May 17, 2021 PAO-05-21-NI-03
Regardless of the technology, all vaccines are designed to elicit a strong immune response so that, once vaccinated, a person’s immune system can effectively detect and destroy the pathogen of interest. Initially, viral vaccines were formulated using the actual virus — sometimes weakened, or attenuated, live viruses and sometimes killed viruses. Newer vaccines rely on recombinant technology and include subunit, virus-like particle (VLP), and conjugate vaccines. The newest genetic technologies include viral vector vaccines (which can also be considered recombinant) and those based on plasmid DNA (pDNA) and messenger RNA (mRNA).1
One of the biggest differences between traditional and genetic vaccines is their mode of action. Conventional vaccines elicit a direct immune response, while genetic vaccines instruct cells (relying on cellular transcription and/or translation) to produce a viral antigen (or antigens) that then stimulates the immune response.
Virus-based vaccines have been safely and effectively used to treat and, in some cases, eradicate deadly diseases for decades. They have the advantage of immediately provoking an immune response, but do involve response to the full viral load, which can be difficult for immunocompromised individuals. Live viruses, which are genetically mutated versions of wild-type viruses, also have the potential to replicate in unexpected ways, spread to others, or revert to their original disease-causing form, so great care is needed during development to ensure the right combination of potency and safety.1
Non-live vaccines, including recombinant and genetic versions (which do not in fact alter human DNA), do not have these same limitations. Some require adjuvants to improve their immune responses, however.1 For viral vector–based vaccines, there is also the potential for an undesired immune response to the viral vector itself. DNA and mRNA vaccines are typically encapsulated in lipid nanoparticles (LNPs) to enable delivery.
Genetic vaccines also have the advantage of containing only the small part of the virus’ genome that causes cells to produce the desired antigen(s).1 Doses are typically smaller as well, because the cell is making the antigen itself. For DNA vaccines, that involves transcription of the DNA into mRNA, which is then translated into the target antigen(s). For mRNA vaccines, only the transcription step is required. With viral vector vaccines, the nucleic acid encoding the antigen(s) is delivered to the cell by the virus and then transcribed to mRNA and translated into the protein(s).
One downside to these genetic vaccines is the need for a booster dose.1 This requirement arises because the RNA and viral vectors involved in antigen production are eliminated from the body after a reasonably short period of time. While the potential for adverse effects is lower than with traditional vaccines, so is the initial immune response. The booster dose helps achieve complete immune protection for a longer period of time.
The first genetic vaccine to receive approval from the U.S. FDA was the viral vector vaccine for Ebola from Janssen (Ad26.ZEBOV, followed by a second dose of MVA-BN-Filo from Bavarian Nordic). The mRNA vaccines against COVID-19 received emergency use authorizations (EUAs) in late 2020, while Janssen’s viral vector vaccine against COVID-19 was granted an EUA by the FDA in early 2021.
The COVID-19 pandemic has clearly highlighted the need for technologies that enable rapid development and scale-up of vaccine manufacturing processes that can be prequalified and readily transferred to multiple locations, including in low- and middle-income countries.
Traditional vaccines based on viruses typically take 7–10 years to develop and bring to market. For each target pathogen, optimum methods for propagation, attenuation, killing, and other steps. that still afford sufficient immunogenicity must be developed. Manufacturing of virus-based vaccines also requires higher biosafety protections and cannot be performed in facilities for the production of other drugs. Recombinant vaccines (subunit, VLP) require less time, but still take several years. Methods for antigen propagation must be developed, as well as appropriate carrier or conjugation molecules.1 The simplest situations arise for similar, known viruses, such as mutations of the flu virus.
Processes that are implemented in egg systems generally must be scaled out rather than scaled up, and thus require large manufacturing footprints to produce large quantities of vaccine. Cell culture processes require cell line development and process optimization. They can be scaled, but are often performed in large, stainless-steel equipment, also with large footprints.1 These large facilities take time and a lot of money to construct and validate. In all cases, it can take many months to produce each batch of vaccine.
The second big difference between genetic and traditional vaccines is the ability to leverage platform manufacturing processes. For DNA, mRNA, and viral vector vaccines, once a process is established, it can readily be applied for the production of vaccines against other viruses without the need for new qualification. “Reverse vaccinology” is a key component of this approach.2
Advanced bioinformatics tools help predict T and B cell epitopes and enable the rational design of the specific antigen or set of antigens needed to generate a sufficient immune response once the genetic sequence of the pathogen is known.2 Next-generation sequencing (NGS) is the primary technology used to identify genomic targets for vaccine development.3 Special kits are also available that provide gene expression insights into host–pathogen interactions that can be used to refine genomic targets.
DNA vaccines can be expressed and scaled up to clinical and commercial scales very quickly using established processes. For mRNA vaccines, transcription of DNA can also be performed using an established in vitro process. For viral vector vaccines, only the gene that encodes the antigen needs to be inserted into the initial vector backbones, after which production of multiple vectors can be achieved using common manufacturing processes.
No cell or viral seed banks are needed. In many cases, the same or very similar supporting analytics can be used. Downstream purification and packaging operations are also generally the same. The smaller doses of genetic vaccines also impact development. Smaller commercial volumes make it possible to leverage single-use technologies and modular facilities for faster scale-up and facility construction/start-up.1
As an example, Moderna shipped its first clinical drug product just six weeks after selection of the antigen sequence, and Lonza scaled up and commercialized Moderna’s process at its Visp, Switzerland site in only eight months.1 Lonza leveraged an existing suite in its manufacturing area that was previously preapred for fit-out and has access to supporting infrastructure.
The ability to use an existing process for other antigens is also valuable in rapidly addressing new virus variants.1 With traditional vaccines, substantial if not complete vaccine redevelopment may be necessary. Genetic vaccines have a good chance of protecting against variants unless major mutations occur in the protein fragment used to trigger the immune response. If significant mutations do occur, a new vaccine can be readily produced using a modified DNA template, potentially without any modifications to the process.
Multiple DNA vaccines have received approval, most for veterinary use. DNA vaccines are being evaluated for many different diseases, including numerous conditions caused by viruses, including various forms of cancer.
DNA vaccines are based on plasmid DNA. The FDA has issued guidance (November 2007)4 for the development of DNA vaccine products covering preclinical testing recommendations and information that should be provided in an IND application, such as sequencing data, manufacturing controls, identify, purity, quality, potency, and safety testing methods for product release and results, and when integration studies are warranted.
Issues with delivery and expression have been tackled with solutions including the use of transdermal, needle-free patches and electroporation and codon modification, respectively.5 One example of the latter is LAMPVaxTM technology from Immunomic Therapeutics, Inc.6 When added to DNA vaccines, the lysosomal-associated membrane protein nucleic acid coding sequence ensures that the vaccines are delivered to the lysosome in dendritic cells, making them readily available to form antigen–MHC-II complexes necessary for generating immune responses.
Manufacturing of plasmid DNA can be challenging, however.7 Plasmids are quite large and have a high negative charge and high viscosity. They are also sensitive to shear, and, during the production process, many impurities with structures similar to the desired product are produced. These issues create particular difficulties for producing high-purity plasmid DNA at large scale.
Plasmid DNA is produced via fermentation in bacteria. Many factors can affect the yield, including the quality of the master cell bank, the culture medium (particularly the carbon-to-nitrogen ratio), and various process conditions (e.g., feed and growth rates, pH, temperature).7 Conditions need to be chosen to provide optimal yield and plasmid stability.
Fed-batch protocols that involve cell expansion followed by DNA production have been reported to afford dramatically increased yields.5 The cells are harvested (centrifugation or microfiltration via tangential-flow filtration (TFF)) and then lysed using chemicals, enzymes, or physico-mechanical means to release the plasmids. Alkaline lysis (followed by neutralization) is the most common.7 Addition of an endonuclease and/or complexing agent to remove unwanted DNA is also a widespread practice.
Next, the harvest is subjected to clarification through depth filters, rather than centrifugation, which was the traditional method.7 Concentration and buffer exchange via TFF is then performed to prepare the plasmid DNA for purification. All of these operations must be implemented with careful control of shear conditions to avoid damage to the plasmids. They can, in fact, have a significant impact on plasmid yield.
A series of chromatography steps (e.g., anion exchange, size exclusion, hydrophobic interaction) removes impurities such as undesired DNA isoforms, host-cell proteins, high-molecular-weight RNA, and endotoxins followed by concertation/buffer exchange into the final formulation via TFF and sterile filtration.5,7 Advances in chromatographic resins designed specifically for plasmid DNA purification have helped improve throughput and yield.
More recently, strains of bacteria have been developed with the ability to increase plasmid production from 2–7 times that of currently used strains.9 Using these strains, vaccine capacity can be increased without the need to install more equipment, because the strains produce plasmid DNA more rapidly.
Even with these many advances, demand for plasmid DNA is outstripping capacity owing to the explosive growth of gene and gene-modified cell therapies, which currently rely on viral vectors made from plasmids, the demand for authorized COVID-19 vaccines, all of which use some form of DNA, and the many COVID and non-COVID genetic vaccines advancing through the clinic. Significant investments are being made in additional capacity to meet the growing need for large quantities of GMP plasmid materials.9
Multiple viral vector vaccines have been approved for veterinary applications, and, as noted above, Janssen received approval of a viral vector vaccine for humans for its Ebola vaccine in 2020. Several other companies are also developing viral vector vaccines against many different viruses and various cancers. They are also used for the production of gene and gene-modified cell therapies.
Production of viral vectors can be achieved via a few different routes. The most common approach — transient transfection of HEK293 cells with DNA plasmids — offers the shortest development times but is far from optimum with respect to yield.10 Coinfection processes have better performance in that respect but take a much longer time to develop. Packaging and producer cells lines have the potential to provide the best performance, but also take longer to develop, and there are still issues that must be addressed before they can become the technology upon which platforms ultimately are likely to be established.
For the transient transfection process, the number of plasmids required depends on the virus type. Once an optimized cell line is developed, the cells are expanded to inoculate the bioreactor. Cell culture was traditionally performed in two-dimensional flatware using adherent cells, which is labor intensive and not practically scalable.9 Today, larger, multi-layer, scalable single-use (SU) products provide cell growth areas up to more than 25,000 cm2 and are designed to provide similar growth kinetics to those observed in laboratory-scale equipment suitable for clinical production and small-scale commercial manufacturing.11–13
For large-scale production, suspension-adapted cell lines designed to perform in traditional stirred-tank SU bioreactors are commonly used.9 More recently, SU fixed-bed bioreactors that mimic adherent conditions have been introduced (by Pall14 and Univercells Technologies15) that provide more scalable solutions with higher yields.17 Transfection requires use of a transfection agent (commonly PEI-based compounds).
The upstream process typically takes 3–5 weeks to complete.9 Downstream processing for viral vectors is much faster (1–2 days) and involves unit operations similar to those required for conventional biologics (harvest, clarification, buffer exchange, chromatography, ultrafiltration/diafiltration, and sterile filtration). As with plasmid DNA, chromatographic purification is particularly challenging due to the need to separate very similar species, most notably partial and empty capsids. Removal of adventitious agents can be challenging as well.
In addition, the specific process parameters for each type of viral vector will be different. Current COVID-19 vaccines are mostly based on adenovirus (AV) or different serotypes of adeno-associated viruses (AAVs).
The analytics can pose hurdles too, given that viral vector capsids include both proteins and genetic material.17 Critical quality attributes (CQAs) for viral vectors include viral potency, identity, quantity, process residuals, aggregation, empty capsids, protein content, and product safety. The use of multiple orthogonal methods, many of them newly developed to address specific properties of viral vectors, are needed to fully characterize them and provide assurance of their quality.
The development of automated, end-to-end integrated, scalable platforms leveraging single-use technologies and continuous bioprocessing are helping to speed the development of viral vector vaccines (and gene therapies).9 Some pDNA providers are also developing standardized yet flexible viral vector components that can be readily produced at scale.18 New process control technologies designed for viral vector production systems are also leading to more consistent processes and higher-quality products.
In one recent example, Pall Biotech leveraged its viral vector manufacturing template, which comprises 80% standard and reusable elements, to develop a scalable process for the AstraZeneca/Oxford University viral vector vaccine against SARS-CoV-2 in eight weeks — an 80% reduction compared to the typical 40-week timeline.19 This approach simplified and compressed the end-to-end value stream design, supply-chain planning, and re-ordering requirements, enabling rapid transfer of the process to multiple CDMOs.
Cytiva has also developed a scalable, standardized suspension process for AV production from upstream cell culture to downstream purification that also includes two surface plasmon resonance assays for determination of virus titer.20 A process economic simulation revealed the process, particularly when single-use equipment was employed, to be cost-efficient at all investigated scales and scenarios compared with an established reference process.
Univercells Technologies has, meanwhile, introduced a modular, integrated solution for the upstream manufacture of viral vectors to enable rapid deployment of scalable production solutions that deliver high performance with low capital and operating expenditures.21 The system includes a bioreactor for intensified viral vector production integrated with a TFF cartridge for clarification followed by ultrafiltration (UF)/diafiltration (DF) to produce a highly concentrated harvest for downstream processing.
These solutions are of particular importance given the shortage of capacity for viral vector manufacturing, which, as is the case for plasmid DNA, is impacted by the growth of gene therapies and the need for large quantities of COVID-19 vaccines. There was, in fact, a shortage of capacity before the pandemic.22
Several companies are developing mRNA, including prophylactic vaccines to prevent numerous viral infections, as well as cancer vaccines, cancer immunotherapies, and therapies for the treatment of a number of diseases. Messenger RNA vaccines are produced from DNA. Plasmid DNA has been traditionally used as the starting material. The pDNA template contains the sequence for the RNA construct along with a DNA-dependent RNA polymerase promoter.23
Once the target sequence of the pathogen has been identified, it is possible to design a lead candidate within just a few hours.24 The DNA template can then be produced in just a few days using oligonucleotide synthesis, gene assembly, and PCR amplification.
The pDNA is produced as the circular isoform and thus must first undergo linearization via treatment with a restriction enzyme in order to avoid transcription errors.23 The linearized pDNA is then purified, typically via TFF and chromatography. In vitro transcription using RNA polymerase and nucleotide triphosphates provides the mRNA. To ensure good in vivo stability and function, a 5’ cap structure is added either cotranscriptionally or enzymatically.
Purification of the mRNA to remove endotoxins, immunogenic double-stranded RNA (dsRNA), residual DNA template, RNA polymerase, and elemental impurities can be achieved using a number of techniques.23 TFF can be used to remove smaller impurities but can lead to the formation of additional impurities, so a combination of chromatography steps (reverse-phase ion pair, anion exchange, affinity) is generally employed.
The purified mRNA is then formulated into an LNP using one of several techniques, the most common of which is solvent injection.23 Lipids in a solvent (e.g., ethanol) are mixed into a low pH aqueous buffer solution containing the mRNA using crossflow mixing or microfluidic mixing to generate the LNPs. TFF, performed immediately to avoid lipid degradation, provides the final mRNA formulation at the correct concentration.
This process creates scale-up issues, however, as solvents are flammable. The stability of mRNA also creates challenges for large-scale production and requires careful control of manufacturing areas.
The shortage of pDNA could be a significant issue for the growth of the mRNA market. Some alternatives have been developed, however. IDT offers double-stranded DNA fragments of 125–3000 bp in length that are non-plasmid based.17 They are high-quality, defined sequences that can be produced more quickly and cheaply than plasmid-based fragments. They can also be produced in the large quantities required for mRNA manufacture.
Touchlight has also developed an alternative to plasmid DNA.25 Its synthetic DNA is produced using a proprietary in vitro dual enzyme process to generate “doggybone” or dbDNA™, which is classified by regulatory authorities as a chemical rather than a biological product. The company has manufacturing sites in Spain and England and is expanding its London capacity threefold in 2021. The new capacity will boost the production of dbDNA to > 1 kg per month.
Ensuring the safety, identity, strength, purity, and potency (SISPQ) of genetic vaccines requires special expertise in methods for genetic sequencing and product release.1 Many of the existing techniques were developed for research use and must be modified for GMP applications. Sequencing, infectivity, and transfection assays, among others, are often time-intensive and more robust, rapid, and GMP-compliant alternatives are needed.
Fortunately, advances in liquid chromatography, capillary electrophoresis, and other separation methods have been achieved that allow, when combined with state-of-the-art detection methods, particularly mass spectrometry, the development of rapid, accurate and reproducible results for the analysis of the various components of genetic vaccines.
Work on genetic vaccines over the last decade has been one of the largest factors in the rapid development of mRNA, viral vector, and DNA vaccines against COVID-19. Tremendous advances in understanding of the critical quality attributes and how production conditions affect them have contributed to the development of the first platform manufacturing processes.
Much remains unknown, however, and significant improvements are still needed to ensure truly robust, scalable solutions are available for future vaccine manufacturing. Issues around empty/partial capsid and dsRNA removal for viral vector and mRNA vaccines must be more comprehensively addressed, for instance. Improved upstream and downstream processes are needed to increase yields, particularly for viral vectors.
Further advances are also needed to improve the stability of mRNA and DNA vaccines. Use of more appropriate excipients, and potential novel ones (if the FDA established a separate approval pathway for them) is one area of opportunity. Cryopreservation could potentially be a solution as well. Developing better delivery solutions for mRNA and DNA vaccines is another important goal.
The industry will also need to address the supply chain issues facing genetic vaccine manufacturing.1 In addition to the need for additional capacity to produce both pDNA and synthetic DNA and viral vectors, it will also be essential to resolve the supply constraints surrounding single-use components, which can significantly impede the advance of genetic vaccine development projects.
Dr. Challener is an established industry editor and technical writing expert in the areas of chemistry and pharmaceuticals. She writes for various corporations and associations, as well as marketing agencies and research organizations, including That’s Nice and Nice Insight.
.svg)