Multiscale Coarse-Grained Approach to Investigate Self-Association of Antibodies.

Self-association of therapeutic monoclonal antibodies (mabs) are thought to modulate the undesirably high viscosity observed in their concentrated solutions. Computational prediction of such a self-association behavior is advantageous early during mab drug candidate selection when material availability is limited. Here, we present a coarse-grained (CG) simulation method that enables microsecond molecular dynamics simulations of full-length antibodies at high concentrations. The proposed approach differs from others in two ways: first, charges are assigned to CG beads in an effort to reproduce molecular multipole moments and charge asymmetry of full-length antibodies instead of only localized charges. This leads to great improvements in the agreement between CG and all-atom electrostatic fields. Second, the distinctive hydrophobic character of each antibody is incorporated through empirical adjustments to the short-range van der Waals terms dictated by cosolvent all-atom molecular dynamics simulations of antibody variable regions. CG simulations performed on a set of 15 different mabs reveal that diffusion coefficients in crowded environments are markedly impacted by intermolecular interactions. Diffusion coefficients computed from the simulations are in correlation with experimentally measured observables, including viscosities at a high concentration. Further, we show that the evaluation of electrostatic and hydrophobic characters of the mabs is useful in predicting the nonuniform effect of salt on the viscosity of mab solutions. This CG modeling approach is particularly applicable as a material-free screening tool for selecting antibody candidates with desirable viscosity properties.

Mutational landscape of antibody variable domains reveals a switch modulating the interdomain conformational dynamics and antigen binding.

Somatic mutations within the antibody variable domains are critical to the immense capacity of the immune repertoire. Here, via a deep mutational scan, we dissect how mutations at all positions of the variable domains of a high-affinity anti-VEGF antibody G6.31 impact its antigen-binding function. The resulting mutational landscape demonstrates that large portions of antibody variable domain positions are open to mutation, and that beneficial mutations can be found throughout the variable domains. We determine the role of one antigen-distal light chain position 83, demonstrating that mutation at this site optimizes both antigen affinity and thermostability by modulating the interdomain conformational dynamics of the antigen-binding fragment. Furthermore, by analyzing a large number of human antibody sequences and structures, we demonstrate that somatic mutations occur frequently at position 83, with corresponding domain conformations observed for G6.31. Therefore, the modulation of interdomain dynamics represents an important mechanism during antibody maturation in vivo.

In silico selection of therapeutic antibodies for development: viscosity, clearance, and chemical stability.

For mAbs to be viable therapeutics, they must be formulated to have low viscosity, be chemically stable, and have normal in vivo clearance rates. We explored these properties by observing correlations of up to 60 different antibodies of the IgG1 isotype. Unexpectedly, we observe significant correlations with simple physical properties obtainable from antibody sequences and by molecular dynamics simulations of individual antibody molecules. mAbs viscosities increase strongly with hydrophobicity and charge dipole distribution and decrease with net charge. Fast clearance correlates with high hydrophobicities of certain complementarity determining regions and with high positive or high negative net charge. Chemical degradation from tryptophan oxidation correlates with the average solvent exposure time of tryptophan residues. Aspartic acid isomerization rates can be predicted from solvent exposure and flexibility as determined by molecular dynamics simulations. These studies should aid in more rapid screening and selection of mAb candidates during early discovery.

Cathepsin B Cleavage of vcMMAE-Based Antibody-Drug Conjugate Is Not Drug Location or Monoclonal Antibody Carrier Specific.

Antibody-drug conjugates (ADCs) require thorough characterization and understanding of product quality attributes. The framework of many ADCs comprises one molecule of antibody that is usually conjugated with multiple drug molecules at various locations. It is unknown whether the drug release rate from the ADC is dependent on drug location, and/or local environment, dictated by the sequence and structure of the antibody carrier. This study addresses these issues with valine-citrulline-monomethylauristatin E (vc-MMAE)-based ADC molecules conjugated at reduced disulfide bonds, by evaluating the cathepsin B catalyzed drug release rate of ADC molecules with different drug distributions or antibody carriers. MMAE drug release rates at different locations on ADC I were compared to evaluate the impact of drug location. No difference in rates was observed for drug released from the V(H), V(L), or C(H)2 domains of ADC I. Furthermore, four vc-MMAE ADC molecules were chosen as substrates for cathepsin B for evaluation of Michaelis-Menten parameters. There was no significant difference in K(M) or k(cat) values, suggesting that different sequences of the antibody carrier do not result in different drug release rates. Comparison between ADCs and small molecules containing vc-MMAE moieties as substrates for cathepsin B suggests that the presence of IgG1 antibody carrier, regardless of its bulkiness, does not impact drug release rate. Finally, a molecular dynamics simulation on ADC II revealed that the val-cit moiety at each of the eight possible conjugation sites was, on average, solvent accessible over 50% of its maximum solvent accessible surface area (SASA) during a 500 ns trajectory. Combined, these results suggest that the cathepsin cleavage sites for conjugated drugs are exposed enough for the enzyme to access and that the drug release rate is rather independent of drug location or monoclonal antibody carrier. Therefore, the distribution of drug conjugation at different sites is not a critical parameter to control in manufacturing of the vc-MMAE-based ADC conjugated at reduced disulfide bonds.

Auristatin antibody drug conjugate physical instability and the role of drug payload.

The conjugation of hydrophobic cytotoxic agents such as monomethyl auristatin E (MMAE) to the interchain sulfhydryl groups of monoclonal antibodies (Mabs) through a protease-labile linker generates a heterogeneous drug load distribution. The conjugation process can generate high-drug-load species that can affect the physical stability of antibody-drug conjugates (ADCs). In this study, the mechanism of physical instability of ADCs was investigated by formulating the ADC pool as well as isolated drug load species in high and low ionic strength buffers to understand the effect of ionic strength on the stability of drug-conjugated Mabs. The results showed that the presence of high ionic strength buffer led to time-dependent aggregate and fragment formation of ADCs, predominantly ADCs with high-drug-load species under stress conditions. In addition, differential scanning calorimetry (DSC) results confirmed that there is a direct correlation between thermal unfolding and drug payload and that specific changes in the DSC thermogram profiles can be assigned to modifications by MMAE.

Opposite effects of polyols on antibody aggregation: thermal versus mechanical stresses.

PURPOSE: To investigate the physical stability of antibody-polyol formulations under thermal and mechanical stresses. METHODS: mAb-U was analyzed in buffer, trehalose, sucrose, glycerol and ethylene glycol solutions at pH 7.0. T(m1) of mAb-U was determined using DSC. Thermal stress studies were performed by incubating mAb-U-polyol solutions at 40°C (2 months), 50°C (3 weeks) and 65°C (5 days). Mechanical stress studies were conducted by shaking mAb-U-polyol solutions at 200 rpm for 5 days at 25°C. RESULTS: Trehalose and glycerol increased the T(m1) of mAb-U, whereas ethylene glycol decreased it. The trend observed in the order of increasing aggregation of mAb-U after thermal stress (40°C and 50°C) was buffer = trehalose = sucrose

Construction, MD simulation, and hydrodynamic validation of an all-atom model of a monoclonal IgG antibody.

At 150 kDa, antibodies of the IgG class are too large for their structure to be determined with current NMR methodologies. Because of hinge-region flexibility, it is difficult to obtain atomic-level structural information from the crystal, and questions regarding antibody structure and dynamics in solution remain unaddressed. Here we describe the construction of a model of a human IgG1 monoclonal antibody (trastuzumab) from the crystal structures of fragments. We use a combination of molecular-dynamics (MD) simulation, continuum hydrodynamics modeling, and experimental diffusion measurements to explore antibody behavior in aqueous solution. Hydrodynamic modeling provides a link between the atomic-level details of MD simulation and the size- and shape-dependent data provided by hydrodynamic measurements. Eight independent 40 ns MD trajectories were obtained with the AMBER program suite. The ensemble average of the computed transport properties over all of the MD trajectories agrees remarkably well with the value of the translational diffusion coefficient obtained with dynamic light scattering at 20 degrees C and 27 degrees C, and the intrinsic viscosity measured at 20 degrees C. Therefore, our MD results likely represent a realistic sampling of the conformational space that an antibody explores in aqueous solution.

Vibrational spectroscopy and chemometrics to characterize and quantitate trehalose crystallization.

Investigating the phase behavior of sugars in ice and lyophilized solids is of significant interest in the pharmaceutical industry. In this study, Raman and near infrared (NIR) spectroscopy are used to characterize and quantitate trehalose crystallization using several chemometric models. The predictive behaviors of partial least squares (PLS), principal component analysis (PCA), and multiple linear regression (MLR) models are compared. In general, PLS and PCA outperform linear and MLR models. Changes in specific vibrational modes associated with several coupled motions are described and assigned as a function of crystal content. In addition to characterization and quantitation, our method may be used to localize gradients of amorphous and/or crystallized trehalose within a sample.

Crystallization of trehalose in frozen solutions and its phase behavior during drying.

PURPOSE: (i) To study the crystallization of trehalose in frozen solutions and (ii) to understand the phase transitions during the entire freeze-drying cycle. METHOD: Aqueous trehalose solution was cooled to -40°C in a custom-designed sample holder. The frozen solution was warmed to -18°C and annealed, and then dried in the sample chamber of the diffractometer. XRD patterns were continuously collected during cooling, annealing and drying. RESULTS: After cooling, hexagonal ice was the only crystalline phase observed. However, upon annealing, crystallization of trehalose dihydrate was evident. Seeding the frozen solution accelerated the solute crystallization. Thus, phase separation of the lyoprotectant was observed in frozen solutions. During drying, dehydration of trehalose dihydrate yielded a substantially amorphous anhydrous trehalose. CONCLUSIONS: Crystallization of trehalose, as trehalose dihydrate, was observed in frozen solutions. The dehydration of the crystalline trehalose dihydrate to substantially amorphous anhydrate occurred during drying. Therefore, analyzing the final lyophile will not reveal crystallization of the lyoprotectant during freeze-drying. The lyoprotectant crystallization can only become evident by continuous monitoring of the system during the entire freeze-drying cycle. In light of the phase separation of trehalose in frozen solutions, its ability to serve as a lyoprotectant warrants further investigation.

Carboxylate-dependent gelation of a monoclonal antibody.

PURPOSE: This paper shows the first ever assembly of monoclonal antibody using multivalent carboxylate ions into highly ordered structures that feature viscoelastic properties reminiscent of other filamentous proteins. METHODS: A monoclonal antibody was assembled into filamentous networks by adding multivalent carboxylates to the protein solution. Gelation and characterization of these networks were monitored using mechanical rheometry, electron microscopy, Fourier transform infra-red and Raman spectroscopy. RESULTS: Electron microscopy and mechanical rheometry suggest the formation of rigid filament bundles that feature strong interfilament interactions. Filament network elasticity increased with multivalent carboxylate and protein concentrations, hinting at the importance of multivalent carboxylates in the mechanism of assembly. CONCLUSION: Assembly is not triggered by high ionic strength but with multivalent carboxylates. A high protein concentration is required for filament formation and the elasticity of the networks are weakly dependent on concentration. The exact mechanism of assembly is still elusive, although we speculate that carboxylates could act as a bridge to crosslink antibody monomers. These monoclonal antibody monomers could be linked either through Fab-Fab or Fc-Fab regions, although previous reports have shown evidence of reversible self-association mediated through the Fab regions.

Polysorbate 20 prevents the precipitation of a monoclonal antibody during shear.

Monoclonal antibodies (MAbs) are widely used as therapeutic proteins and they are frequently exposed to a high degree of stress during manufacturing or delivery. MAbs shear thin upon increasing shear rates. After undergoing multiple shear cycles, with a cone-and-plate rheometer, the solution viscosity of high concentration antibodies increases due to the formation of insoluble aggregates. These shear-induced insoluble aggregates do not form when polysorbate 20 is present in solution. We hypothesize that monoclonal antibodies form a thin protein layer at the air-water interface. MAbs at the interface expose their hydrophobic core to air leading to unfolding, multiple non-specific intermolecular interactions and, upon continuous high shear, precipitation. Surface tension analysis confirms that monoclonal antibodies are surface active and that polysorbate 20 can prevent their interaction with the air-water interface. In addition, we complement these findings with a viscometer that measures bulk viscosity without the influence of an air-liquid interfacial viscosity and find that the bulk viscosity increases slightly when Mab solutions contained polysorbate 20. These methods of analysis could be used when designing manufacturing systems in which a protein solution is subject to shear forces.

Split-Cakes, Still Delicious.

"Elegant" lyophilized pharmaceutical product cakes are expected to appear as uniform foamy plugs with little shrinkage and minimal cracking. While studying internal cake structures, we have on occasion observed some cakes that were very sharply split horizontally, roughly in halves, with foamy top and lamellar bottom regions. After many years and numerous experiments, we can finally propose a mechanism for the formation of these cakes with unusual internal structures. This phenomenon involves a complex interplay of momentum, heat, mass transfer, and phase equilibria.LAY ABSTRACT: Freeze drying (lyophilization) is a common unit operation in the manufacturing of pharmaceutical drugs. The typical final lyophilized product is expected to look like a uniform porous plug, or cake, that has foamy (sponge-like) morphology. However, we have occasionally observed cakes that were split horizontally, with the top and bottom layers exhibiting very distinctive and totally different structures. This intriguing phenomenon has not been discussed in the literature. In this report, we present experimental results that lead us to a mechanism by which split-cakes form.

In vitro model for predicting bioavailability of subcutaneously injected monoclonal antibodies.

Monoclonal antibodies (mAbs), which are now more frequently administered by subcutaneous (SC) injection rather than intravenously, have become a tremendously successful drug format across a wide range of therapeutic areas. Preclinical evaluations of mAbs to be administered by SC injection are typically performed in species such as mice, rats, minipigs, and cynomolgus monkeys to obtain critical information regarding formulation performance and prediction of PK/PD outcomes needed to select clinical doses for first-in-human studies. Despite extensive efforts, no preclinical model has been identified to date that accurately predicts clinical outcomes for these SC injections. We have addressed this deficiency with a novel in vitro instrument, termed Scissor, to model events occurring at the SC injection site and now further validated this approach using a set of eight mAbs for which clinical PK/PD outcomes have been obtained. Diffusion of these mAbs from the Scissor system injection cartridge into a large volume physiological buffer, used to emulate mAb movement from the SC injection site into the systemic circulation, provided distinct profiles when monitored over a 6h period. Curve-fitting analysis of these profiles using the Hill equation identified parameters that were used, along with physicochemical properties for each mAb, in a partial least squares analysis to define a relationship between molecule and formulation properties with clinical PK outcomes. The results demonstrate that parameters of protein charge at neutral pH and isoelectric point (pI) along with combined formulation properties such as viscosity and mAb concentration can dictate the movement of the mAb from the injection cartridge to infinite sink compartment. Examination of profile characteristics of this movement provided a strong predictive correlation for these eight mAbs. Together, this approach demonstrates the feasibility of this in vitro modelling strategy as a tool to identify drug and formulation properties that can define the performance of SC injected medicines and provide the potential for predicting clinical outcomes that could be useful for formulation selection and a first-in-human clinical dosing strategy.

An Improved Method of Predicting Extinction Coefficients for the Determination of Protein Concentration.

Concentration determination is an important method of protein characterization required in the development of protein therapeutics. There are many known methods for determining the concentration of a protein solution, but the easiest to implement in a manufacturing setting is absorption spectroscopy in the ultraviolet region. For typical proteins composed of the standard amino acids, absorption at wavelengths near 280 nm is due to the three amino acid chromophores tryptophan, tyrosine, and phenylalanine in addition to a contribution from disulfide bonds. According to the Beer-Lambert law, absorbance is proportional to concentration and path length, with the proportionality constant being the extinction coefficient. Typically the extinction coefficient of proteins is experimentally determined by measuring a solution absorbance then experimentally determining the concentration, a measurement with some inherent variability depending on the method used. In this study, extinction coefficients were calculated based on the measured absorbance of model compounds of the four amino acid chromophores. These calculated values for an unfolded protein were then compared with an experimental concentration determination based on enzymatic digestion of proteins. The experimentally determined extinction coefficient for the native proteins was consistently found to be 1.05 times the calculated value for the unfolded proteins for a wide range of proteins with good accuracy and precision under well-controlled experimental conditions. The value of 1.05 times the calculated value was termed the predicted extinction coefficient. Statistical analysis shows that the differences between predicted and experimentally determined coefficients are scattered randomly, indicating no systematic bias between the values among the proteins measured. The predicted extinction coefficient was found to be accurate and not subject to the inherent variability of experimental methods. We propose the use of a predicted extinction coefficient for determining the protein concentration of therapeutic proteins starting from early development through the lifecycle of the product.LAY ABSTRACT: Knowing the concentration of a protein in a pharmaceutical solution is important to the drug's development and posology. There are many ways to determine the concentration, but the easiest one to use in a testing lab employs absorption spectroscopy. Absorbance of ultraviolet light by a protein solution is proportional to its concentration and path length; the proportionality constant is the extinction coefficient. The extinction coefficient of a protein therapeutic is usually determined experimentally during early product development and has some inherent method variability. In this study, extinction coefficients of several proteins were calculated based on the measured absorbance of model compounds. These calculated values for an unfolded protein were then compared with experimental concentration determinations based on enzymatic digestion of the proteins. The experimentally determined extinction coefficient for the native protein was 1.05 times the calculated value for the unfolded protein with good accuracy and precision under controlled experimental conditions, so the value of 1.05 times the calculated coefficient was called the predicted extinction coefficient. Comparison of predicted and measured extinction coefficients indicated that the predicted value was very close to the experimentally determined values for the proteins. The predicted extinction coefficient was accurate and removed the variability inherent in experimental methods.

Molecular simulations of micellar aggregation of polysorbate 20 ester fractions and their interaction with N-phenyl-1-naphthylamine dye.

Micellar aggregation behavior of polysorbate 20 (PS20) has generated significant interest because of the wide use of PS20 as a surfactant to minimize protein surface adsorption and mitigate protein aggregation. Thus, there is a need for better molecular understanding of what drives the biophysical behavior of PS20 in solution. We observe that a complex amphipathic PS20 molecule, which contains both hydrophobic tail and relatively large hydrophilic head, self-associates strongly within the course of a molecular dynamics simulation performed with a fully atomistic representation of the molecule and an explicit water solvent model. The in silico behavior is consistent with micellar models of PS20 in solution. The dynamics of this self-association is rather complex involving both internal reorganization of the molecule and diffusion to form stable micelle-like aggregates. The micellar aggregates of PS20 are long-lived and are formed by the balance between the large hydrophobic interactions associated with the aliphatic tail of PS20, and the steric repulsion of the hydrophilic sorbitan head structure. In the present work, molecular models of PS20 that represent naturally occurring PS20 fractions were produced and characterized in silico. The study investigated the monoester and diester fractions: PS20M, and PS20D. These fractions present differences in the strength of their hydrophobic effect, which influences the aggregation behavior. Adaptive biasing force (ABF) simulations were carried out with the PS20M and PS20D molecular constructs to calculate the free energy of their pairwise interaction. The free energy barrier for the dissociation is higher for PS20D compared with PS20M. The results show that hydrogen bonds can form when head groups are in close proximity, such as in the PS20 aggregate assembly, and the free energy of interaction can be used to predict the morphology of the micellar aggregate for the different PS20 fractions. We were also able to simulate PS20 in the presence of N-phenyl-1-naphthylamine (NPN) to study the solution behavior of the hydrophobic molecule and of the mechanism in which it is sequestered in the hydrophobic core of the PS20 micellar aggregate.

A Spectral Method for Color Quantitation of a Protein Drug Solution.

Color is an important quality attribute for biotherapeutics. In the biotechnology industry, a visual method is most commonly utilized for color characterization of liquid drug protein solutions. The color testing method is used for both batch release and on stability testing for quality control. Using that method, an analyst visually determines the color of the sample by choosing the closest matching European Pharmacopeia reference color solution. The requirement to judge the best match makes it a subjective method. Furthermore, the visual method does not capture data on hue or chroma that would allow for improved product characterization and the ability to detect subtle differences between samples. To overcome these challenges, we describe a quantitative method for color determination that greatly reduces the variability in measuring color and allows for a more precise understanding of color differences. Following color industry standards established by International Commission on Illumination, this method converts a protein solution's visible absorption spectra to L*a*b* color space. Color matching is achieved within the L*a*b* color space, a practice that is already widely used in other industries. The work performed here is to facilitate the adoption and transition for the traditional visual assessment method to a quantitative spectral method. We describe here the algorithm used such that the quantitative spectral method correlates with the currently used visual method. In addition, we provide the L*a*b* values for the European Pharmacopeia reference color solutions required for the quantitative method. We have determined these L*a*b* values by gravimetrically preparing and measuring multiple lots of the reference color solutions. We demonstrate that the visual assessment and the quantitative spectral method are comparable using both low- and high-concentration antibody solutions and solutions with varying turbidity. LAY ABSTRACT: In the biotechnology industry, a visual assessment is the most commonly used method for color characterization, batch release, and stability testing of liquid protein drug solutions. Using this method, an analyst visually determines the color of the sample by choosing the closest match to a standard color series. This visual method can be subjective because it requires an analyst to make a judgment of the best match of color of the sample to the standard color series, and it does not capture data on hue and chroma that would allow for improved product characterization and the ability to detect subtle differences between samples. To overcome these challenges, we developed a quantitative spectral method for color determination that greatly reduces the variability in measuring color and allows for a more precise understanding of color differences. The details of the spectral quantitative method are described. A comparison between the visual assessment method and spectral quantitative method is presented. This study supports the transition to a quantitative spectral method from the visual assessment method for quality testing of protein solutions.

Validation of a Spectral Method for Quantitative Measurement of Color in Protein Drug Solutions.

A quantitative spectral method has been developed to precisely measure the color of protein solutions. In this method, a spectrophotometer is utilized for capturing the visible absorption spectrum of a protein solution, which can then be converted to color values (L*a*b*) that represent human perception of color in a quantitative three-dimensional space. These quantitative values (L*a*b*) allow for calculating the best match of a sample's color to a European Pharmacopoeia reference color solution. In order to qualify this instrument and assay for use in clinical quality control, a technical assessment was conducted to evaluate the assay suitability and precision. Setting acceptance criteria for this study required development and implementation of a unique statistical method for assessing precision in 3-dimensional space. Different instruments, cuvettes, protein solutions, and analysts were compared in this study. The instrument accuracy, repeatability, and assay precision were determined. The instrument and assay are found suitable for use in assessing color of drug substances and drug products and is comparable to the current European Pharmacopoeia visual assessment method. LAY ABSTRACT: In the biotechnology industry, a visual assessment is the most commonly used method for color characterization, batch release, and stability testing of liquid protein drug solutions. Using this method, an analyst visually determines the color of the sample by choosing the closest match to a standard color series. This visual method can be subjective because it requires an analyst to make a judgment of the best match of color of the sample to the standard color series, and it does not capture data on hue and chroma that would allow for improved product characterization and the ability to detect subtle differences between samples. To overcome these challenges, we developed a quantitative spectral method for color determination that greatly reduces the variability in measuring color and allows for a more precise understanding of color differences. In this study, we established a statistical method for assessing precision in 3-dimensional space and demonstrated that the quantitative spectral method is comparable with respect to precision and accuracy to the current European Pharmacopoeia visual assessment method.

Modeling of protein-anion exchange resin interaction for the human growth hormone charge variants.

Modeling ion exchange chromatography (IEC) behavior has generated significant interest because of the wide use of IEC as an analytical technique as well as a preparative protein purification process; indeed there is a need for better understanding of what drives the unique behavior of protein charge variants. We hypothesize that a complex protein molecule, which contains both hydrophobic and charged moieties, would interact strongly with an in silico designed resin through charged electrostatic patches on the surface of the protein. In the present work, variants of recombinant human growth hormone that mimic naturally-occurring deamidation products were produced and characterized in silico. The study included these four variants: rhGH, N149D, N152D, and N149D/N152D. Poisson-Boltzmann calculations were used to determine surface electrostatic potential. Metropolis Monte Carlo simulations were carried out with the resulting variants to simulate IEC systems, examining the free energy of the interaction of the protein with an in silico anion exchange column represented by polylysine polypeptide. The results show that the charge variants have different average binding energies and the free energy of interaction can be used to predict the retention time for the different variants.

Protein Aggregation in Frozen Trehalose Formulations: Effects of Composition, Cooling Rate, and Storage Temperature.

This study was designed to assess the effects of cooling rate, storage temperature, and formulation composition on trehalose phase distribution and protein stability in frozen solutions. The data demonstrate that faster cooling rates (>100°C/min) result in trehalose crystallization and protein aggregation as determined by Fourier Transform Near-Infrared (FT-NIR) spectroscopy and size-exclusion chromatography, respectively. Conversely, at slower cooling rates (≤1°C/min), trehalose remains predominantly amorphous and there is no effect on protein stability. Evaluation of storage temperatures demonstrates that aggregation increases more rapidly at -14°C compared with higher (-8°C) and lower (-20°C) storage temperatures; however, a relatively higher amount of cumulative aggregation was observed at lower (-20°C) temperature compared with higher storage temperatures (-14°C and -8°C). Further evaluation of the effects of formulation composition suggests that the phase distribution of amorphous and crystallized trehalose dihydrate in frozen solutions depends on the ratio of trehalose to mAb. The results identify an optimal range of trehalose-mAb (w/w) ratio, 0.2-2.4, capable of physically stabilizing mAb formulations during long-term frozen storage-even for fast cooled (>100°C/min) formulations.

A novel in vitro method to model the fate of subcutaneously administered biopharmaceuticals and associated formulation components.

Subcutaneous (SC) injection is becoming a more common route for the administration of biopharmaceuticals. Currently, there is no reliable in vitro method that can be used to anticipate the in vivo performance of a biopharmaceutical formulation intended for SC injection. Nor is there an animal model that can predict in vivo outcomes such as bioavailability in humans. We address this unmet need by the development of a novel in vitro system, termed Scissor (Subcutaneous Injection Site Simulator). The system models environmental changes that a biopharmaceutical could experience as it transitions from conditions of a drug product formulation to the homeostatic state of the hypodermis following SC injection. Scissor uses a dialysis-based injection chamber, which can incorporate various concentrations and combinations of acellular extracellular matrix (ECM) components that may affect the release of a biopharmaceutical from the SC injection site. This chamber is immersed in a container of a bicarbonate-based physiological buffer that mimics the SC injection site and the infinite sink of the body. Such an arrangement allows for real-time monitoring of the biopharmaceutical within the injection chamber, and can be used to characterize physicochemical changes of the drug and its interactions with ECM components. Movement of a biopharmaceutical from the injection chamber to the infinite sink compartment simulates the drug migration from the injection site and uptake by the blood and/or lymph capillaries. Here, we present an initial evaluation of the Scissor system using the ECM element hyaluronic acid and test formulations of insulin and four different monoclonal antibodies. Our findings suggest that Scissor can provide a tractable method to examine the potential fate of a biopharmaceutical formulation after its SC injection in humans and that this approach may provide a reliable and representative alternative to animal testing for the initial screening of SC formulations.