BLOG: COVID-19 Vaccine Scaling Up and Scaling Out

My previous blog was written at the start of April, shortly after the UK had, like many other countries, moved into a period of “lockdown”. At the time there was in the region of 1.25m reported cases of COVID-19 and just over 60,000 deaths. Despite extensive lockdown scenarios across the globe, the numbers have now increased to over 14.5m cases, currently growing at rate of over 1m cases a week and over 600,000 deaths, however these numbers are known to be significant under-estimates of the true extent of the disease. It is also apparent that for many of those who have recovered from more severe forms of COVID-19, there are likely to be long term health impacts due to the damage this disease inflicts, not only to the lungs but to many other organs.

Currently, whilst there are emerging approaches for the treatment of severely ill patients in intensive care, the only means of stopping the spread of the disease once it is widespread within communities and over burdening health services, is through social distancing. Whilst this approach has significantly reduced disease transmission levels in many countries, it has also had severe economic impacts across the globe, the true of extent of which is still emerging. The pandemic has left many countries’ economies weakened to such an extent that a second round of lockdowns would not be feasible should the need arise. Consequently, the need for a vaccine to protect against COVID-19 increases by the day.


A Global Fight


A worldwide effort is underway to develop potential vaccine candidates. Whilst many of these have emerged from academic and small biotech companies, we now see the involvement of large pharmaceutical companies, with billions of dollars being pumped into these efforts from governmental and NGO’s such as the Gates Foundation.  Rightly or wrongly, it is also very clear that these products have taken on a political dimension with regards to access and ownership.  Several “lead candidates” have already emerged, essentially those that have been the first to progress into clinical trials, with at least three of these have now in Phase III studies.  Currently, there is no guarantee that any of these candidates will be successful, or of the length of protection they will give. Ideally vaccines should give life-long protection, but that is often not the case, and there may be requirements for re-administration. It may also be that the initial vaccine candidates are a “stop gap” while vaccines that give long protection are developed, assuming there is a sufficient level of infection present to be able to demonstrate their efficacy. However, the fact that so many vaccine candidates have progressed so far in such a short period of time is a remarkable achievement and a great credit to all those involved.


Different Approaches

 

In my previous blog, I discussed the potential of gene mediated vaccine candidates, which are able to build upon platform approaches, and when we look at potential lead candidates it is interesting to see that a number are platform based, with adenovirus viral vector systems, such as the AstraZenca, Cansino Biologics and J&J candidates,  RNA based systems, such as those from Modena, BioNTech, CureVac and Imperial College London, as well as a DNA based, such as the Inovio candidate. All these vaccines are based on relatively unproven approaches, although it is interesting to note that two Ebola vaccines from Merck and J&J have been licensed in recent months, paving the way for other vaccines based on their own internal viral vaccine platforms (rVSV and adenovirus respectively). Additionally, there are approaches based on protein therapies including one supported by Sanofi/GSK and a classical candidate from SinoVac based in China but is it the platform vaccines that have progressed the furthest to date.


Challenges to Overcome

 

It is interesting to not only reflect on the why and how these products have managed to progress so rapidly, but also to look at the many challenges that will need to be overcome before these products can be made available to people around the world. In terms of why, essentially these approaches are based on the delivery of small elements of genetic code through existing delivery vehicles, whether that be a viral or a lipid envelope and are produced through an already established production platform. Therefore, the only novel element is the identification of the coding sequence for the antigens required to generate a required immune response, a task which is increasingly being performed through in-silico modelling studies, based on sequence knowledge of the disease target, and often building on knowledge gained from similar vaccines candidates. The synthesis of novel gene constructs is again a task that can be achieved in a matter of days, this allows for rapid progression to the generation of potential vaccine candidates for studies in animal models using platform production processes, for the identification of lead candidates.


Development of vaccine candidates and putting them forward for clinical evaluation is just the first step in a long journey towards making a vaccine available on a global scale. Conventionally the manufacture of vaccines, like most therapeutics is scaled up and developed based on the achievement of clinical milestones, with commercial supply gradually rolled out, post licensing, to different territories. Additionally, manufacturing processes tend to be retained in-house or licensed to partners rather than being outsourced to CDMOs. However, COVID-19 is forcing all the standard processes to be challenged, clinical timelines are being accelerated with active support from regulatory bodies, as much as feasibly possible, and there is an expectation of roll-out across the globe once a vaccine has been shown to be safe and effective. This in turn places the focus on the ability to not only manufacture vaccines to support both early and late phase studies within multiple territories, but also produce stock levels which will allow immediate roll out across the globe.  Achieving this goal brings a whole new set of challenges to those which have already been addressed in order to get to clinic and will in no small part be linked to the intrinsic properties of the platforms for the different lead candidates. It is also fair to point out that a lot of modeling around the development of vaccines for emerging diseases, have in reality sought to be able to deal with limited outbreaks of diseases such as Ebola, rather than one that has affected virtually every country in world within a matter of a few months.


Factors for Consideration

 

When looking at the suitability of vaccine production platforms, there are several factors that need to be looked at. Firstly, how much product will you need to produce? This will depend on  dose sizes, the productivity of the production system (essentially how many doses can I make for a set process volume) whether this is for a cell culture system or reaction volume for synthetic processes such as the RNA based vaccines, and will protection be achieved will a single dose or will multiple doses be required. Secondly, is the alignment of the supply chain associated with the vaccine production? This relates not only to the supply of raw materials in for the process, but increasingly these days the supply of single use components. Thirdly, how will I achieve this required production scale? A decision has to be made whether to primarily scale vertically or horizontally and if production will be at single or multiple production site. Finally, the vaccine needs to be produced at an acceptable cost of goods, including those associated with the filling / vialing, test / release and distribution of the vaccine products.


In terms of scales required for the different approaches: with the viral vaccine it is possible to produce one dose from around 0.2-0.4ml of cell culture volume, meaning that a 1000L scale bioreactor would generate up to 5 million doses, whereas the RNA vaccines vary between around 100L reaction volume to produce an equivalent 5 million doses, and with the newer self-amplifying rRNA vectors, it is estimated that it will be possible to produce the same number of doses from a 10L reaction volume.


Viral Vaccines

 

When looking at the different production platforms, there are somewhat differing knowledge bases. Taking the viral vaccines first; whilst it is only in the last eight months that viral vaccines have been licensed, research and development around production processes for these products has been on-going for over 20 years. So, there is a reasonably strong baseline of knowledge around the production of these vectors and they are also simple to produce in that production is based on self-replication from an established viral seed stock, followed extraction from the cells and recovery and purification through a combination of membrane and chromatographic based operations. The challenge is production scale required; to date, the maximum production scales for this kind of products is around 2000L and often performed in single use bioreactor systems.  These systems have relatively straight forward supply chains, the critical element being the generation of the initial viral constructs and enough viral seed stocks to support extensive manufacturing campaigns. Conventionally, this activity can take several months, although there are approaches to accelerate them. Once these are established the supply chains are essentially based around the supply of cell culture media and single use components.


RNA Vaccines

 

RNA vaccines are produced in a very different manner, being constructed via “synthetic” based processes starting with a linearised plasmid template using restriction enzymes and a series of polymerase enzyme-based reactions to generate the RNA construct from free nucleotides, this is then purified via chromatographic routes.  In theory, this is a straight-forward and simple “cell free” approach, performed in simple single use reaction vessels. However, the manufacturing processes requires not only the supply of plasmid DNA as a template to base production on, but also the supply of multiple enzymes in the required amounts, produced within the required quality systems. This necessitates from animal-free sources as well as the required controls and segregation processes to allow them to be used to produce clinical material, and furthermore the required nucleotide bases to feed into the reaction process. Additionally, these products also require formulation in a lipid “envelope” to achieve effective delivery which again requires specialist raw materials. To produce a vaccine on a level to meet the needs of a global pandemic, this is not a simple ask, for example, the enzymes required for these processes, will have never been required in anything like these scales before, so in order to produce the required manufacturing process, the scale of the production of these enzymes will also need to be increased to support the production of the planned vaccine products.


Scaling Up and Scaling Out

 

The second element in producing the required amounts of vaccines, is that of both scale up and scale out, and the overall production strategies used to produce the required amount of vaccine product. Whilst some biologic products have been produced at large scale in 20,000L+ fermenters, an order of magnitude higher for some microbial based products, these newer types vaccines have not been produced at large-scale. Viral vaccines for example have been produced at around the 2000L scale, whilst RNA vaccines have to date only been produced in small scale for relatively limited clinical studies.


So, there are clear challenges here, with regards to being able to produce the required amount of vaccine, whilst the obvious approach is to increase the batch size by scaling vertically, the second approach will be to scale out horizontally by creating multiple production streams. It is likely that to achieve required scales of manufacturing, a combination of the two approaches will need to be used. Scaling up is attractive as it should result in lower cost of goods, but given that, to date, these products have not been produced at large scale, there will be limits on process scales that can be achieved without significant process development being performed, which will take time as well as money. Additionally, this may also require the construction of new production facilities which, again, will require time to build. Generating multiple production streams is not only an approach to increasing production scale but also a necessity to de-risk the manufacturing and will be required from a regulatory perspective to ensure continual supply of the product. This also applies to the supply of critical raw materials and consumables used in the production process. Whilst in normal circumstances, it may be possible for companies to establish duplicate production facilities in house, the reality is that, especially for such novel vaccine products, this will require the outsourcing of processes to CDMO’s or partner companies with the required production facilities. Furthermore, it is increasingly apparent that either individual countries or trading areas such as the EU, will wish to have vaccine production established within their region, to allow them to have an increased level of ownership and control of vaccine supply to their own communities.


Process Transferability

 

The impact of this is that the manufacturing processes for vaccine candidates, will need to be highly transferable with regards to the process robustness, the facilities required and potentially the IP associated the production processes.  This clearly makes significant demands both with regards to the process itself, the knowledge and skill base required to produce, test and release the vaccine, the equipment required, and the facility used for the production processes. When looking at the production of viral vaccines compared with RNA based vaccines, we see somewhat different requirements. The production of viral vaccines is a relatively simple biological process, the challenge is the need for required levels of biological containments for handling of BSL level 1 live viruses impacting the need to disinfect liquid and solid waste, and the ability to decontaminate production facilities when required. Whilst this is outside the capabilities of antibody facilities, this kind of process can be readily applied within existing gene therapy facilities such as the ones at Cobra Biologics. The challenge here is that most of these facilities have relatively limited production capacities.


RNA vaccines manufacturing processes are more akin to chemical processes. Whilst relatively small in scale, there is potentially the need for solvent handling capability with purification processes using techniques such as reverse phase chromatography and the need for solvents in the lipid formulation processes; so such processes may be better suited to chemical rather than biologics facilities which have limited solvent handling capabilities. There is surprisingly little, if anything, in the scientific literature around the production of RNA vaccines from which you can draw the conclusion that the processes are far from straight forward and with limited companies working in this space, it is likely that there could be significant IP and specialist knowledge still around these processes. The consequence of this it that whilst it may be possible to make products in a single specialist facility, transferring these processes out to multiple sites on a global basis may be challenging, which may restrict the access to vaccines produced from this type.


With both vaccine strategies, there an additional challenge of manufacturing the filled drug product from which ever production route is used, both have their own challenges not least being the very large-scale filing of a live viral product. As with the manufacturing of the vaccine drug substance it is likely that this will have to be performed at multiple sites across the globe and will create a huge logistical challenge with regards to the distribution of such a large amount of vaccines to the global population.

 

In conclusion, it is very clear that the scientific and pharmaceutical communities are in a desperate race to produce a protective vaccine, and with every passing day that it takes to achieve this, thousands of lives are being lost and global economies are being severely impacted. To date, huge progress has been made towards the generation of a vaccine and Cobra Biologics, like all the other companies involved in the production of COVID-19 vaccines, are working flat out to achieve this. However, we are still in the very early days and now there needs to be an increasing focus on not only getting the vaccine candidates through clinical trials but also establishing the manufacturing platforms required to deliver the successful vaccines to patients across the globe.


Author: Tony Hitchcock, Technical Director