Reply Brief — Sandoz Inc., et al., Petitioners v. Immunex Corporation, et al.

No. 20-1110

IN THE

Supreme Court of the United States

SANDOZ INC., SANDOZ INTERNATIONAL GMBH,

SANDOZ GMBH,

Petitioners,

v.

IMMUNEX CORP., AMGEN MANUFACTURING, LTD.,

Respondents.

On Petition for a Writ of Certiorari to the United

States Court of Appeals for the Federal Circuit

REPLY BRIEF FOR PETITIONERS

DAVID J. ZIMMER

GERARD J. CEDRONE

GOODWIN PROCTER LLP

100 Northern Avenue

Boston, MA 02210

GEORGE C. LOMBARDI

MAUREEN L. RURKA

JULIA MANO JOHNSON

WINSTON & STRAWN LLP

35 W. Wacker Drive

Chicago, IL 60601

April 26, 2021

WILLIAM M. JAY

Counsel of Record

BRIAN T. BURGESS

GOODWIN PROCTER LLP

1900 N Street, N.W.

Washington, DC 20036

wjay@goodwinlaw.com

(202) 346-4000

CINDY CHANG

GOODWIN PROCTER LLP

620 Eighth Avenue

New York, NY 10018

Counsel for Petitioners

RULE 29.6 STATEMENT

The corporate disclosure statement included in the

petition for a writ of certiorari remains accurate.

i

TABLE OF CONTENTS

Page

Rule 29.6 Statement ....................................................i

Introduction ................................................................ 1

Argument .................................................................... 2

I.

The Federal Circuit decided a purely

legal question, and its answer

undermines the essential protection

against double patenting............................... 2

II.

The question presented is important

and warrants this Court’s attention. ............ 7

III. Nothing prevents this Court from

reaching the question presented. .................. 9

IV. At the very least, this Court should

call for the views of the Solicitor

General......................................................... 11

Conclusion ................................................................ 12

ii

TABLE OF AUTHORITIES

Page(s)

Cases:

Geneva Pharm., Inc. v.

GlaxoSmithKline PLC,

349 F.3d 1373 (Fed. Cir. 2003) ............................ 10

Kimble v. Marvel Entm’t, LLC,

576 U.S. 446 (2015) ................................................ 6

In re Longi,

759 F.2d 887 (Fed. Cir. 1985) .............................. 10

Luminara Worldwide, LLC v. Liown

Elecs. Co.,

814 F.3d 1343 (Fed. Cir. 2016) .............................. 6

In re Mann,

47 F.2d 370 (C.C.P.A. 1931)................................. 10

Speedplay, Inc. v. Bebop, Inc.,

211 F.3d 1245 (Fed. Cir. 2000) .............................. 5

United States v. Stitt,

139 S. Ct. 399 (2018) ............................................ 10

Statutes:

35 U.S.C. § 102 ............................................................ 8

35 U.S.C. § 102(a) (2006) ............................................ 8

35 U.S.C. § 102(b)(2) ................................................... 8

35 U.S.C. § 102(e) (2006) ............................................. 8

35 U.S.C. § 112 ............................................................ 8

35 U.S.C. § 122(b) ........................................................ 8

iii

Leahy-Smith America Invents Act, Pub.

L. No. 112-29, 125 Stat. 284 (2011) ....................... 8

§ 3(n), 125 Stat. 293 .............................................. 8

Other Authority:

Dennis Crouch, Buying Up Overlapping

Patents—And Double Patenting,

PATENTLY-O, http://patentlyo.com

/patent/2021/02/overlapping-patentspatenting.html (Feb. 15, 2021) .............................. 8

iv

INTRODUCTION

A single person or entity may not receive more

than one patent on the same invention, or obvious variants of it. That is the rule against obviousness-type

double patenting that this Court has long recognized.

But the Federal Circuit has now walked back that crucial protection. Under the decision below, a patentee

may enjoy full control over two patents covering the

same technology in essentially the same way and

avoid ODP scrutiny so long as its second patent is

nominally owned by someone else. How little does the

nominal owner need to retain? A secondary right to

sue for infringement—even if that secondary right exists only on paper and is meaningless in practice. As

Sandoz has explained (see Pet. 15-29), that decision

warrants this Court’s attention because it guts the Patent Act’s one-patent-per-invention requirement and

creates a blueprint for other patentees to follow Immunex’s straw-owner strategy.

Immunex argues that the Federal Circuit’s decision is a fact-bound one-off, but only by reimagining

the majority opinion as a “totality-of-the-circumstances assessment” that “did not rely solely on

Roche’s retained right to enforce its patents.” Opp. 1213. That is incorrect. The panel majority deemed

Roche’s illusory right to sue, without more, to be “thoroughly inconsistent” with an assignment of ownership. Pet. App. 21a. That holding answers a purely

legal question—what is the minimum needed to avoid

a finding of ownership and skirt ODP scrutiny?—and

the resulting new pathway around ODP will have implications far beyond this specific case.

2

Immunex also advances various prudential reasons to deny certiorari, but its arguments do not withstand scrutiny. For example, Immunex argues (at 2223) that facts like these are unlikely to recur following

a change to patent term resulting from the Uruguay

Round of the General Agreement on Tariffs and Trade

(GATT). But “the patent regime Immunex exploited

still governs an enormous number of patents.”

AAM/AHIP Amicus Br. 10. And, regardless, patentees can deploy Immunex’s straw-owner gambit on

post-GATT patents, too. As for Immunex’s attempt

(at 24-28) to interpose new threshold questions or alternative grounds for affirmance, Immunex has identified no issue that would prevent the Court from

reaching the question presented.

The Court should grant certiorari and reverse. At

a minimum, this Court should not deny certiorari

without inviting the United States to give its views on

this new pathway around ODP.

ARGUMENT

I.

The Federal Circuit decided a purely legal question, and its answer undermines

the essential protection against double

patenting.

According to the Federal Circuit, Immunex was

not the effective owner of the patents-in-suit because—and only because—Roche maintained a secondary right to sue that would vest only if Immunex

allowed it to. See Pet. App. 21a. That decision flouts

a long line of precedent, in both this Court and the

Federal Circuit, enforcing the Patent Act’s limitations

on double patenting. See Pet. 15-22.

3

Immunex goes to great lengths to avoid defending

the holding that Roche’s entirely hypothetical secondary right to sue was sufficient to keep Roche the effective owner. But Immunex’s account of the decision below does not line up with the court of appeals’ actual

analysis.

1. Immunex first argues (at 11-13) that the Federal Circuit adopted a “totality-of-the-circumstances

assessment,” and so the decision below is limited to

this specific agreement. But the decision below did

not rest on an amorphous totality analysis. Before the

Federal Circuit, Immunex pointed to four separate

“rights” that, in its view, showed that Roche was the

effective patentee. See Pet. App. 21a. The court

trained its attention on just one of them—Roche’s secondary right to sue, which Immunex could prevent

from ever vesting—and deemed it “‘thoroughly inconsistent’ with a conclusion that the patents-in-suit

were effectively assigned to Immunex.” Pet. App. 21a.

The court made that determination independent of its

consideration of any other provision of the 2004

Agreement. That square legal holding—that a secondary and voidable right to sue for infringement can

defeat a finding of ownership—extends beyond the

facts of this specific agreement. See Pet. 23-29;

AAM/AHIP Amicus Br. 8-10.

Immunex identifies other terms of the 2004 Agreement, but the Federal Circuit did not treat them as

necessary to its decision. For example, Immunex

points (at 12) to Roche’s right to veto any assignment

of Immunex’s interests. But the panel majority said

only that this restriction on assignment was “a further

indication” that Roche had not transferred all substantial rights—not that it was an independent basis

4

for its holding. Pet. App. 23a. Nor could the majority

have rested on that restriction on alienation, because

it was reciprocal. If Immunex’s argument were taken

seriously, no one would own the patents-in-suit because Roche, too, lacked unilateral assignment authority. See C.A. App. 25849. And even this restriction on assignment lasts only so long as Immunex

desires to maintain the fiction of Roche’s ownership:

Immunex has the absolute right to assign the patents

to whomever it wishes, as long as it first pays Roche

$50,000 for its remaining rights. Roche has no right

to say no. Pet. 10; Pet. App. 40a (Reyna, J., dissenting); p. 6, infra.

Immunex also cites (at 12-13) restrictions on Immunex’s ability to terminate the 2004 Agreement and

Roche’s right to practice the patents for internal research purposes. But the Federal Circuit simply did

not rely on these provisions in its all-substantialrights analysis. See Pet. App. 21a-24a. In the court’s

view, Roche’s secondary right to sue was sufficient, all

by itself, to support the court’s holding.

2. As a fallback, Immunex argues (at 13) that

Roche’s secondary right to sue was not illusory. But

like the Federal Circuit, Immunex never grapples

with the key fact: Immunex could prevent Roche from

exercising that right by granting a royalty-free sublicense before the right ever vested. See Pet. 13-14, 22.

Immunex observes (at 13) that, according to the court

of appeals, Immunex could not issue a sublicense after

180 days, “once Roche’s secondary right to sue is triggered.” But that is beside the point: “Immunex can

issue a royalty-free sublicense within 180 days of receiving Roche’s written request to correct infringement and can thus prevent Roche’s secondary right to

5

sue from even vesting.” Pet. App. 43a (Reyna, J., dissenting) (emphasis added). The Federal Circuit never

held otherwise—and Immunex simply ignores the

point. 1 Immunex’s argument is thus like arguing that

the President lacks power to veto a bill once ten days

go by: true, but no obstacle to the President’s vetoing

any bill he wants.

In a footnote, Immunex argues (at 14 n.4) that the

Federal Circuit has previously held that a secondary

right to sue “will not defeat a transfer of ownership”

where the secondary right can be avoided by the grant

of a sublicense. See Speedplay, Inc. v. Bebop, Inc., 211

F.3d 1245, 1251-1252 (Fed. Cir. 2000). But Speedplay

is now a dead letter: The Federal Circuit in this case

held that sublicenses are irrelevant unless they can be

given even to a defendant the nominal owner has sued.

Pet. App. 22a. That is a null set, as just discussed,

because of the power to sublicense before the right to

sue vests. Neither an illusory right to sue nor an illusory limit on the sublicensing power gives Roche any

substantial rights.

3. Immunex also claims (at 16-20) that it is correct

on the merits, but its arguments only underscore just

how illusory Roche’s “ownership” was. For example,

Immunex notes (at 18) that Roche retained a right to

practice the patents for internal, non-clinical re-

1 To be clear, the Federal Circuit was wrong to construe the 2004

Agreement to bar Immunex from issuing a sublicense after 180

days: The court fashioned that atextual rule from a provision requiring Immunex to “cooperate” with any Roche-initiated suit.

Pet. App. 20a, 24a. For the reasons described above, however,

Roche’s right is illusory even accepting the Federal Circuit’s

flawed construction.

6

search. But although the Federal Circuit “mention[ed]” that right (Opp. 18), it ultimately did not rely

on it. And for good reason: Precedent makes clear that

“this is not a substantial right.” Luminara Worldwide, LLC v. Liown Elecs. Co., 814 F.3d 1343, 1351

(Fed. Cir. 2016).

Immunex emphasizes (at 19) that it could convert

its supposed license into an outright assignment for

an additional $50,000. But that trivial amount confirms that Roche’s “ownership” was illusory. Immunex argues (at 19) that the $50,000 amount must

“be viewed in the context of the entirety of the agreement.” Exactly right. Immunex paid $45 million to

acquire the rights it obtained in the 2004 Agreement;

Roche was willing to give an outright assignment for

the same price. See Pet. 9-10, 15. That Immunex

could obtain any remaining rights for just $50,000 (or

0.1% more) shows that the remaining rights were insubstantial. See Pet. 10, 21. Immunex’s own counsel

acknowledged (C.A. Oral Arg. 41:01) that the $50,000

sum was a mere “peppercorn”—consideration for consideration’s sake. It could have been one dollar.

A peppercorn is a small price to pay for the power

to dodge the protection against double patenting. Yet

that is exactly what the Federal Circuit has allowed

clever patentees to do: to enjoy the upside of a second

patent that remains nominally in the hands of a straw

owner, and thus to extend their patent exclusivity

past the statutory cut-off date for the term of “a” patent. “This Court has carefully guarded that cut-off

date,” Kimble v. Marvel Entm’t, LLC, 576 U.S. 446,

451 (2015), but the Federal Circuit has now dug a tunnel past the guard post. See Pet. 15-22. Immunex will

not be the last to use it.

7

II.

The question presented is important and

warrants this Court’s attention.

The Federal Circuit’s decision provides a clear

roadmap for patentees to circumvent the one-patentper-invention requirement. See Pet. 23-26. As amici

have explained, “[a]bsent review by this Court, little

stands in the way of other companies applying respondent’s blueprint to their own expiring patents.”

AAM/AHIP Amicus Br. 3. Downplaying the effect of

the decision below, Immunex argues (at 22-24) that its

success in extending its patent term stems from a

unique combination of facts that is unlikely to recur.

But this case is not as unique as Immunex suggests.

1. Immunex’s principal argument (at 22-23) is

that its patent-extension gambit is impossible to replicate post-GATT. But that is wrong for at least two

reasons.

First, the well of pre-GATT patent applications has

not yet run dry. As Sandoz explained (Pet. 26 & n.10),

a number of pre-GATT applications are still pending

in the Patent Office. With respect to those outstanding patent applications, a patentee could replicate Immunex’s exact same patent-extension strategy to avoid

ODP scrutiny.

Second, even putting aside those pre-GATT applications, a patentee could readily extend its exclusivity

and dodge ODP using a post-GATT patent application. Immunex notes (at 23) that, in a post-GATT

world, an earlier-filed patent application will always

expire before a later-filed application—regardless of

when the two patents issue. See Pet. App. 144a-145a.

But that is of no moment: An inventor can still deploy

Immunex’s strategy. After filing for its own patent,

8

an Immunex imitator could take over and repurpose a

later-filed patent application by another inventor,

making sure to leave that other inventor with an illusory secondary right to sue. So long as the first patent

application is not “prior art” to the second patent application, both patents will issue—and the second will

effectively extend the term of the first, because the Patent Office and later the Federal Circuit will not apply

ODP. 2

In short, unwarranted extensions can still be engineered post-GATT, even if the timing differs. Thus,

as one commentator has observed, there is still a

“steady demand for” ODP protections in a post-GATT

regime. Dennis Crouch, Buying Up Overlapping Patents—And Double Patenting, PATENTLY-O, http://

patentlyo.com/patent/2021/02/overlapping-patentspatenting.html (Feb. 15, 2021).

2. Immunex also argues (at 23) that “familiar

statutory requirements” like the written-description

requirement of § 112 and the novelty requirement of

§ 102 will prevent any gamesmanship. But this case

demonstrates otherwise. Neither the written-description requirement nor the rules against obviousness

There are any number of reasons why the first patent application may not be “prior art” to the second. For applications claiming priority to before the Leahy-Smith America Invents Act

(AIA), Pub. L. No. 112-29, 125 Stat. 284 (2011), an earlier patent

application would not be prior art depending on when or whether

it was published. See 35 U.S.C. § 102(a) and (e) (2006); see also

id. § 122(b) (describing when patent applications are published);

AIA § 3(n), 125 Stat. at 293 (effective date of amendment). For

post-AIA applications, an earlier patent application would not be

prior art if it falls within one of the exclusions set forth in 35

U.S.C. § 102(b)(2).

2

9

stood in the way of Immunex’s ability to secure and

repurpose a second patent and thereby extend its patent term. See Pet. App. 24a-34a. To be clear, Sandoz

disagrees with the district court’s findings and Federal Circuit’s affirmance on those issues. But even accepting those holdings, the prohibition on obviousness-type double patenting is supposed to be an independent check—barring a single player from gaming

the patent system even when the written-description

and nonobviousness requirements do not. And yet the

Federal Circuit weakened that critical bulwark.

III.

Nothing prevents this Court from reaching the question presented.

Immunex spends the balance of its brief raising a

host of vehicle objections, but nothing prevents the

Court from addressing this important issue in this

case.

1. First, Immunex argues (at 24-26) that if the

Court grants certiorari, “it would first need to address

the threshold question whether the all-substantialrights inquiry . . . is consistent with the Patent Act.”

But Immunex is not questioning the unanimous holding below (Pet. App. 16a-17a) that the way to assess

common ownership was with the all-substantialrights test. Indeed, Immunex did not propose any different test for common ownership. Pet. C.A. Reply 4.

Rather, as Immunex acknowledges (at 25), it is disputing decades of precedent about the timing of common ownership. Courts have long recognized that the

rule against double patenting extends to situations in

which the double patentee received the second patent

by way of an assignment, rather than as the named

inventor. See Pet. 18 (citing cases). But Immunex

10

suggests (at 26) that even if double patenting can be

grounded in common ownership, it must be common

ownership at the time of invention. That argument,

too, is foreclosed by well-established precedent. See

In re Longi, 759 F.2d 887, 893-895 (Fed. Cir. 1985);

Geneva Pharm., Inc. v. GlaxoSmithKline PLC, 349

F.3d 1373, 1377, 1386 (Fed. Cir. 2003); In re Mann, 47

F.2d 370, 371-72 (C.C.P.A. 1931). The timing question

is merely a distraction; it does not block this Court

from reaching the question presented.

2. Immunex offers a second distraction, arguing

(at 26-28) that it might win on alternative grounds

based on other findings made by the district court.

But the Federal Circuit did not reach these alternative holdings: It rested its decision solely on its common-ownership determination. And so this Court, “a

court of review, not of first view,” lacks a basis to consider those alternative holdings. E.g., United States

v. Stitt, 139 S. Ct. 399, 407 (2018). The fact that Immunex might press alternative bases for affirmance

on a future remand from this Court is irrelevant.

In any event, Immunex’s arguments are wrong.

Immunex claims (at 27) that the patents-in-suit are

patentably distinct from the ’690 Patent. Sandoz disputed that finding below, but even accepting it for the

sake of argument, it does not affect the ’225 Patent.

That patent is not patentably distinct from the patents-in-suit. As Judge Reyna explained, the district

court concluded otherwise only by applying an erroneous legal standard. See Pet. App. 43a-45a (dissenting

opinion).

Immunex also argues (at 9, 27-28) that the district

court held that the ’225 Patent could not serve as a

11

reference patent for ODP purposes as a matter of law,

but here Immunex is just misreading the district

court’s opinion. The district court concluded that, as

a general matter, “the statutory term for the Patentsin-Suit [could] not be cut short to mirror the statutory

term for the Finck Patents” (i.e., a family of patents

that included the ’225 Patent). Pet. App. 146a. But

the court went on to hold that “the ’225 Finck Patent

. . . could be properly considered an ‘earlier patent’ for

an obviousness-type double patenting analysis.” Pet.

App. 146a n.43 (emphasis added).

3. Finally, Immunex throws up a host of factual

issues that are immaterial to either the question presented or the alternative grounds on which it would

seek affirmance. Indeed, many of them are findings

that (even while ruling for Immunex) the court of appeals held the district court “should not have made” in

answering the legal question of ODP, Pet. App. 19a,

such as the district court’s finding that the subjective

purpose of the 2004 Agreement was to create a license.

Opp. 6-7. Others come from portions of the opinions

below that did not involve ODP, such as the district

court’s finding that the Roche application could be

read to describe etanercept for written-description

purposes. Opp. 3-5, 7. Sandoz disputes each of these

findings. But in the end they are legally irrelevant to

the question presented—again, Immunex simply

raises them to distract from the issue at hand.

IV.

At the very least, this Court should call

for the views of the Solicitor General.

For all the foregoing reasons, the Court should

grant the petition now. At the very least, however,

the Court should invite the views of the United States.

12

The Patent Office assesses ODP during examination,

and the Federal Circuit’s decision can be read to allow

parties like Immunex to control prosecution without

necessarily disclosing their ownership of all substantial rights. The Court would benefit from the government’s views on whether the Federal Circuit’s rule is

either workable or correct, particularly for patent-examination purposes.

CONCLUSION

The petition for a writ of certiorari should be

granted.

Respectfully submitted.

DAVID J. ZIMMER

GERARD J. CEDRONE

GOODWIN PROCTER LLP

100 Northern Avenue

Boston, MA 02210

GEORGE C. LOMBARDI

MAUREEN L. RURKA

JULIA MANO JOHNSON

WINSTON & STRAWN LLP

35 W. Wacker Drive

Chicago, IL 60601

April 26, 2021

WILLIAM M. JAY

Counsel of Record

BRIAN T. BURGESS

GOODWIN PROCTER LLP

1900 N Street, N.W.

Washington, DC 20036

wjay@goodwinlaw.com

(202) 346-4000

CINDY CHANG

GOODWIN PROCTER LLP

620 Eighth Avenue

New York, NY 10018

Counsel for Petitioners