By Kerstin Pohl

Kerstin Pohl
Kerstin Pohl
Senior Manager, Cell & Gene Therapy and Nucleic Acids, SCIEX

Therapeutics has entered the genomic medicine era, with gene therapies, DNA vaccines, and other novel genomic medicines in development. These new gene therapies and vaccines require more complex process related impurity analyses compared with classical biologics, such as monoclonal antibody therapies. With different biomanufacturing processes and components that can introduce contaminants—such as host cell proteins (HCPs)1 and other impurities—into the final drug product, gene therapies and vaccines present unique challenges unlike those seen with classical biologics.

COVID-19 vaccine–related side effects

In recent times, vaccine-induced immune thrombotic thrombocytopenia (VITT), or thrombosis with thrombocytopenia syndrome (TTS), has been associated with the Vaxzevria (AstraZeneca), Jcovden (Janssen, Johnson & Johnson), and Sputnik V (Gamaleya) COVID-19 vaccines.2 VITT is characterized by the concurrent presence of two clinical conditions: thrombosis and cytopenia.

Although the mechanism of disease is still unclear, research suggests that, in some rare individuals, the vaccine may trigger the production of antibodies that attack a blood cell protein called platelet factor 4 (PF4), resulting in the cytopenia.3 The cytopenic reduction in platelets, which leads to the impaired function of platelets in the blood clotting process, is thought to be linked to the formation of blood clots or thrombi.

The incidence of COVID-19 vaccine–related VITT has been estimated to be about 1 case per 100,000 exposures. Despite this low rate, approximately 4 out
of 10 people who got VITT died.3 Moreover, millions of people across the world have been dosed with Vaxzevria or Jcovden. Consequently, COVID-19 vaccine–related VITT represents an enormous challenge, one that scientists hope to address through continuing investigations into the syndrome’s causes and risk factors, as well as into potential treatments.

VITT associated with HCPs

Researchers in Germany reported this year a possible association between VITT and HCPs found in the vector-based vaccines, Vaxzevria and Jcovden.4 HCPs are critical quality attributes (CQAs) that must be assessed and controlled to meet stringent regulatory requirements regarding acceptable quantities present in final drug products. If HCPs exceed these limits, the safety, efficacy, and stability of the medicine may be compromised, and patients may be put at risk.

HCP impurities come from biological expression systems, that is, host cells used to manufacture biologics. (HCPs in biomanufacturing processes may be accompanied by other protein impurities, including those introduced from the cell culture media, nutritional supplements, and viral components.) HCPs can be immunogenic themselves, or they can include enzymes that catalyze the degradation or modification of the active drug substance or its excipients, thereby damaging the final drug product.

When scientists analyzed Vaxzevria and Jcovden, Vaxzevria was found to contain significant amounts of HCP impurities, including functionally active proteasomes, and adenoviral proteins, whereas Jcovden was found to contain a much smaller quantity of HCPs. Moreover, constituents of the Vaxzevria vaccine, but not whole Vaxzevria virions, were found to bind to PF4. (Each virion consists of a complete adenovirus vector particle encapsulating DNA encoding the SARS-CoV-2 spike protein.) With Jcovden, neither the vaccine’s constituents nor its virions were found to bind to PF4.

Scientists also discovered that Vaxzevria, but not Jcovden, increased the permeability of blood vessels, meaning that Vaxzevria can potentially enter the bloodstream and be distributed around the body, elevating the risk that antibodies against PF4 will be raised. This may in turn contribute to the higher incidence rate of VITT associated with Vaxzevria than Jcovden.

HPC challenges posed by gene-based drugs

Whether classical biologics or the new gene-based drugs are being manufactured, certain general HCP analysis challenges must be addressed. They include the detection of low-abundance HCPs, the comprehensive detection of all the HCP impurities present in the drug, and the detection of changes of the HCP profiles during upscaling or other changes in manufacturing.

These challenges, however, may need to be addressed in different ways depending on which kind of product is being manufactured. Impurities in the gene-based drugs that incorporate adeno-associated virus (AAV) and lentiviral (LV) vectors cannot be thoroughly assessed using the same methods that are applied for classical biologics—namely, ligand-binding assays (LBAs) such as enzyme-linked immunosorbent assays (ELISAs).

Moreover, in the manufacture of gene-based drugs, additional challenges arise. They include determining when HCP analyses should be performed during drug development and production processes, which are often fast-tracked processes. It is also necessary to adapt to an evolving regulatory landscape, as regulators come to grips with a maturing modality and an expanding array of analytical technologies. HCPs associated with AAV/LV-based therapeutics and other types of gene-based drugs are more complex than those associated with classical biologics because of the greater system complexity and number of HCP sources involved.

Recently developed viral vector vaccines are based on a number of different viruses, and they are grown on a number of different cell lines, including plant and insect species (Figure 1). This is a point that has been stressed by Ejvind Mørtz, PhD, co-founder and COO of Alphalyse, a contract research organization. He adds that in the manufacture of such vaccines, growth media contain many different additives.

“There are also large variations in upstream and downstream manufacturing processes,” Mørtz elaborates. “Altogether, this represents a wide range of very complex manufacturing systems. Moreover, these drug products can contain other impurities, such as polyethylene glycol, glycerol, or lipids, which may interfere with the analytics and thus limit the kinds of analyses that can be performed.”

LC-MS for HCP analysis

Although LBAs/ELISAs remain the gold standard in protein-based impurity analysis, they lack the flexibility to cope with manufacturing processes that are being optimized. For example, when a process is modified, a once-relevant ELISA may become irrelevant. Worse, no commercially available ELISA may qualify as a suitable alternative. It often takes too much time to develop a new customized, process-specific ELISA when the drug development has been fast-tracked.

More important, ELISAs do not provide any details regarding the characterization of individual HCPs. Accordingly, ELISAs are blind toward any changes of specific HCPs if the overall HCP burden remains consistent. In most cases, ELISAs are used to measure HCPs in overall amounts. This approach means that changes in the concentration of individual HCPs are not measured, which can have severe consequences if an increase in concentration of a problematic HCP (that is, an immunogenic or enzymatically active HCP) is missed.

ELISAs rely on antibodies, which are generated by raising an immune response in species such as rabbits and chickens. This is a highly empirical approach, and the variation between batches can be considerable, resulting in differences in antibody specificity and sensitivity. Furthermore, ELISAs struggle to provide comprehensive detection coverage of all the HCPs present in a sample.

An orthogonal approach, such as liquid chromatography–mass spectrometry (LC-MS), can address these challenges and complement the ELISA analyses by adding unique capabilities. Alphalyse has developed an LC-MS method based on a data-independent acquisition (DIA) strategy called SWATH DIA. The method has been optimized for HCPs and is capable of detecting and characterizing HCPs from multiple species in one assay.5

SWATH DIA is sensitive enough to detect low-abundance HCPs alongside high-abundance HCPs, and it is capable of detecting all the HCPs present in the sample. Indeed, SWATH DIA can accomplish these tasks robustly and with high reproducibility.6 Once the data has been acquired and analyzed, it can be compared with data in annotated databases, enabling the identification and characterization of HCPs.

Thus, problematic HCPs flagged by regulators can be detected and quantified, such as immunogenic proteins or enzymes that are likely to degrade critical components of the drug product. Missing a critical HCP can have major implications for drug safety and efficacy. It can also lead to costly and time-consuming setbacks, especially if the HCP is missed until late in the drug development process. A critical HCP that is detected too late could end an entire project. Data-independent LC-MS approaches can reduce such risks.

Further optimizing production with LC-MS

HCP analyses should be implemented early during the development of a gene-based drug, using LC-MS to augment any ELISA data. Resulting information can then be used not only to establish the HCP profile of the final drug product, but also to inform the development of the drug production process itself.

“What we typically see with our clients is that they want to use LC-MS for analyzing their HCPs in their final drug substance, but actually they find out that this tool is also very good for process development; it’s just perfect for the comprehension, data-driven optimization, and documentation of the process,” says Thomas Kofoed, PhD, co-founder and CEO of Alphalyse. “We have seen clients using this to increase their yield by 10–20% by just testing different settings, equipment, and techniques. And once the medicine is on the market, a 10–20% increase in yield can mean a lot.”

Another way to optimize process development is to minimize sources of HCP contamination in the first place. Possibilities for doing so can be inserted into the target product profile (TPP), a planning tool for therapeutic candidates based on FDA guidelines. For example, the TPP can address how drug design considerations could prevent or reduce HCP contamination, as well as how HCP analyses could be set up to optimize HCP clearance and assure the quality of the final drug product.

“The first step on the arduous journey toward a new gene-based drug often starts small, in an academic or commercial laboratory,” says Venkata Indurthi, PhD, CSO of Aldevron. “We must pool our expertise. Choosing the right partner can dramatically help drug developers advance and navigate the challenges of regulatory-compliant drug development, scaled up production, and commercial release—including HCP control. Not only is it likely to save time and money, but it could also mean being first to market with a new therapy for patients who desperately need it.”


Kerstin Pohl is Senior Manager, Cell & Gene Therapy and Nucleic Acids, SCIEX.