Notes on Antibody Manufacturing
I was helping a company with understanding where the bottlenecks and costs go towards antibody manufacturing and decided to publish my notes. I’d love to hear any thoughts or opinions from others who spent some time thinking about the space!
Assumption #1: Antibody manufacturing is expensive
How do you make antibodies at scale? What is the cost breakdown and why is it so expensive?
To my knowledge, there are four methods to produce them in lab or the clinic at a small scale:
mouse hybridoma
phase display
transgenic mouse
single B cell
The overall production of monoclonal antibodies involves several key steps:
immunization of a host
hybridoma production
screening and cloning
monoclonal antibody purification
In large scale production (typically in pharma), a typical workflow for monoclonal antibody production at scale could look something like this workflow:
Seed train
where media is prepared and filtered to ensure sterility before being used in small seed bioreactors
seed bioreactors are scaled up through stages, moving from a first seed bioreactor (300 liters) to a second seed bioreactor (2,000 liters), and eventually to the Production Bioreactor (15,000 liters) where large-scale antibody production occurs
Primary recovery phase
involving centrifugation and filtration to remove cells and debris, followed by storage
protein A affinity chromatography for initial purification, where antibodies bind to protein A columns, and viral inactivation and filtration steps ensure safety and purity
Ion-exchange chromatography and hydrophobic-interaction chromatography
further purify the antibodies by separating them based on charge and hydrophobic properties.
Final filtration
includes diafiltration for buffer exchange and final polishing filtration to achieve the desired product quality
purified antibodies are cooled and prepared for final formulation and packaging
Limitations of large scale manufacturing
Three things need to happen in order to successfully scale production:
cells used must express the antibody abundantly
often resource intensive and require intensive training to grow and maintain cell lines and consistent growth
mAbs must remain pure when produced in cell lines
impurities can affect the efficacy and safety of the final product
purification steps must be scalable to handle large volumes
needs to achieve high-quality and highly abundant production during final QC
mAb degradation and other biological contaminants need to be removed while ensuring downstream processes avoid physically and chemically degrading mAb
Assumption #2: Raw materials are expensive
The raw costs in monoclonal antibody (mAb) production are also significant. High optimization costs are incurred, associated with selecting a subclone that exhibits the best growth and antibody production characteristics. These subclones must be capable of thriving in low-serum or serum-free conditions, which is critical for in vitro production. These fixed costs, while substantial, are spread over a larger production volume, potentially making the cost per gram competitive with in vivo production methods. However, achieving this efficiency requires substantial initial investment in optimizing and maintaining the production process. Below are the detailed raw costs associated with each stage of mAb production
Approximate costs of raw materials:
high-yield stable cell production: ~ $500,000 to $3,000,000 for developing and maintaining productive cell lines
cell culture media formulations: often customized, leading to expenses typically ranging from $50 to $100 per liter
growth factors: costly to produce, costing between ~$1,000 and $10,000 per gram
bioreactors: ~$500,000 to $1,500,000 each, depending on capacity and features
purification: depending on purification type and needs, ~ $5,000 to $20,000 per batch
quality control: ~ $100,000 to $500,000 annually depending on production scale
clean rooms and controlled environments: ~$250,000 and $1,000,000 annually
Some fixed costs can come from optimization process, which can add to expenses overtime. Selecting a subclone that exhibits the best growth and antibody production characteristics can be costly as these subclones must be capable of thriving in low-serum or serum-free conditions. While these costs are substantial, they are spread over a larger production volume, potentially making the cost per gram competitive with in vivo production methods.
Purification is a huge bottle neck
During the clinical phase of monoclonal antibody production, costs are driven by the use of expensive chromatographic resins:
antibodies are typically produced using a process called Protein A affinity chromatography, where the antibody binds to Protein A, which is immobilized in a chromatography column
the impurities can be washed away from the bound antibody, after which the pH level is adjusted to recover the purified antibody product
slight modifications can affect the binding site of the antibody, making it hard to predict the variability of the binding site pocket between different monoclonal antibodies within the same batch
single Protein A chromatography column can cost more than $10 million
Bioreactor titre limits
current bioreactor titre limits for monoclonal antibody production are around 3-5 g/L, with expectations to improve to 10-15 g/L in the next decade
variable costs in this process include experimenting with expanded bed chromatography, synthetic affinity ligands, and rigid chromatography matrices
Alternative production methods
Projected demand currently impossible to meet using mammalian cell cultures in stirred tank bioreactors, the current gold standard. Mammalian expression system are used to produce full-length antibodies, as lower organisms lack the necessary cellular machinery for production. However, at scale, isolating and quality testing hundreds of thousands of antibodies takes 3-6 months in mammalian systems. Chinese hamster ovary (CHO) cell cultures, the most common cell type used, remain fragile and difficult to grow and maintain at scale.
plant-based production
plant-based farming units (VFU) and fermentation platforms are interesting to explore in the future
downstream capture and polishing steps in plant-based production would be similar to those used for purifying mAbs from CHO cells, with comparable cost implications
mAbs must be released from intact plant tissues through mechanical disruption, which also releases many dispersed particles and soluble impurities that need to be removed before chromatography
bacterial production
potential titres of up to 3-15 g/L and shorter fermentation cycles in E. coli systems are expected to drive down costs
only a few bacterial species can form disulfide bonds, which are essential components of antibodies
e. coli and p. pastoris are favored for producing monoclonal antibody fragments due to ease of use, high yield, and lower manufacturing costs compared to mammalian expression systems
yeast production
can glycosylate but with high amount of mannose which can cause immunogenicity and reduced half-life
Assumption #3: Reducing downstream steps can reduce costs
To optimize monoclonal antibody production, there’s a few ideal outcomes:
Increasing yield and reducing raw material
Adopting continuous bioprocessing techniques over conventional batch methods can significantly improve productivity and scalability
Optimizing alternative production systems such as plant-based platforms for humanized therapeutic mAb production
Predicting cell line productivity and yield by identifying optimal environments and culture conditions → determining the best combinations for bioreactor conditions
Discovering novel antibodies and improving existing ones, with in silico antibody-antigen modeling
Reducing the number of downstream steps
Optimizing plant-based methods during the purification step offers significant advantages over traditional mammalian systems. A precipitation-based process can reduce costs throughout the therapeutic antibody lifecycle by minimizing the number of downstream steps and avoiding the need for buffer exchanges.
A continuous precipitation-based purification process coupled with a continuous perfusion production system, efficiency and cost-effectiveness are greatly improved. Here’s a paper that goes into the system in depth.
Upstream Processing
Fed-batch bioreactor
Continuous perfusion bioreactor
Includes a cell retention device to continuously separate cells from the culture medium, allowing continuous removal (harvest) of the product
Downstream Processing
Traditional process:
After centrifugation, the product undergoes single-batch chromatography
Hybrid process:
Combines features of both batch and continuous processes for improved efficiency and yield
Continuous process:
uses Simulated Moving Bed Chromatography (SMBC):
similar purification steps (Protein A, Ion Exchange, Hydrophobic interaction) are performed continuously
allows for continuous and efficient separation and purification of antibodies
Assumption #4: There is demand for producing antibodies at scale
The market for monoclonal antibodies (mAbs) is experiencing significant pressures to scale up production due to the high dosages required per patient and the large potential patient populations. For example, some treatments require up to 1 gram per patient per year and have potential markets of up to 500,000 patients. This scenario has catalyzed efforts to drastically reduce manufacturing costs, aiming to decrease from thousands of dollars per gram to hundreds or even tens of dollars per gram.
The economic drive is focused on reducing the production costs at commercial scales. The goal is to bring down the cost from thousands of dollars per gram to a more manageable range of hundreds or even tens of dollars per gram. This cost reduction is crucial for making these therapies more accessible and sustainable in the long term.
Top drugs in the market are primarily mAbs
Currently, over a hundred monoclonal antibodies have received approval from the US Food and Drug Administration (FDA) for use in humans. In 2013, nearly 10 tons of diagnostic and therapeutic monoclonal antibodies were manufactured. This production volume is projected to double by 2024 if the annual growth rate continues. The demand for mAbs is poised to escalate even further, especially for more common diseases.
The median price of mAbs by therapy area in US dollars:
Oncology/Hematology: $142,833 per year
Immunology: $52,969 per year
Infectious Diseases/Allergy: $29,808 per year
Ophthalmology: $22,464 per year
Cardiology/Endocrinology: $15,624 per year
In 2020, adalimumab had around $19.2 billion in sales. Keytruda, an mAb for cancers, followed with $11.1 billion. Other significant mAbs include Opdivo, Avastin, and Rituxan, all showing substantial sales figures.
Consider the case of rheumatoid arthritis, where monoclonal antibodies such as adalimumab (Humira) is a commonly used treatment. For patients, adalimumab is administered as an injection of 40 mg every two weeks, equating to approximately 1.04 grams per patient per year. Consequently, for every million patients, around 1,040 tons of adalimumab would be required annually.
Global antibody manufacturing facilities are expanding globally, with notable developments outside North America.
Concentration of global biopharmaceutical manufacturing
US and Canada: 36.6%
Europe: 25.9%
Japan and other Asia: 9.6%
China: 8.5%
India: 8.1%
Latin America: 6.5%
Russia and Eastern Europe: 2.8%
Africa: 1.4%
In Argentina, mAbxience operates a single-use bioreactor. Brazil hosts several facilities, including PlantForm with plant transgenic production, Butantan with continuous processing, and Libb with single-use bioreactors. China is a significant player, featuring WuXi Biologics, Pfizer, and Boehringer Ingelheim, all utilizing single-use bioreactors, and HJB with continuous processing. India’s Biocon and Cipla employ hybrid models and continuous processing, respectively. In Japan, Fujifilm uses continuous processing, while Singapore has Lonza and WuXi Biologics. South Africa’s iBio and Cipla use plant transgenic methods and single-use bioreactors.
Overall, there are significant market pressures around the need for more scalable and cost-effective production methods as more treatments start to get approved in the antibody space. mAbs used in cancer therapies alone is projected to reach $88 billion by 2024. Scalable and efficient production methods are needed to keep up with the pace of demand and treatments being approved, while reducing costs for both manufacturers and patients.
Additional reading:
https://www.sciencedirect.com/science/article/pii/S0734975017300290
https://www.ncbi.nlm.nih.gov/books/NBK100189/#
https://wellcome.org/sites/default/files/expanding-access-to-monoclonal- antibody-based-products.pdf
https://jbiomedsci.biomedcentral.com/articles/10.1186/s12929-019-0592-z
https://encyclopedia.pub/entry/165
https://www.frontiersin.org/articles/10.3389/fctls.2021.810779/full
https://translational-medicine.biomedcentral.com/articles/10.1186/s12967-019-1804-8
https://www.evitria.com/journal/monoclonal-
antibodies/monoclonal-antibody-production/