Case 2:15-cv-01455-WCB Document 190-1 Filed 10/26/16 Page 1 of 13 PageID #: 7093
`Case 2:15-cv-01455-WCB Document 190-1 Filed 10/26/16 Page 1 of 13 PageID #: 7093
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`EXHIBIT A
`EXHIBIT A
`
`

`

`Case 2:15-cv-01455-WCB Document 190-1 Filed 10/26/16 Page 2 of 13 PageID #: 7094
`
`IN THE UNITED STATES DISTRICT COURT
`FOR THE EASTERN DISTRICT OF TEXAS
`MARSHALL DIVISION
`
`
`
`ALLERGAN, INC.,
`Plaintiff,
`
`
`v.
`
`TEVA PHARMACEUTICALS USA, INC.,
`AKORN, INC., MYLAN
`PHARMACEUTICALS INC., and MYLAN
`INC.,
`
`Defendants.
`ALLERGAN, INC.,
`Plaintiff,
`
`v.
`
`INNOPHARMA, INC.,
`Defendant.
`ALLERGAN, INC.,
`Plaintiff,
`
`v.
`
`FAMY CARE LIMITED,
`Defendant.
`ALLERGAN, INC.,
`Plaintiff,
`
`v.
`
`TWI PHARMACEUTICALS, INC. AND
`TWI PHARMACEUTICALS USA, INC.
`Defendant.
`
`
`
`Civil Action No. 2:15-cv-1455-WCB
`LEAD
`JURY TRIAL DEMANDED
`
`Civil Action No. 2:15-cv-1504-WCB
`
`Civil Action No. 2:16-cv-0401-WCB
`
`Civil Action No. 2:16-cv-0820-WCB
`
`SUPPLEMENTAL RESPONSIVE DECLARATION OF ERNING XIA, Ph.D.
`
`

`

`Case 2:15-cv-01455-WCB Document 190-1 Filed 10/26/16 Page 3 of 13 PageID #: 7095
`
`
`
`I.
`
`1.
`
`Introduction
`
`
`I, Erning Xia, Ph.D., submit this supplemental declaration on behalf of
`
`Defendants Akorn, Inc. (“Akorn”), Teva Pharmaceuticals USA, Inc. (“Teva”), Mylan
`
`Pharmaceuticals Inc., and Mylan Inc. (collectively, “Mylan”), InnoPharma, Inc. (“InnoPharma”),
`
`Famy Care Limited (“Famy Care”), and Twi Pharmaceuticals, Inc. and Twi Pharmaceuticals
`
`USA, Inc. (“Twi”) (collectively, “Defendants”) in the above-captioned actions.
`
`2.
`
`I am the same Erning Xia, who submitted an opening declaration on August 5,
`
`2016 (“my Opening Declaration”). I incorporate herein by reference my Opening Declaration
`
`and supporting exhibits.
`
`3.
`
`I have been asked to submit this supplemental responsive declaration to respond
`
`to opinions raised in the declaration of Dr. Thorsteinn Loftsson and Allergan’s Opening
`
`Supplemental Brief submitted on in this case on September 26, 2016. The additional materials
`
`that I have considered beyond those already identified in my Opening Declaration, are found
`
`below.
`
`II.
`
`4.
`
`Response to Dr. Loftsson’s Declaration
`
`A. Meaning of the Numerical Values in the Claims
`
`Dr. Loftsson opines that it is “difficult or impossible to make a drug product with
`
`an ‘exact’ or ‘precise’ amount of each ingredient in that formulation.” Loftsson Decl. ¶22 I
`
`disagree as Dr. Loftsson has ignored that each numerical integer has a certain degree of margin
`
`based on rounding and thus it would not be difficult to make a drug product using the ranges
`
`inherent in a numerical number.1
`
`
`1 See Ex. 1,, Catherine W. Johnson et al., Essential Laboratory Mathematics (2d ed. 2003) ("Numbers
`obtained from a measurement are approximate values. There is always some uncertainty due to the
`- 1 -
`
`

`

`Case 2:15-cv-01455-WCB Document 190-1 Filed 10/26/16 Page 4 of 13 PageID #: 7096
`
`5.
`
`For instance, a person of ordinary skill in the art would understand that the
`
`ordinary number defined with two decimal points such as “1.25”% means a weight amount
`
`between 1.245% to 1.254% based on the reasoning that numbers in this range would be rounded
`
`to 1.25%. Similarly, 0.05% may encompass an amount between 0.045% to 0.054% based on
`
`conventional scientific rounding.
`
`6.
`
`Contrary to Dr. Loftsson’s opinions, it would not have been “impossible” to
`
`manufacture a drug with ingredients having a given numerical value since that number will
`
`inherently have a range based on rounding. For example, even if one of ordinary skill in the art
`
`would not have been able to consistently manufacture a drug with 1.250% w/w of a particular
`
`ingredient, such individuals would have been able to manufacture a drug at an amount within a
`
`range of 1.245 to 1.254 % w/w allowed by the ordinary value of the numbers itself.
`
`7.
`
`Although a person of ordinary skill in the art would typically define numbers by
`
`rounding off, I understand that an exception to this rule applies in cases where the patentee has
`
`itself limited the interpretation ordinarily given to the number. As discussed in my first
`
`declaration, during prosecution of the Patents-in-Suit, the patentees argued criticality of the
`
`weight percentages and made other statement which created utter confusion as to the scope of the
`
`disputed terms.
`
`B. The FDA Tolerances Are Not Relevant to the Meaning of the Claim Terms
`
`
`
`limitations of the measuring devices used and the skill of the individual making the measurement. The
`figures used to report a result should reflect the precision of the test and the sensitivity of the
`measuring device that produced the value. To express this precision, the number should contain all the
`digits that are known plus one digit that is estimated. These are the significant figures (or significant
`digits). For instance, a measurement described by the number 2.54 mL has an actual value
`greater than or equal to 2.535 mL but less than or equal to 2.544 mL. By simply writing 2.54 mL,
`we indicate our uncertainty about exactly where in that range the measurement falls.") (emphasis
`added)
`
`- 2 -
`
`

`

`Case 2:15-cv-01455-WCB Document 190-1 Filed 10/26/16 Page 5 of 13 PageID #: 7097
`
`8.
`
`Dr. Loftsson opines that a person of ordinary skill in the art would have looked to
`
`manufacturing tolerances and shelf-life specification for guidance on the appropriate limits for
`
`“about” in the Patents-in-Suit. Loftsson Decl. at ¶22. I disagree.
`
`9.
`
`First, one of ordinary skill in the art would not have understood that there is a
`
`universal rule to receive a 5 % or 10% “add on” to any claim term having a numerical value in
`
`addition to the range already given to the ordinary value of the number based on rounding. For
`
`instance, a person of ordinary skill in the art would not have understood the claim term 0.05% by
`
`weight cyclosporin to encompass the range provided by rounding (0.045 to 0.054) as discussed
`
`above and ± 10% of that range (0.0405 to 0.0594) as Dr. Loftsson requires. There is simply
`
`nothing in the record which would lead a person of ordinary skill in the art to such a conclusion.
`
`10.
`
`Second, I disagree with Dr. Loftsson that a tolerance of ±5% manufacturing and
`
`±10% over the shelf-life of a product is “standard” in the pharmaceutical industry. Loftsson
`
`Decl. ¶27. Indeed, nothing in the ICH Guidelines (Ex. C of Loftsson’s Declaration) or the Q6A
`
`Specification (Ex. D of Loftsson’s Declaration) cited by Dr. Loftsson requires a ±5% or ±10%
`
`tolerance for all drugs. To the contrary, specifications are quality standards proposed by the
`
`manufacturer which must be approved by the FDA as conditions of approval. Ex. C to
`
`Loftsson’s Declaration at AGN_RES1085590. In fact, the ICH Guidelines provide decision trees
`
`to determine the acceptable criteria for a given drug substance. Therefore, the release
`
`specifications may differ for a given product.
`
`11.
`
`Notably, the ICH guidelines and Q6A specification cited by Dr. Loftsson relate to
`
`“solid oral products, liquid oral drug products, and parentals (small and large volume).” See e.g.,
`
`Ex. C at AGN_Res1085599. Ophthalmic emulsions, such as Restasis, are not required to strictly
`
`follow the ICH guidelines.
`
`- 3 -
`
`

`

`Case 2:15-cv-01455-WCB Document 190-1 Filed 10/26/16 Page 6 of 13 PageID #: 7098
`
`12.
`
`Dr. Loftsson also cites to the New Drug Application (“NDA”) for Restasis® as
`
`support for the ±5 or ±10% tolerances for cyclosporin. Loftsson Decl. ¶¶25-26. However, the
`
`NDA is a highly confidential document, which is not available to the public at the time the
`
`Patent-in-Suit were filed. Therefore, a person of ordinary skill in the art would not have
`
`considered such documents in determining the meaning of the numerical claim terms.
`
`13.
`
`Even less convincing, Dr. Loftsson cites to an FDA Guidance (Ex. G to Dr.
`
`Loftsson’s Declaration) for support that products which are Q1/Q2 to one another are allowed to
`
`be 5% different in both active and inactive ingredients. However, the FDA Guidance cited by
`
`Dr. Loftsson relates to nasal aerosols, not ophthalmic emulsions. Moreover, the draft guidances
`
`refer only to the percentage of inactive ingredients, not active ingredients. Ex. G at
`
`AGN_RES1085682. The percent tolerability provided by the FDA guidance for nasal aerosols
`
`for such inactive ingredients is not designed for RLD production, rather it was designed for
`
`Q1/Q2 or reverse engineering evaluation of RLD at generic companies. In my opinion, the FDA
`
`guidance cited by Dr. Loftsson is completely irrelevant to the meaning of the disputed terms.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`- 4 -
`
`

`

`Case 2:15-cv-01455-WCB Document 190-1 Filed 10/26/16 Page 7 of 13 PageID #: 7099
`
` I
`
` declare under penalty of perjury that to the best of my knowledge the foregoing is true
`
`and correct.
`
`
`
`Date: October 26, 2016
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`__________________________
`
`Dr. Erning Xia
`
`- 5 -
`
`

`

`Case 2:15-cv-01455-WCB Document 190-1 Filed 10/26/16 Page 8 of 13 PageID #: 7100
`
`Exhibit 1
`
`

`

`Case 2:15-cv-01455-WCB Document 190-1 Filed 10/26/16 Page 9 of 13 PageID #: 7101
`
`
`
`

`

`r
`
`-I
`
`
`
`Case 2:15-cv-01455-WCB Document 190-1 Filed 10/26/16 Page 10 of 13 PageID #: 7102
`
`•
`IC
`Concepts and Applications
`fo:r the Chemical and Clinical
`Laboratory Technician,
`Second Edition
`
`Catherina W. Johnson, MAediMath
`Alamance Community College
`Graham,. NC
`
`Daniel L Timmous, MSed/Physies
`Alamance Community College
`Graham, NC
`
`Pamela E., Hall. MA, MT(ASCP), SBB
`Alamance Community College
`Graham, NC
`
`

`

`Case 2:15-cv-01455-WCB Document 190-1 Filed 10/26/16 Page 11 of 13 PageID #: 7103
`
`THOMSON . ...
`
`DELMAR LEARNING
`
`&w11tial LaboottoJy M~ Concepts and
`Appticlltitms ft1t thf. Chemield and Oinka! ·~ Tocbnl~t.m. Soc()nd Hdtllon
`by t"Mherine w. JObllSOt.t.. 1>1\nte~ L Tirnmom:, and Pamela a Hail
`F.ucutt'\ ... >trutor. Hesltb ea,. BasiMI:5lll Unit
`WUJiam BroumiJicr
`Ex:Kutit~ •:.tltor:
`C411h)' l ... R<tpmi
`Al'qtdfitJont F.dittW:
`M~ Rt;K~net'
`
`YtorW~•~
`Matthew Thouin
`EuaJUw M.tlrldio& MllftiiiC'J':
`O.wn F. 0cmUn
`Chtmd&~
`Jtmifcr McAJey
`Project Editor;
`Mary EJitn Cox
`
`~~nt.lf!AUtor:
`MarahBtU~
`
`rroductloa Coordl.Wor:
`Anne Shennan
`Art/Dnlgn CooniiNitor:
`JnyPu~tU
`
`COPYRICJH'f C 2())3 b)' ll¢!mat t~l'lling:. a dil>1sioo nf~ ~ng.
`Inc. lbonw.m Lecamlngrn 15 a
`m;<knwl u.1ed hm:in under lletn".
`Print!Jd in Calllldil
`l4SXXX06(}504
`For n~ Information;. c•mtut:t tklmu Lc~ing.
`S M111wt~U Dt. PO.Bo11 8007
`Clifton P..ut. NY l206S-tl007
`Or find us 00 the WOOd Wide Web ut .bltp:J/W'II,'W .dclmaf,CQID
`ALL RIGiffS RESERVED. No p;ut of tlus wt)d; covi!I'Od by the copyriJbt
`~ my oo rqmldoo.t!d or u.~d tn an)' form« by any
`IM'llM-~.ehl!cuunk, or trutebankal, including ~opying.
`~ina. tl!pir~g. Web distri~•ron c;r mf~tioo stmp
`and rv~rit"'~l $)'~1-witl:w;nn wrh~n pcnniy,lioo of~ pu.bli~.
`
`fat pennluion II)'* ~~ from tb:i$ ttxt tlt produe~. wntar.:t us by
`Td (800) 7JO.llJ4
`Fu (800) 730-211 s
`www.thQmwnriab~.eom
`UllnrJ ol COCIJI'•Cat.tdooafq.U..Pu~ Dati
`JohMtm. Catherine W .. MAed!Math.
`EucmW labonulf)' maltletnatia: com:epcs lind. ~pplieation:J for tbt
`chl:mkul a.nd ~Unlcallaboratory tedmk:ian I Catherine W. Jaftnson,
`~ll .. nm~mJ. p;tmdi E. Halt-2nd ro.
`p.;cm.
`lnd\Jib bibfi~bl ~and i.ndcx.
`ISBN 0· 7668-3Jn6.9 (alt. ~.)
`l. Mc:dicaiiiOOI'atory flldmok!Jy..-M.itbematk:s.
`(DNLM: t. M~ l. Cli!J.ical ubora~nry T«tmiques. l.
`Tecfmol(:}Sy. M«<bl. QT 35 166¢ 200,'\J r. Timmoo5, O.niell., Jl H#lt
`Pa.mdA B. tiL Tttk.
`RB3&3 J647 2003
`610' .284..<k:21
`
`NOTJ.C£ 'TOTR£ IUADF.lt
`Tbe publl*.- doeli nm wm-m11t t)lf ~ntee any uf the product!S dell."ribed hm:in or perform any i~ analysis in ~tlon with an)' m· the produd
`infonnation ~~ained herein. "J'be publisher does not usumt~ and Uprt~UI)! disclaims. any obliptiwllo obtain and ini:lude infOI"'''iation· other tiWI dlat pro.
`'~<'Kkd to it by the manuf3Citl~t.
`The ~r •~ exprc""'Y wmtl.td 10 Cffll~ and lldopc i.llliafdy pm:tWtiml 11W might be lndkat!Jd by the Klh•ltiti bmin tnd to awoid aU potMtial tmutd:i.
`By fullowintc tbt lmtnn:UonK c.onmilwd hmin, tbe reader wUli!til)' .Uume'il an risb tn (<()ftfiCC'tiOO w.ltb ·me'b i.nsii'UetionJ.
`The pubUwr make• no repreuntation or W1\m:ltl~11 of an)' kind, ine!wting but 004 limited to, tho·~~ of fi~s (or pmk."'Ular ~or mcr(cid:173)
`~abilicy. !Wf are !MY such r~~m .. im'ls hnpJ\fd witl! rupect 10 tho~ ~ fomt herein. 1M the pvbli*t ~no ~J>OO$ibillly with mpca w
`sw:b mtm-riat T'be publiwr s:haU not be liable fm any Sf.!CiCW. CllMequential, or cx~.mplary damages mulling. in whole Qf part, from the ~ use of. Of
`rcUanr:c up:m, this nwmid.
`
`

`

`Case 2:15-cv-01455-WCB Document 190-1 Filed 10/26/16 Page 12 of 13 PageID #: 7104
`
`Margin Problems
`How many significant figures?
`1. 2.62m
`2. 0.00268L
`3. 2.3600dL
`4. 0.036050 s
`
`2.3 Significant Figures
`
`Numbers obtained from a mcatiurement are approximate values. There is at~
`ways some uncenainty due to tbe limitations of lhe measuring devices used and
`the skill of the indh•idual making the measurement. The figures used to report a
`result should reftect the precision of the test and the sensitivity of the measuring
`device that produced the value. To expre.'ls this precision. the number should
`contain aU the digits that are known plus one digit that is estimated. These are
`the signlftamt tlgures (or significant digits).
`For instance. a me&urement described by the number 2.54 mL has an actual
`value greater than or equal to 2.535 ml .. but less than or equal to 2.544 ml ... By
`simply writing 2.54 mL. we i.ndicate our uncertainty about exac..*tly where in that
`range tbe tnc3Suretnent falls. Our mea,o;urement is accurnrc to 3 si.gnificant figures.
`Some nu.mbers arc exact and occu.r in sin1ple counting operations. For exam·
`pJe. you can count exactly 25 oranges. Defined numbers such as I 2 inches =
`I foot are also considered to be exact. Significant figures are used for me.asured
`values, not counted or exact values.
`
`Rules for Determining Significant Figures
`t. In any measurement. aU nonuro numbers are significant.
`2. Zeroes may or may not be significant as follows:
`a. Zeroes between nonzero digits are significant.
`7,005 has four significant figures
`6.0035 ha.~ five significant figures
`b. Zeroes at the end of a number that include a decimal poinl are
`significant.
`13.500 bas five significant figures
`41.0 ha.~ three significant figures
`60. has two significant figures
`c. Zeroes at the end of a number without a decimal are considered
`ambiguous and are not signHicruu.
`S.OOO bas one significant figure. the 5
`350 bas two significant figures
`d. Zeroes th.at precede the first nonzero digit are twt significant.
`These zeroes are u.~d to locate l.be dedrnal point.
`0.0083 ha.\ two significant figures. the 8 and 3
`0.0103 has three signifi<.-ant figures. the 1. the 3, and rhe 0 between
`the 1 and 3
`
`Answen
`
`'· 3
`2. 3
`3. 6
`4.6
`
`In the event that a whole number contains zeroes that are significant, a bar as
`placed over the rightmost significant zero. For example, for the two significant
`fig.ures. the number 5,000 is written as 50oo. The numbers to the right of the 0
`are not significant. The number so.oOo bas 4 significant figures.
`In c,alculations we often obtain answers that have more digits than we are
`justified in using. When digits are dropped from a number, the value of the last
`digit retained is determined by a proce.iiS known as rounding o,ffm1mbers. This
`process is discussed in Chapter 1. so we will just review here the rules that will
`be used throughouc the book.
`
`

`

`Case 2:15-cv-01455-WCB Document 190-1 Filed 10/26/16 Page 13 of 13 PageID #: 7105
`
`Rules: for Roundtng Off Numbers
`Rule l: When the first digit to the right of the one you want to retai.n is 4
`or less. that digit and all others to its right are dn:.pped. The last digit re·
`tained is not changed.
`For example:
`74.593 rounds off to 74.59 (the 3 is dropped)
`3.00249 rounds off to 3.002 (the 4 and 9 are both dropped)
`Rule %: When the first digit to the right of the one you want to retain is 5
`or greater. that digi.t and all others to the right of it are dropped and the
`last digit retained is increased by 1.
`For example;
`1.0268 rounds offto L027 (the 8 is dropped and the 6 is increased to a 7)
`
`Numbers that are the result of a measurement are rounded to the correct
`number of significant figures using these rules for rounding. Exact numbers do
`not need to be rounded to significant figures since they are not approxima.te.
`The results of a calculation based on measurements cannot be more· precise
`than the measu.rements used t.o achieve that result. Because of the arirhmetic
`rules for the placement of decimals in multiplication and division problems, an~
`swers may seem to be more precise than they actually are. Therefore. in calcuJa~
`ti.ons in .. ·otving the multiplication or division of measured amounts. the answer
`should contain the same number of significant figures as in the measurement
`that has the least number of significant figures. For example. the product shown
`on a calculator when multiplying 2.54 X 3.213 is 8.16102. However. the most
`accurate value we can give for the answer is 8J6 because tbe least precise
`factor (2.54) ha.~ only three significant figures ..
`
`Rule for Multiplying or Dividing Measurements
`When multiplying or dividing measurements that have different numbers
`of digits. round the answer so that it ba." the same nuJnbi~r
`measurement with the least number of significant
`
`Example 1: Multiply the following~ assuming that they all represent mea·
`sured amounts. Round the answer to the correct number of
`significant figures.
`
`(0.0211)(2.53)(13.82)
`1,200
`
`The calculated result is 0.()00614794.
`Examine the individual f'acrors to determine rounding:
`0.02 J 1 has tbree significant figures
`2.53 has three significant figures
`13.82 bas four significant flgures
`1.200 has two srgniflcant figures
`The factor 1.200 contains the least number of significant
`figures (two) and therefore the calculated answer is rounded.
`foUowing the previously listed rules. to two signiflcan.t figures.
`The correctly rounded answer is 0.0006·1 .
`
`