Back to the future: where are we taking lentiviral vector manufacturing?Published: December 18, 2018
Separation & Expansion Technologies
Hanna P Lesch
From more than 200 clinical trials involving lentiviral vectors, only a handful of products have reached marketing approval. One reason for this may be the technical bottleneck in large-scale lentiviral vector manufacturing. Today there are several upstream and downstream technology solutions, which claim to support clinical manufacturing at large scale. These still have several limitations, such as a complex production methodology and the relatively high cost of the goods. The fragile nature of the vector further causes its own challenges. No one knows yet where the future will take us. This insight covers an overview of the current technology and discusses the possible future solutions for lentivirus manufacturing.DOI: 10.18609/cgti.2018.109
Submitted for review: October 18, 2018
Citation: Cell Gene Therapy Insights 2018; 4(11), 1137-1150.
Rough road from research to commercial advanced medicine
Although KymriahTM and YescartaTM, two advanced cell therapeutic products where T cells are ex vivo modified with lentiviral vector (LVV) to express chimeric antigen receptors (CAR), have achieved marketing approval, few other LVV products have even managed to reach late stage clinical trials. A major challenge holding back progress in the field is related to technical difficulties with the manufacturing of LVVs. Up to now, production for research purposes has been based on the co-transfection of the adherent Human Embryonic Kidney (HEK) 293 variant, 293T cells with different plasmid constructs [1,2]. As a straightforward method, this provides an attractive small scale manufacturing strategy for early clinical trials, but it is not suitable for commercial stage production, where plasmid transfections at GMP manufacturing scale, small batch sizes with relatively low titres, high cost of goods, the fragile nature of the lipid-enveloped virus and safety concerns are some of the challenges still to be overcome. In addition, there is the fragile nature of the lipid-enveloped virus.
Scale-up of an adherent cell production system
For a long time, most production applications have relied on adherent cells cultured in two-dimensional (2D) flask-type approaches, such as Cell Factories or hyperflasks/hyperstacks [3,4]. However, the controllability in flasks is limited as is the scalability. To expand the production area means multiplying the units, which makes them impractical to handle. Microcarriers dispersed in suspension were considered an option for adherent cell culturing but they did not succeed as a popular technology for viral vectors . Problems were reported in the handling of the carriers, expansion of a large cell mass and the need for labour-intensive operations for their separation from the vector later during the downstream process .
Pall brought to the market their iCELLis® technology, a disposable fixed-bed bioreactor with an integrated perfusion, developed for scaling up adherent cell-based production systems (Figure 1). There are two different size bioreactors available: iCELLis Nano, providing a culture area up to 4 m2, and iCELLis 500, with a culture area from 66 to 500 m2. iCELLis allows a closed and controlled environment and meets the current GMP requirements. The iCELLis Nano has been used for a range of vector applications, such as for adenovirus , retrovirus  and AAV . Previously, we evaluated for the first time the iCELLis 500 large-scale bioreactor for the manufacturing of Ad5 vectors . Recently, we developed an adherent cell-based manufacturing process also for LVVs. A small-scale process development was originally done in iCELLis Nano system. More recently we have tested different fixed-bed sizes and compaction, and optimized the cell culture parameters, such as cell density and perfusion and transfection conditions . iCELLis 500 100 m2 and 333 m2 scale-up runs for two different constructs were successful with a high yield of vector produced. Compared e.g to Cell Factories , the iCELLis500 systemcan provide up to 30 times bigger scale to produce vector per single batch in a controlled manner. Such a system may provide one manufacturing solution for future clinical applications. Nevertheless, the adherent production systems are still not trouble-free. In most cases, adherent cell cultures use animal-derived products, mainly fetal bovine serum (FBS), as the culture medium constituents. The serum provides an attachment factor for adherent cells but it also contains other constituents, e.g. growth factors, hormones, additional amino acids, vitamins, trace elements, fatty acids and lipids . FBS has benefits in virus production. It can enable a remarkable increase in the production of LVV and ultimately, stabilize the produced virus . However, there are number of reasons why the use of FBS is problematic, e.g. high lot-to-lot variability, introduction of animal components to the cell culture medium and the risk of potential microbial contamination. The serum can also complicate downstream purification. Serum is very expensive and resources need to be directed to thorough tests . Last but not least, ethical issues need to be considered too. Therefore, the demand to reduce the use FBS or to replace it totally has becoming an ever urgent issue. Future development should explore chemically defined media that can maximally support LVV production. Several commercial serum-free mediums are available for HEK293-derived cells, but most of these have been developed and optimized for suspension cells. Whilst some media may well support cell growth, they do support the productivity and vice versa. Pro293A-CDM (Lonza) is an example of a serum-free chemically defined medium for adherent cells that has shown promise for LVV production with PEIpro transfection . The adaptation of the cell line into serum-free conditions is generally time-consuming and the removal of FBS can decrease the titer. Luckily, virus production can be increased again by adding supplements, e.g. Thermo Fisher Scientific supplies LV-MAX-Supplement and LV-MAX-enhancer to boost LVV production in serum-free conditions. Lipids were shown to be one key serum component during retroviral vector production that could increase the yield and vector stability . Several lipid formulations are currently on the market, e.g. from Thermo Fisher Scientific, GE Healthcare and Sigma-Aldrich. The problem with the serum-free media and supplements is still their relatively high price.
Suspension cells have benefits
HEK293-derived cells might be adapted into serum-free suspension growth as another option for upstream development. For a long time, suspension bioreactors have been the only option available for closed, scalable, cost-effective production where culture conditions, such as the pH, pO2 and pCO2 can be controlled. The first successful PEI-mediated plasmid transfection-based method for producing LVVs in suspension cultures was published by Segura et al. in 2007 . Since then, there have been other publications showing that LVV production in a suspension bioreactor is feasible [16,17]. An important step in the development towards this resulted from  crosstalk between scientists in the field and technology suppliers. Today, several suppliers provide serum-free media suitable for 293-derived cells, such as Ex-CELL 293 (Sigma-Aldrich), Pro293S-CDM (Lonza), Freestyle CD293 and Expi293TM (Thermo Fisher Scientific). Many suppliers also provide large-scale disposable bioreactors. Examples are e.g. stirred tank bioreactors from Mobius CellReady (Merck Millipore), HyPerforma (Thermo Fisher Scientific), XCellerex XDR (Ge Healthcare), Biostat SRT (Sartorius Stedim) and Allegro STR (Pall). Another option is wave type bioreactors, called WAVE (Ge Healthcare) or Cultibag RM (Sartorius Stedim) [18,19].
Suspension cells do not require the use of FBS so are a natural choice in the future. They also provide an alternative approach for the transfection as suspension cells can be driven through flow electroporation and transfection without the need for expensive transfection reagents . The suspension mammalian cell electroporators, such as 4D-NucleofectorTM (Lonza) and Maxcyte VLX® (MaxCyte) are interesting options but capacity may become limited in very high cell numbers (eg. >1 x1012 cells in 200L bioreactor). Suspension platforms are in the pipeline for many companies in the field, such as Oxford Biomedica  and Genethon . This is the direction manufacturing is developing for the future.
Many processes described in the literature are batch or fed-batch mode production. If fresh medium is provided continuously while removing the metabolites from the culture, the cell density and viability can be increased. Up to a certain level, this has a direct impact on the increased productivity. Perfusion technologies integrated into suspension bioreactor based have been based on tangential flow filtration (TFF) or alternative flow filtration (ATF) (Repligen) and conditions needs to be optimized for both producer cells and LVVs as the shear stress could break lipid membrane of the cells or the fragile virus .
The transfection step has several process parameters (e.g. cell density, transfection reagent, plasmid amount, reagent versus plasmid amount ratio and complex formation time) that need to be optimized. Micro-scale automated bioreactors can speed up the optimization in a controlled environment and in a very small scale (ambr15, Sartorius Stedim or Micro-24 MicrReactor, Pall) usually in a very cost-effective manner. Rapid characterization of conditions are, of course ideally supported by a proper statistical design of experiments (Doe) .
Are plasmids the way to go?
The most commonly used plasmid constructs are third-generation packaging plasmids , self-inactivating transfer construct [1,25] and the vesicular stomatitis virus g protein as an envelope protein for pseudotyping . Traditional calcium phosphate-based transfection is cheap and practical in the small scale but has limitations in terms of its reliability, repeatability and scalability [4,27]. Polymer-based approaches, such as polyethylenimine (PEI)-mediated, or lipid-based transfections appear efficient, better for scalability, less toxic for the cells and suitable for suspension cells [15,28,29]. Raw material for GMP manufacturing must also be sourced from qualified suppliers with appropriate documentation of the batch manufacturing and quality control. Hence, the use of commercial quality assured transfection reagents, such as PEIpro (Polyplus) or Lipofectamine (Invitrogen), is increasing.
We and others have shown that the large-scale plasmid transfection can be successful as a technique, and it is not that much of a ‘showstopper’ anymore. However, many issues remain with its use. Large batch production can require hundreds of milligrams or even grams of plasmid. The production of these large quantities of plasmid can be very expensive with all the requirements for the master, working cell banks, large fermentation, thorough purification, fill and finish, quality control and documentation . Furthermore, plasmid DNA can affect vector downstream processing  and there is a risk of unwanted plasmid recombination . Furthermore, the human immune system can recognize plasmid DNA and induce an inflammatory response and silencing of the transgene expression, in the unlikely event that CpG contaminants can be part of the LV vector prep . In addition, the use of antibiotic selection pressure during the plasmid production is a major regulatory concern. Antibiotic-resistant bacteria are global health problems. The risk with plasmids is not only the antibiotic itself, but the antibiotic resistance gene might, in the worst case, end up in the patient in a scenario called “horizontal genetic transfer” (reviewed ). Major risks have already been seen with ß-lactam antibiotics, such as penicillin, streptomycin and ampicillin, whereas antibiotics, such as kanamycin, tetracycline and neomycin, are still accepted. Many players in the field are today using kanamycin-selection in their plasmid constructs instead of the traditional ampicillin. In the future, a serious attempt could be regulated to use mini circles or nanoplasmids to effectively result in antibiotic-free plasmids during viral vector production.
There are different types of antibiotic-free systems available for supporting plasmid replication/selection in bacteria cells without the need for the addition of antibiotics . Many of the approaches, however, have not yet been extensively applied to DNA production for Gene Therapy Minicircle technology (Plasmid Factory) has been proven to improve the quality of adeno-associated vectors. Intramolecular recombination of the parental plasmid leads to minimal plasmid DNA with basically only the gene of interest. With this technology there is a minimal risk for encapsidation of the bacterial backbone sequence, a lack of need for an antibiotic gene and improved transfection efficiency for a small plasmid and improved transduction efficiency of AAV [35,36]. The development of minicircles, nanoplasmids or equivalent structures for LVV production is ongoing.
Stable cell lines
Stable producer cell lines are naturally desired to facilitate a cost-effective way to produce LVVs and increase reproducibility, quality and safety. Scalability is improved when these cell lines can be cultured in suspension serum-free systems . The toxicity of lentiviral protease  and the envelope protein VSV-G  has hindered the development of stable cell lines for LVVs and constitutive production has yielded only low titres. The development of stable cell lines is very time-consuming. The most frequently used inducible system is the tetracycline/doxycycline antibiotic system, where transcription is regulated through the tetracycline response element (TRE) either by adding or removing the antibiotics in the cell culture medium (reviewed ). Constitutive packaging cell lines provide another option, but the traditional VSV-g envelope is replaced with a less cytotoxic option, such as RD114-TR [39,40]. A recent approach was to introduce a mutation into viral protease to minimize its cytotoxicity for the producer cells . Up to 1 × 107 Tu/ml titer has been achieved with many different producer cell lines . A new approach to create producer cells is genome editing, where the characteristics of the produced virus can be modified in the desired direction, such as when Milani et al. modified producer cells to be free of MHC, which reduced the vector immunogenicity .
Downstream process of lentiviral vectors
The lentivirus is sensitive to high temperature, pH and salt concentration changes and it is well known to be very shear sensitive. The stability of the vector can be strengthened by pseudotyping the vector with other envelope proteins, such as VSV-g. Its fragile nature makes downstream purification very challenging. The traditional way is to use an ultracentrifugation-based process , but this is not nowadays considered as a valid option. Typical large-scale purification consists of steps for clarification to remove the cell debris, tangential flow filtration (TFF) to pre-concentrate and diafiltrate the product, capture and polishing by chromatographic methods and final formulation and concentration (reviewed in ) (Figure 2).
Clarification by depth filters is a straightforward process step performed using a peristaltic pump and the technique does not need very complicated equipment. In the TTF step, the product is circulated as a retentate through the filter (cassette or hollow fiber), while the permeate goes to waste through the membrane pores. The recovery of LVV can be high (>97%) in TFF as long as the shear stress (pressure and flow rate) are kept reasonably low . Anion exchange chromatography, either resin , membrane [4,23], monolith-based  or affinity-based chromatography  have been developed for LVVs. However, the chromatographic step is still the bottleneck in the field as the recovery of the product is typically very low (<50%). Overall, the downstream recovery for LVVs have been only c. 30% , so further process optimization is still needed. Today several systems are available to support large-scale purification. Improved technology has involved single-use columns and flow paths, which save time and eliminate the need for cleaning and cleaning validation. Several single-use technologies are available, e.g. ÄTKATM family (GE Healthcare), AllegroTM (Pall) and Mobius® (Merck Millipore).
Downstream processing includes also the endonuclease step to remove/decrease the level of host cell DNA and plasmid DNA . The standard method to date has been to add Benzonase (Merck Millipore) to the harvested product or later during the purification. Competitors to Benzonase are now emerging with a number of new endonucleases entering the field, e.g. Denarase® (C-Lecta) and SAN High Quality (ArcticZymes), the latter of which is said to be more efficient in higher salt conditions.
The latest approach in the field of biopharma is a move to continuous processes where the production is long term. The product is harvested continuously, followed by direct downstream processing until the end-product is obtained. The continuous production of monoclonal antibodies can take weeks, even months per batch. However, virus cytotoxicity towards the producer cells is hindering the development of processes taking months for viral vectors. In an ideal system, the whole process would be integrated, fully automated and a product coming from the controlled upstream system would continue directly to purification. This would naturally lead to increased productivity, cost reductions and increased flexibility.
Upstream, continuous production would require monitoring the culture using automated sampling systems or probes and controlling the run based on the monitored parameters. The typical process parameters that are monitored are pH, DO, pCO2, the number of viable cells and the viability, nutrient availability and waste metabolite concentration and viral titer. pH, DO and pCO2 probes have been used for a long time in all existing bioreactors. pH measurement is still troublesome as a shifting of the measured value often occurs and thus this can require separate sampling and off-line calibration during runs. Cell growth has been traditionally analyzed by sampling the cells and counting them using a haemocytometer, but improved technology is now providing alternatives. Growing cells with an intact plasma membrane is one option as this acts as a capacitor under the influence of an electric field. This capacitance can be measured and converted to the live biomass reading, typically cells/ml (ABER Instruments). Capacitance can be used not only for monitoring the cell growth but also as a basis for controlling the system, e.g. the perfusion rate based on the cell number [10,46]. Such a system is suitable for suspension cells and adherent cells, though in fixed-bed bioreactors only the top carriers can be monitored. Nutrient availability and the amount of produced metabolite side products require sampling of the medium. This can be done using an auto-sampling system, which takes the samples from the bioreactor and creates the process parameter measurements using external equipment, such as the Bioprofile family (Nova Biomedical) or Cedex family (Roche) equipment. Also glucose and lactate probes directly installed into the bioreactors have been developed. More advanced solutions could also involve real-time systems that would analyze the metabolic activity of the culture (e.g. Ranger technology, Stratophase).
The implementation of a continuous downstream system has been increasing in the recombinant protein and monoclonal antibody fields and might be a future direction in the Gene Therapy field as well. Traditional TTF operates by circulating the retentate through the filter. Merck Millipore and Pall have launched single-pass TFF systems, which runs at constant operating conditions, and no retentation return is needed as the product is sufficiently concentrated/diafiltrated. The new Cadance TM Inline Concentrator (Pall Life Sciences) utilizes single-pass technology and allows direct flow-through and volume reduction of the in-process product. Single-pass technology reduces the shear damage, improves the recovery, reduces the hold-up volume and is theoretically simple to operate ( , Merck Millipore, Application note; Pall Life Sciences, Application note). Traditional chromatography is a batch process whereby the product is loaded into one column, so scaling up means using a larger size column. Such process steps can become very expensive and time-consuming. The recovery can be also lower because suboptimal usage of the chromatography column may lead to a loss of important breakthrough products into the waste. Multi-column chromatography using the CadenceTM BioSMB platform is the first scalable, GMP compliant, disposable, continuous multi-column chromatography system (Pall Life Science, Application note). Multi-column chromatography operations are performed in a small-scale column with maximal recovery. The breakthrough virus from the first column is loaded and captured by the second column, while other columns can be for elution, washing, CIP or a regeneration stage for re-use. BioSMB has been used for viral vaccine purification with promising results (Tarpon Biosystems Inc.). A continuous downstream eliminates the hold steps, decreases the process time and buffer consumption and reduces the footprint and product costs. This can be also very crucial for LVVs, which are not stable at room temperature . A continuous process would make the operating hours more reasonable than stepwise batch downstream.
LVVs are typically stored at below -70?C. However, the LVV is not particularly stable and its activity decreases during storage or freezing/thawing . Formation of ice crystals inside the virus, pH changes and osmolarity can affect the viral membrane and proteins, and decrease the activity [48–51]. Such decreases can be reduced by optimal formulation of the final buffer. Sugars and salts are the main buffer components . Tris-based buffer has been frequently used for LVV [17,23,28] but problems are related to changes in temperature, which can lead to severe changes in pH. (4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid) HEPES buffer has smaller pH range and thus has been preferred choice in some of the applications [23,49,50]. LVVs for ex vivo transduction have often been formulated in the X-VIVOTM medium, which contains proteins and sugars to protect the vector . Nilsson tested different sugars and lyophilization for LVVs  and Camro et al. showed that recombinant human albumin and lipoproteins were one of the most promising for protecting especially reverse transcription . In Gene Therapy approaches, the human immune response is also a key issue to be considered as some of the components can induce the immune response of the patient, but this is an application-dependent issue.
Modern analytical tools are needed
Thorough testing of the vector is done already in preclinical stage. Also the GMP produced final product is heavily analysed by pattern of release assays (Table 1). The characterization and biologic activity of LVV for Gene Therapy follow the regulatory guidelines for other medicinal products, containing also the safety assay for Replication competent lentivirus (RCL) testing. The safety testing will be always unique and based on relevant biology of the product [53–55]. Nevertheless, modern technologies are needed to deepen the understanding, robustness and quality control of the process, but also the final product attributes. Scientists involved in the process development should know their process and its demands as the better the understanding you have, the better you can optimize your process. Standard titering, total DNA and total protein measurements are no longer giving adequate answers. Next-generation sequencing (NGS) technologies allow the possibility for greater understanding e.g. viral genome integrity, identification and integration sites in host cells, virus interactions, purity and quantifying transcripts. Equally, recent advances in mass spectrometry together with proteomics are providing broad information on the composition of virions, the structure, viral protein interactions and the effect of the infection on the cellular proteome . Raman spectroscopy provides a structural fingerprint by which molecules can be identified. With TEM imaging and high-throughput quantitative analysis, MiniTEM (Viranova) offers innovative solutions to analyze the morphology, particle integrity, particle size distribution, purity, aggregation and empty versus full virus particle ratio from TEM pictures . Digital-droplet PCR is replacing traditional PCR methods because it quantifies the absolute amount of target DNA and is an excellent method for accurate virus tittering without the need for applying a standard curve . These advanced analytics involve the use of bioinformatics, biostatistics and data management and will be key for the success in this field in the future. For LVVs, a commercial standardized reference standard is not available yet. Well-characterized reference standards are important for a comparison of the internal results and also the results between different sites and can be used for the establishment of appropriate pre-clinical and clinical dosing. LVV transduced cells for the reference purpose have been developed  and can be used for standardizing integration copy number analysis. Another worldwide LVV reference standard is under development for standardizing the determination of particle concentration and infectious titer (Lentivirus Vector Reference Standard Initiative – ISBioTech). Nevertheless, the comparison between the different sites can be still troublesome if target cell or purpose varies.
|Table 1||An example of releases assays for lentiviral vectors.|
|Safety/sterility||RCL (following infection of a susceptible cell line and detecting amplification of the vector by p24 Gag increase or RTase assay)|
|Insertional mutagenesis (integration site analysis by sequencing)|
|In vitro assay for the detection of viral contaminants|
|Determination of pH|
|Determination of osmolality|
|Determination of particle size and count|
|Purity||Host cell proteins (ELISA)|
|Detection of residual host cell DNA (qPCR)|
|Detection of residual plasmid DNA (qPCR)|
|Detection of residual BSA|
|Analysis of process contaminants (e.g. Bensonase ELISA)|
|Activity||Total particle number (amount of capsid protein taken into account that 1 pg GAG is approximate 104 particles or RTase activity or number of LV genomes by RT-qPCT)|
|Transducing activity/integration capacity/incorporation of vector proviral DNA into target cells (qPCR analysis of transduced cells)|
|Transgene expression (ELISA)|
|Ratio between viral particles and functional viral vectors|
|Functionality/potency of the product (application specific assay)|
Cost of goods
Ultimately, money drives change, and one big challenge in the current production is the very high costs. Production consumables, such as bioreactors, filters and columns, can be tens of thousands of dollars per batch. LVV manufacturing typically still uses plasmids and special reagents, such as FBS, transfection reagents or endonuclease, where the price tag is getting higher and higher as quality standard rise. Good quality control can easily take again tens of thousands more for all the analytics and when one also takes into account the very high overhead costs (facility, supporting actions, human resources), the final price can be suprising. StrimvelisTM, a stem cell therapy of ADA-SCID patients, was initially priced at USD 665 000 per treatment. What can be done about this? Costs per dose can naturally be decreased by increasing the productivity and maximizing the recovery by efficient manufacturing process development. Automation and optimal raw material supply should theoretically decrease expense. Serum-free production systems would also be the preferred economic choice. Stable cells lines overcome the need for transfection reagents. More supplier and technology options bring increased competition, which has a positive influence on costs. Finding a balance between the optimal batch size and cycle times without forgetting about the risk factors can also create cost reductions. Ultimately, well-known Lean philosophy could bring a significant financial value . This means that cost saving could be achieved by optimizing the procedures and manufacturing processes and making them as simple and straightforward as possible, while at the same time systematically minimizing all the waste from the process. These cost-related issues should always be considered early on in the product and process development pipeline.
LVVs can be manufactured today at a clinical scale, but significant effort is still needed to increase the yield and to make the production more cost effective. This will need deep collaboration between the science field and service providers, as manufacturing needs new technologies for the production. A future challenge is also the availability of production facilities. Not every company has the financial capability or knowhow to build its own production suites and so rely on contract manufacturing organizations, booking for production slots months ahead. Success stories have been published showing the technologies are there and the knowhow is there. The industry challenge is to put more effort in to optimizing manufacturing technologies in the future, step by step to get LVV products to phase III and beyond. This may be achievable with existing systems and technologies, but science is often not predictable and who knows what breakthroughs will emerge and radically change the paradigm for LVV production.
financial & competing interests disclosure
The author has no relevant financial involvement with an organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock options or ownership, expert testimony, grants or patents received or pending, or royalties. No writing assistance was utilized in the production of this manuscript.
1. Zufferey R, Dull T, Mandel RJ et al. Self-inactivating lentivirus vector for safe and efficient in vivo gene delivery. J. Virol. 1998; 72: 9873–9880.
2. Dull T, Zufferey R, Kelly M et al. A third-generation lentivirus vector with a conditional packaging system. J. Virol. 1998; 72: 8463–8471.
3. Merten O-W, Charrier S, Laroudie N et al. Large-Scale Manufacture and Characterization of a Lentiviral ector Produced for Clinical Ex Vivo Gene Therapy Application. Hum. Gene. Ther. . 2011; 22: 343–356. CrossRef
4. Kutner RH, Puthli S, Marino MP et al. Simplified production and concentration of HIV-1-based lentiviral vectors using HYPERFlask vessels and anion exchange membrane chromatography. BMC Biotechnol. 2009; 9: 1–7. CrossRef
5. Wu S-C, Huang GY-L, Liu J-H. Production of retrovirus and adenovirus vectors for Gene. Ther. apy: a comparative study using microcarrier and stationary cell culture. Biotechnol. Prog. 2002; 18: 617–22. CrossRef
7. Lesch HP, Heikkilä KM, Lipponen EM et al. Process Development of Adenoviral Vector Production in Fixed Bed Bioreactor: From Bench to Commercial Scale. Hum. Gene. Ther. . 2015; 26: 560–571. CrossRef
8. Feldman SA, Goff SL, Xu H et al. Rapid Production of Clinical-Grade Gammaretroviral Vectors in Expanded Surface Roller Bottles Using a “Modified” Step-Filtration Process for Clearance of Packaging Cells. Hum. Gene. Ther. 2011; 22: 107–115. CrossRef
9. Powers AD, Piras BA, Clark RK et al. Development and Optimization of AAV hFIX Particles by Transient Transfection in an iCELLis ® Fixed-Bed Bioreactor. Hum. Gene. Ther. Methods 2016; 27: 112–121. CrossRef
11. Gstraunthaler G. Alternatives to the use of fetal bovine serum: serum-free cell culture. ALTEX 2003; 20: 275–281.
14. Guerin-Peyrou G, Vjetrovic J, Cuzange A et al. PEIpro®: Optimized PEI-based transfection reagents for production of clinical grade viral vectors. In: ESACT Conference. 2017.
15. Segura MM, Garnier A, Durocher Y et al. Production of lentiviral vectors by large-scale transient transfection of suspension cultures and affinity chromatography purification. Biotechnol. Bioeng. 2007; 98: 789–799. CrossRef
16. Ansorge S, Lanthier S, Transfiguracion J et al. Development of a scalable process for high-yield lentiviral vector production by transient transfection of HEK293 suspension cultures. J. Gene. Med. 2009; 11: 868–876. CrossRef
21. Miskin J, Biomedica O, Annual A et al. Lenti Production , Scale-up and Commercialisation Forward-looking statements and Disclosures. ASGCT 2017 Annu Meet.
22. Marceau N, Gasmi M. Scalable Lentiviral Vector Production System Compatible with Industrial Pharmaceutical Applications. 2013.
24. Guy HM, McCloskey L, Lye G et al. Characterization of lentiviral vector production using microwell suspension cultures of HEK293T-derived producer cells. Hum. Gene. Ther. Methods 2013; 24: 125–39. CrossRef
25. Miyoshi H, Blömer U, Takahashi M et al. Development of a self-inactivating lentivirus vector. J. Virol. 1998; 72: 8150–7.
26. Burns JC, Friedmann T, Driever W et al. Vesicular stomatitis virus G glycoprotein pseudotyped retroviral vectors: concentration to very high titer and efficient gene transfer into mammalian and nonmammalian cells. Proc. Natl Acad. Sci. 1993; 90: 8033–8037. CrossRef
37. Manceur AP, Kim H, Misic V et al. Scalable lentiviral vector production using stable HEK293SF producer cell lines. Hum. Gene. Ther. Methods 2017; 28:hgtb.2017.086.
38. Haselhorst D, Kaye J, Lever A. Development of cell lines stably expressing human immunodeficiency virus type 1 proteins for studies in encapsidation and gene transfer. J. Gen. Virol. 1998; 79 ( Pt 2):231–237. CrossRef
39. Sanber KS, Knight SB, Stephen SL et al. Construction of stable packaging cell lines for clinical lentiviral vector production. Sci Rep. 5 . Epub ahead of print 2015. DOI: 10.1038/srep09021. CrossRef
40. Marin V, Stornaiuolo A, Piovan C et al. RD-MolPack technology for the constitutive production of self-inactivating lentiviral vectors pseudotyped with the nontoxic RD114-TR envelope. Mol. Ther. – Methods Clin. Dev. 2016; 3: 16033. CrossRef
47. Dizon-Maspat J, Bourret J, D’Agostini A et al. Single pass tangential flow filtration to debottleneck downstream processing for therapeutic antibody production. Biotechnol. Bioeng. 2012; 109: 962–970. CrossRef
49. Baicu SC, Taylor MJ. Acid-base buffering in organ preservation solutions as a function of temperature: New parameters for comparing buffer capacity and efficiency. Cryobiology 2002; 45: 33–48. CrossRef
50. Amorij JP, Meulenaar J, Hinrichs WLJ et al. Rational design of an influenza subunit vaccine powder with sugar glass technology: Preventing conformational changes of haemagglutinin during freezing and freeze-drying. Vaccine 2007; 25: 6447–6457. CrossRef
52. Nilsson SM. Process development of lentiviral vector expression , purification and formulation for Gene Therapy applications. PhD thesis.
53. White M, Whittaker R, Stoll EA. A guide to approaching regulatory considerations for lentiviral-mediated Gene. Ther. apies. Hum. Gene. Ther. Methods 2017; 28: hum.2017.096.
54. EMA. Guideline on Development and Manufacture of Lentiviral vectors.
55. FDA/CBER. Testing of Retroviral Vector-Based Human Gene Therapy Products for Replication Competent Retrovirus During Product Manufacture and Patient Follow-Up – Draft Guidance for Industry. Fda.
57. Turkki V, Lesch HP, Martin R et al. Viral Vector Particle Integrity and Purity Analyses in Early Process Development. Bioprocess Int. 2012; 15: 34–37.
59. Zhao Y, Stepto H, Schneider CK. Development of the First WHO Lentiviral Vector Standard: Towards the Production Control and Standardisation of Lentivirus based Gene Therapy Products. Hum. Gene. Ther. Methods 2017; 28:hgtb.2017.078.
60. Dos H, Leite R, Vieira GE. Lean philosophy and its applications in the service industry: a review of the current knowledge. Production 2015; 529–541.
Dr Lesch is the Gene Therapy Unit Director of Kuopio Center for Gene and Cell Therapy (KCT) in Finland. Her research interest is focused on gene therapy and translational development, including early stage analyticals and the development of scalable, robust manufacturing processes operating under current regulatory guidelines. She also has research and development affiliations with Finvector and FKD Therapies. She has wide experience of all the most commonly used viral vectors (lentivirus, AAV, adenovirus and baculovirus) and a research interest in vascular biology and cancer therapy. Her PhD was in Molecular Medicine, obtained from the University of Kuopio. She followed up her PhD with post doc work at the University of California San Diego UCSD, CA, USA, and the University of Eastern Finland, Kuopio, Finland, concentrating on the understanding of gene regulation and the role of enhancer RNAs in macrophages. She has several patents related to vector manufacturing.