Imperial researchers say integrated modelling and flexible planning will be essential for manufacturers to meet the global COVID-19 vaccine demand.

Amidst the current global crisis all eyes are on the scientists racing to find a safe and effective vaccine for COVID-19. For most, a vaccine offers the best prospect of overcoming the current pandemic and a return to normal life. However, once a formulation has been approved new challenges for manufacturers will present themselves. A group of experts in biomanufacturing and distribution from Imperial College London explain the challenges of manufacturing a vaccine at an unprecedented scale and speed, and how they can be addressed.

Fast tracking a new vaccine

International academic and industrial teams are currently focusing all their efforts towards the development of SARS-CoV-2 vaccines. Development and testing of a new vaccine normally takes years, and sometimes decades, to complete. However, the COVID-19 pandemic is forcing this to change, and scientists are now working hard to fast-track vaccine development to just 12-18 months.

Although the biggest battle at the moment is to produce an effective formulation and complete clinical trials, manufacturers must begin planning for the challenges of producing and distributing the vaccines.

These include making modifications to existing facilities for new lines of production and significantly larger quantities, all in a short space of time. Added to this, the current climate means normal supply chain routes and distribution networks may face disruption. Underpinning this is the desperate need to get it right first time, to save lives and enable society to return to a regular pace.

Companies must be innovative and flexible if they are to successfully manufacture a viable vaccine at the speed and scale required to meet the global demand, and they will need to act quickly to rise to the challenge. In the UK we have seen how the Government has faced difficulties in trying to secure personal protective equipment (PPE) and materials needed for widespread coronavirus testing when normal supply chains fail. If a vaccine is to be one of the key elements enabling us to return to normal life, lessons must be learned from this experience to ensure that the UK is not left behind when a vaccine is approved.

Researchers in the Department of Chemical Engineering at Imperial College London are using their expertise to offer solutions to these challenges. Through partnerships including the Imperial Network for Vaccine Research, the Future Targeted Healthcare Manufacturing Hub, and the two Future Vaccine Manufacturing Research Hubs led by Imperial and UCL/Oxford, our chemical engineers have long-since been at the heart of projects developing more efficient and cost-effective methods of manufacturing and distributing vaccines and other medicines at scale. Their work is vital now more than ever, as they turn towards the fight to suppress and contain COVID-19.

Engineering solutions

Chemical engineers can support production planning and decision-making once a vaccine has received regulatory approval and moves into the manufacturing phase, in addition to collaborating with teams in the lab to develop new vaccine platform technologies, such as UK initiatives by Imperial College and Oxford.

Using computer modelling and other digital tools they simulate vaccine manufacturing processes, which enables them to evaluate the whole system - from raw material sourcing to all stages of production and distribution of the vaccine. From this information they are able to identify and address potential bottlenecks, and provide manufacturers, governments and stakeholders with accurate predictions on how to optimise their systems.

By improving the performance of a whole chain of tasks, activities and processes results, chemical engineers can ensure the required vaccine availability in reduced amounts of time, without risking vaccine quality.

Professor Nilay Shah, Head of Chemical Engineering and Deputy Director of the Future Vaccine Manufacturing Research Hub at Imperial explained: “A systems approach to manufacturing uses integrated modelling to evaluate entire processes from end to end. This ensures that, whatever the problem, the important factors are taken into consideration and resources are targeted at the most critical aspects. This enables us to make recommendations to improve the speed and efficacy of production, which will be crucial to delivering a vaccine for COVID-19.” 

How ready are we to meet the global demand?

There are approximately 200 vaccine manufacturing facilities worldwide, but none of them are ready to produce a SARS-CoV-2 vaccine, in large part because the final formulation is still being developed. With so many different vaccine technologies being trialled, manufacturers face the dual challenge of trying to prepare their facilities to produce an unknown product in a very short space of time. 

“Timing is essential if countries are to prevent subsequent pandemic waves in populations through vaccination programmes." Dr Maria Papathanasiou Lecturer in Chemical Engineering

Dr Maria Papathanasiou is an expert in biopharmaceutical manufacturing systems optimisation and has previously modelled solutions for cost-effective and timely delivery of personalised cancer therapies (CAR-T cell therapy) to support the growth of this industry, which will need to rapidly upscale its production in the next ten years to meet the increasing demand for this pioneering therapy.

Using her expertise in this field, she has turned her attention to the scale-up of manufacturing technologies for a COVID-19 vaccine and identifying good candidate supply chain networks to ensure vaccine availability.

She said: “Timing is essential if countries are to prevent subsequent pandemic waves in populations through vaccination programmes. With little room for long-term investment in new facilities, manufacturers must explore “smart” solutions to ensure they’re able to meet the global demand.”

Dr Papathanasiou explains how she and her colleagues have developed models to support manufacturers with this. They run computer simulations of a wide range of vaccine production processes and scenarios and use these to identify the best performing production setups. They can make recommendations based on these to ensure facilities are constructed more effectively, saving manufacturers substantial time and resources for experimenting with developing new processes. 

These models can also be applied to existing production processes as well as new facilities. Without being able to invest in long-term developments or new, larger facilities, most companies will need to optimise manufacturing processes using the equipment they already have and/or scaling-out process steps that seem to create capacity bottlenecks. This could involve retrofitting or re-purposing existing production facilities to produce the type and quantity of vaccine needed for COVID-19.

Modelling can also help manage the added complication of the sheer number of different vaccine technologies on trial. Each of these has specific requirements for upstream production, downstream and formulation, making it hard for manufacturers to reconfigure their existing facilities while clinical trials are ongoing. Simulations based on the two types of RNA vaccine, subunits and viral vectors can help provide manufacturers with recommendations on potential production setups, which will save crucial time once a vaccine has been approved.

Optimising production

In addition to making decisions around the type of vaccine being produced, manufacturers will need to consider whether to scale-up production, where production takes place in larger facilities, or to scale-out and prepare several small-scale facilities or production trains. Each has its advantages and disadvantages. 

In the current situation where billions of doses will be needed to vaccinate the global population, being able to produce large volumes of a vaccine at speed will be crucial. But that’s just one set of challenges. In this situation, one of the main bottlenecks is expected to be in the fill-to-finish process, which can be understood as the ‘bottling’ of a product. Suitable ‘containers’ are required in terms of size, sterility, stability of the product and, crucially, safety to patients.

Dr Cleo Kontoravdi is a member of the Future Vaccine Manufacturing Research Hub and Vaccine Research Network, with expertise in bioprocess modelling and optimisation. Commenting on the challenge of scaling up production she said: “No matter which vaccine platform technology is approved, fill-to-finish will slow down capacity. In the past, larger vials containing 10 or even 20 doses of a vaccine have been used, but this approach will not be enough for the current crisis.”

It is essential that manufacturers tackle these capacity limitations and identify an efficient strategy that will enable them to be effective. Dr Kontoravdi notes that the Coalition for Epidemic Preparedness Innovations (CEPI) is currently evaluating the use of a 200-dose bag setup in parallel to conventional multidose vials to speed up vaccine roll-out. Another solution could be a more efficient utilisation and prioritisation of existing fill-to-finish capacity around the world.

Scale-out versus scale-up

An alternative to scaling-up operations is scaling-out. This involves building or adapting a greater number of small-scale facilities, which can be geographically distributed. This approach addresses some of the challenges involved in manufacturing vaccines at scale, as well as some of the issues caused by disruptions to supply chains, by creating a broader network of facilities.

Under normal circumstances, supply chains have built-in robustness to manage the supply of raw materials and distribution of a finished vaccine. However, in the current climate these networks could face significant disruption from closed borders and limited international travel and transportation. During a pandemic there are additional pressures caused if personnel cannot work due to health issues caused by the outbreak, or if production processes break down due to contamination or other technical issues.

Scale-out of manufacturing reduces the risk of losing production and supply capacity, by increasing the availability of operational facilities over a wider geographic area. This means that if one facility fails, others will be able to continue with production. As we have already seen with protective and medical testing equipment, reliance on centralised production facilities has proven unreliable in times of uncertainty and high demand. There is also the advantage of flexibility, as operations can be more easily scaled-down once the current wave has passed, and possibly scaled back up if future COVID-19 wave were to occur. 

“The use of scenario and risk analysis combined with model-based optimisation can help identify the major bottlenecks in the production and delivery of a given vaccine, making the best possible use of resources". Dr Benoit Chachuat Imperial Co-Investigator within the UCL/Oxford Vax-Hub

Dr Benoit Chachuat is an Imperial researcher working on the development of safe, efficient and sustainable processes through the synergistic use of advanced computational modelling and optimisation methods, and process data. He is a co-investigator within the UCL/Oxford Vax-Hub, focusing on enhanced operational and economic tools for uninterrupted, low-cost supply of vaccines.

In the context of the current crisis, he explains that process and supply-chain modelling can empower comparisons between various deployment strategies, such as scale-up versus scale-out, for different vaccine candidates. He said: “The use of scenario and risk analysis combined with model-based optimisation can help identify the major bottlenecks in the production and delivery of a given vaccine, making the best possible use of resources, maximising speed, reducing costs, and minimising wastage.”

Some vaccines are better suited to scaled-out operations, such as those with established technologies, which have the advantage of existing know-how shared among several organisations. This makes technology transfer to multiple manufacturing partners in parallel possible. Other, newer, technologies are showing promise for this type of scaling.

Dr Zoltan Kis is a researcher working in the Imperial Future Vaccine Manufacturing Research Hub with Professor Nilay Shah and Dr Cleo Kontoravdi. He is evaluating options for producing a very high number of vaccine doses rapidly, with focus on using new vaccine platform technologies, such as the RNA vaccine platform. He explains that the high productivity of the RNA platform makes it economically viable even when implemented at small scale.

He said: “An RNA vaccine like the one being developed and tested at Imperial offers an increased production rate and capacity, at around 100-10,000 fold more than conventional vaccines in terms of number of doses per day and per unit volume of the production process. Due to this very high productivity, RNA vaccine production processes are suitable for scaling out, and for being set up quickly at small scale, addressing the need for both speed and large quantities for COVID-19 vaccine manufacturing”.

Re-routing supply chains and distribution networks

Another important factor for success is that of supply chains and distribution networks, which are vital to the delivery of raw materials and delivering viable vaccines. Traditionally, a vaccine supply chain considers three levels of storage - central, regional and district - with an increasing number of locations as the process moves down the chain. For planning to be effective in the current crisis, it will have to happen in real-time and be responsive to need, especially in the early months of any vaccination programme.

Uniquely in the case of COVID-19, vaccination urgency will lead to a highly-distributed network at the administration end which will have to be tightly coordinated. This will include safely delivering vaccines to GP practices, care homes and vaccination centres as quickly as possible, via robust distribution networks which can ensure uninterrupted vaccine availability and stable transport conditions so that vaccine safety and efficacy are not jeopardised. 

Professor Nilay Shah explained the importance of timely delivery to ensuring that vaccines are still viable when they reach their destination. He said: “When a vaccine has passed through the final stages of production it has a limited time in which it is active, so it is essential it is administered before it expires. Our modelling accounts for this and ensures that distribution networks will deliver it within this time frame, taking into account possible disruption and delays.”

In the current climate, it’s vital that supply chain networks are agile and flexible to mitigate risks related to route, and to be responsive to needs. In the early stages, vaccine availability will be insufficient to immediately cover global demand, so strategies will have to be decided at governmental and global levels to decide who will be prioritised for the initial doses. This will define the target for the supply chain and will change over time. 

"Integrated modelling can be used to support flexibility in the supply chain by considering vaccine manufacturing and supply together as opposed to separate activities.” Professor Nilay Shah Deputy Director of the Future Vaccine Manufacturing Research Hub at Imperial

Any strategy will consider factors including disease spread, age- and health- group vulnerability and climatic conditions impacting disease spread. It is highly likely that first priority will be given to front-line healthcare staff, other first responders, vulnerable groups and pregnant women.  Highly populated urban areas may also have to be placed high on the priority list as disease spread likelihood is generally higher here.

As the wider population begins to receive the vaccine and herd immunity is created, the supply needs will shift again to consider boosters that may be required to increase immunity.

Professor Shah added that this the supply chain may need to be re-optimised to increase vaccine availability to meet these needs. He said: “This may involve increasing manufacturing capacities and bypassing storage or collection points or using alternative routes in case some of them become unavailable. Integrated modelling can be used to support flexibility in the supply chain by considering vaccine manufacturing and supply together as opposed to separate activities.”

A flexible approach to multifactorial problems

All the above-mentioned challenges form a highly complex, multifactorial problem for which the global community has been mobilised to find robust solutions as fast as possible. For manufacturers to be in a position to rise to the challenges of mass vaccine production and distribution, a systems approach which uses integrated modelling will be key. By using these tools to approach manufacturing and vaccine distribution decisions together, our engineers will be able to propose candidate manufacturing setups and supply chain networks that can respond to the global demand in a flexible manner with uninterrupted vaccine delivery. 

Looking forward, their work will continue to promote the development of rapid-response vaccine production platforms. Combined with computational modelling tools, this will further accelerate quality assurance and the regulatory approval process. Thinking ahead, past the current COVID-19 crisis, this will support planning on how to best prepare for being in a much better position to overcome possible future epidemics and pandemics with less cost to human life and damage to wider society.

--