11
`
`LOWER DRUG PRICES FOR CONSUMERS, LLC
`Exhibit 1021-1
`IPR2016-00379
`
`

`
`~·
`
`-~
`
`~ r j
`-1 .
`
`I 1--
`
`........
`l
`I
`l I
`i
`r j
`
`~
`+ !
`
`I""
`I
`
`'.,
`
`_J
`
`LOWER DRUG PRICES FOR CONSUMERS, LLC
`Exhibit 1021-2
`IPR2016-00379
`
`

`
`INTRODUCTION
`This booklet has been prepared tor the users of CYCLO(cid:173)
`BOND columns to aid them in developing and optimizing
`separations of their compounds. Since CYCLOBOND col(cid:173)
`umns perform their separation via a new and unique
`primary mechanism, and a handful of secondary mech-
`. anisms, these will be briefly described so they might be
`invoked to get the most out of any CYCLOBOND column.
`Varying the conditions of the separation to aid in the op(cid:173)
`timization will be discussed as will the care of the column
`to ensure its long and reproducible use. Please take time
`to familiarize yourself with the use of the columns by
`reading this text thoroughly. By doing so you will get bet(cid:173)
`ter and faster results. If you still have questions, do not
`hesitate to call us to obtain help.
`
`CYCLOBOND COLUMNS
`CYCLOBOND columns are packed with unique stationary'
`phases of chemically bonded cyclodextrtns to a high
`purity, 5 inicron, spherical silica gel. The cyclodextrins are
`linked through a -CH2- chain of optimum length via a
`patented process which yields a stable non-hydrolytic,
`non-nitrogen containing bond. Five CYCLOBOND col(cid:173)
`umns are now avialable called CYCLOBOND L CYCLO(cid:173)
`BOND I ACETYLATED, CYCLOBOND IL CYCLOBOND ill and
`CYCLOBOND III ACETYLATED.
`CYCLOBOND I is bonded with beta-cyclodextrin (f3-CD)
`and from available separations performed thus far, ap(cid:173)
`pears to be the most versatile of the products. The 8-CD is
`a macrocyclic molecule that contains 7 glucopyranose
`units arranged in the shape of a hollow truncated cone
`in which the interi.or cavity is relatively hydrophobic
`being comprised of essentially methylene and 1,4-gluco- ·
`side linkages producing a high electron density for exten-
`. sive interaction with analyte. The exterior faces, on the
`other hand, are hydrophilic. The large;r opening is sur(cid:173)
`rounded by the secondary hydroxyls. The 2-hydroxyls are
`rotated clockwise and the 3-hydroxyls are rotated coun(cid:173)
`. terclockwise. The primary hydroxyls constitute the smaller
`end of the cone. The tunctional structure and dimensions
`of beta-cyclodextrin are given in Figure 1.
`
`-I-
`
`CYCLOBOND I ACETYLATED is formed by acetylating the
`fixed secondary hydroxyls on the rim of the $-CD cavity.
`This has an effect of changing the size of the cavity andh
`changing the interaction between the hydrogen bonding'-·
`site of the cyclodextrin and the specific functional group
`'
`on the asymmetric carbon. Figure 2 shows the separation
`of dJ-Norphenylephrine on this column. The acetyl group
`has shown a high degree of selectivity for geometric
`separations based on the position of double bonds.
`
`Figure 2
`Sepaxation oJ d,l-Nozphenyleph.rille
`
`ANA!Yl'.ES'
`l. 1-Ncrphenylephrine
`2. d-Norphenylephiine
`CONDIDONS,
`COLUMN,
`SIZE•
`MOBil.E PHASE:
`FLOW RATE,
`PRESSURE,
`CHART SPEED,
`DE'IECTION,
`CONCENT.R.l>JION1
`
`CYCLOBOND I ACETYLATED
`250x4.-6mm
`0.5% Acetic Acid. pH 6.1
`0.5 mlhnin
`600 psi
`0.5 cm/min
`254 nm
`l.Oµg/j.il
`
`CYCLOBOND II is the 'Y-CD form consisting of 8
`glucopyranose units arranged in the same truncated
`cone shape·and is useful for isomeric compounds based
`on anthracene, chrysene and pyrene ring structures.
`CYCLOBOND ill is the .a-CD form c;;onsisting of 6 gluco(cid:173)
`pyranose units, also a truncated shape with a smaller in-
`temal diameter. The smaller diameter of the CYCLOBOND
`ID allows it to be most useful tor molecules smaller than
`benzerle, many underivatized cirnino acids, prostagland(cid:173)
`ins., and inorganic ions. CYCLOBOND III ACETYLATED
`bridges the cavity sizes between fl-CD and S-CD, and in
`addition offers specific functional group interaction as in '
`CYCLOBOND I ACETYL.ATED.
`
`,
`
`,
`
`LOWER DRUG PRICES FOR CONSUMERS, LLC
`Exhibit 1021-3
`IPR2016-00379
`
`

`
`In Figure 3 the decreased retention observed for chrysene,
`phenanthrene and anthracene on CYCLOBOND II is
`thought to be the result of less interaction due to the linear
`nature of these aromatic structures. The more bulky
`pyrene shows expected behavior. Further, due tci the
`bulky nature of Y-CD, less is bonded to the silica surface.
`The naphthalene structure enters the gamma cavity hor(cid:173)
`izontally and the beta cavity vertically so the total bon(cid:173)
`ding energy is equivalent The slight dif1erence in reten(cid:173)
`tion is the result oi diminishecty:.cn on the CYCLOBOND II.
`It should be noted that not all of the solute structure has
`to fit into the cyclodextrtn cavity for the separation to oc-
`cur, but at least a portion of it must fit well enough to form
`a stable complex.
`
`.
`
`f
`
`.
`.;
`
`Figure 4 shows o-, m-, and p-nitroanilines separated on
`each of the CYCLOBOND I, II and III columns with
`MeOH/H201 40/60. These separations of structural isomers
`are based primarily on the fit ot these molecules into the
`CD cavity. The same elution order of m-, o-, and p- isomers
`results regardless of the substituents. With CYCLOBOND II
`
`I
`1
`
`POSSIBLE MECHANISMS
`WITH CYCLOBOND COLUMNS
`~structures of these cyclodextrtn bonded columns have
`.an discussed since tfiey constitute the basis on which
`the separation mechanism operates. The selectivity of
`each of these phases is very different and as with any
`·liquid chromatographic separation, the exact mechan(cid:173)
`ism changes as a result of the structure of the solute, their
`solubility, the mobile phases, and all of these in relation(cid:173)
`ship to the phase in the column. These columns are par(cid:173)
`ticulmly effective for: molecules having a hydrophilic por(cid:173)
`tion which can tit the cyclodextrin cavity reasonably well.
`Because of the unusual shape of the cyclodextrin
`molecules (whether ·a, f3, or Y) and the different polarities
`of the various surfaces, ditterent interactions with solutes
`· are possible. One of the first observed phenomenon with
`cyclodextrins was their ability to form inclusion com- ·
`plexes. In order for a solute to be able to do this it must
`meet certain requirements. It must have a hydrophobic
`segment that will fit snugly into the CD cavity. The com(cid:173)
`plex can also be more stable by having polar groups that
`hydrogen bond to the secondary hydroxyls at the edge
`of the cone. Structural dillerences greatly effect retention
`of the analyte. It for instance, oxygen atoms are in the
`structure of an analyte, retention is decreased (see ref. 18),
`due to electrostatic repulsion. The position of the oxygen
`can also effect the retention of the analyte. Increasing the
`number ot methylene groups enhances retention so that
`4, 5, 6, 8, etc. carbon rings can easily be separated.
`Chlorine atoms have a high affinity for the cavity and
`their position and number greatly effect retention.
`Tf the diam~ter of the solute is too large,· no complex can
`'.ll because the hydrophobic segment cannot enter into
`-+e mouth of the CD cavity. If too small, the solute can
`enter, but the complex is not as stable (smaller formation
`constant). As general rules, single ring compounds such
`as benzene or smaller, tit well into a-CD (CYCLOBOND ill);
`compounds such· as substituted benzenes, naphthalene
`or biphenyl derivatives tit well into S-CD (CYCLOBOND I);
`and, larger compounds such as pyrene fit well into y:.cn
`(CYCLOBOND II). Figure 3 illustrates the retention of these
`compounds on the various CYCLOBOND columns.
`
`CONDmONS,
`CYCLOBOND m. I and n
`COLUMN,
`250x4.6mm
`SJZE,
`MeOHIH,O, 40160
`MOBILE PHASE,
`1.0 ro.\knln
`FWW RATE,
`CHART SP!0£0,
`1.0 cmhn!n
`264 run
`OETEcnON.
`CONCENTRATION. 0.5)Jlg1W
`
`-2-
`
`LOWER DRUG PRICES FOR CONSUMERS, LLC
`Exhibit 1021-4
`IPR2016-00379
`
`

`
`(c), the Y-CD cavity is too large to allow sufficient interac(cid:173)
`tion of the aromatic ring vvith the internal surfaces of the
`CD so little difference is seen between ortho and meta
`isom·ers. With CYCLOBOND I (b) the best overall resolution
`· is seen because there is good interaction of the aromatic
`ring with the CD internal surface and the p- isomer fits very
`well into the 13-CD cavity allowing for stable hydrogen
`· bonding of the nitro group with the primary hydroxyls,
`
`' .....
`
`thereby increasing its retention. Them- isomer is too bulky
`for a good fit. hence the earliest resolution. This is shown
`diagrammatically in Figure 5. With CYCLOBOND ill'(a),
`longer retention times are observed in this_ case due to
`tighter fit for that portion of the aromatic group interactinr
`with the cavity.
`
`I
`I
`I
`I r
`
`- - - - Hydrogen bondlng
`• • •· Electrostatic association
`X
`Hydrophobic groups that can penetrate cavity. ier -N03• -Cl. -OH
`Polar groups that can hydrogen bond -coon. -NH2
`Y
`It is also possible to operate the CYCLOBOND column as
`a reversed phase column using only the nonpolar
`character of the internal surface of the CD and the
`solvated surfaces of the entire structure. It can be con(cid:173)
`sidered a Iather unique reversed phase. very diiferent
`from anything in the LC market today. When attempting
`new separations via reversed phase, a CYCLOBOND col(cid:173)
`umn should also be tried. Likewise, because of the
`
`prevalence of the primary and secondary hydroxyl
`groups, it can be used as a high density, very efficient diol
`column. These latter types of columns have found wide
`use in the separation of proteins and peptides. CYCLO(cid:173)
`BOND columns can be used as all purpose columns as
`seen in the typical reversed phase separations shown j
`Figure 6A and 6B.
`
`l
`
`CONDmONS.
`COLUMN,
`SJZE.
`MOBll.EPHASE,
`FJ..OV'J RA'.l'Eo
`
`CYCLOBONDl
`100x4.6mm
`CH,CNJ0.1% IT.AA pH 4.0, 15/!15
`l.Omlitron
`
`CONDITTONS,
`COUJlvlH,
`SJZE,
`MOB!l£ PHASE,
`FLOW RATE,
`
`CYCLOBONPI
`250x4.6mm
`CH,OHJH20. Jono
`J.OmLnlln
`
`-3-
`
`.I
`·I
`
`LOWER DRUG PRICES FOR CONSUMERS, LLC
`Exhibit 1021-5
`IPR2016-00379
`
`

`
`MOBILE PHASE DESIGN
`Since inclusion complexation is encouraged in CD bond(cid:173)
`ed phases by water and waterbrganic mobile phases,
`these are most often used in CYCLOBOND separations.
`"lee there is competition between the solutes and the
`----:1.ganic component ot the mobile phase for the CD cavi(cid:173)
`ty, the selectivity of the organic solvent used becomes im-
`. portant. Methanol is the lowest solvent strength organic
`modifier, followed by other alcohols whose solvent
`strength increases as the chain length increases. ke(cid:173)
`tonitrile has approximately tour times the eluting power
`of methanol. Other organic solvents like DMF, THF and
`chloroform can and have been used to obtain useful sep(cid:173)
`arations. Little is known of the mechanism here except
`these types of solventS have strong affinity for the cavity
`and the mechanism is more likely partition rather than in(cid:173)
`clusion complexing.
`Because acetonitrile has a much greater affinity for the ,
`CD cavity, less is required to obtaill comparable retentions
`to water/methanol mixtures. As with many LC separations.
`there is no way of telling which modifier will give the best
`separation, but the choice is usually to try methanol first.
`To decrease retention or increase efficiency or selectivi(cid:173)
`ty, acetonitrile may be substituted in whole or in part. The
`dramatic effect of such a substitution can be seen in
`Figure 7 in the separation of doxepin isomers.
`
`ANA1.YT£51
`1. Z~Doxeptn
`2. l>Doxeptr.
`
`CONl>mONS,
`COw:viN.
`SIZE,
`MO.BILEPliAS!!,
`fl.OW RA'll:'.·
`
`C'lCLOBONDI
`250x4.6mm
`McOl!llt,O, .ia/60
`I.~ mLrni.'"\
`
`~: i:~it~nCY.CPCd
`.) :Xr~nom-
`•· %-OoAO!k.
`5 .F , (cid:173)
`o.~elhYICo:x"l'pio
`7. E-Pe<!Mtl!yl llOxepll\
`
`COrnxtlONS,
`~~·
`M::nm..£PHAS£,
`F".Dl'JIW£
`
`A more subtle effect of the differences in solvent selectivi(cid:173)
`ty is seen in Figures 8A and 8B showing a comparison of
`the separation of dansyl-D,L-leucine and dansyl-D,L(cid:173)
`phenylalanine on 1 o cm CYCLOBOND I columns. It can
`be seen that each performs the separation but the
`methanol gives better separation be1ween the enantio(cid:173)
`meric pairs while acetonitrile gives better efficiency.
`
`(~i:W~ ;~:~!J:i;k'.~~\"})1~ JHi\;}i;·.~~'.~.~;~:::~~::r.~=':fflit~ -~~~£,~.
`F1.guro BB
`Soparat!r>n <>I I>ansrJ-Dk.Leuc;iJle and
`Dcmsyl-D,L-Pb.e11ylalcmine
`Acetonilzile ~lectinty
`
`CON:\.ilONSi
`~·
`M0811£f'llAS<,
`
`C\'Cl.ODC,.1)1
`lt'O;t;4.0mm
`~~·~.M.(0.J%)pH4.0,
`
`IUJW!Uot£,
`
`C..Sml.tnln
`
`cormrr.oNS.
`~fN·
`MOEJJ.E~
`
`FWW:LJ\'.IE.
`
`In working with CYCLOBOND columns to perform a re(cid:173)
`versed phase separation, not only methanol but increas(cid:173)
`ed alcohol chain length, i.e. EtOH, 1-PrOH, etc., and
`acetonitrile, as well as_ t~trahydroturan can be tried dur(cid:173)
`ing solvent optimizati.Qn. These organic modifiers, some(cid:173)
`times in combination with each other, are the most useful
`for solvating and allowing separations in the reversed
`phase mode ..
`Finally, when attempting a separation using CYCLOBOND
`columns as multi-al, that is, a multi-hydroxyl containing
`bonded phase, a tu1l range of solvent polarities have
`been used, from polar to nonpolar. This, of course,
`depends upon the solubility .of the solutes which need to
`be separated. For proteins, the most often separated
`species in such a column, methanol/water or pro(cid:173)
`panol/w'ater combinations with trifluoacetic acid (fFA) or
`triethylammonium acetate (TEAA) are used. For nonpolar
`species such as carotenes, the mobile phase consists of
`hexaneeth.c;;mol.
`Note that the use of more nonpolar modttl.ers, such as pro(cid:173)
`panol or hexane, can be used on CYCLOBOND columns.
`They, however, perform a different role since they occupy
`the CD cavity so tenaciously. In these instances the solute
`retention is occurring due to the adsorption of the solute
`on: the outside of the cyclode:xtrin cavity by interaction
`with the surface hydroxyls andbr on the solvated surfaces
`of this struchue via a unique partition mechanism.
`
`STARTING MOBILE PHASES
`The beginning mobile phase for inclusion complexing.is
`Methanol/Water: 50/50. After an initial chromatographic
`run, retention is adjusted by adjus!ing this ratio. If the.
`retention is too short, the percent methanol is decreased;
`if it is too long, the percent of methanol is increassd. This
`
`LOWER DRUG PRICES FOR CONSUMERS, LLC
`Exhibit 1021-6
`IPR2016-00379
`
`

`
`is the usual procedure when optimizing a reversed phase
`separation. Make the percent changes in smaller degrees
`than with a reversed phase column since the CYCLO-
`. BOND column is more responsive to these changes.
`
`Gradients can and have been used in CYCLOBOND sep(cid:173)
`arations. It is important, however, to use a slower gradient
`ratio or range since equilibration of the CD cavity takes
`longer than with a comparable C 8 or C 18 reversed phase
`column.
`
`SALT AND pH EFFECTS
`As with many reversed phase separations, the optimiza(cid:173)
`tion of the solvent ratios is only part of the process to a final
`separation of most solute mixtures. It is frequently
`necessary to change the pH or add competing salt mix(cid:173)
`tures to enhance the separation. This has also been found
`to be useful in CYCLOBOND optimization where pH has
`been demonstrated to be one of the most effective'
`separation parameters. By increasing the hydrophobic
`
`character of the analyte, retention time increases. In ad(cid:173)
`dition, pH control can mask the interaction of a functional
`group with the cyclodextrin or can enhance its hydrogen
`bonding ability . .Anionic and cationic species have beep ·
`separated, generally by increasing the alcohol conten
`Also, association effects can be realized using saltto form-(cid:173)
`adducts with spacially favorable isomers thereby de(cid:173)
`creasing the retention of one form over the other. The
`details of all of these techniques cannot be :tully discussed
`in such a limited space ·and the reader is referred to
`papers listed in the reference list at the end 01 tl11s booklet
`especially references 2, 9 and 23.
`
`With the addition of a salt, retention and -resolution can
`improve markedly. The improved resolution is observed
`because of a narrowing of the peak width. Up to a four(cid:173)
`fold increase in efficiency can often be seen. Figure 9
`shows the separation of dansyl-D,L-leucine with and
`without a salt. If present in the mobile phase during
`separatiC?n, the salt occupies a portion of the cavity.
`·
`
`COLUMN,
`SJZE,
`MOBll.El'HASE.
`FLOWRA'.l'E,
`
`CYCLOBONDI
`250x4.6nun
`MeOHJHp, 80120
`1.5 mlknln
`
`CYCLOBONDI
`250x4.6mm
`MeOttfI'E.6.A (0. I%) pH 4.0. 60120
`1.5 m!.hn1n
`
`CYCLO!!OND I
`250x4.6mm
`MeOH/0.02M NH.NO, pH 5.2, 80/20
`1.5 ml1mln
`
`Retention times are usually decreased with salt addition,
`but this is variable depending upon the pH and the salt
`co;n.centration. The effect of vmying the salt concentration ..
`and the salt are seen in Tables I and II.
`·
`
`ex
`Compound
`Na2HP041 M k
`Rs
`Dansyl-D.L-valine
`0
`1:"16 0.4
`3.9
`0.005
`Dansyl-D ,L-valine
`1.20 0.5
`0.6
`Dansyl-D ,L-valine ·
`0.7 1.20 0.6
`0.01
`Dansyl-D,L-valine
`0.02
`1.4 1.17
`1.1
`1.2
`Dansyl-D ,L-y-aline
`1.18
`1.1
`0.03
`The mobile phase for all separations was MeOH/vvater
`(25: 75), I ml/min. The capacity factor, k, is for the first
`. eluting L-enantiomer.
`
`-5-
`
`Salt
`
`e
`ex
`Rs
`1.7
`1.3
`LlC104
`1.16
`1.17
`1.8
`NaC104
`1.4
`1.16
`2.2
`KC104
`1.3
`l.5
`2.4
`Cs Br
`1.16
`2.4
`NH4Cl
`1.16
`1.5
`1.2
`1.7
`1.15
`Ca(N'03)2
`2.3
`1.14
`1.4
`Cd(N03)2
`The mobile phase for all separations was MeOH/vvater
`(25: 75), 1 mlh:nin. The capacity factor, k. is for the
`first eluting L-enantiomer. The salt concentration was
`0.02M.
`
`'1
`
`i
`ii
`
`II
`
`1
`'.I
`
`.. I
`I
`
`ii
`'
`
`LOWER DRUG PRICES FOR CONSUMERS, LLC
`Exhibit 1021-7
`IPR2016-00379
`
`

`
`As seen from these tables, as little as 0.005 molar smt can
`affect the retention and resolution to a marked degree.
`The usual range of salt concentration used with CYCLO(cid:173)
`BOND columns is from 0.005 to 0.1 molar. Triethylam-
`1onium acetate buffer is made up by taking a O .1 to 1 %
`_,..;ilutlon of triethylamine and adjusting to pH 4.1 (or other
`desired pH) with glacial acetic acid. Its concentration is
`. more easily described in terms of the % solution.
`Other buffers that can be used with the CYCLOBOND col(cid:173)
`umns are trifluroacetic acid and ammonium acetate. If
`the salts from which these buffers are made are not of the
`highest quality, which is often the case, it is strongly
`recommended that they be made up, and filtered
`througp. a 0.45 µm membrane. This should be followed by
`paSSing them through a spare C 18 column to further
`remove organic impurities which would otherwise collect
`on a CYCLOBOND column. The collected impurities can
`be washed from the spare Cl 8 column with 25 - 50 ml of_
`100% methanol. It is necessary to wash buffer from the col(cid:173)
`umn when not in use. Storing a column in buffer will
`diminish its life.
`As can be seen from the information given, the stability
`of the cyclodextrin complexes is also dependent on the
`charge of the guest molecule, hence the large effect with
`the addition of the salt, which concurrently changes this
`charge. In general, the binding strength of a charged
`species·is smaller than with its neutral counterpart. This is
`presumably because of the decreased hydrophobic in(cid:173)
`teractions between the solute molecule and the nonpolar
`cyclodextrin cavity. Therefore, the retention and selectivi(cid:173)
`ty of ionizable species can be effected strongly by alter(cid:173)
`ing the pH at which they are separated on a CYCLOBOND
`column. A strategy of noting the pKa differences of sam-
`ule components, adjusting to neutralize a key species to
`:lhance its retention and rn.inim.1ze the retention ot the
`-<>ther species in the sample. has proven to be very
`successful.
`TEMPERATURE EFFECTS
`As with all complexes,· there is a
`temperature
`dependence on the binding constants. With cyclodextrin
`complexes, there is a much stronger binding at lower
`temperatures. Table ill shows an increase of both reten(cid:173)
`tion and selectivity with decreasing temperature resulting
`
`in increased resolution. Mass transfer and band broaden(cid:173)
`ing can become problems because of the increasing viS- ·
`cosity of the mobile pha?e. To minimize viscosity and
`band broadening problems which are often critical in
`methanol/water combinations, temperature effects are
`usually conducted "\Ali.th acetonitrile/water combinations
`when lowering the temperature.
`
`;!!&••~1~~~~·
`
`k
`Temp, C
`Compound
`Rs
`a
`2.4 1.16 1.5
`23
`Dansyl-D,L-valine
`3. 7
`1.23 1 . 7
`4
`Dansyl-D ,L-valine
`The mobile phase for all separations was MeOH/O. 02 M
`NH4Cl (40:60), 0.5 ml/min. The capacity factor, k, is for
`the first eluting L-enantiomer.
`'
`
`l.
`I
`}
`
`If on the other hand it is found that a very strongly form(cid:173)
`ed complex results in extremely long retention times, then
`the temperature of the CYCLOBOND column can be
`elevated to weaken the complex. Most complexes are
`broken between 60° and 90°C.
`
`FLOW RATE EFFECTS AND PRESSURE
`The usual flow rates employed when operating a CYCLO(cid:173)
`BOND column are 0.5 to 2.0 ml/min. A 250 x 4.6 mm
`CYCLOBOND column "With 5 µm spherical packing will
`exhibit about 2000 psi when operated at 1 ml/min with
`methanol/water (40:60). If the back pressure seems to be
`increasing it probably indicates that particulate matter is
`collecting on the inlet frit because of poorly filtered
`samples or mobile phases. Change frit or reverse column.
`
`Without buffers, many inclusion complex formation
`separations will show increased peak broadening be(cid:173)
`cause of poor mass transfer involved in the mechanism
`when running at taster flow rates of 1 to 2 ml/min.
`Whenever-increased efficiency is required for a separa(cid:173)
`tion, decreasing the flow rate by one quarter will approx(cid:173)
`imately double the efficiency. This increase is seen in
`Figure 10.
`
`COl.Ulv!N,
`SIZE,
`MOBll.E PHASE.
`
`CYCLOEOND I
`250x4.6mm
`Me0HI0.02M Nli,NOJ,80120
`
`LOWER DRUG PRICES FOR CONSUMERS, LLC
`Exhibit 1021-8
`IPR2016-00379
`
`

`
`Remember also to adjust the chart speeGl so the separa(cid:173)
`tion will not look worse since it will be taking longer for
`the sample to elute. Except tor instances when doing mul(cid:173)
`tiple samples, time is usually available to do a better
`separation, especially since you do not have to change
`anything other than the flow rate.
`All Astec columns can be operated from either direction
`without loss of performance because of the uniform pack(cid:173)
`ing procedure used. Thus, backflushing the column can
`be used when reconditioning the column. If performance
`has decreased, frequently changing the direction of the
`column will bring back the separation previously
`observed.
`COLUMN STABILITY
`Since the CYCLOBOND columns all have a backbone of
`silica gel to give them the maximum strength, they should
`only be operated at a pH below 7.5. At pH's lower than·
`3.0 the cyclodextrin groups can be cleaved. Such de(cid:173)
`terioration is also noted with prolonged use with butter
`salts. If buffers are to be used constantly with a CYCLO(cid:173)
`BOND column, it is recommended that a precolumn be
`used since it will presaturate the mobile phase with ppm
`ot silica sparing the CYCLOBOND column.
`A precolum is a pre-saturator column placed between
`the pump and the injector and is filled with 40 µm silica
`gel. Do not contuse this column with the guard column,
`which is placed between the injector andthe analytical
`column and acts as the final tilter for the sample. If using
`biological extracts, samples from fermentation broths, or
`other samples derived from complex matrices, even
`though cleaned, extracted, andbr filtered, always use a
`guard column ahead of the analytical or prepara±!ve col(cid:173)
`umn to ensure the longest life of the working column.
`At the end of each day, always wash any salts, acids,
`or bases (these might be amine moditiers - never any
`strong base!) from the column with the mobile phase
`minus the modifiers - about 25 ml will su1fice to clean
`the column. Not allowing any of these remaining in
`static contact with the CYCLOBOND columns will pro(cid:173)
`long their life.
`To minimize buildup in backpressure, always filter
`samples and butter containing mobile phase. In-line filters
`are also recommended, especially those through which
`the moblle phase or its components are drawn. Always
`have spare filters and frits for the columns ayffil:able so
`these can be rapidly changed so work can proceed.
`REGENERATION
`Any CYCLOBOND column showing decreased resolution
`can be regenerated by passing several column volumes
`of water followed by pure ethanol. The ethanol is more ef(cid:173)
`ficient for displacing substances from the cyclodextrin
`cavity than is methanol.
`STORAGE
`Following the quality control testing of the CYCLOBOND
`column, it is conditioned with methanol for storage and
`shipment. When the column is to be taken from the
`system and stored, it should be regenerated, then stored
`in methanol. Never store a column in a buffer at any pH!
`
`-7-
`
`COLUMN ASSESSMENT PARAMETERS
`Column Assessment Parameters provide useful data as to
`the repeatability of column performance and can be
`used to evaluate your column for s~gn.S ot deterioration.
`Each column is individually tested and assigned a seria
`number tor the traceability of all column components."'------ ·
`Since virtually every molecule, organic and inorganic,
`can be 'included', no useful void volume marker has
`been found. The retention volume of m-nitroaniline in
`methanol/water ( 50 :50) at 1 . 0 ml/min is used as a relative
`measure of consistent packing.
`
`IMPROVED CYCLOBOND SERIES
`Current CYCI.:.OBOND columns (since Oct. 1986) have
`been stabilized tor maximum· cyclode.xtrin loading and
`are produced under stringent GMP guidelines. The
`CYCLOBOND series, however, has undergone two im(cid:173)
`provements since first being offered commercially by Ad(cid:173)
`vanced Separation Techn9logies in l 984. The aim has
`been to produce a line of products with the greatest possi(cid:173)
`ble stability and selectivity. The current· series have
`substantially more cyclodextrin bonded; to the silica gel.
`This increases the retention times while also increasing the
`resolution of many enantiomers. To decrease the reten(cid:173)
`tion times found on the newer columns to those obtained
`with the older CYCLOBOND columns, increase the lip(cid:173)
`ophilic solvent by 20-25%. Thus a separation of Dansyl-L,
`D-amino acids published in 1984 using MeOHffEAA
`(0.1 %), pH 4.0 (40,60)would be changed to (80,20)torcol(cid:173)
`urnns manuiactured since Oct. 1986. This change has al(cid:173)
`lowed for a broader range of separations not possible with
`the earlier CYCLOBOND columns.
`
`. I
`
`I
`
`LOWER DRUG PRICES FOR CONSUMERS, LLC
`Exhibit 1021-9
`IPR2016-00379
`
`

`
`OTHER CYCLOBOND SEPARATIONS
`AND REFERENCES .
`To allow the CYCLOBOND column users to fully appreciate
`·--e wide variety of compound classes and types able to
`
`be separated on these columns and to suggest ot11er
`research areas, a partial listing is furnished in the follow(cid:173)
`ing Tables. Details on the separations can be found in the
`reterences from which this data was abstracted.
`
`Compound
`8-Adrenergic Blockers
`Propranolol
`Meroprolol
`
`Antihistamine
`Chlorpheniramine
`Calcium Channel Blockers
`Verapamil
`Nisolidipene
`Nimodipene
`Diuretic
`Chlorthalidone
`Anticonvulsant
`Hexobarbital
`Mephobarbital
`Mephenytoin
`Triazoline
`Phensuximide
`Anijcorticosteroid
`Aminoglutethimide
`Anttirrtlam.rnatory
`Ketoprofen
`
`Narcotic Analgesic
`Methadone
`Central Nervous System Stimulant
`Methylphenidate·
`Amino Acid Derivs
`L(D)-alanine-bNA
`L(D)-methionine-bNA
`L(D)-alanine-bNE
`Dansyl-L(D)-valine
`Dansyl-L(D)-threonine
`Do:nsyl-L(D)-norleucine
`Dansyl-L(D)-phenylalanine
`Dansyl-L(D)-leucine
`
`k
`
`Mobile Phase
`
`Column
`
`Ref
`
`2.78 (2.89)
`3.51 (3.62)
`
`25:75
`32.68
`
`CB I (50)
`CB I (50)
`
`5.86 (6'.27)
`
`15:85*
`
`CB I (25)
`
`2.94 (3.02)
`4.13 (4.25)
`5.09 (5.34)
`
`+
`30:70
`30:70
`
`CB I (25)
`CB I (50)
`CB I (50)
`
`0.50 (0.72)
`
`30:70
`
`CB I (25)
`
`9.39 (10.7)
`14.8 (16.9)
`0.48 (0.64)
`5.00 (5.75)
`1.97 (2.26)
`
`15:85
`20:80
`40:60
`40:60
`10:90*
`
`7.49 (7.71)
`
`+
`
`CB I (10)
`CB I (10)
`CB I (25)
`CB I (25)
`CB I (25)
`
`CB I (25)
`
`7.67 (8.13)
`
`27:73
`
`CB I (50)
`
`2.38 (2.48)
`
`+
`
`CB I (25)
`
`1.17 (1.33)
`
`10:90*
`
`CB I (50)
`
`~·
`
`5.1 (6.1).
`2.7·(3.o)
`. 1.0 (1.8)
`. 2.2 (2.6)
`1.7(2.1)
`1.9 (2.4)
`3.1 (3.8)
`3.0 (4.0)
`
`50:50
`50:50
`50:50
`sm5o
`50:50
`50:50
`55:45
`50:50
`
`CB I (10)
`CB I (10)
`CB I (10)
`CB I (10)
`CB I (10)
`CB I (10)
`CB I (10)
`CB I (10)
`
`19
`19
`
`19
`
`19
`19
`19
`
`19
`
`19
`19
`19
`19
`19
`
`19
`
`19
`
`19
`
`19
`
`12
`12
`12
`12
`12
`12
`12
`12
`
`Notes: All mobile phase ratios are Me0H/H20 except* where acetonitrile is substituted for the
`methanol. The number in parenthesis after the column· is the length in cm used to obtain the
`separation.
`+
`bNA
`bNE
`
`gradient: Acetonitrile/l % TEAA pH 4.1: 10/90 to 20/80 in 20 min.
`beta-naphthylamide
`beta-napthyl ester
`
`LOWER DRUG PRICES FOR CONSUMERS, LLC
`Exhibit 1021-10
`IPR2016-00379
`
`

`
`Compound
`
`Mobile Phase
`
`Column
`
`....
`
`""';
`
`···-~-"!
`
`Positional Isomers
`o-Cresol
`m-,p-
`o-Xylene
`m-. p-
`a-Nitro phenol
`m-. p-
`o-Nitroaniline
`·m-, p-
`o-Bromobenzoic acid
`m-,p-
`o-Arninobenzoic acid
`m-, p-
`1-Methylindole
`2-, 3-, 5-, 7-
`
`Benzo(a)pyrene
`Benzo(e)pyrene
`1.2.3,4-Dibenzanthracene
`1,2,5,6-Dibenzanthracene
`Phenanthrene
`Anlliracene
`a-Naphthol
`(3-Naphthol
`o, o'-Biphenyl
`P,P' -Biphenyl
`Quinoline
`Isoquinoline
`1,2-Naphthoquinone
`1,4-Naphthoquinone
`a-Naphihoflavone
`13-Naphtho:tlavone
`ct-Ethylphenethyl alcohol
`13-Ethylphenethyl alcohol
`Prostaglandin A 1
`Prostaglandin B1
`Prostaglandin ~
`Prostaglandin B2.
`Vitamin D2
`Lumisterol
`Previtamin D
`
`Geometric Isomers
`cis-Clomiphene
`trans-Clomiphene
`cis-Stilbene
`trans-Stilbene
`cis-Benzo( a)pyrene-7, 8-diol
`
`trans-Benzo(a)pyrene-7, 8-d.iol
`
`cis-3-Hexene-l-ol
`trans-3-Hexene-l-ol
`syn-Azobenzene
`cmti-Azobenzene
`
`2.3
`1.9. 3.5
`3.1
`2.5, 3.8
`1.3
`0.6, 2.4
`5.9
`5.4. 14.0
`5.0
`7.7. 9.1
`4.7
`3.8, 4.8
`1. 7
`1.4, 1.1, 1.6. 0.8
`
`7.9
`6.2
`3.8
`4.2
`2.6
`3.4
`9.0
`11.5
`0.1
`2.2
`1.7
`2.2
`o·.7
`0.9
`3.4
`5.3
`l.7
`2.0
`
`3.8
`. 2.6
`4.6
`2.0
`0.4
`0.7
`1.2
`
`5.4
`3.6
`7.3
`4.5
`18
`19
`20
`24
`0.3
`0.5
`2.6
`4.0
`
`30:70
`
`. 40:60
`
`60:40
`
`40:60
`
`60.40
`
`30.70
`
`50:50
`
`50:50
`
`55:45
`
`55,45
`
`40:60.
`
`55:45
`
`40,60
`
`45,55
`
`50:50
`
`50:50
`
`50:50
`
`50:50
`
`90.10
`
`65.35
`
`55:45
`
`35:65
`30:70
`35:65
`30:70
`50:50
`
`55:45
`
`CB I (25)
`
`CB I (25)
`
`CB I (10)
`
`CB I (10)
`
`CB I (25)
`
`CB I (10)
`
`CB I (25)
`
`CB I (10)
`
`CB I (25)
`
`CB I (10)
`
`CB I (10)
`
`CB I (10)
`
`CB I (25)
`
`CB I (25)
`
`CB I (25)
`
`CB I (25)
`
`CB I (10)
`
`CB I (10)
`
`CB I (25)
`
`CB I (25)
`
`CB I (10)
`
`CB I (25)
`CB IT (10)"
`CB I (25)
`CB Il (10)
`CB I (25)
`
`CB I (10)
`
`i
`~-/ !
`
`"-.__./
`
`60.40
`
`CB I (10)
`
`..... ~
`
`Epimers
`3.8
`Estriol
`11.5
`16-Epiestriol
`5.8
`17-Epi~striol
`2.8
`16, 17-Eptestriol
`9.0
`Testosterone
`3.5
`7.:.Epitestosterone
`1.9
`17 a-Estradiol
`5.0
`1 7 8-Estrad.iol
`2.3
`11 a -Hydroxyprogesterone
`3.7
`11 ~-Hydroxyprogesterone
`8.5
`20 C!-Hydroxy-4-pregnen-3-one
`10.7
`20 B-Hydroxy-4-pregnen-3-one
`8.9
`5 a-Androstan-3, 17-dione
`4.4
`5 S-Ancirostan-3, 17-d.ione
`3.1
`, 5 8-And.rostan -3$-ol-17-one
`5.2
`5 $-And:rostan-3a-ol-d-Aldosterone
`4.9
`17-Isoaldosterone
`2.0
`5 f3-Androstan-3 f3-ol-l 7-one
`10.l
`5 a-Androstcm-3 f3.-ol-17-one
`1.8
`5 a-Androstan-3a-ol-17-one
`6.4
`5 8-Androstan-3a-ol-l 7-one
`Note, All mobile phase ratios are Me0H/H20. The nu,mber in parentheses atter the column is the length in cm used to obtain
`the separation.
`
`.. --
`
`80:20
`
`65:35
`
`55:45
`
`55:45
`
`65.35
`
`65:35
`40:60
`
`75125
`
`75:25
`
`CB I (10)
`
`CB I (IO)
`
`CB I (10)
`
`CB I (10)
`
`CB I (10)
`
`CB I (10)
`CB I (10)
`
`CB I (10)
`
`'CB I (IO)
`
`!
`
`-9-
`
`LOWER DRUG PRICES FOR CONSUMERS, LLC
`Exhibit 1021-11
`IPR2016-00379
`
`

`
`• I
`I!
`
`I
`I
`..
`
`
`
`1.
`
`!
`:
`I
`
`I
`
`I
`
`I
`I
`I
`·J
`11
`
`Compound/Class
`Jubstituted Benzoic Acids
`· Nitrosoamines
`Cyclic
`Acyclic
`Aspartame
`Substituted Phenols
`Hydantoins
`Tarnoxifen
`Quaternary Amines
`Metallocenes
`2,2'-Binaphthylcliyl Crown Ethers
`
`23
`25
`22
`18
`29
`32
`26
`8
`13
`
`CYCLOBOND BIBLIOGRAPHY
`1. New HPLC Column Technology: Inclusion Complex(cid:173)
`ing, THE ASTEC INFORJ\1ER. Vol. 4, Issuo L 1984.
`2, HPLC Inclusion Complexing - An Update, THE ASTEC
`INFORivIER, Vol. 4, Issue 2, 1984.
`3. Armstrong, D.W.