Developability Assessments of Monoclonal Antibody Candidates to Minimize Aggregation During Large-Scale Ultrafiltration and Diafiltration (UF-DF) Processing

Assessing the Manufacturability and Critical Quality Attribute Profiles of Anti-IL-8 Immunoglobulin G Mutant Variants

Early-phase manufacturability assessment of high-concentration therapeutic monoclonal antibodies (mAbs) involves screening of process-related risks impacting their translation into the clinic. Manufacturing a mAb at scale relies on cost-effective and robust approaches to derisk manufacturability parameters, such as viscosity, conformational stability, aggregation, and process-related impurities. Using a panel of model anti-IL-8 IgG1 mutants, we investigate upstream and downstream processability, phase behavior, and process-related impurities. We correlate trends in the biophysical properties of mAbs with their cell growth, expression, filtration flux, solubility, and post-translational modifications. We find significant trends in increased relative free light chain expression with heavy chain mutants and detect a requirement for adjusted operation pH for cation exchange polishing steps with charge-altering variants. Moreover, trends between phase stability and high-concentration viscosity were observed. We also investigated unique correlations between increased glycosylation and biophysical behavior. Further in-depth analysis and modeling are required to elucidate the impact of the mAb sequence on the metabolism of the expression system, solubility limits, and alternative gelation models as future directions.

Correlating Differences in the Surface Activity to Interface-Induced Particle Formation in Different Protein Modalities: IgG mAb Versus Fc-Fusion Protein

The propensity of protein-based biologics to form protein particles during bioprocessing can be related to their interfacial properties. In this study, we compare the surface activity and interfacial film properties of two structurally different biologics, an IgG and Fc-fusion, in the absence and presence of interfacial dilatational stresses, and correlate these differences to their tendency to form interface-induced protein particles. Our results show that interface-induced particle formation is protein-dependent, with the Fc-fusion demonstrating greater interfacial stability. This observation can be correlated with faster adsorption kinetics of the Fc-fusion protein, and formation of a less incompressible film at the air-liquid interface. The addition of polysorbate 80 (PS80), commonly added to mitigate protein particle formation, led to a surfactant-dominant interface for quiescent conditions and coadsorption of protein and surfactant for the Fc-fusion when exposed to interfacial stress. On the other hand, for the IgG molecule, the surface always remained surfactant dominant. Image analysis demonstrated that PS80 was more effective in mitigating particle formation for the IgG than Fc-fusion. This suggests that a surfactant-dominant interface is necessary to prevent interface-induced protein particle formation. Further, while PS80 is effective in mitigating particle formation in the IgG formulation, it may not be the best choice for other protein modalities.

A Carrier Phase Ultrafiltration and Backflow Recovery Technique for Purification of Biological Macromolecules

A simple carrier phase based ultrafiltration technique that is akin to liquid chromatography and is suitable for medium-to-large volume sample preparation in the laboratory is discussed in this paper. A membrane module was integrated with a liquid chromatography system in a "plug and play" mode for ease of sample handling, and recovery of species retained by the membrane. The sample injector and pump were used for feed injection and for driving ultrafiltration, while the sensors and detectors were used for real-time monitoring of the separation process. The concentration of retained species was enriched by utilizing controlled concentration polarization. The recovery of the retained and enriched species was enhanced by backflow of carrier phase through the membrane using appropriate combination of valves. The backflow of carrier phase also cleaned the membrane and limited the extent of membrane fouling. Proof-of-concept of the proposed technique was provided by conducting different types of protein ultrafiltration experiments. The technique was shown to be suitable for carrying out protein fractionation, desalting, buffer exchange and concentration enrichment. Adoption of this approach is likely to make ultrafiltration easier to use for non-specialized users in biological research laboratories. Other advantages include enhanced product recovery, significant reduction in the number of diavolumes of buffer needed for conducting desalting and buffer exchange, minimal membrane fouling and the potential for repeated use of the same module for multiple separation cycles.

Using Passive Microrheology to Measure the Evolution of the Rheological Properties of NIST mAb Formulations during Adsorption to the Air-Water Interface

The development of novel protein-based therapeutics, such as monoclonal antibodies (mAbs), is often limited due to challenges associated with maintaining the stability of these formulations during manufacturing, storage, and clinical administration. An undesirable consequence of the instability of protein therapeutics is the formation of protein particles. MAbs can adsorb to interfaces and have the potential to undergo partial unfolding as well as to form viscoelastic gels. Further, the viscoelastic properties may be correlated with their aggregation potential. In this work, a passive microrheology technique was used to correlate the evolution of surface adsorption with the evolution of surface rheology of the National Institute of Standards and Technology (NIST) mAb reference material (NIST mAb) and interface-induced subvisible protein particle formation. The evolution of the surface adsorption and interfacial shear rheological properties of the NIST mAb was recorded in four formulation conditions: two different buffers (histidine vs phosphate-buffered saline) and two different pHs (6.0 and 7.6). Our results together demonstrate the existence of multiple stages for both surface adsorption and surface rheology, characterized by an induction period that appears to be purely viscous, followed by a sharp increase in protein molecules at the interface when the film rheology is viscoelastic and ultimately a slowdown in the surface adsorption that corresponds to the formation of solid-like or glassy films at the interface. When the transitions between the different stages occurred, they were dependent on the buffer/pH of the formulations. The onset of these transitions can also be correlated to the number of protein particles formed at the interface. Finally, the addition of polysorbate 80, an FDA-approved surfactant used to mitigate protein particle formation, led to the interface being surfactant-dominated, and the resulting interface remained purely viscous.

Correlating Surface Activity with Interface-Induced Aggregation in a High-Concentration mAb Solution

Interface-induced aggregation resulting in protein particle formation is an issue during the manufacturing and storage of protein-based therapeutics. High-concentration formulations of therapeutic proteins are even more prone to protein particle formation due to increased protein-protein interactions. However, the dependence of interface-induced protein particle formation on bulk protein concentration is not understood. Furthermore, the formation of protein particles is often mitigated by the addition of polysorbate-based surfactants. However, the details of surfactant-protein interactions that prevent protein particle formation at high concentrations remain unclear. In this work, a tensiometer technique was used to evaluate the surface pressure of an industrially relevant mAb at different bulk concentrations, and in the absence and presence of a polysorbate-based surfactant, polysorbate 20 (PS20). The adsorption kinetics was correlated with subvisible protein particle formation at the air-water interface and in the bulk protein solution using a microflow imaging technique. Our results showed that, in the absence of any surfactant, the number of subvisible particles in the bulk protein solutions increased linearly with mAb concentration, while the number of protein particles measured at the interface showed a logarithmic dependence on bulk protein concentration. In the presence of surfactants above the critical micelle concentration (CMC), our results for low-concentration mAb solutions (10 mg/mL) showed an interface that is surfactant-dominated, and particle characterization results showed that the addition of the surfactant led to reduced particle formation. In contrast, for the highest concentration (170 mg/mL), coadsorption of proteins and surfactants was observed at the air-water interface, even for surfactant formulations above CMC and the surfactant did not mitigate subvisible particle formation. Our results taken together provide evidence that the ratio between the surfactant and mAb molecules is an important consideration when formulating high-concentration mAb therapeutics to prevent unwanted aggregation.