Amgen Inc. et al v. Hospira, Inc.

Filing 386

MEMORANDUM OPINION. Signed by Judge Richard G. Andrews on 8/27/2018. (nms)

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 1 of 47 PageID #: 16168

IN THE UNITED STATES DISTRICT COURT
FOR THE DISTRICT OF DELAWARE

AMGEN INC. and AMGEN
MANUFACTURING, LIMITED,
Plaintiffs,
V.

Civil Action No . 15-cv-839-RGA

HOSPIRA, INC.,
Defendant.

MEMORANDUM OPINION

Robert W. Whetzel and Jason J. Rawnsley, RICHARDS, LAYTON & FINGER, P.A. ,
Wilmington, DE; Kevin M. Flowers, Mark H. Izraelewicz, John R. Labbe, Julianne M . Hartzell,
Benjamin T. Horton, Tiffany D. Gehrke, and Douglas G. Bolesch, MARSHALL, GERSTEIN &
BORUN LLP, Chicago, IL; Nicholas Groombridge, Eric Alan Stone, Jennifer H. Wu, and Stephen
A Maniscalco, PAUL, WEISS, RIFKIND, WHARTON & GARRISON LLP, New York, NY.
Attorneys for Plaintiffs
Dominick T. Gattuso, HEYMAN ENERIO GATTUSO & HIRZEL LLP, Wilmington, DE;
Thomas J. Meloro, Michael W. Johnson, Heather M. Schneider, Dan Constantinescu, M. Diana
Danca, Tara L. Thieme, and Philip F. DiSanto, WILKIE FARR & GALLAGHER LLP, New York,
NY.
Attorneys for Defendant.

August

r{_,2018

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 2 of 47 PageID #: 16169

On September 18, 2015, Amgen, Inc. and Amgen Manufacturing, Limited (collectively,
"Amgen") sued Hospira, Inc. for infringement of U.S. Patent No. 5,856,298 under 35 U.S.C.
§§ 271(a) and 271(e)(2)(C) and for infringement of U.S. Pat. No. 5,756,349 under§ 271(a). (D.I.
1). The '298 patent and the ' 349 patent cover erythropoietin ("epoetin" or "EPO") isoforms and
aspects of their production. Hospira submitted Biologic License Application ("BLA") No. 125545 to the FDA in December 2014, seeking FDA approval for Hospira' s epoetin biosimilar
product. (D.I. 290-1 at 1). Amgen asserts that Hospira' s manufacture of drug substance for its
epoetin biosimilar drug product infringes claims 24 and 27 of the ' 298 patent and claims 1-7 of
the ' 349 patent. (D.I. 290 at 1).
I held a jury trial from September 18-22, 2017. (D.I. 328-332 ("Trial Tr.")). 1 The jury
found each of the asserted claims not proved invalid, decided that the asserted claims of the ' 349
patent were not infringed, and returned a verdict of infringement of all asserted claims of the ' 298
patent. (D.I. 325 at 2). Of Hospira' s twenty-one accused drug substance batches, the jury found
seven batches entitled to the safe harbor defense. (Id. at 3). The jury awarded Amgen $70 million
in damages for Hospira's infringement. (Id. at 4).

Presently before the Court are Hospira' s Rule 50(a) Motion for Judgment as a Matter of
Law on the Issues of Safe Harbor, Noninfringement, Invalidity, and Damages and related briefing
(D.I. 336, 337, 348, 351), Hospira' s Motion for Judgment as a Matter of Law Under Rule 50(b)
and, in the Alternative, For Remittitur or New Trial Under Rule 59 and related briefing (D.I. 355 ,
357, 374, 381), Hospira' s Motion to Seal Confidential Exhibits Admitted at Trial and related
briefing (D.I. 361 , 369, 370), Amgen' s Renewed Motion for Judgment as a Matter of Law of

1

The trial transcript is consecutively paginated. References to the trial transcript will refer to "Trial Tr." in lieu of the
docket item reference number.

1

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 3 of 47 PageID #: 16170

Infringement of the '349 Patent or, in the Alternative, for a New Trial and related briefing (D.I.
356,358, 373 , 380), and Amgen's Motion for Prejudgment and Post-judgment Interest and related
briefing (D.I. 352, 376, 382).
I.

LEGAL STAND ARDS
A. Judgment as a Matter of Law

Judgment as a matter of law is appropriate if "the court finds that a reasonable jury would
not have a legally sufficient evidentiary basis to find for [a] party" on an issue. FED. R. C1v. P.
50(a)(l). "Entry of judgment as a matter oflaw is a 'sparingly' invoked remedy, granted only if,
viewing the evidence in the light most favorable to the nonmovant and giving it the advantage of
every fair and reasonable inference, there is insufficient evidence from which a jury reasonably
could find liability." Marra v. Phi/a. Haus. Auth. , 497 F.3d 286, 300 (3d Cir. 2007) (citation
omitted).
In assessing the sufficiency of the evidence, the Court must give the nonmovant, "as [the]
verdict winner, the benefit of all logical inferences that could be drawn from the evidence
presented, resolve all conflicts in the evidence in his favor and, in general, view the record in the
light most favorable to him." Williamson v. Consol. Rail Corp., 926 F.2d 1344, 1348 (3d Cir.
1991 ). The Court may "not determine the credibility of the witnesses [nor] substitute its choice
for that of the jury between conflicting elements in the evidence."

Perkin-Elmer Corp. v.

Computervision Corp., 732 F.2d 888, 893 (Fed. Cir. 1984). Rather, the Court must determine

whether the evidence reasonably supports the jury's verdict. See Gomez v. Allegheny Health Servs.
Inc., 71 F.3d 1079, 1083 (3d Cir. 1995); 9B Charles Alan Wright & Arthur R. Miller, Federal
Practice and Procedure § 2524 (3d ed. 2008) ("The question is not whether there is literally no

evidence supporting the party against whom the motion is directed but whether there is evidence

2

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 4 of 47 PageID #: 16171

upon which the jury might reasonably find a verdict for that party.").
Where the movant bears the burden of proof, the Third Circuit applies a stricter standard.

Fireman 's Fund Ins. Co. v. Videfreeze Corp. , 540 F.2d 1171 , 1177 (3d Cir. 1976). To grant
judgment as a matter of law in favor of a party that bears the burden of proof on an issue, the Court
"must be able to say not only that there is sufficient evidence to support the [movant' s proposed]
finding, even though other evidence could support as well a contrary finding, but additionally that
there is insufficient evidence for permitting any different finding. " Id.
B. New Trial
Federal Rule of Civil Procedure 59(a)(l)(A) provides, in pertinent part: "The court may,
on motion, grant a new trial on all or some of the issues-and to any party-

.. . after a jury trial,

for any reason for which a new trial has heretofore been granted in an action at law in federal
court ... ." Among the most common reasons for granting a new trial are: "(1) when the jury' s
verdict is against the clear weight of the evidence, and a new trial must be granted to prevent a
miscarriage of justice; (2) when newly discovered evidence exists that would likely alter the
outcome of the trial; (3) when improper conduct by an attorney or the court unfairly influenced the
verdict; or (4) when the jury' s verdict was facially inconsistent." See Zarow-Smith v. NJ Transit

Rail Operations, Inc., 953 F. Supp. 581 , 584-85 (D.N.J. 1997) (citations omitted).
The decision to grant or deny a new trial is committed to the sound discretion of the district
court. Allied Chem. Corp. v. Daiflon, Inc., 449 U.S. 33 , 36 (1980); Olefins Trading, Inc. v. Han

Yang Chem Corp., 9 F.3d 282, 289 (3d Cir. 1993). Although the standard for granting a new trial
is less rigorous than the standard for granting judgment as a matter of law- in that the Court need
not view the evidence in the light most favorable to the verdict winner-a new trial should only
be granted where "a miscarriage of justice would result if the verdict were to stand" or where the

3

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 5 of 47 PageID #: 16172

verdict "cries out to be overturned" or "shocks [the] conscience." Williamson, 926 F.2d at 135253.
II.

HOSPIRA'S SO AND SO MOTIONS
(a)
(b)

Hospira's Rule 50(a) motion raises the same issues as its Rule 50(b) motion.2 Having
considered and decided the issues in ruling on Hospira's Rule 50(b) motion, I will dismiss
Hospira's Rule 50(a) motion as moot.
Hospira seeks judgment as a matter of law on the issues of the applicability of its safe
harbor defense, noninfringement and invalidity of the '298 patent, and damages. (D.I. 357, pp. 122). Alternatively, Hospira seeks a new trial based on what it characterizes as improper jury
instructions on the safe harbor defense and third party liability, improper claim construction, and
contradictory infringement and validity verdicts. (Id. pp. 22-30). Finally, Hospira argues that it
is entitled to a remittitur of the damages award. (Id. p. 28).
A. JMOL

I. Safe Harbor
The parties dispute whether any reasonable jury could have found some, but not all, of
Hospira's drug substance batches protected by the "safe harbor" defense. (Id. p. 1; D.I. 374, p. 2).
The Biologics Price Competition and Innovation Act of 2009 ("BPCIA") "create[s] an
artificial 'act of infringement,' similar to that of 35 U.S.C. § 271(e)(2)(A), and [allows]
infringement suits to begin based on the filing of a biosimilar application prior to FDA approval
and prior to marketing of the biological product." Amgen Inc. v. Sandoz Inc. , 877 F .3d 1315, 1321
(Fed. Cir. 2017) (citing 35 U.S.C. § 271(e)(2)(C), (e)(4), (e)(6)). Section 271(e)(l) carves out an

2

Hospira's Rule 50(a) motion, filed based on information known at the close of Amgen 's case-in-chief, raises the
issue ofnoninfringement of the '349 patent. (D.I. 337 at 10). This issue was mooted when Hospira received a verdict
ofnoninfringement of the '349 patent at trial, and Hospira's Rule 50(b) motion does not raise it. (D.I. 325 at 2).

4

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 6 of 47 PageID #: 16173

exception to this rule, creating a "safe harbor" defense for defendants when their otherwiseinfringing activities are "solely for uses reasonably related" to obtaining FDA approval. 35 U.S.C.

§ 271(e)(l) ("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 a Federal law which regulates the
manufacture, use, or sale of drugs or veterinary biological products."); Abtox, Inc. v. Exitron Corp. ,
122 F.3d 1019, 1027 (Fed. Cir. 1997) ("By its terms, this shield from infringement permits use of
'patented invention[s]' to acquire information for regulatory approval of 'drugs or veterinary
biological products. " ') (brackets in original). "As long as the activity is reasonably related to FDA
approval, [a party' s] intent or alternative uses are irrelevant to its qualification to invoke the section
271(e)(l) shield." Abtox, 122 F.3d at 1030.
Hospira asserts that no reasonable jury could find that the safe harbor defense did not
protect each of its twenty-one drug substance batches. (D.I. 357, p. 1). Additionally, Hospira
contends that Amgen's arguments improperly limited the applicability of the safe harbor defense
to batches required for FDA approval. (Id. p. 5). Since each batch was used for one or more of
biosimilarity3 testing, updating product specifications, process validation, stability testing, or
continued process verification, Hospira insists that no reasonable jury could have found that each
of the batches was not reasonably related to obtaining FDA approval. (Id. pp. 4-8). According to
Hospira, the jury improperly second-guessed the number of batches that Hospira manufactured,

3

Hospira cites draft guidance from the FDA, published on September 22, 2017 and distributed for comment purposes
only, as further support for its argument that each of the 2013 drug substance batches were reasonably related to FDA
approval. (D.I. 357, p. 4). The draft guidance "recommend[s] a minimum of 10 reference product lots be sampled"
in order "to establish meaningful similarity acceptance criteria." (D.I. 357-1 at 7).
Hospira seeks to rely on the draft guidance to support its argument that no reasonable jury could have
concluded that the safe harbor did not protect each drug substance batch that Hospira used for biosimilarity testing.
The draft guidance was not publicly available at the time of trial, let alone at the time Hospira manufactured the drug
substance batches at issue. The draft guidance was not presented to the jury. I thus find Hospira' s reliance on the
draft guidance misplaced.

5

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 7 of 47 PageID #: 16174

when "the subjective reason why any batch was made is not relevant" to whether the safe harbor
applies. (D.I. 381 , pp. 3-4). Essentially, Hospira argues that since Hospira generated test data for
each batch prior to FDA approval, each batch could conceivably have been used to respond to
inquiries from the FDA, and each batch was reasonably related to FDA approval.
Amgen disagrees and maintains that substantial evidence supports the jury' s reasonable
conclusion that Hospira failed to prove that the safe harbor defense applies to each of Hospira' s
drug substance batches. (D.I. 374, p. 2). According to Amgen, the evidence presented at trial gave
the jury ample reason to reject Hospira' s arguments about biosirnilarity, product specifications,
process validation, stability, and continued process verification; credit Amgen' s witnesses; and
conclude that the safe harbor applied to only seven of Hospira' s twenty-one drug substance
batches. (Id. pp. 2-4).
Regarding biosimilarity, even accepting as true that ten reference product lots are required
to establish biosimilarity, Amgen points out that Hospira performed biosimilarity testing on drug
product batches, not drug substance batches, and that Hospira had previously manufactured
twenty-six drug product batches from four drug substance batches. (Id. p. 12 (citing Trial Tr.
811 :24-812:8; DTX-266 at 3-4)).

Therefore, Amgen argues, though Hospira performed

biosimilarity testing on nine drug substance batches, "the jury reasonably concluded that the final
two of those batches were not made for uses reasonably related to seeking FDA approval where
Hospira had made 26 drug product batches from just 4 of those drug substance batches for
biosimilaritytesting." (Id. pp. 12-13).
Amgen also argues, "Hospira' s witnesses admitted, and its submissions to the FDA
confirmed, that the FDA never required Hospira to manufacture any additional batches of its drug
substance to support its narrowed release specifications." (Id. p. 10 (citing Trial Tr. 823 :4-824:1)).

6

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 8 of 47 PageID #: 16175

Regardless, Hospira would be required to perform release testing on all batches manufactured
before or after FDA approval to ensure that the batches complied with the release specifications in
place at the time of manufacture. (Trial Tr. 819:11-22). Each of the batches at issue in this case
"were released against specifications that were in place at the time of manufacture, not against
revised specifications," and they remain available to Hospira for future use, since they comply
with the release specifications at the time of their manufacture (D.I. 374, p. 10 (citing Trial Tr.
820:24-822: 1)). Therefore, Amgen asserts, Hospira' s revised product specifications do not justify
the conclusion that product specification testing on each of the batches at issue was reasonably
related to obtaining FDA approval.
Regarding process validation, Amgen submits that the "Process Validation and/or
Evaluation" section of Hospira's BLA does not refer to batches other than those admitted by
Amgen or found by the jury to fall within the safe harbor. (Id. p. 11 (citing DTX-250)). Amgen
points out that even Hospira's updated BLA does not list any of the fourteen batches that the jury
found to fall outside the safe harbor defense. (Id.). Additionally, Amgen notes, cleaning validation
is "the only specific process validation that Hospira raises in its motion," and "Hospira' s FDA
expert, Dr. Levine, admitted that cleaning validation need not be completed before FDA
approval." 4 (Id. (citing Trial Tr. 1102:14-24, 1153:16-18); see also Trial Tr. 878:5-18 (Dr.
Billingham acknowledging same)).
Stability testing would not have required a reasonable jury to conclude that each of
Hospira's batches was protected by the safe harbor, Amgen argues, because FDA guidance

4

Amgen maintains that Hospira relied on a cleaning validation document authored by GSK for Hospira (DTX-252),
in lieu of presenting any cleaning validation data submitted to the FDA. (D.I. 374, p. 12). In reply, Hospira responds,
"Dr. Billingham clearly testified that all of the 2013, 20 14, and some 20 15 batches were used in several cleaning
validation studies to be completed before FDA inspection," and that the FDA requires that such cleaning validation
studies be performed. (D.1. 381, p. 7).

7

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 9 of 47 PageID #: 16176

requires only three batches to demonstrate stability before obtaining FDA approval. (D.I. 374, p. 9
(citing PTX-492, p. 3)). Amgen argues that the jury was therefore free to credit Dr. Martin-Moe ' s
testimony that Hospira's five batches from 2009 to 2012 would have provided sufficient stability
data. (Id. (citing Trial Tr. 1329:6-1331 :9)). Additionally, stability testing was required each time
a new drug substance batch was made, regardless of the future uses for the batch. (Trial Tr. 1338:21339:5). As further support, Amgen cites an internal Hospira Risk Authorization document
confirming Hospira' s belief that material from drug substance batches manufactured in 2009 and
2012 was sufficient to "support [the drug substance] ' shelf life and commercial saleability of
material produced in subsequent campaigns."' (D.I. 374, pp. 9-10 (quoting PTX-342 at 1)). The
Risk Authorization further states, "The balance of the material from the 2013 campaign
(approximately 50%) and most of the material from the 2014 and 2015 campaigns will serve as
commercial inventory to support single dose vial launch stock." 5 (PTX-342 at 1).
Like stability testing, Amgen's witnesses testified that continued process verification is "an
ongoing program ... during routine commercial production" that sometimes "can take many years
to complete." (D.I. 374, p. 8 (citing Trial Tr. 1336:21-1337:9)). Though the FDA requires that
applicants have committed to a continued process verification program before approval,
completing continued process verification is not required to obtain FDA approval. (Trial Tr.
1337:10-13; see also PTX-435, p. 14). Hospira's witnesses confirmed that continued process
verification need not be completed before FDA approval, and Hospira made no commitment to
manufacture the thirty batches tested for continued process verification prior to FDA approval.
(Trial Tr. 752 :7-11 (Ms. Dianis), 747:17-748:3 (Ms. Dianis), 883:3 -6 (Dr. Billingham), 1095:8-24
(Dr. Levine)).

5

The remaining 50% of the 2013 drug substance material was to be "allocated to continuing CMC and Clinical
development work." (PTX-342 at 1).

8

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 10 of 47 PageID #: 16177

Amgen further argues that the jury was free to credit Amgen' s witnesses over Hospira' s
witnesses given the evidence presented. (D.I. 374, pp. 4-5). First, though Hospira argues that it
manufactured each of the 2013 , 2014, and 2015 drug substance batches for use in obtaining FDA
approval, Amgen notes that Ms. Dianis, the regulatory lead for Hospira' s EPO product, "admitted
that she did not know why Hospira made its 2015 batches, or why Hospira made as many batches
as it did, and that she assumed Hospira' s supply team (not the regulatory team) made those
decisions." (Id. (citing Trial Tr. 738:22-740:2)). Second, though Hospira informed the FDA in
2014 that its 2013 and 2014 batches were for "commercial inventory," Hospira' s 2015
resubmission (after litigation began) designated these batches for use for continued process
verification. (Id. at 5 (citing PTX-250 at 4-6; DTX-255 at 5-8); see also Trial Tr. 748:9-751:23).
Third, Dr. Levine admitted that he did not consider whether Hospira made any batches for
commercial inventory, and that "simply submitting data [to the FDA] isn' t a justification" for
manufacturing a batch of drug substance. (Trial Tr. 1075: 18-1076: 1, 1098: 1-10).
I agree with Amgen and conclude that substantial evidence supports the jury' s verdict that
not all of Hospira' s drug substance batches are protected by the safe harbor. To demonstrate
entitlement to judgment as a matter of law on its safe harbor defense, Hospira must demonstrate
that "there is insufficient evidence for permitting any other finding. " Fireman 's Fund Ins. Co.,
540F.2d at 1177. Hospira has not met that burden.
A reasonable jury could have concluded that fewer than all of the batches were protected
by the safe harbor defense. Testimony by Ms. Dianis and Dr. Levine either called into question
or contradicted Hospira's argument that each of the batches at issue fell within the safe harbor
defense. Amgen' s presentation of FDA guidance documents, admissions in Hospira's internal
documents, and post-litigation changes to Hospira' s representations to the FDA also challenged

9

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 11 of 47 PageID #: 16178

Hospira' s assertion that each batch at issue was covered by the safe harbor. Finally, Hospira' s
argument that the jury impermissibly focused on Hospira' s intent in manufacturing the batches
does not stand up to further scrutiny. Though all of the 2015 batches were designated for use as
"commercial inventory" in Hospira' s Risk Authorization, the jury nonetheless found that some of
those batches were protected by the safe harbor. (D.I. 325 at 3). This suggests that the jury did
not improperly base its verdict on Hospira' s intent. I therefore conclude that substantial evidence
supports the jury' s verdict that only some batches at issue are covered by the safe harbor defense.
I will deny Hospira' s motion for JMOL on this ground.

2. Noninfringement of the '298 Patent
Hospira submits that I should grant its motion for JMOL that it does not infringe claims 24
or 27 of the '298 patent.
Hospira contends that it is entitled to JMOL of noninfringement of claim 24 of the 298
patent because "Amgen failed to prove that Hospira' s process ' selectively elutes' isoforms as
required by claim 24 and as construed by the Court." (D.I. 357, p. 12). According to Hospira,
elution of all biologically active isoforms does not qualify as selective elution. (Id.). Specifically,
"Hospira' s process does not achieve a precise set of isoforms;" instead, it "results in a variable
number of different isoforms [i.e., five to eight] , and a variable amount of each isoform in the drug
substance." (Id. (citing Trial Tr. 984:5-989: 17 (Dr. Levine))). Dr. Levine opined that such
variability is not consistent with selective elution, because one would expect consistent levels of
each isoform across batches in a selective elution process. (Trial Tr. 989:3-9, 1580:4-23). Finally,
Hospira argues that Amgen failed to prove infringement of the ' 298 patent because it did not
provide any analysis of the starting material that Hospira puts into the chromatography column.
(D.I. 381 , p. 9).

10

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 12 of 47 PageID #: 16179

Amgen maintains that substantial evidence supports the jury' s conclusion that Hospira' s
process "selectively elutes" isoforms as required by claim 24 of the ' 298 patent. Dr. Cummings
testified that Hospira's process "selectively elutes" isoforms because it "first elute[s] more basic
isoforms from the chromatography column, then elute[s] the remaining desired isoforms." (D.I.
374, p. 13 (citing Trial Tr. 468: 11-469:4)). During trial, Amgen argued that Dr. Levine ignored
the first step in the elution process, and that Dr. Levine admitted that he "was intentionally not
showing those [more basic isoform elution] steps because [he] had already discussed this" the day
before. (Trial Tr. 1156:22-1157:13). Amgen also notes that contrary to Hospira' s argument, the
Court' s claim construction for claim 24 does not "require a 'precise set of predetermined
isoforms."' (D.I. 374, p. 14; see also D.I. 180 at 2). Further, Dr. Strickland, the inventor on the
'298 patent, testified that the process he invented "select[ s] isoforms by-well, specific mixtures
of isoforms[,] by selective elution of an ion exchange chromatography column." (Trial Tr. 373: 1420). As further evidence that Hospira' s process met the "selectively elute" limitation, Amgen
offered Hospira' s lot release specifications, which specified five isoforms that must be present,
and three additional isoforms that may be present, with specific ranges for each isoform. (DTX141 , p. 7; Trial Tr. 470:13-472:19). Finally, Amgen asserts that Hospira' s "starting material"
argument is frivolous because "Hospira admitted in its BLA that its ion-exchange chromatography
process first removed the 'more basic' isoforms .. . from the column (DTX-116 at 58), a step that
would not be necessary if the starting material did not contain isoforms that were ' more basic' than
the isoforms required by Hospira' s release specification." (D.I. 374, p. 15). Similarly, Hospira' s
BLA test results reveal that the only isoforms present in the material leaving the column in
Hospira's process are the same isoforms present in Hospira's drug substance. (Id. (citing DTX139, p. 102; DTX-141 , p. 7; Trial Tr. 474:19-476:11)).

11

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 13 of 47 PageID #: 16180

I agree with Amgen that substantial evidence supports the jury's verdict that Hospira
infringes claim 24 of the ' 298 patent. Dr. Strickland' s testimony, Hospira's release specifications,
and Dr. Cummings' testimony provided the jury substantial evidence to conclude that Hospira' s
process met the "selectively elute" limitation and infringed claim 24 of the ' 298 patent.
Additionally, I think Dr. Levine' s admission that he did not include all steps of the process in his
demonstratives for the jury provided a basis for the jury to question the reliability of his
conclusions and discount his testimony.
Claim 27 of the '298 patent requires "preparing a mixture of two or more erythropoietin
isoforms of claim 1." Though I construed claim 27 as an independent claim, Hospira argues that
it is entitled to JMOL because Amgen did not mention claim 1 during trial, nor did it present
evidence that "isoforms are isolated during Hospira' s manufacturing process." (D.I. 357, p. 13).
Amgen responds that since claim 27 is an independent claim, Amgen was not required to present
evidence that each of the limitations of claim 1 was met in order to prove infringement of claim
27. (Id. p. 15). As to the isolation ofisoforms, Amgen notes that the Court's construction of claim
27 "does not require the isoforms of Claim 1 to be separately prepared prior to making the
mixture." (D.I. 374, p. 15 (citing D.I. 308 at 2)). Regardless, Amgen urges, "the evidence at trial
showed that the limitations of claim 1 were satisfied, that is, that Hospira' s product contains
'biologically active' EPO." (Id. (citing DTX-270, p. 17; Trial Tr. 394:1-4 (admission by Dr.
Strickland that "all EPO isoforms have biological activity."))).

I agree with Amgen.

Though Hospira may be correct that Amgen never explicitly

mentioned claim 1 at trial, Hospira does not discuss substantively how Amgen failed to prove that
the limitations of claim 1 were met. Amgen has also offered citations to Hospira' s BLA and to
the trial transcript to support the conclusion that the limitations of claim 1 were satisfied.

12

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 14 of 47 PageID #: 16181

Claim 27 also has a limitation requiring the creation of an EPO composition with a
"predetermined in vivo specific activity." Hospira argues that Amgen failed to prove that Hospira
infringed this limitation because "Dr. Cummings did not provide any evidence of the actual in vivo
specific activity of Hospira's product," and Hospira's product targets a range of in vivo specific
activities, rather than targeting a specific activity. (D.I. 357, p. 14). Hospira also contends, "Dr.
Strickland testified that in order to achieve a predetermined specific in vivo activity, one selects
individual isoforms and prepares them in such a way to know what biological activity they are
going to get." 6 (D.I. 381 , pp. 9-10 (citing Trial Tr. 375:12-376:2)). Since there is no evidence that
Hospira separates and selects individual isoforms, Hospira argues, there was insufficient evidence
for the jury to conclude that Hospira infringed claim 27. (Id. p. 10). Amgen responds that Dr.
Cummings relied on Hospira's BLA, which stated that 100% oflots fell within an in vivo specific
activity of93-147 U/ µg. 7 (D.I. 374, p. 16 (citing DTX-270, p. 17)).
I agree with Amgen. I do not find Dr. Strickland's testimony, cited by Hospira, inconsistent
with Hospira's BLA, which identifies a predetermined in vivo specific activity-93-147 U/µg.
(DTX-270, p. 17; see Trial Tr. 375 : 12-376:2). Hospira' s BLA provided the jury with substantial
evidence to conclude that Hospira' s process achieved an EPO composition having a predetermined
in vivo specific activity.

6

I do not think Dr. Strickland's testimony is as clear as Hospira makes it out to be. The transcript reflects that Dr.
Strickand testified as follows:
Q. And how does that experiment relate to the inventions in claim 24 and 27?
A. Well, it's directly related to both of them in that in claim 24, it's selective elution ofisoforms on
an ion exchange column, which is what this is an ion exchange column, and it's related to claim 27
in that if we - we can select the fractions from that column to give us a mixture of predetermined
biological activity if it was desired since in the background experiments, we determined what the
biological activity was of each isoform. Now, since this method separates the isoforrns, then we can
recombine them and know what biological activity we're going to get.
(Trial Tr. 375:12-376:2 (discussing the '298 patent at 4:12-22)).
7
EPO is measured in "activity units or international units," represented by " U." (Trial Tr. 208:8-10).

13

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 15 of 47 PageID #: 16182

I therefore conclude that substantial evidence supports the jury's verdict that Hospira
infringes claims 24 and 27 of the '298 patent. I will deny Hospira's JMOL on this ground.

3. Invalidity of the '298 Patent
Hospira argues that it is entitled to JMOL that the '298 patent is invalid because no
reasonable jury could have found that claims 24 and 27 were not anticipated or obvious over U.S.
Pat. No. 4,667,016 ("Lai"). (D.I. 357, p. 14). Since it was Hospira' s burden to prove invalidity,
to prevail on its JMOL, Hospira must demonstrate that "there is insufficient evidence for
permitting any different finding" than that the disclosures in Lai render invalid claims 24 and 27
of the ' 298 patent. See Fireman 's Fund Ins. Co., 540F.2d at 1177.
The parties dispute whether Hospira adequately proved that the Lai process inherently
anticipates claim 24 of the '298 patent. Specifically, they dispute whether Hospira proved that Lai
meets the "selectively eluting" and "predetermined number of sialic acids" limitations of claim 24.
Hospira asserts that Dr. Levine' s testimony, Dr. Strickland's testimony, and Dr.
Cummings' testimony conclusively established that Lai includes a "selective elution" step. (Id.
pp. 15-16). Dr. Levine did not dispute that Lai does not refer to the removal of biologically active
EPO in ion exchange chromatography. (Trial Tr. 1010: 10-16). Based on "the fundamental
principles on which ion exchange chromatography works, and the difference in pKa8 between high
sialic and low sialic acid containing isoforms," however, Dr. Levine opined that Lai's step 2
example 2 "low pH, low salt wash will remove proteins that have a pKa greater than the
biologically active EPO ... [which] will include the isoforms ofEPO that have a small number of
sialic acid[ s] and are therefore not biologically active, or less biologically active." (Id. 1010: 116). As further support, Hospira points to Dr. Levine ' s and Dr. Strickland' s discussions of the

8

pKa is "related to the isoelectric point" of a substance. If a substance "has a low pKa, it' s more basic." (Trial Tr.
422 :6-22).

14

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 16 of 47 PageID #: 16183

' 298 patent's disclosure that the starting material for the fourth isoform of EPO is the material
removed from the column in Example 2 of Lai, which contains EPO isoforms with less than or
equal to nine sialic acids. (Id. 1034:5-14, 393:14-21). Dr. Cummings confirmed that when Dr.
Strickland replicated the experiment reported in Example 2 of Lai, the result was that EPO
isoforms containing nine to fourteen sialic acids were retained on the chromatography column
after the first acid wash step. (Id. 1508:15-1509:9). Therefore, Hospira argues, Lai inherently
discloses selective elution ofEPO with less than nine sialic acids. (D.I. 357, p. 16).
Amgen responds that the jury declined to credit Hospira's argument that "practicing
Example 2 of Lai ' necessarily and inevitably' resulted in 'selectively eluting ' EPO molecules with
a 'predetermined' number of sialic acids." (D.I. 374, pp. 16-17). Amgen points to Dr. Cummings,
who testified that, contrary to Dr. Levine 's assertion, "all EPO is biologically active," and the
purpose of Lai was to purify EPO, not to separate EPO isoforms. (Trial Tr. 1494:12-1495:4).
Amgen maintains that despite testimony that "some isoforms may be removed in Example 2 of
Lai, none of the witnesses "testified that Example 2 in Lai ' necessarily and inevitably' results in"
selectively eluting EPO isoforms with a predetermined number of sialic acids, as would be required
to prove inherent anticipation. (D.I. 374, p. 18). Amgen argues that Dr. Levine 's admission that
several factors could affect which isoforms are present in the starting material (including cell
culture conditions and the components of the cell culture medium) further supports that the Lai
process does not "necessarily and inevitably" meet the limitations of claim 24. (Trial Tr. 1128: 191130:1). Finally, Amgen notes, Dr. Levine agreed that, "Lai couldn' t have selectively eluted
isoforms having a predetermined number of sialic acids because Lai eluted all bound isoforms at
the same time." (Id. 1127:4-9).
Despite the '298 patent disclosures, I think that Dr. Levine's admission that Lai could not

15

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 17 of 47 PageID #: 16184

have eluted isoforms having a predetermined number of sialic acids, combined with the absence
of testimony from other witnesses that Lai process "necessarily and inevitably" met each of the
limitations of claim 24, adequately supports the jury' s conclusion that Hospira had failed to prove
by clear and convincing evidence that Lai anticipated claim 24.
The issue underlying the parties' dispute over whether Hospira adequately proved that the
Lai process inherently anticipates claim 27 of the '298 patent is whether Hospira proved that the
Lai process necessarily resulted in an EPO composition "having a predetermined in vivo specific
activity."

Hospira alleges that testimony by Dr. Levine established that "Lai discloses a

composition having a predetermined in vivo specific activity" because Lai "disclosed how to create
compositions of the high sialic acid, biologically active EPO." (D.I. 357, p. 18 (citing Trial Tr.
103 9 :2-1040: 16)). Additionally, Hospira asserts that Amgen' s witness, Dr. Cummings, essentially
conceded that Lai anticipated claim 27 because he opined in the context of infringement that
claim 27 may be satisfied by a process that results in a variable amount of the most biologically
active isoforms, thus achieving a broad range of in vivo specific activity. (D.I. 381 , p. 11 ). Amgen
responds that Dr. Cummings opined that Lai does not disclose a "predetermined in vivo specific
activity" because Lai provides no indication of any "finding ahead of time for select mixtures of
isoforms."

(Trial Tr. 1496:16-1497:16).

Amgen also maintains that Lai's disclosure that

"biologically active" EPO was eluted does not mean that Lai disclosed an EPO composition with
predetermined in vivo specific activity, because all EPO isoforms have some biological activity,
and "Lai never refers to a composition with a predetermined in vivo activity." (D.I. 374, p. 19).
The evidence at trial regarding anticipation of claim 27 by Lai consisted primarily of expert
testimony. The jury was free to assess the credibility of the experts and weigh their testimony
accordingly. Hospira's argument about Dr. Cummings' concession ignores the "predetermined"

16

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 18 of 47 PageID #: 16185

portion of the limitation, and does not account for the role ofHospira' s BLA to establish an in vivo
specific activity in the infringement analysis. It seems to me that a reasonable jury could have
credited Dr. Cummings' testimony over that of Dr. Levine, and decided that Lai did not disclose a
"predetermined in vivo specific activity," particularly since Hospira was required to prove
anticipation by clear and convincing evidence.
Hospira also argues that no reasonable jury could have found that claims 24 and 27 of the
'298 patent were non-obvious over Lai in view of Lukowsky. During trial, Dr. Levine offered his
opinion that claim 24 would have been obvious because (1) it was known that more sialylated
forms of EPO were more biologically active (Trial Tr. 1047:16-23); (2) it was known that sialic
acid "add[ed] negative charge to the" EPO molecules to which it was attached (Id. 957:3-8); (3) ion
exchange chromatography was a well-known method for separating protein molecules by their net
charge (Id. 967 :4-1 O); and (4) the Beeley reference taught that glycoproteins could be separated
by charge using ion exchange chromatography (Id. 1052:24-1054:23). Therefore, Hospira argues,
a POSA would have been motivated to separate isoforms and create a preparation of EPO with a
predetermined in vivo specific activity, and have a reasonable expectation of success in doing so.
(D.I. 357, p. 19).
Amgen submits that the Patent Office considered both Lai and Lukowsky during
prosecution, and the "PTO examiner acknowledged that the '298 patent taught the unexpected
advantage of combinations of isoforms, and the ability to prepare EPO compositions with
predetermined EPO isoforms." (D.I. 374, p. 20 (citing PTX-4B, pp. 11-12; Trial Tr. 1500:221502:23)). Amgen further points to Dr. Cummings' explanation that Lukowsky does not supply
the limitations missing from Lai, because Lukowsky does not disclose EPO "isoforms," a
"predetermined mixture of [EPO] isoforms," or "predetermined specific activity" of any EPO

17

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 19 of 47 PageID #: 16186

isoforms. (Id. (citing Trial Tr. 1489:14-1490:2)).
I think that substantial evidence supports the jury' s conclusion that neither claim 24 nor
claim 27 of the '298 patent is obvious over Lai in view of Lukowsky and Beeley. Hospira bore
the burden by clear and convincing evidence to prove that the asserted claims of the '298 patent
were invalid. Since neither Lukowsky nor Lai discloses EPO isoforms, or predetermined mixtures
or in vivo specific activities of EPO isoforms, and the PTO acknowledged unexpected results
produced by the ' 298 patent, I cannot say that the jury was unreasonable in deciding that Hospira
had not met its burden to prove obviousness. I also note that in the relevant briefing, Hospira's
statements of a POSA' s motivation to combine these references lack explanation. (D.I. 357, p. 19
("[A] person of ordinary skill in the art would have been motivated to create a preparation of EPO
with a predetermined specific activity, and would have had a reasonable expectation of success in
doing so.")).
I therefore conclude that substantial evidence supports the jury' s verdict that claims 24 and
27 of the ' 298 patent are not anticipated by or rendered obvious by Lai. I will deny Hospira' s
motion for JMOL on this ground.

4. Damages
Hospira also moves for JMOL on the ground that the jury' s damages award is not
reasonable, challenging both the amount and the lump sum nature of the award. (D.I. 357, p. 20).
Based on Dr. Bell' s analysis, Hospira asserts, "Dr. Bell' s proposed royalty of $1.5 million per
batch, when the batch is sold, is the only damages methodology that properly accounts for the
expectations of the hypothetical negotiators at the time concerning FDA approval, and the reality
of what occurred afterwards." (Id. p. 22).
Relying on Dr. Bell' s trial testimony, Hospira contends, "Hospira, as a willing licensee,

18

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 20 of 47 PageID #: 16187

would not have been willing to pay more than the replacement cost of the batches, which was from
$4.1 to $4.6 million per batch." (Id. p. 20 (citing Trial Tr. 1241:1-1243:13)). Considering the
Georgia-Pacific factors, Dr. Bell adjusted this downward to 35% of the replacement cost. (Trial
Tr. 1246:12-1248:21). Dr. Bell further opined that due to the uncertainties associated with FDA
approval, Hospira would not willingly pay an up-front lump sum royalty. (Id. 1252: 13-19).
According to Dr. Bell, if the FDA never approves Hospira' s biosimilar, then Hospira has no
opportunity to sell the product and realizes no gain, and Amgen has no losses. (Id. 1252:1-9).
Hospira further criticizes Amgen' s damages theory because it "requires Hospira to bear all the
'risk' of the license," and it reflects an award "more than the twenty-year net present value of the
entire EPO project." (D.I. 357, pp. 20-21). According to Hospira, this is inconsistent with a
willing licensor and a willing licensee. (Id.). Hospira also criticizes Amgen' s damages theory as
"based entirely on the cost to Hospira of the supposed delay that would have occurred if it had to
wait until after patent expiration to manufacture its EPO substance for launch," when "no such
delay ever occurred." (D.I. 381 , p. 13). Finally, Hospira argues that the Vifor Agreement cited by
Amgen "is a non-comparable marketing and distribution agreement with an upfront payment that
can be refunded if Hospira does not obtain FDA approval." (D.I. 357, pp. 21-22).
Amgen responds, "Dr. Bell' s testimony provides the lowest reasonable royalty that may be
supported by the evidence," not the only reasonable royalty. (D.I. 374, p. 21). The jury' s award
was supported by the evidence, Amgen argues, because "[t]he jury [i]s entitled to choose a
damages award within the amounts advocated by the opposing parties."9

(Id. (quoting

Spectralytics, Inc. v. Cordis Corp., 649 F.3d 1336, 1347 (Fed. Cir. 2011), abrogated on other
grounds by Halo Elecs., Inc. v. Pulse Elecs., Inc., 136 S. Ct. 1923 (2016)) (brackets in original)).

9

Here, the difference between the parties ' positions is $ 116 million, as Hospira proposed $2 1 million and Amgen
proposed $137 million. (D.l. 374, p. 24).

19

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 21 of 47 PageID #: 16188

Amgen presented the Vifor Agreement as an example of a lump sum agreement involving Hospira,
asserting that Amgen would not agree to a refund in this case because Amgen and Hospira are
competitors. (Id. p. 23 (citing Trial Tr. 665:4-666:1)). Additionally, Amgen argues, "Hospira
cites no legal support for its contention that economic harm is required for a jury to award a royalty
as a lump sum." (Id.). "The reasonable royalty determined in a hypothetical negotiation does not
compensate for lost sales but rather the lost opportunity of a reasonable royalty before
infringement." (Id. (citing AstraZeneca AB v. Apotex Corp., 782 F.3d 1324, 1334 (Fed Cir.
2015))).
Amgen maintains that Dr. Heeb "provided numerous bases in addition to the Vifor
agreement" to support his damages theory. (Id.). Dr. Heeb opined that since Amgen and Hospira
are competitors, Amgen would not agree to a running royalty that required Amgen to share any
risk associated with Hospira's manufacture ofEPO. (Trial Tr. 640:12-641:15). Additionally, Dr.
Heeb cited administrative advantages of a lump sum royalty, such as there being no need to track
sales. (Id. 641 :16-23).
Regarding Hospira's arguments that the award does not reflect events occurring after the
hypothetical negotiation (i.e., the lack of FDA approval), Amgen notes that the jury was instructed
that it could consider such events. (D.I. 374, p. 22). Amgen argues, "It is not error that the jury
did not agree with Hospira" about the effect of such events on the reasonable royalty rate. (Id.).
Contrary to Hospira' s argument that Dr. Bell ' s theory is the only one under which the jury could
have found a per-batch royalty (D.I. 357, p. 29), Amgen notes that Dr. Heeb offered testimony on
the value of the license if the jury found only some batches to infringe. (D.I. 374, p. 22). For
example, if fourteen batches were found to infringe, the value to Hospira of a license would have
been $137 million. (Id. (citing Trial Tr. 645:22-646:6)). Amgen would have valued a license at

20

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 22 of 47 PageID #: 16189

$170 million. (Trial Tr. 636 :2-11). Regarding Hospira's criticism that Dr. Heeb's analysis
resulted in an award that exceeded the net present value of Hospira' s EPO project, Amgen points
to testimony by Dr. Bell acknowledging Hospira documents that stated the net present value of its
EPO project as up to $297 million. (/d.1273:10-1274:5).

I conclude that substantial evidence supports the jury's $70 million damages award.
Regarding the lump sum payment, both parties' experts provided testimony to support their
positions on whether a lump sum would be appropriate, and the jury was free to determine the
experts' credibility and weigh their testimony accordingly. I decline to substitute my judgment
for that of the jury.
Hospira essentially argues that Dr. Heeb's methodology was not supported by substantial
evidence because no launch delay ever materialized. Indeed, Hospira's expert Dr. Bell testified
that he considered only the hypothetical negotiation scenario in which Hospira does not launch a
product prior to mid-2017, "because it's the one that we happen to be in." (Trial Tr. 1293:161294: 15). In other words, Dr. Bell' s analysis focuses solely on a hypothetical negotiation in which
the parties have knowledge of all subsequent events. Amgen's analysis appears to amount to a
hypothetical negotiation in which the parties do not have the benefit of subsequent knowledge that
Hospira did not receive FDA approval.

I cannot say that it was unreasonable for the jury to find neither expert struck the proper
balance in considering how post-negotiation events would have affected the reasonable royalty.
The [reasonable royalty] methodology encompasses fantasy and flexibility; fantasy
because it requires [the jury] to imagine what warring parties would have agreed to
as willing negotiators; flexibility because it speaks of negotiations as of the time
infringement began, yet permits and often requires [the jury] to look to events and
facts that occurred thereafter and that could not have been known to or predicted
by the hypothesized negotiators.
Fromson v. W Litho Plate & Supply Co., 853 F.2d 1568, 1575 (Fed. Cir. 1988), overruled on other

21

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 23 of 47 PageID #: 16190

grounds by Knorr- Bremse Systeme Fuer Nutzfahrzeuge GmbH v. Dana Corp., 383 F.3d 1337
(Fed.Cir.2004) (en bane); see also Sinclair Ref Co. v. Jenkins Petroleum Process Co., 289 U.S.
689, 698 (1933) ("[A] different situation is presented if years have gone by before the evidence is
offered. Experience is then available to correct uncertain prophecy. Here is a book of wisdom that
courts may not neglect."). Though the parties would have recognized the possibility, as of the time
of the hypothetical negotiation, that Hospira may not receive FDA approval before the expiration
of the patents, Hospira did not expect such a result. 10 In essence, Hospira wants me to do what the
Federal Circuit has expressly held was error, to "replace[] the hypothetical inquiry into what the
parties would have anticipated, looking forward when negotiating, with a backward-looking
inquiry into what turned out to have happened." Aqua Shield v. Inter Pool Cover Team, 774 F.3d
766, 772 (fed. Cir. 2014). I therefore cannot say that the consideration in Dr. Heeb's analysis of
the value to Hospira of avoiding launch delay was not supported by substantial evidence. The
parties' experts each provided an endpoint for the range of potential hypothetical reasonable
royalties, and as Amgen points out, the jury was free to choose a damages award within the
amounts advocated by the opposing parties. Therefore, I will deny Hospira's motion for JMOL
on this ground.
B. New Trial

I. Safe Harbor Instruction
Hospira asserts that it is entitled to a new trial because the safe harbor jury instruction was
"legally erroneous and prejudicial." (D.I. 357, p. 22). 11 Specifically, Hospira argues that the
instructions "confused the 'manufacture' and 'use' of the batches in a way that misrepresents the

10

Hospira projected that it would obtain FDA approval in the fourth quarter of 2015. (PTX-342, p. 1).
Citing its JMOL arguments, Hospira also argues that the jury' s verdict that fourteen batches were not protected by
the safe harbor defense is against the clear weight of the evidence. Having already addressed these arguments in
Hospira's JMOL, I will not address them again in considering Hospira' s motion for a new trial.
11

22

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 24 of 47 PageID #: 16191

statute," and Hospira challenges the instructions' "fail[ ure] to instruct the jury that ulterior motives
and intent are irrelevant to the Safe Harbor." (Id.) . Amgen submits that the instructions were
adequate, and that any instructional error was "at most harmless error." (D.I. 374, p. 24).
Hospira contends that the verdict form and jury instructions were erroneous because they
did not use the terms "make" and "use" consistently with the statute, thus failing to "clarify the
difference between 'manufacture' and 'use' under the Safe Harbor." (D.I. 357, pp. 23-24; see also
Trial Tr. 1404:11-1405:15 (Hospira objecting to the safe harbor jury instruction and stating, "We
would urge the broader standard for uses, but focus on the instruction should be on the uses and
not the motives or purposes in making [a] batch. The statutory exemption is premised on the use
aspect."); Trial Tr. 1445 :12-1449:18, 1452:8-23 (Hospira's continuing objections)).

The

instructions initially refer to Hospira's burden to prove "uses reasonably related to obtaining FDA
approval," and Hospira's burden to prove "that the safe harbor defense applies to Hospira's use of
Amgen's patented invention," but they subsequently "ask[] the jury to determine whether the
manufacture is covered." (D.I. 357 p. 24). According to Hospira, the verdict form "compounded
this error[] by asking the jury to find whether the 'Safe Harbor Defense applied to the manufacture
of the following batches."' (Id.). Amgen responds that the jury instructions and verdict form track
the statute because, "The Court correctly instructed the jury that the alleged infringing activity was
Hospira' s making of its drug substance, which needed to be 'for uses reasonably related' to seeking
FDA approval. " (D.I. 374, p. 25 (citing Trial Tr. 1522:5-20, 1553:3-1554:13)).
I agree with Amgen. The safe harbor defense provides,
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 (other than a
new animal drug or veterinary biological product (as those terms are used in the
Federal Food, Drug, and Cosmetic Act and the Act of March 4, 1913) ... ) solely
for uses reasonably related to the development and submission of information under
a Federal law which regulates the manufacture, use, or sale of drugs ....

23

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 25 of 47 PageID #: 16192

35 U.S.C. § 271(e)(l). The safe harbor exempts activity that would ordinarily constitute an act of
infringement if that activity is undertaken "solely for uses reasonably related" to obtaining FDA
approval. Asserted claims 24 and 27 of the '298 patent are method claims. Hospira is correct in
pointing out that the instructions and the safe harbor statute refer to "use" of a patented invention.
Here, Hospira' s potentially infringing "use" of Amgen' s patented invention is Hospira's
manufacture of the EPO drug substance referred to in its BLA (i.e., Hospira's performance of the
steps of Amgen 's method claims), not Hospira's subsequent use of the EPO drug substance (i.e.,
Hospira's subsequent use of the product obtained by practicing Amgen's method claims). See Joy
Techs, Inc. v. Flakt, Inc. , 6 F.3d 770, 775 (Fed. Cir. 1993) ("A method claim is directly infringed

only by one practicing the patented method." (emphasis omitted)); Roberts Dairy Co. v. United
States, 530 F.2d 1342, 1354 (Ct. Cl. 1976) (" It is well established that a patent for a method or

process is not infringed unless all steps or stages of the claimed process are utilized."). Therefore,
the safe harbor defense applies in this case only ifHospira' s manufacture ofits EPO drug substance
is reasonably related to obtaining FDA approval. Though Hospira' s subsequent uses of its EPO
drug substance are probative in determining whether Hospira's manufacture of its EPO drug
substance was reasonably related to obtaining FDA approval, it is the manufacture itself (not
Hospira's subsequent uses ofEPO drug substance) that is the potentially infringing act which must
be evaluated for safe harbor protection. I thus think the jury instructions and the verdict form were
proper in asking the jury to determine whether Hospira' s potentially infringing act, i.e., its
manufacture of the EPO drug substance, was covered by the safe harbor defense.
According to Hospira, "the Court's Safe Harbor instructions were erroneous for a second
reason- they omitted any discussion of how intent related to the Safe Harbor analysis." (D.I. 357,
p. 23 ; Trial Tr. 1404:11-1405:15 (Hospira's charge conference objection to the safe harbor jury

24

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 26 of 47 PageID #: 16193

instruction); D.I. 318, pp. 30-31 (Hospira's objection to the Court' s proposed jury instruction on
safe harbor); Trial Tr. 1445:12-1449:18). Though the instructions stated that "Hospira's additional
underlying purposes for the manufacture and use of that batch do not remove that batch from the
safe harbor defense" (D.I. 323 , p. 19), Hospira contends that "this language was ambiguous and
did not explicitly state that intent does not matter." (D.I. 357, p. 23). Per Hospira, "This error,
coupled with the Court' s denial of Hospira' s motion in limine to preclude Amgen[] from
introducing the Risk Authorizations allowed the jury to hear evidence of alleged commercial intent
based on the highly prejudicial Risk Authorizations." (Id. (citation omitted)). To support its
assertion that its internal Risk Authorizations are unduly prejudicial and should not have been
admitted into evidence, Hospira repeats its argument that intent is irrelevant to evaluating safe
harbor protection. (Id.). Essentially, Hospira asserts that I should have used its proposed jury
instructions on the safe harbor defense. (Id.). Amgen responds, "Hospira's proposed instruction
[] also omitted any discussion of ' intent' ; it did not even contain that word." (D.I. 374, p. 24 (citing
D.I. 304, pp. 3-4). Regardless, Amgen submits that the instructions adequately addressed intent
because the court "instruct[ ed] the jury that ' Hospira' s additional underlying purposes for the
manufacture and use of [a] batch do not remove that batch from the safe harbor defense. "' (Id.
pp. 24-25).
Hospira appears to argue that intent is entirely irrelevant to the safe harbor analysis, but I
do not think the cases Hospira cites in its brief stand for such a broad proposition. In Abtox, the
court found a competitor' s limited testing during development of a device "consistent with the
collection of data necessary for filing an application with the [FDA] ," despite allegations that that
the actual purpose of the tests was "to promote the [device] and other equipment to potential
customers." 122 F.3d at 1027. Accordingly, the Abtox court concluded that the safe harbor defense

25

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 27 of 47 PageID #: 16194

"allows [a party] to use its data from the tests for more than FDA approval" and "does not look to
the underlying purposes or attendant consequences of the activity (e.g., tests led to the sale of the
patent), as long as the use is reasonably related to FDA approval." Id. at 1030. Notably, the Abtox
court did not state that intent was irrelevant in determining whether an activity is reasonably related
to obtaining FDA approval.
Hospira' s citations to other cases quote language clarifying that the mere existence of some
intent, "ulterior motives," or "alternate purposes" to commercialize does not preclude a party from
successfully invoking the safe harbor defense. (D.I. 357, pp. 22-23 (citing Intermedics, Inc. v.

Ventritex Co., 1993 WL 87405, at *5 (Fed. Cir. Feb. 22, 1993) ("Reliance on section 271(e)(l) is
not precluded by manifestation of an intent to commercialize upon FDA approval."); Amgen, Inc.

v. Hoechst Marion Roussel, Inc., 3 F. Supp. 2d 104, 108 (D. Mass. 1998) ("[U]lterior motives or
alternate purposes do not preclude application of the section 271(e)(l) exemption."); Intermedics,

Inc. v. Ventritext Co., 775 F. Supp. 1269, 1280 (N.D. Cal. 1991), aff'd 991 F.2d 808 (Fed. Cir.
1993) ("[I]f a party were to lose the exemption every time a business purpose was detectable in its
otherwise infringing activities, the exemption would virtually never be available and thus would
fail to achieve Congress' objective."))). I think that evidence of intent can be a relevant factor in
determining whether an activity is reasonably related to obtaining FDA approval, and that these
cases stand for the proposition that evidence of commercial intent is not determinative of the safe
harbor inquiry. In my view, they do not support Hospira's assertions that intent is irrelevant to
determining whether an activity is reasonably related to obtaining FDA approval and that intent
may not be considered at all. But once it is determined that "the activity is reasonably related to
obtaining FDA approval, [] intent or alternative uses are irrelevant to its qualification to invoke
the section 271(e) shield." See Abtox, 122 F.3d at 1030.

26

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 28 of 47 PageID #: 16195

Additionally, adopting Hospira's interpretation of the safe harbor defense would expand
the defense beyond recognition and create a loophole that would make it virtually impossible to
prove infringement in cases involving products regulated by the FDA.

Since Hospira's

interpretation requires ignoring intent in deciding whether the safe harbor applies, a party could
manufacture 200 drug substance batches and earmark them for future use as commercial inventory
without infringing, so long as the party used each of those batches for at least one test to generate
data of the type used by the FDA in determining whether to approve the drug. In that scenario,
each batch would be tested to generate data that could conceivably be used to respond to inquiries
from the FDA, making each batch reasonably related to obtaining FDA approval. Essentially,
Hospira' s interpretation allows a single "token" submission of information derived from a
potential infringing act to exempt that act from infringement, without regard to the realities
surrounding the potentially infringing act. It seems to me that Hospira' s interpretation reads the
words "solely" and "reasonably" out of the statute, and that a party's stated intent may be
considered as part of whether the manufacture or use of a patented drug was "solely for uses
reasonably related to" obtaining FDA approval. I think that the jury instructions properly recited
the role of intent in the safe harbor analysis.
Hospira also argues that FDA draft guidance on statistical approaches to evaluate analytical
similarity, published on September 22, 2017, constitutes new evidence that "proves the uncertainty
of the regulatory landscape" and warrants a new trial. (D.I. 357, p. 24; see also D.I. 357-1 ).
Amgen responds that the draft guidance is cumulative of Hospira' s other evidence of regulatory
uncertainty that would not have changed the outcome at trial. (D.I. 374, p. 26). Setting aside the
fact that the draft guidance was distributed for comment purposes only, Amgen points out that the
guidance "recommend[s] a minimum of 10 reference product lots be sampled" to "establish

27

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 29 of 47 PageID #: 16196

meaningful acceptance similarity criteria." (Id. (citing D.I. 357-1 at 2, 7)). The guidance would
not have changed the outcome, Amgen maintains, because, "Here, Hospira tested 26 drug product
batches to demonstrate biosimilarity, all made using material from 4 drug substance batches on
which the jury did not award damages." (Id. (citing DTX-266, pp. 3-4)). Finally, Amgen notes
that Hospira did not present evidence that the FDA required Hospira to manufacture a certain
number of batches to demonstrate biosimilarity. (Id.).
I agree with Amgen that the FDA draft guidance does not constitute new evidence that
would justify a new trial. Since Hospira submitted test data from twenty-six drug product batches
manufactured from four drug substance batches that the jury found not to infringe, the jury could
have found the remaining drug substance batches to infringe even if the draft guidelines had
constituted a final regulation requiring Hospira to submit data from at least ten drug product
batches to prove biosimilarity. Hospira has not shown that the FDA draft guidance would likely
alter the outcome of the trial.
Having concluded that neither the safe harbor instructions nor the FDA draft guidance
warrants a new trial, I will deny Hospira' s motion for a new trial on safe harbor grounds.
2. Contradictory Verdicts
Hospira argues that it is entitled to a new trial because the jury' s verdicts on infringement
of the ' 298 patent and validity of the ' 298 patent are inherently contradictory. (D.I. 357, p. 24).
Having found each of the jury' s infringement and validity verdicts supported by substantial
evidence, and having found that Hospira failed to meet its burden of proof that the ' 298 patent is
invalid, I will deny Hospira' s motion for a new trial based on contradictory infringement and
validity verdicts.
3. Claim Construction

28

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 30 of 47 PageID #: 16197

Hospira also argues that "the jury received a claim construction order with errors that
warrant a new trial." (Id. p. 26). Specifically, Hospira argues, "For the reasons advanced by
Hospira during claim construction," the Court' s construction of claims 24 and 27 of the ' 298 patent
were incorrect. (Id.). Amgen responds that Hospira' s assertion amounts to an improper request
for reconsideration of the Court' s claim constructions, since "Hospira raises no new arguments"
to support its contention that the Court's claim constructions were improper. (D.I. 374, pp. 2728).
Hospira frames its claim construction arguments as repetitions of the arguments it made
during claim construction. I decline at this late stage to reconsider my constructions based on the
same arguments considered and addressed in my previous claim construction opinions. (Se e D.I.
162, 177). Therefore, I will deny Hospira' s motion for a new trial based on claim construction.
4. Third Party Liability Instruction

Hospira further submits that it is entitled to a new trial because the jury was erroneously
instructed that "Hospira is responsible for the manufacturing activities of GlaxoSmithKline, or
GSK, as they relate to Hospira' s epoetin drug substance."

(D.I. 357, p. 27 (citing Trial Tr.

1552:23-1553:2)). The third party liability jury instruction challenge is proper, Hospira submits,
because Hospira "vigorously disputed the jury instruction on third party liability and confirmed
that its objections to the jury instructions were preserved." (D.I. 381 , p. 15 (citing Trial Tr.
1389: 15-1390: 11, 1524:11-19)). According to Hospira, the erroneous instruction "allowed Amgen
to circumvent the requirements to show induced infringement under 35 U.S .C. § 271(b)[, when]
Amgen never pled induced infringement of the ' 298 patent nor amended its pleadings to do so."
(D.I. 357, p. 27).
Amgen asserts that Hospira's motion for a new trial represents an improper vehicle to

29

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 31 of 47 PageID #: 16198

challenge the Court's JMOL decision during trial that "Hospira was responsible for GSK's
manufacture of Hospira's EPO drug substance for purposes of direct infringement." (D.I. 374,
p. 28; see Trial Tr. 1391:18-1392:6). According to Amgen, since "Hospira has not sought
reconsideration of that ruling and did not object to the jury instruction," Hospira's motion "cannot
overturn the Court's decision to grant JMOL on Hospira' s responsibility for GSK's activities."
(D.I. 374, p. 28).
I will deny Hospira' s motion for a new trial based on the third party liability instruction.
Hospira acknowledges that I granted JMOL that Hospira was responsible for the activities of GSK.
(D.I. 357, p. 27; see Trial Tr. 1392:14-24). Assuming JMOL was properly granted, 12 I think it was
proper to instruct the jury that, "Hospira is responsible for the manufacturing activities of
GlaxoSmithKline, or GSK, as they relate to Hospira's epoetin drug substance." (Trial Tr. 1553:14). Though Hospira reserves its right to appeal the grant of JMOL on this issue (D.I. 357, p. 27
n.1), Hospira does not object to the grant of JMOL in its post-trial briefing. (See id. pp. 27-28;
D.I. 381, p. 15). Regardless, I do not think Hospira's citations to the trial transcript demonstrate a

12

As I noted at trial, I think the evidence supports the grant of JMOL that Hospira is responsible for the activities of
GSK. (Trial Tr. 1392:17-1393:7). As the Federal Circuit stated in Akamai Techs. , Inc. v. Limelight Networks, Inc.,
797 F.3d 1020, 1023 (Fed. Cir. 2015), it has "held that an actor is liable for infringement under§ 27l(a) if it acts
through an agent (applying traditional agency principles) or contracts with another to perform one or more steps of a
claimed method." Under Delaware law, "If the principal assumes the right to control the time, manner and method of
executing the work, as distinguished from the right merely to require certain definite results in conformity to the
contract, a master/servant type of agency relationship has been created." Fisher v. Townsends, Inc., 695 A.2d 53 , 59
(Del. 1997).
Here, Hospira reserves particular time slots with GSK for manufacturing, "sets the overarching specifications for the
manufacture of the drug substance" in accordance with Hospira's BLA, and owns the drug substance on completion
of manufacturing. (Trial Tr. 296: 1-297: 12, 554 :9-555 : 14, 886 :3-887:6). Additionally, Hospira employees worked
with GSK during the manufacturing process and were present at the GSK facility during the FDA's pre-approval
inspection. (Trial Tr. 834:3-835 :3, 864 :3-866:3). I therefore think that Hospira exercises sufficient direction and
control over the manufacturing process such that GSK qualifies as Hospira's agent. Accordingly, even if Hospira
could not be held liable for direct infringement based on its contract with GSK to perform the steps of the claimed
method, Hospira could be held liable for direct infringement based on GSK's actions under an agency theory.

30

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 32 of 47 PageID #: 16199

clear objection by Hospira to the third party liability jury instruction. (See Trial Tr. 1390: 151394:24 (discussing third party liability issue, discussing proposed "infringement by agent"
instruction ultimately not given, granting JMOL that Hospira is responsible for the activities of
GSK, and instructing Amgen to submit a sentence on this issue for the final jury instructions),
1519:9-1524:21 (indicating Hospira's objection to the safe harbor jury instruction; demonstrating
no specific objection by Hospira to revised Instruction 5.2, which included a sentence drafted by
Amgen stating that Hospira is liable for the activities of GSK; and reflecting both parties'
preservation of prior objections)).
5. Remittitur

Finally, Hospira argues that remittitur to $1.5 million per batch, if sold, is appropriate
because "the $70 million damages award contradicts the weight of the evidence." (D.I. 357, pp. 28,
30). As support, Hospira restates in part and incorporates by reference its arguments raised in its
motion for JMOL regarding damages. (Id. pp. 28-30). I will deny Hospira's request for remittitur
or a new trial for the same reasons already discussed with respect to Hospira' s motion for JMOL
regarding damages.

III.

HOSPIRA'S MOTION TO SEAL CONFIDENTIAL EXHIBITS
Hospira requests that the Court seal three exhibits admitted at trial which it asserts contain

confidential business information. The exhibits are DTX-138 (the Vifor Agreement), DTX-177
(the GlaxoSmithKline ("GSK") Agreement), and DTX-178 (amendments to the GSK Agreement).
(D.I. 361 at 2). Amgen opposes Hospira' s motion. (D.I. 369).
The Third Circuit recognizes "a strong presumption that material introduced into evidence
at trial should be made available for public access." Littlejohn v. Bic Corp., 851 F.2d 673 , 678 (3d
Cir. 1988) (citation omitted). "It is well established that the release of information in open court

31

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 33 of 47 PageID #: 16200

is a publication of that information and, if no effort is made to limit its disclosure, operates as a
waiver of any rights a party had to restrict its future use." Id. at 680 (citation omitted). "The party
seeking the closure of a hearing or the sealing of a transcript bears the burden of showing that the
material is the kind of information that courts will protect and that there is good cause for the order
to issue. Good cause is established on a showing that disclosure will work a clearly defined and
serious injury to the party seeking closure." Publicker Indus., Inc. v. Cohen , 733 F.2d 1059, 106970 (3d Cir. 1984) (citations omitted).
According to Hospira, it is appropriate to seal these exhibits because they contain
"confidential commercial and technical information" that "would damage the competitive standing
of the parties and the third parties named in those agreements." (D.I. 361 at 1). Hospira contends
that it "took efforts to keep the contents [of these exhibits] confidential" during trial because only
limited portions of the exhibits were discussed or shown. (Id. at 3). Specifically, Hospira' s
witnesses discussed these documents only in general terms, and Amgen's witness Dr. Heeb
referenced only one page of DTX-178 in live testimony. (Id.). Though " [t]he exhibits were
discussed in deposition testimony of Mr. Noffke and Mr. Pinnow," they "were only shown on a
split screen as the testimony was played," and Hospira had marked the deposition transcripts and
exhibits "confidential." (Id.).

Finally, Hospira argues, "Redaction of the documents is not

practicable, as the information discussed at trial was often intertwined with other sensitive
information and because the organization and structure of the documents themselves are
confidential." (Id.).
Amgen submits that the exhibits should not be sealed because " [d]uring trial, the parties
introduced exhibits into evidence without restriction." (D.I. 369 at 1). Since " [t]he Court never
sealed the courtroom, and Amgen's corporate representative, members of the press, and other

32

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 34 of 47 PageID #: 16201

members of the public were present throughout the trial," Amgen maintains that the exhibits are
in the public record. (Id.). Additionally, Amgen argues that the exhibits should not be sealed
because "several witnesses testified about the three agreements, including about specific details in
those agreements." (Id. at 2 (citing Mr. Noffke's testimony about GSK's reserve capacity in the
GSK Agreement (Trial Tr. 565:8-568:5); Dr. Bell' s testimony about cost of manufacture
information derived from the GSK Agreement (Trial Tr. 1244:6-1245:14); Mr. Pinnow's and Dr.
Bell's testimony about payment terms in the Vifor Agreement (Trial Tr. 306:19-22, 309:3-312:2,
1254:1-15))). Finally, Amgen expresses concern that Hospira seeks to seal portions of the GSK
Agreement on which Hospira intends to rely in raising the argument that GSK is not an agent of
Hospira. (Id. at 3-4).
I agree with Amgen that the exhibits should not be sealed. On the first morning of trial I
indicated my preference that the parties "redact out [the] portions that aren' t relevant," and
suggested that the parties ought to think about limiting disclosures of portions of the exhibits to
those that are "critical to the testimony" at trial. (Trial Tr. 18:13-19). Though Hospira may have
taken some measures to keep the exhibits confidential, Hospira published portions of the exhibits
in open court and relied on information from the exhibits in presenting its case. That portions of
the exhibits "were only shown on a split screen" does not change the fact that they were published
and admitted into evidence without any request by Hospira to seal them at the time. (Id. at 317:513, 594:16-595:19).

Hospira' s broad argument about the impracticality of redaction is not

sufficient for me to conclude that it would be impractical to redact the exhibits.
Accordingly, I will not seal the exhibits. Since the exhibits were not published in their
entirety in open court and contain sensitive information about ongoing commercial agreements, I
will allow Hospira to submit proposed redactions within ten days of the date of the order

33

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 35 of 47 PageID #: 16202

accompanying this opinion. The parties should meet and confer on the proposed redactions before
the submission is made.
IV.

AMGEN'S RENEWED JMOL
Amgen seeks judgment as a matter of law that Hospira infringed the ' 349 patent. (D.I.

358). Alternatively, Amgen seeks a new trial on infringement of the ' 349 patent because the jury' s
verdict was against the weight of the evidence, and was based on what Amgen characterizes as
Hospira' s improper argument to the jury that the ' 349 claims require RIA evidence for
infringement. (Id.) .
A. JMOL
The only limitations in the ' 349 patent at issue during trial required cells "capable of
producing" EPO at a rate of 100 U, 500 U, and 1000 "U of erythropoietin per 10 6 cells in 48 hours
as determined by radioirnrnunoassay." (Id. p. 5). It was Arngen' s burden at trial to prove by a
preponderance of the evidence that Hospira' s EPO-producing cells met this limitation. To prevail
on its motion, Amgen must demonstrate that "there is insufficient evidence for permitting" a
finding that Hospira does not infringe the asserted claims of the ' 349 patent. Fireman 's Fund Ins.
Co., 540 F.2d at 1177.

Amgen contends, "The only evidence introduced at trial about the production rate of
Hospira' s cells established that they were capable of producing EPO at a rate of more than 3500
Units per million cells in 48 hours," because "Hospira' s expert, Dr. Hamilton, did not offer any
affirmative evidence or an opinion that Hospira' s cells were not capable of producing EPO at the
rates recited in the ' 349 claims." (Id. pp. 1-2). As further support, Amgen points to a Hospira
report submitted to the FDA as part of Hospira' s BLA stating that Hospira' s cells produce EPO
"in the range of 100 µg or higher" per day based on dot-blot analysis. (Id. p. 7 (citing PTX-293 ,

34

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 36 of 47 PageID #: 16203

p. 28)).

Amgen's expert, Dr. McLawhon, testified that Hospira's 100 µg/mL rate could be

converted into the "U" of EPO recited in the '349 patent claims, and that his calculations resulted
in a rate of 3,534 U of EPO per 106 cells in 48 hours. (Id. (citing Trial Tr. 525:1-536:11)).
Essentially, since Hospira' s expert, Dr. Hamilton, admitted that it was theoretically possible to
convert µg/mL EPO production to U of EPO, and since Dr. Hamilton "did not offer any affirmative
evidence ... that Hospira's cells were incapable of producing EPO at the claimed production
levels," Amgen asserts that no reasonable jury could have found that Hospira's cells did not
infringe. (Id. pp. 9-10).
Amgen also argues that no reasonable jury could have found Dr. Hamilton's testimony
adequate to rebut Amgen' s evidence of infringement. (D.I. 358, pp. 11-14). First, Amgen argues
that Dr. Hamilton' s criticisms are "inconsistent with the written description of the '349 patent, in
which Dr. Lin described using an RIA to assay crude ' culture fluids,' and converted the resulting
data to 'U' of EPO using a purified EPO standard." (D.I. 380, p. 3). 13 Amgen's responds to Dr.
Hamilton's testimony that the RIA and dot blot testing at issue here are not comparable by arguing
that Dr. Hamilton "failed to tie these alleged deficiencies to any limitation recited in the claims:
the '349 claims do not require the same standard or antibody used by Dr. Lin when he tested his
inventions, nor are the production-rate limitations limited to testing a purified EPO sample." (D.I.
358, pp. 12-13). Amgen further argues that Dr. Hamilton's testimony should not be credited
because he "did not interpret the claims, or opine on whether a person of ordinary skill in the art
would have understood the claims to require the use of the same EPO standard or anti-EPO
antibody preparation that the ' 349 inventor, Dr. Lin, used." (Id. p. 13). Essentially, Amgen argues,

13

In response, Hospira points to Dr. Hamilton's testimony that "the '3 49 patent examples use a standard curve," but
"that curve was not compared to Hospira's sample in this case." (D.I. 373, p. 13 (citing Trial Tr. 1226:3-23, 1227:2 11228: 10)). Therefore, the ' 349 patent examples further demonstrate a " lack of comparability [that is] another reason
that the dot blot from Hospira's BLA does not prove infringement." (id.).

35

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 37 of 47 PageID #: 16204

Hospira did not offer any evidence to suggest that the differences identified by Dr. Hamilton
actually matter.
Hospira responds that the jury was reasonable in declining to credit Amgen's
circumstantial evidence of infringement of the '349 patent. (D.I. 373, p. 1). Hospira maintains
that Amgen failed to carry its burden to prove infringement because Amgen never established the
comparability of the dot-blot and RIA testing at issue. (Id. p. 4). First, Dr. Wall offered no
testimony on the comparability of dot-blot and RIA testing. (Id. (citing Trial Tr. 271 :5-22, 275: 19276:23)). Second, Dr. McLawhon did not "explain[] how the dot blot works or how it compares
to an RIA," instead converting mass to units "like a ' currency converter."' (Id. p. 5 (citing Trial
Tr. 527:24-530:1)). Third, "Dr. McLawhon d[id] not know whether the same antibody was used
in the standard he used for the conversion and the dot blot, although he admit[ted] that should be
done if one is going to make a comparison and that he would have known which antibody was
used if he had run an experiment." (Id. (citing Trial Tr. 545:3-546:2)). Fourth, a former Amgen
scientist who conducted RIA testing on Amgen' s EPO project testified that the same standard
should be used to compare assays, because the standard sets the potency measurement. (Id. p. 6
(citing Trial Tr. 1163:5-24)).
Hospira notes that Dr. Hamilton "provided several reasons why [RIA and dot blot] do not
yield similar results." (Id.). First, "Hospira' s BLA contained rough information that some of the
vertebrate cells tested could produce 100 µg of EPO per mL of cell-culture medium, [i.e.,
supernatant]." (Id. p. 7; Trial Tr. 540:18-541:3). Second, Hospira argues that Dr. Hamilton
"explained that the dot blot was done as a rough measurement of the amount of EPO." (Id. p. 8
(citing Trial Tr. 1193:18-1194:8)). Third, as Dr. Hamilton explained, "converting from mass units
to biological activity units as measured in the claims is not like converting currency because these

36

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 38 of 47 PageID #: 16205

units are not standardized like money." (Id. p. 7). Fourth, "a sample taken directly from the
supernatant, as done in the dot blot assay, will likely contain biologically inactive impurities and
EPO fragments that indiscriminately bind to the antibody, and thus will not be an appropriate
sample to use in the conversion calculation." (Id. p . 8). Fifth, one "can't rely on a purified standard
to give[] an estimate of what' s present in the crude mixture of impurities and isoforms as well as
active EPO on the dot blot." (Trial Tr. 1198 :2-6). Sixth, the standards used in Hospira's BLA and
in the ' 349 patent were not the same. (Id. 1198:19-1198:22). In summary,
[T]he standard was different, the antibodies were different, the assay design was
different, the relative degree of quantification was different, and based on all of
those variables, one can' t accurately assess the level ofEPO in a culture supernatant
cell preparation, which is really what claims 1 through 6 in the patent are requiring.

(Id. 1198:23-1199:6).
I agree with Hospira that substantial evidence supports the jury' s verdict that Hospira does
not infringe the asserted claims of the '349 patent. That Amgen' s calculation was the only one
offered at trial did not mean that the jury was obligated to credit it. And the fact that the ' 349
patent describes using a purified EPO standard to convert RIA assay data obtained from a crude
sample to "U" does not compel the conclusion that one could do the same with dot-blot data
obtained from a crude sample. Additionally, Amgen' s argument that Hospira' s BLA documents
disclosing dot-blot measured rates of 100 µg/mL constitute an admission that Hospira's cells meet
the production limitations is not persuasive. It ignores that the "admission" is explicitly dependent
on the measurement technique used, and does not address issues of whether the dot blot testing
and RIA testing at issue are comparable. Though Dr. Hamilton acknowledged that one could
theoretically convert the dot-blot µg/mL measurement to a measurement in "U," he qualified that
statement by saying that the material resulting from Hospira's dot-blot test would need to be
purified to get a proper conversion against the pure standard. (Trial Tr. 1214:18-21). Dr. Egrie, a

37

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 39 of 47 PageID #: 16206

former Amgen scientist, confirmed that the same standard should be used to compare assays. (Id.
1163:21-1164:5).

Dr. McLawhon admitted that he did not run any experiments to convert

Hospira's dot-blot results to a measurement in "U." (Id. 545:3-546:2).
I think that the testimony from Dr. Hamilton and Dr. Egrie provides substantial evidence
from which the jury could have concluded that the evidence presented did not establish that
Hospira' s dot-blot results could be reliably converted to RIA results. Without comparability, the
dot-blot production rate in Hospira's BLA would be meaningless to establish infringement, leading
the jury to the reasonable conclusion that Amgen had failed to carry its burden.

Amgen's

assertions that Hospira failed to provide affirmative evidence of noninfringement do not change
this result. (See D.I. 373 , p. 5). Amgen's contention that Dr. Hamilton's comparability testimony
should not be credited because he failed to tie it to the claim language also does not change the
result. Rather, Amgen's arguments represent an attempt to improperly shift the burden of proving
noninfringement to Hospira. To rebut Amgen's infringement argument, Dr. Hamilton need only
have presented testimony that called into question Dr. McLawhon's testimony such that a
reasonable jury could conclude that Amgen failed to prove infringement by a preponderance of
the evidence. 14 Dr. Hamilton's testimony was corroborated in part by Dr. Egrie. (Trial Tr.
1163:21-1164:17). I decline to supplant the jury's determinations of credibility with my own.
Thus, I conclude Amgen has failed to show that there is insufficient evidence for the jury' s verdict
that Hospira did not infringe the asserted claims of the '349 patent.
Accordingly, I will deny Amgen' s motion for JMOL that Hospira infringed the asserted
claims of the '349 patent.
B. New Trial

14

Alternatively, if the jury found Dr. Hamilton at least as credible as Dr. McLawhon, Amgen would not have proven
that it was more likely than not that Hospira infringed the ' 349 patent.

38

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 40 of 47 PageID #: 16207

First, Amgen asserts that it is entitled to a new trial because, "The great weight of the
evidence provided by Amgen established that Hospira's dot blot assays were comparable to an
RIA, and that Hospira's cells satisfied every limitation in the '349 claims." (D.I. 358, p. 15). This
argument essentially repeats Amgen's JMOL arguments.
Hospira responds that Dr. McLawhon "improperly converted the dot blot assay results into
biological activity units without testifying as to why the dot blot assay and the RIA are similar or
comparable." (D.I. 373, p. 14). According to Hospira, the jury's verdict is supported by Dr.
Hamilton's testimony "present[ing] several reasons why the two tests are not comparable,"
including that "the dot blot assay used an unpurified sample whereas the standard used in Amgen' s
calculations was a purified sample of EPO," and the dot blot assay and the RIA in the '349 patent
did not use the same standard or the same antibody. (Id. (citing Trial Tr. 1197:18-1198:2)).
I conclude that the jury's verdict was not against the great weight of the evidence for the
same reasons expressed with respect to Amgen's motion for JMOL of infringement of the ' 349
patent. Accordingly, I will deny Amgen's motion for a new trial on infringement of the '349 patent
on this ground.
Second, Amgen asserts that a new trial is warranted based on Hospira's statements during
its closing argument that "based on the evidence [] what is inside the fence is as determined by
RIA," while"outsidethefenceisdotblot." (D.I. 358, p. 15 ; Trial Tr.1641:11-1642:8). According
to Amgen, these statements were "legally improper, because [they] asked the jury to construe the
claims to require evidence produced during an RIA to prove infringement: that is, construing the
claims in such a way that they could never be infringed based on evidence from a dot-blot assay."
(D.I. 358, p. 15). 15 By contrast, Hospira characterizes its statements as "not ask[ing] the jury to

15

Amgen also cites Amgen Inc. v. F. Hoffman-La Roche Ltd. , 580 F.3d 1340, 1385 (Fed. Cir. 2009) as further support
for its assertion that Hospira' s statements in closing arguments were improper, because that case held that proof of

39

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 41 of 47 PageID #: 16208

make a claim construction decision" and "merely point[ing] out, correctly, that 'as determined by
radioimmunoassay' is literally recited in the claims." (D.I. 373, p. 15). As support for its "outside
the fence" statements, Hospira's counsel pointed to Dr. Hamilton's testimony that Amgen had not
adequately established that the dot-blot testing was sufficiently comparable to the RIA testing
recited in the '349 claims for the dot-blot results to prove infringement. (Id.; Trial Tr. 1641:241642:8).

Further, "Hospira's counsel never said that dot blot could not be used or that

circumstantial evidence was not allowed. " (D.I. 373, p. 15).
I agree with Hospira. Taken in context, I do not think that Hospira's "outside the fence"
statements request that the jury engage in claim construction.

Rather, Hospira used these

statements to highlight for the jury Amgen's failure of proof of infringement of the '349 patent.
Accordingly, I will deny Amgen' s motion for a new trial based on Hospira's counsel 's statements
during closing argument.

V.

AM GEN'S MOTION FOR PREJUDGMENT AND POST-JUDGMENT INTEREST
A. Prejudgment Interest
Asserting "Prejudgment interest on a damages award for patent infringement 'is the rule'

under 35 U.S .C. § 284," Amgen moves to amend the judgment under Fed. R. Civ. P. 59(e) to
award prejudgment interest to Amgen. (D.I. 352, pp. 1-2 (citing Sensonics, Inc. v. Aerosonic

Corp., 81 F.3d 1566, 1574 (Fed. Cir. 1996); see also General Motors Corp. v. Devex Corp., 461
U.S. 648 (1983))). Hospira requests that I deny Amgen's request for prejudgment interest, arguing
that since "Hospira has not received FDA approval or sold its proposed EPO product, and Amgen

infringement of the '349 patent did not require RIA evidence. (D.1. 358, p. 16). Hospira counters, "The Roche case
does not provide any support for having a new trial because it merely held that Amgen could present its evidence to a
jury- it did not say the jury had to rule in Amgen's favor." (D.I. 373, p. 15). I agree with Hospira. For the reasons
stated with respect to Amgen's motion for JMOL of infringement of the '349 patent, substantial evidence supports the
jury's conclusion that the dot-blot evidence presented by Amgen was insufficient to prove infringement of the ' 349
patent.

40

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 42 of 47 PageID #: 16209

has suffered no economic harm," a prejudgment interest award would create a windfall for Amgen.
(D.I. 376, p. 1). In other words, Hospira appears to argue that since Amgen has suffered no
economic harm, prejudgment interest is not required to make Amgen whole. Hospira' s argument
does not fully account for the purpose of prejudgment interest "to ensure that the patent owner is
placed in as good a position as he would have been had the infringer entered into a reasonable
royalty agreement." General Motors , 461 U.S. at 655-56 (" An award of interest from the time that
the royalty payments would have been received merely serves to make the patent owner whole,
since his damages consist not only of the value of the royalty payments but also of the foregone
use of the money between the time of infringement and the date of the judgment."). To make
Amgen whole, I will award Amgen prejudgment interest.
Assuming that prejudgment interest should be awarded, the parties dispute whether interest
should be calculated as a lump sum or on a per-batch basis, and the appropriate interest rate.
Hospira asserts that prejudgment interest should be awarded on a per-batch basis, because
discrete acts of infringement (i.e., manufacture) occurred on identifiable dates, justifying
incremental payments of a reasonable royalty.

(D.I. 376, pp. 1-2).

According to Hospira,

prejudgment interest on a lump sum royalty payment is inappropriate, because it "assume[s] a
royalty would have been paid on EPO batches well before they even existed." (Id. p. 1). Amgen
counters that prejudgment interest on a lump sum royalty is appropriate because, "Based on expert
testimony that the lump-sum royalty would have been determined at the time of the hypothetical
negotiation in late 2013, the jury awarded Amgen $70 million as lump-sum reasonable-royalty
damages." (D.I. 382, p. 1). The jury could not have premised its award on Hospira' s per-batch
damages theory at trial, Amgen maintains, because Hospira' s theory required that sales take place
to trigger any damages award, and Hospira has made no sales to date. (Id.). I agree with Amgen.

41

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 43 of 47 PageID #: 16210

The verdict form reflects that the jury awarded Amgen $70 million in damages, without any
mention of batches or indication that the award was calculated on a per-batch basis. (See D.I. 325
at 4).
With respect to the applicable interest rate, Amgen submits that an award of prejudgment
interest calculated at the prime rate, compounded quarterly is appropriate. (D.I. 352, p. 2). Since
Hospira would have needed a license to begin manufacturing EPO batches, Amgen argues that I
should award Amgen prejudgment interest starting on November 10, 2013, the date of manufacture
of the first infringing batch. (Id. pp. 2-3). Hospira asserts that prejudgment interest should be
awarded at "Amgen' s average debt rate on a per-batch basis," instead of the prime rate. (D.I. 376,
pp. 2, 10). 16 According to Hospira, since "Amgen's 10-Ks show loans at rates at significantly
below the prime rate," the prime rate "is not supported by evidence and would create a windfall
for Amgen." (Id. p. 2). Amgen responds that Hospira bases its calculation on "two instances in
Amgen' s corporate filings, identifying a term loan entered in 2013 and a revolving credit
agreement entered in 2014." (D.I. 382, p. 5). This is improper, Amgen submits, because "Hospira
is simply speculating that Amgen would have used the awarded royalty, had it been paid when
due, to pay off these loans; or, alternatively, that Amgen would have borrowed the money to invest
in its business in anticipation that one day Hospira would pay the $70 million owed." (Id.).
Therefore, Amgen asserts, the prime rate should apply, because "awarding prejudgment interest at
the prime rate is one way of compensating Amgen that numerous courts, including this Court, have
found to be fair and reasonable." (Id. p. 1).

I agree with Amgen. "Courts have recognized that the prime rate best compensate[ s] a

16

Hospira calculates the total prejudgment interest due under its average debt rate theory on a per-batch basis at
$4,843 ,492. (D.I. 376, p. 10). Hospira calculates the total prejudgment interest due under its average debt rate theory
on a lump-sum basis at $6,276,396. (Id. p. 12).

42

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 44 of 47 PageID #: 16211

patentee for lost revenues during the period of infringement because the prime rate represents the
cost of borrowing money, which is ' a better measure of the harm suffered as a result of the loss of
the use of money over time. "' IMX, Inc. v. LendingTree, LLC, 469 F. Supp. 2d 203 , 227 (D. Del.
2007). Therefore, I will set prejudgment interest at the prime rate, compounded quarterly. See,

e.g. , Ironworks Patents, LLC v. Apple, Inc., 2017 WL 2535877, at *14 (D. Del. June 12, 2017);
LG Display Co. v. AU Optronics Corp. , 722 F. Supp. 2d 466, 475 (D. Del. 2010).
Accordingly, I will award Amgen prejudgment interest using the prime rate compounded
quarterly and applied against the $70 million lump-sum reasonable royalty award beginning on
November 10, 2013 .
B. Post-judgment Interest

The parties agree that post-judgment interest should accrue at the statutory rate as specified
in 28 U.S.C. § 1961(a). (D.I. 352, p. 3; D.I. 376, pp. 12-13). They disagree, however, regarding
when post-judgment interest begins to accrue on the judgment and when post-judgment interest
begins to accrue on the prejudgment interest.
Amgen asserts, "prejudgment interest [should] be awarded through the date of the final
judgment and [] post-judgment interest (on the jury award and the prejudgment interest) [should]
be awarded after that date." (D.I. 382, p. 6). Hospira argues that post-judgment interest should
begin to accrue on the damages award as of September 25, 2017, the date the Court entered
judgment on the jury' s verdict, and that post-judgment interest should begin to accrue on the
prejudgment interest as of the date that the prejudgment interest is quantified. (D.I. 376, p. 14).
Section 196l(a) provides, "Interest shall be allowed on any money judgment in a civil case
recovered in a district court. . . . Such interest shall be calculated from the date of the entry of the
judgment .... " 28 U.S.C. § 196l(a). The Third Circuit addressed the accrual of post-judgment

43

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 45 of 47 PageID #: 16212

interest under§ 1961(a) in Eaves v. Cty. of Cape May, 239 F.3d 527 (3d Cir. 2001). See Travelers
Cas. & Sur. Co. v. Ins. Co. of N Am. , 609 F.3d 143, 174-75 (3d Cir. 2010) (declining to limit
Eaves to the attorneys' fees context). "Given the plain language and structure of the statute, it is

clear that ' the judgment' referred to in the third quoted sentence is the 'money judgment' specified
in the first." Eaves, 239 F.3d at 532. "[T]he phrase 'any money judgment' in§ 1961(a) [] requires
that the judgment at issue award a fixed amount of fees to the prevailing party in order to trigger
the post-judgment interest period." Id. at 534. Therefore, "post-judgment interest begins to run
on a judgment awarding attorney' s fees where that judgment fixes the amount owed to the
prevailing party." Id. "The statute does not, by its terms, mandate that the judgment from which
interest is calculated must be a final judgment." In re Lower Lake Erie Iron Ore Antitrust Litig. ,
998 F.2d 1144, 1177-78 (3d Cir. 1993); see also Skretvedt v. E.I DuPont De Nemours, 372 F.3d
193, 216 (3d Cir. 2004) ("The fact that the December 13, 2001,judgment was not a final order for
purposes of appeal would not otherwise prevent postjudgment interest from running under
§ 1961 ... .").
On September 25, 2017, I entered judgment for Amgen and against Hospira on the jury' s
verdict in the amount of $70 million. (D.I. 327). As of that date, I entered a "money judgment"
for Amgen that "include[d] both 'an identification of the parties for and against whom the
judgment [wa]s being entered,' and 'a definite and certain designation of the amount ... owed."'
See Travelers , 609 F.3d at 175 (quoting Eaves, 239 F.3d at 533) (brackets added). Accordingly, I

will award Amgen post-judgment interest on the $70 million damages award beginning on
September 25, 2017. Prejudgment interest, however, will not have been quantified in a money
judgment until the date of the final judgment accompanying this opinion. Accordingly, I will
award Amgen post-judgment interest on the prejudgment interest commencing on the date of entry

44

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 46 of 47 PageID #: 16213

of the final judgment quantifying the amount of prejudgment interest owed to Amgen.

See

Travelers, 609 F.3d at 175 (holding that though the district court entered judgment in favor of
Travelers on August 14, 2006, "post-judgment interest on Travelers' award of prejudgment
interest did not begin to run until the December 5, 2007 order was entered quantifying the amount
in prejudgment interest owed to Travelers.").
Amgen cites several cases from this District in which "prejudgment interest has [] been
awarded through the date of entry of final judgment, rather than the date of the jury's verdict."
(D.I. 382, pp. 6-8 (citing LG Display Co., Ltd v. AU Optronics Corp. , 722 F. Supp. 2d 466, 475
(D. Del. 2010); Telecordia Techs. , Inc. v. Cisco Sys. , Inc., 592 F. Supp. 2d 727, 748-49 (D. Del.
2009), vacated in part on other grounds, 612 F.3d 1365, 1379 (Fed. Cir. 2010); Northeast

Controls, Inc. v. Fisher Controls Intern. , LLC, 2008 WL 678701 , at *2 (D. Del. Mar. 12, 2008),
rev 'd on other grounds, 373 F. App'x 162 (3d Cir. 2010); Tristrata Tech. , Inc. v. Mary Kay Inc.,
423 F. Supp. 2d 456, 471 (D. Del. 2006))). Each of these cases, with the exception of LG Display,
was decided before the Third Circuit's Travelers decision. Additionally, none of these cases
provide any explanation for their selection of the date through which prejudgment interest was
awarded, or any indication that the parties disputed the date through which prejudgment interest
would accrue. Amgen also asserts, "The cases that Hospira cites state that post-judgment interest
may be awarded on a judgment that sets the amount of the damages, but they do not address the
appropriate timing for switching from the prejudgment rate to the post-judgment rate." (D.I. 382,
p. 7). Notably, Amgen' s discussion of Hospira's cited cases omits any mention of Travelers. In
fact, Travelers is not cited anywhere in Amgen's brief. Therefore, I do not find convincing
Amgen' s argument that pre-judgment interest should be awarded on the $70 million award through
the date of final judgment.

45

Case 1:15-cv-00839-RGA Document 386 Filed 08/27/18 Page 47 of 47 PageID #: 16214

Therefore, I will award Amgen prejudgment interest using the prime rate compounded
quarterly and applied against the $70 million lump-sum reasonable royalty award beginning on
November 10, 2013 . I will award Amgen post-judgment interest on the $70 million award
beginning on September 25, 2017, the date judgment was entered on the award. I will award
Amgen post-judgment interest on the prejudgment interest beginning on the date of entry of the
final judgment quantifying the amount of prejudgment interest owed to Amgen in accordance with
this opinion.
VI.

CONCLUSION
For the reasons set forth above, Hospira' s Rule 50(a) Motion for Judgment as a Matter of

Law on the Issues of Safe Harbor, Noninfringement, Invalidity, and Damages (D.I. 336) is
dismissed as moot. Hospira' s Motion for Judgment as a Matter of Law Under Rule 50(b) and, in
the Alternative, For Remittitur or New Trial Under Rule 59 (D.I. 355), Hospira' s Motion to Seal
Confidential Exhibits Admitted at Trial (D.I. 361), and Amgen' s Renewed Motion for Judgment
as a Matter of Law oflnfringement of the ' 349 Patent or, in the Alternative, for a New Trial (D.I.
356), are each denied. Amgen's Motion for Prejudgment and Post-judgment Interest (D.I. 352) is
granted-in-part.

46