A critical quality parameter for all AAV products is the proportion of capsids, which contain the gene of interest in its full length, the so-called ‘full-to-empty ratio’. It will be shown that no single currently available analytical method allows an accurate determination of this parameter. As such, a panel of assays should be used to fully characterise the AAV product. This talk will discuss several of these assays, along with their advantages and possible pitfalls.

MolMed has developed an analytical panel, lentiviral platform based, in order to release and characterize vectors and LV genetically modified cell products. All the assays have been studied and optimized in order to achieve the best optimization in term of volume of sample used, time for the analysis, applicability of different projects and GMP suitability according ICH Q2 (R1) guidelines. Each test is developed in order to be suitable for clinical trial application and to be completely validated in case of marketing submission. Increase robustness, accuracy and throughput by automation and decreasing method variability are goals, which are in continuous improvement.

The complexity of multispecific antibodies requires a comprehensive set of analytical techniques to guide lead discovery and optimization. An overview of these techniques will be presented with a focus on mispairing analysis and functional characterization of multispecific drug candidates. Furthermore, the integrated developability concept at Sanofi Biologics will be presented along with show cases highlighting the challenges in characterization and developability of multispecifics.

The safety and efficacy of complex biopharmaceuticals, such as antibodies and antibody-drug conjugates, not only depend on their purity, but also on their higher-order structure (HOS). Consequently, the characterization of molecular structures is of utmost importance throughout the pharmaceutical development trajectory. To investigate the thermal stability of an antibody and its corresponding drug conjugate, a series of orthogonal analytical techniques, both 'classical' spectroscopic, as well as mass spectrometry-based, was applied. In this way, not only the performance of the methods applied could be studied, but the data obtained also provided insight into degradation pathways of the molecules investigated. In addition to providing a feel for the sensitivity, accuracy, linearity, and precision of biophysical, i.e. spectroscopic, characterization methods, the addition of mass spectrometric data potentially reveals the actual regions of the molecules impacted, which is essential information in the process of assessing whether a change in structure is not only significant, but also scientifically relevant. The results will be presented that together paint a comprehensive picture of: a) the performance and orthogonality of the methods routinely used for HOS characterization in the pharmaceutical industry; and b) the HOS of both antibodies and antibody-drug conjugates, how it differs amongst them, and how it responds to heat.

Rexgenero-France is producing bald-lentivectors lacking the VSV-G immunogenic protein which are later encapsulated in targeting ligand incorporated biodegradable, oligo-peptide modified poly(beta amino ester)s to obtain T-cell targeting hybrid nanoparticles (NPs). Further, these hybrid NPs are used as therapeutic agents to generate in vivo CAR-T therapy.  

AUC is widely used as an orthogonal technique to confirm the aggregation content routinely measured by size-exclusion chromatography. In order to optimize accuracy and precision, it is typically employed at moderate concentrations around 0.5 mg per ml in a simple PBS-like buffer. The aim is to optimize the sample composition in order to meet the requirements of the technique best. However, the advent of high-concentration formulations of biopharmaceuticals at 100 mg per ml or even higher, and many of them self-buffering, renders this approach questionable. Furthermore, predictions on long-term stability and aggregation propensity are only valid if the intact system, the undiluted protein in its native formulation, is analyzed. The recent development of new fitting algorithms for non-ideal sedimentation in Sedanal and sedfit allows the determination of self- and cross-sedimentation terms, as well as the non-ideality parameters, kS and kD. In combination with 3D printed centerpieces, this allows the accurate description of protein solutions up to 100 mg per ml, and even the discovery of previously unrecognized self-association (Chaturvedi et al. 2019). We have explored the capacity of AUC to accurately quantify the aggregation content of a marketed antibody formulation (Humira) using a model protein mimicking an antibody dimer. We extended this approach on a nano-particle drug, using 1 mm centerpieces to a concentration of 50 mg/ml.

Topics that will be covered in the presentation include: 1) native MS platform robustness, including impact of transferring from one MS platform to another; 2) streamlining of sample preparation, with a focus on simplicity, robustness, and a hands-off approach; and 3) case studies from stability, lead selection, and clone selection studies.

Recent years have seen great advances in interfacing formerly MS-incompatible separation modes to mass spectrometric detection. In this course, we developed MS-friendly, pH-gradient-based cation- and anion-exchange chromatography methods, which have proven exceptionally powerful for the characterisation of monoclonal antibodies and other protein formats. Additionally, the CE-MS-based ZipChip platform was explored and found to be a highly potent and complementary technique for the sensitive and comprehensive characterisation of complex biopharmaceuticals.

switchSENSE® tackles the challenge of quantifying association and dissociation rates of antibodies in their native environment. The new heliX Biosensor delivers in vitro data on target and off-target kinetic rates measured on cell surfaces with defined epitope distributions. Multiplexed resolution of affinity and avidity speeds up development of your bi- and multi-specific, or locked, formats.

Over the past decade, a wide variety of separation methods have been successfully coupled to native MS to characterize the heterogeneity of mAbs and related products. Here, we report the development of several novel native LC-MS techniques and showcase their applications in drug product characterization. In addition, we introduce an integrated native LC-MS solution that offers large dynamic range, high robustness and great versatility.

Identifying the "Critical Quality Attributes" of a lead mAb candidate early in the research phase is mandatory for selecting "Right the First Time" molecule and avoiding late-stage failures. Among the various intrinsic properties that require in-depth characterization, the top quality attributes of biotherapeutics, such as conformational and serum stability, non-specific binding, and aggregation propensity, need immediate characterization. Integrated biophysical screening of these attributes, in combination with sequence quality assessment, offers a successful predictive tool for selecting risk-free candidates early on. This talk is designed to provide case studies highlighting the criticality of developing such an early tool kit in the discovery phase. 

Top and middle-down biologics characterization approaches aim at capturing a direct snapshot of all proteoforms with their combinatorial distribution. This presentation will focus on a new data analysis workflow to reveal relevant diagnostic information in middle-down MS spectra, missed by the classical middle-down approach. This new workflow enabled the localization of a deamidation event and a sequence variant site directly from the gas-phase fragmentation of a mAb light chain.

To ensure the safety and efficacy of viral vector based drugs, it is crucial to characterize their critical quality attributes (CQA) throughout product development, manufacturing, and QC. We will discuss how SEC-MALS can quantify three CQAs of AAVs in one single assay: capsid concentration, payload content and degree of aggregation.