Slow and steady wins the race: Optimizing lentiviral vector production for efficient clinical-scale manufacturing

Lentiviral (LV) vectors are important gene-delivery tools for cell and gene therapy applications. Approved therapies now using LV vectors include Zynteglo for beta-thalassemia and Skysona for cerebral adrenoleukodystrophy.1Locatelli F. Thompson A.A. Kwiatkowski J.L. Porter J.B. Thrasher A.J. Hongeng S. Sauer M.G. Thuret I. Lal A. Algeri M. et al.Betibeglogene Autotemcel Gene Therapy for Non–β0/β0 Genotype β-Thalassemia.N. Engl. J. Med. 2022; 386: 415-427Crossref PubMed Scopus (0) Google Scholar,2Keam S.J. Elivaldogene Autotemcel: First Approval.Mol. Diagn. Ther. 2021; 25: 803-809Crossref PubMed Scopus (2) Google Scholar Numerous clinical trials using LV vector-based therapies are also currently underway, with over 100 studies registered (https://www.clinicaltrials.gov). Growing momentum in the use of LV vectors for therapeutics has resulted in a push for better production methods that are suited to clinical and commercial manufacturing phases. Researchers have moved away from traditional flask-based production approaches and have been adopting bioprocessing technologies to improve scalability. There have also been ongoing efforts to optimize production conditions to reduce costs, minimize reliance on animal-derived products such as serum, increase yields, and enhance clinical acceptability. In their study, Carme Ripoll Fiol and colleagues lead by Qasim Rafiq at the University of London optimized a method of LV vector production that employed a fixed-bed bioreactor (iCELLis Nano, Pall) and produced high titers without the use of animal-derived components.3Fiol C.R. Collignon M.L. Welsh J. Rafiq Q.A. Optimizing and developing a scalable, chemically defined, animal component-free lentiviral vector production process in a fixed-bed bioreactor.Mol. Ther. Methods Clin. Dev. 2023; 30: 221-234Abstract Full Text Full Text PDF PubMed Scopus (0) Google Scholar This manuscript contributes to a growing body of research that aims to resolve challenges that have previously limited the production of clinically acceptable LV-based bioproducts at large scale. Traditional methods of LV vector production typically consist of using large quantities of cell culture flasks or multi-layer cell factories to cultivate adherent human embryonic kidney (HEK) 293T cells in the presence of growth media that contain animal-derived serum.4McCarron A. Donnelley M. McIntyre C. Parsons D. Challenges of up-scaling lentivirus production and processing.J. Biotechnol. 2016; 240: 23-30Crossref PubMed Scopus (73) Google Scholar Newer production methods now favor the use of bioreactors, including both stirred-tank and packed or fixed-bed configurations. The move toward bioreactor systems offers several benefits for production including amenability to automation, availability of online monitoring, tight regulation of conditions such as pH and oxygen, ease of scaleup, reduced labor input, and a lower risk of batch contamination due to fewer manual manipulations. Stirred-tank and fixed-bed systems are both being pursued for large-scale LV vector production. Employing suspension cells in stirred-tank reactors enables scaleup to be performed in a linear fashion by increasing the vessel size to thousand-liter capacities. However, the main difficulty with this approach is the need for compatible downstream processes that can efficiently separate the cells from the LV vector. Fixed beds, on the other hand, contain a substrate for cell adhesion, which physically retains the producer cells, separating them from the LV product and enabling simplified purification, as less intensive clarification steps are required. Fixed-bed bioreactor systems tend to have large surface area-to-volume ratios; therefore, the productivity is high, and the volume of supernatant to be processed is reasonably low. Increasing production capacity with fixed-bed systems can typically be achieved by addition of parallel production units in a scaling-out approach.5McCarron A. Donnelley M. McIntyre C. Parsons D. Transient Lentiviral Vector Production Using a Packed-Bed Bioreactor System.Hum. Gene Ther. Methods. 2019; 30: 93-101Crossref PubMed Scopus (17) Google Scholar In addition to adopting newer technologies, there has also been an effort to eliminate animal-derived serum from the production process to improve the clinical acceptability, reduce dependence on unreliable supply chains, minimize batch-to-batch variability, and eliminate serum carryover into the final product. Removing serum also has the additional benefit of reducing production costs. Early work in attempting to remove serum from the production process found that LV vector yields significantly diminished. Ongoing optimization efforts, including the work described here and by others,6Bauler M. Roberts J.K. Wu C.C. Fan B. Ferrara F. Yip B.H. Diao S. Kim Y.I. Moore J. Zhou S. et al.Production of Lentiviral Vectors Using Suspension Cells Grown in Serum-free Media.Mol. Ther. Methods Clin. Dev. 2020; 17: 58-68Abstract Full Text Full Text PDF PubMed Scopus (48) Google Scholar have now bridged this gap in titers, making serum-free production a feasible approach. Fiol et al. performed a series of optimizations at small scale in flasks to establish a method that could be successfully transferred to a bioreactor system. They identified a number of conditions that significantly improved LV titers including lowering the cell seeding density and reducing the amount of plasmid DNA used for transfection. They also found a positive effect from performing a post-transfection media replacement and a benefit from the addition of sodium butyrate, which, interestingly, was only observed when production was performed in serum-free media. Following these optimizations, production was moved from flasks to the iCELLis Nano fixed-bed bioreactor system. Eight different bioreactor runs were performed, with each run exploring different conditions in an attempt to improve LV titers. The highest titers were achieved when perfusion mode was employed, whereby media are continuously exchanged, supplying fresh nutrients to the cells and removing waste, as opposed to batch mode operation, where the same media remain in the vessel for the entire duration of the run. Using perfusion mode proved beneficial, as it allowed the temperature-sensitive LV vector to be stored in the refrigerator during the production phase, thus minimizing losses in viability. Importantly, the authors demonstrated that titers obtained in serum-free conditions were comparable to those without the use of serum, highlighting the feasibility of this approach for clinical-phase production. In summary, the findings presented by Fiol et al. contribute to a growing body of work designed to streamline the LV vector production process and increase yields. Each iteration and optimization, while often seemingly small, is important, as it improves the cell productivity (i.e., number of viral particles produced by each cell), increasing the overall LV vector yield and significantly reducing the volume of raw product needed to achieve comparable titers to previous production methods. Of particular importance is their work demonstrating that LV production in a serum-free environment can generate comparable yields to serum-containing conditions. This important step forward will eliminate reliance on expensive and unreliable supply chains and will allow for the generation of a purer product, which is desirable for clinical settings. The authors were able to successfully transfer their LV vector production method to a fixed-bed bioreactor system, highlighting the feasibility of translating existing methods to new technologies. The study employed the “nano” size bioreactor configuration, which has a small surface area of 4 m2 and a working volume of up to 1 L. There is, however, potential for up-scaling manufacturing by using larger fixed-bed bioreactor systems such as the iCELLis 500+ (Pall) with a 500 m2 surface area and 70 L capacity, the Scale-X Nitro platform (Univercells), which has similar specifications, or the TideCell-100 system (CESCO Bioengineering), boasting a 1,560 m2 surface area and 100 L volume.7Lesch H.P. Valonen P. Karhinen M. Evaluation of the Single-Use Fixed-Bed Bioreactors in Scalable Virus Production.Biotechnol. J. 2021; 16: e2000020Crossref Scopus (0) Google Scholar,8Leinonen H.M. Lepola S. Lipponen E.M. Heikura T. Koponen T. Parker N. Ylä-Herttuala S. Lesch H.P. Benchmarking of Scale-X Bioreactor System in Lentiviral and Adenoviral Vector Production.Hum. Gene Ther. 2020; 31: 376-384Crossref Scopus (11) Google Scholar A limitation of this study is that it only described one bioreactor run for each condition; therefore, future work should focus on establishing the reproducibility of the method, which will be particularly important for moving to clinical and commercial manufacturing stages. Over the last few decades, the LV production field has been working toward establishing optimized methods to increase yield, improve purity, and ensure suitability for human use. It has been a slow, steady, and iterative process, which has begun to yield successful results, with manufacturers adopting these techniques in clinical and commercial production scenarios. The groundwork laid by these researchers, and many others, has been essential in supporting the rapidly growing gene and cell therapy sector, which continues to generate an ever-increasing demand for clinical-grade viral vector products. As we look forward, substantial investment will be needed to build infrastructure and facilities that can accommodate commercial-scale LV vector manufacturing. Such steps will be critical to prevent bottlenecks to the clinical trialing and the adoption of potentially transformative gene and cell therapies. The author declares no competing interests.