A Case Study based on the Covid-19 vaccine being developed by the Jenner Institute, Oxford University.
An affordable, global vaccine for Covid-19 is still seen as the best chance of protecting vulnerable people and populations at the same time as getting everyone back to normal activity.
There are now 79 candidates on the list of potential vaccines to be used in the fight against Covid-19 (1) One of the frontrunners is the vaccine currently being developed by the Jenner Institute at Oxford University. It is a viral vector vaccine based on an adenovirus vector (AdV) platform with the SARS-CoV-2 spike protein substituted. Together with Merck kGaA, they have developed a rapid production platform for small GMP batches of non-replicating Adv, in preparation for response to emerging pathogen outbreaks (2). This platform is based on a scalable suspension culture with virus expression titres ranging between 5×1013 VP/L and 1×1014 VP/L and on overall purification yields varying between 13% and 25%. We have used Biosolve Process software to model this platform process to estimate capacity and costs to support the ongoing clinical trials, which is currently in late Phase II/III. The Covid-19 vaccine dose was based on quantities being administered in these trials of 5×1010 viral particles (VP) per patient (3). Based on these assumptions, we have estimated that between 1,300 and 5,000 doses could be produced from a 10L batch (4). In the best-case scenario this would be equivalent to having an annual capacity to manufacture half a million doses a year with a cost of $40 per dose, including 30% of fill-finish costs.
The aim is to manufacture billions of doses of low-cost vaccine for global access. Astra Zeneca and Serum Institute of India are building extra capacity at risk to supply each over 1 billion Covid-19 vaccine doses, and do not intend to make a profit, during the pandemic (5). To enable a manufacturing cost of Covid-19 in the range of $1 per dose (see 7), with similar dosage as in trials, assuming average upstream titres of 7.5×1013 VP/L and a lower downstream process yield of 16%, then a 4 x 30,000L bioreactor capacity would be required. If the final dose requirements post Phase III trial, end up being in the lower dose range of 109 virus particles (as patented, ref 6) and the downstream process yield is the highest, then a 1,000 litres bioreactor capacity would be required (titre of 1×1014VP/L and 25% DSP yield), resulting in a manufacturing cost per dose lower than $1. The cost targets are important (8) and they put restrictions on the manufacturing process. From the modelling, the cost targets can only be achieved by process improvements or by scaling up to large bioreactors (20-30kL). This has important implications. There is limited bioreactor capacity at this scale (9) and building new capacity usually takes >1 year. Relying on existing smaller scale capacity could produce 100’s of millions of doses, but the manufacturing cost could be high and may not reach the cost targets for global access.
These latter costs are associated with the production of bulk vaccine. What about the resources needed for dispensing doses? The model showed pre-filled syringes would add a minimum $0.9 per syringe/dose and a single-dose vial would add a minimum of $1.2 per vial. However, given the potential lack of vials to support billion dose vaccine production scales and syringes to support mass vaccination programs (10) alternative types of vaccine formulation such as the use of multi-dose vials need to be considered, in line with reports for influenza pandemic preparedness (11). The RNA/DNA-based vaccines may offer quicker access to low cost doses within current manufacturing capacity (12), but delivery to patients will be a common consideration for all vaccines in development.
Billions of doses of low cost COVID-19 vaccines based on safe viruses can be produced, even at higher virus doses, but the planning of suitable manufacturing capacity to achieve global access needs to be evident in the public domain. AstraZeneca has entered into several agreements to manufacture 100’s of millions of doses, with costs seemingly in the $2-$3/dose range (13) and using existing, relatively small bioreactor capacity; so it looks like they have enabled higher production titres and higher overall yields. Or maybe, the current vaccine supply is seen as a loss-leader, which raises many other questions?
Photo taken by Josh Sorenson from Pexels
- Fedosyuk, Sofiya, Thomas Merritt, Marco Polo Peralta-Alvarez, Susan J Morris, Ada Lam, Nicolas Laroudie, Anilkumar Kangokar, et al. “Simian Adenovirus Vector Production for Early-Phase Clinical Trials: A Simple Method Applicable to Multiple Serotypes and Using Entirely Disposable Product-Contact Components.” Vaccine, Vaccine Technology VII: Promises and achievements of the Decade of Vaccines, 37, no. 47 (November 8, 2019): 6951–61. https://doi.org/10.1016/j.vaccine.2019.04.056
- Merck website. Merck Supports Jenner Institute to Reach First Milestone in Covid-19 Vaccine Manufacturing. News Release April 14, 2020
- AstraZeneca press release June 13, 2020 – https://www.astrazeneca.com/media-centre/press-releases/2020/astrazeneca-to-supply-europe-with-up-to-400-million-doses-of-oxford-universitys-vaccine-at-no-profit.html
- Patent: WO2018215766A1
- A similar estimate for 1 billion dose manufacturing was given as $10/dose (BARDA/Johnson & Johnson – see Khamsi, Roxanne. “If a Coronavirus Vaccine Arrives, Can the World Make Enough?” Nature 580, no. 7805 (April 2020): 578–80. https://doi.org/10.1038/d41586-020-01063-8.
- Yamey, Gavin, Marco Schäferhoff, Richard Hatchett, Muhammad Pate, Feng Zhao, and Kaci Kennedy McDade. “Ensuring Global Access to COVID-19 Vaccines.” The Lancet 395, no. 10234 (May 2020): 1405–6. https://doi.org/10.1016/S0140-6736(20)30763-7.
- BioProcess International. “Forecasts for Biomanufacturing Capacity,” February 7, 2020. https://bioprocessintl.com/business/economics/forecasts-for-biomanufacturing-capacity/.
- The Guardian, 12 Jun 2020. Exclusive: Bottlenecks? Glass vial makers prepare for COVID-19 vaccine. https://www.theguardian.pe.ca/news/world/exclusive-bottlenecks-glass-vial-makers-prepare-for-covid-19-vaccine-461291/
- Vaccine presentation in the USA: economics of prefilled syringes versus multidose vials for influenza vaccination. Expert Rev. Vaccines 9(11), 1343–1349 (2010)
- Lonza press release- https://www7.lonza.com/~/media/Files/japan/News/200501_Press_Release__Moderna_Lonza_COVID_FINAL.ashx?la=en
- $750M for 300 M doses CEPI/GAVI – https://www.astrazeneca.com/media-centre/press-releases/2020/astrazeneca-takes-next-steps-towards-broad-and-equitable-access-to-oxford-universitys-covid-19-vaccine.html