Abstract
The quantification of lentiviral vector (LVV) potency via titration is a critical quality control step for cell and gene therapies. However, standard functional titre assays are fundamentally limited by an inability to detect multiple integration events, procedural variations, and, most critically, a mass transport limitation created by the fluid overlay in conventional well plates. These issues, compounded by a lack of standardisation, lead to significant inter-laboratory variability and a systematic underestimation of true vector potency. In this study, we employed a microfluidic approach to create a more precisely engineered assay environment. We systematically evaluated channel depth, incubation time, vector concentration, and multiplicity of infection (MOI) for their impact on assay linearity, sensitivity, limit of detection, and reproducibility. A 0.2 mm deep channel provided linearity and reproducibility comparable to 96-well plates yet offered shorter incubation periods, and enhanced sensitivity, detecting activity down to a MOI of 0.0625 (corroborated by qPCR analysis) - a level at which conventional well plates fail. This establishes our microfluidic platform as a device-based analytical standard that transforms functional titre quantification from a variable protocol into a more reliable engineering solution for quality testing in cell therapy manufacturing.
1 Introduction
Cell and gene therapies (CGTs) have emerged as promising approaches for treating various neoplastic disorders, with Chimeric Antigen Receptor T (CAR-T) cells showing particularly strong efficacy in haematological malignancies such as Acute Lymphoblastic Leukaemia (ALL) (). These immunotherapies rely on the stable ex vivo modification of human T cells to express a CAR on their surface, enabling them to recognise a tumour-specific antigen (). Among the different delivery methods, lentiviral vectors (LVVs) are one of the most suitable due to their ability to stably integrate transgenes into both dividing and non-dividing cells (). As of 2022, LVV vectors accounted for up to 66% of all ex vivo gene modifications employed in UK clinical trials (). LVV demand is projected to increase significantly with 50–75 CGTs expected to be approved in the US alone by 2030 (). Moreover, recent reports have documented the use of LVV expanding beyond the CAR-T cell therapy field, with considerable evidence currently supporting their potential efficacy in the treatment of solid tumours or autoimmune diseases (; ). The success of these therapies is fundamentally dependent on the consistent production of high-quality vectors, necessitating robust analytical methods to define, monitor, and ensure their quality.
Analytical assays typically evaluate identity, potency, safety and purity of a lentiviral sample (). A particularly relevant measure of potency is the functional titre, which quantifies the number of lentiviral particles capable of integrating the gene of interest and driving transgene expression in target cells in vitro. Functional titre directly determines the vector dose required for successful CAR-T transduction, which often necessitates Multiplicity of Infection (MOI) values resulting in less than 5 vector copies per cell (). Recent functional titres are reported at approximately 2 × 107 TU mL-1 for transiently transfected systems () and 2 × 108 TU mL-1 (; ) for stable producer cell lines. Accurate, reproducible measurement of this parameter is therefore essential to ensure the quality of LVV products.
The standard method for quantifying LVV functional titre is a cell-based flow cytometry assay typically performed using HEK293 cells. Serial dilutions of the LVV sample are incubated with a known quantity of target cells. Following the expression of the protein of interest, which may require immunostaining if the transgene does not express a reporter protein, the fraction of positive cells is quantified by flow cytometry (). A key limitation of this assay is its inability to detect multiple integration events within a single cell, which can lead to an underestimation of the (true) functional titre. To address this underestimation issue, functional titre calculations are typically based on a transduction efficiency of ≤ 30%, assuming a Poisson distribution of infections (). Additionally, the total assay timeframe varies, with reports ranging from 72 h () to approximately 120 h (). This timeframe also includes an extended incubation of the LVV particles with the cells, widely ranging from 4 h () to 18 h () or 72 h () in the instance when fresh media is not supplemented to the cells post-transduction.
Fundamentally, the assay is also compromised by physical constraints understood for decades. Seminal work established that LVVs are only able to diffuse an effective distance of ∼600 µm before their half-life at 37 °C leads to degradation (; ). This physical reality is in direct conflict with the geometry of the standard assay environment: a microwell plate with a culture fluid overlay of 1–3 mm. The thickness of this overlay forces most viral particles to traverse distances where they are likely to decay before reaching a target cell. This challenge was long recognised in classical virology, where it was standard practice to “lay as thin a layer of a viral preparation as possible over the target cells for as long as practical” to minimise titre underestimation (), with data suggesting layers in the tenths of a millimetre (). When combined with the other issues, such as the inability to detect multiple integration events and variation in procedural standards, this results in a field where titres from different labs become difficult to compare, presenting a significant challenge to the development and quality control of CGT products (; ; ; ).
Microfluidics offers an approach to re-engineer the assay environment by precisely controlling the cellular microenvironment (; ; ), minimising the diffusion path length between vector and cell to a scale where transport is no longer limiting. This principle has been previously employed to enhance LVV gene transfer kinetics (), to explore cost-effective scalability for manufacturing (), and to intensify transduction of hard-to-transduce cells (). Critically, reported vector “utilisation efficiencies” greater than 100% in microchannels, a paradox that directly implicates that mass transport limitations in conventional well plates relate to the underestimation of vector potency. We hypothesised this paradox of “greater than 100% efficiency” points to an opportunity for analytics, where a microfluidic environment, which reveals the hidden potency masked by the diffusion limitations of standard plates, could be leveraged to create a more sensitive and accurate functional titre assay. In this contribution, we therefore systematically evaluate channel depth, viral vector concentration, and MOI, and rigorously analyse assay linearity, sensitivity, limit of detection, and reproducibility. Furthermore, we compare our flow cytometry results with qPCR to further validate a microfluidics-based analytical assay. Our results demonstrate that a microfluidic approach can create a functional titre assay that is not only more sensitive and economical but also more reproducible and standardised.
2 Materials and methods
2.1 Routine cell culture maintenance
Human Embryonic Kidney cells (HEK293T) were obtained from the UCL tissue culture facility and cultured in Dulbecco’s Modified Eagle Medium (DMEM) (Thermo Fisher Scientific, United Kingdom) supplemented with 10% Foetal Bovine Serum (FBS), 1% Penicillin/Streptomycin, and 1% GlutaMAX (Thermo Fisher Scientific, United Kingdom). Routine passaging was performed through the addition of trypsin-EDTA 0.25% solution (Thermo Fisher Scientific, United Kingdom). This media composition was used across all experiments involving HEK293T cell culture unless otherwise stated. Further cell line authentication was not performed.
2.2 Transient transfection using the third-generation lentiviral plasmid system
HEK293T cells were seeded in a T175 CellBind flask (Corning, United Kingdom) at 7.5 × 104 cells cm-2, and incubated for 16 h at 37 °C and 5% CO2. A proportion of 1.5 µg plasmid DNA per 1.0 ×106 cells seeded was used for transfection, summing up a total of 19.5 µg using an equal ratio of the 4 plasmids comprising a third-generation lentiviral plasmid system: pRSV-Rev, pLJM1-EGFP, pMDLg/pRRE, pMD2.G. The plasmids were a gift from Didier Trono (Addgene plasmids #12253, #19319, #12251, #12259) (). A transfection solution was made with a 1:1 volumetric ratio between PEIpro (Polyplus, United Kingdom) and the mixture of plasmid DNA diluted in OptiMEM (Thermo Fisher Scientific, United Kingdom). The transfection solution is added to the flask and incubated for 24 h at 37 °C and 5% CO2. A media change was performed after 24 h, using complete media supplemented with 5 mM sodium butyrate (Cambridge Bioscience, United Kingdom). The media collected at 48 h post-transfection was filtered through a 0.45 µm polyether sulfone (PES) syringe filter (Merck, Germany) and concentrated using the Lenti-X Concentrator reagent (Takara Bio, France) according to the manufacturer’s instructions. The LVV material was aliquoted and stored at −80 °C.
2.3 Functional titre quantification in 96-well plates
HEK293T cells were seeded in a flat-bottom 96-well plate (Thermo Fisher Scientific, United Kingdom) at 7.5 × 104 cells cm-2 and incubated for 24 h at 37 °C and 5% CO2. Each lentiviral dilution and a non-transfected control were measured in triplicates. After 24 h, the cell count was determined by trypsinization and quantification on the NucleoCounter NC-3000 (ChemoMetec, Denmark). The plate was then incubated with 50 µL lentiviral sample for 48 h at 37 °C and 5% CO2. The cells were harvested, fixed using 4% paraformaldehyde in Phosphate Buffered Saline (PBS) (Thermo Fisher Scientific, United Kingdom) and run on a LSRFortessa™ flow cytometer (BD Biosciences, United Kingdom) with a 10,000 events limit. The data was analysed using FlowJo v10.8.1, and the functional titre of the lentiviral vector material was calculated using Equation 1.
2.4 Polydimethylsiloxane (PDMS) casting and characterisation by optical profilometry
Microchannel moulds were designed in Fusion v2.0.20981 (Autodesk, United States), manufactured using CNC milling from 3 mm thick polycarbonate (PC) sheets. PDMS was cast into the PC mould using a SYLGARD™ 184 Silicone Elastomer Kit (Dow, United States) 10:1 w/w ratio of elastomer to curing agent preparation. The moulds were degassed for 30 min in a vacuum chamber, cured at 80 °C for 2 h, and then cut out using a scalpel. The channel depths (PC mould and corresponding cast PDMS layer) were measured with a 3D optical profilometer Contour GT-100 using a 5× objective and analysed in the Vision64 v5.41 software (Bruker, United States) through the Step Height function.
2.5 Chemical bonding of cyclo-olefin copolymer (COC) and PDMS layers
A 1 mm thick COC microscope slide (Microfluidic ChipShop, Germany) was treated by a corona plasma treater (Elveflow, France) for 2 min, followed by incubation in a 2% (3-Aminopropyl)-triethoxysilane (APTES) (Merck, Germany) in PBS (Thermo Fisher Scientific, United Kingdom) solution for 20 min at room temperature. The slide was then washed in distilled water (Thermo Fisher Scientific, United Kingdom) and dried by air stream. The surface modified microscope slide and the PDMS layer forming the microfluidic channels were treated by a corona plasma treater (Elveflow, France) for 2 min before being manually placed in contact and incubated at 80 °C for 16 h to reinforce the bonding.
2.6 Validation of device sterility
All microchannels were submerged in distilled water (Thermo Fisher Scientific, United Kingdom) and sterilised by autoclaving at 121 °C for 20 min. The PDMS and COC parts were separated and incubated overnight at 37 °C in LB Broth (Thermo Fisher Scientific, United Kingdom) alongside non-sterilised counterparts and a negative control. The supernatant was then analysed by spectrophotometry at 600 nm to verify the extent of bacterial growth in each culture using a Jenway UV/Visible Single Beam Spectrophotometer (Fisher Scientific, United Kingdom). An additional sterility validation experiment was performed by incubating the PDMS and COC parts for 14 days at 35 °C in Fluid Thioglycolate Medium (FTM) (Fisher Scientific, United Kingdom) according to the United States Pharmacopeia (USP) guidelines. The results were evaluated through visual observation of the culture media.
2.7 Routine cell culture maintenance and lentiviral transduction in microchannel format
HEK293T cells were seeded at a density of 7.5 × 104 cells cm-2 and media changes performed every 24 h. The cell viability and total cell count were obtained by harvesting cells at specified time intervals, and quantified using the Viability and Count protocol on the Nucleocounter NC-3000 (ChemoMetec, Denmark). Microscopy was performed using an EVOS M5000 microscope (Thermo Fisher Scientific, United Kingdom) for cell morphology observation. Lentiviral vector material diluted in complete media was added at 24 h post cell seeding at the concentration corresponding to the specified MOI. Media changes are performed at 24 h post-transduction, unless otherwise stated. The cells are then harvested at 48 h post-transduction and analysed via flow cytometry using the procedure stated in 2.3. To assess assay reproducibility, the transduction efficiency obtained across separate experimental occasions was analysed using the coefficient of variation (CV) (Equation 2). For linearity evaluation, the transduction efficiency data was normalised using Equation 3.
2.8 Quantification of vector genome integration
Cells from microchannels or from 96-well plates were collected by centrifugation at 300 RCF for 5 min, washed once in sterile PBS, and the cell pellets stored at −80 °C until further sample processing. Genomic DNA was isolated from cell pellets according to an automated protocol, using a QIAamp 96 DNA kit (Qiagen, Germany) with a QIAcube HT (Qiagen, Germany). The elution was then stored at −20 °C until further processing. The number of vector copies per cell was identified by quantifying the vector packaging signal (Ψ) in extracted DNA samples and normalising this value to the number of copies of the housekeeping gene, ribonuclease P RNA component H1 (RPPH1), via duplex qPCR. Both sequences were quantified by interpolation from a corresponding 5-points DNA standard curve with known number of target copies. The 96-well qPCR plates (Thermo Fisher Scientific, United Kingdom) were prepared by combining the PCR Master Mix (Thermo Fisher Scientific, United Kingdom) with genomic DNA and corresponding primers as well as generating the standard curves using a Hamilton STAR Automated Liquid Handler (Hamilton, United States). Corresponding oligonucleotides were added for each target sequence, the HIV-1 Ψ and RPPH1 copies being quantified using specific primers, showed in Table 1. Data was generated and analysed using a QuantStudio 7 Pro Real-Time PCR System (Thermo Fisher Scientific, United Kingdom). Assay controls included a non-template control (NTC) to ensure the lack of template DNA contamination during the extraction procedure.
TABLE 1
| Target | Sequence type | Sequence |
|---|---|---|
| A region of the integrated HIV-1 packaging signal (Ψ) | Forward primer | 5′ TGGGCAAGCAGGGAGCTA 3′ |
| Reverse primer | 5′ TCCTGTCTGAAGGGATGGTTGT 3′ | |
| Probe | 5′ FAM-AACGATTCGCAGTTAATCCTGGCCTGTT-TAMRA 3′ | |
| RPPH1 | Forward primer | 5′ CCCTAGTCTCAGACCTTCCCAAG 3′ |
| Reverse primer | 5′ GCGGAGGGAAGCTCATCAG 3′ | |
| Probe | 5′ VIC-CCACGAGCTGAGTGCGTCCTGTCATAMRA 3′ |
Primer and probe sequences used for qPCR against the HIV-1 Ψ and RPPH1 targets.
2.9 Statistical analysis
All data is presented as individual data points with their mean ± one standard deviation. All statistical analysis was performed using GraphPad Prism v8.4.2 (GraphPad, United States). According to the experiment, either a paired or un-paired T-test was used or a one-way ANOVA with either Dunnett’s or Tukey’s post hoc analysis with a 95% confidence interval and statistical significance reported as *: p < 0.05; **: p < 0.01; ***: p < 0.001; ****: p < 0.0001.
2.10 Equations
Formula for calculating the functional titre, where Ncells is total number of cells, P is the percentage of GFP positive cells, V is the sample volume, D is the LVV dilution factor (Equation 1).
Formula for calculating the coefficient of variation (CV %), where SD is the standard deviation and the mean is the average of triplicate measurements (Equation 2).
Normalisation of transduction efficiency using the Poisson distribution, where P is the percentage of GFP positive cells (Equation 3).
3 Results
The observation of “greater than 100% utilization efficiency” in microfluidic transduction systems () not only suggested that standard titration assays are failing to capture the full potency of lentiviral vectors but also pointed to a clear opportunity to use microfluidics to develop a more insightful analytical tool. We therefore explored whether a microfluidic environment, which minimises the diffusion path between vector and target cell, could be engineered to create a more sensitive and accurate–and ultimately more standardised - functional titre assay. To test this, we systematically characterized the impact of channel depth, incubation time, and vector concentration on key assay performance metrics.
3.1 Microchannel fabrication and characterization by optical profilometry
A set of microfluidic channels of 0.2-, 0.6- and 1.0-mm depth was manufactured using polydimethylsiloxane (PDMS) casting and chemical annealing onto a cyclo-olefin co-polymer (COC) substrate (Figure 1A). To allow for a direct comparison in the case of an adherent cell line transduction, the surface area of the microchannels was designed to be similar to a 96-well plate well, measuring approximately 0.32 cm2 (). The enclosed structures of microfluidic channels offered reduced fluid volumes (relative to 96-well plates), which helped to mitigate media evaporation over long incubation periods at 37 °C. Consequently, the operating volumes corresponding to the 3 channel sizes were approximately 42, 25 and 8 μL, as derived from the dimensions of the Computer Aided Design (CAD) design. The minimum channel depth of 0.2 mm was selected to facilitate accurate culture operations through pipetting while minimizing the impact of media evaporation on the culture characteristics. Smaller channel depths were avoided due to the increased pressure required for media displacement and propensity for channel deformation or collapse during operation.
FIGURE 1
Polydimethylsiloxane (PDMS) is a commonly used substrate within microfluidic cell culture applications, with desirable properties such optical transparency and high capacity for gas exchange (). A chemical surface modification approach was employed to seal the PDMS structure onto a COC microscope slide. This modification involved air plasma activation followed by surface treatment using APTES solution. Similar approaches have been previously reported (; ), although they have not been validated for cell culture applications. Due to the nature of the manually operated application, the system was expected to be under negligible pressure. The efficiency of the chemical bonding process was confirmed through visual observation and the absence of leakage during the typical fluidic operation and incubation over approximately 96-h. In addition, in the absence of APTES treatment, HEK293T cells have the capacity to attach and grow on both untreated COC and COC treated with plasma ().
Overall, the assembly was compatible with monitoring through phase contrast and fluorescence microscopy due to its favourable optical transparency, low autofluorescence and conventional dimensions of a microscope slide (). In addition, COC was identified as suitable substrate due to its biocompatibility () and capacity to support the adherence and maintenance of HEK293T cells in culture (). Furthermore, both substrates can withstand autoclaving conditions (steam at 15 psi, 121 °C, 20 min), therefore being suitable for cell-based applications. A simple architecture enabled a cell culture region placed between the 2 inlet ports, as detailed in Figure 1B. The aim of this device architecture was to provide a simple method for probing the reliance of transduction efficiency upon a series of cell culture fluid overlays. In addition, this design facilitated high-throughput device production, resulting in single-use cell culture structures. This approach allowed for generating microchannels of various depths in an inexpensive, rapid and high-throughput manner, relying on a polycarbonate (PC) mould design obtained through Computer Numerical Control (CNC) milling. The PC mould as well as the corresponding PDMS channels were characterized using optical profilometry to evaluate the channel depth obtained for each of the 3 selected instances.
Figure 1C compares the depths of the PC mould and PDMS channels with the expected depth according to the CAD. As expected, the results indicated no statistically significant difference (p > 0.05, unpaired t-test) between the measured depths of the PC mould and the corresponding PDMS channels for all three depths tested ().
3.2 Validation of device sterility
Microchannel sterility prior to cell seeding is essential for successful cell culture experiments. PDMS and COC were the two materials comprising the microchannel and were therefore in direct contact with the cell culture. Therefore, to evaluate the suitability of autoclaving and validate the sterility of the device, we first compared the levels of bioburden present on both autoclaved and unsterilised device parts. The results demonstrated that both substrates were sterile after one autoclave cycle (Figure 2). Negative (LB media incubated in the absence of any device parts) and positive (LB media containing device parts which have not been autoclaved) control experiments supported this conclusion. The differences between the autoclaved and non-sterilized parts was statistically significant (p < 0.0001, one-way ANOVA) To further validate device sterility, we incubated the device parts for 14 days at 35 °C in Fluid Thioglycolate Medium (FTM) according to US Pharmacopeia (USP) guidance. Visual observation confirmed no growth in the cultures (Supplementary Figure S1).
FIGURE 2
3.3 Lentiviral transduction performed in microchannel format
Initial cell culture experiments were conducted to assess the suitability of the microchannels with various culture fluid overlays for maintaining HEK293T cells over 72 h. All 3 channel depths kept both total cell count (Figure 3A) and viability (Figure 3B) to a level comparable to 96-well plates for up to 48 h (p > 0.05, one-way ANOVA, Tukey’s post hoc analysis). No detrimental effects were observed on the cell morphology, suggesting suitability for routine cell maintenance and cell-based assays (Figure 3C).
FIGURE 3
Accurate functional titre quantification requires a consistent cell count at the time of viral addition. We confirmed no statistically significant difference for 48 h after cell seeding (Figure 3A, p > 0.05, one-way ANOVA, Tukey’s post hoc analysis). In addition, there was no discrepancy in cell viability when the 1 mm microchannel was compared with the plate (p > 0.05, one-way ANOVA, Tukey’s post hoc analysis. However, there was a statistically significant decrease in cell viability and count when comparing the 0.2 mm microchannel to the 96-well plate well at 72 h after cell seeding (p < 0.05 and p < 0.01, one-way ANOVA, Tukey’s post hoc analysis). This might point an insufficient rate of media exchange, which was performed manually using pipettes. The decrease was more pronounced for the shallower microchannels. However, since cells are harvested for flow cytometry 72 h post-seeding in a typical assay protocol, the microchannels were considered suitable for performing the cell-based functional quantification assay up until this time-point.
Due to the varying volumes (42, 25 and 8 µL) in the microchannels, adding the same number of viral vector particles would result in different LVV concentrations; smaller culture fluid overlays, for example, would yield higher LVV concentration. Therefore, when conducting lentiviral transductions in microchannels of different depths (hence, volumes), we either kept the multiplicity of infection (MOI) or the LVV concentration constant.
Analysing first the impact of different culture fluid overlays in the microchannels, the results in Figure 4A showed a statistically significant increase between the 1- and 0.2-mm channels when transduction is performed at constant MOI (p < 0.05, one-way ANOVA, Dunnett’s post hoc analysis). Figure 4B indicates that when the LVV concentration was kept constant at 6.3 × 105 TU mL-1, the transduction efficiency correlated positively with the channel. In particular, there was a statistically significant decrease between the 1 and 0.2 mm microchannels when the LVV concentration was maintained constant (p < 0.05, one-way ANOVA, Dunnett’s post hoc analysis). Conversely, the percentage of GFP-positive cells correlated negatively with the channel depth when the transduction was performed at a constant MOI of 2. Consequently, this suggests that the total amount of particles loaded into a microchannel has a greater impact on the transduction efficiency than the concentration of LVV particles for the samples tested.
FIGURE 4
When evaluating a range of MOI values and comparing the microchannel results with the 96-well plates, a statistically significant increase in transduction efficiency in 0.2 mm channels was observed compared with 96-well plates, for all tested MOIs (0.5–4), and with p-values ranging from <0.001 to <0.05 (Figure 4C). The increase was approximately 2-fold in terms of the fraction of GFP-positive cells is observed across all MOI values tested relative to the 96-well plate instance. (MOI 0.5: 13% relative to 29%, representing an approximately 2-fold increase; MOI 1: 25% relative to 46%, 2-fold increase; MOI 2: 46% relative to 64%, 1.5-fold increase and MOI 4: 1.3-fold increase). In contrast, the 0.6 mm depth shows an increase in transduction efficiency only at high MOI values of 4. Similarly, the transduction performed at a fluid overlay of 1 mm shows relatively reduced transduction levels to the 96 well plate when tested at MOIs above 1 (p < 0.05, p < 0.01, one-way ANOVA, Dunnett’s post hoc analysis). This indicates that a significant increase in transduction efficiency across a range of MOIs can be obtained when constraining the interaction between lentiviral particles and target cell to a greater extent. By spatially constraining transduction events to depths equal to or smaller than 0.2 mm, a transduction efficiency of ≤20–30% () can be obtained from a 50% reduction in the number of particles. In the context of a cell-based functional titre quantification assay, this suggests an improvement in assay sensitivity of up to 50% across the 4 MOI values tested. This enables the detection of transduction efficiency and quantification of functional titre from samples with reduced LVV content (Supplementary Table S1).
To exclude that the microchannel data represent microfluidic artefacts, we performed LVV transductions in a 1.5 mm microchannel in parallel to 96-well plates with 50 µL volume per well; this volume corresponds to a fluid overlay of ∼1.5 mm. The results showed no statistically significant difference in transduction efficiency (p > 0.05, one-way ANOVA, Dunnett’s post hoc analysis) across a range of MOIs (0.5–4) (Figure 4D). The results from microchannels can therefore be directly compared with those from conventional well plates.
3.4 Evaluating assay linearity, reproducibility and effect of incubation time
Assay linearity was evaluated across a range of MOI values for different culture fluid overlays (96-well plate and microchannels). The transduction efficiency was normalized by using Poisson regression and fitted with a simple linear model. The goodness of fit, indicated by the R2 value, for a 96-well plate (R2 = 0.9593) was comparable to the transductions in either 0.6 mm (R2 = 0.9270) and 0.2 mm (R2 = 0.9798) microchannels (Figure 5). A less favourable R2 was identified for the 1 mm channel (R2 = 0.8721), suggesting that potentially a spatial constraint to a greater extent is desirable in this case. Overall, the increase in normalised transduction efficiency followed a linear pattern across MOI values ranging from 0.5 to 4, indicating the suitability of microchannels for performing a functional titre quantification assay with a 0.2 mm culture fluid overlay representing the best performer.
FIGURE 5
The reproducibility of obtaining an increase in transduction efficiency across a series of culture fluid overlays was investigated by conducting 3 independent experimental occasions at constant MOI (Figure 5B). The 0.6 and 0.2 mm microchannels show a low coefficient of variation (CV) of 3.4% and 3.9%, respectively, indicating higher reproducibility. In contrast, the 1 mm channel had a CV of 11.0%. Though this CV is significantly higher than for other microchannels, the value still falls below the typically acceptable limit of 20% for cell-based assays (). An increase in CV could be expected due to the inherent nature of cell-based assays, reflecting biological variation between experimental occasions performed on different days. In addition, the CV can be impacted by a series of external factors, including but not limited to operator error, imperfect microchannel microfabrication, bonding and operation or the impact of evaporation on the cell culture. However, this could represent an indication that by spatially constraining the interaction between target cells and viral particles to a greater extent, a potentially improved inter-assay precision can be achieved.
The effect of channel depth on transduction efficiency was evaluated for three different incubation times, representing strictly the variation of the ‘contact time’ between the LVV and target cells. As soon as the incubation time had elapsed, the LVV solution in the microchannel was replaced with fresh media. To allow for the expression of the gene of interest, the cells were cultured for at least an additional 48 h, followed by the quantification of transgene-positive cells by flow cytometry. For this experiment, the efficiency is evaluated across a wide range of culture fluid overlays in microchannels with depths of 0.2, 0.6 and 1 mm and in a 96-well plate with volumes of 50, 100 and 150 µL (corresponding to overlays of approximately 1.5, 3 and 4.5 mm, respectively) (Figure 5C). Transduction efficiency was higher in microchannels compared to a 96-well plate, across all incubation times. Notably, there appears to be an increase in transduction efficiency when comparing the 0.2 mm culture fluid overlay relative to the 96-well plate for a 4 h incubation. These results suggest the potential for a reduction in the overall time frame required to measure the transduction efficiency when quantifying the functional titre of an LVV sample. However, this would be applicable exclusively to the incubation time between LVV particles and target cells, while the time required to allow for gene expression remains constant.
3.5 Comparing assay sensitivity with a standard 96-well plate format
The results obtained thus far strongly suggest that minimising diffusion path lengths between adherent target cells and lentiviral particles increases transduction efficiency. We hypothesised that the enhanced proximity potentially also promoted a higher number of vector integrations per cell. The latter is crucial for analytical development as this would mean underestimating functional titre. While this effect can be corrected by accounting for the Poisson distribution (), exploring this hypothesis was relevant to further characterise microchannel-based assays. In addition, clinical cell therapy programmes aim for fewer than 5 vector copies per cell, ideally <1 or 2, as recommended by regulatory guidelines, to avoid risks like insertional mutagenesis and cell population heterogeneity ().
The quantification of the vector copies per cell by qPCR is considered a more sensitive measure of transduction relative to determining the fraction of transgene-positive cells by flow cytometry (). Therefore, to properly assess the sensitivity of the microchannel format, both methods were used to measure transduction levels at very low MOIs (between 0.5 and 0.0625). With a constant cell seeding density of 7.5 × 104 cells cm-2 in both microchannel and well plate, the cells had to be pooled from two consecutive transduction experiments to ensure sufficient cell material for both analytical methods. Consequently, the inter-assay variability introduced by performing the transductions on two separate occasions may confound the direct comparison of the two quantification methods.
The number of vector integrations was quantified by measuring the vector packaging signal in extracted genomic DNA samples from microchannels at various MOI values, using a plate control as reference. As anticipated, due to the low MOI values employed in this experiment, the number of integrations per cell was very low, with all values below 1 copy per cell (Figure 6A). This finding aligned with the flow cytometry data (Figure 6B), with the highest value of 20% GFP-positive cells for the highest MOI in the thinnest culture fluid overlay. Together, these results indicate reduced transduction levels across the cell population due to the low MOIs tested.
FIGURE 6
The microchannels with a 0.2 mm overlay demonstrated a higher level of transduction efficiency and vector genome copies per cell, indicating enhanced sensitivity to detect minimal amounts of viral particles in the sample to be analysed, with MOIs as low as 0.0625. The 0.2 mm microchannel showed a statistically significant increase in transduction compared to the plate control, for both qPCR (p < 0.01, one-way ANOVA, Dunnett’s post hoc analysis) and flow cytometry (p < 0.05, one-way ANOVA, Dunnett’s post hoc analysis). This trend was observed across all MOIs when using a culture overlay of 0.2 mm (Figures 6A,B). Given that lower MOIs result in fewer particles targeting the adherent cell monolayer, a higher degree of spatial constraint may be necessary to achieve a statistically significant difference in transduction efficiency. Together, the results indicate that the 0.2 mm microchannel format has the highest sensitivity and the only case detecting MOIs as low as 0.0625.
4 Discussion
We have successfully developed and validated a novel microfluidic assay to quantify the functional titre of a lentiviral vector (LVV). Our devices, composed of chemically annealed PDMS on COC substrates (Figure 1), were confirmed to be sterile by autoclaving via a bioburden assay and a 14-day USP-compliant sterility test (Figure 2; Supplementary Figure S1). A systematic investigation of channel depths (0.2, 0.6, and 1.0 mm) confirmed that channel depth–and therefore mass transport - is a critical parameter of transduction performance (; ; ). At a constant multiplicity of infection (MOI), shallower channels significantly increased transduction efficiency (Figure 4A). Conversely, at constant LVV concentration, shallower channels significantly decreased transduction efficiency (Figure 4B) These effects were most pronounced for the 0.2 mm channel. When evaluated across a standard MOI range (0.5–4) and benchmarked against well plates, only the 0.2 mm channel provided a consistently higher transduction efficiency while maintaining a comparable reproducibility (Figure 4C) At low MOI values, the transduction efficiency increase amounted to approximately two-fold, pointing to a substantially higher sensitivity of the microfluidic assay.
To further validate the microfluidic approach, we evaluated the linearity of the assay for each channel depth, by analysing the normalised transduction efficiency (across MOIs from 0.5 to 4.) The 0.2 mm channel showed the strongest linearity (R2 = 0.9798), comparable to that of well plates (Figure 5A). The 0.2 mm channel furthermore had a coefficient of variation of (3.9%) from three technical replicates (three separately fabricated channels) within a single assay run (MOI = 1), highlighting the potential for improved inter-assay precision with greater spatial constraint (Figure 5B). As an additional operational advantage, the functional titre assay in 0.2 mm channels also provided a reduction in incubation time to just 4 h (Figure 5C).
Furthermore, the 0.2 mm channel exhibited the highest sensitivity, determined by two complimentary experiments: Firstly, it produced an approximate two-fold increase in transgene-positive cells across the standard MOI range (0.5–4) (Figure 4C). Secondly, the 0.2 mm channel was the only configuration–well plates included–to reliably detect transduction at very low MOIs (down to 0.0625) (Figure 6). This enhanced low-level detection was corroborated by qPCR analysis of vector genome integration, which confirmed the 0.2 mm channel was uniquely capable of detecting a signal at this limit. Collectively, these findings established the 0.2 mm microchannel as a new performance benchmark, delivering superior sensitivity alongside assay reproducibility and linearity.
The increased sensitivity is particularly attractive for in-process testing during process development steps, where detecting low levels of LVV in samples with small amounts of viral vector content is essential. An enhanced assay sensitivity is also advantageous in cell line development, for the generation of stable LVV producers, as it enables sequential high throughput screening steps at lower titres, typical for operation at small scale.
Our study further highlighted an opportunity for assay standardisation. The 1.5 mm microchannel and the 50 µL well plate performed identically (Figure 4D). This result extends beyond serving as a control: it not only confirms that the substantial variations observed in TE across different channel depths are not a microfluidic artefact, nor simply validates that comparable results can be achieved when identical workflows are used in both microfluidic or well plate format. Rather, it highlights that a microfluidic approach enables controlled and systematic variation of the culture environment to identify an improved assay standard.
That an optimal culture fluid overlay thickness lies within a few tenths of a millimetre is consistent with both our findings and historical virology studies. Early investigations of host-virus, such as plaque assays, employed very thin fluid overlays; published protocols for plaque assays employed fluid overlays on the order of a 10th of a millimetre () to minimise titre underestimation (). For larger overlays, previous reports pointed out that the effective diffusion distance for LVV particles at 37 °C is approximately 0.6 mm before substantial decrease in particle half-life occurs (). This aligns with our results: we observed higher transduction efficiency and better reproducibility in 0.2 mm channels compared with 0.6 mm. Even thicker culture fluid overlays, such as in well plates, further exacerbate titre underestimation by reducing the likelihood of productive virus-cell interactions (). There is also likely a lower limit. reported a decline in transduction efficiency below approximately 0.1 mm, i.e., when the volume dropped below 3 μL at a given MOI. Our own results indicate challenges of maintaining cell viability in very thin microchannels for longer than 72 h (Figure 3). Reduced overlay heights limits nutrient availability for the cells, creating a practical limit for very thin microchannels.
To fully standardise the assay, additional factors must be considered. In line with our findings (Figure 5C), variations in the incubation time between LVV particles and cells substantially impact TE, leading to discrepancies when determining functional titre. Beyond this, cell-based functional assays inherently reflect the biological variability of cell cultures (; ). Variability is further compounded by a lack of standardisation across LVV titration methodologies. Differences in target cell type, incubation duration, MOI selection, and gating strategies used during downstream analysis can all strongly affect the resulting functional titre. Together, all these effects have historically limited direct comparisons across studies conducted under different experimental conditions (; ; ; ).
Current regulatory guidance similarly states that infectious titre determination should be product-specific and reflective of the mechanism of action, without prescribing a standardised assay format or experimental conditions (). While this regulatory framework allows for flexibility, it concomitantly perpetuates methodological heterogeneity and limits inter-laboratory comparability. Consequently, the need for developing international reference standards for viral vectors has been widely acknowledged, with an LVV reference material made recently available ().
Further development of this approach could focus on workflow automation. The current microfluidic devices relied on extensive manual pipetting, and automating liquid handling would remove operator-dependent variability. In addition, replacing flow cytometry with fluorescence microscopy-based quantification of adherent cells would streamline readout and enable real-time monitoring (). Coupling this with AI-driven image analysis could further accelerate data processing and reduce assay burden (). Together, these advances would further enhance reproducibility and standardisation, thereby accelerating LVV development and improving inter-laboratory comparability.
In summary, we have successfully demonstrated that microfluidics can be leveraged to deliver a reliable analytical assay for functional titre quantification. By systematically characterising the thickness of the culture fluid overlay, i.e., the microchannel depth, we have identified this parameter as a key design variable for achieving a better-defined, more sensitive analytical system. The 0.2 mm channel provides linearity and reproducibility comparable to 96-well plates, whilst reducing incubation periods. Critically, the enhanced analytical sensitivity (limit of detection down to MOI 0.0625) enables robust detection of very low viral particle-to-cell ratios at which well plates fail. Thus, the microfluidic offers a more controlled and engineered microenvironment than conventional well plates, creating an opportunity to shift the functional titre assay from a protocol-based procedure to a device-based analytical standard.
Statements
Data availability statement
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.
Ethics statement
Ethical approval was not required for the studies on humans in accordance with the local legislation and institutional requirements because only commercially available established cell lines were used.
Author contributions
DF: Data curation, Writing – original draft, Investigation, Conceptualization, Visualization, Methodology, Formal Analysis. CM-K: Supervision, Conceptualization, Writing – review and editing, Project administration. PS: Investigation, Writing – review and editing. RR: Conceptualization, Writing – review and editing, Project administration, Supervision. MM: Project administration, Conceptualization, Supervision, Writing – review and editing. NS: Funding acquisition, Writing – review and editing, Project administration, Conceptualization, Supervision.
Funding
The author(s) declared that financial support was received for this work and/or its publication. The work was funded by a UKRI BBSRC and Oxford Biomedica (OXB)-funded Advanced Bioscience of Viral Products (ABViP) Collaborative Training Partnership (CTP), grant number: BB/W009420/1. ABViP is a doctoral training programme between Oxford Biomedica, University College London and the University of Oxford and includes financial and in-kind support from the UK university partners. Engineering and Physical Sciences Research Council (EPSRC) for the Future Targeted Healthcare Manufacturing Hub hosted at University College London with UK university partners (Grant Reference: EP/P006485/1) is also acknowledged.
Acknowledgments
The authors thank Jamie Evans (Division of Medicine, University College London) for flow cytometry training and guidance.
Conflict of interest
Authors CM-K and RR were employed by Oxford Biomedica (UK) Limited. NS is Managing Director of Young Owl Microfluidics, Ltd.
The remaining author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declared that generative AI was not used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
Supplementary material
The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fbioe.2026.1720882/full#supplementary-material
References
1
AghaA.WaheedW.AlamoodiN.MathewB.AlnaimatF.Abu-NadaE.et al (2022). A review of cyclic olefin copolymer applications in microfluidics and microdevices. Macromol. Mater. Eng.307 (8), 2200053. 10.1002/mame.202200053
2
AghaA.DawaymehF.AlamoodiN.AlazzamA. (2024). Enhancing fabrication of hybrid microfluidic devices through silane-based bonding: a focus on polydimethylsiloxane-cyclic olefin copolymer and PDMS-Lithium niobate. Appl. Res.3 (4), e202300116. 10.1002/appl.202300116
3
AhmadiS. E.ShabestariR. M.kojabadA. A.SafaM. (2023). A straightforward microfluidic-based approach toward optimizing transduction efficiency of HIV-1-derived lentiviral vectors in BCP-ALL cells. Biotechnol. Rep.38, e00792. 10.1016/j.btre.2023.e00792
4
AndreadisS.LaveryT.DavisH. E.Le DouxJ. M.YarmushM. L.MorganJ. R. (2000). Toward a more accurate quantitation of the activity of recombinant retroviruses: alternatives to titer and multiplicity of infection. J. Virology74, 1258–1266. 10.1128/jvi.74.3.1258-1266.2000
5
Aranda HernandezJ.HeuerC.BahnemannJ.SzitaN. (2022). “Microfluidic devices as process development tools for cellular therapy manufacturing,” in Microfluidics in biotechnology. Editors BahnemannJ.GrünbergerA. (Springer International Publishing), 101–127. 10.1007/10_2021_169
6
AsadianE.BahramianF.SiavashyS.MovahediS.KeçiliR.HussainC. M.et al (2024). “A review on recent advances of AI-integrated microfluidics for analytical and bioanalytical applications,” 181. TrAC - Trends Anal. Chem.10.1016/j.trac.2024.118004
7
BernardM.JubeliE.BakarJ.TortolanoL.SaunierJ.YagoubiN. (2017). Biocompatibility assessment of cyclic olefin copolymers: impact of two additives on cytotoxicity, oxidative stress, inflammatory reactions, and hemocompatibility. J. Biomed. Mater. Res. - Part A105 (12), 3333–3349. 10.1002/jbm.a.36199
8
BezeljakU. (2022). “Cancer gene therapy goes viral: viral vector platforms come of age,” 56. Radiology Oncol., 1–13. 10.2478/raon-2022-0002
9
BroussauS.LytvynV.SimoneauM.GuilbaultC.LeclercM.Nazemi-MoghaddamN.et al (2023). Generation of packaging cells for lentiviral vectors using the cumate and coumermycin gene induction systems and nanowell based single-cell cloning. Mol. Ther. - Methods and Clin. Dev.29, 40, 57. 10.1016/j.omtm.2023.02.013
10
CGT Catapult (2022). Cell and gene therapy catapult UK ATMP clinical trials report 2022.
11
ChoeS. H.KrofahE.ConlinA. (2022). “Cell and Gene therapies: looking ahead to 2022,” in Milken Institute. Editor INSTITUTE, M. Available online at: https://milkeninstitute.org/content-hub/research-and-reports/reports/cell-and-gene-therapies-looking-ahead-2022 (Accessed October 02, 2025).
12
ChuckA. S.PalssonB. O. (1996). Consistent and high rates of gene transfer can be obtained using flow-through transduction over a wide range of retroviral titers. Hum. Gene Ther.7, 743–750. 10.1089/hum.1996.7.6-743
13
ComiselR. M.KaraB.FiesserF. H.FaridS. S. (2021). Lentiviral vector bioprocess economics for cell and gene therapy commercialization. Biochem. Eng. J.167, 107868. 10.1016/j.bej.2020.107868
14
CorteseB.MowlemM. C.MorganH. (2011). Characterisation of an irreversible bonding process for COC-COC and COC-PDMS-COC sandwich structures and application to microvalves. Sensors Actuators, B Chem.160 (1), 1473–1480. 10.1016/j.snb.2011.07.040
15
DanL.Kang-ZhengL. (2025). “Optimizing viral transduction in immune cell therapy manufacturing: key process design considerations,” 23. J. Transl. Med.10.1186/s12967-025-06524-0
16
Department of Health and Human Services. (2011). Guidance for industry: potency tests for cellular and gene therapy products.
17
DuB.QinJ.LinB.ZhangJ.LiD.LiuM. (2025). CAR-T therapy in solid tumors. Cancer Cell, 43 (4) 665–679. 10.1016/j.ccell.2025.03.019
18
DulbeccoR.VogtM. (1954). Plaque formation and isolation of pure lines with poliomyelitis viruses. J. Exp. Med.99, 167–182. 10.1084/jem.99.2.167
19
DullT.ZuffereyR.KellyM.MandelR. J.NguyenM.TronoD.et al (1998). A third-generation lentivirus vector with a conditional packaging system. J. Virology72 (11), 8463–8471. 10.1128/JVI.72.11.8463-8471.1998
20
GeraertsM.WillemsS.BaekelandtV.DebyserZ.GijsbersR. (2006). Comparison of lentiviral vector titration methods. BMC Biotechnol.6, 34. 10.1186/1472-6750-6-34
21
Gouvarchin GhalehH. E.BolandianM.DorostkarR.JafariA.PourM. F. (2020). “Concise review on optimized methods in production and transduction of lentiviral vectors in order to facilitate immunotherapy and gene therapy,” 128. Biomed. Pharmacother., 110276. 10.1016/j.biopha.2020.110276
22
HanS.YouJ. Y.EomM.AhnS.ChoE.YoonY. (2024). From pixels to information: artificial intelligence in Fluorescence Microscopy. Adv. Photonics Res.5 (9), 2300308. 10.1002/adpr.202300308
23
HarveyJ. (2023). Using a microfluidic approach to test for critical quality attributes of viral vectors for gene therapy applications. [PhD thesis] London, United Kingdom: University College London. Available online at: https://discovery.ucl.ac.uk/id/eprint/10168866/ (Accessed March 2, 2026).
24
HigashikawaF.ChangL. J. (2001). Kinetic analyses of stability of simple and complex retroviral vectors. Virology280 (1), 124–131. 10.1006/viro.2000.0743
25
KneitzJ.RotterM.DekevicG.KäßerL.CzermakP.SalzigD. (2022). Analyzing single-use polymers for cell culture processes comparison of cell growth and viability Test procedures.
26
LabischJ. J.Philip WieseG.BarnesK.BollmannF.PflanzK. (2021). Infectious titer determination of lentiviral vectors using a temporal immunological real-time imaging approach. PLoS ONE16 (7 July), e0254739. 10.1371/journal.pone.0254739
27
Lin-GibsonS.SarkarS.ElliottJ.PlantA. (2016). Understanding and managing sources of variability in cell measurements. Cell Gene Ther. Insights2 (6), 663–673. 10.18609/cgti.2016.073
28
ManceurA.RodenbrockA.JacobD.PatelM.PuttaswamaiahS. G.Dorion-ThibaudeauJ.et al (2025). Large-scale production and purification of a lentiviral vector reference material. BioProcessing J.24. 10.12665/J24OA
29
McCarronA.DonnelleyM.McIntyreC.ParsonsD. (2016). “Challenges of up-scaling lentivirus production and processing,” 240. J. Biotechnol., 23–30. 10.1016/j.jbiotec.2016.10.016
30
MertenO. W.HebbenM.BovolentaC. (2016). Production of lentiviral vectors. Mol. Ther. Methods Clin. Dev. (Vol. 3, p. 16017). 10.1038/mtm.2016.17
31
MirandaI.SouzaA.SousaP.RibeiroJ.CastanheiraE. M. S.LimaR.et al (2022). Properties and applications of PDMS for biomedical engineering: a review. J. Funct. Biomaterials. 13 (1). 10.3390/jfb13010002
32
MooreN.ChevilletJ. R.HealeyL. J.McBrineC.DotyD.SantosJ.et al (2019). A microfluidic device to enhance viral transduction efficiency during manufacture of engineered cellular therapies. Sci. Rep.9 (1), 15101. 10.1038/s41598-019-50981-9
33
NaldiniL.BlomerU.GallayP.OryD.MulliganR.GageF. H.et al (1996). In vivo gene delivery and stable transduction of nondividing cells by a lentiviral vector. Science272, 263–267. 10.1126/science.272.5259.263
34
Nyberg-HoffmanC.ShabramP. L.LiW.GirouxD. (1997). Sensitivity and reproducibility in adenoviral infectious titer determination. Mol. Ther.3, 808–811. 10.1038/nm0797-808
35
PerryC.RayatA. C. M. E. (2021). Lentiviral vector bioprocessing. Viruses13 (2), 268. 10.3390/v13020268
36
PoorebrahimM.AbazariM. F.SadeghiS.MahmoudiR.KheirollahiA.AskariH.et al (2020). Genetically modified immune cells targeting tumor antigens. Pharmacol. Ther. (Vol. 214). 107603. 10.1016/j.pharmthera.2020.107603
37
RissT. L.MoravecR. A.DuellmanS. J.NilesA. L. (2021). Treating cells as reagents to design reproducible assays. SLAS Discov.26 (10), 1256–1267. 10.1177/24725552211039754
38
ShabramP.Aguilar-CordovaE. (2000). Multiplicity of infection/multiplicity of confusion. Mol. Ther.2 (5), 420–421. 10.1006/mthe.2000.0212
39
ShiR.JiaS.LiuH.NieH. (2022). “Clinical grade lentiviral vector purification and quality control requirements,” 45. J. Sep. Sci., 2093–2101. 10.1002/jssc.202100937
40
SinnP. L.SauterS. L.McCrayP. B. (2005). Gene therapy progress and prospects: development of improved lentiviral and retroviral vectors - design, biosafety, and production. Gene Ther.12, 1089–1098. 10.1038/sj.gt.3302570
41
Thermo Fisher Scientific (2006). Useful Numbers for Cell Culture. Available online at: https://www.thermofisher.com/uk/en/home/references/gibco-cell-culture-basics/cell-culture-protocols/cell-culture-useful-numbers.html?ef_id=CjwKCAjwpcTNBhA5EiwAdO1S9m5kLcUSjirQ7igAX9iXxVWMv69I8LYEpM2mrywFMGNlzKSWU-u57hoC6d4QAvD_BwE:G:s&s_kwcid=AL!3652!3!733018704372!p!!g!!useful%20numbers!1765073902!69284776499&cid=bid_clb_cce_r01_co_cp0000_pjt0000_bid00000_0se_gaw_nt_awa_awa&gad_source=1&gad_campaignid=1765073902&gbraid=0AAAAADxi_GR7rqt53oIsnyfqHP9WBHeqm&gclid=CjwKCAjwpcTNBhA5EiwAdO1S9m5kLcUSjirQ7igAX9iXxVWMv69I8LYEpM2mrywFMGNlzKSWU-u57hoC6d4QAvD_BwE(Accessed October 1, 2025).
42
TorinoS.CorradoB.IodiceM.CoppolaG. (2018). Pdms-based microfluidic devices for cell culture. MDPI Multidiscip. Digit. Publ. Inst.3 (Number 3), 65. 10.3390/inventions3030065
43
TranR.MyersD. R.DenningG.ShieldsJ. E.LytleA. M.AlrowaisH.et al (2017). Microfluidic transduction harnesses mass transport principles to enhance gene transfer efficiency. Mol. Ther.25 (10), 2372–2382. 10.1016/j.ymthe.2017.07.002
44
TridgettM.MuletM.JohnyS. P.AbabiM.RaghunathM.FustinoniC.et al (2024). Lentiviral vector packaging and producer cell lines yield titers equivalent to the industry-standard four-plasmid process. Mol. Ther. - Methods and Clin. Dev.32 (3), 101315. 10.1016/j.omtm.2024.101315
45
WagnerD. L.KoehlU.ChmielewskiM.ScheidC.StripeckeR. (2022). Review: sustainable clinical development of CAR-T cells - switching from viral transduction towards CRISPR-Cas gene editing. Front. Immunol.13. 10.3389/fimmu.2022.865424
46
Williams-FegredoT.DaviesL.KnevelmanC.MiskinJ.MitrophanousK.RafiqQ. A. (2024a). Auto-transduction in lentiviral vector bioprocessing: a quantitative assessment and a novel inhibition strategy. Biotechnol. Bioeng.121, 3728–3741. 10.1002/bit.28834
47
Williams-FegredoT.DaviesL.KnevelmanC.MitrophanousK.MiskinJ.RafiqQ. A. (2024b). Degradation of specific glycosaminoglycans improves transfection efficiency and vector production in transient lentiviral vector manufacturing processes. Front. Bioeng. Biotechnol.12, 1409203. 10.3389/fbioe.2024.1409203
48
WoodB.JevremovicD.BénéM. C.YanM.JacobsP.LitwinV. (2013). Validation of cell-based fluorescence assays: Practice guidelines from the ICSH and ICCS - Part v - performance criteria. Cytometry. Part B, Clinical Cytometry84(5), 315–323. 10.1002/cyto.b.21108
49
YinX.WeiH.GuoL.LiuB.PengY.ZhouM.et al (2025). Therapeutic effect of miR-30b-5p-loaded lentivirus on experimental autoimmune uveitis via inhibiting Notch signaling activation. J. Transl. Med.23 (1), 426. 10.1186/s12967-025-06438-x
50
ZaninelliS.MeliC.BorleriG.QuaroniM.PavoniC.GaipaG.et al (2024). Optimization and validation of in vivo flow cytometry chimeric antigen receptor T cell detection method using CD19his indirect staining. Cytom. Part A105 (2), 112–123. 10.1002/cyto.a.24796
Summary
Keywords
analytics, CAR-T cells, cell and gene therapies (CGT), functional titre, lentiviral vectors, microfluidics, multiplicity of infection (MOI), transduction
Citation
Farcas DA, Moore-Kelly C, Stevenson P, Raposo RAS, Marques MPC and Szita N (2026) Rethinking viral vector quantification: a microfluidic approach to standardised functional titre assays. Front. Bioeng. Biotechnol. 14:1720882. doi: 10.3389/fbioe.2026.1720882
Received
08 October 2025
Revised
06 March 2026
Accepted
13 March 2026
Published
13 April 2026
Volume
14 - 2026
Edited by
Volker Huppert, Glycostem Therapeutics B. V., Netherlands
Reviewed by
Wei-Xiang Sin, Singapore-MIT Alliance for Research and Technology (SMART), Singapore
Dirk Geerts, Amsterdam University Medical Center, Netherlands
Lightson Ngashangva, Rajiv Gandhi Centre for Biotechnology, India
Updates
Copyright
© 2026 Farcas, Moore-Kelly, Stevenson, Raposo, Marques and Szita.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Nicolas Szita, n.szita@ucl.ac.uk
Disclaimer
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.