CMC Regulatory Considerations for Antibody-Drug Conjugates.

Antibody-drug conjugates unite the specificity and long circulation time of an antibody with the toxicity of a chemical cytostatic or otherwise active drug using appropriate chemical linkers to reduce systemic toxicity and increase therapeutic index. This combination of a large biological molecule and a small molecule creates an increase in complexity. Multiple production processes are required to produce the native antibody, the drug and the linker, followed by conjugation of afore mentioned entities to form the final antibody-drug conjugate. The connected processes further increase the number of points of control, resulting in necessity of additional specifications and intensified analytical characterization. By combining scientific understanding of the production processes with risk-based approaches, quality can be demonstrated at those points where control is required and redundant comparability studies, specifications or product characterization are avoided. Over the product development lifecycle, this will allow process qualification to focus on those areas critical to quality and prevent redundant studies. The structure of the module 3 common technical document for an ADC needs to reflect each of the production processes and the combined overall approach to quality. Historically, regulatory authorities have provided varied expectations on its structure. This paper provides an overview of essential information to be included and shows that multiple approaches work as long as adequate cross-referencing is included.

Streamlined Characterization of an Antibody-Drug Conjugate by Two-Dimensional and Four-Dimensional Liquid Chromatography/Mass Spectrometry.

This study describes the use of a multidimensional HPLC (2D and 4D) system for a faster and more effective characterization of an antibody-drug conjugate (ADC) product, compared to the standard off-line approach of fraction collection and off-line variant characterization. The size variants of an interchain cysteine-linked ADC were characterized to understand the effect of the different drug-to-antibody ratio (DAR) species on aggregate formation. For this purpose, the ADC product and a full panel of stressed samples were analyzed. The dimeric ADC species were baseline resolved from the main peak (Rs = 2.7) by UHP-SEC (ultra-high-performance size exclusion chromatography) under nondenaturing conditions using a buffered mobile phase containing 5% 2-propanol. A 2D-LC (SEC-HIC) method was then developed to compare the average DAR values of the main peak species vs the aggregates. A 4D-LC/MS method (SEC-reduction-digestion-RPHPLC) was also developed to determine levels of potential critical quality attributes (pCQAs) including aggregation, average DAR, oxidation, and deamidation, in a 2 h run. An average DAR value of 3.5-3.6 was found for the main peak using both 2D-LC and 4D-LC methods, and these values were consistent with DAR determined by the in-house reference hydrophobic interaction chromatography (HIC) method. The multidimensional LC approaches also showed an increase in the content of high-DAR species in the SEC fractions containing the aggregates. Overall the entire workflow of data acquisition is completed within a day using the multidimensional on-line approach, in comparison to multiple days required with the traditional off-line approaches.

Statistical characterization of therapeutic protein modifications.

Peptide mapping with liquid chromatography-tandem mass spectrometry (LC-MS/MS) is an important analytical method for characterization of post-translational and chemical modifications in therapeutic proteins. Despite its importance, there is currently no consensus on the statistical analysis of the resulting data. In this manuscript, we distinguish three statistical goals for therapeutic protein characterization: (1) estimation of site occupancy of modifications in one condition, (2) detection of differential site occupancy between conditions, and (3) estimation of combined site occupancy across multiple modification sites. We propose an approach, which addresses these goals in terms of summarizing the quantitative information from the mass spectra, statistical modeling, and model-based analysis of LC-MS/MS data. We illustrate the approach using an LC-MS/MS experiment from an antibody-drug conjugate and its monoclonal antibody intermediate. The performance was compared to a 'naïve' data analysis approach, by using computer simulation, evaluation of differential site occupancy in positive and negative controls, and comparisons of estimated site occupancy with orthogonal experimental measurements of N-linked glycoforms and total oxidation. The results demonstrated the importance of replicated studies of protein characterization, and of appropriate statistical modeling, for reproducible, accurate and efficient site occupancy estimation and differential analysis.

Structural Characterization of Cross-Linked Species in Trastuzumab Emtansine (Kadcyla).

The antibody-drug conjugate, trastuzumab emtansine (Kadcyla), is produced by attachment of the antitubulin drug, DM1, to lysine amines via a heterobifunctional linker, SMCC (succinimidyl 4-(N-maleimidomethyl)cyclohexane-1-carboxylate). Following the reaction of the N-hydroxysuccinimide activated linker with antibody lysines to produce a linker-modified intermediate (Tmab-MCC), DM1 is added to yield the desired product. In addition to the expected distribution of drug-linked forms (from 0 to 8), mass spectrometry also demonstrates the presence of a second distribution shifted by about +222 Da. This series is consistent with the presence of a population containing a bound linker without DM1 ("unconjugated linker"). Extended characterization of trastuzumab emtansine was performed using capillary isoelectic focusing, CE-SDS, peptide mapping, and LC/MS following (18)O labeling of peptide digests to identify this family of product variants. These studies demonstrate that the presence of these +222 Da species is due to an unexpected reaction of the maleimide moiety in the MCC linker with antibody lysine residues to produce cross-linked species that cannot conjugate to DM1.

Investigation of low recovery in the free drug assay for antibody drug conjugates by size exclusion-reversed phase two dimensional-liquid chromatography.

Antibody drug conjugates (ADCs) are complex therapeutic agents combining the selectivity of monoclonal antibodies and highly efficacious small molecule drugs to successfully eliminate tumor cells while limiting the general toxicity and side effects of the therapeutic to protect patient safety. One unique attribute critical to the safety of ADCs is the residual content of unconjugated small molecule drug present from either incomplete conjugation or degradation of the ADC. Typically for quality control assays, quantifying the amount of the free drug is performed through precipitation of the protein species using an organic solvent and then assaying the amount of free drug left in the supernatant. During the validation of an assay of this type for a maleimide based linker drug, issues were experienced with low and variable recovery in the spiked samples of the drug substance and drug product. A two-dimensional heart-cutting method coupling Size Exclusion Chromatography (SEC) with Reverse Phase (RP) chromatography was utilized to explore possible mechanisms leading to the low recovery of the free linker drug. The results of the investigation indicated that the spiked linker drug reacts with residual reactive groups on the ADC; a conclusion which was confirmed by the observed increase of average Drug to Antibody Ratio (DAR) determined by Hydrophobic Interaction Chromatography (HIC). Finally, several approaches were evaluated to minimize the recovery loss. Capping the residual reactive groups on the ADC with maleimide containing reagents effectively mitigated the low recovery issue.

Statistical modeling of the drug load distribution on trastuzumab emtansine (Kadcyla), a lysine-linked antibody drug conjugate.

Trastuzumab emtansine (Kadcyla) is a recently approved antibody-drug conjugate produced by attachment of the anti-tubulin drug, DM1, to lysine amines via the SMCC linker. The resulting product exhibits a drug load distribution from 0 to 8 drugs per antibody that can be quantified using mass spectrometry. Different statistical models were tested against the experimental data derived from samples produced during process characterization studies to determine best fit. The Poisson distribution gives the best correlation for samples manufactured using the target process conditions (yielding the target average drug to antibody ratio (DAR) of 3.5) as well as those produced under conditions that exceed the allowed manufacturing ranges and yield products with average DAR values that are significantly different from the target (i.e., ≤3.0 or ≥4.0). The Poisson distribution establishes a link between average DAR values and drug load distributions, implying that measurement and control of the former (i.e., via a simple UV spectrophotometric method) could be used to indirectly control the latter in trastuzumab emtansine.

Phase I study of trastuzumab-DM1, an HER2 antibody-drug conjugate, given every 3 weeks to patients with HER2-positive metastatic breast cancer.

PURPOSE: Trastuzumab-DM1 (T-DM1) is an antibody-drug conjugate that uses trastuzumab to specifically deliver the maytansinoid antimicrotubule agent DM1 to HER2-positive cells. This first-in-human study of T-DM1 evaluated safety, pharmacokinetics, and preliminary activity of T-DM1 in patients with advanced HER2-positive breast cancer. PATIENTS AND METHODS: Successive cohorts of patients who had progressed on trastuzumab-based therapy received escalating doses of T-DM1. Outcomes were assessed by standard solid-tumor phase I methods. RESULTS: Twenty-four patients who had received a median of four prior chemotherapeutic agents for metastatic disease received T-DM1 at 0.3 mg/kg to 4.8 mg/kg on an every-3-weeks schedule. Transient thrombocytopenia was dose-limiting at 4.8 mg/kg; the maximum-tolerated dose (MTD) was 3.6 mg/kg. The half-life of T-DM1 at the MTD was 3.5 days, with peak DM1 levels < 10 ng/mL. Clearance at doses < 1.2 mg/kg was faster than at higher doses. Common drug-related adverse events (AEs) included grade < or = 2 thrombocytopenia, elevated transaminases, fatigue, nausea, and anemia. No grade > 1 nausea, vomiting, alopecia, or neuropathy events and no cardiac effects requiring dose modification were reported. The clinical benefit rate (objective response plus stable disease at 6 months) among 15 patients treated at the MTD was 73%, including five objective responses. The confirmed response rate in patients with measurable disease at the MTD (n = 9) was 44%. CONCLUSION: At the MTD of 3.6 mg/kg every 3 weeks, T-DM1 was associated with mild, reversible toxicity and substantial clinical activity in a heavily pretreated population. Phase II and III trials in patients with advanced HER2-positive breast cancer are under way.

Charge-based analysis of antibodies with engineered cysteines: from multiple peaks to a single main peak.

THIOMABs are antibodies with an engineered unpaired cysteine residue on each heavy chain that can be used as intermediates to generate antibody-drug conjugates. Multiple charge variant peaks were observed during cation-exchange chromatography (CEX) and imaged capillary isoelectric focusing (cIEF) analysis of several different THIOMABs. This charge heterogeneity was due to cysteinylation and/or glutathionylation at the engineered and unpaired cysteines through disulfide bonds formed during the cell culture process. Cysteine treatment followed by analysis using CEX, LC/MS and electrophoresis demonstrates that cysteine is a mild reductant that can remove glutathione and cysteine bound to the engineered cysteines without disrupting the inter- or intra-chain disulfide bonds of antibodies. We further demonstrated that using a cysteine/cystine redox pair (rather than cysteine alone) can not only effectively remove glutathione at the engineered cysteines, but also generate homogeneously cysteinylated species, which resulted in one main peak in both CEX-HPLC and imaged cIEF assays for antibodies with engineered and unpaired cysteines.

Anti-CD22-MCC-DM1 and MC-MMAF conjugates: impact of assay format on pharmacokinetic parameters determination.

CD22 represents a promising target for antibody-drug conjugate therapy in the context of B cell malignancies since it rapidly internalizes, importing specifically bound antibodies with it. To determine the pharmacokinetic parameters of anti-CD22-MCC-DM1 and MC-MMAF conjugates, various approaches to quantifying total and conjugated antibody were investigated. Although the total antibody assay formats gave similar results for both conjugates, the mouse pharmacokinetic profile for the anti-CD22-MCC-DM1 and MC-MMAF appeared significantly different depending on the conjugated antibody assay format. Since these differences significantly impacted the PK parameters determination, we investigated the effect of the drug/antibody ratio on the total and conjugated antibody quantification using multiple assay formats. Our investigations revealed the limitations of some assay formats to quantify anti-CD22-MCC-DM1 and MC-MMAF with different drug load and in the context of a heterogeneous ADC population highlight the need to carefully plan the assay strategy for the total and conjugated antibody quantification in order to accurately determine the ADC PK parameters.

Protein aggregation and bioprocessing.

Protein aggregation is a common issue encountered during manufacture of biotherapeutics. It is possible to influence the amount of aggregate produced during the cell culture and purification process by carefully controlling the environment (eg, media components) and implementing appropriate strategies to minimize the extent of aggregation. Steps to remove aggregates have been successfully used at a manufacturing scale. Care should be taken when developing a process to monitor the compatibility of the equipment and process with the protein to ensure that potential aggregation is minimized.