Case 1:15-cv-00697-RGA Document 109 Filed 10/23/17 Page 1 of 12 PageID #: 1636
`
`IN THE UNITED STATES DISTRICT COURT
`FOR THE DISTRICT OF DELAWARE
`
`
`HOSPIRA, INC.,
`
`
`
`
`
`AMNEAL PHARMACEUTICALS LLC,
`
`
`
`
`
`v.
`
`
`
`
`
`
`
`
`
`
`
`Plaintiff,
`
`Defendant.
`
`)
`)
`)
`)
`)
`)
`)
`)
`)
`)
`
`
`
`
`
`
` Civil Action No. 15-697-RGA
`
`
`
`
`
`
`
`
`
`
`
`
`HOSPIRA’S POST-TRIAL REPLY BRIEF ON INFRINGEMENT
`
`

`

`Case 1:15-cv-00697-RGA Document 109 Filed 10/23/17 Page 2 of 12 PageID #: 1637
`
`TABLE OF CONTENTS
`
`AMNEAL INFRINGES THE 2% LIMITATION ...............................................................1 
`A. 
`Amneal Meets The 2% Limitation As A Matter Of Law. .......................................1 
`B. 
`Amneal Infringes The 2% Limitation As A Matter Of Fact. ...................................3 
`1. 
`Dr. Bloch’s Opinions Are Not Relevant To The 2% Limitation. ................4 
`2. 
`Amneal’s Stability Data Prove Infringement. ..............................................6 
`a. 
`HPLC Is The Standard Tool For Measuring Potency Loss. ............6 
`b. 
`Dr. Linhardt Considered The Appropriate Rates of Loss. ...............7 
`c. 
`All Of Amneal’s Stability Data Show Infringement. .......................7 
`d. 
`Amneal Cannot Disavow Its Conclusion About Stability. ..............8 
`CONCLUSION ....................................................................................................................8 
`
`
`
`
`i
`
`
`I. 
`
`II. 
`
`
`
`
`
`

`

`Case 1:15-cv-00697-RGA Document 109 Filed 10/23/17 Page 3 of 12 PageID #: 1638
`
`
`
`
`
`CASES
`
`TABLE OF AUTHORITIES
`
`Page(s)
`
`Allure Energy, Inc. v. Nest Labs., Inc.,
`2015 WL 11110634 (E.D. Tex. Apr. 13, 2015) .........................................................................5
`
`Ferring B.V. v. Watson Labs., Inc.,
`764 F.3d 1382 (Fed. Cir. 2014)..............................................................................................2, 3
`
`Med. Co. v. Mylan, Inc.,
`853 F.3d 1296 (Fed. Cir. 2017)..................................................................................................3
`
`Nikken USA, Inc. v. Robinsons-May, Inc.,
`51 F. App’x 874 (Fed. Cir. 2002) ..............................................................................................2
`
`Novartis AG v. Actavis Elizabeth LLC,
`2017 WL 1398347 (D. Del. Apr. 17, 2017) ...............................................................................5
`
`Phillips v. AWH Corp.,
`415 F.3d 1303 (Fed. Cir. 2005)..................................................................................................5
`
`SEB S.A. v. Montgomery Ward & Co.,
`594 F.3d 1360 (Fed. Cir. 2010)..................................................................................................5
`
`Sundance, Inc. v. DeMonte Fabricating Ltd.,
`550 F.3d 1356 (Fed. Cir. 2008)..........................................................................................4, 5, 6
`
`Sunovion Pharm., Inc. v. Teva Pharm. USA, Inc.,
`731 F.3d 1271 (Fed. Cir. 2013)..........................................................................................1, 2, 3
`
`OTHER AUTHORITIES
`
`FEDERAL RULE OF EVIDENCE 702 .......................................................................................4, 5
`
` 
`
`
`
`ii
`
`

`

`Case 1:15-cv-00697-RGA Document 109 Filed 10/23/17 Page 4 of 12 PageID #: 1639
`
`Amneal takes numerous contradictory positions regarding the 2% limitation. For
`
`obviousness, Amneal proffers a select few stability data points from Hospira’s internal work and
`
`asserts that they prove by clear and convincing evidence that all dexmedetomidine compositions
`
`meet the 2% limitation. (D.I. 100 at 14-16.) Then, on indefiniteness, it takes a single stability
`
`data point from Example 6 of the patent and contends that it shows by clear and convincing
`
`evidence that the product discussed there does not meet the limitation. (Id. at 23-24.) Now,
`
`Amneal argues that its entire stability study—submitted to the FDA to establish the “stability
`
`characteristics” of its products (PTX 93.4)—is insufficient to establish infringement of the 2%
`
`limitation by a preponderance of the evidence. (D.I. 105 at 13.) This comes after Amneal earlier
`
`averred that its ANDA “sufficiently describes” its product for purposes of this case such that test
`
`samples would be merely “duplicative” of information in the ANDA. (JTX 83.13-14.)
`
`Amneal’s house-of-cards defense cannot stand. The 2% limitation, along with all of the
`
`limitations from the remainder of the asserted claims, are valid and infringed.
`
`I.
`
`AMNEAL INFRINGES THE 2% LIMITATION
`As described in Hospira’s Opening Brief (D.I. 101 at 4-14), Amneal infringes the 2%
`
`limitation as a matter of both law and fact.1
`
`A.
`
`Amneal Meets The 2% Limitation As A Matter Of Law.
`
`Amneal cannot escape Sunovion here. In Sunovion, the Federal Circuit held that a claim
`
`to ‘less than 0.25%’ impurity was infringed by an ANDA specification providing for less than
`
`0.6% of the impurity. Sunovion Pharm., Inc. v. Teva Pharm. USA, Inc., 731 F.3d 1271, 1278
`
`
`1 The relevant claim language is “wherein the liquid pharmaceutical composition when stored in
`the glass container for at least five months exhibits no more than about 2% decrease in the
`concentration of dexmedetomidine.” (JTX 4.15.)
`
`
`
`1
`
`

`

`Case 1:15-cv-00697-RGA Document 109 Filed 10/23/17 Page 5 of 12 PageID #: 1640
`
`(Fed. Cir. 2013). Here, the claim to ‘not more than about 2% decrease’ is infringed by Amneal’s
`
`ANDA specification of not more than 10% decrease. (D.I. 101 at 5-7.)
`
`Amneal argues that Sunovion does not apply because its ANDA specification is for
`
`twenty-four months of storage whereas the claim is directed to five months of storage. (D.I. 105
`
`at 3-4.) This is incorrect. First, its specification of no more than 10% loss after twenty-four
`
`months necessarily specifies no more than 10% loss after five months—there cannot be more
`
`loss at five months than is permitted over the product’s entire shelf life because the amount of
`
`dexmedetomidine does not increase over time. (See Tr. 286:17-23; 447:1-7.) Second, the claim
`
`recites “at least five months” of storage, and so covers no more than about 2% loss after twenty-
`
`four months of storage. Third, Amneal’s argument implies that a claim requiring no more than
`
`about 2% loss at twenty-four months would be infringed, but that a broader claim requiring only
`
`no more than about 2% loss after five months would not be infringed. This cannot be. See, e.g.,
`
`Nikken USA, Inc. v. Robinsons-May, Inc., 51 F. App’x 874, 882 (Fed. Cir. 2002) (noting that
`
`infringement of narrower claim necessarily results in infringement of broader claim).
`
`Amneal’s reliance on the far-afield Ferring case is instructive. (D.I. 105 at 3-4.) That
`
`case is inapposite because the ANDA specification there was silent on certain claim limitations.
`
`Ferring B.V. v. Watson Labs., Inc., 764 F.3d 1382, 1387 (Fed. Cir. 2014). Specifically, the
`
`claims recited a gradually-dissolving drug whose dissolution matched the drug’s absorption rate
`
`in the body. Id. at 1384. They required that less than 40% of the drug dissolve after 15 minutes,
`
`less than 70% of the drug dissolve after 45 minutes, and more than 50% of the drug dissolve
`
`after 90 minutes (one claim had only the 45-minute requirement). Id. at 1385. By contrast, the
`
`ANDA provided only that more than 80% of the drug dissolve in 60 minutes. Id. at 1385-86.
`
`
`
`2
`
`

`

`Case 1:15-cv-00697-RGA Document 109 Filed 10/23/17 Page 6 of 12 PageID #: 1641
`
`Because it said nothing about whether the dissolution amount would be above or below the
`
`claimed maximums at 15 and 45 minutes, it did not address the claims’ limitations. Id. at 1387.
`
`Amneal’s citation to Medicines Company also misses the mark. In that case, the claim
`
`required not only a maximum impurity amount but also a process of “efficient mixing.” Med.
`
`Co. v. Mylan, Inc., 853 F.3d 1296, 1309-10 (Fed. Cir. 2017). The ANDA specification did not
`
`address this additional “efficient mixing” limitation, so Sunovion did not apply. Id. at 1310.
`
`Here, on the other hand, Amneal’s ANDA specification speaks directly to the single
`
`claim limitation at issue. It defines a set of products that includes a product having no more than
`
`about 2% decrease in dexmedetomidine concentration after at least five months of storage. So,
`
`as in Sunovion, the ANDA specification defines infringement. See Sunovion, 731 F.3d at 1278.
`
`In fact, Amneal’s infringement is even more compelling than in Sunovion because its ANDA
`
`specification of 10% loss after twenty-four months mirrors the claimed loss of about 2% after at
`
`least five months. (See D.I. 101 at 6-7.) This equivalency derives from a linear model of
`
`dexmedetomidine loss over time. (See id. at 6.) Dr. Linhardt’s unrebutted testimony confirmed
`
`that this is the best model to evaluate loss. (Tr. 294:7-23.)
`
`Finally, Amneal argues that Sunovion cannot apply because HPLC measurement
`
`variability means that its ANDA “cannot possibly be interpreted to directly address” the 2%
`
`limitation. (D.I. 105 at 5.) Amneal is wrong. The ANDA specification squarely addresses the
`
`claimed limitation: both use HPLC to measure dexmedetomidine loss; both contemplate
`
`measurement variability. (JTX 76.4 (providing for up to 110% potency measured by HPLC);
`
`JTX-4 at 5:31-35, Examples 1, 4-6 (defining “about” to account for measurement error and using
`
`HPLC to measure potency).)
`
`B.
`
`Amneal Infringes The 2% Limitation As A Matter Of Fact.
`
`Even setting aside Sunovion, Amneal infringes the 2% limitation. (See D.I. 101 at 7-14.)
`
`
`
`3
`
`

`

`Case 1:15-cv-00697-RGA Document 109 Filed 10/23/17 Page 7 of 12 PageID #: 1642
`
`1.
`
`Dr. Bloch’s Opinions Are Not Relevant To The 2% Limitation.
`
`With no evidence from a POSA, and no expert opining that it does not infringe, Amneal
`
`turns to its statistician, Dr. Bloch. But Amneal acknowledges that Dr. Bloch is not a POSA.
`
`(D.I. 105 at 6.) And because infringement and invalidity are “analysed in great part from the
`
`perspective of a person of ordinary skill in the art,” a witness cannot “testify as an expert on the
`
`issues of non-infringement or invalidity unless that witness is qualified as an expert in the
`
`pertinent art” because “where an issue calls for consideration of evidence from the perspective of
`
`one of ordinary skill in the art, it is contradictory to Rule 702 to allow a witness to testify on the
`
`issue who is not qualified as a technical expert in that art.” Sundance, Inc. v. DeMonte
`
`Fabricating Ltd., 550 F.3d 1356, 1361, 1363 (Fed. Cir. 2008).
`
`Amneal does not account for this. Instead, it makes much of Sundance’s note that non-
`
`POSA experts may testify at trial, but context shows that the court was alluding to permissible
`
`testimony from a non-POSA on background issues, not opinions on infringement. See id. at
`
`1363 n. 5 (“[P]atent lawyers might offer testimony in contexts other than non-infringement and
`
`invalidity, such as patent office practice . . . a chemist not skilled in the particular art at issue
`
`might nevertheless appropriately testify as an expert as to the consequences of reacting two
`
`chemicals in a particular environment.”). Dr. Bloch plays no such permissible role here.
`
`Notwithstanding Sundance, Amneal urges that Dr. Bloch’s opinions are relevant because
`
`he is “an expert in the relevant technical field of biostatistics.” (D.I. 105 at 6.) But this is not a
`
`relevant technical field here. Nowhere does the patent say anything about statistical analysis for
`
`the 2% limitation. Rather, the patent teaches that the inventors determined stability by
`
`conducting stability tests. (See JTX-4 at Examples 1, 3, 6.) Neither the patent nor the
`
`prosecution history says anything about ascribing statistical significance to these stability results.
`
`
`
`4
`
`

`

`Case 1:15-cv-00697-RGA Document 109 Filed 10/23/17 Page 8 of 12 PageID #: 1643
`
`This is unsurprising because, as an unrebutted Dr. Linhardt explained, a POSA does not delve
`
`into biostatistics for this type of stability testing. (Tr. 288:12-289:15, 351:2-352:5.)
`
`Amneal fails to identify any instance where courts have found testimony like Dr. Bloch’s
`
`useful. It first tries to piggyback on SEB’s quotation that “this case comes nowhere close to the
`
`unusual situation in Sundance,” but provides no description of that case. (D.I. 105 at 6.) SEB is
`
`nothing like this case—it involved an expert who was a POSA at least on some aspects of the
`
`technology. 594 F.3d 1360, 1373 (Fed. Cir. 2010). Amneal’s second case similarly fails to
`
`support it. See Allure Energy, Inc. v. Nest Labs., Inc., 2015 WL 11110634, at *2 (E.D. Tex. Apr.
`
`13, 2015) (finding expert “more than meets the requirements set forth for a PHOSITA” and
`
`noting that he must be an expert on “the technology about which he or she is opining”).
`
`Amneal cites one case where a non-POSA statistician was not precluded from testifying
`
`at trial. (D.I. 105 at 7 (citing Novartis AG v. Actavis Elizabeth LLC, 2017 WL 1398347, at *1
`
`(D. Del. Apr. 17, 2017).) But there, two examples in the patent specifically discussed
`
`“statistically significant” clinical effects, and the statistician opined that the patent was non-
`
`enabled because the examples did not provide sufficient data to “verify or repeat” this statistical
`
`significance. (See U.S. Patent No. 5,604,229 at col. 278; D.I. 287-7 at 8 & D.I. 287-2 at 23, Case
`
`No. 14-1487-LPS.) Here, the patent says nothing about statistics or statistical significance.
`
`Dr. Bloch’s disconnect from the subject matter at issue serves only to sow confusion. As
`
`a non-POSA, he raises issues with no basis in the patent or art, thereby constructing an
`
`inappropriate burden to prove infringement. Cf. Phillips v. AWH Corp., 415 F.3d 1303, 1318-19
`
`(Fed. Cir. 2005) (instructing that extrinsic evidence cannot “change the meaning of claims in
`
`derogation” of the intrinsic record). His testimony distracts from the appropriate analysis for the
`
`2% limitation and is unhelpful. Compare Fed. R. Evid. 702 (requiring that expert’s knowledge
`
`
`
`5
`
`

`

`Case 1:15-cv-00697-RGA Document 109 Filed 10/23/17 Page 9 of 12 PageID #: 1644
`
`“help the trier of fact to understand the evidence or to determine a fact in issue”), with Tr.
`
`447:20-448:1 (alleging that Dr. Linhardt may have used an incorrect kinetic model but admitting
`
`that “I’m not a POSA. I can’t judge what is right.”); 490:16-491:17 (admitting that it would be
`
`“a good idea” to consider the stability data as a whole but “I didn’t do that, because I would have
`
`to ask the scientist”); 470:11-471:13 (testifying that “I never really thought about whether is it
`
`reasonable or not” for Dr. Linhardt to rely on Amneal’s data because “I don’t know how to judge
`
`whether it’s reasonable or not”). Under Sundance, Dr. Bloch’s testimony should be disregarded.
`
`2.
`
`Amneal’s Stability Data Prove Infringement.
`
`Dr. Linhardt provided unrebutted testimony that a POSA determines whether the 2%
`
`limitation is met by considering stability data and performing regression analysis. (See D.I. 101
`
`at 9.) In addition to relying on his decades of experience and teaching POSAs, Dr. Linhardt cited
`
`literature assessing kinetics through this analysis. (Tr. 405:6-410:2.) Amneal’s wayward
`
`criticisms of his analysis disregard how stability data is interpreted in the art.
`
`a.
`
`HPLC Is The Standard Tool For Measuring Potency Loss.
`
`Amneal first argues that data variability prevents one from knowing whether Amneal’s
`
`stability data show less than about 2% loss over at least five months of storage. (D.I. 105 at 8-9.)
`
`It argues that the data were collected by HPLC, whose error range of 2-5% “would swallow the
`
`2% limitation.” (Id.) But the patent teaches the use of HPLC to measure stability, and allows for
`
`data variability within the scope of the claim by reciting “about 2%” loss. (JTX 4 at 5:31-43,
`
`13:43-48, 20:37-50.) And there is no dispute that HPLC is a standard technique in the art for
`
`measuring concentration changes. (Tr. 256:8-18, 257:12-20.) Amneal’s own stability study does
`
`not discuss data variability when representing to the FDA that its data demonstrate “no
`
`significant change” in concentration over eighteen months of long-term storage. (JTX 56.17.)
`
`
`
`6
`
`

`

`Case 1:15-cv-00697-RGA Document 109 Filed 10/23/17 Page 10 of 12 PageID #: 1645
`
`Amneal provides no evidence that ‘data variability’ affects the results of Dr. Linhardt’s
`
`analysis. (See D.I. 105 at 8 (proffering only “a hypothetical illustration” with no supporting
`
`testimony).) Nor could it. Dr. Linhardt accounted for variability by considering the data in
`
`Amneal’s stability study as a whole, and finding that all of the data met the 2% limitation across
`
`every means of analysis that he employed. (Tr. 289:16-291:11; D.I. 101 at 10-11.)
`
`b.
`
`Dr. Linhardt Considered The Appropriate Rates of Loss.
`
`The possible losses of active ingredient were discussed extensively at trial: absorption,
`
`adsorption, degradation, and contamination. (Tr. 86:3-87:17, 374:12-375:7, 925:18-926:12.)
`
`For dexmedetomidine in a sealed glass container, the most likely losses are adsorption and
`
`degradation. (Tr. 405:10-406:4.) Dr. Linhardt determined that adsorption “fit[s] very well” with
`
`a linear rate model, while degradation can exhibit a first order rate model. (Tr. 295:16-296:18.)
`
`To account for both possible losses, he analysed Amneal’s stability under both models, and
`
`found that the loss was not more than about 2% under either model. (Tr. 296:19-297:1.)
`
`Amneal suggests that a “mixed” linear-first order loss model might apply instead. (D.I.
`
`105 at 10.) But either model shows infringement; a mix of the models would necessarily show
`
`the same. (D.I. 101 at 14.) In any event, Amneal offers no evidence to rebut Dr. Linhardt’s use
`
`of linear and first order models: Dr. Bloch admitted he had no idea which model applies. (Id.)
`
`c.
`
`All Of Amneal’s Stability Data Show Infringement.
`
`In view of Dr. Linhardt’s explanation that all of Amneal’s stability data showed
`
`infringement (D.I. 101 at 11), and Dr. Bloch offering no opinion of non-infringement (id. at 12),
`
`Amneal argues that Dr. Linhardt’s calculations cannot mean what they say. Amneal contends
`
`that a calculation showing, for example, 1.8% loss after five months could equally mean some
`
`other loss based on “confidence intervals.” (D.I. 105 at 11-14.) This is yet another statistical
`
`parameter nowhere contemplated by the patents or a POSA. Amneal never explains how these
`
`
`
`7
`
`

`

`Case 1:15-cv-00697-RGA Document 109 Filed 10/23/17 Page 11 of 12 PageID #: 1646
`
`“confidence intervals” could be managed such that its own stability data has any meaning. (See
`
`id.) Perhaps it is suggesting that more data is needed. (Id. at 13.) But Amneal cannot assert that
`
`an entire FDA-worthy stability study is insufficient to assess infringement by a preponderance of
`
`the evidence while maintaining that select data points suffice to prove inherency and
`
`indefiniteness by clear and convincing evidence. Nor can Amneal reconcile this argument with
`
`its statement that its ANDA “sufficiently describes” its products such that test samples would be
`
`“duplicative.” (See D.I. 101 at 12.) Amneal now recasts this position as a “discovery dispute”
`
`(D.I. 105 at 15), but it plainly acknowledges that its ANDA is dispositive. Equally implausible is
`
`Amneal’s assertion that its ‘sufficiently describes’ position “merely reflects its contention that
`
`the 2% limitation is inherent” (id.)—nothing in Amneal’s statement concerns inherency.
`
`d.
`
`Amneal Cannot Disavow Its Conclusion About Stability.
`
`Finally, Amneal seeks to downplay its conclusion that its ANDA stability study
`
`demonstrates “no significant change observed from initial to 6 Months at accelerated and from
`
`initial to 18 Months at long term study results.” (JTX 56.17; D.I. 105 at 14-15.) This conclusion
`
`matters. First, the 6-month data point for one of the two accelerated tests and both 18-month
`
`data points for the long-term tests show no more than about 2% potency loss, corroborating Dr.
`
`Linhardt’s analysis at 5 months. (See JTX 56.9, 56.11, 56.15.) Second, Amneal concluded this
`
`without any of the statistical assurances that it now seeks from Hospira. Finally, the conclusion
`
`is made in the context of demonstrating a twenty-four month shelf life with no more than 10%
`
`potency loss; this loss is commensurate with the 2% limitation. (Supra p. 3; D.I. 101 at 11.)
`
`II.
`
`CONCLUSION
`Hospira requests judgment that the asserted claims are valid and infringed. 2
`
`
`2 Amneal provided no response to Hospira’s proof of infringement of “sealed.” (D.I. 101 at 2-4.)
`
`
`
`8
`
`

`

`Case 1:15-cv-00697-RGA Document 109 Filed 10/23/17 Page 12 of 12 PageID #: 1647
`
`Dated: October 23, 2017
`
`
`
`
`
`
`
`
`Respectfully Submitted,
`
`
`
`HOSPIRA, INC.
`
`
`By: /s/ Ryan P. Newell
`
`CONNOLLY GALLAGHER LLP
`Arthur G. Connolly, III (# 2667)
`Ryan P. Newell (# 4744)
`The Brandywine Building
`1000 West Street, Suite 1400
`Wilmington, Delaware 19801
`Telephone: (302) 757-7300
`aconnolly@connollygallagher.com
`rnewell@connollygallagher.com
`
`Bradford P. Lyerla
`Sara T. Horton
`Yusuf Esat
`Chad J. Ray
`JENNER &BLOCK LLP
`353 N. Clark Street
`Chicago, IL 60654-3456
`Telephone: 312 222-9350
`Facsimile: 312 527-0484
`blyerla@jenner.com
`shorton@jenner.com
`yesat@jenner.com
`
`Attorneys for Plaintiff Hospira, Inc.
`
`9
`
`