The Trojan ADC Challenge
Often touted as ‘magic bullets’ against cancer, antibody-drug conjugates (ADCs) aim to combine the effectiveness of chemotherapy with the specificity of antibodies. But the same complexity that imbues them with desirable qualities adds many hurdles to development and manufacturing processes. Fortunately, solutions – and expertise – are on the rise.
Chemotherapeutics are very effective at killing rapidly dividing tumor cells, but there is a major drawback: they lack specificity and also kill other cells in the body. Over the last decade, targeted anticancer treatments have been developed, including several recombinant monoclonal antibodies (mAbs). The specific binding properties of antibodies allow them to differentiate between cancer cells and healthy cells; however, they are rarely curative in anticancer therapy. Since the approval of the first mAb (Orthoclone) in 1986, only 18 naked mAbs have reached the oncology market. Many fail clinical testing due to lack of efficacy (1).
It is possible to enhance the functionality of mAbs by coupling diverse moieties to the antibody. One subclass of antibody-related therapeutics that is seeing increased interest for oncology applications is antibody-drug conjugates (ADCs). An ADC is a mAb that has been covalently linked to cytotoxic agents, combining the potency of chemotherapy with the specificity of antibodies. The concept is simple: the mAb delivers the cytotoxic payload to the correct location. After being taken up by tumor cells, the drug is released intracellularly, killing the cell in a targeted and effective way.
Next-generation ADCs are further enhancing this concept; for example, by increasing the homogeneity and broadening the linkers that can be used. Efforts are also being made to improve the ADC payload by having the cytotoxin present as an inactive prodrug in the intact ADC molecule. Through internalization of the ADC, the drug enters the cell in its inactive shape and only becomes activated after intracellular processing in the endosomal pathway. It can be thought of as a Trojan horse.
The concept of targeted therapy, selectively delivering a cytotoxic drug to a tumor via a targeting agent was postulated by Paul Ehrlich more than 100 years ago, but it’s only fairly recently that they have become valuable therapeutic agents – mainly because of recent advances in linker, drug and antibody technologies. The use of higher drug potency, more stable linkers to prevent early release of the toxin in the blood stream, better control of the amount of toxin per antibody and more selective antibodies lead to more successful treatment (2).
Despite their promise and potential in the fight against cancer, few ADCs have been approved. The first ADC approved by the FDA was Mylotarg (gemtuzumab ozogamicin) in 2000, but was withdrawn in 2010 after failing a post approval study. It is, however, still available in Japan. At the moment, only two ADCs are marketed in both the US and Europe – Adcetris (brentuximab vedotin) for Hodgkin lymphoma and anaplastic large cell lymphoma, and Kadcyla (ado-trastuzumab emtansine) for breast cancer (the latter has caused some controversy, particularly in the UK, because of its high cost of around £90,000 per patient). But the number of ADCs in clinical trials is growing.
The combined challenges of ADCs
The low number of ADC approvals is testament to their associated development and manufacturing challenges. It’s common knowledge that biopharmaceuticals present manufacturing and characterization challenges – and naked mAbs are no exception. MAbs are produced using mammalian cell culture in stainless steel vessels or disposable bags. The product is excreted by the cells and, after removal of the cells, purified to a more than 99-percent pure product – usually using chromatographic techniques. Although mAbs are not the largest molecules produced using recombinant technologies, they have a significant molecular mass (around 145 kDa) and also have complex glycan structures and other post-translational modifications that create additional characterization challenges, demanding a broad spectrum of analytical techniques. Although the manufacture of mAbs is a complex activity, the whole process, including aseptic techniques and purification steps, is relatively well understood.
ADCs are trickier because they are biological products that require chemical transformation. The ADCs currently on the market or in clinical trials are predominantly based on two drug classes. The first comprises auristatins and maytansinoids; both are tubulin binders that block the cell in its progression through mitosis. As a result, only rapidly-dividing cells are attacked. The second drug class is formed by DNA-alkylating drugs, such as duocarmycins, which induce cell death in both dividing and non-dividing cells (3). These drugs are far more cytotoxic than standard chemotherapy methods, with potencies in the picomolar range, and require chemical facilities that are equipped for manufacturing highly potent toxins.
Top Challenges of ADC Manufacture
- Use of highly potent cytotoxic compounds requires additional safety and environmental precautions, demanding complex facilities
- Limited availability of CMOs that have sufficient experience with both cytotoxic compounds, and proteins
Complex manufacturing process, involving linker, cytotoxic and mAb components and a conjugation process - Analytical complexity – all components require characterization, including the linker, which makes up a very small part of the molecule
- Complex supply chain, involving multiple suppliers
- Linker-drug technologies are limited in number
To form an ADC, you need the mAb, a linker and a highly potent cytotoxic drug, followed by the conjugation of all three components to form the final drug substance (DS). Although the mAb and the linker-drug are considered intermediates, they generally have to meet the same level of specifications as if they were a separate DS. As a result, the ADC is considered a biological entity, but both ICH Q6A (Specifications: Test Procedures And Acceptance Criteria For New Drug Substances And New Drug Products: Chemical Substances) for the linker and drug, and Q6B (Specifications Test Procedures And Acceptance Criteria For Biotechnological/Biological Products) for the mAb and ADC apply. The DS also needs further sterile filtration, filling, and often lyophilization, to obtain the final drug product for intravenous treatment.
When manufacturing mAbs, Grade D and C cleanroom facilities are needed for cell culture and purification, respectively. A pressure regime is applied to prevent cross-contamination where, in general, a positive pressure to the outside environment is applied to keep any particles and bugs out of the manufacturing areas. For the cytotoxic components used in ADCs, however, the same GMP regulation is applicable, but exposure of highly potent material to the environment needs to be prevented, leading to a negative pressure in the manufacturing area. This dichotomy creates additional challenges to meet GMP in relation to cleanroom contamination – and these should be addressed during the initial facility design.
It is not only exposure to air that must be considered; exposure via wastewater streams and other waste must also be prevented. Cleaning activities often create significant amounts of rinse volumes, but can be eliminated with single-use materials (disposable reactor vessels, filters, chromatographic fluid paths, and so on). Unfortunately, as solvents are applied during the conjugation process, the use of single-use materials could become an issue in relation to leachables. You can also help eliminate the risk for environmental exposure by using a leak-proof floor and the removal of all drains. All waste materials must be packed in closed containers and incinerated, or taken for validated chemical inactivation.
Another important aspect when manufacturing ADCs is characterization and release. The linker drug represents only about 1 percent of the total mass of the ADC, but is the main driver for its potency. Indeed, it can have a significant impact on the behavior of the mAb as some drugs are very hydrophobic, which could result in increased aggregation and increased plasma clearance (4) – a problem that becomes more relevant at higher drug-to-antibody ratios (DAR) as higher DAR will result in higher hydrophobicity of the ADC. Moreover, multiple binding sites for the linker drug results in heterogeneity of the molecule and additional complexity for characterization. DAR is a critical quality attribute and is quantified using UV for lysine conjugation and hydrophobic interaction chromatography or reversed-phase high-performance liquid chromatography for cysteine conjugation. Drug load distribution, the levels of unconjugated mAb and free drug, the charge heterogeneity, aggregation, higher order structures, and potency must all be tested (5). Any structural changes of the mAb or the linker drug caused by the conjugation process demand additional scrutiny.
Overcoming the hurdles
As previously discussed, ADCs require mAbs, linker drug and conjugation manufacturing activities – and these all need their own facilities and technologies. It is not unusual for four different contract manufacturing organizations (CMOs) to be used – a logistical nightmare! In-house manufacturing for all these aspects solves issues relating to supply chain and provides much better control over quality, cost and timelines, but is only appropriate in the case of a robust pipeline since it requires significant investments and expertise. Most companies therefore go down the CMO route.
Many CMOs can deal with either cytotoxic compounds or proteins, but many lack the experience to work with both at the same time, or may lack the required infrastructure and ability to produce ADCs in suitable quantities (multi-kilogram scale). A variety of different linkers and payloads are used in ADCs, which all require sophisticated know-how. Multiple technologies also exist for conjugation. Initially, naturally-occurring lysines or cysteines liberated from interchain disulfides were used for conjugating the linker drug to the mAb. To facilitate manufacturing and create a more homogeneous product, so-called site-directed conjugation is being applied. Site-directed conjugation introduces engineered cysteines or non-natural amino acids into the mAb for use as linker sites, which also has an impact on the design of cell lines used for manufacturing ADCs.
Having so much know-how under one roof is no easy feat, particularly given that the ADC field is relatively new, so when seeking contract manufacturing partners it’s wise to look closely at both the manufacturing facilities and expertise on offer. You also need to look at cleaning validation to prevent cross contamination. If the CMO addresses contamination issues by using dedicated equipment, then significant upfront cost and time could be required to qualify the equipment. If single-use equipment is used, then extractable and leachable studies performed by the CMO are also useful. There are also several other questions you need to ask. Is the CMO really capable of developing the conjugation process from scratch, or would you be better off using in-house development and then subsequently transferring the process to the CMO? How well are environmental, health and safety aspects under control? And, of course, you must take into account all standard GMP aspects, as well as cultural fit, financials, conditions and CMO reliability.
To date, we’ve seen few commercial ADCs, but with increased attention, and more research and technologies pouring into the field, including better understanding of their modes of action, it’s likely that we’ll be seeing new linker designs and more drugs reaching the market in the near future. Overcoming the manufacturing hurdles, however, should not be underestimated. It takes time and expertise, and timely strategic decisions to successfully develop and manufacture these highly potent biologicals.
- L. Iamele, L. Vecchia and C. Scotti, Antibody-Drug Conjugates: Targeted Weapons Against Cancer. Antibody Technology Journal, 5, 1-13 (2015).
- H.L. Perez et al., “Antibody–Drug Conjugates: Current Status and Future Directions,” Drug Discovery Today 19(7), 869-881 (2014).
- W. Dokter et al., “Preclinical Profile of the HER2-Targeting ADC SYD983/SYD985: Introduction of a New Duocarmycin-Based Linker-Drug Platform,” Mol. Cancer Ther. 13(11), 1–12 (2014).
- R.P. Lyon et al., “Reducing Hydrophobicity of Homogeneous Antibody-Drug Conjugates Improves Pharmacokinteics and Therapeutic Index,” Nat. Biotechnol. 33(7), 733-736 (2015).
- A. Wakankar et al., Analytical Methods for Physicochemical Characterization of Antibody Drug Conjugates,” mAbs 3(2), 161-172 (2011).
Aad van de Leur has been working at Synthon Biopharmaceuticals BV, Nijmegen, the Netherlands, since January 2009. In his role as chief operations officer, he is responsible for all biopharma operational activities, including process development activities from cell line development to formulation and related analytical development, as well as manufacture and supply of clinical material.
