The ability to turn cells into miniature drug plants has revolutionized medicine, allowing developers to create better targeted and more effective therapies for a broader range of diseases. But establishing cell-based production systems has come with a price—the potential for viral contamination.
Despite industry efforts, the cell lines used in biopharmaceutical production can carry viruses which, if not removed from the finished product, can impact patients. To counter viral contamination, companies have developed various removal and inactivation strategies.
Workflows and timelines
The effectiveness of a viral safety strategy depends on the drug that is made, the cell line and reagents that are used to produce the drug, and the efficacy of the processing steps, says Shada Warreth, senior bioprocessing trainer and training coordinator at Ireland’s National Institute for Bioprocessing Research and Training.
“Several methods are available for viral clearance,” she continues. “However, the most common methods are pH inactivation, which is generally used for enveloped viruses, and viral filtration, which is generally used for non-enveloped viruses.
“The biggest challenge to these viral clearance methods [involves the passage of a processed] solution through a viral filter, which will have a very small pore size. We are talking nanometers. [Another challenge] is ensuring that the processed solution maintains its virus-free state up to the point where it’s filled and sealed in the final container, be it a vial or syringe, for example.”
Drug companies use virus studies to determine if removal steps are effective. And there are practical challenges involved in this type of research, says Andrew Bulpin, PhD, head of process solutions at MilliporeSigma.
“The biggest challenge is managing the timelines for these studies,” he details. “A study cannot be scheduled until the downstream purification process has been finalized and sufficient intermediate materials are available for the viral clearance spiking studies.
“The viral clearance studies then need to be scheduled with the contract research organization. Managing these timelines can become even more of a challenge when there is any urgency to start clinical trials.”
Bulpin adds that viral clearance studies have been performed in a similar manner for over 25 years with very few changes, and that the methods used in these studies are typically very well characterized. “However,” he warns, “regulators will not allow generic clearance data to be used to reduce the costs or shorten the timelines associated with process validation. There are a limited number of contract research organizations who perform these studies, and capacity is limited with few new players entering this market.”
Continuous manufacturing issues
For batch-based production, viral clearance and inactivation protocols are well established and effective, even if they are demanding in terms of expertise and capacity. Less familiar viral safety challenges are being posed by continuous production, which has come to be used more widely over the past decade by biomanufacturers seeking to boost efficiency and reduce costs.
According to Warreth, continuous production comes with technical requirements that can be a challenge for systems developed for batch-mode manufacturing. “The processing times are longer in traditional manufacturing, where there are multiple discrete processing steps, than in continuous manufacturing, where holdup times between different steps is reduced,” she explains. “However, should contamination occur, difficult questions arise. How much process fluid could be lost? Where in the continuous process did the contamination occur? Where does the batch start and end?”
Viral inactivation is a particular challenge in continuous manufacturing, agrees Moo Sun Hong, a bioprocess engineering researcher at the Massachusetts Institute of Technology (MIT).
“Continuous manufacturing of monoclonal antibodies and other biologics is attracting significant interest due to a number of benefits over traditional manufacturing systems, including lower costs, reduced processing times, and improved product consistency and quality,” he says. “However, an important challenge during continuous biomanufacturing is the integration of effective, continuous viral inactivation systems.
“Viral inactivation techniques, including a low pH hold to inactivate enveloped viruses, are critical for minimizing viral contamination risks and ensuring patient safety. Previously invented approaches for integrating low pH holds into continuous processes lack sufficient control over key process parameters, such as operating pH and residence time. There is a need for tightly controlled, viral inactivation systems that can be integrated into continuous biomanufacturing processes and ensure product quality and safety.”
To address this need, Hong and colleagues developed a system that allows for precise control of pH and residence time during viral inactivation.1,2 The input solution containing the biologic is mixed with acid and pumped into a glass-packed column, where it is incubated for a time sufficient to inactivate enveloped viruses.
The pH of the fluid is monitored by sensors and tightly regulated using a model-based, feedback control scheme. In addition, residence time distributions are periodically measured through inverse tracer experiments and used to adjust feed flow rates.
Hong and colleagues have shown that the approach has a significant impact on viral inactivation. “We demonstrated,” he points out, “that the feedback control scheme enables fast and accurate regulation of the solution pH, with rapid start-up and effective disturbance rejection. We further determined that the system can correctly identify residence time changes and adjust feed flow rates to meet residence time set points.
“Such suppression of pH and residence time disturbances is essential for ensuring effective viral clearance with minimal degradation of product quality. Incorporating this invention into continuous biomanufacturing systems can increase productivity, improve product quality, and enhance patient safety.”
For some biopharmaceutical products, standard approaches to viral inactivation or clearance are simply not applicable, irrespective of the production mode. Hydrophobic monoclonal antibodies, for example, can be unstable and prone to aggregation in the pH range used for virus inactivation, that is, pH 3.4–3.6.
As a result, researchers working with glycosylated antibodies or IgG2- and IgG4-based products need alternative ways of deactivating viruses, says Green Guihang Zhang, PhD, associate director of large molecule purification at Incyte.
“Hydrophobic monoclonal antibodies are usually stable at pH 4.0 or above,” he explains. “However, virus inactivation at pH 4.0 is not robust and usually requires more than 120 minutes of incubation time for complete virus inactivation, in comparison with less than 15 minutes at pH 3.4–3.6 for an effective virus inactivation.”
Currently, inactivation alternatives include detergents such as Triton X-100 and Triton CG-110 or the solvent/detergent combination that is tri-n-butyl phosphate/polysorbate 20. These alternatives can inactivate enveloped viruses at near neutral pH. Caprylic acid, arginine, or an arginine derivative may be effective at slightly acidic pHs, where hydrophobic monoclonal antibodies are usually stable.
Incyte has developed a hybrid approach that melds low-pH inactivation with physical processing methods. This approach, Zhang asserts, is more efficient than other options.
“Our proposal is to combine or integrate the low-pH virus inactivation at pH 4.0 with nanofiltration, which usually lasts more than 120 min, as a continuous process,” he elaborates. “This way, we can eliminate the incubation time in the regular low-pH virus inactivation process. The continuous virus inactivation and nanofiltration will simplify the overall purification process by eliminating the separate virus inactivation incubation, the container transfer needed to overcome the liquid drop ‘hanging effect,’ and the post virus inactivation neutralization.”
Combining technologies is also key to a new purification platform developed by Samsung Biologics, a contract development and manufacturing organization. According to Beomkyu Kim, PhD, the company’s lead scientist in downstream process development, the new platform consolidates six processes into a two-step purification procedure. It succeeds a three-stage purification procedure that the company deemed effective, but time consuming and costly.
With the new platform, the first step of the purification procedure involves a protein A column and accomplishes virus inactivation and depth filtration. The second step involves a multimodal (MMX) column and accomplishes nanofiltration and ultrafiltration/diafiltration (UF/DF).
“Compared to other purification platforms, the differences are at the second column,” Kim says. “MMX chromatography is used, and a Design of Experiments approach is used to determine the critical operating parameters for protein A and MMX chromatography.”
Kim reports that the two-step procedure allowed Samsung Biologics to cut the cost of purifying a 200-L production run by 20% and to reduce processing time by 25%.
Furthermore, according to Kim, in process streams passed through the two-step system, high-molecular-weight contaminants were below 2% and host cell protein levels were less than 100 ppm.
“The two-step purification platform was applied for two projects, and this year the platform will be used for clinical manufacturing production,” Kim adds. “As far as we know, the application of the two-step purification platform is not a common approach at the moment. Therefore, the application of the platform helps reduce our development timelines and costs with a high yield and purity.”
Viral clearance and COVID-19
The COVID-19 pandemic has disrupted operations across all industries in all countries, and the bioprocessing sector has been hit harder than most. Besides disrupting the complex supply chains on which the biopharmaceutical industry relies, the pandemic has also increased demand for technologies and services, as developers rush to create vaccines and therapies.
MilliporeSigma’s Andrew Bulpin says, “We’ve seen an increased demand for rapid execution of viral clearance studies to allow vaccines and treatments to start clinical trials. Normally, these studies are planned six to nine months in advance to allow sufficient time to execute these studies and for regulatory review.
“The pandemic has heightened the need to accelerate this timeline. Most of the new therapies for COVID-19 treatment are recombinant proteins and as such have a generic downstream processing stream that incorporates well-studied viral reduction steps.
“An argument can be made that computer modeling can be used to estimate some or even all of the likely viral log reduction steps from downstream processing. If this were shown to be robust, then the need for laboratory-based studies could be eliminated or at least dramatically reduced.
“Alternatively, one of the issues with viral clearance is the supply of an adequate volume of virus to perform these spiking studies. For years, other than live mammalian viruses, many alternatives have been proposed. Models such as bacteriophage or nanoparticles can be more rapidly manufactured and have an ability to yield higher reduction values. These have not been adopted because there has been sufficient time to use the more ‘appropriate’ mammalian model viruses; however, under emergency circumstances, these models may need to be reassessed.”
1. Hong MS, Severson KA, Jiang M, et al. Challenges and opportunities in biopharmaceutical manufacturing control. Comput. Chem. Eng. 2018; 110: 106–114.
2. Jiang M, Severson KA, Love JC, et al. Opportunities and challenges of real-time release testing in biopharmaceutical manufacturing. Biotechnol. Bioeng. 2017; 114(11): 2445–2456.