Working to Overcome Analytical Challenges Posed by Genetic Vaccines
Jun 4, 2021PAO-06-21-NI-01
Traditional vaccines comprise weakened or attenuated live viruses, or dead viruses. Administration of these vaccines is intended to elicit an immune response from the body, most notably production of antibodies against the virus. Newer recombinant vaccines (e.g., subunit, virus-like particle) are also designed to elicit a strong immune response.
The issue with these technologies is the length of time required to develop and commercialize effective products. Traditional vaccines can take up to 10 years or more, while recombinant vaccines generally require several years. The time is needed to develop unique vaccine products from scratch for each target virus.1
Genetic vaccines, including those based on naked or plasmid DNA and RNA (messenger RNA (mRNA), in particular, and viral vectors have different modes of action. Rather than cause the immune system to respond to the presence of the virus, they instruct cells to directly produce the desired antigens. This approach eliminates the need to work with live viruses and to achieve sufficient antigen uptake.2
As importantly, they can be manufactured using scalable platform technologies that are applicable to multiple vaccine products. All that is needed is the genetic sequence of the virus of interest, specifically the section(s) of the sequence to be targeted by the vaccine. The formulation, production, and packaging operations are the same. The safety profile is generally similar as well.2 These factors contribute to dramatically reduced development timelines, as was observed for the Pfizer/BioNTech, Moderna, Johnson & Johnson, and AstraZeneca/Oxford University mRNA and viral vector vaccines against COVID-19.
In addition, because these vaccines cause direct antigen production, they also generally require lower doses, allowing for smaller manufacturing footprints — or the production of much more material in the same footprint needed for traditional vaccines.
Even with platform manufacturing processes, however, the rapid development of effective genetic vaccines cannot be achieved without rapid, robust, accurate, sensitive, and scalable analytical methods for use during process development and for in-process and product-release testing. Fit-for-purpose analytics are essential for ensuring the quality, purity, potency, and safety of genetic vaccines.1
Given the relative newness of genetic vaccine technologies and accelerated development timelines as well as the need for rapid turnaround of analytical results, access to ideal analytical tools is often not possible. Each type of genetic vaccine has specific analytical requirements, some of which are unique. Viral vectors are particularly complex, comprising both protein and genetic materials.
In many cases, vaccine developers are forced to rely on cell-based assays initially developed for use during traditional vaccine development with turnaround times of days or weeks, when manufacturing decisions must be made in a day or less.
Simpler, more rapid methods with sufficient sensitivity and precision, combined with automated sample preparation and data analysis, are needed to reduce analysis times. This also has to be done while providing the information needed to comprehensively characterize genetic vaccines and ensure monitoring — and control — of key process parameters.3
The common approach –– as is the case with developers of next-generation therapies, is to adapt existing analytical technologies initially developed for protein therapeutics to meet the specific needs of the different types of genetic vaccines. In fact, for DNA and viral vector vaccines, recent developments in the gene therapy field are being leveraged.
Often, these modified methods do not meet the unique requirements of genetic vaccines, not the least of which is the need for methods that require very minimal sample quantities. There is, as a result, a concerted effort by instrument manufacturers and software developers to leverage existing techniques and equipment in novel ways, including linking separation and detection methods through state-of-the-art technologies that contribute to simpler methods and provide more data in less time.3
Some of these efforts are also geared to the development of valuable functional assays that can replace slower existing assays.3 For genetic vaccines, some of the most important developments include liquid chromatography (LC) and capillary electrophoresis (CE)-based separation methods, which can be combined with mass spectrometry (MS) and a wide array of other detection technologies. These new methods provide rapid, accurate, and reproducible results for the protein and genetic components of DNA and viral vector vaccines, as well as sequence confirmation, separation and sizing, high-resolution analysis of important fragments, and confirmation of LNP composition for mRNA vaccines.
For instance, the direct coupling of capillary isoelectric focusing (cIEF) charge-variant analysis with high-resolution MS enables rapid analysis of intact viral vector capsid proteins. Acoustic ejection mass spectrometry (AEMS), meanwhile, eliminates the need for extensive sample prep and LC method development and run times, providing the performance of MS at the high-throughput speeds associated with plate readers. These new approaches demonstrate how the industry is working to create solutions that help vaccine developers make informed, timely decisions that result in shortened timelines for the development of effective, novel vaccines.3
For vaccines based on plasmid DNA, it is important to investigate the structure and purity of the plasmid, as both properties can directly impact vaccine quality, safety, and efficacy.4 Generally, plasmid DNA for clinical applications must be >80% supercoiled (vs. open or circular), and each must have the correct length and sequence.4
The combination of CE with laser-induced fluorescence detection (CE-LIF) has been demonstrated to be an automated and reproducible method for the rapid quantitative analysis of plasmid DNA isoforms and the purity analysis of plasmid DNA.5 Automated systems have also been introduced for Sanger genome sequencing of the recombinant DNA in plasmids.1
Viral vectors must be fully characterized before proceeding with the transfection process. The desired vectors themselves contain both capsid proteins and the genetic information, and, in an ideal world, only full capsids of the right size and with the right peptide sequence and posttranslational modifications (PTMs) would be produced with minimal contaminants (e.g., host-cell proteins, DNA).6 Confirmation of vector properties and purity is essential, because the capsid viral proteins are involved in cell entry, transport within the cell, and genome release.7
Analysis of the capsid proteins can be rapidly achieved using advanced MS techniques, such as quadrupole time-of-flight MS. Proprietary acquisition technologies and data analysis software allow fairly comprehensive peptide mapping using a single sample, including for low-abundance compounds.8 LC-MS/MS solutions are also available for the identification and quantification of thousands of contaminants in a single run. Other automated CE-based methods provide rapid capsid protein purity determination, even at very low concentrations, providing analysis with good sensitivity and high resolution, quantitation, and reproducibility.7
One of the biggest challenges for viral vector analysis is the determination of the percentage of full, partial, and empty capsids.7 Existing methods have various issues: quantitative polymerase chain reaction (qPCR)/ELISA and spectrophotometry lack sufficient accuracy, electron microscopy takes too long, ion-exchange chromatography (IEX) has insufficient resolution, and analytical ultracentrifugation is costly and complex and requires large sample sizes.
Capillary isoelectric focusing (cIEF) and CE-LIF are two simpler options that are robust and rapid and provide reliable and reproducible results.7 Multi-angle light scattering (MALS) and dynamic light scattering (DLS), when combined with size-based separation methods, have also been shown to be useful for evaluating the size, size distribution, aggregation, and particle concentration of viral vectors.9
Continued advances in analytical technologies for viral vector analysis are still needed, however, because many methods in use for product release are not sufficiently robust to compare the properties of vectors manufactured using different processes.10 In addition, the industry’s understanding of which vector properties directly impact safety and efficacy is still developing, further complicating the situation. Platform analytical technologies will ultimately be needed.
In December 2020, the World Health Organization published a draft white paper on regulatory considerations for the evaluation of RNA-based vaccine quality, safety, and effiicay.11 At the time of its publication, 17 mRNA COVID-19 vaccines around the world were either close to being submitted for regulatory review or already had received emergency use authorization.
The paper was issued by the WHO because no formal regulatory guidances exist with regard to assessing the quality, safety, and efficacy of mRNA vaccines. In addition to the information contained therein, mRNA vaccine developers must apply the principles of existing regulations, such as FDA CFR Title 21 for CMC information and ICH guidelines on cGMPs, method validation, elemental impurities, residual solvents, and setting specifications.
Like plasmid DNA and viral vector vaccines, in addition to confirmation of structure and purity,12 there are assessment requirements unique to mRNA vaccines. Of particular note are the 5´ cap and 3´ polytail, which impact both stability and function in vivo.13,14 Evaluation of the composition of the lipid nanoparticles (LNP) that encapsulate the mRNA is also essential, as the correct LNP is necessary for effective delivery of the mRNA into cells.15
Sequencing has increasingly been achieved using long-read, single-molecule RNA-Seq approaches (rather than short-read) to improve de novo transcriptome assembly and isoform quantification.16 These can be single-cell RNA-Seq and direct RNA-Seq methods.
All of the newer methods used for other genetic vaccines have applicability for mRNA vaccines, including Q-TOF MS, LC-MS/MS with advanced and targeted acquisition and software capabilities, and CE-based solutions. These methods can be used for detection, separation, sizing, and sequence determination, along with high-resolution analysis of 3´ and 5´ RNA fragments and lipid identification and quantification.1
Automated CE-based methods are particularly attractive, because they are fairly simple to implement, rapid, highly sensitive, and reliable, and require minimal sample quantities, unlike traditional manual assessment methods (i.e., PCR and microarray analysis), which require large quantities of material and can be negatively affected by sample degradation due to their long analysis times.17
DLS and size-exclusion chromatography (SEC) combined with MALS can, meanwhile, readily provide biophysical and structural information including molecular weight, size, conformation, and percentage of aggregates, data that can be difficult to obtain given the large size and high charge density of mRNA.18 Field-flow fractionation (FFF)-MALS can be used for high-resolution determination of LNP size distributions and quantification of particle concentrations.19 Real-time MALS enables continuous monitoring of liposome size and molar mass during manufacturing.
Robust, reliable, and rapid potency and safety assays are essential, but perhaps some of the most challenging methods to develop for any vaccine, let alone vaccines based on novel mechanisms of action. There is a trend toward substituting in vitro assays for traditional in vivo techniques for assessing vaccine potency and safety.
This movement is in response to the fact that in vivo methods are often not available, and those that are tend to suffer from inherent variability.20 In vitro assays tend to be not only more reproducible, but also precise and less time-consuming. There are several examples in which in vivo and in vitro tests have been run in parallel during both clinical and stability testing to demonstrate the advantages of in vitro methods.
With genetic vaccines, it is necessary to demonstrate that they function properly, which includes entering the correct cells and causing expression of the correct antigens. It is also essential to demonstrate that the expressed antigens afford sufficient T cell responses.
Many of the advanced analytical techniques used during genetic vaccine production can be employed to confirm the function-generated immune responses of DNA, mRNA, and viral vector vaccines, including MS, CE, and multiplex genome analyses.1
For instance, researchers at Bristol University leveraged advances in genetic sequencing and protein analysis technology to develop a method for validating the functioning of the Oxford COVID-19 vaccine (ChAdOx1 nCoV-19, also known as AZD1222).21 With this method, the scientists are able to directly check the transcription performance following entry of the viral vector vaccine into cells, as well as the structure of the spike protein expressed by the cells, thus validating that the instructions are copied and implemented correctly and accurately.
One newer approach to vaccine safety evaluation is the monitoring of gene expression biomarkers correlated with immunotoxicity and immunogenicity.22 This technology can be used to evaluate multiple biomarkers simultaneously, as well as to detect RNA directly in lysates or tissue homogenates without the need for sample purification. As a result, it can be particularly useful for assessing the biodistribution of mRNA in serum and tissues following dosing of mRNA vaccines.
Given the challenges to analytical method development for genetic vaccines, the goal is to develop rapid, robust, and reliable platform solutions that can be applied alongside the platform production technologies to shorten development timelines.
One way to achieve this goal is to leverage an analytical quality-by-design (aQbD) approach to method development. This approach was successfully used by the Analytical Assays group within Janssen Vaccines and Prevention (Johnson & Johnson) during the development of new high-throughput, high-sensitivity analytical methods with minimal matrix incompatibility for the analysis of viruses and viral proteins during vaccine production processes.23
Examples include a capillary gel electrophoresis (CGE) method for the quantification of influenza virus proteins and virus-like particles, a reversed-phase-ultra-high-performance liquid chromatography-ultraviolet (RP-UHPLC-UV) method for quantitative adenovirus protein profiling, and a capillary zone electrophoresis (CZE) method for precise and accurate analysis of adenovirus samples containing variable amounts of cell debris, cell lysate, host cell proteins, host cell DNA, salts, detergents, and additives.
The aQbD approach was adopted after input from analytical scientists revealed that a lack of clearly defined goals, process complexity, and minimal standardization often led to the use of analytical techniques that were not fit-for-purpose and often required redevelopment and troubleshooting. aQbD offers a standardized approach. Defining an analytical target profile enables selection of the optimum technology for a given method, while defining the critical method parameters ensures optimum method development.
In the case of the CZE method, the first challenge was to convince others within the organization that capillary electrophoresis, which was not used and, thus not well understood, is a robust method suitable for use in a quality control environment. After two years the CZE method was qualified in a QCD laboratory for in-process control of virus particle concentration. The new method provides results within 2 hours, compared with 1–3 days for older techniques, and is used at multiple facilities.
As importantly, many of the methods developed at Johnson & Johnson using the aQbD approach are applicable to any adenovirus-associated vaccine, including the one the company developed for COVID-19. Access to such platform methods reduces development and validation efforts and timelines, and, thus, overall vaccine development.
In 2019, even before the emergence of the COVID-19 pandemic and the SARS-CoV-2 virus, approximately 50% of vaccine candidates in phase II clinical development for pathogens identified as priority targets by the WHO were based on novel technologies, including DNA, viral vectors, and mRNA.24
These vaccine technologies provide many advantages over more traditional approaches, including access to platform production methods for reduced development timelines. They do create analytical challenges, however. Fortunately, key players in the pharma value chain have and continue to work on developing state-of-the art solutions that enable rapid, robust analysis of DNA, viral vector, and mRNA vaccines across the entire development cycle through commercialization and large-scale, ongoing production.
Many of these methods enable high-throughput analysis, and some are suitable for close to real-time monitoring. They are reproducible and provide the high sensitivity and accuracy required, given the minimal sample quantities often available. The optimum solutions are platform technologies that can be readily adopted for similar vaccines.
Continued success requires continued collaboration between vaccine developers, their contract research, development and manufacturing partners, instrument and software companies, academia, and regulators.20 This collaboration is essential to help move new methods using new instruments and techniques into the QC environment. Collaboration between different regulatory authorities can further facilitate this process by expediting comparison studies between existing and new methods.
