Pharmaceutical Processing

be both relatively inexpensive and highly productive if samples are collected appropriately and questions are designed well. To complement the expensive bench-to-bedside research, academic institutes should focus on the studies to investigate the mechanisms of treatments already part of standard clinical care, which will generate income for the institution. These may simultaneously provide material for biological studies and help support a research infrastructure to sustain the more costly development of novel therapeutics (9). Overcome regulatory accrual limitations and patient stratification Current regulatory trend is to limit phase 1 and phase 2 studies to small cohorts of individuals in whom toxicity and therapeutic results can be efficiently tested, minimizing the risk and expense of large trials. The downside is that it decreases the quality of the scientific information resulting from the study because conclusions are drawn from smaller population samples, limiting the ability of the results to be generalized (10). Careful selection of biomarkers for patient stratification is crucial, as exemplified by the recognition that the expression of epithelial growth factor receptor (EGFR) is an accurate predictor of response to anti-EGFR therapy only when its ligand transforming growth factor-a, which stimulates cancer growth, is co-expressed (11). The epidemiology of the biomarker in the targeted population and its biological relevance are often not known, indicating the need for prospective collection of clinical material to identify with high-throughput technology new biomarkers and validate known ones. Cross-link of clinical information to high-quality sample collections High-throughput technologies enable researchers to study human disease worldwide, accounting for generic variability of individuals and the heterogeneity of their diseases. Samples obtained through venipuncture, fine-needle biopsies and cytological smears can be studied using DNA and RNA amplification techniques and high-sensitivity proteomic tools. There is a limited ability to link clinical information to highquality sample collection. Samples collected retrospectively are often unusable for analysis because materials degrade quickly after tissues or fluids are removed from the organism. Clinical trials are often designed without knowledge of these limitations and the collection and preservation of clinical samples does not follow the strict guidelines that would optimize their usefulness (12). The opportunity to take specific steps to preserve the ex vivo profile is often lost unless researchers identify the samples for use in a particular clinical study each time. Prospective collections are possible but require a great degree of cooperation between scientists and

clinicians. Good examples of broad-based efforts are starting to emerge, such as the integration of translational research in the European Organization for Research and Treatment of Cancer and the Cooperative Breast Cancer Tissue Resource (13-14) . These consortia have already collected large libraries of tumor samples prospectively linked to clinical information. Researchers can access standardized collections of tumor samples and relevant clinical data without violate the patient privacy.
Conclusions

Characterize formulation for developing a lyophilization process

deqian Wang

ABSTRACT

The field of translational research offers tremendous challenges and rewards. Such a large commitment cannot be made without a full understanding of the infrastructure, programs, and research to be supported. The goal is to meet the future challenge of bringing discoveries to practical application for the benefit of people. The cancer research enterprise involves not only futuristic research but also exciting efforts that will soon yield results. The translational research, whether it yields outcomes immediately or in the future, is conducted with the following purposes: to advancing science at the reasonable cost, to develop critical medicines that save peoples lives from the deadly illness, and to help human beings live healthier and longer.
References

One of the most important requirements for developing a lyophilization process is to thoroughly understand the formulation. This article describes how to characterize a crystalline matrix type of formulation in order to successfully develop a lyophilization process. Techniques discussed here for characterizing the formulation include Differential Scanning Calorimetry (DSC), lyophilization microscopy, water adsorption studies, and moisture optimization. A thorough understanding of a formulation provides the basis for developing the freeze-drying cycle parameters, such as freezing temperature, annealing temperature, shelf temperature, vacuum level, and time duration. It also assists in the validation of the cycle, particularly in bracketing the process parameters.
dr. deqian Wang received his Bs degree in Engineering in China and Ms & Ph.d degree also in Engineering in the u.s. after spending 5 years working in a star-up bioengineering company, d.Q. joined Bayer 11 years ago and is current the Head of Formulation, Freezedrying and drug delivery in Process sciences. d.Q. was the Chairman of Industrial advisory Board for the u.s. National science Foundation Research Center for Pharmaceutical Processing in 2004, an adjunct professor in Industrial Pharmaceutics at Purdue university; a member of the hall of fame in the school of Engineering at oregon state university. he was also the President of saPaWEsT in 2003.

Introduction

1. Horig, H., Marincola, E., and Marincola, F.M. (2005) Nature Medicine 11:705-708 2. http://dtp.nci.nih.gov/docs/raid/raid_pp.html 3. http://nihroadmap.nih.gov/clinicalresearch/overviewtranslational.asp 4. McGee, P. (2005) Drug Discovery & Developmet 8: 20-28 5. Roadblocks to cancer cures. (2004) Nature Medicine 10: 1003. 6. Monsurro, V. et al. (2003) Seminary Cancer Biology 13: 473-480. 7. Martin, J.B. & Kasper, D.L. (2000) New England Journal Medicine 243: 1646-1649. 8. Parks, M.R. & Disis, M.L. (2004) Journal of Translational Medicine 2: 28. 9. Mankoff, S.P., Brander, C., Ferrone, S. & Marincola, F.M. (2004) Journal of Translational Medicine 2: 3714-3721. 10. Hammond, M.E. & Taube, S.E. (2002) Seminary Oncology 29: 213-221. 11. Arteaga, C.L. & Baselga, J. (2003) Clinical Cancer Research 9: 1579-1589. 12. Lotze, M.T. et al. (2005) J. Immunotherapy 28: 79-119. 13. Lehmann, F., Lacombe, D., Therasse, P. & Eggermont, A.M. (2003) Journal of Translational Medicine 1: 2. 14. Glass, A.G. et al. (2001) Clinical Cancer Research 7: 1843-1849.

The most important step in developing a lyophilization cycle is to characterize the formulation. This can not be over emphasized because understanding the formulation will assist in developing a lyophilization cycle with a scientific rationale, instead of just depending on trial-and-error1. Characterization of a lyophilization formulation has a relatively long history. In 1960, Rey reported the use of DTA and a freezing microscope to investigate freezing process of biological products2. In 1964, MacKenzie reported that the first model of a lyophilization microscope that was constructed by R.J. Williams in 1962, enabled him to develop a second model as an improvement3. In 1972, MacKenzie also discussed how the lyophilization process could be affected by different formulations by measuring melting temperature (Tm), eutectic temperature (Te), glass temperature (Tg) and glass transition temperature (Tg) of a water/sucrose system and collapse temperatures (Tc) of NaCl/sucrose system4. In the 90s, probably due to numerous publications on the glass transition theory and the wide use of lyophilization in the biopharmaceutical industry, the importance of characterizing a lyophilized formulation gained recognition. Pikal reported that the eutectic and collapse temperatures could vary over a wide range, thus determination of the maximum allowable product temperature was extremely important and was the first step in formulation and process development5-7. Franks in 1991 pointed out that an understanding of a lyophilization formulation could remove most of the empiricism from the lyophilization cycle development1,8. In addition, the instrumentation used for characterizing formulation has also been improved significantly. Nail et al developed a further improved model of a lyophilization microscope that offered better temperature control, faster temperature responses, and sophisticated video recording and image capturing, among many others9. In the thermal analysis field, TA Instrument (New Castle, DE, USA) developed MDSC which extends the thermal analysis to the freeze-dried cakes in order to better understand and predict the stability of a lyophilized product. In this article, we will focus on techniques that have been used for characterizing lyophilized formulations. We will subsequently discuss
23

22

TRENds IN BIo/PhaRMaCEuTICaL INdusTRY | Winter 2005

TRENds IN BIo/PhaRMaCEuTICaL INdusTRY | Winter 2005

Pharmaceutical Processing

how to interpret data from these characterization studies for developing a lyophilization cycle. A lyophilized formulation can consist of various excipients. Sugars such as sucrose and trehalose are commonly used to stabilize protein conformation against denaturation which is caused by water removal during lyophilization. Polymers or proteins such as human albumin serum (HAS) also function similarly. These excipients usually remain amorphous during and after lyophilization. A formulation consisting of mainly amorphous excipients is often called amorphous formulation. On the other hand, bulking agents such as mannitol and glycine are commonly used to improve cake appearance. Such excipients are usually expected to be in crystalline structure after lyophilization. Totally crystalline cakes are not commonly used for protein lyophilization, since the crystallization of all excipients will tend to remove any stabilizing effect of the excipients on the protein10-12. For pharmaceutical protein products, a formulation termed crystalline matrix is widely used. In such a formulation, crystalline component is added at a relatively high level so that a crystalline matrix is formed for the amorphous components to collapse upon. In such a way, the crystalline component provides excellent cake appearance, good reconstitution characteristics, and ease in lyophilization. On the other hand, the amorphous component stabilizes proteins during processing and storage. Lyophilizing such a formulation allows partial collapse (also termed Microcollapse) without affecting cake appearance. As a result, the product can be lyophilized at a relatively higher product temperature if protein activity is not compromised. In this article, the discussion mainly focuses on the crystalline matrix type formulation. A lyophilization process typically consists of three steps. First, the aqueous solution that has been filled into containers (such as vials and trays), is frozen to below -40 C. Next, the primary drying takes place. The freeze dryer chamber is evacuated and the shelf temperature is elevated to sublimate bulk water (also termed free water) out of the system. Finally, the shelf temperature is further increased to remove bound water by desorption. This is called secondary drying. The principles of the lyophilization process have been well documented1, 5. Our development process for lyophilization is composed of five steps. First, the formulation is characterized. Second, based on the understanding of the formulation, the process is optimized. Third, the range of the critical process parameters is found. Fourth, the process is scaled up and transferred to production. Fifth, the process at the production scale is validated and qualified. In this article, we focus our discussion on the formulation characterization for a lyophilization

process development, particular on a crystalline matrix type formulation.

Utilize DSC to characterize formulation

Differential Scanning Calorimeter (DSC) measures heat flow as a function of temperature applied to a sample going through freezing, melting, crystallization, and glass transition. The thermal properties that can be measured by using DSC include eutectic crystallization temperature (Tx), eutectic melting temperature (Te), glass transition temperature (Tg) and ice melting temperature (Tim). Among these critical temperatures, the glass transition temperature, Tg, is one of the most important thermophysical properties of the formulation. For a formulation that forms an amorphous cake after being freeze-dried, the Tg is also the collapse temperature, which is the most critical factor in ensuring the success of the primary drying. For example, Mackenzie (1985) measured the collapse temperature of the formulations with different ratios of Sucrose versus Sodium Chloride. He demonstrated that at certain rang of ratios, the lyophilization became impractical as the sodium chloride depressed the collapse temperature to a point below 40C 13. Many other reports are available in determining the Tg by DSC 14, 15, 16. To measure thermal properties of a formulation, the sample is first frozen to -60C and then warmed up to 20C. During the warming, a glass transition temperature (Tg) can be identified. The determined glass transition temperature is the highest allowable product temperature during primary drying. That is because that the cake will collapse if the product temperature is higher than this highest allowable temperature during sublimation so that the glass transition temperature is also the collapse temperature. Generally speaking, the lower the allowable product temperature is, the less efficient the lyophilization cycle will be. It may become even impractical for large scale production if the highest allowable temperature approaches -40C. For the crystalline matrix type formulation, however, adding an annealing step in freezing may make a significant difference in the lyophilization process. Annealing is a step that after sample is fully frozen, the sample temperature was slowly increased to a point in the range of 10 to 20C and then held for one hour at this temperature. Annealing enables crystalline components in the formulation crystallize out; and as a consequence it will allow primary drying to occur at a much higher product temperature. In addition to DSC, other instruments, such as differential thermal analysis (DTA) and electrical resistance analysis (ER), are also commonly used in determining the thermophysical

properties of a lyophilized formulation, such as glass transition temperature (Tg) and collapse temperature, etc. One of the examples using DTA and ER was recently reported by X. Ma et al on a crystalline matrix formulation 15. Other literatures on the application of DAT 18, 19, 22 and ER 3, 17, 18, 19, 20, 21, in formulation characterization are also very informative. However, DSC is the most common and reliable means, and also the easiest.
Confirm DSC Results with a Lyophilization Microscope

A lyophilization microscopy provides real-time images of freezing, melting, crystallization, collapse, and meltback during the freezing and lyophilization processes. A lyophilization microscope is shown in Figure 1, consisting of two major parts of an optical microscope and a microscope stage. To investigate the lyophilization behavior of a formulation, a small aliquot (10ml) of formulation is frozen to 50C between quartz coverslips in the lyophilization stage, either with or without annealing. The behavior of the material during sublimation is then observed using a polarizing microscope. Micrographs are recorded using a video camera that is mounted on the microscope. The magnification used for observation of the lyophilization front is 150-fold. More details about this type of lyophilization microscope can be found from Nail 9. Studies of lyophilization microscopy were also reported by others 3, 4, 13, 22, 23, 24.

cycle and primary drying cycles. For example of a crystalline matrix type formulation, annealing is necessary to crystallize the crystalline component so the highest allowable product temperature is elevated and therefore, primary drying can be conducted with a much higher efficiency. In addition, during primary drying, both heat transfer and mass transfer are important in affecting product temperature. In other words, we can achieve a product temperature lower than the highest allowable temperature by using many different combinations of shelf temperature and chamber pressure. Figure 2 demonstrates the possible combinations for a crystalline matrix type formulation. This provides very important information not only for designing a primary drying process, but also for bracketing process parameters during validation. For example, if -20C is the highest allowable temperature for the formulation, the figure illustrates that all combinations of shelf temperature and chamber pressure resulting in a product temperature lower than -20C would be acceptable process parameters. In practice, we usually take at least 2-50C as a safety margin, and we always optimize the process by looking for the combinations that result in the highest sublimation rate and consequently, the shortest drying time.

Figure 2. Relationship of product temperature as a function of shelf temperature and chamber pressure Study Water Sorption

Figure 1. A photo of a lyophilization microscope

The lyophilization microscope has become an excellent tool to confirm DSC results. The advantage of a lyophilization microscope is that it provides real time images, which can be visually seen, during freezing and lyophilization. With these two different methods, we can confidently design the freezing

Water sorption, including desorption and adsorption, provide the degree of hygroscopicity of the formulation by plotting the increase in water content of a dry cake as a function of relative humidity to which the sample has been exposed at a given temperature. The curve indicates water vapor sorption characteristics, in other words, the degree of ease in secondary drying. Although the desorption process is not necessary a straight-forward reversal of the adsorption, the trend is predictive. In general, the easier the water adsorption, the easier the water desorption.

24

TRENds IN BIo/PhaRMaCEuTICaL INdusTRY | Winter 2005

TRENds IN BIo/PhaRMaCEuTICaL INdusTRY | Winter 2005

25

Pharmaceutical Processing

The simplest way to conduct a water sorption study is to store the product in desiccators with solid salt or saturated salt solutions for 48 hours equilibration at an ambient temperature. The salts included phosphorus pentoxide, lithium chloride, potassium acetate, magnesium chloride, potassium carbonate, and sodium chloride, which generated relative humidities of approximately 0, 11%, 23%, 33%, 43% and 75%, respectively. The product vials were sealed immediately after equilibration. The moisture in the lyophilized product was determined by the Karl Fischer method. Then a water adsorption curve that illustrates a progressive increase in water content as increase of the relative humidity was generated to provide information on the affinity of water for a dried product. Another tool for determining water sorption is Dynamic Vapor Sorption (DVS, Surface Measurement System, London). It measures weight changes due to moisture gain at different relative humidity conditions. This instrument also provides information on water sorption kinetics.
Optimize Moisture Content

long secondary drying cycle to make a bone-dry product. In addition, partial production loads can be comfortably bracketed in the validation because over-drying is not a concern for such a formulation.

moisture contents the relationship between Tg and moisture content can be quantified. Again, samples for Tg measurement are prepared similarly to those for adsorption study using desiccators described previously.

With DSC and lyo-microscopy, we are able to determine the thermophysical properties of a formulation such as freezing, melting, crystallization, and glass transition temperatures. This information allows us to precisely design freezing and primary drying parameters, such as annealing temperature, shelf temperature and pressure. By generating a water adsorption curve and conducting a moisture optimization study, we can optimize the secondary drying process with regard to its drying temperature, pressure and time. In summary, we have demonstrated that a good understanding of formulation characteristics will assist in developping an optimal lyophilization cycle with a scientific rationale.
References

Lyophilization is a process to remove water so that the water activity, or the mobility of the water, in the dried product can be reduced to an optimal level. This is achieved by using a secondary drying cycle. As a result, fully understanding the effects of the moisture content on product stability becomes the most important information necessary to develop a secondary drying. Generally speaking, for any formulation there is an optimal range of moisture content that would result in the best stability of the lyophilized product. Hsu et al. (1991) reported that there is an optimum residual moisture range for a lyophilized recombinant protein25. Overdrying may result in opalescence in the product upon reconstitution, while underdrying leads to a greater protein activity loss upon storage under temperature stress conditions. Greiff (1971) 26 and Liu et al. (1991) 27 revealed that aggregation or insoluble proteins can be induced by moisture content that was higher than the optimal range. Samples for a moisture optimization study are usually generated in the same desiccators that incubate samples for the water adsorption study. That is, samples were exposed to different relative humidities for 48 hours to allow them to adsorb various amounts of moisture. An accelerated stability program was then conducted at 40C storage temperature. For example, monomer percentage of a protein product, shown in Figure 3, was measured by the SEC-HPLC method. Data showed a clear trend that as moisture increased, monomer percentage dropped rapidly, indicating protein aggregation increased drastically. The data demonstrates a trend of the dryer, the better, which suggests having a

Figure 3. Percent Monomer Measured by SEC-HPLC for Samples Stored at 40 C

Measure Tg of dry cakes

Figure 4. A typical thermogram of Tg measurement using MDSC. A typical thermogram and testing conditions are given in Figure 4. Glass transition values were determined from the reversing heat flow and then corrected to the total heat flow. The relationship of Tg and moisture content for the crystalline matrix formulation is given in Figure 5. It shows if the relationship between Tg and the moisture content of a dry cake is known, an optimum moisture content can be chosen based upon the desired storage conditions. Product stability at a higher storage temperature can be achieved by lowering the moisture content of the cake as much as possible.

Another way to determine the optimal moisture content for stability is to develop a relationship between moisture content and glass transition temperature (Tg) of lyophilized dry cakes. Tg can be measured using either DSC or MDSC. Tg is a measure of the molecular motion of water, which is a major factor in denaturing proteins 28, 29. In order to increase the long-term storage stability of a pharmaceutical protein, the motion of water can be limited (frozen storage) or water can be removed (lyophilization). However, even for a lyophilized biopharmaceutical product, residual moisture in the dry cake could be mobile depending on the level of the residual moisture content and storage temperature. High residual moisture content and high storage temperature may result in more mobility of water29, and as a consequence, may damage the product protein significantly as well as harm the cosmetic properties of the lyophilized cake. The glass transition temperature of a freeze-dried cake (Tg) marks an increase in the mobility of the remaining water28. Freezedried products stored below their Tg will exhibit much greater stability and structural integrity than those stored above Tg. Due to the cost of storing a product at low temperatures, it is economically more desirable for the Tg of a lyophilized product to be relatively high. Each product formulation will have a different relationship between moisture content and glass transition 29. Modulated Differential Scanning Calorimetry (MDSC) can be used to characterize the glass transition of a dry cake 30, 31. By analyzing cakes of different

Fig.5. Tg of dry cakes as a function of moisture content

Conclusion

1. F. Franks. Freeze-Drying: From Empiricism to Predictability. The Significance of Glass Transitions. In J. May and F. Brown (eds.), Developments in Biological Standardization, Karger, Basel, Vol. 74, 1991, pp. 9-38. 2. L. Rey. Thermal analysis of eutectics in freezing solutions. Ann. N.Y. Acad. Sci. 85 (part 2): 510-534 (1960). 3. A.P. Mackenzie. Apparatus for microscopic observations during freeze-drying. Biodynamica, 1964, 9 (186), 213 -222. 4. A.P. MacKenzie. Freezing, Freeze-drying and Freezesubstitution. Proceedings of the workshop on biological specimen preparation techniques for scanning electron microscopy, IIT Research Institute, April, 1972. 5. M.J. Pikal. Freeze-drying of proteins. Part I: Process Design. BioPharm. 1990, 3(8), 18-27. 6. M.J. Pikal. Freeze-drying of proteins. Part II: formulation selection. BioPharm. 1990. 10: 26-30. 7. M. Pikal and S. Shah. The collapse temperature in freezedrying: dependence on measurement methodology and rate of water removal from the glassy phase. International Journal of Pharmaceutics, 62, 165-186 (1990). 8. F. Franks. Improved freeze-drying: an analysis of the basic scientific principles. Proc. Biochem. 2: iii-vii (1989). 9. S.L. Nail, L.M. Her, P.B. Christopher, and L.L. Nail. An improved microscope stage for direct observation of freezing and freeze drying. Pharm. Res., 1994, 11 ( 8), 1098 -1100. 10. K. Izutsu, S.Yoshioka, and S. Kojima. Physical stability and protein stability of freeze-dried cakes during storage at elevated temperatures. Pharm. Res., 1994, 11(7), 995-9. 11. T. Arakawa, S.J. Prestrelski, W.C. Kenney, and J.F. Carpenter. Factors affecting short-term and long-term stabilities of proteins. Adv. Drug Deliv. Rev., 1993, 10, 1-28. 12. J.F. Carpenter, S.J. Prestrelski, and T. Arakawa. Separation of freezing- and drying-induced denaturation of lyophilized proteins using stress-specific stabilization. Archiv. Biochem. Biophys., 1993, 303(2), 456-64. 13. A.P. MacKenzie. A current understanding of the freezedrying of representative aqueous solutions. In Refrigeration

2

TRENds IN BIo/PhaRMaCEuTICaL INdusTRY | Winter 2005

TRENds IN BIo/PhaRMaCEuTICaL INdusTRY | Winter 2005

27

Science and Technology. 1985, pp. 21-34 14. L.M. Her and S.L. Nail. Measurement of glass transition temperatures of freeze-concentrated solutes by differential scanning calorimetry. Pharm. Res., 1994, 11 (1), 54 - 9. 15. X. Ma, D.Q. Wang, R. Bouffard, and A. MacKenzie. Characterization of Murine Monoclonal Anitbody to Tumor Necrosis Factor (TNF-MAb) Formulation for Freeze-Drying Cycle Development. Pharmaceutical Research Vol. 18, No. 2: 196-202 (2001). 16. R.H.M. Hatley. The effective use of differential scanning calorimetry in the optimization of freeze-drying processes and formulations. In Developments in Biological Standardization., 17. T. Jennings. Thermal-analysis instrumentation for lyophilization. MD&DI, 11: 49-57 (1980). 18. L.R. Rey. Thermal analysis of eutectic in freezing solution. International Institute of Refrigeration and Laboratoire de Physiologie, Ecole Normale, Paris, France, pp. 510 -534. 1975. 19. L. Rey. Study of the freezing and drying of tissues at very low temperatures. In Recent research in freezing and drying, A. Parkes and A. Smith (eds.) Blackwell, Oxford. 1960, pp. 40-62 20. P. DeLuca and L. Lachman. Lyophilization of pharmaceuticals, I: effect of certain physical-chemical properties. J. Pharm. Sci. 54: 617-624 (1965) 21. A.P. MacKenzie. Changes in electrical resistance during freezing and their application to the control of the freezedrying process. In Refrigeration Science and Technology 1985, pp. 155-163 22. A.P. MacKenzie. Factors affecting the transformation of ice into water vapor in the freeze-drying process. Ann. N.Y. Acad. Sci., 125: 522-547 (1965). 23. A.P. MacKenzie, Collapse during freeze-dryingqualitative and quantitative aspects. In Freeze-drying and advanced food technology. S. Goldblith, L. Rey and W. Rothmayr (eds.). Academic Press. 1975, pp.277-307 24. L. Rey, Dispositif pour lexamen microscopique aux basses temperatures. Experientia. 1957, XIII: 201-202 (1957). 25. C.C. Hsu, C.A. Ward, R. Pearlman, H.M. Nguyen, D.A. Yeung, and J.G. Curley, Determinig the optimum residual moisture in lyophilized protein pharmaceuticals. Dev. Biol. Stand., 1991, 74, 257-71. 26. Greiff, D. Protein structure and freeze-drying effects of residual moisture and gases. Cryobiology, 1971, 8, 145-52. 27. W.R. Liu, R. Langer, and A.M. Klibanov. Moistureinduced aggregation of lyophilized protein in the solid state. Biotechnol. Bioeng., 1991, 37, 177-84. 28. S. Yoshioka, Y. Aso, and S. Kojima. The effect of excipients on the molecular mobility of lyophilized formulations, as measured by glass transition temperature and NMR relaxation-based critical mobility temperature.

Pharmaceutical Research, Vol. 16, No. 1, 1999. 29. B. Hancock and G. Zografi. The relationship between the glass transition temperature and the water content of amorphous pharmaceutical solids. Pharmaceutical Research. Vol. 11, No. 4, 1994. 30. D. Miller, J. de Pablo, and H. Corti. Thermophysical properties of trehalose and its concentrated aqueous solutions. Pharmaceutical Research, Vol. 14, No. 5, 1997. 31. P. Royal, D. Craig, and C. Doherty. Characterisation of the glass transition of an amorphous drug using modulated DSC. Pharmaceutical Research. Vol. 15, No. 7, 1998.

28

TRENds IN BIo/PhaRMaCEuTICaL INdusTRY | Winter 2005