Making a difference: manufacturing antibody-drug conjugates
Michal Wlodarski, Camin examines the manufacturing challenge of antibody-drug conjugates
Traditionally afflictions such as cancer, autoimmune diseases and inflammatory disorders have been treated with small molecule compounds that lack cell-targeting and thus have profound systemic implications. A well-known example is a chemotherapeutic agent. These compounds often have detrimental non-specific effects which, together with their prolonged multi-dose administration and systemic exposure, yield severe side-effects. In addition, such drugs often have very low therapeutic indexes where the window between the drug’s effective and toxic dose is narrow, making the optimal dosing a significant challenge for the clinicians, and ultimately deteriorating treatment outcomes and patient’s wellbeing.
Antibody-drug conjugates (ADCs) combine the functionality of biologic preparations with small drug molecule therapeutics to optimise treatment efficacy and thus improve outcomes and eliminate many of the problems associated with non-specific treatments outlined above. The key to this unique functionality is in the antibody-drug complex structure. An ADC is a multipart molecule consisting of a monoclonal antibody fused with a small molecule compound via suitable linker and spacer molecules. The antibody contains tumour-specific epitopes driving the API to the tumour tissue and away from the healthy cells. Upon binding with tumour antigens, the antibody-drug complex is internalised by the cancer cell where the antibody disintegrates and the drug molecules interfere with essential cellular processes to halt the cancerous growth.
Treatment revolution
ADCs have already made a positive impact within the oncology sphere and are soon to revolutionise autoimmune and inflammatory disorders treatments. Adcetris (brentuximab vedotin) marketed by Seattle Genetics in 2011 targets Hodgkin and systemic anaplastic large cell lymphomas while Kadcyla (trastuzumab emtansine) introduced to market in 2013 works against HER2-positive metastatic breast cancer (mBC) non-responsive to conventional treatment lines. Drug development pipelines currently hold more than 250 potential ADC designs, with more than 30 different lead ADC products in clinical development.
Given that many potent APIs suitable for conjugation have long been successfully commercialised and that monoclonal antibody development and production are well understood, the challenge in producing quality ADC products lays on the manufacturing side. Indeed, ADC development can be regarded as an area of manufacturing where very few companies offer all of the drugs’ pieces. This makes for a complex supply chain and calls for diverse partnership collaborations between specialised biotechnology companies, typically larger co-developer pharma companies and contract manufacturing organisations (CMOs).
ADC assembly itself is challenging, requiring providers to simultaneously handle biologic materials in aseptic conditions and manipulate highly potent small molecules under containment. Demand for experienced CMOs is therefore strong. Global CMO market is currently consolidating under heavy maturation forces that include changes to regulatory environment, changing client expectations, and small profitability margins for smaller (< US$ 25m) players. Consequently, in next few years we are likely to observe some of the smaller companies pursuing their competitive value by establishing strong technical niches. ADC assembly presents as one of such niches, one that is bound to attract investments across a wide range of disease areas where drug-target specificity is central to effective therapy.
Main antibody groups
Looking at marketed ADC technologies and those currently in development, small molecules conjugated to antibodies belong to two main groups: drugs that cause damage to the DNA and drugs that disrupt microtubule polymerisation. Common cancer chemotherapeutics such as doxorubicin, methotrexate, vinca alkaloids (auristatins and maytansinoids) as well as DNA-binding (eg calicheamicin) and tubulin-binding antibiotics (eg maytainsine) have also been successfully conjugated to antibodies. Notably, delivery in the form of ADCs allows to exploit the excellent potency of many cytotoxic agents with low therapeutic indexes by decreasing their systemic toxicities 100-1,000-fold as compared to their unconjugated forms.
Small molecule compounds are covalently attached to the antibody via a linker molecule that in some preparations (eg Adcetris) is accompanied by a spacer, crucial for providing enough room for the enzyme to cleave the bond and release the drug inside the target cell. Other methods of drug release include linker hydrolysis at low pH, cleavage of the linker through a redox reaction, and proteolytic degradation of the antibody. Stability of the linker and spacer molecules is central to efficacy and safety profiles of ADC preparations as it ensures delivery of the API load to its target, whereas the choice depends on the treated condition as a consequence of the unique biochemistry of target cells.
Cysteine and lysine are two most common naturally occurring amino acids suited to serve as drug binding sites on the antibody (usually IgG1 or IgG3). The major advantage of using cysteines is the relatively mild condition for the partial reduction of antibody’s inter-disulphate bonds and relatively high efficiency of the conjugation step. In case of lysines, the primary amine on lysine side chain forms a suitable reactive group for conjugation. Importantly, lysines do not commonly reside in the complementarity determining region of the antibody, which minimises possible interference with antigen recognition that is necessary for specific binding. However, with almost 40 lysine amino acids suitable for conjugation by having their side chains exposed on the IgG surface and much lower numbers with conjugated small molecule compounds (typically 3-4), resultant preparations could result in highly heterogeneous mixtures.
Best performance
For the best performance in vivo, however, the drug needs to be evenly distributed on the antibody. This is because a well-defined loading can make an ADC easier to purify and characterise, and possibly more stable and functional as a therapy. It is generally agreed that three to four drug molecules per antibody offer most optimal performance as too high numbers may lead to short ADC serum half-life and increased hydrophobicity and thus reduced solubility. Loading consistency is expressed with the number (usually as the drug-to-antibody ratio, DAR) and positions of drug molecules conjugated to single antibody. Successful conjugation should also allow to maintain antibody’s pharmacokinetics, antigen binding capacities, and sufficiently complex thermostability.
Site-specific conjugation technologies designed to manufacture more homogeneous ADCs are currently under development. Three main approaches exist and include taking advantage of an antibody’s natural structure and the widespread distribution of lysine or cysteine amino acids (eg Seattle Genetics’, ImmunoGen’s technologies), employing mutagenesis or non-natural amino acids to create defined high-affinity linkage sites on an antibody to attach drug molecules (Catalent’s SMARTtag, Innate Pharma’s tagging technologies), or both by using the natural bonding sites but improving binding efficiency and product performance (Meditope Biosciences’ SnAP, Igenica’s SNAP, and PolyTherics’ ThioBridge technologies).
Site-specific conjugation
Catalent’s SMARTteg technology aims for site-specific conjugation to create homogeneous ADC structures in terms of drug loading and positioning. This approach involves introducing natural five-amino acid sequences into the antibody as linkage points equipped with a chemical handle. From there, a formylglycine-generating enzyme recognises the sequence tag and selectively converts it into a unique aldehyde-containig residue suitable for joining with existing or novel linkers at defined locations. The company successfully tested more than 50 such combinations against multiple targets. Innate Pharma’s new conjugation technology uses a bacterial transglutaminase enzyme where single-point mutations in an antibody generate two to four enzyme-recognition sites. The enzyme then attaches a pre-synthesised drug-carrying linker the company has designed to couple specifically at these sites in order to consistently manufacture ADCs with an average DAR of three.
Meditope Biosciences’ ‘meditope’ technology combines what antibodies have to offer with useful genetic engineering. A naturally occurring pocket of space in the Fab region of antibodies can be altered to accommodate another molecule, typically a small peptide, called a meditope. Importantly, using the company’s patented SnAP technology, any antibody can be ‘meditope-enabled’ by removing several amino acids from that pocket of space. Antibodies enabled in this way remain fully functional and can carry an API load attached to the meditope itself. Igenica’s SNAP method on the other hand involves engineering site-specific amino acid changes in the native antibody to enable flexibility in choice of linkers and toxins. The company complements this approach with iTab (in vivo anti-Tumor Antibody) and sTAg (surface Tagged Antigen) methods to identify promising antigens and generate novel antibodies of high therapeutic potential.
Polytherics’ ThioBridge method uses a bis-thiol alkylating reagent to form a three-carbon bridge to which an API load is conjugated while maintaining antibody’s original structure and stability. The company is actively collaborating with drug discovery units of other developers to test a range of drug-linker combinations. Its patented conjugation process enhances conjugate’s solubility profile by reducing ADC aggregation and assembles products of suitably narrow DARs. Another novel technology is the ‘lock and release’ approach developed by ADC Biotechnology where an antibody is immobilised on a solid surface and conjugated prior to its release in a soluble form. The process is suitable for any antibody-linker-drug combinations and can be scaled up by running on columns in a flow mode.
With numerous examples of drug-antibody complexes and various conjugation technologies in pre-clinical and clinical development phases as well as promising commercial success of the field’s pioneers, this therapeutic technology area is expected to grow and diversify significantly in the near future. Areas of extensive development will include antibody-drug complex design, formulation protocols, and manufacturing (assembly) methods. Consequently, given the fast technical advancement in aseptic handling of small molecule compounds, dynamic CMO market restructuration, and evolving CMO-biotech partnerships, this technology market segment is expected to exceed US$ 5bn in sales by 2018 becoming a key component of cancer, and likely autoimmune and anti-inflammatory therapeutics.