINGS BELOW .................................... 17

A. The Trial Proceedings .................................... 17

B. The Jury’s Inducement Verdict and the

District Court’s JMOL .................................... 19

C. The Federal Circuit Reinstates the

Jury’s Verdict, Twice ...................................... 20

REASONS FOR DENYING THE PETITION .......... 23

I. THIS CASE PRESENTS NO BASIS FOR THE

COURT TO REVIEW ANY ASPECT OF INDUCED

INFRINGEMENT ...................................................... 24

iv

TABLE OF CONTENTS—continued

Page

A. Both the Jury and Federal Circuit

Correctly Applied the Well-Settled

Concept of “Active Inducement” in

Determining Teva’s Liability ......................... 24

B. The Factual Finding Below that Teva’s

Partial Label Encourages Infringement

Does Not Conflict with Relevant

Decisions of This Court .................................. 28

II. THE SPECIFIC FACTS OF THIS CASE LIMIT ITS

REACH AND COUNSEL AGAINST REVIEW ................ 30

III.THE SECTION VIII CARVE-OUT REGIME IS

NOT IN DANGER, NOR WILL EVERY CARVEOUT SITUATION PRESENT UNDUE RISK OF

INFRINGEMENT LIABILITY ..................................... 33

CONCLUSION .......................................................... 35

v

TABLE OF AUTHORITIES

Page(s)

Cases

AstraZeneca LP v. Apotex, Inc.,

633 F.3d 1042 (Fed. Cir. 2010) ............................ 33

Bayer Schering Pharma AG v. Lupin

Ltd.,

676 F.3d 1316 (Fed. Cir. 2012) ............................ 29

Caraco Pharm. Lab’ys, Ltd. v. Novo

Nordisk A/S,

566 U.S. 399 (2012) ................................ 1, 8, 14, 31

Commil USA, LLC v. Cisco Systems,

Inc.,

575 U.S. 632 (2015) ................................................ 1

Global-Tech Appliances, Inc. v. SEB

S.A.,

563 U.S. 754 (2011) .................................... 1, 25, 26

HZNP Medicines LLC v. Actavis Lab’ys

UT, Inc.,

940 F.3d 680 (Fed. Cir. 2019) .............................. 29

Limelight Networks, Inc. v. Akamai

Techs., Inc.,

572 U.S. 915 (2014) ................................................ 1

Metro-Goldwyn-Mayer Studios, Inc. v.

Grokster, Ltd.,

545 U.S. 913 (2005) ...................................... passim

vi

TABLE OF AUTHORITIES—continued

Page

Stilz v. United States,

269 U.S. 144 (1925) .............................................. 28

Takeda Pharms. U.S.A., Inc. v. WestWard Pharm. Corp.,

785 F.3d 625 (Fed. Cir. 2015) ........................ 29, 30

Statutes

35 U.S.C. § 251 ............................................................ 7

35 U.S.C. § 252 ............................................................ 7

35 U.S.C. § 271(b) .......................................... 24, 30, 32

Regulations

21 C.F.R. § 314.53 ......................................... 10, 31, 34

21 C.F.R. § 314.127(a)(7) (2016) ............................... 32

68 Fed. Reg. 36,676 (June 18, 2003) ......... 8, 10, 11, 31

80 Fed. Reg. 6,802 (Feb. 6, 2015) .............................. 31

81 Fed. Reg. 69,580 (Oct. 6, 2016) ...................... 10, 32

Other Authorities

Bryan S. Walsh et al., Frequency of

First Generic Drug Approvals With

‘Skinny Labels’ in the United States,

181 JAMA ............................................................. 34

Sup. Ct. R. 10....................................................... 27, 30

INTRODUCTION

This Court has considered the issue of induced infringement at least four times over the last approximately fifteen years. Commil USA, LLC v. Cisco Systems, Inc., 575 U.S. 632 (2015); Limelight Networks,

Inc. v. Akamai Techs., Inc., 572 U.S. 915 (2014);

Global-Tech Appliances, Inc. v. SEB S.A., 563 U.S.

754 (2011); Metro-Goldwyn-Mayer Studios, Inc. v.

Grokster, Ltd., 545 U.S. 913 (2005). It also has considered the inner workings of FDA use codes and “Orange Book” listings under the Hatch-Waxman Act.

Caraco Pharm. Lab’ys, Ltd. v. Novo Nordisk A/S, 566

U.S. 399 (2012). This fact-bound case presents no new

issues in those areas of law, and was properly decided

by the appellate court under those authorities. Certiorari should be denied.

Coreg® (carvedilol) is a life-sustaining treatment

for congestive heart failure developed against the

great weight of scientific consensus. Only through the

counter-intuitive vision of the inventors and hundreds

of millions of dollars of investment did GSK develop

the patented method and obtain FDA approvals for

Coreg® to treat patients with congestive heart failure.

Coreg® pioneered an entirely new category of treatment for heart failure unknown to medicine—the use

of beta blockers to prolong the life of heart failure sufferers.

Teva, for its part, chose to sell a generic version of

Coreg® and capture this market for itself. It did so

years before GSK’s patent on the use of carvedilol to

treat heart failure expired, and without taking appropriate steps to ensure its generic drug was approved

only for non-patented uses. Instead, Teva left one of

Coreg®’s two approved heart failure uses on its generic label and launched its product. Within weeks,

2

Teva captured over 40% of the market for Coreg®, including the market for the use of Coreg® to treat heart

failure, exactly as its trial witnesses admitted it intended.

Based on this record, and as detailed below, a

properly-instructed jury and the Federal Circuit

found liability under this Court’s inducement authorities. Simply put, Teva sold a generic drug with a patented use on its label and marketed the drug, with

the intent to capture the sales for that patented use.

Liability here under the law of induced patent infringement is clear.

Nothing about the potential legal consequences of

this case warrants re-visiting the fact-specific decision

of the Federal Circuit, or reversing the jury’s verdict.

Indeed, for all of Teva’s complaints, it is telling that it

resorts to a blatantly misleading “question presented”

that relies on a premise—that Teva’s label “carved

out” the patented use—rejected by the jury and the

Federal Circuit.

Teva also fails to mention that the “skinny-label”

issue impacts less than 30% of the damages. After

launching with a “carve-out” label, four years later,

Teva went to a full label and never notified GSK, as

required by law.

In short, this case presents no novel issues under

the law and concerns the liability of a party whose

problems are of its own making. It presents no threat

to generic companies who operate properly under the

law of induced infringement as applied to generic drug

labels—“skinny” or not. And, it concerns circumstances highly unlikely to be repeated, as well as regulations and policies FDA has changed multiple times

3

since the events relevant to this case occurred. Certiorari should be denied.

STATEMENT

I.

COREG®—A PIONEERING, LIFE-PROLONGING

TREATMENT FOR CONGESTIVE HEART FAILURE

A. The Coreg® Breakthrough

Congestive heart failure (“CHF” or, simply, “heart

failure”) is a devastating condition affecting millions

of Americans annually. Broadly characterized by a reduced pumping ability of the heart, untreated, heart

failure has a high mortality rate and radically reduces

patients’ quality of life. Symptoms include shortness

of breath, fluid retention and inadequate blood flow to

the limbs and vital organs. C.A.App.11519.

When launched by GSK in 1997, Coreg® was the

first heart failure treatment that significantly prolonged most patients’ lives. Its history—and that of

the ’000 patent claiming that use—is extraordinary.

The story of Coreg® as a heart failure treatment

begins with the discovery of carvedilol four decades

ago. Carvedilol is a beta blocker, and like other beta

blockers, can be used to treat hypertension, as has

been known since the 1960s. Beta blockers work by

reducing heart rate and the heart’s pumping action.

C.A.App.10357; C.A.App.11276. Cardiologists (and

FDA) thus believed beta blockers could kill heart failure patients, and the drugs were therefore contraindicated for these patients. C.A.App.2996; Pet.App.5a;

C.A.App.10357–10358; C.A.App.11274–11275. Because heart failure patients already have reduced cardiac pumping capacity, it was accepted that beta

blockers would further impair blood circulation,

4

thereby putting heart failure patients even more at

risk. C.A.App.10357–10358. Or so everyone thought.

Enter the inventors of the ’000 patent. Based on

observations they made in working with carvedilol,

the inventors conceived carvedilol could treat, rather

than worsen, heart failure, contrary to the medical

consensus. C.A.App.11275–11280. And while it took

some convincing, GSK eventually pursued the inventors’ conception in a full-blown clinical trial beginning

in 1993.

C.A.App.2997; C.A.App.11279–11280;

C.A.App.10436–10438.

This clinical trial was a spectacular success. Rather than killing heart failure patients, carvedilol was

so effective it decreased the risk of death by 65%.

C.A.App.2997–2998;

C.A.App.10373–10374;

C.A.App.11282. As a result, GSK was ordered to stop

the trial early because it would have been unethical to

withhold the life-saving drug from the patients receiving placebo. C.A.App.11282, C.A.App.10371–10373;

C.A.App.3409. When inventor Dr. Ruffalo received

the call that the trial had to be stopped, his first reaction was panic. “[O]h my God,” he testified, “it killed

people, just like everybody said, and it’s my fault.”

C.A.App.11282. But when he learned the real reason

for the call, he “cried for three hours . . . .” Id. Dr.

Lukas, another inventor, recalled being “shocked” at

the trial’s extraordinary results. C.A.App.10374.

Based on these extraordinary results, in May

1997, FDA approved Coreg® to treat “mild to moderate” heart failure patients—the majority of the patients that were in that first trial—and GSK finally

launched the drug, achieving immediate commercial

success.

C.A.App.10377–10379; C.A.App.10795.

GSK then ran additional clinical trials with “severe”

heart failure patients, and FDA approved Coreg® for

5

those patients, too, such that, by 2001, Coreg® was labeled for “the treatment of mild to severe chronic

heart failure . . . .”

C.A.App.10379–10381;

C.A.App.11164. Today, this is indication 1.1 on

Coreg®’s label. C.A.App.7664–7665.

But for patients who had recently suffered a heart

attack and might only be in the earliest stages of heart

failure, with few or no symptoms yet present, Coreg®

was unavailable. In GSK’s clinical trials, individuals

who had suffered a major cardiovascular event within

three months of the trials’ start were excluded.

C.A.App.11515.

Unfortunately, many recent heart attack victims

have heart failure and need immediate care. Accordingly, GSK got back to work, and initiated clinical trials on the use of carvedilol for “post-MI LVD” patients.

C.A.App.10382. These are patients who have suffered

heart attacks (i.e., myocardial infarctions, or MI)

within 21 days, and have left ventricular dysfunction

(LVD), meaning the left ventricle of the heart has a

reduced ability to pump out blood, the main defining

characteristic of heart failure. Id. LVD is quantified

by the heart’s “ejection fraction,” which measures the

amount of blood pumped out of the left ventricle with

each beat. C.A.App.10602–10603. A normal ejection

fraction is typically between 55%–70%; patients with

an ejection fraction of less than 40% have congestive

heart failure, regardless of whether they have developed any symptoms of the disease. C.A.App.10603;

C.A.App.11132; C.A.App.11226.

These additional clinical trials were also successful, and, in March 2003, FDA approved Coreg® for reducing the risk of death in patients with this manifestation of heart failure, post-MI LVD with an ejection

fraction below 40%, “with or without symptomatic

6

heart failure.” C.A.App.7992; C.A.App.10382–10383.

Today, this is indication 1.2 on Coreg®’s label.

C.A.App.7665.

GSK’s trailblazing work and investment in carvedilol changed the standard of care for heart failure patients. C.A.App.10667–10668; C.A.App.11267. As a

result, heart failure is no longer the short-term death

sentence it was just 25 years ago.

B. GSK’s ’000 Reissue Patent Covering the

Use of Carvedilol to Reduce the Risk of

Mortality From Heart Failure

Days after GSK’s first carvedilol heart failure trial

ended early based on extraordinary results, GSK filed

a patent application covering the use of carvedilol to

decrease mortality of heart failure patients.

C.A.App.10373–10374. That application issued as

U.S. Patent No. 5,760,069 (the ’069 patent”) on June

2, 1998, with an expiration date of June 7, 2015.

Rather than wait until this patent covering GSK’s

pioneering work expired, Teva wanted in on the carvedilol heart failure market early, in 2007, when the

patent on the carvedilol molecule expired. And so in

2002 Teva applied for FDA approval to market a

Coreg® copy. As part of this process, Teva provided

GSK a “Paragraph IV” letter notifying GSK of Teva’s

generic drug application (“ANDA”). In this letter,

Teva did not deny that its proposed generic copy of

Coreg®

would

infringe

the

’069

patent.

C.A.App.3003–3019. Instead, Teva asserted that the

’069 Patent was technically invalid over the prior art,

including GSK’s own prior carvedilol publications. Id.

7

Instead of brushing off Teva’s arguments, GSK

took them seriously and asked the Patent Office to reconsider the ’069 Patent in light of Teva’s prior art in

a “reissue” proceeding.

Reissue proceedings allow a patent owner to obtain substitute patent rights where, as a result of an

error, original patent rights are potentially invalid.

35 U.S.C. § 251. Importantly, while patent owners

must agree to surrender their original patent rights

at the conclusion of the proceedings, those rights remain in force up and until the PTO issues the reissue

certificate. 35 U.S.C. § 252.

GSK filed for reissue of the ’069 patent in November 2003 to correct the mistake identified by Teva.

See, e.g.¸ C.A.App.7015–7018; C.A.App.10968. The reissue proceeding progressed slowly, concluding in

January 2008, when the PTO granted the ’000 reissue

certificate with newly worded claims. C.A.App.32;

C.A.App.11039; C.A.App.10968.

While narrower,

these claims still covered GSK’s pioneering invention

of the use of carvedilol to prolong the life of heart failure patients, no matter the patient’s specific heart

failure symptoms or cardiac event history.1 It is these

claims the jury found Teva willfully infringed by marketing its generic copy of Coreg®.

II. TEVA’S BLATANT ATTEMPT TO END-RUN GSK’S

PATENT RIGHTS

Because the reissue proceeding took so long, it

overlapped with Teva’s generic launch in September

1 The re-issued claims require a heart failure patient

be maintained on treatment with carvedilol for at

least six months, a requirement absent from the original claims. C.A.App.45 (8:30–41); (C.A.App.261

(8:27–33).

8

2007. During those years, and even after, Teva engaged in a concerted effort to capture the lion’s share

of the market for the use of carvedilol to treat heart

failure. These efforts include the conduct at the center of this case—Teva’s manipulation of its generic

drug label.

A. GSK’s Proper Listing of the ’069 and ’000

Patents in the Orange Book and FDA’s U233 Use Code

When Teva first notified GSK of its ANDA in

2002, GSK had three patents listed in the Orange

Book related to Coreg®. One was the ’069 patent,

which GSK had listed upon its issuance in 1998. At

the time, applicants listed method of use patents under a 1994 FDA policy that encouraged, but did not

require, NDA applicants to “submit to FDA information on the approved use claimed by the patent.”

68 Fed. Reg. 36,676, 36,682 (June 18, 2003).

While that policy no longer exists, it had two critical aspects: 1) use codes appearing in the Orange

Book were created entirely by FDA; and 2) use codes

were only “intended to alert the ANDA and 505(b)(2)

applicants to the existence of a patent that claims an

approved use. They are not meant to substitute for

the applicant’s review of the patent and the approved

labeling.” Id. at 36,683. Additionally, the “patent certification forms” touted by Teva in its petition—and

discussed in Caraco—did not exist under this policy;

FDA first required those from NDA applicants in August 2003. Id. at 36,677 (the final rule “makes

changes to the patent information required to be submitted and provides declaration forms for submitting

that information to FDA . . . .”); 36,686, 36,707–709

(form 3542a), 36,710–712 (form 3542).

9

Accordingly, when Teva filed its ANDA in 2002,

the Orange Book listed the ’069 patent under the

FDA-created use code U-233. C.A.App.7831. FDA’s

summary for that use code was “decreasing mortality

caused by congestive heart failure.” C.A.App.7833.

This has always been the use code summary for

the ’069 patent and, ultimately, the ’000 reissue patent. Notably, FDA did not change the use code when

it approved Coreg® for post-MI LVD patients in

March 2003, which also occurred under the 1994 policy. Nor should it have. The indication approved by

FDA for these patients explicitly instructs doctors to

treat them with carvedilol “to decrease cardiovascular

mortality,” exactly as the use code describes. The

main relevant clinical distinction between these patients and those with mild to severe heart failure is

that post-MI LVD patients may or may not be symptomatic. As the approved indication for post-MI LVD

patients states:

COREG is indicated to reduce cardiovascular mortality in clinically stable patients who have survived the acute phase of

a myocardial infarction and have a left ventricular ejection fraction of less than or equal

to 40% (with or without symptomatic heart

failure) [see Clinical Studies (14.2)].

C.A.App.7665 (emphasis added).

This use code did not change when FDA began requiring the patent information forms in August 2003.

As an initial matter, Teva omits that the requirements for these forms, as first implemented in 2003,

were very different than they are today. Today, an

NDA holder wanting to list a method of use patent

must identify every section and subsection of its label

10

that describes the method of used claimed by its patent. 21 C.F.R. § 314.53(c)(2)(i)(O)(2) (2019). FDA

only adopted these requirements in 2016, almost 20

years after GSK first listed the ’069 patent in the Orange Book. 81 Fed. Reg. 69,580, 69,599 (Oct. 6, 2016).

In August 2003, when FDA first required patent

information forms, they were fairly sparse. They

simply required the NDA holder to confirm that its

patent claimed an approved use and to both identify

and describe that approved use with reference to the

approved labeling. 68 Fed. Reg. 36,686, 36,707–709,

36,710–712 (June 18, 2003). There was no requirement—nor room on the form—for the NDA holder to

identify everything on its label related to the listed patent. Id. at 36,686. FDA did not impose an obligation

to identify every section and subsection of the labeling

that described the claimed method of use until 2016.

81 Fed. Reg. 69,599.

Teva also omits that there are two different types

of patent information forms, Forms 3542 and 3542a,

which have distinct purposes. Form 3542a is for use

by NDA applicants when applying for approval

(whether original, amended or supplemental). 21

C.F.R. § 314.53(c)(2)(i). Form 3542 is for use by NDA

holders after receiving approval.

21 C.F.R. §

314.53(c)(2)(ii).

Critically, in first requiring these forms in 2003,

FDA announced it would only use Form 3542, and not

Form 3542a, for Orange Book listing purposes. See 68

Fed. Reg. 36,707. As the forms and their instructions

explicitly state: “[t]he information submitted in the

declaration form submitted upon or after approval

will be the only information relied upon by FDA for

listing a patent in the Orange Book” (form)

11

(C.A.App.6894; 68 Fed. Reg. 36,707); “[o]nly information from form 3542 will be used for Orange Book

publication purposes” (instructions) (C.A.App.6897).

Consistent with this, Form 3542a does not even request of the applicant a proposed use code, while Form

3542 does. Compare C.A.App.6894–6896 (form 3542a)

with C.A.App. 6880–6882 at 6882 (form 3542 with box

4.2b “Use Code”).

But, even with these new forms, FDA remained

clear that use codes developed based thereon were no

substitute for the generic filer’s independent evaluation: “[use codes] are not meant to substitute for the

applicant’s review of the patent and the approved labeling.” 68 Fed. Reg. 36,683 (June 18, 2003).

Under these then-applicable policies, FDA never

asked GSK to provide a patent information form in

connection with the post-MI LVD indication, whether

at the time the requirement arose in August 2003, five

months after the post-MI LVD indication had been

approved, or anytime thereafter. During this time,

GSK continuously sold Coreg® with the post-MI LVD

indication on the label, and applications for generic

copies of Coreg®, including Teva’s, remained pending.

If there had been any issue with use code U-233 during this time frame, in relation to the post-MI LVD

indication, surely it would have been raised by FDA

or the highly motivated generic companies. Instead,

there was silence from FDA, and no generic company

ever told GSK that a generic copy of Coreg® labeled

with the post-MI LVD indication would not infringe

the then-existing ’069 patent.

While GSK ultimately did submit the two patent

information forms raised by Teva in its petition, GSK

did so in relation to events having nothing to do with

the contested post-MI LVD indication. The first is a

12

Form 3542a for the ’069 patent GSK submitted on

September 1, 2006, in connection with GSK’s application for pediatric exclusivity. C.A.App.6889–6907.

The second is a Form 3542 GSK submitted on February 6, 2008 to note the issuance of the ’000 reissue patent in place of the already-listed ’069 patent.

C.A.App.6873–6887.

Notably, while Teva claims it relied on these

forms, there is no evidence in the record Teva ever obtained or saw them before this case. We nonetheless

address these forms, so the record before this Court is

clear.

Plainly, from a timing perspective, only the 2006

form could possibly have any relevance to Teva’s initial launch in September 2007. But that form was a

Form 3542a, not a Form 3542, and thus had nothing

to do with any Orange Book listing, the already existing FDA use code in the Orange Book, or Teva’s carveout label. As FDA made clear, only information in a

Form 3542 would ever be used for Orange Book purposes. But even still, nothing on GSK’s 2006 Form

3542a reasonably suggests that the post-MI LVD indication was not covered by the then existing ’069 patent. Rather, GSK described the approved use covered by the patent as “Treatment of Mild-To-Severe

Heart Failure . . . To Increase Survival,” which embraces all heart failure patients covered by the patent

and referenced on the approved label for Coreg®.

C.A.App.6895.

As for the Form 3542 GSK submitted in February

2008 for the ’000 reissue patent, that form repeats the

above description of the patented use, and, in the additional box not present on Form 3542a, adds that the

use code should remain the same as it always had

13

been since FDA created it in 1998: “Decreasing Mortality Caused by Congestive Heart Failure.”

C.A.App.6881–6882. But as noted, GSK submitted

this form six months after Teva’s launch, so it had

nothing to do with Teva’s carve-out label or decision

to launch its infringing product. And while GSK misled no one about the scope of its patent rights, the timing makes it impossible for GSK to have even theoretically done so, let alone be responsible for Teva’s (or

FDA’s) actions. GSK complied with FDA regulations

and policies, and misled no one.

B. Teva’s Game-Playing With Its Generic

Label to Ensure It Obtained All Uses of

Coreg® for Heart Failure

For its part, what did Teva do to comply with

FDA’s policies over the years in relation to the postMI LVD indication that appears on its label? Nothing.

At trial, Teva presented no evidence it ever independently evaluated the ’069 patent or the ’000 reissue patent as required by FDA policy. Far from doing

“everything right,” as Teva proclaims, it did nothing

at all.

Instead, Teva intended and did everything to ensure it would secure sales of carvedilol for all heart

failure patients, regardless of how its product was labeled. This began with Teva’s initial ANDA in 2002,

labeled for all indications. C.A.App.10530. After FDA

approved Coreg® for post-MI LVD heart failure patients, Teva also added that indication. See, e.g.,

C.A.App.5508; C.A.App.10622–10623. Tellingly, Teva

did not revise its Paragraph IV notice to state that

this added indication did not infringe the ’069 patent.

14

Then, in 2004, FDA granted Teva tentative approval. C.A.App.7788–7792. Immediately, Teva began priming the market for the launch of its generic

copy, encouraging doctors to use the product just like

Coreg® to treat heart failure. Teva trumpeted in a

press release, distributed to cardiologists, that its

“Carvedilol Tablets are the AB-rated generic equivalent of Glaxo SmithKline’s Coreg® Tablets and are indicated for treatment of heart failure” and would be

available in 2007. C.A.App.6347. This press release

also referenced GSK’s total Coreg® sales, the vast majority of which were for heart failure, evidence Teva

planned on capturing all sales including those for the

patented use. Id.

As for GSK’s then-existing ’069 patent, Teva

seemed content to rely on invalidity arguments. But,

as its desired launch date drew near, and it appeared

the reissue proceeding might fix GSK’s patent, Teva

apparently got cold feet and hatched a new strategy,

based on a Hatch-Waxman Act “section viii statement.” As described by the Court in Caraco, these

statements “assert[] that the generic manufacturer

will market the drug [only] for one or more methods of

use not covered by the brand’s patents.” Caraco, 566

U.S. at 406. Generic manufacturers that employ them

“propose labeling for the generic drug that ‘carves out’

from the brand’s approved label the still-patented

methods of use.” Id.

But there was a problem with Teva’s strategy: its

“carve-out” was incomplete. Rather than carve-out

both indications covering Coreg®’s heart failure uses,

Teva only carved out the first one, leaving the indication that instructs doctors to use carvedilol to “reduce

cardiovascular mortality” in post-MI LVD patients

15

with an ejection fraction of ≤40% and with symptomatic heart failure.2 C.A.App.5508.

Also remaining on the label was language warning of the risk of “worsening heart failure” when patients first use the product, as well as data from GSK’s

clinical studies showing carvedilol reduced mortality

in the post-MI LVD heart failure patients.

C.A.App.5510; C.A.App.5512–5513. Compounding

matters, Teva did no analysis to support its carve-out;

instead, it copied a carve-out FDA apparently negotiated

with

another

generic

manufacturer.

C.A.App.7793.

None of these stratagems had anything to do with

GSK’s actions. Unlike today, when GSK would have

identified each section of its label for consideration by

FDA, GSK was cut out of this process, the opposite of

Teva’s erroneous claim GSK somehow informed FDA

and the public which parts of its label should be

carved out to avoid infringement. As noted, by the

time Teva concocted its carve-out strategy, GSK had

not yet provided the Form 3542 Teva now claims it

relied upon.

Teva went forward with the carve-out label, and

launched its product on September 6, 2007. In its

press release announcing this launch, Teva nowhere

mentions its product has a carve-out label, but instead

communicated to doctors its product should be used to

treat all heart failure patients. C.A.App.6342. This

press release explicitly stated that Teva was selling a

Both experts agree a patient with an ejection fraction

of ≤40% with symptoms is suffering from heart failure. C.A.App. 11226 (Dr. Zusman); C.A.App.11131–

1132 (Dr. Zusman); C.A.App.10622–10631 (Dr.

McCullough); C.A.App.10696 (Dr. McCullough).

2

16

“Generic version of GlaxoSmithKline’s cardiovascular

agent Coreg®,” which GSK’s expert testified was

shorthand to doctors to use the drug to treat heart failure. C.A.App.11660.

And, just like the 2004 release, the 2007 release

included Coreg®’s total revenue, demonstrating Teva

still intended to capture the entire heart failure market. In fact, Teva included heart failure revenue despite personnel explicitly questioning whether it was

right to do so. C.A.App.10973–10974. This press release remained on Teva’s website throughout the entire infringement period. C.A.App.6353.

Teva’s subsequent marketing materials built upon

those press releases and reinforced that its generic

product should be used exactly like Coreg®, including

for the patented use. Teva’s product catalogs directly

compared Teva’s product to Coreg® without limitation, including by stating it was “AB”-rated to

“Coreg®.” See, e.g. C.A.App.6221; C.A.App.6270.

Teva’s 2009 product guide was even more explicit,

referring to Coreg® as the “Brand Equivalent” of

Teva’s product. C.A.App.6324. As explained to the

jury by GSK’s regulatory expert, according to FDA,

such direct comparisons—where the generic not only

refers to an “AB” rating but also invokes the name of

the branded drug (Coreg®)—communicates both

products are approved for all the same uses.

C.A.App.10545. All of these efforts by Teva achieved

its intended goal—capture the lion’s share of the market for Coreg®, including for the patented use.

Having largely destroyed the market for Coreg®

with its launch in 2007, Teva applied the coup de

grâce in 2011. Teva amended its label to add back the

indication for “mild-to-severe” heart failure and the

17

corresponding clinical trial data and reports, which it

previously omitted. Teva’s “full” label, as well as related Monthly Prescribing References for Healthcare

Professionals, now told doctors to use the product for

all types of heart failure, as its other promotional materials had done for years. C.A.App.5531–5553;

C.A.App.6192–6208.

And, just as in 2007, Teva’s 2011 efforts to comply

with FDA policy and regulations were non-existent.

As GSK’s and Teva’s experts agreed, Teva’s change to

the “full” label in 2011 should have been accompanied

by a Paragraph IV notification. C.A.App.10571–

10572; C.A.App.11049–11050. Instead? Silence from

Teva, and its capture of the market GSK worked for

15 years to create was complete.

III. THE PROCEEDINGS BELOW

A. The Trial Proceedings

After unsuccessfully trying to rebuild the branded

market for Coreg® with an improved carvedilol product, GSK ultimately sued Teva in 2014 to recover at

least some of its damages. In its complaint, GSK alleged Teva willfully induced infringement of the ’000

reissue patent through its sale and marketing of its

generic carvedilol. The case went to trial in June

2017.

In the main, Teva did not try the case based on the

issues described in its petition—i.e., that GSK’s Orange Book listings were somehow the cause of Teva’s

infringement. That defense, which Teva has put forward as an equitable estoppel defense, will be tried on

remand in the District of Delaware.

Instead, in an attempt to avoid liability, Teva

challenged validity and also raised a novel causation

18

theory: that generic drug providers should not be held

responsible for inducing patent infringement because

branded companies, when they first make the market,

teach doctors how to use the drug according to their

patented methods. Under this theory, the cause of

doctors’ infringement with Teva’s product was not

Teva, but, bizarrely, GSK.

As for Teva’s carve-out label, the primary role it

played at trial was a mundane one—Teva challenged

it actually instructed the patented method during the

partial or full label period. C.A.App.11152–11154. To

keep the two separate, the jury rendered its liability

verdict in two time periods: a) the carve-out, or partial

label, period, from 2007–2011; and b) the full-label period, from 2011 to patent expiry in 2015.

C.A.App.205–206.

Far from communicating to the jury that Teva

launched its product because it had purportedly

carved out “all of the language” GSK “identified as

covering its patented uses,” as Teva’s Question Presented claims, Teva relied on the reissue proceedings.

Counsel for Teva told the jury in opening that Teva

launched in 2007 because the re-issue proceeding resulted in a “gap” in patent coverage, and thus Teva

was free to launch—and destroy GSK’s market—without fear of liability. C.A.App.10324; C.A.App.10343.

But, as the jury was instructed, there was no “gap”

that resulted from reissue. GSK’s patent rights were

continuously in force from 1998 forward.

C.A.App.166; C.A.App.11832–11833. When forced to

confront this in closing, Teva’s counsel cavalierly dismissed his earlier claims with a shrug and a glib “My

bad.” C.A.App.11891.

19

B. The Jury’s Inducement Verdict and the

District Court’s JMOL

After hearing all of this, the jury found Teva willfully induced infringement by selling its generic carvedilol with both its partial label and full label, awarding GSK $235 million in damages. Pet.App.9a–10a;

C.A.App.207; C.A.App.211. The jury thus sided with

GSK that Teva’s partial label was not a true carveout, and encouraged infringement, including the indication which instructs doctors to use carvedilol to reduce cardiovascular mortality in post-MI LVD heart

failure patients. See Pet.App.27a. Indeed, the jury

heard experts for GSK and Teva agree that the postMI LVD patients with a left ventricular ejection fraction of ≤40% and symptoms of heart failure were all

suffering from heart failure. Pet.App.16a, 18a–19a;

C.A.App.10602–10606, 10622–10623, 11226. The verdict also means the jury rejected Teva’s defense that

it did not cause infringement.

The district court upended the jury’s liability verdict at JMOL. On JMOL, Teva argued the jury could

not reasonably have found Teva caused doctors to infringe. The district court agreed. Pet.App.160a–161a;

see also Pet.App.95a. In addition, in a footnote the

district court treated the fact question of whether the

post-MI LVD indication instructs the claimed use “as

though it were a legal one for it to decide de novo.”

Pet.App.21a–22a; C.A.App.15–16 at n.9. And, improperly slotting itself in as the fact-finder, “decided

the post-MI LVD portion of Teva’s label was insufficient to find that the [partial] label instructed an infringing use.” Pet.App.21a; C.A.App.15–16 at n.9.

20

C. The Federal Circuit

Jury’s Verdict, Twice

Reinstates

the

At the Federal Circuit, a majority panel reversed

the district court’s JMOL decision and reinstated the

jury’s verdict, finding it supported by substantial evidence. Pet.App.90a. The primary issue addressed on

appeal was Teva’s novel causation argument that had

persuaded the district court to grant its JMOL request. Pet.App.96a. Whereas the majority found causation supported by substantial evidence, in the dissent’s view it was lacking, including based on the dissent’s belief—contrary to the jury’s verdict—that Teva

had marketed “its generic carvedilol for unpatented

uses through a ‘skinny label.’” Pet.App.110a–11a (emphasis in original).

Emboldened by the dissent’s favorable take on its

“skinny” label, Teva sought rehearing, contending the

majority’s decision “could be broadly read to impose

liability on ANDA filers that carved out patented uses

under section viii when seeking approval to market

generic drug products, in direct contravention of the

Hatch-Waxman Act.” Pet.App.10a–11a. Several

amici supported Teva. The Federal Circuit granted

the petition for panel rehearing, vacating its earlier

judgment and opinion. Pet.App.11a.

The rehearing grant had a distinct purpose: to

make clear that Teva’s “skinny” label did not reflect

only unpatented uses, and thus assuage concerns the

decision would have far-reaching consequences to the

section viii carve-out regime.

Indeed, the Federal Circuit’s subsequent per curiam decision, again reinstating the jury’s verdict of

willful induced infringement, explains that the panel

“agreed to rehear this case to make clear how the facts

21

of this case place it clearly outside the boundaries of

the concerns expressed by amici” that the prior decision “upset the careful balance struck with section viii

carve-outs.” Pet.App.11a. It did so with a concise message about the rare nature of the facts of this case that

Teva still refuses to come to terms with—that its

“skinny” label did not carve out the patented use:

This is a case in which substantial evidence

supports a jury finding that the patented use

was on the generic label at all relevant times

and that, therefore, Teva failed to carve out all

patented indications. This narrow, case-specific review of substantial evidence does not upset the careful balance struck by the HatchWaxman Act regarding section viii carve-outs.

Pet.App.12a.

The rehearing decision explains why the evidence

supports the jury’s inducement verdict in both the

partial and full label periods. Most relevant to the issues Teva raises here, the decision reviews the substantial evidence supporting the jury’s conclusion that

the post-MI LVD indication left on Teva’s partial label

encourages the patented use of decreasing mortality

caused by congestive heart failure. Pet.App.15a–16a.

And, it rightly calls out that Teva’s expert conceded

the post-MI LVD indication covers patients with congestive heart failure. Pet.App.16a.

The rehearing decision also addressed the impact

GSK’s submissions to FDA for Orange Book listing

had on the jury. Simply put, none. Pet.App.22a–25a.

This makes sense. While the jury did not hear about,

nor did the rehearing decision (or dissent) reflect, the

various policies in place at the relevant time concern-

22

ing Orange Book listings and use code drafting outlined above, the jury did hear and the rehearing decision did recognize, a critical fact that was in play

when the relevant events occurred: that ANDA applicants like Teva have an obligation to analyze OrangeBook listed patents to determine how to prepare their

carve-out labels. Pet.App.24a–25a. From Teva’s petition, one would never guess that Teva’s counsel

rightly conceded during oral argument that GSK’s

FDA submissions “are ‘not absolutely dispositive of infringement.’” Pet.App.23a.

The rehearing decision also dispensed with Teva’s

attempts to liken its partial label to others found not

to induce infringement. Pet.App.16a–21a. In so doing, it rejected Teva’s claim that GSK’s expert had

merely “cobbled together” disparate portions of the label to drum up infringement instructions, and that

such piecing together of the label cannot give rise to

inducement liability. Pet.App.17a–18a. As the jury

had done, the majority read the actual label, rather

than rely on Teva’s distorted description, finding “[a]ll

of the claim limitations were contained in the Indication section (which amounted to a single sentence),

the Clinical Study section (to which doctors were directly referred by the Indication section), and the Dosage and Administration section (which immediately

follows the Indication section and which says how

much and how often to give the carvedilol).”

Pet.App.17a–18a.

The dissent’s contrary take, parroted by Teva in

its petition, is built on an assumption unsupported by

the factual record—that GSK’s FDA submissions are

somehow responsible for the words in Teva’s partial

label. Pet.App.64a. In the dissent’s view, because

23

GSK did not recite the post-MI LVD indication verbatim in its Form 3542a—which states it will not be used

for Orange Book listing purposes (C.A.App.6893–

6894)—GSK had put a stake in the ground that the

post-MI LVD indication was not a patented use.

Pet.App.57a–58a, 62a, 65a. The dissent then bases its

entire analysis on this incorrect view of “the carve-out

backdrop” of this case. Pet.App.64a, 66a. Moreover,

the dissent—like the district court—wrongly played

fact-finder, relying on its own interpretation of Teva’s

press releases, notwithstanding expert testimony to

the contrary, and also improvidently attacked the circumstantial, as opposed to direct, nature of GSK’s inducement evidence. Pet.App. 69a–74a, 80a–81a. The

dissent further doubled down on lack of causation,

concluding, that even when Teva outright copied

GSK’s entire label Teva did not induce infringement.

Pet.App.76a–77a.

None of the dissent’s concerns or arguments present a sound reason for this Court’s review. We explain why below.

REASONS FOR DENYING THE PETITION

When viewed through the lens of the actual facts

and regulatory landscape, Teva’s petition collapses.

The decision below does not have the far-reaching consequences Teva and amici lament because its outcome

turns on the application of a specific set of facts to a

properly stated rule of law. The decision below—like

others before it addressing carve-out labels—has not

had, and will not have, a chilling effect on the generic

drug industry.

Moreover, review of the decision below would involve considering long-since-changed FDA regula-

24

tions that have no bearing on how the carve-out process operates today. And, the challenged aspect of the

decision below is not even case dispositive, since over

70% of the damages occurred during Teva’s full label

period.

At bottom, Teva is asking this Court to rescue it

from an adverse factual finding of substantial liability

based on a fanciful Question Presented. Denial is appropriate.

I.

THIS CASE PRESENTS NO BASIS FOR THE COURT

TO REVIEW ANY ASPECT OF INDUCED INFRINGEMENT

The law of inducement is not in jeopardy. The record demonstrates both the jury and appellate court

applied the correct construct of induced infringement,

and that the liability determination is amply supported. In suggesting otherwise, Teva ignores the

substantial evidence supporting the finding that its

carve-out, or partial, label encourages a patented use,

and that its inducement liability was also predicated

on marketing messages beyond its labels.

A. Both the Jury and Federal Circuit Correctly Applied the Well-Settled Concept

of “Active Inducement” in Determining

Teva’s Liability

“Whoever actively induces infringement of a patent shall be liable as an infringer.” 35 U.S.C. §

271(b). Though first codified in the 1952 Patent Act,

this concept of induced infringement had long been

recognized at common law. Metro-Goldwyn-Mayer

Studios, Inc. v. Grokster, Ltd., 545 U.S. 913, 935–36;

see also id. at 942 (Ginsburg, J., concurring). To illustrate, at common law, a “patent defendant who not

25

only expected but invoked infringing use by advertisement” was liable for infringement. Id. at 935 (brackets omitted).

As explained in Grokster, “[t]he rule on inducement of infringement as developed in the early cases

is no different today.” Id. at 936. “[A]ctive steps”

taken by the accused infringer “to encourage direct infringement” are required to establish inducement liability, and such steps may be shown through “advertising an infringing use or instructing how to engage

in an infringing use.” Id.; see also id. at 937 (“The

classic instance of inducement is by advertisement or

solicitation that broadcasts a message designed to

stimulate others to commit violations.”).

Global-Tech paints the same picture based on the

statute’s plain language. It explains, “[t]he term ‘induce’ means ‘[t]o lead on; to prevail on; to move by persuasion or influence.’” Global-Tech Appliances, Inc. v.

SEB S.A., 563 U.S. 754, 760 (quoting Webster’s New

International Dictionary (2d ed. 1945)). It continues,

“[t]he addition of the adverb ‘actively’ suggests that

the inducement must involve the taking of affirmative

steps to bring about the desired result.” Id.

Inducement liability must thus be grounded in actions that encourage and foster using a product to infringe rather than for using it for some lawful reason.

Grokster, 545 U.S. at 936–37.

This well-settled law is exactly what the jury considered in determining Teva’s infringement liability.

The jury instruction on “Induced Infringement” included the “active steps” requirement:

Teva is liable for active inducement of a claim

only if GSK proves by a preponderance of the

evidence each of the following:

26

...

2. that Teva took some affirmative action, or

that Teva continued to take an action that began before the ’000 patent issued, after the ’000

patent was issued on January 8, 2008, intending to cause the physicians to directly infringe

by administering Teva’s carvedilol product;

....

C.A.App.168.

An additional jury instruction, entitled “Affirmative Actions Intended to Cause Infringement,” communicates the same point. C.A.App.171. It provides

in relevant part that, “to find inducement, you must

find that Teva intended others to use its products in

at least some ways that would infringe the asserted

claims of the ’000 patent, took affirmative acts to encourage direct infringement, and that those actions actually caused the direct infringement of the asserted

claims of the ’000 patent.” Id. (emphasis added). Yet

further, the jury knew “[i]t is not enough for the accused inducer to lead another to engage in conduct

that happens to amount in infringement. Rather, to

induce infringement, the accused inducer must persuade another to engage in conduct that the inducer

knows – or believes with high probability, but deliberately avoided confirming – is infringement.”

C.A.App.172.

These jury instructions correctly communicate

what is required to find liability for “active inducement,” including based on the statute’s plain language, Grokster and Global-Tech. In fact, Teva has

never challenged them.

27

In reviewing the jury’s verdict for substantial evidence, the Federal Circuit recognized and applied

these same, long-standing principles. It explained

that inducement liability only attaches where the accused infringer takes “active steps” to encourage infringement. Pet.App.12a–13a; see also id. at 28a–29a.

The Federal Circuit then applied that requirement.

For example, it concluded that “substantial evidence

in this case supports the jury’s determination that

Teva’s partial label contained information encouraging each claimed step and the preamble” and also

“finding that Teva’s partial label was evidence that

Teva instructed physicians to use its carvedilol in an

infringing way.” Pet.App.20a, 27a. It also made clear

that Teva’s other marketing materials, in addition to

its labels, support inducement. Pet.App.36a–37a,

42a. The requirement of active steps to encourage infringement was thus unquestionably central to the decision below.

Also central to the decision was the understanding that inducement will not follow where there is

“mere knowledge of infringing potential or of actual

infringing uses” and an absence of active encouragement to infringe—in other words, “legitimate commerce.” Grokster, 545 U.S. at 937. The Federal Circuit made clear it considered this in explaining that

GSK’s inducement expert “did not testify that Teva’s

actions merely describe infringement; he testified that

Teva’s

actions

encouraged

infringement.”

Pet.App.27a; see also id. at 37a.

What this all demonstrates is the law of induced patent infringement was correctly stated and

applied below. No additional look by this Court is

warranted. See Sup. Ct. R. 10 (providing that even

28

when a properly stated rule of law is misapplied, review is rare).

B. The Factual Finding Below that Teva’s

Partial Label Encourages Infringement

Does Not Conflict with Relevant Decisions of This Court

The factual nature of the issue of induced infringement disposes of Teva’s claim that the decision below

erases the line between encouragement to infringe (inducement) and the mere description of an infringing

use.

“Infringement is a question of fact.” Stilz v.

United States, 269 U.S. 144, 147 (1925). A central fact

question the jury decided was whether Teva’s partial

label instructed the patented use, notwithstanding

Teva’s carve-out attempt. The jury’s verdict answered

that question affirmatively. Teva tries to skirt that

problem in its petition in two ways, neither of which

pass muster.

First, Teva rides the dissent’s bandwagon that its

carve-out approach is non-infringing because it traces

back to GSK’s representations to FDA. This proposition falls apart when the applicable FDA regulations

are considered, and because the jury rejected that position. The jury found that even in the face of GSK’s

FDA submissions, Teva failed to carve-out the patented use and heed its affirmative responsibility—

confirmed by its regulatory expert (Pet.App.24a–

25a)—to ensure its actions would not constitute infringement.

Second, Teva re-asserts its rejected factual argument that what was left in its carve-out label does not

amount to encouragement to infringe because, at best,

the infringing use is merely described or inadvertently

29

mentioned in its partial label. But again, whether this

is so is a fact question for the jury to decide. The jury

decided against Teva. Lower court decisions addressing different labels and finding no infringement are no

basis to cast aside the jury’s fact finding. Just as this

case turns on its facts, so did the ones Teva invokes.

In Takeda Pharms. U.S.A., Inc. v. West-Ward

Pharm. Corp., 785 F.3d 625 (Fed. Cir. 2015), the patentee tried to prove a label encouraged the patented

use of a drug to treat acute gout flares based on the

label’s statement “that ‘[i]f you have a gout flare while

taking [the drug], tell your healthcare provider.’” Id.

at 630. The court held that based on the record before

it, the patentee had demonstrated “there may be some

infringing uses” of the drug, but there was insufficient

evidence of intentional encouragement to infringe. Id.

at 633–34. That this case and Takeda both involve a

carve-out label does not mean they should be decided

the same, as Teva advances. Pet. at 30–31; see also

Pet.App.13a (lower decision citing Takeda). Different

facts compel a different result here, as already explained.

Same for HZNP Medicines LLC v. Actavis Lab’ys

UT, Inc., 940 F.3d 680 (Fed. Cir. 2019). There, the

patented method involved three steps. There was no

inducement because the label only required the first

step. The district court serving as the fact finder

found insufficient evidence the label encouraged performing the last two steps. Id. at 702. By contrast

here, substantial evidence demonstrated Teva’s partial label encouraged all steps of the claimed method.

Pet.App.20a.

In Bayer Schering Pharma AG v. Lupin Ltd., 676

F.3d 1316, 1322–24 (Fed. Cir. 2012), the label in question was deemed to lack information demonstrating

30

the safety and efficacy of two of the three simultaneous effects recited in the patented method, and so

failed to recommend achieving that claimed combination of effects. That is not the case here. Pet.App.18a.

As the appellate court rightly stated, “[t]he jury

was entitled to credit expert testimony regarding the

label’s instructions on who should take what drug,

when, why, and how, and to reject the dissent’s claim

that the label describes rather than instructs as to an

infringing use.”

Pet.App.29a n.6; see also

Pet.App.18a.

Thus “inadvertent description” has not been

deemed to satisfy § 271(b) per the decision below, as

Teva claims. Pet. at 31. The issue of inducement vs.

mere description was and remains a fact question.

Teva’s dissatisfaction with this case’s outcome on the

factual record—which plainly supports the conclusion

Teva did not carve out the patented use, unlike in

Takeda—is no compelling basis for review. See Sup.

Ct. R. 10 (providing that even “where the asserted error consists of erroneous factual findings,” review is

rare).

II. THE SPECIFIC FACTS OF THIS CASE LIMIT ITS

REACH AND COUNSEL AGAINST REVIEW

Certiorari should be denied for the additional reason that this case concerns regulatory issues long

since mooted. FDA’s rules concerning disclosure of

patent information and use codes for Orange Book

listing have dramatically changed since GSK first

listed its patent in 1998. If the patent were listed today, GSK would submit a Form 3542 with a use code

and specifically identify whether the patent covered

all or just a portion of each approved use as well as

each section and subsection of the label that describes

31

the method of use. 21 C.F.R. § 314.53(c)(2)(i)(O)(2)

(2019). None of this information was required in 1998,

2003 or 2008. Indeed, in 1998, FDA merely “requested, but [did] not require[] NDA applicants [to]

submit to FDA information on the approved use

claimed by the patent.” 68 Fed. Reg. 36,682 (June 18,

2003).

Moreover, this case does not concern any overriding statutory issue like Caraco. Caraco Pharm.

Lab’ys, Ltd. v. Novo Nordisk A/S, 566 U.S. 399 (2012).

There an NDA holder submitted an overly broad use

code that prevented approval of a section viii carveout label even though it was undisputed the label

would not induce the infringing use. The Court reversed the Federal Circuit and remanded, holding

that “Caraco may bring a counterclaim seeking to ‘correct’ Novo’s use code ‘on the ground that’ [Novo’s

method of treatment patent] ‘does not claim . . . an

approved method of using the drug’— indeed, does not

claim two.” 566 U.S. at 426. As the Court noted, ‘‘[a]n

overbroad use code . . . throws a wrench into the FDA’s

ability to approve generic drugs as the statute contemplates.’’ Id. at 419. The Court went on to say, “[a] fix

is in order, but it must come from Congress or FDA.”

Id. at 428.

FDA heard the Court and proposed changes to the

Orange Book listing process and proposed a process

for challenging use codes submitted by NDA holders.

80 Fed. Reg. 6,802 (Feb. 6, 2015). FDA reiterated that

its “role in patent listing is ministerial and does not

involve substantive review of patents” and that use

codes are meant to alert applicants but not meant to

substitute applicant’s review of the patent and ap-

32

proved labeling. Id. at 6,828. After receiving comment, FDA implemented these changes in 2016. 81

Fed. Reg. 69,580, 69,643–69,644 (Oct. 6, 2016).

Here, there is no suggestion that the FDA-created

use code listed for Coreg® was overbroad. More importantly, FDA’s post-Caraco rule changes further

highlight that the regulatory scheme today and going

forward is dramatically different from the scheme

during the relevant time of this dispute. The Court

should not expend resources reviewing antiquated

FDA rules, the results of which will have no impact on

the future.

Nor does this case present any conflict between

Hatch-Waxman’s “same labeling” requirement and §

271(b), as an amici suggests. See Brief of 42 Professors, 6–8. The Hatch-Waxman Act does not, and cannot, override patent rights, as confirmed by the “same

labeling” provision itself. That provision expressly allows labeling edits when “aspects of the listed drug’s

labeling are protected by patent.” 21 C.F.R. §

314.127(a)(7) (2016). Amici’s reliance on a copyright

infringement case is thus inapposite—the “same labeling” provision contains no exemption for copyrighted material.

Lastly, the record in this case is not yet complete.

The case has been remanded to the district court for

further proceedings, including resolution of equitable

defenses relating to the Orange Book listing. And contrary to the Petition, the section viii issue presented

is not case dispositive. As described above, Teva

amended its label in 2011 to add back the carved-off

portions. The jury and appellate court found this full

label, along with all of Teva’s other acts, induced infringement from that point forward. Pet.App.37a–

39a. In fact, GSK presented evidence at trial that

33

more than 70% of its damages arose during the full

label period. A ruling on since-replaced FDA rules

would not negate Teva’s liability for the full label period.

In sum, this case is an exceptionally poor vehicle

for considering inducement in the context of section

viii carve-out labels. The fact-bound result flowed

from the application of well-settled inducement law to

unique facts under a long-replaced regulatory scheme.

A future dispute under then-applicable FDA rules, to

the extent such a case ever gets to the Court, would be

a far superior vehicle to address inducement in the

context of a section viii label. Moreover, this case’s

incomplete record coupled with Teva’s liability for the

full label period—barely mentioned in the Petition—

further support denial of the Petition.

III. THE SECTION VIII CARVE-OUT REGIME IS NOT IN

DANGER, NOR WILL EVERY CARVE-OUT SITUATION PRESENT UNDUE RISK OF INFRINGEMENT

LIABILITY

The doomsday scenario for section viii carve-outs

Teva’s petition portrays falls apart for a simple reason: the jury found, on the record before it, that Teva’s

partial label did not properly carve out GSK’s patented use. The fact-bound nature of this case is why

the decision below does not spell the end of carve-outs.

As long as generics fully and truly carve out the patented use, they can continue to enjoy the carve-out

statute’s protection.

A decade-old Federal Circuit case addressing an

attempted carve-out label proves the point. In AstraZeneca LP v. Apotex, Inc., 633 F.3d 1042, 1056 (Fed.

Cir. 2010), the generic was deemed not to have carved

out enough from its FDA-approved label to avoid the

34

patented use. Yet that decision had no chilling effect

on carve-outs, as Teva’s statistics show.3

Nor has this decision. The Federal Circuit first

reinstated the jury’s verdict in October 2020. In the

nearly two years since then, GSK is aware of only one

other case asserting a carve-out label failed to carve

out the patented use, and thus could give rise to inducement liability. Teva’s petition thus oversells the

idea that every carve-out will lead to a jury trial. Pet.

at 33.

Moreover, as explained above, today a patentholder must provide much more information to FDA

that FDA will use to assess a carve-out label. The patent-holder’s role in this process undercuts Teva’s

claim that carve-out launches are now accompanied

by prohibitive risk and will have grave consequences

to the market. Additionally, the rules now provide a

fast and convenient mechanism for testing the patent

information provided by the brand for Orange Book

listing. See, 21 CFR 314.53(f).

But, at bottom, the Hatch-Waxman Act’s carveout provision was never meant to snuff out patent

rights and give failed attempts to carve-out a patented

use a free pass just because a generic company might

get it wrong. Such instances are few and far between,

and this Court’s inducement law need not be changed

simply to account for the rare exception. In addition,

the facts here don’t even fit the exception. Contrary

to what Teva and amici argue, Teva’s carve-out was

not the result of any genuine effort to avoid patent

3 Bryan S. Walsh et al., Frequency of First Generic Drug Approv-

als With ‘Skinny Labels’ in the United States, 181 JAMA Intern.

Med.

995–997

(2021),

https://jamanetwork.com/journals/jamainternalmedicine/article-abstract/2777965.

35

rights, but rather was one of expediency, made at the

last minute, without any assessment of GSK’s patent

rights. No responsible generic or biosimilar company’s efforts under section viii could possibly be

threatened by the actual circumstances of this case.

CONCLUSION

Teva’s petition should be denied.

Respectfully submitted,

JUANITA R. BOOKS

Counsel of Record

JONATHAN E. SINGER

Fish & Richardson P.C.

12860 El Camino Real

Suite 400

San Diego, CA 92130

(858) 678-5070

brooks@fr.com

MICHAEL J. KANE

ELIZABETH M. FLANAGAN

Fish & Richardson P.C.

60 South Sixth Street

Suite 3200

Minneapolis, MN 55402

(612) 335-5070

eflanagan@fr.com

August 19, 2022

Counsel for Respondents

GlaxoSmithKline LLC,

and SmithKlineBeecham (Cork) Limited