2 W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60come the instability barrier, proteins often have to cal and chemical instabilities and stabilization ofbe made into solid forms to achieve an acceptable proteins in aqueous and solid states (Manning etshelf life. al., 1989; Cleland et al., 1993); chemical instability The most commonly used method for preparing mechanisms of proteins in solid state (Lai andsolid protein pharmaceuticals is lyophilization Topp, 1999); various factors affecting protein sta-(freeze-drying). However, this process generates a bility during freeze-thawing, freeze-drying, andvariety of freezing and drying stresses, such as storage of solid protein pharmaceuticalssolute concentration, formation of ice crystals, pH (Arakawa et al., 1993); and application ofchanges, etc. All of these stresses can denature lyophilization in protein drug development (Pikal,proteins to various degrees. Thus, stabilizers are 1990a,b; Skrabanja et al., 1994; Carpenter et al.,often required in a protein formulation to protect 1997; Jennings, 1999). Nevertheless, it appearsprotein stability both during freezing and drying that several critical issues in the development ofprocesses. solid protein pharmaceuticals have not been fully Even after successful lyophilization, the long- examined, including various instability factors,term storage stability of proteins may still be very stabilization, and formulation of solid proteinlimited, especially at high storage temperatures. In pharmaceuticals.several cases, protein stability in solid state has To have an up-to-date perspective of thebeen shown to be equal to, or even worse than, lyophilization process and more importantly, itsthat in liquid state, depending on the storage application in formulating solid protein pharma-temperature and formulation composition. For ceuticals, this article reviews the recent investiga-example, a major degradation pathway of human tions and achievements in these exciting areas,insulin-like growth factor I (hIGF-I) is oxidation especially in the past 10 years. Four interrelatedof Met59 and the oxidation rate in a freeze-dried topics are discussed sequentially, lyophilizationformulation in air-ﬁlled vials is roughly the same and its denaturation stresses; cryo- and lyo-pro-as that in a solution at either 25 or 30°C tection of proteins by excipients; design of a ro-(Fransson et al., 1996). Similarly, the oxidation bust lyophilization cycle; and with emphasis,rate of lyophilized interleukin 2 (IL-2) is the same instability, stabilization, and formulation of solidas that in a liquid formulation containing 1 mg protein pharmaceuticals.ml − 1 IL-2, 0.5% hydroxypropyl-b-cyclodextrin(HP-b-CD), and 2% sucrose during storage at 4°C(Hora et al., 1992b). At a high water content 2. Lyophilization and its denaturation stresses( 50%), the degradation rate of insulin is higherin a lyophilized formulation than in a solution 2.1. Lyophilization processwith similar pH-rate proﬁles in both states(Strickley and Anderson, 1996). The glucose-in- Lyophilization (freeze-drying) is the most com-duced formation of des-Ser relaxin in a mon process for making solid protein pharmaceu-lyophilized formulation is faster than in a solution ticals (Cleland et al., 1993; Fox, 1995). Thisduring storage at 40°C (Li et al., 1996). These process consists of two major steps: freezing of aexamples indicate that stabilizers are still required protein solution, and drying of the frozen solidin lyophilized formulations to increase long-term under vacuum. The drying step is further dividedstorage stability. into two phases: primary and secondary drying. In the past two decades, numerous studies have The primary drying removes the frozen water andbeen conducted in the areas of protein freezing the secondary drying removes the non-frozenand drying, and instability and stabilization of ‘bound’ water (Arakawa et al., 1993). The amountproteins during lyophilization and long-term stor- of non-frozen water for globular proteins is aboutage. Many critical issues have been identiﬁed in 0.3–0.35 g g − 1 protein, slightly less than thethis period. These studies and achievements have proteins’ hydration shell (Rupley and Careri,been reviewed elsewhere with emphasis on physi- 1991; Kuhlman et al., 1997). More detailed analy-
W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60 3sis of each lyophilization step is provided in Sec- cluding formation of dendritic ice crystals, in-tion 4. creased ionic strength, changed pH, and phase Lyophilization generates a variety of stresses, separation; and (3) drying stress (removing of thewhich tend to destabilize or unfold/denature an protein hydration shell).unprotected protein. Different proteins toleratefreezing and/or drying stresses to various degrees. 2.2.1. Low temperature stressFreeze-thawing of ovalbumin at neutral pH did The ﬁrst quantitative study on low-temperaturenot cause denaturation (Koseki et al., 1990). Re- denaturation of a model protein was conductedpeated (three times) freeze-thawing of tissue-type presumably by Shikama and Yamazaki (1961).plasminogen activator (tPA) did not cause any They demonstrated a speciﬁc temperature rangedecrease in protein activity (Hsu et al., 1995). in which ox liver catalase was denatured duringSome proteins can keep their activity both during freeze-thawing. Cold denaturation of catalase atfreezing and drying processes, such as a1-an- 8.4 mg ml − 1 in 10 mM phosphate buffer (pH 7.0)titrypsin in phosphate – citrate buffer (Vemuri et started at − 6°C. Loss of catalase activity reachedal., 1994), porcine pancreatic elastase without ex- 20% at − 12°C, remained at this level betweencipients (Chang et al., 1993), and bovine pancre- − 12°C and near − 75°C, then decreased gradu-atic ribonuclease A (RNase A, 13.7 kD) in the ally from − 75 to − 120°C. There was almost nopresence or absence of phosphate (Townsend and activity loss between − 129 and − 192°C. SimilarDeLuca, 1990). results were also obtained for ovalbumin by However, many proteins cannot stand freezing Koseki et al. (1990). Incubation of frozen ovalbu-and/or drying stresses. Freeze-thawing caused loss min solution caused structural change of ovalbu-of activity of lactate dehydrogenase (LDH) min, as monitored by UV difference spectra,(Nema and Avis, 1992; Izutsu et al., 1994b; An- which increased with decreasing temperature be-dersson and Hatti-Kaul, 1999), 60% loss of L-as- tween −10 and − 40°C. Further decrease inparaginase (10 mg ml − 1) activity in 50 mM incubation temperature to − 80°C caused lesssodium phosphate buffer (pH 7.4) (Izutsu et al., structural change, and no change at − 192°C.1994a), and aggregation of recombinant Perlman and Nguyen (1992) reported that inter-hemoglobin (Kerwin et al., 1998). Freeze-drying feron-g(IFN-g) aggregation in a liquid mannitolcaused 10% loss of the antigen-binding capacity formulation was more severe at −20°C than atof a mouse monoclonal antibody (MN12) (Ress- − 70, 5 and 15°C during storage. To preventing et al., 1992), more than 40% loss of bilirubin freezing-induced complication in studying coldoxidase (BO) activity in the presence of dextran or protein denaturation, cold and heat denaturationpolyvinylalcohol (PVA) (Nakai et al., 1998), loss of RNase A has been conducted under high pres-of most b-galactosidase activity at 2 or 20 mg sure (3 kbar). Under this condition, RNase Aml − 1 (Izutsu et al., 1993, 1994a), complete loss of denatured below −22°C and above 40°C (Zhangphosphofructokinase (PFK) and LDH activity in et al., 1995). All these examples are clear indica-the absence of stabilizers (Carpenter et al., 1986, tion of low temperature denaturation rather than1990; Prestrelski et al., 1993a; Anchordoquy and a freezing or thawing effect.Carpenter, 1996), and dissociation of Erwinia L- The nature of cold denaturation has not beenasparaginase tetramer (135 kD) into four inactive satisfactorily delineated. Since solubility of non-subunits (34 kD each) in the absence of any polar groups in water increases with decreasingprotectants (Adams and Ramsay, 1996). temperature due to increased hydration of the non-polar groups, solvophobic interaction in2.2. Denaturation stresses during lyophilization proteins weakens with decreasing temperature (Dill et al., 1989; Graziano et al., 1997). The The lyophilization process generates a variety decreasing solvophobic interaction in proteins canof stresses to denature proteins. These include (1) reach a point where protein stability reaches zero,low temperature stress; (2) freezing stresses, in- causing cold denaturation (Jaenicke, 1990). While
4 W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60normal or thermal denaturation is entropy-driven, centration upon freezing may drastically acceler-cold denaturation is enthalpy-driven (Dill et al., ate protease-catalyzed protein degradation.1989; Shortle, 1996). Oligomeric proteins typicallyshow cold denaturation, i.e. dissociation of sub- 2.2.3. Formation of ice-water interfaceunit oligomers, since association is considered to Freezing a protein solution generates an ice-wa-be a consequence of hydrophobic interaction ter interface. Proteins can be adsorbed to the(Jaenicke, 1990; Wisniewski, 1998). Theoretically, interface, loosening the native fold of proteins andthe calculated free energy of unfolding (DGunf) for resulting in surface-induced denaturationproteins has a parabolic relationship with temper- (Strambini and Gabellieri, 1996). Rapid (quench)ature. This means that a temperature of maximum cooling generates a large ice-water interface whilestability exists, and both high and low tempera- a smaller interface is induced by slow coolingture can destabilize a protein (Jaenicke, 1990; (also see Section 4.2). Chang et al. (1996b)Kristjansson and Kinsella, 1991). ´ demonstrated that a single freeze–thaw cycle with quench cooling denatured six model proteins, in-2.2.2. Concentration effect cluding ciliary neurotrophic factor (CNTF), gluta- Freezing a protein solution rapidly increases the mate dehydrogenase (GDH), interleukin-1concentration of all solutes due to ice formation. receptor antagonist (IL-1ra), LDH, PFK, andFor example, freezing a 0.9% NaCl solution to its tumor necrosis factor binding protein (TNFbp).eutectic temperature of −21°C can cause a 24- The denaturation effect of quench cooling wasfold increase in its concentration (Franks, 1990). greater or equivalent to that after 11 cycles ofThe calculated concentration of small carbohy- slow cooling, suggesting surface-induced denatu-drates in the maximally freeze-concentrated ma- ration. This denaturation mechanism was sup-trices (MFCS) is as high as 80% (Roos, 1993). ported by a good correlation (r= 0.99) foundThus, all physical properties related to concentra- between the degree of freeze-induced denaturationtion may change, such as ionic strength and rela- and that of artiﬁcially surface-induced denatura-tive composition of solutes due to selective tion. The surface was introduced by shaking thecrystallization. These changes may potentially protein solution containing hydrophobic Teﬂondestabilize a protein. beads. In a similar study, a correlation coefﬁcient Generally, lowering the temperature reduces the of 0.93 was found between the tendency of freezerate of chemical reactions. However, chemical denaturation and surface-induced denaturationreactions may actually accelerate in a partially for eight model proteins, including aldolase, basicfrozen aqueous solution due to increased solute ﬁbroblast growth factor (bFGF), GDH, IL-1ra,concentration (Pikal, 1999). Due to solute concen- LDH, maleate dehydrogenase (MDH), PFK, andtration, the rate of oligomerization of b-glutamic TNFbp (Kendrick et al., 1995b). However, thereacid at −20°C was much faster than at 0 or 25°C was no signiﬁcant correlation (r=0.78) betweenin the presence of a water-soluble carbodiimide, freeze denaturation and thermal denaturation1-ethyl-3-(3-dimethylaminopropyl) carbodiimide temperature (Chang et al., 1996b).(EDAC) (Liu and Orgel, 1997). The increase in the rate of a chemical reaction 2.2.4. pH changes during freezingin a partially frozen state could reach several Many proteins are stable only in a narrow pHorders of magnitude relative to that in solution range, such as low molecular weight urokinase(Franks, 1990, 1994). (LMW-UK) at pH 6–7 (Vrkljan et al., 1994). At The reported oxygen concentration in a par- extreme pHs, increased electrostatic repulsion be-tially frozen solution at −3°C is as high as 1150 tween like charges in proteins tends to causetimes that in solution at 0°C (Wisniewski, 1998). protein unfolding or denaturation (Goto andThe increased oxygen concentration can readily Fink, 1989; Volkin and Klibanov, 1989; Dill,oxidize sulphydryl groups in proteins. If a protein 1990). Thus, the rate of protein aggregation issolution contains any contaminant proteases, con- strongly affected by pH, such as aggregation of
W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60 5interleukin 1b (IL-1b) (Gu et al., 1991), human separation due to polymers’ altered solubilities atrelaxin (Li et al., 1995a), and bovine pancreatic low temperatures. Freezing-induced phase separa-RNase A (Townsend and DeLuca, 1990; Tsai et tion can easily occur in a solution containing twoal., 1998). Moreover, the solution pH can signiﬁ- incompatible polymers such as dextran and Ficollcantly affect the rate of many chemical degrada- (Izutsu et al., 1996). During freezing of recombi-tions in proteins (Wang, 1999). nant hemoglobin in a phosphate buffer containing Freezing a buffered protein solution may selec- 4% (w/w) PEG 3350, 4% (w/w) dextran T500, andtively crystallize one buffering species, causing pH 150 mM NaCl, liquid–liquid phase separationchanges. Na2HPO4 crystallizes more readily than occurred and created a large excess of interface,NaH2PO4 because the solubility of the disodium denaturing the protein (Heller et al., 1997). Addi-form is considerably lower than that of the tion of 5% sucrose or trehalose could not reversemonosodium form. Because of this, a sodium the denaturation effect in the system (Heller et al.,phosphate buffer at pH 7 has a molar [NaH2PO4]/ 1999a).[Na2HPO4] ratio of 0.72, but this ratio increases Several strategies have been proposed to miti-to 57 at the ternary eutectic temperature during gate or prevent phase separation-induced proteinfreezing (Franks, 1990, 1993). This can lead to a denaturation during freezing. These include use ofsigniﬁcant pH drop during freezing, which then alternative salts (Heller et al., 1999a), adjustmentdenatures pH-sensitive proteins. For example, of the relative composition of polymers to avoidfreezing of a LDH solution caused protein denat- or to rapidly pass over a temperature regionuration due to a pH drop from 7.5 to 4.5 upon where the system may result in liquid–liquidselective crystallization of Na2HPO4 (Anchordo- phase separation (Heller et al., 1999c), and chemi-quy and Carpenter, 1996). LDH is a pH-sensitive cal modiﬁcation of the protein such as pegylationprotein and a small drop in pH during freezing (Heller et al., 1999b).can partially denature the protein even in thepresence of stabilizers such as sucrose and tre- 2.2.6. Dehydration stresseshalose (Nema and Avis, 1992). The pH drop Proteins in an aqueous solution are fully hy-during freezing may also explain why freezing drated. A fully hydrated protein has a monolayerbovine and human IgG species in a sodium phos- of water covering the protein surface, which isphate buffer caused formation of more aggregates termed the hydration shell (Rupley and Careri,than in potassium phosphate buffer, because 1991). The amount of water in full hydration ispotassium phosphate buffer does not show signiﬁ- 0.3–0.35 g g − 1 protein (Rupley and Careri, 1991;cant pH changes during freezing (Sarciaux et al., Kuhlman et al., 1997). Generally, the water con-1998). tent of a lyophilized protein product is less than The pH drop during freezing can potentially 10%. Therefore, lyophilization removes part ofaffect storage stability of lyophilized proteins. the hydration shell. Removal of the hydrationLyophilized IL-1ra in a formulation containing shell may disrupt the native state of a protein andphosphate buffer at pH 6.5 aggregated more cause denaturation. A hydrated protein, whenrapidly than that containing citrate buffer at the exposed to a water-poor environment during de-same pH during storage at 8, 30 and 50°C (Chang hydration, tends to transfer protons to ionizedet al., 1996c). Similarly, the pH drop of a succi- carboxyl groups and thus abolishes as manynate-containing formulation from 5 to 3 –4 during charges as possible in the protein (Rupley andfreezing appeared to be the cause of less storage Careri, 1991). The decreased charge density maystability for lyophilized IFN-g than that contain- facilitate protein–protein hydrophobic interac-ing glycocholate buffer at the same pH (Lam et tion, causing protein aggregation.al., 1996). Water molecules can also be an integral part of an active site(s) in proteins. Removal of these2.2.5. Phase separation during freezing functional water molecules during dehydration Freezing polymer solutions may cause phase easily inactivates proteins. For example, dehydra-
6 W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60tion of lysozyme caused loss of activity apparently Lyophilization may induce several potentialdue to removal of those water molecules residing changes in the IR spectra of proteins. Disruptionfunctionally in the active site (Nagendra et al., of hydrogen bonds in proteins during lyophiliza-1998). tion generally leads to an increase in frequency Lastly, dehydration during lyophilization may and a decrease in intensity of hydroxyl stretchingcause signiﬁcant difference in moisture distribu- bands (Carpenter and Crowe, 1989). Unfolding oftion in different locations of a product cake. The proteins during lyophilization broadens and shiftsuneven moisture distribution may lead to possible (to higher wave numbers) amide I componentlocalized overdrying, which may exacerbate dehy- peaks (Prestrelski et al., 1993b; Allison et al.,dration-induced protein denaturation (Pikal and 1996). Lyophilization often leads to an increase inShah, 1997). b-sheet content with a concomitant decrease in a-helix content. Conversion of a-helix to b-sheet2.3. Monitoring protein denaturation upon during lyophilization has been observed in manylyophilization proteins such as tetanus toxoid (TT) in 10 mM The most common method for monitoring sodium phosphate buffer (pH 7.3) (Costantino etprotein denaturation upon lyophilization appears al., 1996), recombinant human albumin (rHA) into be infrared (IR) spectroscopy, although other different buffer solutions at different pHsmethods have been used such as mass spec- (Costantino et al., 1995a), hGH at pH 7.8, andtroscopy (Bunk, 1997), and Raman spectroscopy seven model proteins in water, including bovine(Belton and Gil, 1994). In the following section, pancreatic trypsin inhibitor (BPTI), chymotrypsi-IR methodology is discussed in monitoring nogen, horse myoglobin (Mb), horse heart cy-protein denaturation upon lyophilization followed tochrome c (Cyt c), rHA, porcine insulin, andby a discussion on reversibility of protein RNase A (Griebenow and Klibanov, 1995).denaturation. An increase in b-sheet content during lyophilization is often an indication of protein2.3.1. Infrared (IR) spectroscopy aggregation and/or increased intermolecular inter- IR (or FTIR) is probably the most extensively action (Yeo et al., 1994; Griebenow andused technique today for studying structural Klibanov, 1995; Overcashier et al., 1997).changes in proteins upon lyophilization (Susi and Lyophilization-induced increase in b-sheet contentByler, 1986; Dong et al., 1995; Carpenter et al., seems to be a rather general phenomenon as1998, 1999). The lyophilization-induced structural lyophilization or air-drying of unordered poly-L-changes can be monitored conveniently in the lysine induced structural transition to a highlyamide I, II, or III region. For lyophilized protein ordered b-sheet (Prestrelski et al., 1993b; Wolkerssamples, residual water up to 10% (w/w) does not et al., 1998b). Such transition has also been ob-interfere signiﬁcantly in the amide I region, a served in proteins during lyophilization such asfrequently used sensitive region for determination human insulin in water (pH 7.1) (Pikal and Rigs-of secondary structures (Dong et al., 1995). How- bee, 1997). The b-sheet structure after lyophiliza-ever, IR studies on proteins in an aqueous solu- tion shows a higher degree of intermoleculartion need either subtraction of water absorptionor solvent replacement with D2O (Goormaghtigh hydrogen bonding because polar groups must sat-et al., 1994). To make reliable subtraction, high isfy their H-bonding requirement by intra- orprotein concentrations ( 10 mg ml − 1) are rec- intermolecular interaction upon removal of water.ommended to increase protein absorption signal, The intermolecular b-sheet is characterized by twoand a CaF2 (or BaF2) cell with a path length of 10 major IR bands at about 1617 and 1697 cm − 1 inmm or less should be used to control the total solid state, which can be used to monitor proteinsample absorbance within 1 (Cooper and Knut- denaturation (Allison et al., 1996). Similarly, theson, 1995). relative intensity of a-helix band also can be used
W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60 7in this regard (Yang et al., 1999; Heller et al., upon lyophilization (Carpenter et al., 1993), while1999b). loss of BO activity in a PVA-containing formula- The extent of changes in overall IR spectrum of tion was at least partially reversible (Nakai et al.,a protein upon lyophilization reﬂects the degree of 1998). Using IR spectroscopy, Prestrelski et al.protein denaturation. The changes relative to a (1993b) demonstrated that lyophilization-inducedreference spectrum can be measured using a corre- structural changes were irreversible for bFGF,lation coefﬁcient (r) as deﬁned by Prestrelski et al. IFN-g, and bovine a-casein, but essentially re-(1993a), or the extent of spectral area overlap versible for G-CSF and bovine a-lactalbumin.(Heimburg and Marsh, 1993; Allison et al., 1996; The extensive aggregation and precipitation ofKendrick et al., 1996). Using the correlation co- IFN-g and casein upon rehydration conﬁrmed theefﬁcient, Prestrelski et al. (1993b) were able to irreversibility in structural changes. Therefore,measure the relative freeze-drying stability of sev- lyophilization of proteins may lead to three typeseral model proteins, including bFGF, bovine a- of behavior, (1) no change in protein conforma-lactalbumin, bovine a-casein, IFN-g, and tion; (2) reversible denaturation; or (3) irreversiblerecombinant granulocyte colony-stimulating fac- denaturation.tor (rG-CSF) in the presence of different sugars. In many cases, IR-monitored structural changesNevertheless, Griebenow and Klibanov (1995), during lyophilization seem to be reversible.after analyzing secondary structures of seven Griebenow and Klibanov (1995) demonstratedmodel proteins upon lyophilization, concluded that lyophilization (dehydration) caused signiﬁ-that the correlation coefﬁcient was not highly cant changes in the secondary structures of sevensensitive to structural alterations in proteins. In- model proteins in the amide III region (1220–stead, comparison of overlapping area-normalized 1330 cm − 1), including BPTI, chymotrypsinogen,second-derivative or deconvoluted spectra seemed Mb, Cyt c, rHA, insulin, and RNase A. Themore reliable and objective. structure of almost all proteins became more or- Recently, IR has been used in real-time moni- dered upon lyophilization with a decrease in thetoring of freezing and dehydration stresses on unordered structures. Nevertheless, all these struc-proteins during lyophilization. By this method, tural changes were reversible upon reconstitution.glucose at 10% was shown to protect lysozyme Other examples of reversible changes in the sec-both during the freezing and drying processes ondary structures of proteins upon lyophilization(Remmele et al., 1997). include rHA (Costantino et al., 1995a), Humicola lanuginosa lipase (Kreilgaard et al., 1999), IL-22.3.2. Re6ersibility of freezing- or (Prestrelski et al., 1995), and lysozyme (Allison etlyophilization-induced protein denaturation al., 1999). Many proteins denature to various extentsupon freezing, especially at low concentrations( B 0.1 mg ml − 1). Freezing-induced denaturation 3. Cryo- and lyo-protection of proteins bymay or may not be reversible. Freezing lysozyme stabilizersor IL-1ra caused reversible denaturation(Kendrick et al., 1995a). In contrast, recombinant As discussed before, both freezing and dehydra-factor XIII (rFXIII, 166 kD) was irreversibly tion can induce protein denaturation. To protect adenatured upon freezing, and loss of native protein from freezing (cryoprotection) and/or de-rFXIII at 1 mg ml − 1 increased linearly with the hydration (lyoprotection) denaturation, a proteinnumber of freeze –thaw cycles (Kreilgaard et al., stabilizer(s) may be used. These stabilizers are1998b). also referenced as chemical additives (Li et al., Similarly, lyophilization-induced denaturation 1995b), co-solutes (Arakawa et al. 1993), co-sol-can be either reversible or irreversible. In the vents (Timasheff, 1993, 1998), or excipientsabsence of stabilizers, PFK at 25 mg ml − 1 at pH (Wong and Parascrampuria, 1997; Wang, 1999).7.5 and 8.0 was fully and irreversibly inactivated In the following section, a variety of protein
8 W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60stabilizers are presented for cryo- and lyo-protec- For example, freezing rabbit muscle LDH in wa-tion, followed by discussions of their possible ter caused 64% loss of protein activity, and in thestabilization mechanisms. presence of 5, 10 or 34.2% sucrose, the respective losses were 27, 12, and 0% (Nema and Avis,3.1. Stabilizers for cryo- and lyo-protection 1992). Other sugars or polyols that can protect LDH during freeze-thawing to different degrees Nature protects life from freezing or osmotic include lactose, glycerol, xylitol, sorbitol, andshock by accumulating selected compounds to mannitol, at 0.5–1 M (Carpenter et al., 1990).high concentrations ( 1 M) within organisms. Increasing trehalose concentration gradually in-These accumulated compounds are known as cry- creased the recovery of PFK activity duringoprotectants and osmolytes, which are preferen- freeze-thawing and the recovery reached a maxi-tially excluded from surfaces of proteins and act mum of 90% at about 300 mg ml − 1 (Carpenter etas structure stabilizers (Timasheff, 1993). How- al., 1990). A similar stabilizing trend was alsoever, since the dehydration stress is different from observed for sucrose, maltose, glucose, or inositolthose of freezing, many effective cryoprotectants (Carpenter et al., 1986).or protein stabilizers in solution do not stabilize Since freezing is part of the freeze-drying pro-proteins during dehydration (drying). Some even cess, high concentrations of sugars or polyols aredestabilize proteins during lyophilization. For ex- often necessary for lyoprotection. These examplesample, CaCl2 stabilized elastase (20 mg ml − 1) in include the protection of chymotrypsinogen in the10 mM sodium acetate (pH 5.0), but caused the presence of 300 mM sucrose (Allison et al., 1996),lyophilized protein cake to collapse and lose activ- complete inhibition of acidic ﬁbroblast growthity (Chang et al., 1993). factor (aFGF) aggregation by 2% sucrose (Volkin Similarly, effective lyophilization stabilizers (ly- and Middaugh, 1996), increase in glucose-6-phos-oprotectants) may or may not stabilize proteins phate dehydrogenase (G6PDH) activity from 40effectively during freezing. Therefore, in cases to about 90% by 5.5% sugar mixture (glu-when a single stabilizer does not serve as both a cose:sucrose=1:10, w/w) (Sun et al., 1998), com-cryoprotectant and a lyoprotectant, two (or more) plete recovery of LDH by either 7% sucrose or 7%stabilizers may have to be used to protect proteins rafﬁnose, a trisaccharide (Moreira et al., 1998),from denaturation during lyophilization. signiﬁcant improvement of PFK recovery by 400 mM trehalose (Carpenter et al., 1993), and com-3.1.1. Sugars/polyols plete protection of four restriction enzymes by Many sugars or polyols are frequently used 15% trehalose (Colaco et al., 1992). More exam-nonspeciﬁc protein stabilizers in solution and dur- ples can be found in Table 2.ing freeze-thawing and freeze-drying. They have Lower concentrations of sugars or polyols maybeen used both as effective cryoprotectants and or may not have any signiﬁcant effect. At 5 to 100remarkable lyoprotectants. In fact, their function mM, neither trehalose nor glucose could protectas lyoprotectants for proteins has long been LDH or PFK to a signiﬁcant level duringadopted by nature. Anhydrobiotic organisms (wa- lyophilization (Carpenter et al., 1993). To deter-ter content B1%) commonly contain high con- mine the minimum sugar concentration that offerscentrations (up to 50%) of disaccharides, the maximum stabilization effect, Tanaka et al.particularly sucrose or trehalose, to protect them- (1991) studied the lyoprotective effect of saccha-selves (Crowe et al., 1992, 1998). rides on the denaturation of catalase during The level of stabilization afforded by sugars or lyophilization. They demonstrated that saccha-polyols generally depends on their concentrations. rides protected the protein by direct interactionA concentration of 0.3 M has been suggested to with the protein and a concentration of saccha-be the minimum to achieve signiﬁcant stabiliza- rides sufﬁcient to form a monomolecular layer ontion (Arakawa et al., 1993). This has been found the protein surface was the minimum to achieveto be true in many cases during freeze-thawing. the maximum stabilization. Therefore, the stabi-
W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60 9lization of catalase was found to depend not on buffer (pH 7.4) (Table 2), mannitol for LDH in 50the bulk concentration of maltose but on the mM sodium phosphate buffer (pH 7.4) (Izutsu etweight ratio of maltose to catalase (Tanaka et al., al., 1994b), and myo-inositol for PFK during1991). Maximum stabilization of catalase was at a freeze-thawing and freeze-drying (Table 2). Again,ratio of about 0.4. A recent study showed that the decreased protein recovery is probably due tomaximum protection (about 75% recovery) of L- crystallization of these excipients at highasparaginase at 1.45 mg ml − 1 during lyophiliza- concentrations.tion was reached at a saccharide concentration of The level of protein protection afforded byabout 0.5 mg ml − 1, which was about the calcu- different sugars or polyols can be either similar orlated monosaccharide concentration required to signiﬁcantly different, depending on the formula-interact with all exposed highly polar residues of tion composition, concentration and physicalthe protein (Ward et al., 1999). At this concentra- properties of the stabilizer, and its compatibilitytion, the weight ratio of saccharide to L-asparagi- with the protein. Ward et al. (1999) found thatnase is 0.34, which is coincidentally very close to several saccharides, including trehalose, lactose,that of maltose to catalase. maltose, sucrose, glucose, and mannitol, displayed On the other hand, increasing sugar/polyol con- similar level of protection towards tetrameric L-centration to a certain level may eventually reach asparaginase (1.45 mg ml − 1) during lyophiliza-a limit of stabilization or even destabilize a tion at saccharide concentrations up to 0.1%. Atprotein during freeze-drying. For example, actin 2%, glucose or lactose protected L-asparaginasewas maximally stabilized during lyophilization in from dissociation during freeze-drying, but man-the presence of 5% (w/v) sucrose and a further nitol did not, possibly due to its crystallizationincrease in sucrose concentration to 10% did not and loss of intimate interaction with the proteinimprove the protein stability signiﬁcantly, which (Adams and Ramsay, 1996). Probably for thewas apparently attributable to sticky, pliable, and same reason, mannitol at 88 mM inhibited thecollapsed formulation structure (Allison et al., formation of insoluble hGH aggregates in phos-1998). Increasing trehalose concentration to 150 phate buffer (pH 7.4) at a freezing rate of 50°Cmg ml − 1 in a PFK formulation (at 50 mg ml − 1, min − 1, but accelerated hGH aggregation at lowerpH 8.0) increased the freeze-drying recovery of freezing rates of 0.5 and 5°C min − 1 (Eckhardt etPFK activity to about 65%, but further increases al., 1991). In a different study, however, Tanakain trehalose concentration caused a gradual de- et al. (1991) demonstrated that both mannitol andcrease in recovery of the protein activity (Carpen- sorbitol could increase the recovery of catalaseter and Crowe, 1989). At a trehalose activity during lyophilization to a similar level asconcentration of 400 mg ml − 1, basically no PFK that offorded by maltose. They also showed thatactivity was left after freeze-drying. Since tre- different sugars (maltose, glucose, and mal-halose at 400 mg ml − 1 protected about 90% of totriose) at 1 mg ml − 1 could increase the recoverythe protein activity after freeze-thawing, the desta- of catalase activity to the same level (from 35 tobilization of PFK at high concentrations of tre- 90%), but maltopentaose, maltohexaose, and mal-halose occurred in the dehydration step, possibly toheptaose were not as effective (Tanaka et al.,due to crystallization of trehalose, preventing req- 1991). The ineffectiveness of larger saccharidesuisite hydrogen bonding to the dried protein (see suggests that protein stabilization by sugars maySection 3.2) (Carpenter and Crowe, 1989). A sim- depend on their glucoside chain lengths, and ailar trend was observed in the stabilization of long chain length may interfere with intermolecu-several other proteins during lyophilization in the lar hydrogen-bonding between stabilizing sugarspresence of increasing concentrations of excipi- and proteins.ents, including mannitol for L-asparaginase (10 mg In many cases, disaccharides appear to be theml − 1) in 50 mM sodium phosphate buffer (pH most effective and universal stabilizers among7.4) (Izutsu et al., 1994b), mannitol for b-galac- sugars and polyols (Arakawa et al., 1993; Carpen-tosidase (2 mg ml − 1) in 10 mM sodium phosphate ter et al., 1997). For example, the disaccharides
10 W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60trehalose, sucrose, maltose, and lactose, were all 1998b), while sucrose was a better stabilizer thanessentially equivalent to or more effective than trehalose during freeze-drying of Humicola lanugi-monosaccharides such as glucose in stabilizing nosa lipase, a hydrophobic protein (Kreilgaard etPFK during lyophilization (Crowe et al., 1993b). al., 1999). Trehalose at 20 mg mg − 1 protein wasTrehalose at 400 mM increased the recovery of more effective than sucrose in stabilizing H+-AT-PFK activity to greater than 60% during Pase during lyophilization, but at 5 10 mg mg − 1lyophilization whereas glucose at the same con- protein, sucrose was more effective (Sampedro etcentration only recovered less than 5% of the al., 1998). In stabilizing G6PDH duringprotein activity (Carpenter et al., 1993). Similarly, lyophilization, both the glucose/trehalose (1:10,the activity of H+-ATPase upon lyophilization w/w) and glucose/sucrose (1:10, w/w) systemswas increased from 4 to 100, 91, and 84% in the were shown to be equally effective (Sun andpresence of disaccharides trehalose, maltose, and Davidson, 1998).sucrose, respectively, but only 72 and 37% in the Not all proteins can be stabilized by sugars/presence of monosaccharides glucose and galac- polyols. This is still an unsolved puzzle (Carpentertose at 20 mg sugar per mg protein (Sampedro et et al., 1999). For example, sucrose at concentra-al., 1998). tions from 0.1 to 0.5 M showed little effect on the Among disaccharides, sucrose and trehalose ap- aggregation of recombinant hemoglobin in PBSpear to be the most commonly used. In compari- during freeze–thaw cycles (Kerwin et al., 1998).son to sucrose, trehalose seems to be a preferable Addition of 5% sucrose in MN12 formulation didlyoprotectant for biomolecules, because it has a not show signiﬁcant stabilizing effect duringhigher glass transition temperature (Crowe et al., lyophilization (Ressing et al., 1992). Trehalose at1992, 1996). The higher glass transition tempera- 5% actually increased the loss of LDH activity inture of trehalose arises at least partly from the water from 64 to 74% during freezing (Nema andformation of trehalose – protein – water microcrys- Avis, 1992). Although the pH of the trehalosetals, preventing water plasticizing the amorphous solution decreased during freezing, the pH changephase (Librizzi et al., 1999). Other properties of during freezing could not explain the destabiliza-trehalose are also considered to be advantageous, tion of LDH by trehalose because sucrose, whichwhich include (1) less hygroscopicity, (2) an ab- stabilized LDH, also caused the same pH change.sence of internal hydrogen bonds, which allows Therefore, the type of sugar and its subunit orien-more ﬂexible formation of hydrogen bonds with tation might have caused the difference in stabiliz-proteins, and (3) very low chemical reactivity ing LDH (Carpenter et al., 1986).(Roser, 1991). To support these arguments, Roser In a few cases, sugars have to be used with(1991) demonstrated that 35 air-dried restriction another excipient(s) to achieve satisfactory proteinand DNA-modifying enzymes are maximally sta- stabilization. Carpenter et al. (1986) demonstratedbilized by 0.3 M trehalose in comparison to other that freezing rabbit skeletal muscle PFK in liquidnon-reducing sugars, including sucrose, sorbitol, nitrogen for 30 s completely inactivated themannitol, galactitol, etc. as well as reducing sug- protein. Inclusion of 1 mM ZnSO4 or 50 mMars, including glucose, mannose, galactose, mal- sugars (trehalose, sucrose, or maltose) helped totose, lactose, etc. These advantages of using retain less than 13 or 10% of the initial proteintrehalose were later challenged by Levine and activity after freeze-thawing, while a combinationSlade (1992), who contended that sucrose could of 1 mM ZnSO4 and 50 mM trehalose (sucrose orbe equally effective in protecting biomolecules. In maltose) resulted in retention of more than 80%reality, the relative stabilization effect of these two protein activity. More than 85% of protein activ-sugars seems to be depend on both the protein ity was recovered when ZnSO4 was used withand sugar concentration. For example, trehalose glucose or inositol. Thus, sugars and metal ionsat 30 mg ml − 1 was more effective in inhibiting had a synergistic effect in stabilizing PFK duringIL-6 aggregation during lyophilization than su- freezing. Similarly, neither 10 mM sugar (tre-crose at the same concentration (Lueckel et al., halose, lactose or mannitol) nor 1% PEG could
W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60 11improve the lyophilization recovery (46%) of In addition to albumin, other polymers alsoLDH at 2 mg ml − 1 in phosphate buffer (pH 7.5). have been used. The level of protein stabilizationHowever, combined use of 1% PEG and 10 mM afforded by these polymers depends on structurelactose completely protected the protein from in- and concentration of both the polymer and theactivation (Prestrelski et al., 1993a). Also, com- protein. For example, dextran (5%), PVA (2.5%),bined use of 1% PEG and sugar ( 25 mM hydroxypropyl methylcellulose (HPMC) (1%), ortrehalose or glucose) almost completely protected gelatin (0.5%) reduced the loss of rabbit musclethe activity of PFK during lyophilization (Car- LDH activity in water during freezing from 64 topenter et al., 1993). PEG in these formulations 24%, 24, 18, and 9%, respectively (Nema andserved as a lyoprotectant, while the sugars were Avis, 1992). LDH activity was also protectedused against dehydration denaturation. during lyophilization in the presence of different concentrations of polyethyleneimine (Andersson and Hatti-Kaul, 1999). While ovalbumin at 0.01%3.1.2. Polymers had little effect on the stability of catalase (8.4 mg Polymers have been used to stabilize proteins in ml − 1 in 10 mM phosphate buffer, pH 7.0) duringsolution and during freeze-thawing and freeze- freezing, gelatin at the same concentration com-drying (Arakawa et al., 1993). One of the favor- pletely protected the protein activity (Shikamaable polymers used in the history of protein drug and Yamazaki, 1961). Polyvinylpyrrolidone (PVP)development was serum albumin. It has been used (40 kD) increased both the freeze-thawing andboth as a cryoprotectant and lyoprotectant. For freeze-drying recovery of LDH in a concentra-example, bovine serum albumin (BSA) at 1% tion-dependent manner (Anchordoquy and Car-completely protected the activity of rabbit muscle penter, 1996). Addition of 2% dextran (192 kD)LDH in water during freezing (Nema and Avis, into a sucrose-containing actin formulation sig-1992). At much lower concentrations between niﬁcantly increased the protein stability during0.05 and 0.1% (w/v), human serum albumin lyophilization (Allison et al., 1998). Hydroxyethyl(HSA), due to its effective inhibition of protein cellulose (HEC) at 1% completely inhibitedsurface adsorption and general stabilization of lyophilization-induced aggregation of aFGF atproteins during lyophilization, was used in formu- 100 mg ml − 1 in PBS containing 33 mg ml − 1lating freeze-dried hydrophobic cytokines, such as heparin, although reconstitution time was in-interleukin-1a (IL-1a), IL-1b, IL-3, and creased signiﬁcantly (Volkin and Middaugh,macrophage colony stimulating factor (MCSF) 1996).(Dawson, 1992). Increasing BSA concentrations Stabilization of proteins by polymers can gener-to 0.05% gradually increased the activity recovery ally be attributed to one or more of these polymerof LDH at 25 mg ml − 1 from about 30 to 100% properties: preferential exclusion, surface activity,during freeze-thawing and to about 80% during steric hindrance of protein–protein interactions,freeze-drying (Anchordoquy and Carpenter, and/or increased solution viscosity limiting1996). Many protein products on the market, protein structural movement. In recent years, ad-such as Betaseron®, Epogen®, Kogenate®, and ditional properties of polymers have been impli-Recombinate™ contain albumin (Physicians’ cated in stabilizing proteins during freeze-thawingDesk Reference, 1999). However, the ever-increas- and freeze-drying. Polymers such as dextran haveing concern about the potential contamination of been reported to stabilize proteins by raising theserum albumin with blood-borne pathogens limits glass transition temperature of a protein formula-its future application in protein products. There- tion signiﬁcantly and by inhibiting crystallizationfore, rHA has been recommended recently to of small stabilizing excipients such as sucrosereplace serum albumin as a protein stabilizer (Skrabanja et al., 1994). PEG 3350 or dextran(Tarelli et al., 1998). Nevertheless, the ultimate T500 at 4% (w/w) has been found to inhibit a pHsolution is to develop albumin-free formulations drop during freezing of a phosphate-buffered so-for protein pharmaceuticals. lution by inhibiting crystallization of disodium
12 W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60phosphate (Heller et al., 1997). Probably by the 1996). For example, increasing initial concentra-same mechanism, BSA or PVP (40 kD) at 10% tion of rhFXIII from 1 to 10 mg ml − 1 increaseddramatically inhibited the pH drop during freez- the recovery of native rhFXIII during repeateding of a buffered LDH solution (Anchordoquy freeze-thawing (Kreilgaard et al., 1998b). The re-and Carpenter, 1996). At least partly due to this covery of LDH activity gradually increased frominhibition effect, both BSA and PVP increased the 6% at a protein concentration of 10 mg ml − 1 tofreeze –thaw recovery of the protein in a concen- about 65% at concentrations above 175 mg ml − 1tration-dependent manner. The inhibition of crys- after freeze-thawing (Carpenter et al., 1990). Uptallization of small molecules is apparently due to to about 90% LDH activity was recovered whenpolymer-induced viscosity increase (Slade et al., the concentration was increased to 500 mg ml − 11989). (Anchordoquy and Carpenter, 1996). Koseki et al. On the other hand, polymers may cause phase (1990) demonstrated that increasing the ovalbu-separation during freezing, adversely affecting min concentration in the range 0.5–2.5 mg ml − 1protein stability (see Section 2.2). Certain poly- at pH 1.9 decreased the structural changes of themers may destabilize proteins during lyophiliza- freeze-treated (−40°C) protein, as measured bytion due to steric hindrance, preventing efﬁcient UV.hydrogen bonding with proteins. Dextran (40 kD) Similarly, the lyophilization recovery of PFKat concentrations of up to 100 mg ml − 1 failed to activity at 25 and 40 mg ml − 1 was 34 and 64%,inhibit dehydration-induced unfolding of respectively, in the presence of 200 mM trehaloselysozyme because of its inability to form adequate (Carpenter et al., 1987). Increasing the concentra-hydrogen bonding with the protein (Allison et al., tion of rabbit muscle LDH from 10 to 500 mg1999). Similarly, this compound could hardly pre- ml − 1 gradually increased the activity recoveryvent formation of b-sheets in poly-L-lysine during from less than 20% to about 60% duringdehydration (Wolkers et al., 1998b). In fact, Dex- lyophilization (Anchordoquy and Carpenter,tran (162 kD) at 5% (w/v) was shown to destabi- 1996). Increasing the concentration of bovine andlize Humicola lanuginosa lipase during human IgG species markedly decreasedlyophilization, as determined by IR (Kreilgaard et lyophilization-induced protein aggregation (Sar-al., 1999). ciaux et al., 1998). Certain proteins, however, do not show this concentration-dependent protec-3.1.3. Protein itself tion. The percentage of lyophilization-induced de- Protein aggregation in solution is generally con- naturation of catalase in the absence of acentration-dependent. It has been suggested that stabilizer was determined to be about 65%, inde-increasing protein concentration to higher than pendent of the protein concentration in the range0.02 mg ml − 1 may facilitate potential protein 1–5000 mg ml − 1 (Tanaka et al., 1991).aggregation (Ruddon and Bedows, 1997). Increas- The mechanisms of proteins’ self-stabilizationing protein concentration increases aggregation of during freezing and/or lyophilization have notmany proteins in solution, such as LMW-UK in been clearly delineated. Proteins are polymers,the range 0.2–0.9 mg ml − 1 (Vrkljan et al., 1994), and therefore, at least some of the above-dis-IL-1b in the range 100 – 500 mg ml − 1 (Gu et al., cussed stabilization mechanisms for polymers may1991), apomyoglobin in the range 4 – 12 mg ml − 1 be applicable to proteins’ self-stabilization. Re-in the presence of 2.4 M urea (De Young et al., cently, two hypotheses have been reiterated to1993), and insulin (Brange et al., 1992a). explain the concentration-dependent protein sta- In contrast, proteins at higher concentrations bilization upon freezing (Allison et al., 1996).are often more resistant against both freezing- First, unfolding of proteins at high concentrationsand lyophilization-induced protein denaturation/ during freezing may be temporarily inhibited byaggregation. The activity recovery of many labile steric repulsion of neighboring protein molecules.proteins after freeze-thawing correlates directly Second, the surface area of ice-water interfacewith initial protein concentration (Allison et al., formed upon freezing is ﬁnite, which limits the
W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60 13amount of protein to be accumulated and dena- the driving force of protein adsorption and/ortured at the interface. In addition, favorable aggregation at these interfaces. Low concentra-protein–protein interactions (possible formation tions of nonionic surfactants are often sufﬁcientof dimers or multimers) may contribute to the to serve this purpose due to their relatively lowincreased protein stability at high concentrations, critical micelle concentrations (CMC) (Bam et al.,as observed for thermophilic proteins (Mozhaev 1995). Other stabilization mechanisms were alsoand Martinek, 1984). proposed, such as assistance in protein refolding during thawing and protein binding, which may3.1.4. Non-aqueous sol6ents inhibit protein–protein interactions (Carpenter et Non-aqueous solvents generally destabilize al., 1999).proteins in solution. At low concentrations certain Tween 80 is one of the commonly used surfac-non-aqueous solvents may have a stabilizing ef- tants for protein stabilization during freezing.fect. These stabilizing non-aqueous solvents in- Tween 80 at concentrations of ] 0.01% protectedclude polyhydric alcohols such as PEGs, ethylene both LDH and GDH from denaturation duringglycol, and glycerol and some polar and aprotic quench freezing and thawing (Chang et al.,solvents such as dimethylsulphoxide (DMSO) and 1996b). The freeze–thaw recovery of LDH activ-dimethylformamide (DMF) (Volkin and ity was increased from 36 to 57 and 65% in theKlibanov, 1989; Carpenter et al., 1991). presence of 0.002 and 0.005% Tween 80, respec- In fact, polyhydric alcohols are among the com- tively (Nema and Avis, 1992). Maximum freeze–monly used and effective cryoprotectants. For thaw recovery (about 80%) of LDH activity wasexample, in the presence of 0.2 M PEG 400, the reached in the presence of 0.05% Tween 80.loss of rabbit muscle LDH activity upon freezing Tween 80 at concentrations from 0.005 to 0.01%was reduced from 64 to 15% (Nema and Avis, also protected several other proteins from freezing1992). LDH can also be protected from freeze- denaturation, including TNFbp, IL-1ra, bFGF,thawing denaturation to different degrees by eth- MDH, aldolase, and PFK (Kendrick et al.,ylene glycol or 2-methyl-2,4-pentanediol 1995b).(Carpenter et al., 1990). PEG at 1 – 10% (w/v) Other nonionic and ionic surfactants have alsocompletely protected both LDH and PFK at 25 been reported in cryoprotection of proteins. Themg ml − 1 (at pH 7.5 and 8.0, respectively) during following surfactants protected LDH from freez-freeze-thawing, although they were not effective ing denaturation to various degrees, Brij 35; Brijstabilizers during freeze-drying (Carpenter et al., 30 (polyoxyethylene lauryl ether); Lubrol-px; Tri-1993). Glycerol at 0.3% (v/v) prevented freezing ton X-10; Pluronic F127 (polyoxyethylene-poly-denaturation of ovalbumin (0.1%) (Koseki et al., oxypropylene copolymer); and SDS (Nema and1990) and at 1 M, increased the recovery of Avis, 1992; Chang et al., 1996b). SDS at 0.5 mMcatalase activity upon freezing from 80 to 95% increased the activity recovery of catalase (8.4 mg(Shikama and Yamazaki, 1961). ml − 1 in 10 mM phosphate buffer, pH 7.0) from Cryoprotection of proteins by these non- 80 to 90% upon freezing (Shikama and Yamazaki,aqueous solvents may be pH-dependent. Ethylene 1961).glycol stabilized RNase A at pH 2.3 but destabi-lized it at pH 5.5 (Arakawa et al., 1991). This is 3.1.6. Amino acidspartly because proteins may tolerate freezing de- Certain amino acids can be used as cryoprotec-naturation to different degrees at different pHs. tants and/or lyoprotectants. For example, freezing rabbit skeletal muscle PFK in liquid nitrogen for3.1.5. Surfactants 30 s inactivated the protein completely, and sev- The formation of ice-water interfaces during eral amino acids, including glycine, proline, or 4freezing may cause surface denaturation of hydroxyproline, signiﬁcantly increased the recov-proteins (see Section 2.2). Surfactants may drop ery of the protein activity (Carpenter et al., 1986).surface tension of protein solutions and reduce Glycine at low concentrations has been shown to
14 W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60suppress the pH change in 10 or 100 mM sodium Recently, Ramos et al. (1997) demonstratedphosphate buffer during freezing (Pikal and Car- that 2-O-b-mannosylglycerate at 500 mM in-penter, 1998). Therefore, amino acids may protect creased the freeze-drying recovery of LDH activ-proteins from freezing denaturation at least partly ity (at 50 ug ml − 1) from 12 to 85% whileby reducing the rate and extent of buffer salt trehalose only increased the recovery to 54%.crystallization. As lyoprotectants, several amino acids in- 3.2. Mechanisms of protein stabilization duringcreased the lyophilization recovery of LDH from lyophilization22 to about 39–100%, including proline, L-serine,sodium glutamate, alanine, glycine, lysine hydro- Since freezing and drying stresses imposed onchloride, sarcosine, g-aminobutyric acid (Carpen- proteins during lyophilization are different, mech-ter et al., 1990). Glycine alone or in combination anisms of protein stabilization by excipients arewith mannitol inhibited aggregation of an anti- not the same in the two stages of lyophilization.body-vinca conjugate during lyophilization (Royet al., 1992). LDH activity was increased by 20% 3.2.1. Mechanisms of cryoprotectionduring vacuum-drying in the presence of pheny- One of the most widely accepted protein stabi-lalanine:arginine:H3PO4 (1:1:0.5 molar ratio) lization mechanisms in liquid state is preferential(Mattern et al., 1999). interaction. Preferential interaction means that a protein prefers to interact with either water or an3.1.7. Miscellaneous excipients excipient(s) in an aqueous solution. In the pres- Salts and amines have been used as cryoprotec- ence of a stabilizing excipient, the protein preferstants. LDH activity can be protected to various to interact with water (preferential hydration) anddegrees upon freezing in the presence of potas- the excipient is preferentially excluded from thesium phosphate, sodium acetate, ammonium sul- domain of the protein (preferential exclusion). Infate, magnesium sulfate, sodium sulfate, this case, proportionally more water moleculestrimethylamine N-oxide, or betaine (Carpenter et and fewer excipient molecules are found at theal., 1990). Increasing the potassium phosphate surface of the protein than in the bulk. Therefore,concentration from 10 mM to 1 M increased the preferential exclusion of an excipient is usuallyrecovery of LDH upon freezing from less than associated with an increase in the surface tension20% to more than 80% (Arakawa et al., 1993). of water. Detailed discussion of this stabilization Metal ions can protect certain proteins during mechanism can be found elsewhere (Arakawa etlyophilization. In the presence of 100 mM sugars al., 1991, 1993; Timasheff, 1993; Lin andsuch as trehalose, maltose, sucrose, glucose or Timasheff, 1996; Timasheff, 1998).galactose, some divalent metal ions improved the The preferential interaction mechanism appliesrecovery of PFK activity (at 40 mg ml − 1 in 1 mM equally well during freeze–thaw processes (Car-sodium borate, pH 7.8) during lyophilization in a penter et al., 1991; Arakawa et al., 1993; Crowe etconcentration-dependent manner. The relative ef- al., 1993b). Protein stabilizers, which are excludedfectiveness of these metal ions was apparently in from protein surface in solution, can also stabilizethe following order: Zn2 + Cu2 + Ca2 + , proteins during freezing. Nema and Avis (1992)Mn2 + Mg2 + (Carpenter et al., 1987). examined the stabilizing effect of 13 cryoprotec- The activity of LDH can be protected to differ- tants on the recovery of rabbit muscle LDH activ-ent degrees during lyophilization in the presence ity, including trehalose (5%), mannitol (5%),of some amphiphilic excipients, including HP-b- sucrose (5, 10, 34.2%), Brij 30 (polyoxyethyleneCD, 3-[(3-cholamidepropyl)-dimethylammonio]-1- lauryl ether, 0.05%), Tween 80 (0.002– 1%),propanesulfate (CHAPS), sodium cholate, sucrose Pluronic F127 (1%), HPMC (1%), PVP (2.5%),monolaurate (Izutsu et al., 1995). Combinations PEG 400 (0.2 M), gelatin (0.5%), BSA (1%),of sucrose and these amphiphilic excipients in- b-cyclodextrin (0.9%) and dextran (5%). Theycreased the protein stability synergistically. found that the cryoptotectants that increased the
W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60 15stability of LDH in solution at room temperature activity at all during lyophilization (Ramos et al.,also improved the recovery of protein activity 1997).after freeze– thaw. However, no apparent correla- One major mechanism of protein stabilizationtion was found between the increase in surface by lyoprotectants is the formation of an amor-tension induced by the cryoprotectants and their phous glass during lyophilization (Roser, 1991;protective effect on protein recovery during Franks, 1994; Fox, 1995). Formation of a glassfreeze –thaw cycles. Based on this study, several increases the viscosity to 1012 Pa s (1013 P)other stabilization mechanisms were postulated, (Angell, 1995). It is the extreme viscosity at theincluding modiﬁcation of the size of ice crystals, glassy state, that increases protein stability byreduction (instead of elevation) of surface tension, slowing down interconversion of conformationaland restriction of diffusion of reacting molecules. substates and conformational relaxation of aSupposedly, reduction of surface tension is how protein (Hagen et al., 1995, 1996). This stabiliza-most surfactants stabilize proteins during freezing. tion mechanism explains the retention of G6PDH Besides polymers, many cryoprotectants can in- activity during freeze-drying (Sun et al., 1998).crease the viscosity of a solution, restricting diffu- Amorphous materials are structurally more simi-sion of reacting molecules. In fact, the difference lar to a liquid than crystalline materials (Taylorin solution viscosity has explained why trehalose and Zograﬁ, 1998b). Freeze-dried amorphous in-is apparently more effective than sucrose, maltose, sulin is far more stable than crystalline insulinglucose, or fructose in stabilizing liquid pyrophos- against deamidation and dimer formation at dif-phatase and G6DPH (Sola-Penna and Meyer-Fer- ferent water contents up to 15% (Pikal and Rigs-nandes, 1998). On top of this, concentration of all bee, 1997). Izutsu et al. (1995) studied the effect ofsolutes during freezing increases the solution vis- amphiphilic excipients on freeze-drying of LDHcosity rapidly. Therefore, the rate of a chemical and found that only those that remain amorphousreaction may increase initially due to concentra- in the solid state protected the enzyme duringtion of all solutes but then drops gradually as the freeze-drying. These excipients, including HP-b-viscosity increases (Pikal, 1999). The rate of a CD, CHAPS, sodium cholate, sucrose monolau-chemical reaction is minimized at the glassy state rate, showed a concentration-dependentwhen the viscosity is increased to 1012 Pa·s (An- stabilization effect during freeze-drying.gell, 1995). In addition to viscosity increase, some A glass can be roughly divided into two types:of these cryoprotectants stabilize proteins by sup- fragile and strong. The viscosity of a fragile glasspressing pH changes during freezing (Anchordo- increases more deeply than a stronger glass for aquy and Carpenter, 1996). given temperature drop below the glass transition The preferential interaction mechanism does temperature (Angell, 1995). Therefore, excipientsnot fully explain protein cryoprotection by poly- forming fragile glasses are better stabilizing agentsmers or by proteins themselves at high concentra- (Hatley, 1997). Both sucrose and trehalose cantions. These different mechanisms have been form a fragile glass (Hatley, 1997; Duddu et al.,addressed in Section 3.1. 1997). Another interrelated stabilization mechanism is3.2.2. Mechanisms of lyoprotection the water replacement hypothesis (Crowe et al., During lyophilization, the preferential interac- 1993a; Allison et al., 1996, 1998). This mechanismtion mechanism is no longer applicable because involves the formation of hydrogen bonds be-the hydration shell of proteins is removed (Car- tween a protein and an excipient(s) at the end ofpenter et al., 1993; Crowe et al., 1993b; Allison et the drying process to satisfy the hydrogen bond-al., 1996). Thus, many excipients that stabilize ing requirement of polar groups on the proteinproteins in solution do not offer the same effect surface (Carpenter and Crowe, 1989; Carpenter etduring lyophilization. For example, KCl at 500 al., 1990). These excipients preserve the nativemM effectively protected LDH from thermal in- structures of proteins by serving as water substi-activation at 50°C, but did not protect the protein tutes (Carpenter et al., 1990; Arakawa et al., 1991;
16 W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60Carpenter et al., 1993; Prestrelski et al., 1995). In because of their structural differences. Sucrose hasthis way, intra- or interprotein hydrogen bonding been found to form hydrogen bonds with lysozymemay be prevented during dehydration (Leslie et al., to a greater extent than trehalose (Allison et al.,1995; Cardona et al., 1997). Therefore, stabilization 1999) and with PVP than both trehalose andof proteins requires hydrogen bonding with an rafﬁnose (Taylor and Zograﬁ, 1998b). The differ-excipient(s) during freeze-drying or dehydration ence among sugars in stabilization of proteins may(Carpenter and Crowe, 1989; Arakawa et al., 1991; be partially due to the difference in the extent andCarpenter et al., 1991; Crowe et al., 1998). intimacy of hydrogen bond formation. Since an amorphous state of proteins and stabi- In addition to glass formation, many excipients,lizers allows maximal H-bonding between protein especially polymers, can stabilize proteins by in-and stabilizer molecules, crystallization of any creasing Tg of protein formulations, since higheramorphous protein stabilizers during lyophiliza- Tgs generally result in more stable protein formula-tion often causes protein destabilization due to tions during lyophilization. For example,inefﬁcient hydrogen bonding. Mannitol can easily Costantino et al. (1998b) examined six stabilizersbe crystallized and its crystallization is apparently (lactose, trehalose, cellobiose, mannitol, sorbitol,responsible for the destabilization of some proteins and methyl a-D-mannopyranoside) duringduring lyophilization. The aggregation of IL-6 lyophilization of rhGH and found that the higherduring lyophilization could not be inhibited effec- the Tg of the stabilized formulation, the greater thetively in a formulation containing only mannitol degree of structural (such as a-helix) preservation(Lueckel et al., 1998b). In the presence of 1% PEG, in the co-lyophilizate with less protein aggregation.increasing the mannitol concentration above 10 In general, larger carbohydrates form a glass moremM reduced the activity of LDH and PFK, possi- readily with a higher Tg than smaller ones, but havebly due to crystallization of mannitol (Carpenter et more steric hindrance interfering with intimateal., 1993). Mannitol at 300 mM destabilized Humi- hydrogen bonding with a dried protein (Crowe etcola lanuginosa lipase during lyophilization and al., 1993b). Therefore, selection of such an excipientDSC analysis indicated that 85% mannitol was needs balancing both the formation of a glass withcrystallized during lyophilization (Kreilgaard et al.,1999). a high Tg and intimacy of hydrogen bonding. Although it was debatable whether or not hydro- Other mechanisms of protein stabilization alsogen bond was indeed formed between trehalose and seem operable. Sugars may stabilize proteins bylysozyme upon lyophilization (Belton and Gil, inhibiting crystallization of other excipients such as1994), many studies have conﬁrmed hydrogen PEGs during lyophilization (Izutsu et al., 1995), bybonding by IR spectroscopy between carbohy- inhibiting acute lyophilization-induced protein un-drates and freeze-dried proteins, such as lysozyme, folding such as rhIL-1ra (Chang et al., 1996a), orBSA, and PFK (Carpenter and Crowe, 1989; by preserving a protein’s internal mobility such asCrowe et al., 1993b; Remmele et al., 1997; Allison sperm whale Mb (Sastry and Agmon, 1997).et al. 1999) and bFGF, g-IFN, recombinant G- Polyelectrolytes can stabilize a protein duringCSF, bovine a-lactalbumin, and bovine a-casein lyophilization by forming multiple electrostatic(Prestrelski et al., 1993b). The degree of structural interactions with the protein (Gibson, 1996).protection of lysozyme by sucrose and trehalosespectra was shown to correlate with the extent ofhydrogen bonding between the sugars and the 4. Design of a robust lyophilization cycle — aprotein (Allison et al., 1999). Hydrogen bonding step-by-step analysishas also been demonstrated between sucrose andother non-protein polymers, such as poly-L-lysine The purpose of designing a robust lyophiliza-(Wolkers et al., 1998b), and PVP (Taylor and tion cycle for protein pharmaceuticals is to obtainZograﬁ, 1998b). Different excipients may form a consistent, stable, and esthetically acceptablehydrogen bonds with proteins to different extents product. To achieve this goal, a number of
W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60 17parameters that directly determine or characterize eventually changes the solution from a viscousa lyophilization cycle need to be determined or liquid to a brittle glass, which contains aboutdeﬁned. These parameters should include glass 20–50% water (Pikal, 1990b; Hatley et al., 1996).transition temperature (T % )/collapse temperature g The temperature of this reversible transition for(Tcol), cooling rate, drying rate, and residual mois- the maximally freeze-concentrated solution isture content. termed glass transition temperature, T % . This tem- g perature is also called the temperature of vitreous4.1. Characterization of protein formulations prior transformation (Rey, 1999). T % is used to differen- gto lyophilization tiate this transition from the softening point of a true glass transition, Tg of a pure polymer. T % is g In addition to glass transition temperature (T % )/ g one of the most important parameters for opti-collapse temperature (Tcol), several other critical mization of a lyophilization process (Franks,temperatures, including crystallization tempera- 1990).ture (Tcry), eutectic temperature (Teut), and devit- The collapse temperature (Tcol) is the tempera-riﬁcation temperature (Tdev), should be ture at which the interstitial water in the frozendetermined in order to design a robust lyophiliza- matrix becomes signiﬁcantly mobile (Jennings,tion cycle. These temperatures are mostly deter- 1999). Tcol is closely related to T % . In fact, Tcol has gmined by thermal analysis such as DSC, electrical been considered to be equivalent to Tg of anresistance measurements, and direct microscopic amorphous system or to the eutectic melting tem-observation. perature of a crystalline system (Slade et al., 1989; Pikal, 1990a,b). Recent literature indicates that4.1.1. Glass transition temperature (T % ) and g the Tcol of many small carbohydrates is consis-collapse temperature (Tcol) tently higher than their T % by about 12 K (Sun, g Ice formation during cooling of a protein solu- 1997). The discrepancy between T % and Tcol for gtion concentrates all solutes. Solute concentration polymers seems even larger (Roos and Karel, 1991). This is because the decrease in viscosity at T % may not be sufﬁcient enough to cause struc- g tural collapse (Bindschaedler, 1999). For refer- ence, Table 1 lists T % s and Tcols of some g commonly used excipients and buffers. 4.1.2. Crystallization temperature (Tcry) When the temperature of an aqueous protein formulation drops below 0°C, water usually crys- tallizes out ﬁrst. Then, the crystalline component, which usually has the least solubility in the formu- lation, may crystallize out. This temperature is termed crystallization temperature. 4.1.3. Eutectic crystallization/melting temperature (Teut) When the temperature of an aqueous protein formulation further decreases after crystallization of the least soluble component, this component and water crystallize out at the same time as aFig. 1. A theoretical phase diagram showing ice formation, mixture. This temperature is termed eutectic crys-solute crystallization, eutectic point, and glass transition dur- tallization/melting temperature. The relationshiping freezing. between Teut and T % is shown in Fig. 1. Due to g
18Table 1Glass transition and collapse temperatures (°C) of buffers, excipients and proteinsCompounds T% g Tcol Tg ReferencesBuffering agents aCitric acid −54a, −53b 11b Chang and Randall, 1992; bLu and Zograﬁ, 1997Hepes −63 Chang and Randall, 1992 aHistidine −33a, −32b Chang and Randall, 1992; bOsterberg and Wadsten, 1999 8Potassium −76 Franks, 1993acetatePostassium −62 Franks, 1993citratePotassium −55 Chang and Randall, 1992phosphate(KH2PO4)Sodium acetate −64 Chang and Randall, 1992Sodium −52 Franks, 1993bicarbonateSodium citrate −41 Chang and Randall, 1992Sodium −45 Chang and Randall, 1992phosphate(NaH2PO4)Tris base −55 Franks, 1993Tris–HCl −65 Chang and Randall, 1992Excipients, low MWb-Alanine −65 Franks, 1993Arabinose 3 Roos, 1993Arginine 42 Mattern et al., 1999Cellobiose 77 Franks, 1990; Costantino et al., 1998b aFructose −42b 10a, 13b Roos, 1993; bWisniewski, 1998Fucose 31 Roos, 1993 aGalactose −41a 31b, c38 Franks, 1990; bLevine and Slade, 1992; cRoos, 1993 a W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60Glucose −43a −41d 4c, 32b, 39a Franks, 1990; bLevine and Slade, 1992; cPrestrelski et al., 1995; dAdams and Ramsay, 1996Glutamic acid −17 Chang and Randall, 1992 aGlycerol −65a −92b, −93a Franks, 1990; bBell et al., 1995Glycine −37 Chang and Fischer, 1995 b a aHistidine −32 37 Mattern et al., 1999; bOsterberg and Wadsten, 1999 8 b aLactose −30.5 101a, 114c Levine and Slade, 1992; bAdams and Ramsay, 1996; cTaylor and Zograﬁ, 1998bLysine 68 Mattern et al., 1999 aMaltose −30a 43a, 82b, Franks, 1990; bLevine and Slade, 1992; cTaylor and Zograﬁ, 1998b 100cMaltotriose −23 Hatley and Franks, 1991 aMannitol −33a,−27b 13c Meredith et al., 1996; bLueckel et al., 1998a; cKim et al., 1998
Table 1 (Continued)Compounds T% g Tcol Tg References aMannose −41a 30a, 33b Franks, 1990; bLevine and Slade, 1992Melibiose 91 Roos, 1993Octulose 21 Wolkers et al., 1998a aRafﬁnose −26c 108b, 114a Taylor and Zograﬁ, 1998b; bWolkers et al., 1998a; cWillemer, 1999 aRibose −47b −10b, −13a Roos, 1993; bWisniewski, 1998Sodium chloride B−60 Franks, 1993 aSorbitol (glucitol) −44a,b −54c −3a Franks, 1990; bKerr et al., 1993; cAdams and Ramsay, 1996Sucrose −32f −31d, −40h 31c, 61a, 72e, aLevine and Slade, 1992; bHancock et al., 1995; cPrestrelski et al., 1995; dAdams and 77b, 75g Ramsay, 1996; eCostantino et al., 1998c; fKasraian et al., 1998; fLueckel et al., 1998a; g Taylor and Zograﬁ, 1998b; hOvercashier et al., 1999Trehalose −29c, −34h 78a, 46b, 117d, aLevine and Slade, 1992; bPrestrelski et al., 1995; cAdams and Ramsay, 1996; dDuddu and 115c, 105f, 118g Dal Monte, 1997; eMiller et al., 1997; fLueckel et al., 1998a; gTaylor and Zograﬁ, 1998b; h Overcashier et al., 1999 aWater −133a, −137b Bell et al., 1995; bMiller et al., 1997 bXylitol −47b −23a, −39b Franks, 1990; aRoos, 1993 aXylose −48a 9a, 14b Franks, 1990; bRoos, 1993Excipients, high MWCellulose 227 Hancock and Zograﬁ, 1994b-Cyclodextrin 108 Prestrelski et al., 1995 aDextran (10 kD) −10b 91a Prestrelski et al., 1995; bWillemer, 1999 aDextran (40 kD) 94a, 101b te Booy et al., 1992; bPrestrelski et al., 1995Dextran (70 kD) −11 Adams and Ramsay, 1996Ficoll −20 Pikal, 1990b; Izutsu et al., 1996Gelatin −8 Pikal, 1990bHydroxypropylmeth 155 Hancock and Zograﬁ, 1994yl-cellulose aHydroxyethyl −12a −5b 110c Crowe et al., 1993b; bPikal, 1990b; cCrowe et al., 1998starchMaltodextrin 860 169 Taylor and Zograﬁ, 1998bMethocel −9 Willemer, 1999 W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60PEG (6 kD) −13 Willemer, 1999Polydextrose −27 Kerr et al., 1993 aPVP K15 (10 kD) −27b 100a Bell et al., 1995; bIzutsu et al., 1996PVP (40 kD) −24 Adams and Ramsay, 1996PVP K30 (40 kD) 180 Bell et al., 1995 aPVP K90 (1000 185a, 176b Hancock et al., 1995; bTaylor and Zograﬁ, 1998bkD)Sephadex G 200 −10 Willemer, 1999Starch 227 Hancock and Zograﬁ, 1994Proteins a 19BSA −13a, −11b Slade et al., 1989; bChang and Randall, 1992
20Table 1 (Continued)Compounds T% g Tcol Tg Referencesa-Casein −12.5 Slade et al., 1989Globulins (egg) −10.5 Slade et al., 1989HSA −9.5 Pikal, 1990ba-Lactalbumin −10.5 Slade et al., 1989(Sigma)LDH −9 Chang and Randall, 1992 aLysozyme −16.5a, −13b Slade et al., 1989; bChang and Randall, 1992rhuMAb HER2 −20 Overcashier et al., 1999Myoglobin −15 Chang and Randall, 1992 aOvalbumin −11a −10b Chang and Randall, 1992; bWillemer, 1999RNase A −12 Chang and Randall, 1992 W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60
W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60 21excipient interaction(s), many multicomponent ice crystals. The size of crystals determines theprotein formulations do not exhibit Teut (Hatley pore size to be created during subsequent drying.et al., 1996). Large ice crystals create large pores, leading to rapid water sublimation during primary drying184.108.40.206. De6itriﬁcation temperature (Tde6). When (Willemer, 1992), but the secondary drying maythe temperature of the glassy maximally freeze- slow down due to smaller surface areas, limitingconcentrated solution (MFCS) increases, an en- water desorption during secondary drying (Bind-dothermic glass transition ﬁrst occurs. Further schaedler, 1999). To keep a balance, a moderateincrease in temperature above T % may lead to an g degree of supercooling (10–15°C) has been rec-exothermic event, corresponding to the recrystal- ommended (Pikal, 1990a).lization of a component such as mannitol The rate of freezing-induced protein denatura-(Meredith et al., 1996). This temperature is tion is a complex function of both cooling ratetermed devitriﬁcation temperature. Devitriﬁcation and ﬁnal temperature (Franks, 1990). The effectis a process by which a metastable glass forms a of cooling rates on the stability of proteins variesstable crystalline phase on heating above T % g signiﬁcantly. For example, increasing the freezing(Slade et al., 1989; Chang and Randall, 1992; rate from 0.5 to 50°C min − 1 did not signiﬁcantlyRey, 1999). Recrystallization can occur if a solu- affect the formation of soluble aggregates of rGHtion has been cooled rapidly, arresting crystal at 2 mg ml − 1 in 5 mM phosphate buffer (pH 7.4nucleation and/or growth. To detect recrystalliza- or 7.8), but the formation of insoluble aggregatestion, the heating rate should be slower than the (particulates) increased sharply with increasingcritical heating rate, which is deﬁned as the mini- cooling rates, even in the presence of up to 250mum heating rate fast enough to prevent devitriﬁ- mM mannitol (Eckhardt et al., 1991). Rapidcation of the unfrozen fraction of a solution freezing also caused formation of more aggregates(Chang and Randall, 1992). for bovine and human IgG than slow freezing (Sarciaux et al., 1998). This may result from the4.2. Freezing — the ﬁrst step in lyophilization formation of smaller ice crystals and larger ice- water interfaces at higher freezing rates, leading The freezing step during lyophilization is con- to a greater extent of surface-induced proteinsidered to be at least as important as the drying denaturation. On the contrary, faster freezingstep due to its potential effect on proteins (Wille- caused less loss of LDH activity (Nema and Avis,mer, 1992). One critical parameter that needs to 1992) and less change in the secondary structurebe deﬁned during freezing is the cooling rate. The of hemoglobin in a PEG/dextran solution (Hellercooling rate, w, can be deﬁned as et al., 1999a). The lower loss of protein activity is lT(r,t) likely because faster freezing may prevent exten-w= , sive crystal growth, which may substantially hin- lt der solute concentration-induced protein dena-where, T(r,t) is the temperature ﬁeld, a function turation. Therefore, stability of proteins may beof both time, t, and location, r. Therefore, the affected differently at different freezing rates de-rate varies temporally and spatially (Hartmann et pending on protein denaturation mechanisms.al., 1991). In general, a faster freezing rate gener- It should be noted that freezing rate may haveates small ice crystals (Eckhardt et al., 1991; a potential impact on the storage stability ofWillemer, 1992; Wisniewski, 1998). This is be- lyophilized proteins. Hsu et al. (1995) demon-cause water is super-cooled and crystallization strated that faster cooling during lyophilization ofinto ice occurs rapidly, producing small ice crys- tPA resulted in a product cake with larger inter-tals (Pikal, 1990a; House and Mariner, 1996). nal surface area, which led to formation of moreConversely, a slower cooling rate generates larger opalescent (insoluble) particulates upon long-term
22 W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60storage at 50°C. The rate of formation of opales- became higher from the surface to the center ofcent particulates during storage correlated well the container. They found that the optimum loca-(r= 0.995) with internal surface area of the tion where the cooling rate should be measured,lyophilized tPA product cake. was 1/3 away from the center and 2/3 away from As discussed in Section 2.2, buffer species may the inside surface of the sample container. Thecrystallize out selectively and cause pH shifting cooling rate at this point represented 80% ofduring freezing. Therefore, it is preferable to keep the entire sample volume. The geometrical centerall buffer species amorphous during freezing of was apparently the worst location (the least repre-pH-sensitive proteins. A faster cooling rate may sentative) for measuring cooling rate. Althoughprevent nucleation and subsequent crystallization the optimum location for measuring cooling rate(Franks, 1993). Each buffer salt has its own criti- in a cylindrical vial has not been determined, itcal cooling rate, which is deﬁned as the minimum may not be at the center based on the abovecooling rate fast enough to prevent crystallization study.of a solute (Chang and Randall, 1992). Crystal-lization will not occur when the cooling rate is 4.3. Thermal treatment prior to dryinghigher than the critical cooling rate. If a proteinsolution is cooled rapidly with liquid nitrogen, it Very often, a thermal treatment step, annealing,is likely that buffer salts will remain amorphous. is included before the primary drying step. ThereOtherwise, selective crystallization and a subse- are at least two reasons for this. First, during thequent pH change may occur (Chang et al., freezing step, a crystalline component may not be1996b). completely crystallized. Complete crystallization Freezing rate inﬂuences the extent of crystal- may be necessary if this component is to providelization of a formulation excipient, such as manni- necessary cake structure or if the protein is moretol (Hsu et al., 1996). Accordingly, the duration of stable after complete crystallization. Recrystalliza-subsequent thermal treatment after freezing can tion can be promoted by heat treatment at tem-be affected (see next section). In addition, differ- peratures above T % of the formulation. The gent freezing rates may favor formation of certain duration of the treatment depends on the compo-crystalline forms of an excipient, which may po- sition of the formulation and the heating ratetentially affect protein stability and reconstitution. through T % (Getlin, 1991). Second, removal from gIt has been observed that a slower rate of crystal- the amorphous phase of a crystalline componentlization tends to favor formation of g-glycine, which has a low T % , such as glycine (T % = − g gwhereas rapid crystallization seems to favor for- 42°C), can increase the T % of the amorphous gmation of the b-polymorph (Akers et al., 1995). phase (Carpenter et al., 1997). The increased T % gDifferent polymorphs of mannitol were also ob- can allow more efﬁcient primary drying at atained at different concentrations during freezing higher temperature. For example, annealing at(Izutsu et al., 1993). Recently, Kim et al. (1998) − 20°C in a glycine:sucrose (1:1 weight ratio)demonstrated that slow freezing (about 0.2°C formulation increased the T % from − 44 to − gmin − 1) of 10% (w/v) mannitol produced a mix- 33°C and produced a formulation cake of betterture of a and b-polymorphs and fast freezing (by appearance and higher mechanical strengthliquid nitrogen) of the same solution produced (Lueckel et al., 1998a,b). Similarly, incorporationd-form. The reconstitution time (with water) was of an annealing step in the lyophilization of a36 and 78 s, respectively, for the fast-freeze and monoclonal antibody crystallized glycine and en-slow-freeze dried mannitol samples. abled a higher drying temperature (Ma et al., What is the ideal location in a product vial 1998).where the cooling rate should be measured? Hart- However, an annealing step may have an ad-mann et al. (1991) demonstrated that the cooling verse effect on the stability of a protein due torate in a ﬂat plate-shaped freezing container, ver- crystallization of an amorphous stabilizer, losingtically submerged into liquid nitrogen, generally hydrogen bond interaction with the protein (also
W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60 23see Sections 3.2 and 5.2). Such cases occurred to dependent dry layer hinders diffusion of waterseveral proteins during lyophilization, including vapor, increases Rp, and causes product tempera-b-galactosidase (2 mg ml − 1) in mannitol formula- ture to rise, as observed for Erwinia L-asparagi-tion during lyophilization (Izutsu et al., 1993), nase during lyophilization (Adams and Ramsay,interleukin-6 (IL-6) in sucrose:glycine (1:1 at 20 1996). Any temperature rise during lyophilizationmg ml − 1) formulation (Lueckel et al., 1998b), and may potentially cause product collapse. In thisLDH in mannitol formulation (Izutsu et al., case, the shelf temperature needs to be reduced1994b). Annealing at −7°C for 1 or 12 h was also accordingly to prevent product collapse or melt.found to destabilize hemoglobin in PEG/dextrin On the other hand, small-scale product collapse(1:1 weight ratio) system during lyophilization, as during lyophilization may decrease Rp (Over-monitored by IR (Heller et al., 1999c). cashier et al., 1999). Several other factors, such as ice morphology,4.4. Drying crystal size distribution (see Section 4.2), and for- mulation composition, also affect the drying rate The drying step is divided into two phases: or time. Excipients in protein formulation mayprimary and secondary drying. The primary dry- interact with proteins and reduce the availabilitying removes the frozen water (sublimation of ice) of water-binding sites in proteins. At the sameand the secondary drying removes non-frozen time, excipients themselves may interact with wa-‘bound’ water (desorption of water). The amount ter molecules. Thus, the water-binding force ofof non-frozen water for globular proteins is about the formulation components and the associated0.3–0.35 g g − 1 protein, slightly less than the amount of monolayer water molecules coveringproteins’ hydration shell (Rupley and Careri, the protein formulation will be different depend-1991; Kuhlman et al., 1997). ing on the formulation composition, as demon- Different models have been reported to describe strated in the rhGH:sugar or rhIGF-I:sugarthe drying/sublimation rate during lyophilization formulations (Costantino et al., 1998c). Conse-(Jennings, 1999). When the shelf temperature is quently, the drying time will be different depend-ﬁxed, the drying/sublimation rate, w, of a frozen ing on formulation composition. In addition,solid can be expressed as: composition-dependent formation of any excipi- ent hydrates during lyophilization would invari- Ap(Pp −P0)w= ably reduce the drying rate, as implicated in the Rp formation of a mannitol hydrate duringin which Ap is the cross sectional area of a lyophilization (Yu et al., 1999).product, Pp is product vapor pressure at the subli- The driving force for water sublimation duringmation front, P0 is partial vapor pressure in a lyophilization is the temperature difference be-product vial, and Rp is resistance of a dried tween the product and the condenser. The com-product layer to vapor ﬂow (Nail and Johnson, monly used condenser temperature is − 60°C,1992). Rp may be different under different freezing allowing a minimum of 20°C lower than theconditions. Since Ap, Pp, and Rp at a ﬁxed freez- product temperature during primary dryinging rate are usually ﬁxed for a particular protein (Franks, 1990). During secondary drying the con-formulation in a chosen container, drying rate can denser temperature can be set even lower, such asbe changed only through adjustment of P0. − 80°C, for formulations that require very low As indicated in the above equation, drying rate residual moisture (Bindschaedler, 1999). Tois inversely proportional to Rp. If Rp changes achieve a high drying rate, product temperature isduring the drying process, the drying rate and the frequently set as high as possible. Since theproduct temperature will change accordingly. product temperature is controlled by the shelfDuring lyophilization, continuous drying may temperature, effective heat transfer between shelflead to formation of an increasingly dry product and product is essential, which is affected by thelayer (Overcashier et al., 1999). This formulation- degree of vacuum in the drying chamber. A mod-
24 W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60erate increase in chamber pressure often increase temperature. Sheehan and Liapis (1998) recentlythe drying rate due to more effective heat transfer, modeled primary and secondary drying of phar-leading to a higher product temperature (Bind- maceutical products in vials. They found thatschaedler, 1999). In the development of a controlling heat input close to melting and scorchlyophilized vaccine formulation, the convection of (thermally damaging) temperature constraints re-heat between shelf and product was shown to be sulted in not only faster drying time but also moremore effective under a vacuum of 100 – 120 mTorr uniform distribution of temperature and boundthan under maximum vacuum (15 – 25 mTorr) water in the formulation at the end of secondary(House and Mariner, 1996). For this reason, dry- drying. The scorch temperature was deﬁned as theing of the formulation was not conducted under temperature of the top surface that shows thermalmaximum vacuum. Similarly, it has been reported damage and the melting temperature of the frozenthat the maximum drying rate at a chamber pres- phase was considered to be about 10°C below thesure of 400 mmHg is more than twice that at 100 melting point of the ice.mmHg at a shelf temperature of 40°C (Nail and The end of the primary drying process is whenJohnson, 1992). In a more recent report, it was all the frozen water is removed and the rate ofshown that at a constant shelf temperature of water sublimation is signiﬁcantly reduced. Several25°C, the speciﬁc sublimation rate was 0.19, 0.16, methods can be used in monitoring the comple-and 0.11 g h − 1 cm − 2 at chamber pressures of tion of the primary drying cycle. A simple method300, 200, and 100 mTorr, respectively, for a is to observe the changes in product temperatureprotein formulation containing trehalose, his- (or chamber pressure) during freeze-drying. Thetidine, and polysorbate 20 (Overcashier et al., end of the primary drying process is when the1999). Therefore, a balanced degree of vacuum in product temperature approaches the shelf temper-the drying chamber is needed to achieve the de- ature, evidenced by a signiﬁcant change in thesired drying rate. It has been suggested that a slope of the product temperature trace due to achamber pressure at one-fourth to one-half of the reduced sublimation rate. To prevent prematuresaturated vapor pressure over ice usually lead to a ending, an extra 2–3 h may be added to thehigh sublimation rate (Bindschaedler, 1999). drying cycle. A more objective method is the During lyophilization, complete sample collapse pressure rise test. By disconnecting the vacuumresults in both a lower rate of water sublimation/ source, the chamber pressure should rise at a ratedesorption and an inferior product. In addition, depending on the amount of moisture in thecollapsed materials may crystallize more easily product. The end of the drying process would bethan non-collapsed materials during storage when the rate of pressure rise is below a speciﬁed(Darcy and Buckton, 1997). To prevent product value. Another method for determining the end ofcollapse, the product temperature must be kept the primary drying process is the measurement ofbelow the glass transition temperature (T % ) (or g the heat transfer rate (Jennings and Duan, 1995;Tcol) of the formulation or below the eutectic Oetjen, 1999). Timely detection of the end of themelting temperature (Teut) of any crystalline com- primary drying cycle by this method have resultedponent. On the other hand, primary drying should in a more efﬁcient drying process (Jennings andbe operated at a temperature as close as possible Duan, 1995). The duration of the secondary dry-to these temperature limits for high efﬁciency. ing cycle is dictated by the required moistureTherefore, to have an efﬁcient drying step and to content in the ﬁnal product (see next section).reduce probability of product collapse, a formula- To improve drying efﬁciency, a single-dryingtion should be designed such that its T % is as high g step may be designed for certain proteins. Changas possible. To ensure a high T % during secondary g and Fischer (1995) developed such a cycle fordrying, the primary drying cycle should be com- rhIL-1ra formulation. In this single cycle, thepletely ﬁnished so that only a minimum amount shelf temperature was set for the secondary dryingof bound water is left in the formulation. There and the product temperature was controlled to aare other advantages of drying at a relatively high certain level by adjusting the chamber tempera-
W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60 25ture. Freeze-drying of 1 ml of rhIL-1ra formula- is complex (see Section 5.2). Usually, a lowertion could be completed within 6 h with a ﬁnal moisture content leads to a more stable proteinmoisture level of 0.4%. However, the cycle-associ- product, although there may not be any signiﬁ-ated heterogeneity in temperature and moisture cant difference in protein stability between near-distribution within the product may cause poten- zero and an intermediate moisture content oftial damage to other labile proteins. about 1% (Pikal, 1990a). On the other hand, if a lyophilized formulation needs additional viral in-4.5. Residual moisture activation by dry heat, its moisture content needs to be high enough to achieve efﬁcient and effec- The desired residual moisture level in a tive inactivation (Savage et al., 1998). As a gen-lyophilized product dictates duration of the sec- eral rule, a moisture content in a lyophilizedondary drying step. An electronic hygrometer or a protein formulation should not exceed 2%residual gas analyzer may be used to determine (Daukas and Trappler, 1998). This general ruleresidual moisture level during lyophilization and seems applicable at least to a couple of proteinthus, the end-point of secondary drying (Nail and formulations. Lyophilized bFGF with sugars wasJohnson, 1992). The aforementioned pressure rise stable as long as the moisture content was belowtest or the measurement of heat transfer rate can 2% (Wu et al., 1998). Lyophilized monoclonalalso be used for determination of the end of the antibody cA2 IgG remained stable with a mois-secondary drying cycle. If these methods are not ture content of 2.2% or less (Katakam et al.,available, the minimum duration of drying may 1998). However, lyophilized BSA and bovine g-have to be determined systematically using differ- globulin (BGG) formulations were more stable atent combinations of shelf temperatures and dura- a water content of about 10% than at B 1%tions (Greiff, 1992). Moisture content of (Yoshioka et al., 1997). The complicated effect oflyophilized formulations can be determined by water on stability of solid protein formulations isseveral methods, including loss-on-drying, Karl discussed in Section 5.2.Fischer titration, thermal gravimetric analysis To ﬁnd the moisture content that confers the(TGA), gas chromatography (GC), or near IR. maximal stability for a lyophilized proteinThe reproducibility of moisture determination by product, long-term stability studies should be con-the ﬁrst three methods was found to be similar for ducted on protein formulations with differentseveral biological products (May et al., 1989). moisture contents. Only these real-time stability What is the residual moisture in lyophilized studies can determine the optimal moisture con-product? Proteins have both strong and weak tent for the ﬁnal protein product.binding sites to accommodate unfrozen water.The weak binding sites include mostly carbonylbackbone plus hydrophilic – OH and – NH – 5. Instability, stabilization, and formulation ofgroups, while the strong binding sites include solid protein pharmaceuticalsthose ionizable groups in amino acids such asGlu, Asp, Lys, and Arg (Careri et al., 1979). For Lyophilized proteins may lose activity rapidlylysozyme, strongly bound water molecules during storage, even though they may be stableamount up to 10% (g g − 1 protein) at 38°C (Careri during lyophilization. For instance, porcine pan-et al., 1979). Since the residual moisture content creatic elastase without excipients retained its fullfor lyophilized protein products is usually below activity after freeze-drying, but lost about 70%10%, secondary drying removes weakly bound activity in 2 weeks at 40°C and 79% RH (Changand some of strongly bound water molecules. et al., 1993). Therefore, lyophilized proteins stillTherefore, the residual moisture is a small portion need stabilization in the solid state to surviveof strongly bound water molecules in proteins. long-term storage as pharmaceuticals. In the fol- Residual moisture causes a variety of instabili- lowing section, instability pathways of proteins inties in dried proteins and in many cases, the effect solid state during storage are ﬁrst described in
26 W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60brief, followed by a detailed description of factors rHA (Costantino et al., 1995b), BSA and b-lac-affecting stability of solid proteins and various toglobulin (Liu et al., 1990; Jordan et al., 1994).stabilization strategies. In the ﬁnal section, several Both covalent (disulﬁde bond) and non-covalentaspects are discussed in relation to formulation of formation of dimers and trimers were observedacceptable solid protein pharmaceuticals. for a spray-dried anti-IgE monoclonal antibody (Andya et al., 1999).5.1. Instability of solid proteins during storage Often, both soluble and insoluble protein aggre- gates can form at the same time during storage. A variety of instability mechanisms have been This is the case for hGH (Pikal et al., 1992;reported of lyophilized proteins during storage. Costantino et al., 1998b), Humicola lanuginosaThese include aggregation (a major physical insta- lipase (Kreilgaard et al., 1999), and tPA (Hsu etbility) and different chemical degradations such as al., 1995). The relative amounts of soluble anddeamidation, browning reaction, oxidation, hy- insoluble protein aggregates may change withdrolysis, and disulﬁde bond formation/exchange. storage conditions such as lysozyme aggregation under different relative humidities (Separovic5.1.1. Protein aggregation et al., 1998). Both physical and chemical aggrega- Aggregation is one of the major instabilities for tion can lead to formation of insoluble aggregateslyophilized protein pharmaceuticals during stor- such as moisture-induced aggregation ofage (Costantino et al., 1998d). Many lyophilized lyophilized insulin during storage (Costantino etproteins form aggregates easily during storage al., 1994b).under accelerated conditions, such as BSA at 37or 60°C (Liu et al., 1990; Jordan et al., 1994), 5.1.2. Chemical degradationsrHA at 37°C and 96% RH (Costantino et al., Generally speaking, chemical degradations of1995a,b), porcine pancreatic elastase at 40°C and proteins in solid state have not been reported as79% RH (Chang et al., 1993), Humicola lanugi- extensively as in liquid state. Nevertheless, severalnosa lipase at 40 or 60°C (Kreilgaard et al., 1999), chemical degradation pathways have been ob-IL-2 at 45°C (Zhang et al., 1996) or 65°C (Kenney served in lyophilized proteins during storage. Inet al., 1986), rhIL-1ra at 50°C (Chang et al., some cases, multiple degradation processes pro-1996a), insulin at 50°C and 96% RH (Costantino ceed simultaneously in a protein such aset al., 1994b), tPA at 50°C (Hsu et al., 1995), and lyophilized rGH, which undergoes methionine ox-tumor necrosis factor (TNF) at 37°C (Hora et al., idation, asparagine deamidation, and irreversible1992a). Some proteins form aggregates during aggregation during storage (Pikal et al., 1992).storage under ambient conditions. These include Detailed mechanisms of chemical degradations inaFGF (Volkin and Middaugh, 1996), b-galactosi- solid proteins and peptides have recently beendase (Yoshioka et al., 1993), hGH (Pikal et al., reviewed (Lai and Topp, 1999).1992; Costantino et al., 1998b), and antibody- Chemical degradations may not affect the activ-vinca conjugate (Roy et al., 1992). ity of proteins, depending on the location of the Protein aggregation can be physical, chemical, transformed residue(s). Due to the terminal loca-or both. Physical (non-covalent) interaction is the tion of MetB4 and MetB25 in recombinant humancause of protein aggregation for tetanus toxoid relaxin, oxidation of these two residues did not(Costantino et al., 1994a), ovalbumin (chicken egg change the protein bioactivity (Nguyen et al.,albumin), and glucose oxidase (Liu et al., 1990). 1993). The Met1 mono-oxidized recombinant hu-Disulﬁde bond formation or exchange is a major man leptin (16 kD) did not show any detectablechemical (covalent) pathway leading to protein changes in tertiary structure and retained its fullaggregation. Proteins that aggregate by this mech- potency, as compared with the native form (Liu etanism include b-galactosidase (Yoshioka et al., al., 1998). Other protein degradation products1993), insulin (Costantino et al., 1994b; Strickley having essentially the same biological activity asand Anderson, 1996; Pikal and Rigsbee, 1997), the intact proteins include deamidated insulin
W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60 27(Brange et al., 1992b), oxidized IL-2 (Kenney et of the reconstituted protein solution correlatedal., 1986), and deamidated rIL-2 (Sasaoki et al., positively with the activity loss of the protein1992). (Schebor et al., 1997). The lactose or maltose-in- duced browning reaction may explain why these5.1.3. Deamidation two sugars were less effective than sucrose, a Deamidation is one of the major degradation non-reducing sugar, in stabilizing vacuum-driedpathways in lyophilized proteins during storage. restriction enzyme EcoRI during storage at 45°CAsn and Gln are the two amino acids susceptible (Rossi et al., 1997).to deamidation in proteins. Many cases of deami- Although sucrose is a non-reducing sugar, it candation have been reported in lyophilized proteins. be easily hydrolyzed into two reducing sugars,For example, both lyophilized rGH and bFGF especially at low pHs during storage, not only indeamidated during storage (Pikal et al., 1992; Wu liquid state (Reyes et al., 1982; Buera et al., 1987)et al., 1998). Insulin, lyophilized from a solution of but also in solid state (te Booy et al., 1992;pH 3–5, deamidated via a cyclic anhydride inter- Skrabanja et al., 1994). It has been demonstratedmediate at C-terminal AsnA21 in addition to cova- that the rate of color formation in a freeze-driedlent dimerization during storage (Strickley and sucrose/lysine formulation at pH 2.5 at 40°C wasAnderson, 1996). The combined formation of close to that in a glucose/lysine formulation, partlydeamidated insulin and insulin dimers was due to the catalytic effect of the amino acid onshown to be a linear function of square root of sucrose hydrolysis (O’Brien, 1996). Therefore,time (Pikal and Rigsbee, 1997). Storing lyophilized lyophilized proteins in a sucrose-containing formu-IL-1ra (in 2% glycine, 1% sucrose, and 10 mM lation may still have the potential to experience thesodium citrate buffer) at 50°C caused protein browning reaction. This is the case for lyophilizeddeamidation in addition to aggregation (Chang et IL-6 during storage (Lueckel et al., 1998b).al., 1996a,c). 5.1.5. Oxidation5.1.4. Maillard reaction The side chains of Met, Cys, His, Trp, and Tyr Reducing sugars such as glucose can react with residues are potential sites of oxidation (Manninglysine and arginine residues in proteins to form et al., 1989). Methionine residues in proteins cancarbohydrate adduct via the Maillard reaction, easily be oxidized by atmospheric oxygen.which is also called the browning reaction (Paulsen Lyophilized hGH was easily oxidized in a vialand Pﬂughaupt, 1980). The Maillard reaction has containing only 0.4% oxygen during storage atbeen a subject of extensive investigation mainly in 25°C (Pikal et al., 1991, 1992). The methioninethe food industry (Chuyen, 1998). In the develop- oxidation in hGH during storage has been provedment of solid protein pharmaceuticals, this reac- to be insensitive to moisture. Other examples oftion has also been observed in several lyophilized methionine oxidation in proteins includeproteins during storage, including aFGF (Volkin lyophilized hIGF-I (Fransson et al., 1996) andand Middaugh, 1996), bFGF (Wu et al., 1998), IL-2 (Kenney et al., 1986; Hora et al., 1992b).human relaxin (Li et al., 1996), IgG (Hekman et The free sulphydryl groups of cysteines in solidal., 1995), and porcine pancreatic elastase (Chang proteins can also be easily oxidized to formet al., 1993). This glycation reaction also occurred disulﬁde bridges during storage. This is one of theto an anti-IgE monoclonal antibody co-spray- major chemical mechanisms of covalent proteindried with lactose during storage (Andya et al., aggregation as mentioned above.1999). The browning reaction can result in signiﬁcant 5.1.6. Hydrolysisinactivation of a lyophilized protein during stor- Although moisture content is usually low inage. Storing lyophilized invertase in the presence lyophilized protein formulations, hydrolysis canof rafﬁnose, lactose, or maltose at 95°C for 7 days still occur during storage. Hydrolysis of bFGF hasled to signiﬁcant browning and the color intensity been observed recently in a lyophilized sugar
28 W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60formulation (Wu et al., 1998). Li et al. (1996) 1999). High temperatures also accelerate chemicalfound that the loss of relaxin activity (48%) in a degradations of proteins in solid state, such aslyophilized glucose formulation during storage at deamidation of rhIL-1ra at temperatures between40°C for 2 weeks was signiﬁcantly higher than 8 and 50°C (Chang et al., 1996a,c) and dimeriza-that (8%) in either mannitol or trehalose formula- tion of TNF at temperatures between 25 and 80°Ction. The loss of protein activity was apparently (Hora et al., 1992a).due to glucose-induced elimination of serine at the Increasing temperature also affects protein sta-C-terminal of the B chain in relaxin to form bility indirectly. Since proteins in a solid formula-des-Ser relaxin. tion are stabilized in an amorphous phase, crystallization of an amorphous component(s)5.2. Factors affecting stability of solid proteins may destabilize proteins. Crystallization of an amorphous component in a protein formulation Several factors can affect the stability of solid can occur slowly during storage below Tg andprotein pharmaceuticals. These include storage rapidly above Tg (Sun, 1997). The effect of crys-temperature, glass transition temperature, formu- tallization on protein stability will be discussed inlation pH, residual moisture content, type and a following section.concentration of formulation excipients, and crys-tallization of amorphous excipients. 5.2.2. Glass transition temperature (Tg) Glass transition temperature (Tg) of protein5.2.1. Storage temperature formulations is considered to be one of the major Storage temperature is probably one of the determinants of protein stability (Hatley andmost important factors affecting protein stability Franks, 1991). Tg of a polymer is deﬁned as thein solid state. Many lyophilized proteins show transition temperature between the rubbery (orincreased loss of activity at high temperatures, liquid-like) and glassy (solid-like) states. Micro-such as restriction enzyme PstI in a trehalose scopically, a polymer chain may have cooperativeformulation and neutral lactase in a PVP formula- localized motion above Tg, and below Tg, onlytion at temperatures between 37 and 70°C (Co- individual atoms are able to make small excur-laco et al., 1992; Mazzobre et al., 1997). sions about their equilibrium positions (Ravve,Unfortunately, due to the structural complexities 1967). The greater mobility above Tg is due to aof proteins, the temperature effect on stability of greater free volume and higher degree of bothsolid proteins cannot be simply described by a translational and rotational freedom (Slade et al.,single instability mechanism, although in general, 1989). Therefore, formation of a glassy statethe higher the temperature, the lower the protein should result in a drop in molecular diffusion/mo-stability, both physically and chemically. It should tion and thus, increased protein stability (Duddube noted that ﬂuctuating storage temperatures and Dal Monte, 1997).may be more detrimental to a lyophilized protein DSC is the most widely used method for deter-than a single high storage temperature (Ford and mination of Tg (Crowe et al., 1998). However, theDawson, 1994). Tg of a protein formulation may not be easily High temperatures accelerate physical aggrega- detectable or accurately determined by DSC duetion of proteins in solid state. This can be ascribed to its poor reproducibility, limited detectability,to an increased mobility of protein molecules at formulation heterogeneity, and contribution ofhigh temperatures, which facilitates protein – secondary (or b-) relaxation processes (Fan et al.,protein interactions. Increasing temperatures in- 1994; Bell et al., 1995; Pikal and Rigsbee, 1997;creased aggregation of lyophilized rhIL-1ra Yoshioka et al., 1997, 1998). Several other factorsbetween 8 and 50°C (Chang et al., 1996a,c), ag- can affect Tg and/or its determination, such as thegregation of vacuum-dried LDH between 3 and condition of glass formation, moisture content,60°C, and aggregation of vacuum-dried rhG – temperature ramping rate, thermal history of sam-CSF between 3 and 80°C (Mattern et al., 1997, ples, or presence of multiple temperature transi-
W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60 29tions (Franks, 1994; Her and Nail, 1994; Sartor et ferent molecular weights, including glucose (180al., 1994; Shamblin et al., 1998; Lueckel et al., D), sucrose (342 D), trehalose (342 D), rafﬁnose1998a). These factors may partly explain why Tgs (505 D), stachyose (667 D), b-cyclodextrin (1135are so different for the same compound, as re- D), dextran 12 (10 kD), and dextran 40 (39 kD)ported by different investigators (Table 1). (Prestrelski et al., 1995). Positive correlation was Generally, the higher the glass transition tem- also observed between the stability of freeze-driedperature, the more stable the protein formulation. invertase at 90°C and the glass transition temper-Therefore, the glass transition temperature may ature of polymer excipients including maltodex-be used with caution as a guiding parameter to trin and PVPs (PVP 10, 40, and 360) (Schebor etscreen protein stabilizers and formulations. Is al., 1996). The denaturation temperature (Tm) ofthere a minimum Tg, which can considered to be several dried proteins, including b-lactoglobulin,adequate to achieve long-term stability for solid ovalbumin, lysozyme, somatotropin, and ribonu-protein pharmaceuticals? It has been recom- clease A, correlated positively with Tg of the sugarmended that the Tg of a stable protein product or polyol additives (Bell et al., 1995; Bell andshould be at least 20°C above ambient storage Hageman, 1996). The thermal stability of proteinstemperature (Franks, 1994). In other words, the in seed axes strongly depends on Tg of the intra-product Tg should be higher than 40°C to achieve cellular glass (Sun et al., 1998). However, therelong-term stability and to tolerate shipping are many examples where formulations of a lowerstresses (Carpenter et al., 1997). When Tg is above Tg are more stable than those of a higher Tg,this temperature, protein molecular mobility is especially when sugar formulations are comparedrestricted, thus minimizing protein reactivity with polymer formulations. For example, the Tg values of both sucrose and trehalose are signiﬁ-(Pikal et al., 1991). cantly lower than those of maltodextrin and PVP Is there any signiﬁcant difference in molecular (40 kD) but both sugars were more effective thanmobility immediately below and above Tg, which the polymers in stabilizing vacuum-dried restric-leads to a signiﬁcant difference in protein stabil- tion enzyme EcoRI during storage at 37 or 45°City? Oksanen and Zograﬁ (1993) found that there (Rossi et al., 1997). Similarly, the Tg of freeze-was no critical (sharp) change in diffusion coefﬁ- dried invertase formulations containing trehalose,cient (mobility) of water in PVP around Tg and maltodextrin, or PVP was 63, 124, and 119°C,signiﬁcant mobility existed below Tg. Signiﬁcant respectively, and the remaining activity of inver-molecular mobility was also found of in- tase in the respective formulations after incuba-domethacin, PVP, and sucrose at temperatures tion at 90°C for 6 h was 74, 47, and 56%immediately below Tg (Hancock et al., 1995). (Cardona et al., 1997). The inferiority of polymersTherefore, signiﬁcant protein degradation does in stabilizing proteins has been attributed to theiroccur below Tg, such as the degradation of inefﬁcient hydrogen bonding with proteins (seelyophilized human insulin via formation of a Section 3.1). In a different study, Chang et al.cyclic anhydride intermediate at C-terminal (1996a) found that the Tg of a lyophilized rhIL-AsnA21 (Strickley and Anderson, 1996). In a re- 1ra formulation containing either phosphate orcent study, thermal inactivation of lactase in a citrate was 26 and 46°C, respectively, but bothlyophilized PVP formulation was investigated at deamidation and aggregation of the protein intemperatures between 37 and 70°C and the inacti- phosphate-containing formulation were signiﬁ-vation rate constant did not show a step change cantly slower during storage at 0°C. They alsoaround the Tg (55°C) of the formulation (Mazzo- found that increasing the sucrose concentrationbre et al., 1997). from 1 to 10% (w/v) in rhIL-1ra formulation (2% How well can Tg be used to predict stability of glycine and 10 mM citrate, pH 6.5) decreased thesolid protein pharmaceuticals? Mixed results have Tg of the lyophilized formulation from 68.5 tobeen obtained. The storage stability of freeze- 64.6°C, but both aggregation and deamidation ofdried IL-2 correlated positively with Tg of the the protein were signiﬁcantly slower during stor-formulations in the presence of stabilizers of dif- age at 50°C.
30 W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60 Why does Tg sometimes fail to predict protein ered to be a measure of the solid-state microenvi-stability? The reason is that Tg is a critical temper- ronment ‘pH’ (Strickley and Anderson, 1997).ature of molecular mobility of amorphous materi- Therefore, solid-state acidity/basicity may still af-als, but may not be a direct indicator of molecular fect protein stability, both physically and chemi-mobility (Yoshioka et al., 1998). For instance, the cally. For example, the formation ofTg of a sucrose-containing chimeric monoclonal non-dissociable aggregates during storingantibody formulation was lower than that of a lyophilized RNase depended on the pH of thetrehalose-containing formulation, but the molecu- protein solution for lyophilization in the followinglar mobility in the sucrose-containing protein for- order, pH 10.0 4.0 6.4 water (Townsendmulation was apparently lower than that in the and DeLuca, 1990). At a storage temperature oftrehalose-containing formulation at low tempera- 25°C, aggregation of a lyophilized antibody-vincatures (B12°C), possibly due to the differences in conjugate from a solution of pH 8.5 was moreglass fragilities of the two formulations (Duddu et than that from pH 7.1 or 6.1 (Roy et al., 1992).al., 1997). Probably due to the lower molecular Both the remaining activity and dimer formationmobility, protein aggregation in the sucrose for- of lyophilized TNF during storage varied withmulation during storage at 5°C was slightly lower solution pHs from 4.0 to 10.0 (Hora et al., 1992a).than that in the trehalose formulation. Therefore, Some lyophilized proteins, however, are insensi-Tg may not be a direct indicator of protein mobil- tive to pH changes. Lyophilization of ovalbuminity. Instead, the Kauzmann temperature (TK), from solutions of different pHs (3.0, 5.5, 7.3, andwhich is the extrapolated isoentropy temperature 9.0) had little effect on the formation of non-co-below both Tm and Tg, has been suggested as valent protein aggregates during storage (Liu etbeing more of a molecular mobility indicator al., 1990).(Hancock and Zograﬁ, 1997; Hancock et al., In a few cases, the acidity/basicity of a solid1998). This is because TK is frequently coincided formulation affects chemical stability of a protein.with the zero mobility temperature (T0) (Angell, The rate and product distribution (between cova-1995). lent dimer and deamidated form) of insulin degra- Recent studies indicate that Tmc appears more dation in a lyophilized formulation depended onclosely related to protein stability than Tg. Tmc the pH of the reconstituted protein solutionwas deﬁned as the molecular mobility-changing (Strickley and Anderson, 1996, 1997). The rate oftemperature at which protein or excipient protons deamidation of lyophilized IL-1ra (2% glycine andin lyophilized formulations begin to exhibit 1% sucrose in 10 mM sodium citrate buffer) atLorentzian relaxation process resulting from 50°C depended on the solution pH in the follow-higher molecular mobility in addition to Gaussian ing order, pH 7.0 6.5 6.0 5.5 (Chang et al.,relaxation process resulting from lower molecular 1996c).mobility (Yoshioka et al., 1997, 1998, 1999). At One of the major chemical aggregation mecha-Tmc, protein formulations transition from a non- nisms is formation of intermolecular disulﬁdeliquidized state to a microscopically liquidized bonds, which requires the presence of a free thio-state. Tmc has been found to be more closely late ion, the reactive group. A free thiolate ionrelated to stability of lyophilized g-globulin than can be brought about upon ionization of a freeTg (Yoshioka et al., 1997, 1998). The molecular thiol group under alkaline conditions or via b-mobility of proteins can be determined by NMR elimination of an intact disulﬁde, a reaction thator dielectric analysis (Pearson and Smith, 1998). is also accelerated under alkaline conditions. Therefore, the solution pH signiﬁcantly affects5.2.3. Formulation pH protein aggregation through these mechanisms. The pH of a protein solution for lyophilization For example, insulin lyophilized at 1 mg ml − 1 atoften affects the stability of dried protein products pH 10 aggregated completely in just 1 day at 50°Cduring long-term storage. The solution pH of a and 96% RH, whereas insulin lyophilized at pHsolid formulation upon reconstitution is consid- 7.3 exhibited aggregation of less than 50% in 3
W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60 31weeks under the same conditions (Costantino et and Zograﬁ, 1994). Depression of Tg by wateral., 1994b). rHA lyophilized at pH 4.0 was much may reach 10° or more for each percent of mois-more stable than that lyophilized at pH 7.3 or 9.0 ture retained, especially at low-level moisture con-against moisture-induced aggregation (Costantino tents (Angell, 1995; Hatley, 1997; Rossi et al.,et al., 1995a). 1997). Therefore, a lyophilized protein may easily The ‘pH’ of a solid protein formulation also has adsorb sufﬁcient amounts of moisture during stor-indirect effects on protein stability. The pH-in- age to reduce its Tg below the storage tempera-duced change in the hygroscopicity of a protein is ture, to accelerate its instability, and to causesuch an effect. At relatively low relative humidity possible product collapse (Oksanen and Zograﬁ,( 5 75%), the equilibrated moisture content of a 1990). For example, dimerization of lyophilizedlyophilized insulin formulation is similar at pH insulin increases signiﬁcantly when water content3.1 and 5.0 at 35°C, but at higher relative humid- is high enough to cause signiﬁcant decrease inity, the low-pH formulation is much more hygro- glass transition temperature and cake collapsescopic than that of high pH (Strickley and (Strickley and Anderson, 1997). High moistureAnderson, 1996). Sucrose at lower pHs may hy- content also facilitates crystallization of formula-drolyze readily to form fructose and glucose, tion excipients such as various sugars, which willwhich can react with proteins via the Maillard be discussed in the next section.reaction (see Section 5.1). It was found that the Generally, increasing moisture content of aactual sucrose content of freeze-dried sucrose lyophilized protein increases the deterioration ratefrom a solution of pH 3 (in citrate and phosphate of proteins. Sorbed water increases the free vol-buffer) dropped to 79% at 4°C in 1 month while ume of a lyophilized protein, promoting molecu-that at pH 9 remained above 95%, apparently due lar mobility (Towns, 1995; Shamblin et al., 1998).to the difference in the rate of sucrose hydrolysis The denaturation temperature (161°C) of driedat different pHs (te Booy et al., 1992; Skrabanja somatotropin decreased with increasing moistureet al., 1994). content until it reached a plateau of 65°C at moisture contents of 28% (Bell et al., 1995).5.2.4. Moisture content Increasing moisture content of lyophilized tPA The residual moisture content after lyophiliza- from 4.6 to 7.6 or 18.0% increased loss of proteintion often controls long-term protein stability, activity during storage at 50°C (Hsu et al., 1991).both physically and chemically (Franks, 1990; Increasing the moisture content of a lyophilizedHatley and Franks, 1991). The moisture content invertase formulation from 2.1 to 10.9% graduallyof a lyophilized protein formulation may change decreased the stability of the protein upon incuba-signiﬁcantly during storage due to a variety of tion at 90°C for 10 h (Cardona et al., 1997).factors, such as stopper moisture release and leak- Increasing relative humidity accelerated activityage (Section 5.4), crystallization of an amorphous loss of vacuum-dried restriction enzyme EcoRI inexcipient (next section), or moisture release from both sucrose and trehalose formulations duringan excipient hydrate (Yu et al., 1999). Water can storage at 45°C (Rossi et al., 1997). Protein aggre-affect protein stability both indirectly as a plasti- gation often increases with increasing relative hu-cizer or reaction medium and directly as a reac- midity during storage, such as insulin (Costantinotant or a product (Shalaev and Zograﬁ, 1996). et al., 1994b) and lysozyme (Separovic et al., Indirectly as a plasticizer, water drastically de- 1998). The aggregation-induced loss of activity ofcreases glass transition temperature of proteins, bovine pancreas RNase was faster at 9.8% mois-polymers or other formulation excipients (Slade et ture than at 1.9% (Townsend and DeLuca, 1990).al., 1989; Roos and Karel, 1991; Buera et al., Moisture uptake was also the cause of reduced1992; te Booy et al., 1992; Roos, 1993; Wolkers et a-helical conformation and increased b-structureal., 1998a). The quantitative effect of water on (a common theme of protein aggregation) ofglass transition temperature can be easily esti- spray-dried rhG–CSF or recombinant consensusmated by the Gordon – Taylor equation (Hancock interferon (rConIFN) (French et al., 1995).
32 W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60 The increased deterioration in lyophilized Changing moisture content in a protein formu-proteins at high moisture contents may result lation may change degradation mechanisms andfrom increased chemical degradations. The forma- thus, the overall rate of degradation. The mecha-tion of degradation products of insulin in nism of aggregation of lyophilized tetanus toxoidlyophilized formulations was directly proportional (150 kD) was different under different relativeto the moisture content of the formulation from 3 humidities during storage. Under 80% RH, all theto 52% (Strickley and Anderson, 1996, 1997). The aggregates were formed by non-disulﬁde covalentincreased rate of a chemical reaction at high water bonding after storing the lyophilized protein atcontents is due to increased mobility of water 37°C for 10 days, but under 97% RH, 55% of theinvolved in the reaction and its positive effect on aggregates were formed by hydrophobic interac-the mobility of proteins (Hageman, 1988; Towns, tion and scrambling of disulﬁde bonds (Schwen-1995). The diffusion coefﬁcient (mobility) of water deman et al., 1996). Hekman et al. (1995) foundin PVP at 25°C can increase exponentially with that increasing moisture content of lyophilizedincreased motion of PVP side chains as the water diethylenetriaminepentaacetic anhydride (DTPA)-content increases (Oksanen and Zograﬁ, 1993). In conjugated IgG above 20% could change the ma-fact, the rate of hydrogen exchange in lysozyme jor degradation pathway from Maillard reactionpowder containing 0.2 g water g − 1 protein (about to one which also involved precipitation during17% moisture) is the same as that in a dilute storage.solution (Rupley and Careri, 1991). The effect of residual moisture on protein sta- bility can be strongly inﬂuenced by several other In many cases, however, the effect of moisture factors, including temperature, excipient composi-on protein stability is a complex function. Chang tion, and moisture distribution within a productet al. (1996c) studied the stability of lyophilized cake. Pikal et al. (1992) demonstrated thatrhIL-1ra (from 50 mg ml − 1 rhIL-1ra, 2% Glycine, lyophilized rGH, in the presence of glycine andand 1% sucrose in 10 mM citrate buffer at pH mannitol, aggregated almost linearly with increas-6.5) at moisture levels between 0.5 and 3.2% ing moisture content from 0.7 to 2.5% during(w/w) at 30°C. They found that the least stable storage at 40°C, but the curve was bell-shaped atformulation had a moisture level of around 0.8% 25°C. The rate of hGH degradation in theand the more stable formulations had a moisture lyophilized formulation varied greatly with for-level of either 5 0.5 or 3.2%. In a few studies, a mulation composition as well as with storageclear bell-shaped relationship was demonstrated temperature. They also found that hGH degrada-between protein stability and moisture content. tion in a lyophilized formulation containingFor example, maximum aggregation (78%) of glycine and mannitol was more sensitive to highlyophilized tetanus toxoid occurred at a water levels of moisture or headspace oxygen than in acontent of about 36% during storage at 37°C for formulation without these excipients (Pikal et al.,10 days, and less aggregation was observed at 1992). Moisture distribution within a proteinwater contents either below or above that level product is usually quite uneven and this may(Schwendeman et al., 1995). Aggregation of both result in a biphasic loss of activity (initial fast(lyophilized) rHA and BSA during storage at phase and remaining slow phase) during storage37°C had a bell-shaped relationship as a function (Franks, 1990).of water content with maximum aggregation atabout 32 and 28% moisture, respectively (Liu et 5.2.5. Type and concentration of formulational., 1990; Costantino et al., 1995b). Similar bell- excipientsshaped aggregation dependence on water content Many formulation excipients stabilize proteinswas also observed for ovalbumin (chicken egg in solid state in a concentration-dependent man-albumin), glucose oxidase, bovine b-lactoglobulin ner (see Section 5.3). On the other hand, overuse(Liu et al., 1990), and insulin (Katakam and of an excipient(s) may eventually destabilize aBanga, 1995; Separovic et al., 1998). protein. Even sugars/polyols, the universal protein
W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60 33stabilizers, may destabilize a protein if their quan- 5.2.6. Crystallization of amorphous excipientstities are not appropriately used in a protein The potential for crystallization of amorphousformulation (see next section and Section 5.3). excipients in a lyophilized protein formulation Certain excipients, that are used to stabilize during storage always exists because the crys-proteins in liquid state or during lyophilization, talline state is more stable thermodynamicallymay destabilize proteins in solid state. Ascorbic (Hancock and Zograﬁ, 1997). Crystallization usu-acid is a frequently used antioxidant for small ally destabilizes a protein due to a loss of intimatedrugs, but at 5 mM, it reduced the storage stabil- excipient interaction with proteins and a possibleity of lyophilized elastase (20 mg ml − 1 in 10 mM decrease in Tg of the amorphous phase. The de-sodium acetate, pH 5.0) drastically at 40°C and crease in Tg arises from increased moisture con-79% RH (Chang et al., 1993). tent of the amorphous phase because the relative Lyophilized protein formulations often contain amount of the amorphous material(s) is reduceda buffering agent(s). Different proteins may need after crystallization and the total amount of mois-different buffering agents for maximum stabiliza- ture in a sealed container does not change. How-tion in solid state. For example, histidine has been ever, if the amorphous excipient crystallizes asshown to be the best buffer agent for minimizing hydrates, such as trehalose dihydrate or rafﬁnoseaggregation of lyophilized recombinant factor IX pentahydrates, the Tg of the amorphous phase can(rFIX) during storage at 30°C among all the actually increase (Aldous et al., 1995).buffering agents examined, including sodium There are many cases of protein destabilization due to excipient crystallization. Inositol crystal-phosphate, potassium phosphate, and Tris (Bush lization in a lyophilized b-galactosidase formula-et al., 1998). Sodium phosphate has been shown tion destabilized the protein during storageto be a better buffering agent than sodium citrate (Izutsu et al., 1994a). The formation of bFGFand sodium maleate in maintaining the solubility degradants in a lyophilized formulation underof lyophilized rIFN-b-1b (Betaseron®) upon re- ‘acidic’ condition was accelerated by sucrose crys-constitution (Lin et al., 1996). In a different study, tallization (Wu et al., 1998). Mannitol crystalliza-glycocholate buffer was found to be better than tion was apparently the cause of a drasticsuccinate buffer in stabilizing lyophilized IFN-g decrease in storage stability for co-spray-driedas about 50% IFN-g activity was lost in a anti-IgE monoclonal antibody at 5 or 30°Clyophilized succinate-containing formulation dur- (Costantino et al., 1998a). Crystallization of ex-ing storage at 25°C for 4 weeks, while about 28% cipients also causes possible ‘pH’ shifting, affect-activity was lost in glycocholate-containing for- ing protein stability as observed in lyophilizedmulation (Lam et al., 1996). bFGF during storage (Wu et al., 1998). The buffer concentration also inﬂuences storage Many sugar/polyol excipients have a tendencystability of lyophilized proteins. The loss of activ- to crystallize during storage. The rate of crystal-ity of lyophilized bovine pancreatic RNase in lization increases with increasing temperature orsodium phosphate buffer (pH 6.4) during storage relative humidity (Schmitt et al., 1999). Immediateat 45°C increased with increasing buffer concen- crystallization of disaccharides can occur whentration in the range of 0, 0.02, and 0.2 M these sugars are heated to about 50°C above their(Townsend and DeLuca, 1990). One possible Tgs (Cardona et al., 1997). Lyophilized sucrose incause is that buffering agents may change Tg of a LDH formulation (in 10 mM PBS, pH 7.2) islyophilized formulations. The Tg of lyophilized mainly amorphous but tends to crystallize withrhIL-1ra formulation containing 1% sucrose, 4% increasing residual water content or during heat-mannitol, and 2% glycine decreased from 46 to ing or storage at elevated temperatures (Moreira26°C when the buffering agent sodium citrate was et al., 1998). Spray-dried amorphous lactose, su-replaced with sodium phosphate (Chang et al., crose, or trehalose can easily crystallize at 25°C1996a). A lower Tg generally leads to a less stable and under RH of ] 52% (Naini et al., 1998).formulation (see Section 5.2). Both lyophilized and spray-dried rhDNase–lac-
34 W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60tose mixture could easily crystallize upon expo- ents (also see Section 5.4). Inclusion of sodiumsure to high humidity (]60%) (Chan and Gonda, phosphate in a spray-dried formulation contain-1998; Chan et al., 1999). An amorphous sucrose ing mannitol and an anti-IgE monoclonal anti-formulation could change to a crystalline form body inhibited mannitol crystallization, which incompletely in 1 month at 60°C (te Booy et al., turn reduced solid-state protein aggregation dur-1992). Because of sucrose crystallization, trehalose ing storage at 5 or 30°C (Costantino et al.,stabilized lyophilized rFXIII and Humicola lanug- 1998a). Addition of phosphate and citrate in ainosa lipase much more effectively than sucrose lyophilized sucrose formulation inhibited the crys-during storage at 60°C (Kreilgaard et al., 1998a, tallization of sucrose as demonstrated by DSC (te1999). In a more detailed study, it was demon- Booy et al., 1992).strated that increasing moisture content dropped Many polymer excipients (or proteins) haveroughly linearly the crystallization temperature of been shown to inhibit the crystallization of smallsucrose or recombinant human somatotropin carbohydrate excipients. Maltodextrins or PVP(rbSt) (Sarciaux and Hageman, 1997). Therefore, effectively retard the crystallization of sucrosemoisture content in a protein formulation should (Roos and Karel, 1991; Shamblin et al., 1996).be minimized to prevent crystallization of an Dextran or carboxymethyl cellulose sodium saltexcipient(s). (CMC-Na) effectively inhibits the crystallization Crystallization tendency of amorphous excipi- of inositol (Izutsu et al., 1994a). PVP can reduceents is strongly affected by their relative amount the tendency of moisture-induced crystallizationin a protein formulation. For example, amor- of lyophilized sucrose (Shamblin and Zograﬁ,phous mannitol, sorbitol, or methyl a-D- 1999). Both catalase and insulin can inhibit lac-mannopyranoside in a rhGH formulation at an tose crystallization in a co-spray-dried amorphousexcipient:rhGH molar ratio of less than 131:1 did mixture upon exposure to short-term elevated hu-not crystallize easily during storage at 50°C for 4 midity (Forbes et al., 1998).weeks, but crystallization occurred at 1000:1 mo- To examine excipient crystallization in alar ratio (Costantino et al., 1998b). Increasing the protein formulation during storage, X-ray diffrac-relative content of sucrose signiﬁcantly decreased tometry or IR can be used (Izutsu et al., 1994a;its crystallization temperature in a lyophilized su- Kreilgaard et al., 1999). Recently, Raman spec-crose:rhGH formulation (Costantino et al., 1998c) troscopy has been shown to detect 1% amorphousand facilitated its crystallization under moisture in or crystalline content and proved to be anothera spray-dried sucrose:trypsinogen formulation useful method for crystallinity detection (Taylor(Tzannis and Prestrelski, 1999b). Mannitol at 10 and Zograﬁ, 1998a).or 20% stabilized a co-spray-dried anti-IgE mono-clonal antibody during storage at 5 or 30°C, but 5.2.7. Reconstitution mediummannitol at 30% crystallized and drastically desta- The reconstitution step may potentially affectbilized the protein during storage (Costantino et protein stability by several mechanisms. First,al., 1998a). Similarly, inositol at concentrations rapid reconstitution with water may not allow abetween 50 and 160 mM in a lyophilized b-galac- dried protein to rehydrate as slowly as the dehy-tosidase formulation did not crystallize during dration step (Cleland et al., 1993). Therefore, astorage at 70°C for 7 days but crystallized at protein may not be able to refold to its nativeconcentrations above 250 mM (Izutsu et al., form during reconstitution, causing denaturation1994a). DSC data indicated that the crystalliza- and/or aggregation. Second, the pH of water-re-tion temperature of inositol decreased from 110 to constituted protein formulations may be different60°C when the inositol concentration in the for- from those before lyophilization, possibly due tomulation increased from 50 to 500 mM. loss of some formulation components. For exam- Crystallization tendency of amorphous excipi- ple, three insulin solutions have pHs of 2.0, 3.0ents also changes depending on the type and and 4.0 before lyophilization, and the respectiveconcentration of co-existing formulation excipi- pHs of the reconstituted lyophilized insulin were
W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60 35found to be 3.1, 3.3 and 4.1, presumably due to been discussed in Section 3. The following sectionloss of HCl during lyophilization (Strickley and focuses on protein stabilization for long-termAnderson, 1996). Lyophilization-induced evapo- storage.ration of acetic acid can also cause similar pHchanges (Hatley et al., 1996). Last, many proteins 5.3.1. Stabilization mechanismsare temperature-sensitive and the temperature of Stabilization mechanisms for lyophilizedthe reconstitution medium can make a signiﬁcant proteins during long-term storage are similar todifference. For instance, the activity of H+-AT- those for protein lyoprotection (see Section 3.2).Pase was completely recovered when it was rehy- These stabilization mechanisms include formationdrated at 20°C but the recovered activity was of an amorphous glassy state, water replacementreduced to 62% when it was rehydrated at 30°C by excipients, and hydrogen bonding between ex-(Sampedro et al., 1998). cipients and proteins (Fox, 1995). It has been To alleviate these problems, addition of a sur- shown that excipients capable of replacing waterfactant and/or a stabilizer(s) in the reconstitution molecules upon dehydration better preserve themedium or adjustment of medium pH may be native structure of proteins, resulting in enhancedrequired. For example, inclusion of Tween 20 in stability (Prestrelski et al., 1995). Nevertheless, itthe reconstitution medium decreased aggregation is now clear that glass formation alone is notof interferon-g compared with reconstitution with sufﬁcient for protein stabilization in solid state.water alone (Webb et al., 1998). Addition of The supporting evidence for this argument is thatmaltose in the reconstitution buffer increased ac- glass-forming polymers fail to stabilize or eventivity recovery of the dried restriction endonucle- destabilize solid proteins during storage (Colacoase HindIII (Uritani et al., 1995). The amount of et al., 1992; Nakai et al., 1998; Lueckel et al.,aggregates in reconstituted IL-2 formulation 1998b). Therefore, a combination of these mecha-could be reduced by inclusion of one of the nisms is required for maximum protein stabiliza-following excipients in the medium: heparin, dex- tion in solid state (Crowe et al., 1996, 1998).tran sulfate, glycine, lysine – HCl, polylysine, In addition to the above-mentioned mecha-Tween 20, or HP-b-CD (Zhang et al., 1996). nisms, other stabilization hypotheses have alsoLowering reconstitution solution pH from 7.0 to been proposed. One of them is the prevention of4.0 also lowered the amount of aggregates after protein–protein interaction and aggregation byreconstitution. About 11% of recombinant human physical dilution and separation of proteinkeratinocyte growth factor (rhKGF) formed ag- molecules. The reduced aggregation of lyophilizedgregates immediately after reconstitution of the BSA in the presence of NaCl, sodium phosphate,lyophilized KGF with water, but only about 1.5% carboxymethyl-cellulose, dextran, DEAE-dextran,formed aggregates after reconstitution with water or PEG on incubation at 37°C has been attributedcontaining 0.05% (w/v) heparin (16 kD) or su- to a general dilution effect (Liu et al., 1990). Thiscrose octasulfate (SOS) (Zhang et al., 1995). mechanism has also been suggested in stabiliza-Other stabilizers that inhibited reconstitution-in- tion of other proteins during storage such as rHAduced aggregation included dextran sulfate, fu- by dextran (Costantino et al., 1995a), IL-1ra bycoidan, pentosan polysulfate, chondroitin sulfate, sucrose (Chang et al., 1996c), and insulin bymyo-inositol sulfate, sulfated b-cyclodextrin, trehalose (inhibition of covalent dimerization) atpolyphosphoric acid, NaCl, (NH)2SO4, etc. 35°C (Strickley and Anderson, 1997). Another stabilization hypothesis is excipient–5.3. Stabilization of solid proteins by excipients water interactions. Costantino et al. (1995a) demonstrated that rHA co-lyophilized with NaCl Solid protein pharmaceuticals may need to be (NaCl:protein= 1:6 on a weight basis) did notstabilized in two stages, during lyophilization exhibit any aggregation after a 4-day incubation(preparation); and during long-term storage. period at 37°C and 96% RH, while the proteinProtein stabilization during lyophilization has without the excipient lost over 80% solubility after
36 W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60just 1 day under the same conditions. Since inclu- inhibit the aggregation rate of rhIL-1ra signiﬁ-sion of NaCl did not induce any signiﬁcant cantly in a concentration range of 0–10% (w/v)change in the secondary structure of rHA after (Chang et al., 1996a,c), and to stabilize rFXIII atlyophilization, the stabilization effect of NaCl was 100 mM during storage at 40°C (Kreilgaard et al.,apparently due to water uptake by NaCl in the 1998a). Sucrose has also been used to improve thevicinity of rHA, which facilitated protein refold- storage stability of factor IX (FIX) (Bush et al.,ing into its native and more stable conformation. 1998) and rIFN-b-1b (Betaseron®) (Lin et al.,This stabilization mechanism was clearly demon- 1996). Trehalose increased the stability ofstrated in a similar study, in which it was found lyophilized invertase during incubation at 90°Cthat the effectiveness of several excipients (D-glu- (Schebor et al., 1996; Cardona et al., 1997). Bothcaric acid, D-gluconic acid, sorbitol, sodium chlo- sucrose and trehalose inhibited the dimer forma-ride, etc.) in inhibiting aggregation of lyophilized tion in TNF during storage at 37°C (Hora et al.,rHA roughly correlated with their water uptake 1992a). Trehalose or lactose at or above 300:1under 96% RH (Costantino et al., 1995b). The (excipient:protein) molar ratio inhibited aggrega-greater the excipient’s afﬁnity for water, generally, tion of a spray-dried anti-IgE monoclonal anti-the greater the stabilizing effect. However, if a body effectively during storage at 30°C underprotein is very sensitive to moisture, this stabiliza- both 11 and 38% RH (Andya et al., 1999). Man-tion mechanism is not applicable. For example, nitol has been shown to decrease aggregation ofthe aggregation of moisture-sensitive TT in a lyophilized TT during storage at 37°C and 86%lyophilized formulation containing NaCl, D-sor- RH (Costantino et al., 1996) and lyophilized anti-bitol, or PEG 20 000 (at an excipient/protein body-vinca conjugate during storage at 25°C (Royweight ratio of 1:5) did not correlate with water et al., 1992). Sorbitol could prevent aggregationuptake in these formulations during storage at of rHA completely at 1:1 (excipient:protein)37°C and 86% RH (Schwendeman et al., 1995). weight ratio during storage at 37°C for 4 days Polymers may stabilize proteins by increasing (Costantino et al., 1995b) and reduce aggregationTg of a solid protein formulation, because poly- of lyophilized TT effectively at 1:5 weight ratiomers usually have higher Tgs due to their high during storage at 37°C and 86% RH for 6 daysmolecular weights (te Booy et al., 1992; Prestrelski (Schwendeman et al., 1995). Inositol was able toet al., 1995). The Tgs of maltodextrin and PVP protect lyophilized b-galactosidase in a concentra-were found to increase linearly with their molecu- tion-dependent manner during storage at temper-lar weights, as described by the Fox and Flory atures below 50°C (Izutsu et al., 1994a). A varietyequation (Roos and Karel, 1991; Buera et al., of mono-, di-, and trisaccharides or polyols pro-1992). Because of this effect, the Tg of dextran- tected lyophilized restriction enzyme PstI to vari-formulated g-globulin formulations increased sig- ous degrees during storage at 37°C (Colaco et al.,niﬁcantly with increasing molecular weight of 1992). More examples can be found in Table 2.dextran from 10 to 510 kD (Yoshioka et al., As used in lyoprotection, sucrose and trehalose1997). In addition, polymers (or proteins) may seem to be the most commonly used disaccharidesindirectly stabilize proteins by inhibiting the crys- for protection of solid proteins during long-termtallization of other stabilizing excipients in a solid storage. Their relative effect in stabilizing solidformulation (see Section 5.2). proteins is still a much-debated subject. Clearly, trehalose has several advantages over sucrose, as5.3.2. Sugars/polyols discussed in Section 3.1. In addition, formation of Sugars or polyols have commonly been used to a dihydrate during storage of freeze-dried tre-stabilize lyophilized proteins for long-term stor- halose has been demonstrated to prevent mois-age, such as sucrose, trehalose, mannitol, sorbitol, ture-induced decrease in glass transitionetc. Sucrose has been shown to prevent aFGF temperature, which usually destabilizes proteinsaggregation completely at 2% during storage at (Aldous et al., 1995; Crowe et al., 1996). These25°C for 1 year (Volkin and Middaugh, 1996), to advantages might explain why trehalose was
Table 2Examples of protein instability and stabilization by excipientsProteins Formulation compositions Study conditions Remaining Activity References (or as stated)Actin 10 mg ml−1 actin in 1 mM Tris (pH 8.0), Freeze-drying 24% Allison et al., 1998 0.1 mM ATP, 0.25 mM 2 mercaptoethanol, 0.1 mM CaCl2, and 0.0025% NaN3 +1% (w/v) sucrose 33% +5% (w/v) sucrose 67%ADH (baker’s yeast) 50 mg ml−1 in 20 mM potassium phosphate Freeze-drying 55% Ramos et al., 1997 buffer (pH 7.6) +500 mM KCl 34% +500 mM trehalose 98% +500 mM mannosylglycerate 89%Bilirubin oxidase 0.4 mg ml−1 in 0.2 mg ml−1 Triton X-100 Freeze-drying 56% Nakai et al., 1998 and 40 mg ml−1 dextran 0.4 mg ml−1 in 0.2 mg ml−1 Triton X-100 58% and 40 mg ml−1 PVA 0.4 mg ml−1 in 0.2 mg ml−1 Triton X-100 85% and 40 mg ml−1 trehaloseCatalase (ox liver) 8.4 mg ml−1 in 10 mM phosphate buffer Freezing for 10% at temperatures 80% Shikama and Yamazaki, (pH 7.0) between −15 and −70°C 1961 +1 M glycerol 95% +0.5 mM sodium dodecyl sulfate 90%Catalase (beef liver) 1 mg ml−1 in 50 mM phosphate buffer Freeze-drying 35% Tanaka et al., 1991 +0.2 mg ml−1 maltose 70% +1 mg ml−1 maltose 90% +1.3 mg −1ml maltose 90%Elastase Lyophilized formulation at 20 mg ml−1 in 10 Incubation at 40°C, 79% RH for 2 33% Chang et al., 1993 mM sodium acetate (pH 5.0) weeks +10% sucrose 86% W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60 + 10% dextran 40 82%.b-Galactosidase 20 mg ml−1 in 200 mM phosphate buffer (7.4) Freeze-drying 47% Izutsu et al., 1994a and 100 mM inositol +400 mM inositol 90% +400 mM inositol and 1 mg ml−1 dextran 100% +400 mM inositol and 1 mg ml−1 CMC-Na 100%b-Galactosidase 2 mg ml−1 in 10 mM sodium phosphate, (pH 7.4) Freeze-drying 5% Izutsu et al., 1993 +50 mM mannitol 95% +200 mM mannitol 70% +500 mM mannitol 25% 37
38Table 2 (Continued)Proteins Formulation compositions Study conditions Remaining Activity References (or as stated)b-Galactosidase 2 mg ml−1 in 200 mM sodium Freeze-drying 5% Izutsu et al., 1993 phosphate, pH 7.4 +50 mM mannitol 40% +200 mM mannitol 95% +500 mM mannitol 95%rFXIII 2 mg ml−1 in 0.1 mM EDTA and 10 mM Tris Freeze-drying 82% Kreilgaard et al., buffer (pH 8) 1998a +100 mM mannitol 90% +1% (w/v) PEG 115% +3.5% (w/v) dextran 109% +100 mM trehalose 117% +100 mM sucrose 111% +0.002% (w/v) Tween 20 106%GDH 50 mg ml−1 in 20 mM potassium phosphate Freeze-drying 65% Ramos et al., 1997 buffer (pH 7.5) +500 mM KCl 32% +500 mM trehalose 86% +500 mM mannosylglycerate 95%hGH Lyophilized formulation at 2 mg ml−1 Incubation at 50°C Rate constant = 4.5 per day Costantino et al., 1998b +mannitol at 31:1 molar ratio 0.004 per day +trehalose at 31:1 molar ratio 0.002 per day +cellobiose at 31:1 molar ratio 0.0003 per dayH+-ATPase 1 mg ml−1 in 1 mM EGTA-Tris buffer (pH 7.0) Freeze-drying 4% Sampedro et al., 1998 +20 mg trehalose per mg protein 100% +20 mg maltose per mg protein 91% +20 mg sucrose per mg protein 84% +20 mg glucose per mg protein 72% W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60 +20 mg galactose per mg protein 37%rhIL-1ra Lyophilized formulation at 10 mg ml−1 rhIL-1ra, Incubation at 50°C Degradation at 1.8% per Chang et al., 1996c 2% glycine and 10 mM sodium citrate (pH 6.5) week, aggregation at 0.08 (A500nm) per week +1% sucrose (w/v) 0.15% and 0.008, respectively +1% maltose 6.4% and 0.003, respectively +1% sorbitol 2.3% and 0.04, respectively +1% trehalose 0.43% and 0.005, respectivelyIL-2 Lyophilized formulation at 0.8 mg ml−1 in MES Incubation at 45°C for 4 weeks 47% Prestrelski et al., buffer (pH 7) 1995 +0.5% sucrose 86% +5% mannitol 28%
Table 2 (Continued)Proteins Formulation compositions Study conditions Remaining Activity References (or as stated)Insulin Lyophilized formulation at 1 mg ml−1 (pH 4) Incubation at 35°C for 7 days 0.04 mg ml−1 degradant Strickley and (18% dimers) Anderson, 1997 +5 mg ml−1 trehalose 0.02 mg ml−1 degradant (1% dimers)LDH About 0.5 mg ml−1 (0.25 mg ml−1 rabbit muscle Freeze-thawing 76% Anchordoquy and and 0.25 mg ml−1 porcine heart LDH) in 0.1 M Carpenter, 1996 NaCl and 10 mM potassium phosphate, pH 7.5) +0.1% (w/v) BSA 80% +1.0% (w/v) BSA 100% +0.1% (w/v) PVP 79% +1.0% (w/v) PVP 98%LDH (rabbit muscle) 25 mg ml−1 in water Freeze-thawing 19% Alden and ´ + 1% PEG 79% Magnusson, 1997LDH (rabbit muscle) 9 mg ml−1 in water Freeze-thawing 36% Nema and Avis, 1992 +1% (w/v) methocel 82% +1% (w/v) Pluronic F127 87% +2.5% (w/v) PVP 76% +0.2 M PEG 400 85% +0.5% (w/v) gelatin 91% +1% (w/v) BSA 100% +0.9% (w/v) b-cyclodextrin 31% +5% (w/v) dextran 76% +5% (w/v) trehalose 26% +5% (w/v) mannitol 6% +5% (w/v) sucrose 73% +10% (w/v) sucrose 88% +34.2% (w/v, 1 M) sucrose 102% W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60 +0.05% (w/v) Brij 30 100% +0.002% Tween 80 57% +0.005% Tween 80 65% +0.05% Tween 80 79% +0.1% Tween 80 81% +0.5% Tween 80 80% +1% Tween 80 83%LDH 2 mg ml−1 in 50 mM sodium phosphate Freeze-drying 9% Izutsu et al., 1995 buffer (pH 7.4) 39
40Table 2 (Continued)Proteins Formulation compositions Study conditions Remaining Activity References (or as stated) +400 mM mannitol 14% +200 mM sucrose 30% +10 mg ml−1 CHAPS 76% +10 mg ml−1 HP-b-CD 73% +10 mg ml−1 sodium cholate 60% +10 mg ml−1 PEG 3000 30% +10 mg ml−1 PEG 20 000 38%LDH About 0.5 mg ml−1 (0.25 mg ml−1 rabbit muscle Freeze-drying 66% Anchordoquy and and 0.25 mg ml−1 porcine heart LDH) in 0.1 M Carpenter, 1996 NaCl and 10 mM potassium phosphate, (pH 7.5) +0.1% (w/v) BSA 65% +1.0% (w/v) BSA 86% +10.0% (w/v) BSA 92% +0.1% (w/v) PVP 77% +1.0% (w/v) PVP 70% +1.0% (w/v) PVP 97%LDH 50 mg ml−1 in 50 mM Tris–HCl (pH 7.2) Freeze-drying 27% Andersson and Hatti-Kaul, 1999 +0.01% polyethyleneimine (low MW) 43% +0.1% polyethyleneimine (low MW) 47% +1% polyethyleneimine (low MW) 69% +0.01% polyethyleneimine (high MW) 52% +0.1% polyethyleneimine (high MW) 64% +1% polyethyleneimine (high M.W.) 69%LDH Lyophilized formulation at 50 mg ml−1 in Incubation at 36°C for 1 week 6% Andersson and 50 mM Tris–HCl (pH 7.2) Hatti-Kaul, 1999 +0.01% polyethyleneimine (low MW) 27% +0.1% polyethyleneimine (low MW) 31% W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60 +1% polyethyleneimine (low MW) 59% +0.01% polyethyleneimine (high MW) 21% +0.1% polyethyleneimine (high MW) 51% +1% polyethyleneimine (high MW) 64%LDH 2 mg ml−1 in 20 mM potassium phosphate Freeze-drying 46% Prestrelski et al.,(rabbit muscle) buffer (pH 7.5) 1993a +1% PEG 46% +10 mM mannitol 39% +10 mM lactose 43% +10 mM trehalose 39% +1% PEG and 10 mM mannitol 81%
Table 2 (Continued)Proteins Formulation compositions Study conditions Remaining Activity References (or as stated) +1% PEG and 10 mM lactose 100% +1% PEG and 10 mM trehalose 90%LDH (rabbit muscle) 50 mg ml−1 in 20 mM potassium phosphate buffer Freeze-drying 12% Ramos et al., 1997 (pH 7.6) +500 mM KCl 0% +500 mM trehalose 54% +500 mM mannosylglycerate 85%PFK 2 mg ml−1 in 20 mM potassium phosphate Freeze-drying 0% Prestrelski, et al., buffer (pH 8.0) 1993a +1% PEG 20% +10 mM mannitol 5% +10 mM lactose 5% +10 mM trehalose 0% +1% PEG and 10 mM mannitol 30% +1% PEG and 10 mM lactose 103% +1% PEG and 10 mM trehalose 100%PFK 50 mg ml−1 in 10 mM potassium phosphate Freeze-thawing 0% Carpenter and (pH 8.0) and 5 mM dithiothreitol Crowe, 1989 +150 mg ml−1 trehalose 65% +300 mg ml−1 trehalose 90%PFK 25 mg ml−1 in 1 mM sodium borate (pH 7.8), Freeze-drying 0% Carpenter et al., 25 mM K2SO4, 2 mM (NH4)2SO4, and 5 mM 1986 dithiothreitol +1 mM ZnSO4 B13% +50 mM trehalose, sucrose, or maltose B10% +1 mM ZnSO4 and 50 mM trehalose 80% +1 mM ZnSO4 and 50 mM sucrose, or maltose 60% +1 mM ZnSO4 and 50 mM glucose, glycerol or 85% W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60 inositolPFK 40 mg ml−1 in 1 mM sodium borate (pH 7.8), Freeze-drying 0% Carpenter et al., 25 mM K2SO4, 2 mM (NH4)2SO4 and 5 mM 1987 dithiothreitol +ZnSO4 (up to 1 mM) 0% +400 mM trehalose or maltose 80% +400 mM sucrose 60% +400 mM glucose or galactose 0% +0.9 mM ZnSO4 and 100 mM trehalose or 80% maltose +0.9 mM ZnSO4 and 100 mM glucose 90% 41
42Table 2 (Continued)Proteins Formulation compositions Study conditions Remaining Activity References (or as stated) +0.9 mM ZnSO4 and 100 mM galactose 85%PFK 50 mg ml−1 in 10 mM potassium phosphate Freeze-drying 0% Carpenter and (pH 8.0) and 5 mM dithiothreitol Crowe, 1989 +150 mg ml−1 trehalose 45% +300 mg ml−1 trehalose 20%PFK 150 mg ml−1 in 10 mM potassium Freeze-drying 0% Carpenter and phosphate (pH 8.0) and 5 mM dithiothreitol Crowe, 1989 +50 mg ml−1 myo-inositol Freeze-drying 20% +100 mg ml−1 myo-inositol Freeze-drying 5% W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60
W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60 43much more effective than sucrose in stabilizing formulation exhibited a drastic decrease in stabil-lyophilized restriction enzyme PstI during storage ity during storage due to mannitol crystallizationat 37°C (Colaco et al., 1992). Nevertheless, the (Costantino et al., 1998a). Costantino et al.relative effect of the two disaccharides in stabiliz- (1998b) demonstrated that mannitol, sorbitol, oring solid proteins depends on many factors, such methyl a-D-mannopyranoside inhibited aggrega-as the relative concentration of the disaccharides tion of lyophilized rhGH maximally at an excipi-and the storage temperature at which stability ent:rhGH molar ratio of 131:1 during storage atstudies are conducted. Duddu and Dal Monte 50°C for 4 weeks. However, at higher ratios of(1997) demonstrated that the aggregation rate for 300:1 and 1000:1, rhGH showed increased forma-a lyophilized monoclonal antibody (2.5 mg ml − 1) tion of insoluble aggregates, apparently due towas about the same in the presence of either 31.3 crystallization of these excipients. In contrast, dis-mg ml − 1 sucrose or trehalose during storage at 5 accharides (lactose, trehalose, and cellobiose) wereor 40°C for 2 months. However, at 60°C the more effective and stabilized the protein continu-aggregation percentage (6.0%) in sucrose formula- ally at ratios of 131:1 and above. In a recenttion was signiﬁcantly higher than that (1.1%) in study, Tzannis and Prestrelski (1999a) demon-trehalose formulation, apparently because the strated that the thermal stability (Tm) of trypsino-storage temperature was higher than the glass gen in a spray-dried sucrose formulation wastransition temperature of the sucrose formulation highest at a mass ratio (sucrose:trypsinogen) of(59°C by DSC) but lower than that of the tre- 1:1 and varying the mass ratio in either directionhalose formulation (80°C by DSC). Both sucrose decreased the Tm.and trehalose have been shown to stabilize Sugars/polyols may inhibit chemical degrada-lyophilized rFXIII and Humicola lanuginosa lipase tions in solid protein formulations. Sucrose atto a similar level during storage at 40°C for 3 10% signiﬁcantly decreased deamidation ofmonths, but sucrose was much less effective at lyophilized rhIL-1ra at 50°C (Chang et al.,60°C due to signiﬁcant crystallization (Kreilgaard 1996a). Covalent dimerization of lyophilized in-et al., 1998a, 1999). It has been shown that the sulin could be signiﬁcantly suppressed in the pres-tendency of crystallization of sucrose is higher ence of trehalose (5 mg ml − 1) at 35°C (Strickleythan that of trehalose during storage at tempera- and Anderson, 1997).tures above Tg (Hatley, 1997). Partly because ofthe difference in crystallization tendency, the glu- 5.3.3. Polymerscose/trehalose (1:10, w/w) formulation preserved Many polymers can increase the long-term sta-the activity of G6PDH more effectively than the bility of lyophilized proteins. Dextran, CMC,glucose/sucrose (1:10, w/w) formulation at storage DEAE-dextran, and PEG have been shown totemperatures between 33 and 90.5°C (Sun and reduce aggregation of lyophilized BSA signiﬁ-Davidson, 1998). cantly during storage at 37°C (Liu et al., 1990). The level of protein stabilization afforded by HP-b-CD was found to stabilize a mouse mono-sugars/polyols during long-term storage varies sig- clonal antibody during storage at 56°C (Ressingniﬁcantly. In certain cases, they destabilize et al., 1992), to inhibit moisture-induced aggrega-proteins, particularly at high concentrations. A tion of solid insulin (Katakam and Banga, 1995),frequent cause of destabilization is crystallization to stabilize IL-2 against aggregation during stor-of these excipients during storage. Mannitol at age at 5°C (Hora et al., 1992b), and to inhibit the100 mM does not improve the storage stability of dimerization of TNF during storage at 37°Clyophilized rFXIII at 40 or 60°C due to mannitol (Hora et al., 1992a). Dextran 40 at 10% increasedcrystallization (Kreilgaard et al., 1998a). A co- the activity of lyophilized elastase at 20 mg ml − 1spray-dried formulation containing recombinant (in 10 mM sodium acetate, pH 5.0) from 33 tohumanized anti-IgE monoclonal antibody and 82% during storage for 2 weeks at 40°C and 79%10% (or 20%) mannitol was stable during storage RH (Chang et al., 1993). Dextran (162 kD) at 3.5at 5 or 30°C, but the antibody in 30% mannitol and 5% (w/v) improved the storage stability of
44 W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60lyophilized rFXIII and Humicola lanuginosa li- freezing and thawing, but did not provide enoughpase, respectively, at 40 or 60°C (Kreilgaard et al., protection for the lyophilized product during stor-1998a, 1999). Both PVPs and maltodextrin stabi- age. Inclusion of 0.002% Tween 20 in alized lyophilized invertase during incubation at lyophilized rFXIII formulation did not improve90°C (Schebor et al., 1996; Cardona et al., 1997). its storage stability at 40 and 60°C (Kreilgaard etPolyethyleneimine was shown to increase the stor- al., 1998a). Although Tween 80 at 0.05 or 0.1%age stability of lyophilized LDH in a concentra- could inhibit spray-drying inactivation of LDH, ittion-dependent manner at 36°C (Table 2). actually destabilized the dried protein during stor- However, polymers may not always stabilize age at 25, 40 and 60°C (Adler and Lee, 1999).solid proteins and, in certain cases, have adverseeffect, as discussed in Section 3.1. For example, 5.3.6. Miscellaneous compoundsneither insulin nor dextran could stabilize A combination of 2.0% arginine and 2.3% car-lyophilized restriction enzyme PstI during storage nitine signiﬁcantly decreased aggregation ofat 37°C (Colaco et al., 1992). Inclusion of dextran lyophilized IL-2 during storage at 37°C for 440 in a lyophilized IL-6 formulation containing weeks (Hora et al., 1992b). Combined use ofsucrose signiﬁcantly increased protein aggregation phenylalanine, arginine, and a mineral acid inhib-during storage at 40°C for 9 months (Lueckel et ited aggregation of vacuum-dried rhG–CSF oral., 1998b). The destabilization can be attributed LDH during storage at 40°C (Mattern et al.,to a failure of inﬂexible dextran molecules to 1999). Several excipients, such as D-glucaric acidinteract with the protein effectively by hydrogen and D-gluconic acid, have been shown to inhibitbonding. Apparently for the same reason, the aggregation of lyophilized albumin during storageactivity of lyophilized BO in a dextran formula- (Costantino et al., 1995b).tion decreased faster than that in a PVA formula- Calcium ions has been shown to protect solidtion during storage at 70°C (Nakai et al., 1998). rhDNase signiﬁcantly against aggregation during storage at 40°C (Chen et al., 1999).5.3.4. Salts Salts have been shown to stabilize proteins in a 5.4. Formulation of solid protein pharmaceuticalsfew cases. Liu et al. (1990) found that NaCl orsodium phosphate could signiﬁcantly inhibit ag- Although signiﬁcant progress has been made ingregation of lyophilized BSA (in water) on incu- the past decade in protein formulation, there isbation at 37°C. rHA co-lyophilized with NaCl at still no single pathway to follow in formulating aa NaCl:protein weight ratio of 1:6 did not aggre- solid protein product. In most cases, solid proteingate upon incubation at 37°C and 96% RH for 4 products have been developed on a trial-and-errordays, while the protein without NaCl lost over basis.80% solubility in just one day under the same To achieve successful formulation of solidconditions (Costantino et al., 1995b). NaCl at an protein products by lyophilization, one or more ofexcipient:protein weight ratio of 1:5 was also able the following formulation excipients may beto reduce aggregation of lyophilized TT during needed: a buffering agent(s), a bulking agent(s), astorage at 37°C and 86% RH for 6 days (Schwen- protein stabilizer(s), and an antimicrobialdeman et al., 1995). agent(s), although stable proteins, such as recom- binant human tPA, may be lyophilized without5.3.5. Surfactants any of the above agents (Overcashier et al., 1997). Although surfactants may be effective proteinstabilizers during lyophilization, they seem to be 5.4.1. Selection of a formulation pHincapable of stabilizing proteins effectively during Lyophilization starts with preparation of along-term storage, based on a limited number of protein solution. Many proteins in solution arestudies. Bush et al. (1998) demonstrated that stable only in a narrow pH range (see SectionTween 80 could inhibit aggregation of FIX during 2.2). In addition, pH can strongly affect the solu-
W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60 45bility of certain proteins, such as rIFN-b-1b (Be- can cause a signiﬁcant pH drop during freezing,taseron®), a hydrophobic protein (Lin et al., denaturing proteins (see Section 2.2). Instead,1996). Therefore, an optimum pH is needed to potassium phosphate, citrate, histidine, and Triskeep a protein stable and soluble in solution. can be used due to their minimal changes in pH The solution pH has also been shown to affect during freezing (Franks, 1990; Carpenter et al.,stability of a protein during freezing and/or 1997). In addition, decreasing the buffer concen-freeze-drying. Freeze-thawing ovalbumin at neu- tration can also mitigate the pH shift (Pikal,tral pH did not cause denaturation, but induced 1999).signiﬁcant structural changes at pH 1.9 (Koseki et Since the effect of different buffering agents onal., 1990). Freezing human growth hormone long-term stability of lyophilized proteins is usu-(hGH) at pH 7.4 resulted in the formation of ally unpredictable (see Section 5.2), selection of amore insoluble aggregates than at pH 7.8 (Eck- buffering agent(s) can only rely on stability stud-hardt et al., 1991). Lyophilization of IL-2 at pH 7 ies. In addition, the selection of a proper bufferinduced signiﬁcant irreversible protein unfolding concentration is also important, as the bufferand aggregation while at pH below 5, the protein concentration not only affects the storage stabilityremained essentially native (Prestrelski et al., of lyophilized proteins (see Section 5.2) but also1995). Thus, the solution pH must be optimal to plays a critical role in stabilizing proteins duringminimize protein denaturation during lyophilization. For example, mannitol could notlyophilization. protect b-galactosidase (2 mg ml − 1) in water dur- On top of these effects, the formulation pH ing lyophilization due to mannitol crystallizationmay have signiﬁcant impact on long-term stability (Izutsu et al., 1993). In the presence of 10 mMof solid protein pharmaceuticals, as discussed in sodium phosphate buffer (pH 7.4), crystallizationSection 5.2. Therefore, the formulation pH should of mannitol at 50 mM was inhibited, and aboutbe optimal to allow maximum long-term stability 95% of the enzyme activity was protected. In-for lyophilized proteins. creasing the concentration of sodium phosphate To meet all these requirements in different buffer (pH 7.4) to 200 mM completely inhibitedstages, the formulation pH needs to be carefully the crystallization of mannitol (500 mM) in b-chosen. Proper selection of a solution pH is the galactosidase (2 mg ml − 1) solution duringﬁrst step toward stabilization of solid protein lyophilization (Izutsu et al., 1993).pharmaceuticals. Very often, the most stable pHfor proteins in solution does not offer the best 5.4.3. Selection of a bulking agent(s)stability in solid state. This is because the inacti- A crystallizing bulking agent(s) is usuallyvation mechanisms of proteins may well be differ- needed in a solid protein formulation to have oneent in the two different states, as reported for or more of the following functions: to providebovine pancreatic RNase A (Townsend and mechanical support of the ﬁnal cake, to improveDeLuca, 1990). In such cases, a balanced pH must product elegance, to improve formulation dissolu-be used. tion, and to prevent product collapse and blow- out. A bulking agent(s) should have enough5.4.2. Selection of a buffering agent(s) solubility, compatibility with the protein, no or Many buffering agents covering a wide pH minimal toxicity, and high eutectic temperature,range are available for selection in formulating allowing efﬁcient freeze-drying.solid proteins. These agents include acetate, cit- Different bulking agents may affect stability ofrate, glycine, histidine, phosphate, Tris, etc. A solid proteins to different degrees. Therefore,buffering agent, that also stabilizes a protein, is careful selection of a suitable bulking agent maypreferable, such as histidine for freeze-dried FVIII be necessary. For example, both the degradation 8SQ (Osterberg et al., 1997). For pH-sensitive and aggregation rates of lyophilized IL-1ra (pHproteins, sodium phosphate should be avoided 6.5) during storage at 8, 30 or 50°C have beenbecause the selective crystallization of Na2HPO4 shown to be different in three bulking agents:
46 W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60mannitol, glycine or alanine (Chang et al., 1996c). cake appearance and protein stability, but thatAmong these three agents, glycine was apparently containing 1.7% (0.23 M) glycine and 2% sucrosebest in protecting protein stability. only offered reasonable protein stability with Two bulking agents are frequently used: glycine crumbly cake appearance (Bush et al., 1998).and mannitol. Glycine has several advantages, Therefore, proper selection of a suitable bulkingincluding non-toxicity, high solubility, and high agent(s) and its relative amount is critical.eutectic temperature (Akers et al., 1995). Al-though both free or salt form of glycine can be 5.4.4. Selection of a stabilizer(s)used, it has been found that neutral glycine crys- A solid protein product should be stable duringtallizes rapidly, whereas glycine hydrochloride storage at least above 0°C, and preferably undercrystallizes slowly, even at the same pH (Akers et room conditions. To achieve this goal, a proteinal., 1995). As a bulking agent, mannitol may stabilizer(s) is usually needed to protect a proteinstabilize certain proteins. It has been used both as during lyophilization and/or long-term storage.a bulking agent and as a stabilizer in a lyophilized Based on the stabilization mechanisms discussedformulation of transforming growth factor-b1 in Section 5.3, a stabilizer(s) for a solid protein(TGF-b1) (Gombotz et al., 1996). product should be at least partially amorphous Most amino acids are potential bulking agents and able to replace water, forming intimate hy-as they easily crystallize out (Mattern et al., 1999). drogen bonds with the protein. Formation of anHowever, formation of acid salts reduces their amorphous glassy state is considered to be atendency to crystallize (Mattern et al., 1999). An- prerequisite, not a guarantee, for protein stabilityother crystalline agent, NaCl, is not preferable as (Pikal et al., 1991; Skrabanja et al., 1994). Since aa bulking agent due to its low eutectic and glass disordered amorphous material has a lower en-transition temperature (Carpenter et al., 1997). It ergy barrier for dissolution than a structured crys-was, however, used as a bulking agent in a talline solid, use of an amorphous stabilizer(s) canlyophilized factor VIII SQ formulation because it also lead to faster dissolution of a solid protein 8solubilized the protein (Osterberg et al., 1997). product (Miller et al., 1997). An amorphous excipient(s) in a protein formu- The widely used stabilizers in solid proteinlation may inhibit crystallization of the bulking products are sugars. Most sugars do not crystal-agent(s), thus affecting protein stability. Increas- lize under normal operating conditions (Changing the relative amount of sucrose in a mixture of and Randall, 1992). Reducing sugars are generallysucrose and glycine gradually inhibited crystalliza- not preferable due to the possibility of reactingtion of glycine, and at a weight ratio of 1.4:1 with dried proteins via the Maillard reaction.(glycine:sucrose), crystallization of glycine was However, reducing sugars can be used if they arenot detectable (Lueckel et al., 1998a). In a differ- better stabilizers. For example, dextrose, a reduc-ent study, crystallization of glycine was not ob- ing sugar, has been shown to inhibit moisture-in-servable under microscope when its concentration duced aggregation of bovine insulin signiﬁcantlywas below 29% in the same mixture during freeze- at 1:1 protein:excipient ratio, while trehalose diddrying; partial crystallization was observed at not stabilize the protein to a signiﬁcant level43%, but a good lyophilized cake was obtained (Katakam and Banga, 1995).only at a glycine concentration of about 50% or Among sugars, sucrose seems to be the mosthigher (Kasraian et al., 1998). Similarly, mannitol commonly selected. Recently, trehalose started todid not crystallize in the presence of a second gain attention due to several advantages, as dis-non-crystallizing component, such as sucrose, lac- cussed in Section 3.1. Their relative effect in stabi-tose, maltose, or trehalose until the mannitol con- lizing solid proteins for long-term storage iscentration was over about 30% (w/w) (Kim et al., protein-dependent (also see Section 5.3). Here are1998). Because of this effect, lyophilized rFIX some more examples. Sucrose was shown to beformulation containing 2% (0.26 M) glycine and more effective than trehalose in inhibiting chemi-1% sucrose (as a stabilizer) had both excellent cal degradation of lyophilized IL-1ra during stor-
W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60 47age at 8, 30 and 50°C (Chang et al., 1996c), while stabilization. While polymers increase the Tg of atrehalose at 30 mg ml − 1 in lyophilized IL-6 for- protein formulation and inhibit crystallization ofmulation inhibited IL-6 aggregation more effec- other excipients, sugars can form intimate hydro-tively than sucrose during storage at both 25 and gen bonds with proteins. This strategy, however,40°C for 9 months (Lueckel et al., 1998b). Two has not been very effective at least for somepossible reasons have been offered why trehalose proteins. Inclusion of dextran 40 in a lyophilizedwas more effective, (1) trehalose formulation had IL-6 formulation containing sucrose signiﬁcantlya higher glass transition temperature; and (2) increased protein aggregation during storage atsucrose could be hydrolyzed to form two reducing 40°C for 9 months (Lueckel et al., 1998b).sugars for the Maillard reaction. However, the Proteins, like other polymers, have high Tgs.protein molecular mobility in a trehalose formula- Therefore, increasing protein:excipient ratio in-tion was shown to be higher than that in a sucrose creases Tg of protein formulations. Increasing theformulation at certain temperatures, which might content of a lyophilized monoclonal antibodymake trehalose a less effective stabilizer than su- from 2.5 to 25 mg ml − 1 in sucrose (or trehalose atcrose depending on the protein, degradation 31.3 mg ml − 1) formulation increased the Tg of themechanisms, and storage conditions (Miller et al., formulation from 59 (or 80) to 89 (or 100)°C by1997). DSC (Duddu and Dal Monte, 1997). Increasing Certain salts can be used to stabilize solid the relative content of rbSt to 50% in a sucrose/proteins (also see Section 3.1). However, the pres- rbSt mixture gradually increased the Tg fromence of uncrystallized salts in a freeze-concentrate 74°C for pure sucrose to 96°C for the mixtureusually depresses T % , so the salt content in protein g (Sarciaux and Hageman, 1997). Therefore, theformulations should be kept to a minimum relative protein concentration in a formulation(Franks, 1990). Several factors may affect the should be kept relatively high to prevent collapse.extent of salt crystallization, including the nature More importantly, a high protein:excipient ratioof salt, salt concentration, and cooling rate. strongly inhibits excipient crystallization (also seeChang and Randall (1992) have classiﬁed salts Section 5.3). For example, the lyophilized rhD-into three types based on their glass-forming ten- Nase–mannitol formulations were partially crys-dency at a given cooling rate and subsequent talline when the relative protein quantity was 17%thermal history, (1) crystallizing salts such as (w/w) or less, but became amorphous at 80% ormaleic acid, Na2HPO4, Na2SO4, Na2CO3, KCl, more (Chan et al., 1999). At a higherand (NH4)2SO4; (2) partially crystallizing (doubly protein:sucrose ratio, lyophilized Humicola lanug-unstable glass) salts such as NaCl, NaHCO3, inosa lipase was shown to be more stable duringK2HPO4, KH2PO4, CaCl2, MgCl2, glycine, and storage at 60°C, due to effective inhibition ofb-alanine; and (3) glass-forming salts such as sucrose crystallization by the protein (KreilgaardNaH2PO4, sodium/potassium citrate, citric acid, et al., 1999). In a more detailed study, the Tcry ofhistidine, and sodium/potassium acetate. Since sucrose at 4.4% water content was 70°C andglass-forming excipients can inhibit salt crystal- increased to 120°C when 20% rbSt was includedlization, salts can be potential protein stabilizers (Sarciaux and Hageman, 1997). On the otherin the presence of other amorphous excipients hand, increasing the protein concentration too(Hatley and Franks, 1991). high may eventually destabilize a protein due to Some polymers can be chosen as protein stabi- insufﬁcient quantity of a stabilizer, as has beenlizers in solid state, as they can increase Tg of reported for rhIL-1ra (Chang et al., 1996c).protein formulations. On the other hand, they The presence of a bulking agent may affectmay not be as effective as sugars due to inefﬁcient properties of an amorphous excipient. The T % of a ghydrogen bonding with proteins (Schebor et al., mixture of sucrose and glycine (or lysine–HCl)1996; Cardona et al., 1997; Kreilgaard et al., decreased with increasing amino acid:sucrose ra-1998a). An alternative is to use both polymers tio (Lueckel et al., 1998a). Addition of glycine (upand sugars together to achieve enhanced protein to 71%, w/w) in a sucrose solution decreased
48 W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60linearly the T % of the mixture from − 32 to g ment of Tg of a lyophilized formulation in the− 52°C (Kasraian et al., 1998). When glycine presence of different excipients or determinationconcentration was over 86%, no T % was observed. g of IR spectrum of a formulation in comparison toSimilarly, increasing mannitol concentrations up that of a reference (see Section 2.3). Both methodsto 30% decreased T % of sugars, such as sucrose, g should be used with caution, as these parameterslactose, maltose, and trehalose (Kim et al., 1998). may not always reﬂect changes in protein activityTo keep T % at a level high enough for efﬁcient g or stability.lyophilization, the relative quantity of a bulking Recent investigations suggest that combined useagent should be properly chosen and optimized. of sucrose as a protein stabilizer and glycine as a bulking agent may be a good starting point when5.4.5. Selection of other excipients formulating a solid protein product. The combi- Other formulation excipients, such as antimi- nation of these two excipients has been usedcrobial agents or solubilizers may be used depend- successfully in formulating several proteins. Theing on the protein. Both Tween 80 and SDS recently developed albumin-free formulation for(more effective) facilitate solubilization of rIFN- recombinant factor VIII (rFVIII) contains su-b-1b (Betaseron®) (Lin et al., 1996). SDS has also crose, glycine, histidine, CaCl2, and NaCl (Nayar,been used to solubilize IL-2 in Proleukin® (Physi- 1998). Lyophilized rFIX formulation contains su-cians’ Desk Reference, 1999). crose, glycine, histidine, and polysorbate 80 (Bush et al., 1998). Chang et al. (1996c) developed a5.4.6. O6erall consideration of formulation stable lyophilized IL-1ra formulation, which con-excipients tains sucrose, glycine, and sodium citrate. This Generally, the total quantity of solid in protein stable IL-1ra formulation was also able to stand aformulations is between 2 and 10%. While a solid single-step freeze-drying cycle (Chang and Fis-content lower than 2% may not form a strong cher, 1995).cake, higher amount of solid ( 10%) may bedifﬁcult to process (Hatley et al., 1996; Carpenter 5.4.7. Selection of containers and stopperset al., 1997; Jennings, 1999; Willemer, 1999). In In close relation to selection of formulationaddition, higher solid content may affect reconsti- excipients, a compatible container should be used.tution of lyophilized formulations. Breen et al. Type I borosilicate glass (treated or untreated) is(1998) demonstrated that lyophilized cakes pre- usually the material of choice for containers duepared from 110 mg ml − 1 bulk took approxi- to its strong chemical resistance and low level ofmately 60 min to reconstitute as compared with leachables. Nevertheless, proteins can be absorbedless than 5 min for those from dilute bulks. SEM to glass surfaces to different extents (Gombotz etanalysis showed that cakes lyophilized from al., 1996). Since the loss of protein activity fromhighly concentrated bulks had smaller pores with glass surface absorption and surface-induced de-thicker walls than those from more dilute bulks. naturation is protein-dependent, containers need The relative quantity of different excipients is to be evaluated on a case-by-case basis (Burke etalso important because their physical properties al., 1992).are mutually affected. Excipients that provide The volume of a solution for bolus intravenousmore than one function in protein formulations injection generally does not exceed 10 ml. There-should be selected ﬁrst, such as sugars, which may fore, the size of containers for these applicationsbe used as both cryoprotectants and lyoprotec- is usually smaller than 20 ml, assuming the vol-tants. High levels of buffers or salts should be ume of the solution to be lyophilized is ideallyavoided because this may lead to potential pH between about 20 and 50% of the container vol-changes during freezing and likely depression in ume. A relatively small ﬁlling volume is preferableT % and Tg of dried formulations (Pikal, 1990b). g for an efﬁcient freeze-drying cycle and a higher To expedite selection of formulation excipients, protein concentration often results in a lower losstwo screening methods have been used: measure- of activity due to surface absorption and stability
W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60 49during freeze-drying. The lyophilized product can choice for solid protein formulations because theybe reconstituted with different volumes of water have relatively low moisture absorption and vaporor solution based on the protein solubility and transmission rates (DeGrazio and Flynn, 1992).stability. Among halobutyl stoppers, bromobutyl stoppers Container stoppers may signiﬁcantly affect have been shown to be more resistant thanlong-term stability of solid protein pharmaceuti- chlorobutyl stoppers to moisture absorption dur-cals. This effect is often due to a stopper-induced ing storage and steam sterilization (Corveleyn etincrease in moisture content of lyophilized formu- al., 1997). To minimize the moisturizing effect,lations during storage. The percent change in stoppers can be autoclaved at 121°C for 30 minmoisture content depends on both hygroscopicity and immediately dried in an oven at 135°C for 5and quantity of the protein formulation in the h before use. On the other hand, overdried stop-container. Also, oxygen permeation through stop- pers may absorb moisture from the lyophilizedpers or stopper leakage may increase the partial formulation and dropped the moisture content ofpressure of oxygen in a product vial and cause the formulation during storage (Corveleyn et al.,potential oxidation of proteins during storage 1997). The integrity of stoppers can be tested(Wang et al., 1997). under high humidity by monitoring the change in Three stopper-related processes can lead to an moisture content of the protein formulation.increase in residual moisture of protein formula- Therefore, stoppers should be carefully chosentions, (1) transfer of moisture from stoppers to the based on their compatibility with the protein for-formulation; (2) diffusion or transmission of mulation, resistance to formulation pH, excipientsmoisture through the stopper; and (3) microleaks and sterilization, moisture/vapor transfer prop-in the stopper-vial seal (House and Mariner, erty, and resealability.1996). The ﬁrst process is usually dominant (Fordand Dawson, 1994). Stoppers absorb moisture 5.4.8. Stability testing of ﬁnal formulationsduring routine steam sterilization and sterilized Selection of the ﬁnal protein formulation canstoppers can transfer some of its moisture back to only be based on stability studies. A variety ofthe protein product (DeGrazio and Flynn, 1992; formulation parameters in stability studies can beCorveleyn et al., 1997). In fact, the steam auto- used to compare different formulations, includingclaving process is responsible for the majority of protein activity, cake physical attributes (shape,water driven into elastomeric stoppers. Chang et color and texture), particulate formation, mois-al. (1996c) demonstrated that two out of three lots ture content, and reconstitution (or rehydratabil-of siliconized stoppers showed transfer of mois- ity). Another parameter is the volume of proteinture to the lyophilized rhIL-1ra formulation, re- formulation preparation, which should notsulting in a signiﬁcant increase in moisture change much after lyophilization and during stor-content during stability studies. Storage of a age. Although minor shrinkage in volume may belyophilized IL-6 formulation at 25 or 40°C for 9 seen, especially at low solute concentrationsmonths led to an increase in moisture content (Daukas and Trappler, 1998), a signiﬁcant drop infrom 0.5 to 2% (Lueckel et al., 1998b). The in- volume may be an indication of formulation melt-crease in moisture content can cause a signiﬁcant back or collapse.drop in Tg. It has been shown that moisture To expedite selection of the ﬁnal protein formu-transfer from the stopper dropped the Tg of a lation, accelerated stability studies are frequentlysucrose-formulated rFXIII formulation by 11°C conducted. These stressed stability conditions in-after storage at 40°C for 3 months (Kreilgaard et clude high temperature, high humidity, or inten-al., 1998a) and that of a vacuum-dried LDH sive lighting. One key issue in conductingformulation by 35°C after storage at 60°C for 26 accelerated stability studies is whether and howweeks (Mattern et al., 1999). well the data obtained at high temperatures can There are many types of stopper materials be extrapolated to those under real-time condi-available. Butyl or halobutyl rubber is often the tions. Often, protein stability results obtained at
50 W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60high temperatures do not reﬂect or predict what aqueous solvents, polymers, protein itself, surfac-happens under real-time conditions, due to the tants, and amino acids. These cryoprotectants cancomplexity of multiple protein degradation path- also be used as potential lyoprotectants exceptways at different temperatures. For example, Sun non-aqueous solvents. These stabilizers may pro-et al. (1998) found that the temperature depen- tect proteins by one or more of the followingdence of G6PDH inactivation in carbohydrate mechanisms: preferential interaction, replacement(glucose or sucrose) glassy matrix deviated from of water, formation of a glass, hydrogen bonding,Arrhenius relationship between 8 and 93°C. On and steric hindrance.the other hand, if the multiple degradation pro- The design of a lyophilization cycle may havecesses in proteins can be described separately, or signiﬁcant impact on protein stability during andthe rate-limiting degradation step does not change after lyophilization. Lyphilization cycle-relatedwithin a certain temperature range, prediction of stresses include freezing rate and temperature,protein stability based on accelerated stability thermal treatment condition, drying rate and tem-studies is very optimistic. The loss of lactase activ- perature, and the ﬁnal moisture content. There-ity in a lyophilized PVP formulation followed fore, efforts should be made to design a cycle,Arrhenius behavior well in a temperature range which is robust and efﬁcient, and has minimalbetween 37 and 70°C (Mazzobre et al., 1997). adverse effects on protein stability. To design suchYoshioka et al. (1994) examined the stability of a cycle, protein formulations need thorough char-six protein preparations, including a-chy- acterization and all critical temperatures need tomotrypsin troche, a-chymotrypsin tablet, brome- be determined as discussed in Section 4.1.lain tablet A and B, kallikrein capsule, and Solid protein pharmaceuticals may experience ab-galactosidase powder at elevated temperatures variety of instabilities. Major instabilities includebetween 40 and 70°C (50 or 75% RH). They aggregation, deamidation, oxidation, the Maillardfound that all of them had complex kinetics, but reaction, hydrolysis, and disulﬁde bond forma-the inactivation rate exhibited approximately lin- tion/exchange. Many factors affect these instabili-ear Arrhenius relationship. Nevertheless, for ties, including storage temperature, glassproteins with unknown degradation pathways, transition temperature of the formulation, resid-real-time stability testing has to be conducted for ual moisture content, formulation ‘pH’, crystal-selection of the ﬁnal protein formulation. lization of amorphous excipients, and presence of destabilizing excipients or contaminants. These instabilities may be minimized by proper selection6. Summary of formulation pH, residual moisture content, and more importantly, formulation stabilizers, such as Lyophilization (freeze-drying) is the most com- sugars/polyols, polymers, salts, and surfactants.mon process for making solid protein pharmaceu- Formulation of solid protein pharmaceuticalsticals. However, this process can generate a may require not only suitable stabilizers but alsovariety of stresses, and denature proteins to vari- other excipients such as bulking, buffering, andous degrees. These stresses include low tempera- antimicrobial agents. Since the physical propertiesture, formation of dendritic ice crystals, increase of these agents are mutually affected, the relativein ionic strength, pH changes, phase separation, quantity of these agents in a protein formulationand removal of the protein hydration shell. Struc- is critical and should be determined based ontural changes in proteins during lyophilization can sound experimentation and appropriate stabilitybe either reversible or irreversible depending on studies. A proposed protein formulation for initialthe protein, and can be conveniently monitored development trial is 2% glycine, 1% sucrose, andby IR. Proteins sensitive to freezing and/or drying 20 mM buffering agent controlling the formula-stresses can be stabilized by selection of proper tion pH. To expedite selection of formulationcryo- and/or lyoprotectants. The commonly used excipients, two screening methods may be usedcryoprotectants include sugars/polyols, non- with caution: comparison of Tgs or IR spectra of
W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60 51formulations in the presence of different excipients. IR infrared spectroscopyHowever, selection of the ﬁnal protein formulation LDH lactate dehydrogenaserequires real-time stability studies. LMW-UK low molecular weight urokinase In summary, development of a lyophilized Mb myoglobinprotein product usually takes an enormous amount MDH maleate dehydrogenaseof time, labor, and effort, simply because there is Met-hGH methionyl human growth hormoneno single, short, and mature pathway to follow in MS mass spectroscopyformulating such a product, and many experiments NMR nuclear magnetic resonanceare done on a trial-and-error basis. This trend will spectroscopycontinue until a breakthrough is achieved in under- PEG polyethylene glycolstanding the basic behavior of proteins and their PFK phosphofructokinasestabilization. PVA polyvinylalcohol PVP polyvinylpyrrolidone rbSt recombinant bovine somatotropin7. Abbreviations rconIFN recombinant consensus a- interferonaFGF acidic ﬁbroblast growth factor rFIX recombinant factor IXbFGF basic ﬁbroblast growth factor rFXIII recombinant factor XIIIBGG bovine k-globulin rHA recombinant human albuminBO bilirubin oxidase rhCNTF recombinant human ciliary neuro-BPTI bovine pancreatic trypsin inhibitor trophic factorBSA bovine serum albumin rhIL-1ra recombinant human interleukin-1CHAPS 3-[(3-cholamidepropyl) receptor antagonist -dimethylammonio] rhKGF recombinant human keratinocyte -1-propanesulfate growth factorCMC carboxymethyl cellulose rhMCSF recombinant human macrophageCyt c cytochrome c colony-stimulating factorDMF dimethylformamide RNase A ribonuclease ADMSO dimethylsulphoxide SDS sodium dodecyl sulfateDSC differential scanning calorimetry TGF transforming growth factorG-CSF granulocyte colony-stimulating TNFbp tumor necrosis factor binding factor proteinGDH glutamate dehydrogenase tPA tissue plasminogen activatorG6PDH glucose-6-phosphate TT tetanus toxoid dehydrogenaseGM-CSF granulocyte macrophage colony stimulating factorHEC hydroxyethyl cellulosehIGF-I human insulin-like growth factor I AcknowledgementsHLL Humicola lanuginosa lipaseHP-b-CD hydroxypropyl-b-cyclodextrin This review would not be possible without theHPMC hydroxypropyl methylcellulose support and encouragement of Drs Rajiv Nayar,HSA human serum albumin Sheryl Martin-Moe, and Bob Kuhn. I am alsoIFN-b interferon-b indebted to Drs Rajiv Nayar, Jeff Hey, XingHangIFN-g interferon-g Ma, and especially, the two referees for their criticalIL-1a interleukin-1a review of the manuscript and their valuable com-IL-1b interleukin-1b ments. The careful editing of the manuscript byIL-2 interleukin-2 Blanca Martinez is also acknowledged
52 W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60References Bell, L.N., Hageman, M.J., 1996. Glass transition explanation for the effect of polyhydroxy compounds on protein denat-Adams, G.D.J., Ramsay, J.R., 1996. Optimizing the lyophiliza- uration in dehydrated solids. J. Food Sci. 61, 372 – 378. tion cycle and the consequences of collapse on the pharma- Belton, P.S., Gil, A.M., 1994. IR and Raman spectroscopic ceutical acceptability of Erwinia L-asparaginase. J. Pharm. studies of the interaction of trehalose with hen egg white Sci. 85, 1301 – 1305. lysozyme. Biopolymers 34, 957 – 961.Adler, M., Lee, G., 1999. Stability and surface activity of lactate Bindschaedler, C., 1999. Lyophilization process validation. In: dehydrogenase in spray-dried trehalose. J. Pharm. Sci. 88, Rey, L., May, J.C. (Eds.), Freeze-Drying/Lyophilization of 199– 208. Pharmaceutical and Biological Products, vol. 96. MarcelAkers, M.J., Milton, N., Byrn, S.R., Nail, S.L., 1995. Glycine Dekker, New York, pp. 373 – 408. crystallization during freezing: the effect of salt form, pH, Brange, J., Havelund, S., Hougaard, P., 1992a. Chemical and ionic strength. Pharm. Res. 12, 1457–1461. stability of insulin. 2. Formation of higher molecular weightAlden, M., Magnusson, A., 1997. Effect of temperature history ´ transformation products during storage of pharmaceutical on the freeze-thawing process and activity of LDH formu- preparations. Pharm. Res. 9, 727 – 734. lations. Pharm. Res. 14, 426–430. Brange, J., Langkjaer, L., Havelund, S., Vølund, A., 1992b.Aldous, B.J., Auffret, A.D., Franks, F., 1995. The crystalliza- Chemical stability of insulin. 1. Hydrolytic degradation tion of hydrate from amorphous carbohydrates. Cryo-Let- ters 16, 181– 186. during storage of pharmaceutical preparations. Pharm. Res.Allison, S.D., Dong, A., Carpenter, J.F., 1996. Counteracting 9, 715 – 726. effects of thiocyanate and sucrose on chymotrypsinogen Breen, E.D., Costantino, H.R., Hsu, C.C., Shire, S.J., 1998. secondary structure and aggregation during freezing, drying Affect of bulk concentration on reconstitution of lyophilized and rehydration. Biophys. J. 71, 2022–2032. protein formulations. Pharm. Sci. 1 (Suppl.), S540.Allison, S.D., Randolph, T.W., Manning, M.C., Middleton, K., Buera, M.D.P., Chirife, J., Resnik, S.L., Wetzler, G., 1987. Davis, A., Carpenter, J.F., 1998. Effects of drying methods Nonenzymic browning in liquid model systems of high water and additives on structure and function of actin: mechanisms activity: kinetics of color changes due to Maillard’s reaction of dehydration-induced damage and its inhibition. Arch. between different single sugars and glycine and comparison Biochem. Biophy. 358, 171–181. with caramelization browning. J. Food Sci. 52, 1063 –Allison, S.D., Chang, B., Randolph, T.W., Carpenter, J.F., 1067. 1999. Hydrogen bonding between sugar and protein is Buera, M.P., Levi, G., Karel, M., 1992. Glass transition in responsible for inhibition of dehydration-induced protein polyvinylpyrrolidone: effect of molecular weight and dilu- unfolding. Arch. Biochem. Biophys. 365, 289–298. ents. Biotechnol. Prog. 8, 144 – 148.Anchordoquy, T.J., Carpenter, J.F., 1996. Polymers protect Bunk, D.M., 1997. Characterization of the glycation of albumin lactate dehydrogenase during freeze-drying by inhibiting in freeze-dried and frozen human serum. Anal. Chem. 69, dissociation in the frozen state. Arch. Biochem. Biophys. 332, 231 – 238. 2457 – 2463.Andersson, M.M., Hatti-Kaul, R., 1999. Protein stabilizing Burke, C.J., Steadman, B.L., Volkin, D.B., Tsai, P.-K., Bruner, effect of polyethyleneimine. J. Biotechnol. 72, 21–31. M.W., Middaugh, C.R., 1992. The adsorption of proteinsAndya, J.D., Maa, Y.-F., Costantino, H.R., Nguyen, P.-A., to pharmaceutical container surfaces. Int. J. Pharm. 86, Dasovich, N., Sweeney, T.D., Hsu, C.C., Shire, S.J., 1999. 89 – 93. The effect of formulation excipients on protein stability and Bush, L., Webb, C., Bartlett, L., Burnett, B., 1998. The aerosol performance of spray-dried powders of a recombi- formulation of recombinant factor IX: stability, robustness, nant humanized anti-IgE monoclonal antibody. Pharm. Res. and convenience. Semin. Hematol. 35, 18 – 21. 16, 350 – 358. Cardona, S., Schebor, C., Buera, M.P., Karel, M., Chirife, J.,Angell, C.A., 1995. Formation of glasses from liquid and 1997. Thermal stability of invertase in reduced-moisture biopolymers. Science 267, 1924–1935. amorphous matrices in relation to glassy state and trehaloseArakawa, T., Yoshiko, K., Carpenter, J.F., 1991. Protein-sol- crystallization. J. Food Sci. 62, 105 – 112. vent interactions in pharmaceutical formulations. Pharm. Careri, G., Giansanti, A., Gratton, E., 1979. Lysozyme ﬁlm Res. 8, 285– 291. hydration events: an IR and gravimetric study. BiopolymersArakawa, T., Prestrelski, S.J., Kenney, W.C., Carpenter, J.F., 18, 1187 – 1203. 1993. Factors affecting short-term and long-term stabilities Carpenter, J.F., Crowe, J.H., 1989. An infrared spectroscopic of proteins. Adv. Drug Deliv. Rev. 10, 1–28.Bam, N.B., Randolph, T.W., Cleland, J.L., 1995. Stability of study of the interactions of carbohydrates with dried protein formulations: investigation of surfactant effects by proteins. Biochemistry 28, 3916 – 3922. a novel EPR spectroscopic technique. Pharm. Res. 12, 2–11. Carpenter, J.F., Hand, S.C., Crowe, L.M., Crowe, J.H., 1986.Bell, L.N., Hageman, M.J., Muraoka, L.M., 1995. Thermally Cryoprotection of phosphofructokinase with organic so- induced denaturation of lyophilized bovine somatotropin lutes: characterization of enhanced protection in the pres- and lysozyme as impacted by moisture and excipients. J. ence of divalent cations. Arch. Biochem. Biophys. 250, Pharm. Sci. 84, 707 – 712. 505 – 512.
W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60 53Carpenter, J.F., Crowe, L.M., Crowe, J.H., 1987. Stabilization ity of recombinant human deoxyribonuclease I in the of phosphofructokinase with sugars during freeze-drying: aqueous and lyophilized states. J. Pharm. Sci. 88, 477 – 482. characterization of enhanced protection in the presence of Chuyen, N.V., 1998. Maillard reaction and food processing. divalent cations. Biochim. Biophys. Acta 923, 109–115. Application aspects. Adv. Exp. Med. Biol. 434, 213 – 235.Carpenter, J.F., Crowe, J.H., Arakawa, T., 1990. Comparison Cleland, J.L., Powell, M.F., Shire, S.J., 1993. The development of solute-induced protein stabilization in aqueous solution of stable protein formulations: a close look at protein and in frozen and dried state. J. Dairy Sci. 73, 3627–3636. aggregation, deamidation, and oxidation. Crit. Rev. Ther.Carpenter, J.F., Arakawa, T., Crowe, J.H., 1991. Interactions Drug Carrier Syst. 10, 307 – 377. of stabilizing additives with proteins during freeze-thawing Colaco, C., Sen, S., Thangavelu, M., Pinder, S., Roser, B., and freeze-drying. Dev. Biol. Stand. 74, 225–239. 1992. Extraordinary stability of enzymes dried in trehalose:Carpenter, J.F., Prestrelski, S.J., Arakawa, T., 1993. Separa- simpliﬁed molecular biology. Bio. Technol. 10, 1007 –1011. tion of freezing and drying-induced denaturation of Cooper, E.A., Knutson, K., 1995. Fourier transform infrared lyophilized proteins using stress-speciﬁc stabilization. I. spectroscopy investigations of protein structures. In: Her- Enzymatic activity and calorimetric studies. Arch. ron, J.N., Jiskoot, W., Crommelin, D.J.A. (Eds.), Physical Biochem. Biophys. 303, 456–464. Methods to Characterize Pharmaceutical Proteins. PlenumCarpenter, J.F., Pikal, M.J., Chang, B.S., Randolph, T.W., Press, New York, pp. 101 – 143. 1997. Rational design of stable lyophilized protein formu- Corveleyn, S., De Smedt, S., Remon, J.P., 1997. Moisture lations: some practical advice. Pharm. Res. 14, 969–975. absorption and desorption of different rubber lyophiliza-Carpenter, J.F., Prestrelski, S.J., Dong, A., 1998. Application tion closures. Int. J. Pharm. 159, 57 – 65. of infrared spectroscopy to development of stable Costantino, H.R., Langer, R., Klibanov, A.M., 1994a. Solid- lyophilized protein formulations. Eur. J. Pharm. Bio- phase aggregation of proteins under pharmaceutically rele- pharm. 45, 231 – 238. vant conditions. J. Pharm. Sci. 83, 1662 – 1669.Carpenter, J.F., Izutsu, K., Randolph, T.W., 1999. Freezing- Costantino, H.R., Langer, R., Klibanov, A.M., 1994b. Mois- and drying-induced perturbations of protein structure and ture-induced aggregation of lyophilized insulin. Pharm. mechanisms of protein protection by stabilizing additives. Res. 11, 21 – 29. In: Rey, L., May, J.C. (Eds.), Freeze-Drying/Lyophiliza- Costantino, H.R., Griebenow, K., Mishra, P., Langer, R., tion of Pharmaceutical and Biological Products, vol. 96. Klibanov, A.M., 1995a. Fourier-transform infrared spec- Marcel Dekker, New York, pp. 123–160. troscopic investigation of protein stability in theChan, H.K., Gonda, I., 1998. Solid state characterization of lyophilized form. Biochim. Biophy. Acta 1253, 69 – 74. spray-dried powders of recombinant human deoxyribonu- Costantino, H.R., Langer, R., Klibanov, A.M., 1995b. Aggre- clease (rhDNase). J. Pharm. Sci. 87, 647–654. gation of a lyophilized pharmaceutical protein, recombi-Chan, H.K., Au-Yeung, K.L., Gonda, I., 1999. Development nant human albumin: effect of moisture and stabilization of a mathematical model for the water distribution in by excipients. Biotechnology 13, 493 – 496. freeze-dried solids. Pharm. Res. 16, 660–665. Costantino, H.R., Schwendeman, S.P., Griebenow, K.,Chang, B.S., Randall, C.S., 1992. Use of subambient thermal Klibanov, A.M., Langer, R., 1996. The secondary struc- analysis to optimize protein lyophilization. Cryobiology ture and aggregation of lyophilized tetanus toxoid. J. 29, 632 – 656. Pharm. Sci. 85, 1290 – 1293.Chang, B.S., Randall, C.S., Lee, Y.S., 1993. Stabilization of Costantino, H.R., Andya, J.D., Nguyen, P.-A., Dasovich, N., lyophilized porcine pancreatic elastase. Pharm. Res. 10, Sweeney, T.D., Shire, S.J., Hsu, C.C., Maa, Y.-F., 1998a. 1478 – 1483. Effect of mannitol crystallization on the stability and aero-Chang, B.S., Fischer, N.L., 1995. Development of an efﬁcient sol performance of a spray-dried pharmaceutical protein, single step freeze drying cycle for protein formulations. recombinant humanized anti-IgE monoclonal antibody. J. Pharm. Res. 12, 831 – 837. Pharm. Sci. 87, 1406 – 1411.Chang, B.S., Beauvais, R.M., Dong, A., Carpenter, J.F., Costantino, H.R., Carrasquillo, K.G., Cordero, R.A., Mu- 1996a. Physical factors affecting the storage stability of menthaler, M., Hsu, C.C., Griebenow, K., 1998b. Effect of freeze-dried interleukin-1 receptor antagonist: glass transi- excipients on the stability and structure of lyophilized tion and protein conformation. Arch. Biochem. Biophys. recombinant human growth hormone. J. Pharm. Sci. 87, 331, 249 – 258. 1412 – 1420.Chang, B.S., Kendrick, B.S., Carpenter, J.F., 1996b. Surface- Costantino, H.R., Curley, J.G., Wu, S., Hsu, C.C., 1998c. induced denaturation of proteins during freezing and its Water sorption behavior of lyophilized protein – sugar sys- inhibition by surfactants. J. Pharm. Sci. 85, 1325–1330. tems and implications for solid-state interactions. Int. J.Chang, B.S., Reeder, G., Carpenter, J.F., 1996c. Development Pharm. 166, 211 – 221. of a stable freeze-dried formulation of recombinant human Costantino, H.R., Schwendeman, S.P., Langer, R., Klibanov, interleukin-1 receptor antagonist. Pharm. Res. 13, 243– A.M., 1998d. Deterioration of lyophilized pharmaceutical 249. proteins. Biochemistry (Moscow) 63, 357 – 363.Chen, B., Costantino, H.R., Liu, J., Hsu, C.C., Shire, S.J., Crowe, J.H., Hoekstra, F.A., Crowe, L.M., 1992. Anhydro- 1999. Inﬂuence of calcium ions on the structure and stabil- biosis. Annu. Rev. Physiol. 54, 579 – 599.
54 W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60Crowe, J.H., Crowe, L.M., Carpenter, J.F., 1993a. Preserving Fox, K.C., 1995. Putting proteins under glass. Science 267, dry biomaterials: the water replacement hypothesis. Bio- 1922 – 1923. pharm. 6 (1), 28 – 37. Franks, F., 1990. Freeze-drying: from empiricism to pre-Crowe, J.H., Crowe, L.M., Carpenter, J.F., 1993b. Preserving dictability. Cryo-Letters 11, 93 – 110. dry biomaterials: the water replacement hypothesis. Bio- Franks, F., 1993. Solid aqueous solutions. Pure Appl. Chem. pharm. 6 (2), 40 – 43. 65, 2527 – 2537.Crowe, L.M., Reid, D.S., Crowe, J.H., 1996. Is trehalose Franks, F., 1994. Long-term stabilization of biologicals. Bio- special for preserving dry biomaterials? Biophy. J. 71, technology 12, 253 – 256. 2087 – 2093. Fransson, J., Florin-Robertsson, E., Axelsson, K., Nyhlen, C., ´Crowe, J.H., Carpenter, J.F., Crowe, L.M., 1998. The role of 1996. Oxidation of human insulin-like growth factor I in vitriﬁcation in anhydrobiosis. Annu. Rev. Physiol. 60, 73– formulation studies: kinetics of methionine oxidation in 103. aqueous solution and in solid state. Pharm. Res. 13, 1252 –Darcy, P., Buckton, G., 1997. The inﬂuence of heating/drying 1257. on the crystallization of amorphous lactose after structural French, D.L., Arakawa, T., Niven, R.W., Li, T., 1995. FTIR collapse. Int. J. Pharm. 158, 157–164. investigation of hydration-induced conformational changesDaukas, L.A., Trappler, E.H., 1998. Assessing the quality of in spray-dried protein/trehalose powders. Pharm. Res. 12 lyophilized parenterals. Pharm. Cosmetic Quality 2, 21–25. (Suppl.), S83.Dawson, P.J., 1992. Effect of formulation and freeze-drying on Getlin, L.A., 1991. Kinetics of phase transitions in a frozen the long-term stability of rDNA-derived cytokines. Dev. solution. Dev. Biol. Stand. 74, 93 – 104. Biol. Stand. 74, 273 – 282. Gibson, T.D., 1996. Protein stabilization using additives basedDeGrazio, F., Flynn, K., 1992. Lyophilization closures for on multiple electrostatic interactions. Dev. Biol. Stand. 87, protein based drugs. J. Parenter. Sci. Technol. 46, 54–61. 207 – 217.De Young, L.R., Dill, K.A., Fink, A.L., 1993. Aggregation Gombotz, W.R., Pankey, S.C., Bouchard, L.S., Phan, D.H., and denaturation of apomyoglobin in aqueous urea solu- MacKenzie, A.P., 1996. Stability, characterization, formu- tions. Biochemistry 32, 3877–3886. lation and delivery system development for transformingDill, K.A., Alonso, D.O.V., Hutchinson, K., 1989. Thermal growth factor-beta1. In: Pearlman, R., Wang, Y.J. (Eds.), stabilities of globular proteins. Biochemistry 28, 5439– Formulation, Characterization, and Stability of Protein 5449. Drugs. Plenum Press, New York, pp. 219 – 245.Dill, K.A., 1990. Dominant forces in protein folding. Bio- Goormaghtigh, E., Cabiaux, V., Ruysschaert, J.-M., 1994. chemistry 29, 7133 – 7155. Determination of soluble and membrane protein structureDong, A., Prestrelski, S.J., Allison, S.D., Carpenter, J.F., by Fourier transform infrared spectroscopy II. Experimen- 1995. Infrared spectroscopic studies of lyophilization- and tal aspects, side chain structure, and H/D exchange. Sub- temperature-induced protein aggregation. J. Pharm. Sci. cell. Biochem. 23, 363 – 404. 84, 415 – 423. Goto, Y., Fink, A.L., 1989. Conformational states of b-lacta-Duddu, S.P., Dal Monte, P.R., 1997. Effect of glass transition mase: molten-globule states at acidic and alkaline pH with temperature on the stability of lyophilized formulations high salt. Biochemistry 28, 945 – 952. containing a chimeric therapeutic monoclonal antibody. Graziano, G., Catanzano, F., Riccio, A., Barone, G., 1997. A Pharm. Res. 14, 591 – 595. reassessment of the molecular origin of cold denaturation.Duddu, S.P., Zhang, G., Dal Monte, P.R., 1997. The relation- J. Biochem. 122, 395 – 401. ship between protein aggregation and molecular mobility Greiff, D., 1992. Development of cycles for lyophilization. below the glass transition temperature of lyophilized for- Dev. Biol. Stand. 74, 85 – 92. mulations containing a monoclonal antibody. Pharm. Res. Griebenow, K., Klibanov, A.M., 1995. Lyophilization-induced 14, 596 – 600. reversible changes in the secondary structure of proteins.Eckhardt, B.M., Oeswein, J.Q., Bewley, T.A., 1991. Effect of Proc. Natl. Acad. Sci. USA 92, 10969 – 10976. Freezing on aggregation of human growth hormone. Gu, L.C., Erds, E.A., Chiang, H.-S., Calderwood, T., Tsai, K., Pharm. Res. 8, 1360 – 1364. Visor, G.C., Duffy, J., Hsu, W.-C., Foster, L.C., 1991.Fan, J., Cooper, E.I., Angell, C.A., 1994. Glasses with strong Stability of interleukin 1b(IL-1b) in aqueous solution: ana- calorimetric b-glass transitions and the relation to the lytical methods, kinetics, products, and solution formula- protein glass transition problem. J. Phys. Chem. 98, 9345– tion implications. Pharm. Res. 8, 485 – 490. 9349. Hageman, M.J., 1988. The role of moisture in protein stability.Forbes, R.T., Davis, K.G., Hindle, M., Clarke, J.G., Maas, J., Drug Dev. Ind. Pharm. 14, 2047 – 2069. 1998. Water vapor sorption studies on the physical stabil- Hagen, S.J., Hofrichter, J., Eaton, W.A., 1995. Protein reac- ity of a series of spray-dried protein/sugar powders for tion kinetics in a room-temperature glass. Science 269, inhalation. J. Pharm. Sci. 87, 1316–1321. 959 – 962.Ford, A.W., Dawson, P.J., 1994. Effect of type of container; Hagen, S.J., Hofrichter, J., Eaton, W.A., 1996. Germinate storage temperature and humidity on the biological activity rebinding and conformational dynamics of myoglobin em- of freeze-dried alkaline phosphatase. Biologicals 22, 191– bedded in a glass at room temperature. J. Phys. Chem. 100, 197. 12008 – 12021.
W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60 55Hancock, B.C., Zograﬁ, G., 1994. The relationship between House, J.A., Mariner, J.C., 1996. Stabilization of rinderpest the glass transition temperature and the water content of vaccine by modiﬁcation of the lyophilization process. Dev. amorphous pharmaceutical solids. Pharm. Res. 11, 471– Biol. Stand. 87, 235 – 244. 477. Hsu, C.C., Ward, C.A., Pearlman, R., Nguyen, H.M., Yeung,Hancock, B.C., Zograﬁ, G., 1997. Characteristics and signiﬁ- D.A., Curley, J.G., 1991. Determining the optimum resid- cance of the amorphous state in pharmaceutical systems. J. ual moisture in lyophilized protein pharmaceuticals. Dev. Pharm. Sci. 86, 1 – 12. Biol. Stand. 74, 255 – 271.Hancock, B.C., Shamblin, S.L., Zograﬁ, G., 1995. Molecular Hsu, C.C., Nguyen, H.M., Yeung, D.A., Brookes, D.A., Koe, mobility of amorphous pharmaceutical solids below their G.S., Bewley, T.A., Pearlman, R., 1995. Surface denatura- glass transition temperature. Pharm. Res. 12, 799–806. tion at solid-void interface — a possible pathway by whichHancock, B.C., Christensen, K., Shamblin, S.L., 1998. Esti- opalescent particulates form during the storage of mating the critical molecular mobility temperature (TK) of lyophilized tissue-type plasminogen activator at high tem- amorphous pharmaceuticals. Pharm. Res. 15, 1649–1651. peratures. Pharm. Res. 12, 69 – 77.Hartmann, U., Bunner, B., Korber, C.H., Rau, G., 1991. ¨ Hsu, C.C., Walsh, A.J., Nguyen, H.M., Overcashier, D.E., Where should be cooling rate be determined in an extended Koning-Bastiaan, H., Bailey, R.C., Nail, S.L., 1996. De- freezing sample. Cryobiology 28, 115–130. sign and application of a low-temperature Peltier-coolingHatley, R.H., 1997. Glass fragility and the stability of pharma- microscope stage. J. Pharm. Sci. 85, 70 – 74. ceutical preparations-excipient selection. Pharm. Dev. Technol. 2, 257 – 264. Izutsu, K., Yoshioka, S., Terao, T., 1993. Decreased protein-Hatley, R.H.M., Franks, F., 1991. Applications of DSC in the stabilizing effects of cryoprotectants due to crystallization. development of improved freeze-drying processes for labile Pharm. Res. 10, 1232 – 1237. biologicals. J. Therm. Anal. 37, 1905–1914. Izutsu, K., Yoshioka, S., Kojima, S., 1994a. Physical stabilityHatley, R.H.M., Franks, F., Brown, S., Sandhu, G., Gray, and protein stability of freeze-dried cakes during storage at M., 1996. Stabilization of a pharmaceutical drug substance elevated temperatures. Pharm. Res. 11, 995 – 999. by freeze-drying: a case study. Drug Stability 1, 2–12. Izutsu, K., Yoshioka, S., Terao, T., 1994b. Effect of mannitolHekman, C., Park, S., Teng, W.Y., Guzman, N.A., Rossi, T., crystallinity on the stabilization of enzymes during freeze- 1995. Degradation of lyophilized and reconstituted drying. Chem. Pharm. Bull. (Tokyo) 42, 5 – 8. Macroscint. J. Pharm. Biomed. Anal. 13, 1249–1261. Izutsu, K., Yoshioka, S., Kojima, S., 1995. Increased stabiliz-Heimburg, T., Marsh, D., 1993. Investigation of secondary ing effects of amphiphilic excipients on freeze-drying of and tertiary structural changes of cytochrome c in com- lactate dehydrogenase (LDH) by dispersion into sugar plexes with anionic lipids using amide hydrogen exchange matrices. Pharm. Res. 12, 838 – 843. measurements: an FTIR study. Biophys. J. 65, 2408–2417. Izutsu, K., Yoshioka, S., Kojima, S., Randolph, T.W., Car-Heller, M.C., Carpenter, J.F., Randolph, T.W., 1997. Manipu- penter, J.F., 1996. Effects of sugars and polymers on lation of lyophilization-induced phase separation: implica- crystallization of poly(ethylene glycol) in frozen solutions: tions for pharmaceutical proteins. Biotechnol. Prog. 13, phase separation between incompatible polymers. Pharm. 590– 596. Res. 13, 1393 – 1400.Heller, M.C., Carpenter, J.F., Randolph, T.W., 1999a. Protein Jaenicke, R., 1990. Protein structure and function at low formulation and lyophilization cycle design: prevention of temperatures. Philos. Trans. R. Soc. Lond. B Biol. Sci. damage due to freeze-concentration induced phase separa- 326, 535 – 551. tion. Biotechnol. Bioeng. 63, 166–174. Jennings, T.A., 1999. Lyophilization: Introduction and BasicHeller, M.C., Carpenter, J.F., Randolph, T.W., 1999b. Con- Principles. Interpharm Press, Colorado, USA. formational stability of lyophilized PEGylated proteins in Jennings, T.A., Duan, N., 1995. Calorimetric monitoring of a phase-separating system. J. Pharm. Sci. 88, 58 –64. lyophilization. J. Parent. Sci. Technol. 49, 272 – 282.Heller, M.C., Carpenter, J.F., Randolph, T.W., 1999c. Appli- Jordan, G.M., Yoshioka, S., Terao, T., 1994. The aggregation cation of a thermodynamic model to the prediction of of bovine serum albumin in solution and in the solid state. phase separations in freeze-concentrated formulations for protein lyophilization. Arch. Biochem. Biophys. 363, 191– J. Pharm. Pharmacol. 46, 182 – 185. 201. Kasraian, K., Spitznagel, T.M., Juneau, J.A., Yim, K., 1998.Her, L.-M., Nail, S.L., 1994. Measurement of glass transition Characterization of the sucrose/glycine/water system by temperatures of freeze-concentrated solutes by differential differential scanning calorimetry and freeze-drying mi- scanning calorimetry. Pharm. Res. 11, 54–59. croscopy. Pharm. Dev. Technol. 3, 233 – 239.Hora, M.S., Rana, R.K., Smith, F.W., 1992a. Lyophilized Katakam, M., Banga, A.K., 1995. Aggregation of insulin and formulations of recombinant tumor necrosis factor. its prevention by carbohydrate excipients. PDA J. Pharm. Pharm. Res. 9, 33 – 36. Sci. Technol. 49, 160 – 165.Hora, M.S., Rana, R.K., Wilcox, C.L., Katre, N.V., Hirtzer, Katakam, M., Robillard, P., Brown, J., Tolman, G., 1998. P., Wolfe, S.N., Thomson, J.W., 1992b. Development of a Effects of moisture and lyophilization cycle on aggregation lyophilized formulation of interleukin-2. Dev Biol Stand. of a monoclonal antibody product. Pharm. Sci. 1 (Suppl.), 74, 295 – 303. S543.
56 W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60Kendrick, B., Chang, B.S., Dong, A., Kreilgard, L., Carpen- ˚ Leslie, S.B., Israeli, E., Lighthart, B., Crowe, J.H., Crowe, ter, J.F., 1995a. An infrared spectroscopic study on the L.M., 1995. Trehalose and sucrose protect both mem- structural changes induced by cooling and freezing stresses branes and proteins in intact bacteria during drying. Appl. on proteins. Pharm. Res. 12 (Suppl.), 84. Environ. Microbiol. 61, 3592 – 3597.Kendrick, B., Chang, B.Y., Carpenter, J.F., 1995b. Detergent Levine, H., Slade, L., 1992. Another view of trehalose for stabilization of proteins against surface and freezing denat- drying and stabilizing biological materials. Biopharm. 5 uration. Pharm. Res., 12 (Supple), S-85. (5), 36 – 40.Kendrick, B.S., Dong, A., Allison, S.D., Manning, M.C., Li, S., Nguyen, T., Schoneich, C., Borchardt, R.T., 1995a. ¨ Carpenter, J.F., 1996. Quantitation of the area of overlap Aggregation and precipitation of human relaxin induced between second-derivative amide I infrared spectra to de- by metal-catalyzed oxidation. Biochemistry 34, 5762 –5772. termine the structural similarity of a protein in different Li, S., Schoneich, C., Borchardt, R.T., 1995b. Chemical insta- ¨ bility of protein pharmaceuticals: mechanisms of oxidation states. J. Pharm. Sci. 85, 155–158. and strategies for stabilization. Biotechnol. Bioeng. 48,Kenney, W.C., Watson, E., Bartley, T., Boone, T., Altrock, 491 – 500. B.W., 1986. Parameters for the evaluation of IL-2 stability. Li, S., Patapoff, T.W., Overcashier, D., Hsu, C., Nguyen, Lymphokine Res. 5 (Suppl. 1), S23–S27. T.H., Borchardt, R.T., 1996. Effects of reducing sugars onKerr, W.L., Lim, M.H., Reid, D.S., Chen, H., 1993. Chemical the chemical stability of human relaxin in the lyophilized reaction kinetics in relation to glass transition temperature state. J. Pharm. Sci. 85, 873 – 877. in frozen food polymer solutions. J. Sci. Food Agric. 61, Librizzi, F., Vitrano, E., Cordone, L., 1999. Dehydration and 51–56. crystallization of trehalose and sucrose glasses containingKerwin, B.A., Heller, M.C., Levin, S.H., Randolph, T.W., carbonmonoxy-myoglobin. Biophys. J. 76, 2727 – 2734. 1998. Effects of Tween 80 and sucrose on acute short-term Lin, T.-Y., Timasheff, S.N., 1996. On the role of surface stability and long-term storage at − 20°C of a recombi- tension in the stabilization of globular proteins. Protein nant hemoglobin. J. Pharm. Sci. 87, 1062–1068. Sci. 5, 372 – 381.Kim, A.I., Akers, M.J., Nail, S.L., 1998. The physical state of Lin, L.S., Kunitani, M.G., Hora, M.S., 1996. Interferon-b-1b mannitol after freeze-drying: effect of mannitol concentra- (Betaseron®): a model for hydrophobic therapeutic tion, freezing rate, and a noncrystallizing cosolute. J. proteins. In: Pearlman, R., Wang, Y.J. (Eds.), Formula- Pharm. Sci. 87, 931 – 935. tion, Characterization, and Stability of Protein Drugs.Koseki, T., Kitabatake, N., Doi, E., 1990. Freezing denatura- Plenum Press, New York, pp. 275 – 301. tion of ovalbumin at acid pH. J. Biochem. 107, 389–394. Liu, R., Orgel, L., 1997. Efﬁcient oligomerization of negativelyKreilgaard, L., Frokjaer, S., Flink, J.M., Randolph, T.W., charged Beta-amino acids at −20°C. J. Am. Chem. Soc. 1998a. Carpenter JF Effects of additives on the stability of 119, 9192. recombinant human factor XIII during freeze-drying and Liu, W.R., Langer, R., Klibanov, A.M., 1990. Moisture-in- storage in the dried solid. Arch. Biochem. Biophys. 360, duced aggregation of lyophilized proteins in the solid state. 121– 134. Biotechnol. Bioeng. 37, 177 – 184.Kreilgaard, L., Jones, L.S., Randolph, T.W., Frokjaer, S., Liu, J.L., Lu, K.V., Eris, T., Katta, V., Westcott, K.R., Narhi, Flink, J.M., Manning, M.C., Carpenter, J.F., 1998b. Effect L.O., Lu, H.S., 1998. In vitro methionine oxidation of of Tween 20 on freeze-thawing- and agitation-induced recombinant human leptin. Pharm. Res. 15, 632 – 640. aggregation of recombinant human factor XIII. J. Pharm. Lu, Q., Zograﬁ, G., 1997. Properties of citric acid at the glass Sci. 87, 1597 – 1603. transition. J. Pharm. Sci. 86, 1374 – 1378. Lueckel, B., Bodmer, D., Helk, B., Leuenberger, H., 1998a.Kreilgaard, L., Frokjaer, S., Flink, J.M., Randolph, T.W., Formulations of sugars with amino acids or mannitol-inﬂ- Carpenter, J.F., 1999. Effects of additives on the stability uence of concentration ratio on the properties of the of Humicola lanuginosa lipase during freeze-drying and freeze-concentrate and the lyophilizate. Pharm. Dev. Tech- storage in the dried solid. J. Pharm. Sci. 88, 281–290. nol. 3, 325 – 336.Kristjansson, M.M., Kinsella, J.E., 1991. Protein and enzyme ´ Lueckel, B., Helk, B., Bodmer, D., Leuenberger, H., 1998b. stability: structural, thermodynamic, and experimental as- Effects of formulation and process variables on the aggre- pects. Adv. Food Nutr. Res. 35, 237–316. gation of freeze-dried interleukin-6 (IL-6) after lyophiliza-Kuhlman, B., Yang, H.Y., Boice, J.A., Fairman, R., Raleigh, tion and on storage. Pharm. Dev. Technol. 3, 337 – 346. D.P., 1997. An exceptionally stable helix from the riboso- Ma, X., Wang, D.Q., Bouffard, R., MacKenzie, A., 1998. mal protein L9: Implications for protein folding and stabil- Characterization of protein formulation for freeze-drying ity. J. Mol. Biol. 270, 640–647. cycle development. Pharm. Sci. 1 (Suppl.), S543.Lai, M.C., Topp, E.M., 1999. Solid-state chemical stability of Manning, M.C., Patel, K., Borchardt, R.T., 1989. Stability of proteins and peptides. J. Pharm. Sci. 88, 489–500. protein pharmaceuticals. Pharm. Res. 6, 903 – 917.Lam, X.M., Costantino, H.R., Overcashier, D.E., Nguyen, Mattern, M., Winter, G., Rudolph, R., Kohnert, U., Lee, G., T.H., Hsu, C.C., 1996. Replacing succinate with glyco- 1997. Formulation of proteins in vacuum-dried glasses. I. cholate buffer improves the stability of lyophilized inter- Improved vacuum drying using crystallizing amino acids. feron-g. Int. J. Pharm. 142, 85–95. Eur. J. Pharm. Biopharm. 44, 177 – 185.
W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60 57Mattern, M., Winter, G., Kohnert, U., Lee, G., 1999. Formu- Oksanen, C.A., Zograﬁ, G., 1993. Molecular mobility in mix- lation of proteins in vacuum-dried glasses. II. Process and tures of absorbed water and solid poly(vinylpyrrolidone). storage stability in sugar-free amino acid systems. Pharm. Pharm. Res. 10, 791 – 799. Dev. Technol. 4, 199 – 208. 8 Osterberg, T., Wadsten, T., 1999. Physical state of L-histidineMay, J.C., Wheeler, R.M., Grim, E., 1989. The gravimetric after freeze-drying and long-term storage. Eur. J. Pharm. method for the determination of residual moisture in Sci. 8, 301 – 308. freeze-dried biological products. Cryobiology 26, 277–284. 8 Osterberg, T., Fatouros, A., Mikaelsson, M., 1997. Develop-Mazzobre, M.F., Buera, M.P., Chirife, J., 1997. Glass transi- ment of freeze-dried albumin-free formulation of recombi- tion and thermal stability of lactase in low-moisture amor- nant factor VIII SQ. Pharm. Res. 14, 892 – 898. phous polymeric matrices. Biotechnol. Prog. 13, 195–199. Overcashier, D.E., Brooks, D.A., Costantino, H.R., Hsu,Meredith, P., Donald, A.M., Payne, R.S., 1996. Freeze-drying: C.C., 1997. Preparation of excipient-free recombinant hu- in situ observations using cryoenviromental scanning elec- man tissue-type plasminogen activator by lyophilization tron microscopy and differential scanning calorimetry. J. from ammonium bicarbonate solution: An investigation of Pharm. Sci. 85, 631 – 637. the two-stage sublimation phenomenon. J. Pharm. Sci. 86,Miller, D.P., de Pablo, J.J., Corti, H., 1997. Thermophysical 455 – 459. properties of trehalose and its concentrated aqueous solu- Overcashier, D.E., Patapoff, T.W., Hsu, C.C., 1999. tions. Pharm. Res. 14, 578–590. Lyophilization of protein formulations in vials: investiga-Moreira, T., Pendas, J., Gutierrez, A., Pomes, R., Duque, J., tion of the relationship between resistance to vapor ﬂow Franks, F., 1998. Effect of sucrose and rafﬁnose on physi- during primary drying and small-scale product collapse. J. cal state and on lactate dehydrogenase activity of freeze- Pharm. Sci. 88, 688 – 695. dried formulations. Cryo-Letters 19, 115–122. Paulsen, H., Pﬂughaupt, K.-W., 1980. Glycosylamines. In:Mozhaev, V.V., Martinek, K., 1984. Structure-stability rela- Pigman, W., Horton, D., Wander, J.D. (Eds.), The Carbo- tionships in proteins: new approaches to stabilizing en- hydrates Chemistry and Biochemistry. Academic Press, zymes. Enzyme Microb. Technol. 6, 50–59. New York, pp. 881 – 921.Nail, S.L., Johnson, W., 1992. Methodology for in-process Pearson, D.S., Smith, G., 1998. Dielectric analysis as a tool for determination of residual water in freeze-dried products. investigating the lyophilization of proteins. Pharm. Sci. Dev. Biol. Stand. 74, 137–150. Technol. Today 1, 108 – 117.Nagendra, H.G., Sukumar, N., Vijayan, M., 1998. Role of Perlman, R., Nguyen, T., 1992. Pharmaceutics of protein water in plasticity, stability, and action of proteins: the drugs. J. Pharm. Pharmcol. 44 (Suppl. 1), 178 – 185. crystal structures of lysozyme at very low levels of hydra- Physicians Desk Reference, 1999. Medical Economics Com- tion. Proteins: Struct. Func. Gen. 32, 229–240. pany Incorporation, 53rd edn. Montvale, NJ.Naini, V., Byron, P.R., Phillips, E.M., 1998. Physicochemical Pikal, M.J., 1990a. Freeze-drying of proteins. Part I: process stability of crystalline sugars and their spray-dried forms: design. Biopharm. 3 (8), 18 – 27. dependence upon relative humidity and suitability for use Pikal, M.J., 1990b. Freeze-drying of proteins. Part II: formula- in powder inhalers. Drug Dev. Ind. Pharm. 24, 895–909.Nakai, Y., Yoshioka, S., Aso, Y., Kojima, S., 1998. Solid-state tion selection. Biopharm. 3 (9), 26 – 30. rehydration-induced recovery of bilirubin oxidase activity Pikal, M.J., 1999. Mechanisms of protein stabilization during in lyophilized formulations reduced during freeze-drying. freeze-drying and storage: the relative importance of ther- Chem. Pharm. Bull. 46, 1031–1033. modynamic stabilization and glassy state relaxation dy-Nayar, R., 1998. Stabilized albumin-free recombinant factor namics. In: Rey, L., May, J.C. (Eds.), VIII having a low sugar content. US patent c5763401. Freeze-Drying/Lyophilization of Pharmaceutical and Bio-Nema, S., Avis, K.E., 1992. Freeze-thaw studies of a model logical Products, vol. 96. Marcel Dekker, New York, pp. protein, lactate dehydrogenase, in the presence of cryopro- 161 – 198. tectants. J. Parenter. Sci. Technol. 47, 76–83. Pikal, M.J., Rigsbee, D.R., 1997. The stability of insulin inNguyen, T.H., Burnier, J., Meng, W., 1993. The kinetics of crystalline and amorphous solids: observation of greater relaxin oxidation by hydrogen peroxide. Pharm. Res. 10, stability for the amorphous form. Pharm. Res. 14, 1379 – 1563 – 1571. 1387.O’Brien, J., 1996. Stability of trehalose, sucrose and glucose to Pikal, M.J., Shah, S., 1997. Intravial distribution of moisture nonenzymatic browning in model systems. J. Food Sci. 61, during the secondary drying stage of freeze drying. PDA J. 679– 682. Pharm. Sci. Technol. 51, 17 – 24.Oetjen, G.-W., 1999. Industrial freeze-drying for pharmaceuti- Pikal, K.A., Carpenter, J.F., 1998. pH changes during freezing cal applications. In: Rey, L., May, J.C. (Eds.), Freeze-Dry- in sodium and potassium phosphate buffer systems in the ing/Lyophilization of Pharmaceutical and Biological presence of glycine: effect on protein stability. Pharm. Sci. Products, vol. 96. Marcel Dekker, New York, pp. 267– 1 (Suppl.), 544. 335. Pikal, M.J., Dellerman, K.M., Roy, M.L., Riggin, R.M., 1991.Oksanen, C.A., Zograﬁ, G., 1990. The relationship between The effects of formulation variables on the stability of the glass transition temperature and water vapor absorp- freeze-dried human growth hormone. Pharm. Res. 8, 427 – tion by poly(vinylpyrrolidone). Pharm. Res. 7, 654–657. 436.
58 W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60Pikal, M.J., Dellerman, K., Roy, M.L., 1992. Formulation and Roos, Y., Karel, M., 1991. Phase transitions of mixtures of stability of freeze-dried proteins: effects of moisture and amorphous polysaccharides and sugars. Biotechnol. Prog. oxygen on the stability of freeze-dried formulations of 7, 49 – 53. human growth hormone. Dev. Biol. Stand. 74, 21–37. Roser, B., 1991. Trehalose drying: a novel replacement forPrestrelski, S.J., Arakawa, T., Carpenter, J.F., 1993a. Separa- freeze-drying. Biopharm. 4 (9), 47 – 53. tion of freezing and drying-induced denaturation of Ruddon, R.W., Bedows, E., 1997. Assisted protein folding. J. lyophilized proteins using stress-speciﬁc stabilization. II. Biol. Chem. 272, 3125. Structural studies using infrared spectroscopy. Arch. Sampedro, J.G., Guerra, G., Pardo, J.P., Uribe, S., 1998. Biochem. Biophys. 303, 465–473. Trehalose-mediated protection of the plasma membranePrestrelski, S.J., Tedeschi, N., Arakawa, T., Carpenter, J.F., H + -ATPase from Kluyveromyces lactis during freeze-dry- 1993b. Dehydration-induced conformational transitions in ing and rehydration. Cryobiology 37, 131 – 138. proteins and their inhibition by stabilizers. Biophy. J. 65, Sarciaux, J.M., Hageman, M.J., 1997. Effects of bovine soma- 661– 671. totropin (rbSt) concentration at different moisture levelsPrestrelski, S.J., Pikal, K.A., Arakawa, T., 1995. Optimization on the physical stability of sucrose in freeze-dried rbSt/su- of lyophilization conditions for recombinant human inter- crose mixtures. J. Pharm. Sci. 86, 365 – 371. leukin-2 by dried-state conformational analysis using Sarciaux, J.-M., Mansour, S., Hageman, M.J., Nail, S.L., Fourier-transform infrared spectroscopy. Pharm. Res. 12, 1998. Effect of species, processing conditions and phos- 1250 – 1259. phate buffer composition on IgG aggregation duringRamos, A., Raven, N.D.H., Sharp, R.J., Bartolucci, S., lyophilization. Pharm. Sci. 1 (Suppl.), 545. Roswsi, M., Cannio, R., Lebbink, J., vanderOost, J., de- Sasaoki, K., Hiroshima, T., Kusumoto, S., Nishu, K., 1992. Vos, W.M., Santos, H., 1997. Stabilization of enzymes Deamidation at asparagine-88 in recombinant human in- against thermal stress and freeze-drying by mannosylglyc- terleukin 2. Chem. Pharm. Bull. (Tokyo) 40, 976 – 980. erate. Appl. Environ. Microbiol. 63, 4020–4025. Sartor, G., Mayer, E., Johari, G.P., 1994. Calorimetric studies of the kinetic unfreezing of molecular motions in hydratedRavve, A., 1967. Physical properties of macromolecules. lysozyme, hemoglobin, and myoglobin. Biophys. J. 66, Chapter 2 in Organic Chemistry of Macromolecules,@ 249 – 258. Marcel Dekker, New York. pp. 17-36. Sastry, G.M., Agmon, N., 1997. Trehalose prevents myoglobinRemmele, R.L. Jr, Stushnoff, C., Carpenter, J.F., 1997. Real- collapse and preserves its internal mobility. Biochemistry time in situ monitoring of lysozyme during lyophilization 36, 7097 – 7108. using infrared spectroscopy: dehydration stress in the pres- Savage, M., Torres, J., Franks, L., Masecar, B., Hotta, J., ence of sucrose. Pharm. Res. 14, 1548–1555. 1998. Determination of adequate moisture content forRessing, M.E., Jiskoot, W., Talsma, H., van Ingen, C.W., efﬁcient dry-heat viral inactivation in lyophilized factor Beuvery, E.C., Crommelin, D.J., 1992. The inﬂuence of VIII by loss on drying and by near infrared spectroscopy. sucrose, dextran, and hydroxypropyl-beta-cyclodextrin as Biologicals 26, 119 – 124. lyoprotectants for a freeze-dried mouse IgG2a monoclonal Schebor, C., Buera, M.P., Chirife, J., 1996. Glassy state in antibody (MN12). Pharm. Res. 9, 266–270. relation to the thermal inactivation of enzyme invertase inRey, L., 1999. Glimpses into the realm of freeze-drying: classi- amorphous dried matrices of trehalose, maltodextrin and cal issues and new ventures. In: Rey, L., May, J.C. (Eds.), PVP. J. Food Eng. 30, 269 – 282. Freeze-Drying/Lyophilization of Pharmaceutical and Bio- Schebor, C., Burin, L., Buera, M.P., Aguilera, J.M., Chirife, logical Products, vol. 96. Marcel Dekker, New York, pp. J., 1997. Glassy state and thermal inactivation of invertase 1–30. and lactase in dried amorphous matrices. Biotechnol. Prog.Reyes, F.G.R., Poocharoen, B., Wrolstad, R.E., 1982. Mail- 13, 857 – 863. lard browning reaction of sugar–glycine model systems: Schmitt, E.A., Law, D., Zhang, G.G.Z., 1999. Nucleation and changes in sugar concentration, color and appearance. J. crystallization kinetics of hydrated amorphous lactose Food Sci. 47, 1376 – 1377. above glass transition temperature. J. Pharm. Sci. 88,Rossi, S., Buera, M.P., Moreno, S., Chirife, J., 1997. Stabiliza- 291 – 296. tion of the restriction enzyme EcoRI dried with trehalose Schwendeman, S.P., Costantino, H.R., Gupta, R.K., Siber, and other selected glass-forming solutes. Biotechnol. Prog. G.R., Klibanov, A.M., Langer, R., 1995. Stabilization of 13, 609 – 616. tetanus and diphtheria toxoids against moisture-inducedRoy, M.L., Pikal, M.J., Rickard, E.C., Maloney, A.M., 1992. aggregation. Proc. Natl. Acad. Sci. USA 92, 11234 – 11238. The effects of formulation and moisture on the stability of Schwendeman, S.P., Costantino, H.R., Gupta, R.K., Tobio, a freeze-dried monoclonal antibody-vinca conjugate: a test M., Chang, A.C., Alonso, M.J., Siber, G.R., Langer, R., of the WLF glass transition theory. Dev. Biol. Stand. 74, 1996. Strategies for stabilizing tetanus toxoid towards the 323– 339. development of a single-dose tetanus vaccine. Dev. Biol.Rupley, J.A., Careri, G., 1991. Protein hydration and func- Stand. 87, 293 – 306. tion. Adv. Protein Chem. 41, 37–173. Separovic, F., Lam, Y.H., Ke, X., Chan, H.K., 1998. ARoos, Y., 1993. Melting and glass transitions of low molecular solid-state NMR study of protein hydration and stability. weight carbohydrates. Carbohydr. Res. 238, 39–48. Pharm. Res. 15, 1816 – 1821.
W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60 59Shalaev, E.Y., Zograﬁ, G., 1996. How does residual water Sun, W.Q., Davidson, P., Chan, H.S.O., 1998. Protein stability affect the solid-state degradation of drugs in the amor- in the amorphous carbohydrate matrix: relevance to anhy- phous state? J. Pharm. Sci. 85, 1137–1141. drobiosis. Biochim. Biophy. Acta 1425, 245 – 254.Shamblin, S.L., Zograﬁ, G., 1999. The effects of absorbed Susi, H., Byler, D.M., 1986. Resolution-enhanced Fourier water on the properties of amorphous mixtures containing transform infrared spectroscopy of enzymes. Methods En- sucrose. Pharm. Res. 16, 1119–1124. zymol. 130, 290 – 311.Shamblin, S.L., Huang, E.Y., Zograﬁ, G., 1996. The effects of Tanaka, K., Takeda, T., Miyajima, K., 1991. Cryoprotective colyophilized polymeric additives on the glass transition effect of saccharides on denaturation of catalase by freeze- temperature and crystallization of amorphous sucrose. J. drying. Chem. Pharm. Bull. 39, 1091 – 1094. Therm. Anal. 47, 1567–1579. Tarelli, E., Mire-Sluis, A., Tivnann, H.A., Bolgiano, B.,Shamblin, S.L., Hancock, B.C., Zograﬁ, G., 1998. Water Crane, D.T., Gee, C., Lemercinier, X., Athayde, M.L., vapor sorption by peptides, proteins and their formula- Sutcliffe, N., Corran, P.H., Rafferty, B., 1998. Recombi- tions. Eur. J. Pharm. Biopharm. 45, 239–247. nant human albumin as a stabilizer for biological materialsSheehan, P., Liapis, A.I., 1998. Modeling of the primary and and for the preparation of international reference reagents. secondary drying stages of the freeze drying of pharmaceu- Biologicals 26, 331 – 346. tical products in vials: numerical results obtained from the Taylor, L.S., Zograﬁ, G., 1998a. The quantitative analysis of solution of a dynamic and spatially multi-dimensional crystallinity using FT-Raman spectroscopy. Pharm. Res. lyophilization model for different operational policies. Bio- 15, 755 – 761. technol. Bioeng. 60, 712–728. Taylor, L.S., Zograﬁ, G., 1998b. Sugar-polymer hydrogenShikama, K., Yamazaki, I., 1961. Denaturation of catalase by bond interactions in lyophilized amorphous mixtures. J. freezing and thawing. Nature 190, 83–84. Pharm. Sci. 87, 1615 – 1621.Shortle, D., 1996. The denatured state (the other half of the te Booy, M.P., de Ruiter, R.A., de Meere, A.L., 1992. Evalua- folding equation) and its role in protein stability. FASEB tion of the physical stability of freeze-dried sucrose-con- J. 10, 27 – 34. taining formulations by differential scanning calorimetry.Skrabanja, A.T., de Meere, A.L., de Ruiter, R.A., van den Pharm. Res. 9, 109 – 114. Oetelaar, P.J., 1994. Lyophilization of biotechnology prod- Timasheff, S.N., 1993. The control of protein stability and ucts. PDA J. Pharm. Sci. Technol. 48, 311–317. association by weak interactions with water: how do sol-Slade, L., Levine, H., Finley, J.W., 1989. Protein-water inter- vents affect these processes? Annu. Rev. Biophy. Biomol. actions: water as a plasticizer of gluten and other protein Struct. 22, 67 – 97. polymers. In: Philips, R.D., Finley, J.W. (Eds.), Protein Timasheff, S.N., 1998. Control of protein stability and reac- Quality and the Effects of Processing. Marcel Dekker, New tions by weakly interacting cosolvents: the simplicity of the York, pp. 9– 124. complicated. Adv. Protein Chem. 51, 355 – 432.Sola-Penna, M., Meyer-Fernandes, J.R., 1998. Stabilization Towns, J.K., 1995. Moisture content in proteins: its effects and against thermal inactivation promoted by sugars on en- measurement. J. Chromatogr. A705, 115 – 127. zyme structure and function: why is trehalose more effec- Townsend, M.W., DeLuca, P.P., 1990. Stability of ribonucle- tive than other sugars? Arch. Biochem. Biophys. 360, ase A in solution and the freeze-dried state. J. Pharm. Sci. 10–14. 79, 1083 – 1086.Strambini, G.B., Gabellieri, E., 1996. Proteins in frozen solu- Tsai, A.M., van Zanten, J.H., Betenbaugh, M.J., 1998. I. tions: evidence of ice-induced partial unfolding. Biophys. J. Study of protein aggregation due to heat denaturation: a 70, 971 – 976. structural approach using circular dichroism spectroscopy,Strickley, R.G., Anderson, B.D., 1996. Solid-state stability of nuclear magnetic resonance, and static light scattering. human insulin I. Mechanism and the effect of water on the Biotechnol. Bioeng. 59, 273 – 280. kinetics of degradation in lyophiles from pH 2–5 solutions. Tzannis, S.T., Prestrelski, S.J., 1999a. Activity-stability consid- Pharm. Res. 13, 1142 –1153. erations of trypsinogen during spray drying: effects ofStrickley, R.G., Anderson, B.D., 1997. Solid-state stability of sucrose. J. Pharm. Sci. 88, 351 – 359. human insulin I. Effect of water on reactive intermediate Tzannis, S.T., Prestrelski, S.J., 1999b. Moisture effects on partitioning in lyophiles from pH 2–5 solutions: stabiliza- protein-excipient interactions in spray-dried powders. Na- tion against covalent dimer formation. J. Pharm. Sci. 86, ture of destabilizing effects of sucrose. J. Pharm. Sci. 88, 645– 653. 360 – 370.Sun, W.Q., 1997. Temperature and viscosity for structural Uritani, M., Takai, M., Yoshinaga, K., 1995. Protective effect collapse and crystallization of amorphous carbohydrate of disaccharides on restriction endonucleases during drying solutions. Cryo-Letters 18, 99–106. under vacuum. J. Biochem. (Tokyo) 117, 774 – 779.Sun, W.Q., Davidson, P., 1998. Protein inactivation in amor- Vemuri, S., Yu, C.D., Roosdorp, N., 1994. Effect of cryopro- phous sucrose and trehalose matrices: effects of phase tectants on freezing, lyophilization, and storage of separation and crystallization. Biochim. Biophys Acta lyophilized recombinant alpha 1-antitrypsin formulations. 1425, 235 – 244. PDA J. Pharm. Sci. Technol. 48, 241 – 246.
60 W. Wang / International Journal of Pharmaceutics 203 (2000) 1–60Volkin, D.B., Klibanov, A.M., 1989. Minimizing protein in- Wong, D., Parascrampuria, J., 1997. Pharmaceutical excipi- activation. In: Creighton, T.E. (Ed.), Protein Function A ents for the stabilization of proteins. Pharm. Technol. 21, Practical Approach. Information Press, Oxford, UK, pp. 34 – 50. 1–24. Wu, S., Leung, D., Vemuri, S., Shah, D., Yang, B., Wang,Volkin, D.B., Middaugh, C.R., 1996. The characterization, J., 1998. Degradation mechanism of lyophilized basic stabilization, and formulation of acidic ﬁbroblast growth ﬁbroblast growth factor (BFGF) protein in sugar formu- factor. In: Pearlman, R., Wang, Y.J. (Eds.), Formula- lation. Pharm. Sci. 1 (Suppl.), S-539. tion, Characterization, and Stability of Protein Drugs. Yang, T.-H., Dong, A., Meyer, J., Johnson, O.L., Cleland, Plenum Press, New York, pp. 181–217. J.L., Carpenter, J.F., 1999. Use of infrared spectroscopyVrkljan, M., Foster, T.M., Powers, M.E., Henkin, J., Porter, to assess secondary structure of human growth hormone W.R., Staack, H., Carpenter, J.F., Manning, M.C., 1994. within biodegradable microspheres. J. Pharm. Sci. 88, Thermal stability of low molecular weight urokinase dur- 161 – 165. ing heat treatment. Effect of polymeric additives. Pharm. Yeo, S.-D., Debenedetti, P.G., Patro, S.Y., Przybycien, Res. 11 (2), 1004 – 1008. T.M., 1994. Secondary structure characterization of mi-Wang, W., 1996. Oral protein drug delivery. J. Drug Target croparticulate insulin powders. J. Pharm. Sci. 83, 1651 – 4, 195 – 232. 1656.Wang, W., 1999. Instability, stabilization, and formulation Yoshioka, S., Aso, Y., Izutsu, K., Terao, T., 1993. Aggre- of liquid protein pharmaceuticals. Int. J. Pharm. 185, gates formed during storage of b-galactosidase in solu- 129– 188. tion and in the freeze-dried state. Pharm. Res. 10,Wang, Y.J., Chen, H., Busch, M., Baldwin, P.A., 1997. 103 – 108. Headspace analysis for parenteral products: oxygen per- Yoshioka, S., Aso, Y., Izutsu, K.-I., Terao, T., 1994. Appli- meation and integrity test. Pharm. Technol. 21, 108–122. cation of accelerated testing for shelf-life prediction ofWard, K.R., Adams, G.D., Alpar, H.O., Irwin, W.J., 1999. commercial protein preparations. J. Pharm. Sci. 83, 454 – Protection of the enzyme L-asparaginase during lyophili- 456. sation — a molecular modelling approach to predict re- Yoshioka, S., Aso, Y., Kojima, S., 1997. Dependence of the quired level of lyoprotectant. Int. J. Pharm. 187, molecular mobility and protein stability of freeze-dried 153– 162. g-globulin formulations on the molecular weight of dex-Webb, S., Randolph, T., Carpenter, J., Cleland, J., 1998. An tran. Pharm. Res. 14, 736 – 741. investigation into reconstitution of lyophilized recombi- Yoshioka, S., Aso, Y., Nakai, Y., Kojima, S., 1998. Effect nant human interferon-gamma. Pharm. Sci. 1 (Suppl.), 542. of high molecular mobility of poly(vinyl alcohol) onWillemer, H., 1992. Measurements of temperatures, ice evap- protein stability of lyophilized g-globulin formulations. J. oration rates and residual moisture contents in freeze- Pharm. Sci. 87, 147 – 151. drying. Dev. Biol. Stand. 74, 123–134. Yoshioka, S., Aso, Y., Kojima, S., 1999. The effect of excip-Willemer, H., 1999. Experimental freeze-drying: procedures ients on the molecular mobility of lyophilized formula- and equipment. In: Rey, L., May, J.C. (Eds.), Freeze- tions, as measured by glass transition temperature and Drying/Lyophilization of Pharmaceutical and Biological NMR relaxation-based critical mobility temperature. Products, vol. 96. Marcel Dekker, New York, pp. 79– Pharm. Res. 16, 135 – 140. 121. Yu, L., Milton, N., Groleau, E.G., Mishra, D.S., Vansickle,Wisniewski, R., 1998. Developing large-scale cryopreserva- R.E., 1999. Existence of a mannitol hydrate during tion systems for biopharmaceutical products. Biopharm. freeze-drying and practical implications. J. Pharm. Sci. 11, 50 – 60. 88, 196 – 198.Wolkers, W.F., Oldenhof, H., Alberda, M., Hoekstra, F.A., Zhang, M.Z., Wen, J., Arakawa, T., Prestrelski, S.J., 1995. 1998a. A Fourier transform infrared microspectroscopy A new strategy for enhancing the stability of lyophilized study of sugar glasses: application to anhydrobiotic protein: the effect of the reconstitution medium on ker- higher plant cells. Biochim. Biophys. Acta 1379, 83–96. atinocyte growth factor. Pharm. Res. 12, 1447 – 1452.Wolkers, W.F., van Kilsdonk, M.G., Hoekstra, F.A., 1998b. Zhang, M.Z., Pikal, K., Nguyen, T., Arakawa, T., Prestrel- Dehydration-induced conformational changes of poly-L- ski, S.J., 1996. The effect of the reconstitution medium lysine as inﬂuenced by drying rate and carbohydrates. on aggregation of lyophilized recombinant interleukin-2 Biochim. Biophys. Acta 1425, 127–136. and ribonuclease A. Pharm. Res. 13, 643 – 646. .