May 28, 2021 PAO-05-21-CL-06
Viral clearance strategies are essential for the production of safe biologic drug substances and drug products. Holistic approaches that include avoidance through careful selection and testing of raw materials and appropriate upstream process conditions, combined with multiple, orthogonal downstream virus removal mechanisms, such as virus inactivation and filtration, are required by regulatory authorities to ensure that biopharmaceuticals are free of endogenous or adventitious agents.
In a monoclonal antibody (mAb) templated process, viral inactivation typically follows protein A capture and is prior to polishing chromatography. This order of unit operations is attractive since viral inactivation is usually achieved through a 30-60 minute product hold step at low pH (~pH 3.5) and takes advantage of the low pH conditions required for Protein A elution, minimizing the amount of pH adjustment needed.
While pH, inactivation time, buffer, and temperature are important parameters for viral inactivation and can vary between different mAb processes, in traditional batch manufacturing the implementation of this unit operation remains fairly consistent across the industry and is performed in large stainless steel tanks. The protein A elution pool is captured in the first tank, where the pH is adjusted to the inactivation setpoint as needed. This low-pH bioprocess fluid is often transferred to a second tank where it is held for a predetermined, validated inactivation time to achieve the specified level of viral clearance. The pH of the bioprocess pool is then adjusted up to a pH suitable for the subsequent process step.
The harvest from a single bioreactor is generally purified over protein A in multiple cycles, resulting in several large-scale tanks dedicated for virus inactivation of the entire batch. As a result, the overall viral inactivation process is capital and plant footprint intensive, in addition to being relatively slow.
The biopharmaceutical industry is moving towards intensified and continuous processing to realize gains in cost, speed, flexibility, and product quality with the ambition to improve access to lifesaving therapeutics.
To date, the most signifcant progress has been in upstream operations with advances in perfusion processing. There have also been noteworthy developments in downstream, including continuous multicolumn chromatography for protein A capture and flow-through technologies for polishing rather than traditional bind-and-elute approaches.
This shift toward connected and continuous mAb processing, in particular the chromatography steps that precede and follow viral inactivation, necessitate the development of novel approaches to improve or replace this step as well. Continuous viral inactivation solutions are expected to have numerous benefits including: elimination of large hold tanks for inactivation, reduction of facility footprint requirements, reduced processing time, facilitated conversion to single use, and perhaps most importantly, potential improvement in product quality.
Therefore, inline technology that replaces traditional batch operation while enabling effective virus inactivation is expected to play an important role in the development of next-generation mAb processing solutions.
Viral inactivation is a well-understood process and relies on a number of parameters. Two critical parameters are pH and exposure time, with acid selection, protein concentration, and temperature also of high importance. While lower pH setpoints can reduce the time needed for inactivation, this typically incurs greater risk of protein damage.
Strategies to ensure critical parameters are well-controlled in traditional batch viral inactivation processes have been long established. As the industry moves to inline viral inactivation, it will be essential to develop very robust understanding and control of these critical parameters in the new processes, as well.
Design considerations for continuous inline viral inactivation are different than traditional batch operation. In a continuous flow design, the incoming solution requires inline acidification and complete mixing prior to the inactivation chamber, followed by inline pH adjustment and thorough mixing before moving to the next unit operation. In addition to efficient static mixers, pH sensor and control strategies will have increased importance. The pH probes need to be stable over the duration of the process, up to many weeks, while having the ability to rapidly detect and report changes in pH. Should the pH not meet the requirements of the process, control strategies such as diversion to waste are critical to ensure that only solutions that meet the criteria are used for further processing.
Fluid dynamics are also an important consideration when designing a continuous, inline chamber for inactivation. If we consider the simplest continuous incubation chamber, a straight pipe, we can see the immediate impacts of fluid dynamics, with fluid in the middle of the pipe flowing faster than the fluid near the edges or walls of the pipe. This will result in a broad residence time distribution, with shorter residence times potentially correlating to insufficient virus inactivation and longer residence times resulting in increased levels of protein damage.
In consideration of these challenges, significant effort has been invested in developing an effective incubation chamber design that provides a narrow distribution of residence times with an end goal of getting as close as possible to plug flow through the entire chamber. To this end, different designs have been built and evaluated to tighten the flow distribution profile, including a coil flow inverter, packed-bed reactors and serpentine flow designs.
In each case, an understanding of residence time distribution is necessary before robust claims can be made for chambers at manufacturing scale. A key question must be answered: What is the minimum incubation time that your protein is experiencing for the virus inactivation?
MilliporeSigma is developing an inline virus inactivation process leveraging an incubation chamber with a coiled-flow inverter design. In this design, the number of turns, number of 90° bends, and coil radius in the chamber have a direct impact on fluid dynamics and residence time distribution.
To gain a robust understanding of all operating and design parameters, we are conducting extensive testing using both small scale models and manufacturing scale chambers. As part of this effort, we are pursuing a design of experiments (DOE) approach at small scale to evaluate the impact of turns and 90° bends in chamber designs. Residence time distribution is determined for each chamber at a range of operating flowrates by introducing pulses of protein into the chamber and then pushing them through with buffer. The effluent is then monitored with UV and integrated to obtain cumulative distribution curves.
Though this DOE approach, we are able to develop efficient manufacturing-scale chamber designs that offer narrow residence time distributions. This efficiency reduces the need to use oversized chambers to ensure that even the fastest fluid existing the chamber will have met the minimum incubation time. In turn this efficiency also reduces the maximum residence time of the slowest fluid which minimizes the potential for degradation of pH sensitive mAbs.
In addition to the residence time characterization, evaluation of inactivation is important. We have evaluated an enveloped model bacteriophage and observed equivalent inactivation performance when comparing an inline viral inactivation process to a traditional batch process (static incubation) under equivalent conditions (pH and time), and in future work will evaluate mammalian viruses.
Regulatory authorities recognize the societal benefits of continuous processing to reduce cost and expand therapeutic access and are broadly supportive of these efforts. Patient safety will remain the key priority for regulatory authorities, biomanufacturers and suppliers.
Virus inactivation is a well-understood step. The key parameters (pH, time, and temperature) needed to achieve inactivation have been well characterized. As the biopharma industry evolves to inline viral inactivation, regulators are likely to scrutinize system implementation and the means for demonstrating inactivation. Manufacturers will need to validate their process parameter setpoints through small-scale virus spiking studies and then demonstrate that the mAb solution flowing through the viral inactivation system at manufacturing scale consistently remains at those setpoints throughout the course of the continuous run.
By characterizing chamber residence time distribution and system performance within a recommended operating window, the equipment vendor can streamline industry efforts to adopt and implement inline virus inactivation systems.
Solvent detergent inactivation is another batch viral inactivation technique that is commonly used in the plasma industry. This method utilizes a solution comprising an organic solvent and a detergent (surfactant) instead of low pH to inactivate enveloped viruses.
As with inline activation at low pH, it is essential to ensure the correct solvent and detergent concentrations are reached and to achieve the correct incubation time of the treated process solution in the incubation chamber. While less common in recombinant protein processes, this method – or sometimes a detergent-only method (no solvent) - has been evaluated for therapies that are significantly degraded under low-pH conditions.
The Biopharma industry has expressed strong interest in connected and continuous manufacturing in response to market trends and to expand access to lifesaving therapies. This evolution in Biomanufacturing understandably will take time, and will exist alongside legacy processes where investments in existing infrastructure and regulatory approval have been completed. Over time, the industry will transform since the benefits of speed and higher quality are important for all therapeutic proteins. The full promise of next generation processing will require the convergence of process technologies, digital solutions and analytics. This evolution will require advancements in process analytical technologies, sensors, and adaptive process controls. These future technologies will drive a reduction in process deviations and streamline batch release for continuous processes.
Mindful of these future needs, new facilities are being designed with considerations for perfusion bioreactors and connected or continuous downstream solutions such as multi-column capture and flow through polishing operations. These solutions, coupled with advanced approaches to buffer preparation and management, result in more efficient facility utlilization and footprint reduction.
Inline viral inactivation is a key contributor to the future of connected and continuous processing as the integral step between capture chromatography and flow through polishing.
Elizabeth Goodrich is an accomplished biotech engineer with over 30 years of experience in protein purification development and scale-up as well as system design and process automation. She currently leads the Process Technology Applications and Innovation team at MilliporeSigma, working to develop continuous biomanufacturing solutions that incorporate novel process, analytical, and digital solutions. Ms. Goodrich holds a Bachelor of Science degree in chemical engineering from the Massachusetts Institute of Technology.