Untangling the Identity of Charge Variants in Antibody-Drug Conjugates

Antibody-drug conjugates (ADCs) are next-generation therapeutic proteins that hold promise as effective treatment options for progressive diseases. These molecules are constructed by coupling a monoclonal antibody with a small molecule compound and are engineered for precise delivery of targeted cytotoxic compounds to kill tumor cells. During drug development, therapeutic proteins are often produced in heterogeneous form, with post-translational modifications and genetic mutations being a common cause of the observed heterogeneity. Post-translational modifications (PTMs), which include glycosylation, deamidation, and oxidation, can produce charge variants that affect the stability, efficacy, potency, and safety of the protein therapeutics. These PTMs may be critical quality attributes of a protein therapeutic, and so require characterization and control throughout the pipeline. Monitoring these PTMs at an intact protein level can be done through analysis of charge variants and is a fundamental quality control requirement for release and stability purposes.

ADCs Pose an Analytical Challenge

It is important to monitor and manage post-translational modifications of protein therapeutics in early upstream development to avoid carrying over undesired modifications into the downstream development phases. Charge variant analysis of intact proteins is critical for establishing a valid release method and ensuring process consistency for producing protein therapeutics. The industry standard for protein charge analysis for drugs such as monoclonal antibodies (mAbs) is imaged capillary isoelectric focusing (icIEF), a high-resolution method that separates proteins based on their isoelectric point (pI).

With the evolution of the therapeutic pipeline from simple monoclonal antibodies to more complex therapeutic molecules like ADCs, the challenges in monitoring post-translational modifications at the intact level have increased. For ADCs, it is necessary to analyze, interpret, and understand 3 separate components: the antibody, the linker, and the small-molecule payload or drug, all of which impart charge upon the molecule and alter the charge-heterogeneity profile.

With the evolution of the therapeutic pipeline from simple monoclonal antibodies to more complex therapeutic molecules like ADCs, the challenges in monitoring post-translational modifications at the intact level have increased. Image Credit: © huenstructurebio.com - stock.adobe.com

Kristin Shultz-Kuszak, PhD, scientist II at MedImmune, and her team working have routinely analyzed ADCs using imaged capillary isoelectric focusing (icIEF), with known and expected differences observed between the icIEF profiles of the mAb intermediates and the conjugate ADC. However, unexpected peaks and charge profiles arise, and identifying the cause of these changes takes time. In the case of a specific ADC (ADC-1), while Shultz-Kuszak’s team observed the typical peak distribution for the mAb, there were many more peaks for the conjugated molecule (the mAb-ADC pair) than expected. A complication to this discovery was that the molecule appeared to be increasingly unstable over the duration of analysis, suggesting that the standard icIEF assay was not fit for the purpose of characterizing this ADC.

A previous study had shown that ring opening events in the ADC payload influence charge distribution and may contribute to the complexity observed in icIEF profiles.¹ The ADC payload that the team were analyzing had a similar linker chemistry.² The team began to explore platforms to help them resolve the characterization issues encountered with their specific ADC.

Accelerating Results for Charge Variant Analysis of ADCs

Detailed characterization and identification of icIEF peak constituents has typically been performed using orthogonal chromatography methods, such as ion-exchange chromatography (IEX), followed by characterization by mass spectrometry (MS). Characterization by MS typically requires the collection of fractions for injection into the mass spectrometer for mass identification—a process that can run from weeks to months. For complex molecules like ADCs, a single chromatography peak could be focused across multiple peaks in the icIEF profile, increasing the complexity of identifying what is under the icIEF peak and what is causing the shift in charge profile.

To resolve the characterization issues encountered with ADC-1, the team utilized the Intabio ZT System, a single platform that couples icIEF-UV analysis with high-resolution MS detection, to enable mass identification of charge variants without switching instruments. The platform enabled streamlined separation, quantification, and identification of protein charge variants. Furthermore, the ability to compare pI and mass profiles in real-time provided assurance that all icIEF peaks were accounted for by MS (Figure 1). and enables the gathering of multiple pieces of critical data, faster than the conventional and separated workflows of IEX/MS.

Figure 1. Charge profiles for the same mAb show that changes in the UV charge profile, measured by icIEF (panel A), are reflected in the MS charge profile (panel B). Additionally, all charge isoform peaks detected by icIEF have corresponding peaks in the MS profile. Corresponding charge profiles are mirror images because basic peaks are analyzed first by MS.

This integrated process delivered sufficient sensitivity to identify minute structural changes, such as incremental additions of sialic acid, by allowing correlation of the mass shift to the pI shift (Figure 2). With the Intabio ZT System, the team could also use the pI profile for real-time corroboration of MS characterization (Figure 3, Panel C). The researchers observed an 18 Da peak shift for the ADC-1 compared to the mAb intermediate. Using the icIEF-UV/MS workflow helped identify the formation of carboxylic acid from hydrolysis of chemistry in the payload, corresponding with the 18 Da shift. As a result, the team was able to disentangle the contribution of the change in payload structure from the charge variation that is due to antibody modifications (Figure 3). The ability to distinguish between antibody- and payload-related charge heterogeneity provides a new strategy for icIEF specification of ADCs.

Figure 2. The icIEF UV (Panel A) and MS (Panel B) charge profiles for this complex mAb show the intact mass electropherogram of corresponding icIEF peaks. Corresponding charge profiles are mirror images because basic peaks are analyzed first by MS. Panel C shows the intact mass profiles and their identification by charge variant, arranged from top to bottom by increasing pI value. The Basic 1 peak, with the highest pI value of 8.24, as shown in Panel A, consists of proline amidated species. The neutral charge isoform with a pI of 8.13 is composed of 0 to 3 neutral N-linked glycans attached per glycosylation site per molecule. The decrease in pI from Acidic Peak 1 (pI 8.00) to Acidic Peak 3 (pI 7.27) corresponds to the incremental additions of up to six terminal N-acetylneuraminic (sialic acids).

Using Combined icIEF-UV and MS in the Race to Market

As competition ramps up for delivering life-saving ADCs to the clinic, developers of these therapeutics need fast, robust techniques for establishing valid release methods. A single platform that combines icIEF within in-line MS may be the solution to decreasing drug development time and saving resources. Data from drug development teams using this workflow indicate that icIEF-UV/MS can accelerate timelines, especially for characterization of charge heterogeneity in complex protein therapeutics. A robust platform will avoid introducing icIEF artifacts, ensuring that samples subjected to MS are representative of the starting material.

Figure 3. Conjugation of the ADC results in a non-covalently bound light chain (LC). ADC dissociation during icIEF produces two charge envelopes (red trace = free LC; blue trace = heavy-heavy-light chain [HHL]). Ring opening of the payload is evidenced by 18 Da shifts (inset, top = LC; inset, bottom = HHL), resulting in a negative charge and, consequently, reduced pI. The inverse relationship between the pI of the ADC and the relative abundance of the +18 Da isoform shows preservation of noncovalent interactions between light chain and heavy-heavy-light chain subunits during upstream icIEF separation, indicating that the icIEF is not resulting in artifacts on the molecule.

About the Authors

Roxana McCloskey is the senior global marketing manager, biopharma at SCIEX. McCloskey has over 15 years of experience in sales and marketing roles in the MS community and specializes in communicating the advantages of MS and CE-based applications for biopharmaceutical drug development and workflows.

Zoe Zhang, PhD, is a senior manager, biopharma applications at SCIEX. Zhang currently leads the protein therapeutics and pharma group. She is leading her team, playing a pivotal role in driving collaboration with key customers and fostering strong relationships that lay the foundation for groundbreaking advancements. Her team actively supports the development of novel workflows, ensuring that the team stays at the forefront of cutting-edge methodologies. Further, her team also involves in next-generation instrumentation and software development. Overall, her team performs in-depth mass-spec and capillary electrophoresis characterization and quantitation across diverse modalities and next-generation therapies. Prior to joining SCIEX, Zoe distinguished herself as a lead scientist in LCMS protein characterization at Charles River Laboratory, working on comprehensive protein therapeutics analysis relevant to pharmaceutical processing.

REFERENCES

1. Zheng K, Chen Y, Wang J, et al. Characterization of Ring-Opening Reaction of Succinimide Linkers in ADCs. J Pharm Sci. 2019;108(1):133-141. doi:10.1016/j.xphs.2018.10.063
2. SCIEX. A deep dive into the characterization of an ADC using an integrated icIEF-UV/MS system. BioPharmInternational.com. June 21, 2023. Accessed October 24, 2023. https://www.biopharminternational.com/view/we-all-have-baggage-a-deep-dive-into-the-characterization-of-a-complex-cief-electropherogram-from-an-antibody-drug-conjugate-using-a-novel-integrated-icief-uv-ms-system