I9, Number 2
`
`February 2002
`
`PHREEB |9(2) |l5—2I6 (2002)
`ISSN 0724-874!
`
`An Official journal of the American Association
`of Pharmaceutical Scientists
`
`
`
`Cancer Pharmacology
`Thermophysical Properties
`Drug Targeting
`Computational Biopharmaceutics
`
`9) mm
`
`In This Issue:
`
`NERACADEMICl
`
`iUM PUBLISHERS
`
`Page 1 of 11
`
`CSL EXHIBIT 1059
`
`Page 1 of 11
`
`CSL EXHIBIT 1059
`
`

`

`
`
`
`.
`
`
`
`-
`
`..
`
`PHARMACEUTICAL RESEARCH
`An Ollie-M Journal of the American ASSOCIIIIIOII of Pharmaceutical Scientists
`-
`l'hunnun-imtnl Rcu'urth publishes innuialive basic research and leehnulogiml advances in [he phuiniateulital-binnitdicnl WIK‘IICL‘F. RIM-Juli .ircas
`inumzil melude‘ phunnueculiu and dru}I delivery. pharmucnkinctiu iind phiirmnecidynnmies. drug metabolism. pliannaeI-lugi iind ltNltultigy‘ medicinal g
`Ii;iliiral products chcinNI). analytical eliemisll}. chemical kinelio and diug stabllil) . biulechnulug}. phiiniiiiceutical technology. and clinical investigat .
`as articles on lhc want eennnmic. ui management aspects nt lhc pharmaceutical wicnecs
`EIII'I'OIl-lN-CHIEF
`Keaidti lnli. Kyutu l'nivcrsil)‘ Hmpilal. Kit-Iii. Japan
`Vince-I H. L. Lee. Deplirtmenl nil PhllrmJCL'UIIClll Ken-mes. lll'liu‘hlh ul
`MyI-n K. Jacobson. L‘nitcniiy tit Ariznnn. Tiiemn. Arvnna
`Southem (‘nlil'nrnm. Ins Angeles. California
`Rudy L. Jolene. University Ill North (‘umlinu ill (‘hupcl Hill.
`North ('Juilina
`EDITOR—EUROPE
`Telwyl Kamila“. Hokkaido l"fll\'L‘l!~ll\ Jiipiin
`Dun ('rn-melin. I'Ircchi I‘nnersil). l‘lrcehI Hi: Vclhfllldnd\
`Mitts 0. Klrlsmn. l'ppmla L nt\clsll}. l ppuilu. Sweden
`EDITOR—JAPAN
`Kwnng-J'm Kiln. Univemh nl Southern ('alilornia. LIN Angela. 0
`String Wen Kim. llnnersil) lil l'llih. Still Luke Cil). l'luh
`Milm "I‘llldl, Kym-i l'mtersu), Kintn Lipan
`Tmltiltim Klmun. Okayama l nucml). Okayama. Japan
`ASSOCIATE EDITORS
`AnV. Ton) Kong, Ringers—The Shite l'nii'enitt nl New Jersey.
`Khum Purl. l'urdue liniiemily. \Vcil lalaiene. lnllinnn
`New Jersey
`Dnvld E. SIIlIII. lv'liii‘eisiti nl' Mieliigun. Anti Aihui. Miehipnii
`Jintlridt Kopeeek. llniierm) nl l'l.ih. '\'all Lake ('il_\x I'lali
`ASSOCIATE EDITORS—EUROPE
`Darin Kroetr. llrin ers'il) ul ('iililnmin. Sim l:l;II1CINCt). Californi-
`Peter R. Luggltli. lulunnes (nulenheig-l‘mwmli. Mainl. (it:
`I
`Meladert Dlnhol. Leiden Uniwrsil). Leiden. 111i: Netherlands
`(‘Iii-Ho Lee. l‘us‘an \dlinniil University Pusan. Snuth Knrea
`load 1. I'nldler. llirechi I'nnemly. Utrecht
`l‘he Nelherlumh
`m Chitin Lee. bung Ki utiLwnii Universin Simon (’ili. South
`EDITORIAL ADVISORY BOARD
`Set-lg .liri Lee. thwu Wiimcns Unnersin. Seoul, South Knrca
`Mull Jose Aloiun. Liniwrsiiy nl Sziiitiupn dc (‘IimpIMeL-i. (hnlptlh Slll’. Spnin
`Chas-Mlduel Leh. l'ni'i‘ersily l‘l Saurland. Suurhruecken. Ge .
`“mum ‘1. Lndden. (iltihuMnx l.l.(‘. llamwer. Mainland
`Gunilla L. Amidon. l'lIlVL'lSIl} of Michigan. Ann Arbor. Michigan
`Bridle) Andenou. l'nueml) ul Kentucky. leunginn KunlueLy
`Krhfim Lmlinua. Cunt-hing l’nnersily. Sweden
`Per Artumn. l'ppstila l‘niversil). l'ppxailli. Sncden
`Pimm Mather-s. Univeml) at Athens. Alhens. (truce:
`Jennie L-S. AI. ()liui Sliitc l nl\’t.'l.\lly. ("Illumhus. Uhiu
`Rlnthll l. Mun). ( iirtltll l, niwnil)‘. (Sarditl. lfmled Kingdom
`You-Hui Bk. Kunngdu ln~lilule ul Science tk 'leehnnlug), Rhine-Iii.
`Tod-nod MaytIln‘i. ()xnkn l'nii-ersity. OsrikiL annn
`\nulh Knren
`1m} R. Nedelmln. Nuurlii l‘linrnimeulimls, [Lual llanui :1. New
`l‘eter Bonnie. lLLx ()neulugi Sun Amt-Inn. lexnu
`Derek T. O'Hngen. ("himn rfllptlrlllllln. chrWIlle_ California
`Ron-Id T. Ron-lint“. Uniiemiy nl Kansas, Lawrence. Rum-us
`Fermi OIL-no. 1 UK)" Women‘s Medical t'ullege. lukyu. Lipan
`Michael J. Piltnl. Ijniivrsili ul (‘nnneclieuL Slur“. ('nnnet‘lieul
`lulu.- lIonstrl. Leidanmslt-rduni (mm In! Diug Rewnreli. lhe
`Nelherlnntk
`Mull I’niusniu. Georgia Institute u!
`ICL’hIltIlt)g\ Georgia
`Harry Britt-in. ( enter Iiir Pharmaceutical Physics. Mill'uld. New Jersey
`Mm Rellinn. St. Jude Children's Research limpilnl. Memphis. T
`Kim L. R. Bmtrwer. l'mvemtv at North ( urnlma Ell ('hiipcl lllll_ ( hapel llill.
`Mir-bun Rubinstein. The llehym l‘mh‘r‘ill" nl Jeiumlcm. le
`“€0th (‘zlrlilinii
`Wollgung Sldée. Uhiu Stale l'xiiicml). ('nliimliiis. Uliiu
`W. Mull Sullniul. ("time-ll Unnenm. llhuea. New \ urll
`Gulllm Radium. Ulll\‘L‘l',\ll) III LuniJuIi. l.\|1IkltIII. L niled Kingdiiiii
`.Inlin F. Carpenter. l'nnemty Iil (‘ulnrudii~ DemeL (‘oluriilln
`Wei-Chino: Shea. liniveim} nl Suulhern (‘aililliiniiL Lm Angela. 5
`Steven J. Shire. Licneiitceh. lIIL . 5. Sun l'niiieiueu. (frilllumia
`Alien H. l- Clio». l'hincsu l iineml) ul Hung lwng. Sliutin. Hung lwng
`Paolo Clillnllbo. L'nuersil) of Parmn. Pairmii. Iluli
`Kenneth B. Sloan. linii-crsily 0! Honda. (ininesiille. Florida
`Michael Colbo, Bristulfllyen \‘quibh. Hillside, Ne“ Jcm)
`build Ii. Small, L'nneniti t‘l Midi-gun, Aim Arboi. Michigan
`Petriell Couvretir. Unu emit: dc l'airia-Sud. ('lialuntii-Miilzihi). l’lallu.’
`Valentino 1. Stella. linii'erxily ol‘ Knnxm. l awn-nee Knnsui
`Hlflllllll Darentlnrl. l'nivenily nl Flundn t‘minesnlle. l-lur'idn
`Yuldii Sufiynm. Unnerx'll) at Tokyo. Tully). lnplin
`Rulli Duncan. ('iiidil'l' L'nlwixily, (’urdill, United Kingdum
`Sean Sulllvln. l‘iiixeml) Ul Florida. l'mlliesnlle l'luIldd
`Williun E. Evans. St. Jude‘s Children's Research Hospital. \lL‘tllplth. 'leiinexwe
`Ytidiitmbii ‘I‘altlllurn. K)nhl llniiemly. Kintn. .lLll‘llll'l
`Gary Fujil. 'Vlnleculiir I‘VPI'L'SS. lnc,. lm Angeles, l'ulifnrnin
`limo TIM-yum. l’lllhlll l‘III\I.‘l.\tl\. l'nltin. .lzipiiu
`Ho-Ixng Fling. SUNYB Sehwl (It Phuimaei. Anilieisi. Ne“ \‘ui‘L
`Tetsuyl Term-Iii.
`l'uh-Iku liniversili. Sendai. Jar-mi
`
`Sven I’mlljlr. Royal Dnnhh Schmil nl Pliiirm:
`(‘le'nhilgk‘IL Denmark
`Bern-rd Testii. llnivenll) nl LdllNill’IIIL'. Luuunnc. Suit/crlund
`Igor Gouda. Acrul. Mclhoume. .‘tuslinlm
`kennetll IIInInIm-l. l‘iiiiutsil) ul Washington. Scaltlc. “inning“!
`Margaret: Ilmulund-Udeauu. l'lllV‘rhllV ut L ppmla. “ppmlln. Sweden
`Allin Tuiji. Knnnmnn l7nti'crsil\. Knnnnmu. Japan
`In lenrtli. l'niien‘ili III Viiilhern ('tilifiirmn [m {\IlgClCS, (Iililnriiiii
`Arto llrtll. lWIlHu‘sll)‘ ul Kunpiu, Kunpm, Finliind
`Multini Haymhi. [okyti Unneisity ul Phurmnq & LIlL‘ Sue-nu». Tukyu, ,lnpdn
`Keiji Yum-nut". ('liihn l'nl\t.‘l\ll_\. (‘Iiihii .Inpnii
`Win E. Henninlt. lilieelil lnmulc lnr l’llufmilt‘cllllcdl Science-x l'lreehl. The
`BOOK REVIEW EDITOR
`Netherlands
`Kin-Ii I‘nrk. l'urdm; l'nixersiiy. Sent-n] nl I'harnme}. “est Li
`Sunni Ilenliemon. Aliigen lnet. Thuuslind (Lib. ('ulilnrnia
`J71“)?
`Aniline} .l. Illclte). l'nnemI) nf Vlirlh (‘Iimlmn rll f‘hapel l‘llll. ('linpel llill.
`Nnrlh (‘liruliiiu
`EDITORIAL ASS 'I‘AN'IS
`Rntli Ellis-Ballard
`Sung-Jim Huang. l'liuiignnni Nalmnnl l nncmiy, South kurL‘JI
`
`Eliuhetli II. (longunt
`T
`jl lg]. liniwnil) ill 'lvl'k)” llmpilnl. I'Iilt)“, Lipiiii
`lithe": llluui. West l'lIillllIttL’ttlJllL‘dl SUH‘IL‘L‘. Nntluiglium. l niled Kingdum
`l‘omoiui Urlliylui-
`
`\t‘ml-iiiIe'l‘II'iIiini I'iihluliru I’H It.“ t“_
`nun Mi
`|)t|ttltulll
`l'lu Nel
`In MINII
`iimiiilIli
`I’IIARMAI'IJV'I'II‘AL Ill-SEARCH INV‘. Off-I-A‘Jll i~ 'Iiililnlit-l
`sum-cumin" price I' \ siliwui «n [It iglii .unl "\illllllg h} l'uhlimluim l \pcillllllg. Im'
`.‘.II) Melu‘lum Menu.
`l liiioni. .N Y llllifi
`l'mudiul» [\tl‘lllgt. Nltl ll )nmuwmfl'
`Inimnliuml
`l‘hummu-umiil Rru'un Ii 1- dlhll-ll led m .Iitlentl in t'lieiiii...il Alkll’lith ( uni-m ( IIIIIL'IIINl Ile \t‘ieiitc l. MIL—\Sl 1|
`\LL'If‘ltl Medim llltlt’\ \lediein \Il l)l l\l
`
`\thmtis Mt-
`‘II I?
`' .
`Ltll
`llacunienhili-In \en-ite Rt‘lclrntt l'ridzile. Science I Ildlllm |ndt-\
`\eirnn (mill-iii
`lmleI
`I~\p.indeIl nml
`l'lie lSl Alrtllllg \enitew -
`( urpomlion. ‘10 [VIII \‘l lIii~ null; inn} N rel‘mdm‘ed. ‘llllCd in ii Ieliieiul nxicm uI Ir.iil~niiiletl III an} Inrm ur m um nit-4m. ulcelmnn meetinriieul. pliniucuriiine.mieru
`'
`
`m tlll'll mixer mlhnul uriilen permnstnii imm Il'h. Pllblbhcr. min the exeepimn oi .m_\ iii.iicii:il Wmlllcd \IK’L’IIIL’iIIlI In: the [\urpuie ul lwiiig eiileicd and executed on :i u
`eulusuu nu- l“ lhr pnielmwr ul [lie and I'Iiummu-uluul Rruun Ii p.iflitip.ile~ III llh ('npinglii (Immune: t ’eIIIer II 'I '('I li.in~,ielniii.il Reluirliiiy \cnitr llte Appeal!!! -
`i.
`
`In III; hulluni ml the I’mi [hiuc ni m lulu-l. m Ihu .nunul lndILJlL’fi Iht mmnym .mnu ~ elunsenl Il‘hil cupiu nI
`lhe .Irin-Ie mas h mud. Iur pews-until nr Iflll'llhll I!“ Ito-
`.“T Rllw“lihl [)nw Dunn-n,
`uii th t'illlLlIlltil’l that ill: nip [my .i
`|I.iI [lee [‘L‘I cup} pet .II’IIL’ll‘ mu -IlelllI|I\.Il pet page Ieesi duet-Ill In Ilil ('umiiglil lle.iI.im‘el rum I'll
`
`IilliLi Ivuun LII'I')n)‘lIlILl'ml. Im arming IIIIli.‘\I\lIC1ll\ penniiitd h_\ \‘telium lil‘ nr luh ml the l‘ 5 (upiriglil law Hie ('t ('i— .‘I iwnpmlil clemnehuuw In: lllL payment tit
`ltic
`iilw
`hm:
`(Iii-i uinscni di-ex n-II ulciid Iii ullieI Iiiid~ Ill eirliiiiie. «ugh as turning Iill ewlwul disluhul
`lee~ by “harm and ullIei uficn iegixiered min Ihe t
`t
`l‘
`
`pII-iiinliuiul I‘quMx Limiting new etillcclm. limb. or I-ii Iu~;ile. iiul In the repiiiiling ui Ileuin ”lilo. and It“! nu nth. UTLLK’JI I!)
`I\\l\df|l~ll\g Inqilinex \l’lt‘lllll he liddrened in :\tl‘crll~lng Stiles, I'lluniitln-nimil KI mm It. \lc'Velll ('liuup. lllL . WI (Mind \ .illei Rlldd. WK, \ ll'kllL‘). l'enlmluiiiii NWT
`Cl UNI) and In (2"! ,‘Il ‘Ifllh
`
`\quanu-n mquuiu .ind sulnuipinm urdrn stuml-l Ix .iddiem-tl III "R [llll‘lhllt'l .n Inllmu I m .1“ uilIIIIIIL‘~ unwinding ~iih~euplmn .ii-eiih in ‘uirih and l .itin kinericai'
`[)Nuhinmn ('t-iilu PUB-n ‘3: Hal) AH l)nrdreehl
`III: Nellierlandx tel
`1.]
`’H h‘7ti1illl
`l.i\ Kl
`'H M70474 L‘-I'I'|.i|ll nrderdeplumkqml
`I'iihli~lie r- J-iiiiiiit'h I )epm Imem
`
`nun-men iiiili
`in Nnrlh .Nnulh .imH eiilrnl America. kluiicr .\elidrmit l'iihlii-lirl\,1i~iiiiiu|~ IICPHIIIIIL‘III, Illl Philip Driie, -\.\~iilliipi Pail. N‘umulli M \ IIZMI
`l‘5l\, Id, I
`l 7.“ rm ”ll-HM; filtlll' kliineiwuklipmnu Sulheiipliim lulu
`Lil
`\ uluim W. 3m: l l.‘ mum I‘vmi III (‘lL‘xlIHllIL H R lli‘nl IiIi l ‘\H l mum l‘rInI .md elthiIIIiL’
`l
`l H I Illmi‘l ‘xli LIMP“ Imlnulimi \ilhnri'u’u turn-lung iimi lb:
`lam-ind ux (prinl unliI I
`l R Is'illll .Nl)
`lKTllI
`IIIIIXJ‘ITX
`LS. I’OSTM \h'l'l’lR: Send .iddlc» L'hJIIL'e“ In I'lnmuuiwmmf [\‘1 mm It Khmer Academicll'lt-iiiim Piililltlwis. 2'.“ Swing SiIeei New. \ lurk. N \
`l‘riiill d III the l \x\
`
`
`
`
`
`Page 2 of 11
`
`
`'
`
`—<
`
`
`
`

`

`PHARMACEUTICAL RESEARCH
`
`l ucial Journal of the American Association of Pharmaceutical Scientists
`
`
`
`‘ 19 Number 2
`
`
`
`CONTENTS
`
`February 2002
`
`’ CH PAPERS
`
`.1
`
`.ting
`
`,'
`
`
`
`
`-- of the Structure of Drug Moieties on the in Virm Efficacy of HPMA Copolymer-Geldanamycin
`Derivative Conjugates
`1""iKasuya. Zheng-Rong Lu. Pavia Koper‘km'a'. S. Esmail Tahihi, and Jindfieh Kopec’ek
`
`
`1!
`
`termination of Binding of Cisplatin to DNA in the Presence of Biological Thiols: Implications
`of Dominant Platinum-Thiol Binding to Its Anticancer Action
`: Volckovu. Lea P. Dadones, and Ralhindra N. Bose
`
`
`
`
`. ndency of DL-Lactide/Glycolidc Copolymcr Particulales for lntra-Articular Delivery System
`on Phagocytosis in Rat Synovium
`Ufra Harirawa, Kalsaaki Kubom. Izumi Tubal, Keiichi Sara. Hiromirsa Yamamoro.
`11‘ :fumi Takeuehi. and Yoshiuki Kawashima
`
`
`
`
`
`
`
`
`m traaki Nakamaru, Fumihikn Ushigome, Nnrikn Knyahu. Sim/i Samh, Kiynmi Tsukinmri,
`.' uooNakano. Hisakazu Ohmni. and Yasufimri Sawada
`
`-;-ili1y Profiles of M-Alkoxysubstituted Pyrrolidinoethylesters of Phenylcarbamic Acid across
`Caco—2 Monolayers and Human Skin
`" , 1 ka Gyiirfisiovd, Leena Lailinen. Johanna Rairmm. Jozef Ciz'nuirik. Eva Sell/dram. and
`'auni Hinl'onen
`
`“1 : Processing of Poly(Ethylene 1mine)/Ribozyme Complexes Can Be Observed in Living
`Cells Using Confocal Laser Scanning Microscopy and Inhibitor Experiments
`I Merdan. Klaus Kalmlh. Dagmar Fischer. lindrich Knpecek. and Thomas Kissel
`
`-
`
`_
`
`3‘
`
`i.|ll1
`
`'tive Inhibitory Effects of Different Compounds on Rat Oatpl
`(SlethLil-Mediated Transport
`=1 oshihisa Shiram. Daisuke Sugiyama. Himyaki Kusuhara. Yukiu Karo. Takaaki Abe.
`_" er]. Meier. Tommi huh. and Yuiehi Sugiyuma
`
`(Slc21al)- and 021th
`
`'- Gradient~Dependent Transport of Valproic Acid in Human Placental Brush-Border Membrane
`Vesicles
`
`
`
`‘~ ntial of Chitosan in Enhancing Peptide and Protein Absorption across the 'l‘R146 Cell
`Culture Model-An in Vim: Model of the Buccal Epithelium
`I Porrero. Carmen Remarizin-Lépez. and Hanne Merck Nielsen
`
`
`
`
`
`NW»
`
`, Acylation by Poly(a-H_vdroxy Esters)
`rea Lucke. Josef Kiermaier. and Aehim (iopferieh
`
`Page 3 of 11
`
`115
`
`124
`
`140
`
`147
`
`154
`
`162
`
`169
`
`175
`
`

`

`
`
`CONTENTS (Continued)
`
`Experimental and Computational Screening Models for Prediction of Aqueous Drug Solubility
`Christel A. S. Bergstrom, UlfNorinder, Kristina Luthman, and Per Artursson
`
`
`
`
`Pharmaceutical Engineering
`
`Production and Characterization of a budesonide Nanosuspension for Pulmonary Administration
`Claudia Jacobs and Rainer Helmur Muller
`
`Thermophysical Properties of Phannaceutically Compatible Buffers at Sub-Zero Temperatures: Implications
`for Freeze-Drying
`Evgenyi Y. Shalaev, Tiffany D. Johnson-Elton, Liuquan Chang, and Michael]. Pikal
`
`PharmacokinetieslPhu-macodynamics
`
`Effect of Testosterone Suppression on the Pharmacokinetics of a Potent GnRI-I Receptor Antagonist
`Eugenia A. latsimirskaia, Margaret L. Gregory, Kenna L. Anderes, Rosemary Castillo.
`K. Eric Milgram, David R. Luthin, Ved P. Pathak, Lance C. Christie. Haresh Vazir.
`Mark B. Anderson, and John M. May
`
`Ignoring Pharmacokinetics May Lead to Isoboles Misinterpretation: Illustration with the Norfloxacin~
`Theophylline Convulsant Interaction in Rats
`Miren Cadart. Sandrine Marchand, Claudine Pariat. Serge Bouquet. and William Coue!
`
`AAPS ELECTRONIC SCIENTIST
`
`Debbie Werfel
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Page 4 of 11
`
`

`

`
`
`mtucnuiml Research.
`
`l"nl. I9, No. .7. Fehrmln' 3002 RD 2002)
`
`Research Paper
`
`
`
`ermophysical Properties of
`harmaceutically Compatible Buffers
`
`. Sub-Zero Temperatures:
`... plications for Freeze-Drying
`
`
`'
`
`_ nyi Y. Shalaev”. Tiffany D. Johnson-Elton".
`. uan Chang‘. and Michael J. ritualJ
`
`veil October 10. 200i; accepted November I. .7001
`
`. To evaluate crystallization behavior and collapse tempera-
`I
`.
`ng') of buffers in the frozen state. in view of its Importance in
`
`development of lyophilized formulations.
`
`'
`v
`v
`I
`Sodium tartrate. sodium malate. potassium citrate. and so-
`u citrate buffers were prepared with 3 pl! range within their
`
`-' ‘dual buffering capacities. Crystallization and the Tg‘ were de.
`‘
`I during healing of the frozen solutions using standard DSC‘ and
`-n led DSC.
`
`
`i
`Ls. Citrate and mobile did not exhibit crystnllizution. while suc-
`. e and tartrate crystallized during heating of the frozen solutions.
`
`citrate buffer had a higher Tg‘ than malatc and tartrate buffers at
`
`same pH. Tg' vs. pH graphs for citrate and malate buffers studied
`. a similar shape. with a maximum in Tg‘ at pH ranging from 3 to
`
`- Tg' maximum was explained as a result of n competition
`
`- a two opposing trends: an increase in the viscosity of the armor
`.
`~ phase because of an increase in electrostatic interaction. and a
`
`u
`in the 'l'g‘ because of an increase in a water concentration of
`freeze-concentrated solution.
`
`~
`‘ damn. Citrate buffer was identified as the preferred buffer for
`
`.-' ’ ed pharmaceuticals because of its higher 'l’g' and it lower
`
`-
`ization tendency.
`
`t WORDS: lyophilimtton; freezing; hul'lers; collapse: glass Iran-
`-
`'DSC.
`
`' ODUCI‘ION
`
`
`
`Many pharmaceuticals contain a buffer to control pH. to
`n ' optimal chemical and physical stability of a drug mol-
`. Buffering capacity and a possibility of a buffer-specific
`‘
`
`is are the major buffer properties which are usually
`‘1')
`into consideration in development of liquid pharma-
`'-:
`
`I'l fonnulations (I). For lyophilized formulations. there
`No additional physical chemical parameters to consider.
`
`buffer crystallization potential at sub-ambient tempera—
`
`- and the collapse temperature. A buffer component may
`
`.V‘ ' ' during freezing producing significant pH changes
`
`':, tare usually undesirable and should be avoided. Crys-
`lion and pH changes of phosphate buffer at suhvzero
`
`’
`tures were studied in detail in the presence of differ—
`wetal ions and in a wide range of pH and concentration
`
`3 Systematic studies of equilibrium freezing behavior of
`
`
`
`. m Laboratories. Pfizer Inc.. Groton. Connecticut 06340.
`
`.. of Pharmacy. University of Minnesota. Minnesota.
`
`.
`i of Pharmacy. University of Connecticut. Sturrs. Connecticut
`
`Rhom correspondence should be addressed.
`
`‘_y,shalaevtg'grotonplizencom)
`
`te-mnil:
`
`in
`phosphate buffer were performed by van den Berg 6! ill.
`1950—1960. In these studies. liquid (unfrozen) portions of a
`frozen solution was physically separated at sub-zero tempera-
`tures. and the pll and composition of the liquid portion were
`measured at room temperature (2—4). Later. other methods
`were used such as measurements of pH at sub-zero tempera-
`tures with a low-temperature electrode (5.6.8.9) and pH in-
`dicators (10). X—ray diffraction measurements at sub-ambient
`temperatures (7). and DSC studies (ll-13). Significant pH
`changes were observed depending on a metal ion type and
`experiment setup (sample size. cooling rate). Based on these
`results. phosphate buffer is generally regarded to be undesir-
`able for lyophilized formulations. at least if high buffer con-
`centrations are required to maintain high buffer capacity ( l4).
`There are some studies of other buffers of pharmaceutical
`interest (citrate. glycine. succinate. carbonate) in very narrow
`ranges of solution pH and concentration (9.l0.15).
`Another physical chemical parameter critical for lyophi-
`limtion is the collapse temperature (16). Freeze drying above
`the collapse temperature produces loss of the cake-like struc-
`ture that one desires. Obviously. materials with a higher col~
`lapse temperature can be freeze dried at a higher tempera
`ture. hence providing a faster and more robust lyophilization
`cycle. If the collapse temperature ofa formulation is relatively
`low. it is more difficult and sometimes impossible to lyophilize
`such a fomiulation in a practical process. As a rule. presence
`of amorphous buffer in a formulation decreases the collapse
`temperature resulting in recommendations to minimize buffer
`concentration in lyophilized formulations (l7). The collapse
`temperature can be measured by different techniques. with
`freeze drying microscopy and DSC being the methods of
`choice in most cases. With DSC. a thermal transition denoted
`
`Tg~ is measured as the temperature of an endothermic step
`which precedes the melting endotherm on DSC heating
`curves of frozen solutions (Io).
`It should be stressed that
`interpretation of the physical nature of the Tg‘ thermal event
`is still controversial. There are two alternative interpretation
`of the Tg'. The Tg‘ thermal event has been explained as either
`a glass transition of the freeze-concentrated solution (18.19),
`or onset of ice melting marked as Ts (softening temperature)
`(20). or Tm (21). Despite of this controversy. there is a com-
`mon agreement that the 'Tg‘ corresponds closely to the col-
`lapse temperature. the collapse temperature normally being
`higher by l-3°C (16). Frequently. one detects a second very
`weak apparent glass transition roughly 20“C lower than Tg'.
`This lower transition is denoted Tg“. and does not appear to
`be related to the collapse phenomena. The Tg' of several
`acids and bases (ascorbic acid. citric acid. glycine. HEPES.
`TRIS) and buffers (citrate. TRIS. acetate. glycine. and Mati-
`dine) has been determined in (12.15.22-25).
`ln majority of
`these studies. the Tg' was determined at a single pH value
`(with exceptions (25) for histidinc and ( l5) for glycine). There
`is a lack of systematic data on collapse temperatures and
`crystallization behavior as a function of pH for buffers of
`pharmaceutical significance.
`In the present study. crystallization behavior and col-
`lapse temperature of several buffers (citrate. succinate. ma-
`late. tartrate) have been studied using DSC. Each buffer was
`prepared at different pH to cover the buffering range of the
`particular buffer. It should be emphasized that variation of
`
`
`
`Page 5 of 11
`
`195
`
`lr‘:-‘-f‘74l.‘il='lllfll<lll"5"“ V" :lKl.’ Plenum Publishing (foil-motion
`
`

`

`I96
`
`Shnlacv. Johnson-Elton. Chang. and Pin
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`§E 5L
`
`L 5OI
`
`§5 E E:
`
`t:
`
`40
`
`-20
`
`Temperature (°C)
`Fig. I. Representative DSC healing curves of citric acid/Nat‘m (A)
`and citric acid/KOH (B) solutions. Magnified low—temperature porv
`tions of the DSC scans are shown. Numbers present solution pH.
`Scanning rates: It)"'(‘lmin. The experiments were run with Prrkin'
`Elmer Pyris-l DSC.
`
`cndothcrm is not shown). Two consecutive endothermic
`events. Tg" and Tg'. were observed in majority of cases which
`is typical for frozen aqueous solutions (18-21). There is a
`common agreement that a higher temperature event (Tg')
`corresponds to the collapse temperature.
`A "clip“ in the baseline was observed on DSC heatina
`curves of sodium citrate at pH's from S to 7. Such "dip" on I
`DSC heating curve could be due to an exothermic event sudl
`as crystallization; if this is the case. assignment of the TB.
`event is uncertain. To determine if crystallization occurred in
`these samples.
`two types of DSC experiments were [369
`formed. In the first experiment. a thermal cycling study “’35
`performed; thermal cycling allows one to separate reversiblfi
`thermal transitions (such as glass transition and melting) from
`irreversible transitions (such as crystallization) and artifacf‘
`(such an event associated with a change in a sample shape a
`the DSC pan) (20.26). In this thermal cycling experiment. “I"
`frozen solution was first heated to «145°C (which is the Uni“
`temperature of the thermal event under consideration) f0"
`lowed by cooling to —65‘C. and then heated from —65 to 25°C
`If the thermal event under consideration is the (ice) rrySml‘
`lization exothenn. the second DSC heating curve should hit“
`a different appearance because crystallization would 0‘79“"
`only during the first run and thus would be irreversible. 1"
`particular. the temperature of the first endothermic 51"
`would be shifted to higher temperature (if ice crystalli/c‘l 0‘
`to the lower temperature (if solute crystallizes). Results Of
`thermal cycling experiment are shown in Fig.2u. It can M 5
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`pH is equivalent to variation of the molecular/ionic species in
`solution as pH variations alters the extent of ionization of
`buffer. Significant changes in the crystallization behavior. and
`the Tg‘ were observed as a function of solution pH. An un-
`expected pattern of Tg‘ changes with solution pH was ob»
`served for all buffers studied with Tg‘ having a maximum
`:tround pH 4. In addition. it appeared that a metal ion type
`(i.e.. Na vs. K) has a significant impact on Tg‘ of a citrate
`buffer.
`
`MATERIALS AND METHODS.
`
`Materials
`
`Reagent grade succinic acid and DL-malic acid were pur-
`chased from Fisher Scientific and Sigma. respectively. Citric
`acid of USP grade. L-(+)-tartaric acid of NF grade. and so»
`dium hydroxide of NF grade were obtained from JT Baker. In
`addition. DL-tartaric acid of reagent grade from EM Science
`was used. Dcionized water was used to prepare all of the
`buffer solutions that were studied. Citric acid. succinic acid.
`
`tartaric acid. malic acid were prepared as 0.25M solutions.
`The acids were titrated with 0.25M sodium hydroxide to the
`desired buffer pll. In addition. citric acid/potassium hydrox-
`idc solutions were prepared by the same methods. Weight of
`the added base was measured. The pH range studied was
`chosen as to be within the pH range where the buffer system
`had significant buffering capacity (1).
`
`DSC Experiments
`
`DSC experiments were performed with a Perkin-Elmer
`Pyris l instrument and TA Instruments modulated DSC 2920
`instrument equipped with Refrigerated Cooling System. Ap-
`prox. 15 pl of solution were placed in aluminum pans. and
`empty aluminum pans were used as a reference with both
`instruments. Other details of the experiments performed with
`the Perkin-Elmer instrument are as follows. The instrument
`
`was calibrated using melting points of indium at heating rate
`10°C/min. The calibration was checked using de-ionized wa-
`ter. The uncertainty in the temperature calibration was esti~
`mated to be within 1.5’C. Samples were cooled to -6(l°(‘ at
`10°Clmin. then held at —60°C for 5 min. and then heated from
`430°C to 25"(‘ at ”PC1min. The Tg‘ and Tg" temperatures
`were dctcmtined as extrapolated onset temperatures using
`Pyris software. Experiments were performed with the TA
`instrument as follows. Calibrations were performed using in-
`dium as standard at a heating rate NFC/min and l"'(‘/min for
`standard and modulated method. respectively. The purge gas
`used was nitrogen with a flow rate at 50 ml/min. Samples were
`run in two different modes: (i) Standard DSC mode. which
`the samples were cooled to -60"’C at 10"C/min. cquilibrate at
`-6()“C for 5 min. and then heated to 25°C at UPC/min. (ii)
`Modulated DSC mode. which the samples were cooled to
`—6(l°C and then heated to 25°C at the same heating and coolv
`ing rate of 1”C/min. The run was modulated with an ampli-
`tude xtlj‘C and a period 100 seconds. The Tg‘ and Tg" tem—
`peratures were determined as extrapolated onset tempera—
`tures using TA universal analysis software.
`
`RESULTS
`
`Typical DSC curves of sodium and potassium citrate
`buffer are shown in Figs. la and lb. respectively (ice melting
`
`% P
`
`age 6 of 11
`
`

`

`
`
`.
`
`i ophysical Properties of Buffers at Sub-Zero Temperatures
`
`197
`
`
`
`-3o
`
`8 4o
`
`
`s
`Temparature(°C) is8
`
`-50
`
`430
`
`2
`
`3
`
`4
`
`5
`
`s
`
`7
`
`Solution pH
`
`22(
`
`U-
`
`a g
`
`l-
`
`Solution pH
`Fig. 3. Tg‘ and Tg" of citric acid/NaOH (A) and citric acid/KOH (B)
`
`solutions as u function of pH. C: Tg' measured with 'l'A DSC: I: Tg'
`measured with Perkin-Elmer DSC: l3: Tg" measured with TA DSC;
`O: Tg" measured with PerkinvElmer DSC: A: Tg‘ measured with TA
`MDSC; V: Tg“ measured with TA DSC: O: Tg' measured with TA
`DSC in themal cycling experiment: X: Tg" measured with TA DSC
`in thermal cycling experiment. Each data point corresponds to a
`single DSC run. Lines are given as an visual aid.
`
`different sample preparations. For sodium citrate. hoth Tg'
`and Tg“ passed through a maximum at pH ~4. Similarly. for
`potassium citrate. Tg' passed through maximum between pH
`3 and 4.
`DSC curves of malic acid/NaOH solutions had a similar
`appearance to the citrate buffer solutions with one or two
`endothermic steps (Tg" and Tg') followed by ice melting peak
`(curves are not shown). Tg" and Tg' of malate buffer as a
`function of solution pH are shown in Fig. 43. Again, Tg' goes
`through a maximum at pH 4.
`Succinic acid/NaOH and tartaric acid/NaOH solutions
`
`demonstrated a different thermal behavior. Representative
`DSC heating curves of succinic acid/NaOH and tartaric acid/
`NaOH systems are shown in Figs. 5 and 6. respectively. Buffer
`crystallization occurred in all three succinic acid/NaOH mix-
`tures studied (pH 4. 5. 6) as evidenced from observation of
`both the exothermic peak (D) and endothermic peaks (M)
`prior to ice melting. A weak endothermic step immediately
`before the crystallization cxotherm. which was observed at
`pH 5 and 6 in succinatc buffer (Fig. 5. inset). may be assigned
`to either Tg' or Tg". We did not attempt to characterize this
`transition in more detail. Crystallization was not detected in
`pure succinic acid. Lack of crystallization observed for the
`free acid indicates that a salt (not the free acid) crystallized
`during heating of frozen succinic acid/NaOH solutions. In
`
`
`
`
`
`
`Temperature (°C)
`Thermal cycling (A) and modulated DSC (B) runs of citric
`‘ 80H solution with pH 6. The data were obtained with TA
`
`)
`
`first and second healing curves are practically identical.
`1 e position of the first endothermic step did not change
`. second scan. Hence. the results of the themial cycling
`ent suggested that crystallization did not occur during
`
`
`
`
`
`Modulated DSC allows separation of irreversible
`- as crystallization) and reversible (such as glass transi-
`
`Ihermal events (19). Modulated DSC heating curves are
`.v
`in Fig. 2b. The reversing heat flow curve shows two
`tive endothermic events Hg" and Tg'). similar to a
`
`DSC scan. The nonreversing heat flow curve shows
`
`re is perhaps some crystallization. as evident from the
`exothermic peak centered at -45"‘C: however the mag-
`- of the exotherm is very small. and the modulated DSC
`' Ie not consistent with a significant amount of crystalli-
`.‘ occurring during heating the frozen solution. Tg‘ tem-
`.
`- e determined from the modulated DSC run is slightly
`
`' than determined from the regular DSC scan: however.
`
`“i rence is close to the estimated experimental error.
`ventral cycling and modulated DSC experiments indi-
`l“ t regular (non-modulated) DSC scans can be used to
`
`Tg' and Tg" of the sodium citrate solutions at pH 5
`"' e did not attempt to investigate the origin of the ap-
`
`' “dip“ which was observed prior to the Tg' event for
`I
`-- citrate solutions at pH 5 to 7 in more detail. Figure 3
`
`Tg' and Tg" as a function of pH for sodium citrate and
`
`‘ m citrate. There is good agreement between results
`i with the two different DSC instruments and with
`
`'
`
`
`
`
`
`
`Page 7 of 11
`
`

`

`198
`
`Shalaev. Johnson-Elton. Chung, and Film
`
`3
`
`4
`
`5
`
`6
`
`(°C)
`Temperature
`
`
`
`
`aTemperature(°C) isU)
`
`‘6
`
`a
`
`4
`
`5
`
`Solution pH
`
`Fig, 4, Tg' and 'l'g" of malic acid/NaOH (A) and tartaric acid/NaOH
`(B) solutions as a function of pH. All data point but open squares in
`Fig. 4B correspond to solutions prepared with L-(~»)-tartaric acid. and
`open squares correspond to solutions prepared with DL-tartaric acid.
`See Fig. 3 for other symbols. Each data point corresponds to a single
`DSC run. Lines are given as a visual aid
`
`agreement with this hypothesis. both crystallization exo-
`thcmis and melting endothem'is were stronger in succinate
`solution with pH 6 (solution with a higher salt content). In
`addition. multiple melting peaks were observed at pH 4 and 5.
`indicating that several crystalline succinate salts were formed
`(i.e.. perhaps mono- and di-sodium succinate and their by-
`dratcs).
`A Tg‘ event followed by a strong crystallization peak was
`Observed in a solution of L-(+)-tartaric acid/NaOH at pH 3
`(Fig. (in). In addition. a weaker exotherm was observed in
`solution of pH 4. We note that a solution of DL—tartaric acid/
`NaOH at pH 3 had a much weaker exothermic event (scan is
`not shown) indicating that the racemic reagent had a lower
`crystallization tendency. Fig. 4b shows Tg' and Tg" vs. pH for
`tartaric acid/NaOH system. Despite of the difference in crys-
`tallization behavior. there was no significant difference in the
`Tg' between solutions prepared with either L-(+)-tartaric acid
`or DL.-tartaric acid. Both Tg' and Tg" for tartaric acid/NaOH
`solutions were slightly higher at more acidic pH (pH 3) than
`at pH 5.
`
`DISCUSSION
`
`Tg‘ vs. pH profiles with a maximum were observed for
`citrate and malate buffers. Such behavior has not been re-
`ported in the literature. The lack of literature precedent is not
`
`Page 8 of 11
`
`Heat
`
`Flow(mW)
`
`Temperature (°C)
`Fig. 5. Representative DSC heating curves of suoeinic acidifs'nOH
`solutions. lnsct shows magnified portions of DSC curves with :i non-
`idcntilicd thermal event (which may be either Tg' or Tg"). Scanning
`rates: l0”Ca'min.
`
`surprising. Indeed. there is only one systematic study of Tg'
`vs. pH described in the literature (25). It was observed that
`Tg' of frozen histidine solutions decreased with an increase in
`pH (25). Such a decrease in 'l‘g‘ was explained as due to
`increased salt concentration.
`
`in this work. an opposite trend in Tg‘ vs. pH was ob-
`served. i.e., Tg' first increased with increasing salt concenm'
`tion. To illustrate this point. Tg‘ is presented as a function of
`NaOHJcitric acid mole ratio (Fig. 7). An increase in sodium
`content initially produced a significant increase in Tg’ (Which
`is opposite to the trend reported in (25) for histidincl. fol-
`lowed by a gradual decrease in Tg‘ with further increase in
`sodium content.
`
`An increase in Tg‘ with increasing sodium content can be
`rationalized in terms of increasing ionic interactions .1» the
`concentration of ionic species increases. Indeed. salts fre'
`quently have higher glass transition temperatures than the
`corresponding free acids. For the organic acid. indomclhaciflv
`Tg of the free acid was ~75"C lower than that of the sodium
`salt (27). Hence. the initial increase in Tg' in citrate and ma-
`late buffers when pH increased from 2.5 to 4 can be explained
`on the same basis. That is, ionization allows electrostatic ifl'
`leractions between species and corresponding higher viscos'
`ity. However. this "electrostatic interaction" concept ulonc
`cannot account for the maximum of Tg' vs. pH. Clearly Bn'
`other effect must be important