Making MAbs: Bioprocess Advancements Challenge Platform Assumptions

https://stock.adobe.com

Still the largest sector of the industry, monoclonal antibodies (MAbs) have dominated the biopharmaceutical stage for over 30 years. Some observers might think there’s nothing new to say about these molecules; others point to antibody derivatives as a more exciting alternative. But MAbs are far from an outdated technology. From biosimilar developments to cell-free synthesis to yeast display, immunogenicity improvements, and fully human antibodies — as well as improvements in process efficiency and cost reductions, as discussed herein — the MAb future not only is bright, but also could be amazing.

Our fellow Informa publication, the journal mAbs, started out this year with a cost-benchmarking study from authors at University College London and AstraZeneca (1). Farid et al. analyzed the contribution of process development and manufacturing to overall research and development (R&D) costs. The authors’ model captured cost, time, and risk and took into account the interdependencies among clinical, process development, and manufacturing activities to estimate required budgets needed for each phase of development. Focusing on MAbs in particular, the collaborators sought to elucidate how different clinical success rates affect the overall costs of process development and manufacturing at each stage. Based on a moderate overall clinical success rate of ~12%, they concluded that a biopharmaceutical company needs to budget about US$60 million for preclinical to phase 2 material preparation and about $70 million for manufacturing materials for phase 3 and regulatory review. Lower overall clinical success rates (~4%) characterize treatments for conditions such as Alzheimer’s disease — and according to the benchmarking model, that lack of success significantly raises costs to ~$190 million for early phase and ~$140 million for late-phase material preparation. However, the authors estimated that process development and manufacturing costs were just 13–17% of the overall R&D budget for each market success.

In the same mAbs issue, authors from Bristol-Myers Squibb (BMS) suggested process intensification as a means of both reducing those costs and increasing biomanufacturing productivity (2). They described a case study transforming a conventional 1,000-L MAb manufacturing process to intensified processing schemes at both 1,000-L and 2,000-L scales, all based on a familiar Chinese hamster ovary (CHO) cell line. For upstream production, BMS implemented an intensified scheme to shorten the seed train by use of enriched seed-culture media and perfusion technology. That yielded dense inoculant for a fed-batch production bioreactor, which substantially increased expression titers of comparable-quality product by 400–800%. The downstream team made multiple changes to accommodate those increased titers (2):

New high-capacity resins were implemented for the protein A and anion-exchange chromatography (AEX) steps, and the cation-exchange chromatography (CEX) step was changed from bind–elute to flow-through mode. . . Multicolumn chromatography was developed for protein A capture, and integrated AEX–CEX poolless polishing steps allowed semicontinuous [processing] with increased productivity as well as reductions in resin requirements, buffer consumption, and processing times. A cost-of-goods analysis on consumables showed 6.7–10.1 fold cost reduction from the conventional Process A to the intensified Process C.

As the authors surmised, such process intensification approaches could provide an intermediate step toward developing fully continuous manufacturing with even higher productivities in the future. Incrementally advancing technologies for analysis, production, and processing may not provide a truly “disruptive” revolution in MAb development and manufacturing, but they do indicate where innovation can make a difference in the near future.

Upstream Production
Advancements in cell-line engineering and cell culture process optimizations since 2000 have made dramatic improvements in the productivity of mammalian cell lines for producing antibodies. We watched expression titers go from measurements in milligrams per liter of culture to multiple grams per liter, drastically improving production efficiencies and concomitantly driving down manufacturing costs and facility size requirements. Using similar strategies, biopharmaceutical companies now are focusing on ways to speed up process development and improve product quality while increasing productivities even further.

Cell-Line Engineering: Getting MAb product candidates into clinical testing as soon as possible is a goal for all developers, especially those making medicines that would address unmet medical needs. Cell-line development based on single-cell cloning is time consuming, especially because lead clones must be screened for cell-line stability and suitability for manufacturing operations. To ensure safety and minimize risk, materials for both preclinical (toxicology) and early clinical (first-in-human) studies typically have been generated using the same clonal-derived cells. To shorten development timelines, a team at BMS used pools of up to six clones for early production of drug substance for regulatory-filing–enabling toxicology studies, then selected a final single clone for production of clinical materials (3). The company uses platform processes across all stages of early development: expression vectors, host-cell lines, media, and production processes. Comprehensive analyses demonstrated that the toxicology and clinical materials were comparable across six MAb programs — a vital aspect of such speed-to-clinic efforts.

Other groups are studying ways to engineer cells to express antibodies with improved functional characteristics. For example, removing fucose sugar from glycosylation sites can improve both therapeutic efficacy, complement-dependent cytotoxicity (CDC), and antibody-dependent cell-mediated cytotoxicity (ADCC). A MedImmune team developed a platform for producing afucosylated MAbs by engineering CHO cells to coexpress them with an enzyme that blocks fucose synthesis (4). Using GlymaxX technology from ProBioGen, the team generated a high-yielding cell line that secreted afucosylated MAbs at ~6.0 g/L and maintained such production over 60 generations. Once purified, those MAbs demonstrated increased binding in a cellular assay, with enhanced ADCC compared with a fucosylated control. “Furthermore,” the authors concluded, “afucosylation in these MAbs could be controlled by simple addition of l-fucose in the culture medium, thereby allowing the use of a single cell line for production of the same MAb in fucosylated and afucosylated formats for multiple therapeutic indications.”

A group at Roche Innovation Center Munich in Germany addressed the same issue with a method to engineer CHO cell lines coexpressing two other enzymes that interfere with antibody fucosylation (5). Over 60% of the resulting MAbs were afucosylated and exhibited enhanced Fc-mediated effector function, as determined with a novel FcgRIIIa-affinity chromatography method. The team tested different cultivation conditions to identify the trace element manganese as a key component essential for production of afucosylated antibodies.

Amino-acid sequence variation requires close monitoring during production process development. A cross-functional team at Pfizer worked together to create a practical, reliable testing strategy for sequence variants without requiring additional project resources or time (6). They combined next-generation sequencing (NGS) with amino-acid analysis (AAA) to identify mutated mammalian cell clones and cell culture process media/feed conditions that induce misincorporations. This replaced mass spectrometric (MS) techniques with faster, equally informative, and more efficient approaches — freeing up MS resources for other purposes (e.g., characterization of products from commercial cell lines and culture processes). “Once an industry-wide challenge,” the team wrote, “sequence variation is now routinely monitored and controlled at Pfizer (and other biopharmaceutical companies) through increased awareness, dedicated cross-line efforts, smart comprehensive strategies, and advances in instrumentation/software, resulting in even higher product quality standards for biopharmaceutical products.”

Cell Culture Optimization: During upstream process development, improvements in expression titer can reduce drug manufacturing costs. While doing so, however, ensuring patient safety and efficacy by delivering comparable quality attributes of therapeutic proteins is important. Having found that relatively high concentrations of iron in culture media improved expression but caused unacceptable oxidation/discoloration of a MAb fusion protein, process engineers at BMS used an alternative long-term passaging approach in low-iron media (7). Because that increased charged variance as well as expression titer, the team evaluated other options to discover that removing β-glycerol phosphate from the media eliminated basic variants without lowering titer. The process scaled well to 500-L bioreactors.

Protein production in cellular systems inherently generates molecular variants that must be controlled. Monitoring the presence and characteristics of MAb variants during production can enable cell culture engineers to “tune” process parameters that will maintain optimal conditions. An industry–academic partnership including both a technology supplier and a biosimilar developer in Europe has published an efficient workflow for that purpose (8). With minimal sample preparation (centrifugation) and 15-minute analysis by high-performance liquid chromatography with MS (HPLC-MS), the collaborators were able to identify glycosylation and truncation variants of intact proteins as well as absolute quantification of expression titer. Enzymatically removing N-glycans then allowed for determination of glycation levels. The strategy could enable companies to monitor protein expression continuously for enhanced in-process control purposes.

Another industry–academic collaboration led by Bristol-Myers Squibb scientists used a process intensification approach to increase the viable cell density (VCD) of a fed-batch production bioreactor inoculation as a means of improving manufacturing output by shortening the necessary duration of the production culture (9). A typical starting VCD is ~0.5 × 10⁶ cells/mL; using enriched basal medium, this team was able to increase that to 3–6 × 10⁶ cells/mL without the increasing complexity and media requirements of perfusion. “Although only three stable MAbs produced by CHO cell cultures are used in this study,” the team wrote, “the basic principles of the non-perfusion n – 1 seed strategies for shortening seed train and production culture duration or improving titer should be applicable to other protein production by different mammalian cells and other hosts at any scale biologics facilities.”

Downstream Processing
Whenever you think about separation and purification of antibody products from cell culture supernatant, inevitably protein A affinity chromatography comes to mind. For decades, it has served as the cornerstone of a solid downstream processing platform for nearly all MAb products on the market. That status is well deserved based on the powerful affinity of the ligand for the Fc-region heavy chain, providing for as much as 99% yield after harvest clarification and capture. Generally, that protein A capture step is followed by “polishing” operations based on other chromatographic chemistries such as ion exchange and/or hydrophobic interaction to remove charge variants, host-cell proteins and DNA, and other potential contaminants that remain in trace but not insignificant amounts. Intermediate steps for viral removal/inactivation and further refinements will involve membrane technologies (e.g., tangential-flow, ultrafiltration, diafiltration) and solvent/detergent or temperature treatments.

The high cost of protein A chromatography media has made some process development groups long for an alternative MAb purification platform. As cited below, recent publications highlight extraction methods, activated carbon, multimodal chromatography, and other affinities as potentially disruptive technologies. Each option has its own strengths and limitations, and no single one yet has risen above the rest.
Researchers at the Swedish Royal Institute of Technology developed a purification matrix based on a single domain from protein A that displays calcium-dependent antibody binding (10). This ligand is intended to provide an alternative to the acidic elution conditions required for protein A affinity chromatography for purification of pH-sensitive antibodies and other Fc-based molecules. With elution at neutral pH, this “Z_Ca” affinity method provides, as the team concludes, “effective and mild purification of antibodies and Fc-fusion proteins that cannot be purified under conventional acidic elution conditions due to aggregation formation or loss of function.”

Scientists at Astellas Pharma and Merck/MilliporeSigma in Japan have pushed the envelope farther from the familiar to propose an integrated platform for MAb purification based on activated carbon and flow-through cation- and anion-exchange chromatography steps (11). They reported robust clearance of impurities at high loadings, with an overall MAb yield >80%. Host-cell proteins and DNA were removed by activated carbon, and a new flow-through cation-exchange resin removed MAb aggregates.

An industry–academic collaboration in Israel steps even farther outside the column to eschew ligands and chromatography entirely (12). Their technology uses detergent aggregation to capture antibodies while rejecting hydrophilic impurities. Then those antibodies are extracted in pure form without dissolving the aggregates, thus leaving hydrophobic impurities behind. This “engineered micelles” approach combines Tween 20 nonionic detergent with bathophenanthroline (a hydrophobic metal chelator from MilliporeSigma) and positive iron ions. It removes human and mouse IgGs from bovine serum albumin in serum-free media for a 70–80% overall yield.

Phase separation is another technique of interest. However, it’s more likely to be a problem in downstream processing than a purification strategy itself. An AstraZeneca team reported on just such a case study in which electrostatically driven self-associations between the light chains of one MAb in development caused unintended liquid–liquid phase separation (LLPS) (13). Process parameter optimization efforts that worked at small scale did not solve the problem for large-scale manufacturing. In silico modeling aided in identifying problematic charged residues in the light chain’s complementarity-determining regions. Molecular engineering substitution of those with neutral or oppositely charged amino acids solved the problem without compromising the MAb’s antigen-binding affinity. “This study demonstrates the critical nature of surface charged resides on LLPS,” the authors wrote, “and highlights the applied power of in silico protein design when applied to improving physiochemical characteristics of therapeutic antibodies. Our study indicates that in silico design and effective protein engineering may be useful in the development of MAbs that encounter similar LLPS issues.”

MAb Trends and Technologies
Every year the mAbs journal publishes an “Antibodies to Watch” report for the coming 12 months, and 2020 was no different — at least until February or March came around. The series has documented the global biopharmaceutical industry’s efforts to bring innovative antibody therapeutics to patients in need, focusing specifically on unique antibodies in late-stage clinical studies or regulatory review and those approved for marketing as the previous year came to a close. And as the authors of this year’s installment put it (14),

Then came SARS-CoV-2. The coronavirus pandemic has challenged a typically careful and deliberate approach to MAb development and disrupted supply chains for development and manufacturing — not only for those antibodies deemed most promising a year ago, but also others that have been thrust into the spotlight as potential COVID-19 treatments. Authors in this featured report highlight some of those candidates as well as others with antibody–drug conjugate potential against a “new” class of targets — and elucidate the current state of manufacturing capacity for the MAb industry. Together, these articles provide a much-needed update on the MAb landscape as this very anomalous year begins to draw to a close with all too few answers to the many questions that have arisen over the past several months.

References
1 Farid SS, et al. Benchmarking Biopharmaceutical Process Development and Manufacturing Cost Contributions to R&D. mAbs 12(1) 2020: 1754999; https://doi.org/10.1080/19420862.2020.1754999.

2 Xu J, et al. Biomanufacturing Evolution from Conventional to Intensified Processes for Productivity Improvement: A Case Study. mAbs 12(1) 2020: 1770669; https://doi.org/10.1080/19420862.2020.1770669.

3 Bolisetty P, et al. Enabling Speed to Clinic for Monoclonal Antibody Programs Using a Pool of Clones for IND-Enabling Toxicity Studies. mAbs 12(1) 2020: 1763727; https://doi.org/10.1080/19420862.2020.1763727.

4 Roy G, et al. A Novel Bicistronic Gene Design Couples Stable Cell Line Selection with a Fucose Switch in a Designer CHO Host to Produce Native and Afucosylated Glycoform Antibodies. mAbs 10(3) 2018: 416–430; https://doi.org/10.1080/19420862.2018.1433975.

5 Popp O, et al. Development of a Pre-Glycoengineered CHO-K1 Host Cell Line for the Expression of Antibodies with Enhanced Fc Mediated Effector Function. mAbs 10(2) 2018: 290–303; https://doi.org/10.1080/19420862.2017.1405203.

6 Lin TJ, et al. Evolution of a Comprehensive, Orthogonal Approach to Sequence Variant Analysis for Biotherapeutics. mAbs 11(1) 2019: 1–12; https://doi.org/10.1080/19420862.2018.1531965.

7 Xu J, et al. Improving Titer While Maintaining Quality of Final Formulated Drug Substance Via Optimization of CHO Cell Culture Conditions in Low-Iron Chemically Defined Media. mAbs 10(3) 2018: 488–499; https://doi.org/10.1080/19420862.2018.1433978.

8 Regl C, et al. Dilute-and-Shoot Analysis of Therapeutic Monoclonal Antibody Variants in Fermentation Broth: A Method Capability Study. mAbs 11(3) 2019: 569–582; https://doi.org/10.1080/19420862.2018.1563034.

9 Yongky A, et al. Process Intensification in Fed-Batch Production Bioreactors Using Non-Perfusion Seed Cultures. mAbs 11(8) 2019: 1502–1514; https://doi.org/10.1080/19420862.2019.1652075.

10 Scheffel J, et al. Optimization of a Calcium-Dependent Protein A-Derived Domain for Mild Antibody Purification. mAbs 11(8) 1492–1501; https://doi.org/10.1080/19420862.2019.1662690.

11 Ichihara T, et al. Integrated Flow-Through Purification for Therapeutic Monoclonal Antibodies Processing. mAbs 10(2) 2018: 325–334; https://doi.org/10.1080/19420862.2017.1417717.

12 Dhandapani G, et al. A General Platform for Antibody Purification Utilizing Engineered-Micelles. mAbs 11(3) 2019: 583–592; https://doi.org/10.1080/19420862.2019.1565749.

13 Du Q, et al. Process Optimization and Protein Engineering Mitigated Manufacturing Challenges of a Monoclonal Antibody with Liquid–Liquid Phase Separation Issue By Disrupting Intermolecule Electrostatic Interactions. mAbs 11(4) 2019: 789–802; https://doi.org/10.1080/19420862.2019.1599634.

14 Kaplon H, et al. Antibodies to Watch in 2020. mAbs 12(1) 2020: 1703531; https://doi.org/10.1080/19420862.2019.1703531.

Cheryl Scott is cofounder and senior technical editor of BioProcess International, part of Informa Connect, PO Box 70, Dexter, OR 97431; 1-646-957-8879; [email protected].