CAR-T cell therapy faces well-documented manufacturing challenges that constrain clinical accessibility, scalability, and cost-effectiveness. This meta-analysis synthesises evidence from 74 peer-reviewed and preprint sources (2013–2026), of which 68 reported extractable quantitative results, to characterise the principal production bottlenecks and evaluate the degree to which each is addressable through automation. The evidence indicates that cell expansion capacity, manufacturing time, transduction efficiency, and batch-to-batch variability constitute the most clinically significant bottlenecks, and that all four are substantially addressable through currently available automation technologies. Stirred-tank bioreactors achieve 5–15-fold higher viable cell densities than static culture, and closed-system automated platforms (notably CliniMACS Prodigy) demonstrate >90% batch success rates with reduced contamination risk. Microfluidic approaches to cell isolation and transduction show early but compelling evidence for improving starting material quality and transgene delivery efficiency. A medium-severity debate persists over whether non-viral transduction methods are yet suitable for clinical-scale implementation, and whether decentralised point-of-care manufacturing is economically equivalent to centralised GMP facilities — questions that remain incompletely resolved in the current evidence base.
Key supporting observations:
- Automated closed-system manufacturing reduced batch failures threefold and costs by up to 45% versus manual open processing (Woods et al., 2025).
- Manufacturing time can be reduced from 10–14 days to 6–9 days through process optimisation and automation across multiple platforms.
- Microfluidic enrichment achieves 80–91% T-cell purity with ≥80% recovery, removing contaminant cells with minimal disruption to downstream expansion.
- Real-time, label-free quality-control assays requiring <10,000 cells and delivering results within 10 minutes are technically feasible and represent an emerging frontier for adaptive process control.
A systematic search was conducted across nine databases: Semantic Scholar, Crossref, PubMed, arXiv, OpenAlex, Europe PMC, CORE, DOAJ, and Wikipedia. The search covered publications from 2012 to 2026 using 14 query strings targeting automation, bioreactor systems, microfluidics, closed-system manufacturing, transduction, cryopreservation, and quality control in CAR-T cell production. The initial discovery pipeline identified 2,694 candidate papers. Following gate-level filtering, 66 papers passed the primary gate, 547 failed, and 127 were flagged for review. A total of 190 papers entered extraction, 133 underwent classification, and 74 papers received full credibility and evidence-grounding assessment.
Papers were included if they directly described or evaluated an automation technology, platform, or process for CAR-T cell manufacturing (including production, transduction, expansion, cryopreservation, or quality control); reported quantitative outcome data (viability, expansion fold-change, transduction efficiency, production time, cost per dose, yield, or cell count); and constituted a clinical study, preclinical validation, engineering study, or process development study involving CAR-T cells published in peer-reviewed or conference proceedings from 2012 onwards. Papers were excluded if they addressed solely clinical outcomes without a manufacturing component, described purely manual processes, provided only qualitative descriptions, were reviews or editorials without primary data, focused on non-CAR T-cell engineering as primary content, or investigated CAR design without manufacturing process automation.
| Pipeline Stage | Papers Entering | Papers Passing |
|---|---|---|
| Discovery (all queries) | 2,694 | — |
| Primary gate filter | 613 reviewed | 66 pass / 547 fail / 127 review |
| Data extraction | 190 | — |
| Classification | 133 | — |
| Credibility + evidence grounding | 74 | 74 (full analysis) |
| Papers with extractable quantitative results | 74 | 68 |
74 papers form the basis of this meta-analysis. Of these, 44 (59%) were rated medium credibility, 24 (32%) low credibility, and 6 (8%) uncertain; no papers attained the highest credibility tier, reflecting the dominance of small-cohort feasibility studies in this literature. The mean evidence grounding coverage across all papers was 61% (535 of 873 total claims grounded), with only 2 papers falling below 30% coverage.
Key Point: Automated bioreactor platforms produce substantially higher and more reproducible CAR-T cell yields than static culture, addressing one of the most clinically significant production bottlenecks.
Key Evidence:
"At agitation speeds of 200 rpm and greater (up to 500 rpm), the CAR-T cells are able to proliferate effectively, reaching viable cell densities of >5 × 10⁶ cells ml⁻¹ over 7 days." (Costariol et al., 2020)
"The authors successfully produced functional human CAR-T cells from multiple donors under dynamic conditions in a stirred-tank bioreactor, resulting in overall cell yields which were significantly better than in static T-flask culture." (Costariol et al., 2020)
Constant agitation at 75 RPM produced a 15-fold T-cell expansion, compared to 10-fold under variable agitation (Gatla et al., 2022). Perfusion approaches using alternating tangential flow maintained viability at 96.2% with zero cell loss to waste. The Quantum Flex hollow-fiber bioreactor achieved 150- to 200-fold expansion over 7 days from low seed numbers, with viability remaining >93% at harvest across all four donor replicates (Marshall et al., 2025). Wave bioreactor platforms similarly demonstrated faster proliferation and stronger cytotoxicity compared to static systems (Zhao et al., 2026), while the 2-litre perfusion stirred-tank reported by Springuel et al. (2025) achieved >90% product recovery via automated harvesting with a nine-fold volume reduction. A quality-by-design study found that a single activation step with a 3-day seed train yielded a 3-fold increase in cell yield and 30% reduction in exhaustion markers compared to a double-activation, 7-day seed train protocol (Hood et al., 2024).
Interpretation: The evidence across ≥5 papers consistently places bioreactor automation at the apex of addressable bottlenecks. The maturity of commercial platforms and the breadth of validated process parameters (agitation rate, pH, dissolved oxygen, perfusion mode) position automated stirred-tank and hollow-fiber systems as the most immediately scalable solutions for expansion constraints.
Key Point: Automated platforms and process optimisation reduce CAR-T manufacturing from a conventional 10–14 day baseline to as few as 6–9 days, with fresh-infusion protocols offering further reductions.
Key Evidence:
"This study reports a procedure for generating CD19 CAR-T cells in 6 days, using a functionally closed manufacturing process." (Lu et al., 2016)
"shortened duration of production from 12 to 8 days followed by fresh infusion into patients." (Jackson et al., 2020)
Malakhova et al. (2024) demonstrated that a 7-day manufacturing cycle on the CliniMACS Prodigy achieved equivalent cell expansion and functional properties to an 11-day cycle (median transduction efficiency 41.0% vs. 43.5%, not significantly different), supporting the feasibility of shortened timelines without quality loss. Castella et al. (2020) reported mean expansion times of 8.5 days (range 7–10 days) with a 96.4% product success rate (27/28 products). Transitioning from T-flasks to G-Rex bioreactors reduced operator hands-on time from 21–28 days to 14–17 days (Pisani et al., 2025).
Interpretation: The convergence of evidence across multiple platforms and manufacturing strategies demonstrates that the time bottleneck is tractable. The clinical significance is high: shorter vein-to-vein time is particularly critical for patients with aggressive or rapidly progressing disease.
Key Point: The quality and purity of the starting T-cell population profoundly affects transduction efficiency and expansion kinetics; automated microfluidic and magnetic enrichment approaches address this bottleneck at the leukapheresis processing stage.
Key Evidence:
"Following CD4/CD8 TCS, the CD3 T-cell content in the starting product increased from 47% to 90%... T cell proliferation increased from 1.97-fold to 24.2-fold." (Shah et al., 2017)
"In healthy donors used to optimize the process, the lymphocyte purity was enriched from 65% (SD ± 0.2) to 91% (SD ± 0.06)." (Elsemary et al., 2024)
Two-stage inertial microfluidic processing achieved 87% T-cell enrichment with 80% recovery from apheresis-like material, with no detectable effect on T-cell proliferation (Elsemary et al., 2026). For acute lymphoblastic leukaemia samples, the same platform achieved 80% B-cell depletion, 89% monocyte depletion, and 74% leukemic blast depletion (Elsemary et al., 2024). Deterministic lateral displacement (DLD) microfluidic processing yielded 88% leukocyte recovery versus 58% with Ficoll density gradient centrifugation, while also producing twofold greater central memory cells and reduced baseline T-cell activation (Skelley et al., 2025; Campos-González et al., 2018). Monocyte depletion via magnetic sorting reduced CD14+ cells from 9.3–32% to 0.1–1.1%, directly improving downstream activation efficiency (Noaks et al., 2021).
"In monocyte-depleted cultures, CD14+ cells underwent at least a 10-fold reduction, from 9.3%–32% to 0.1%–1.1%." (Noaks et al., 2021)
Interpretation: Starting material heterogeneity — particularly in heavily pretreated patients — is a frequently underappreciated bottleneck. Automated enrichment addresses contaminating cell populations that directly suppress T-cell activation and transduction, with downstream effects on yield and product quality that propagate through the entire manufacturing workflow.
Key Point: Traditional viral transduction methods yield wide efficiency ranges (8–77%) with significant process dependence; automated spinoculation, microfluidic platforms, and non-viral delivery methods all demonstrate quantifiable improvements.
Key Evidence:
"The MFD geometry allowed reaching a transduction rate of 27% (for MOI 3) compared to 17% and 8% transduction (MOI 3) in 48- and 6-well plates, respectively, used as controls." ("Scalable and ultrafast CAR-T..." 2025)
"Transduction efficiencies from Sepax prepared cells were not significantly different from the 2-h bag spinoculation for CD19/22 CAR T-cells transduction efficiency, with the Sepax spinoculation 1-h results being significantly better than the static transduction (83.4% vs. 35.7%)." (Remley et al., 2021)
Closed automated spinoculation achieved consistent, operator-independent results across three different lentiviral vectors and both healthy donor and patient samples (Remley et al., 2021). Non-viral hydroporation yielded 1.7- to 2-fold more CAR-T cells than electroporation and nucleofection, with superior viability and faster proliferation by day 5 (Sytsma et al., 2025; Sytsma et al., 2024a; Sytsma et al., 2024b). Multidimensional double spiral (MDDS) inertial microfluidic sorting removed up to 85% of unactivated T cells post-activation, resulting in an approximately 2-fold increase in CAR transduction efficiency in enriched populations (Jeon et al., 2024). Culture pH is an underappreciated automation-controllable variable:
"higher pH can significantly improve CAR T-cell transduction and proliferation." (Lamas et al., 2022)
Interpretation: The transduction step is technically addressable through multiple automation vectors — including bioreactor-controlled environmental parameters, automated spinoculation, microfluidic separation of activation-responsive cells, and non-viral delivery platforms. However, the clinical translation of microfluidic and non-viral approaches remains incomplete, and lentiviral integration continues to dominate GMP-approved manufacturing.
Key Point: Closed-system and fully automated manufacturing platforms dramatically reduce contamination risk, batch failures, and operator-dependent variability — all factors that directly affect regulatory compliance and patient safety.
Key Evidence:
"Transitioning to closed-system manufacturing and cryostorage has been shown to reduce batch failures threefold... manufacturing costs by up to 45%... tripling throughput and maintaining cell viability." (Woods et al., 2025)
"Independent of the starting material, operator, or device, the process consistently yielded a therapeutic dose of highly viable CAR T cells." (Lock et al., 2017)
The CliniMACS Prodigy platform across multiple independent academic sites consistently achieved >90% batch success rates in populations including heavily pretreated patients (Castella et al., 2020; Luanpitpong et al., 2024; Keyer et al., 2025). A global survey found that 73% of institutions cited variability in product quality and 70% cited cost constraints as barriers to local manufacturing (Gupta & Shaz, 2025). Automated manufacturing demonstrated cross-site consistency:
"We demonstrated consistency in cell composition and functionality of the products manufactured at two different production sites." (Aleksandrova et al., 2024)
Interpretation: Closed-system automation reduces the three dominant causes of batch failure — microbial contamination, operator error, and environmental variability — simultaneously. The regulatory advantage is substantial: GMP compliance is more reliably achieved and documented in automated versus manual workflows.
Key Point: Novel analytical tools requiring minimal cell numbers and delivering rapid results are technically feasible and could enable adaptive, closed-loop process control — a currently unmet but automatable need.
Key Evidence:
"the CTM assay requires fewer than 10,000 cells and delivers results within 10 minutes... correlates with CD4:CD8 ratio, memory subtypes, and cytotoxic activity... Validated across multiple donors and culture platforms." (Zeming et al., 2025)
"The use of 405-nm visible light to substitute for the genotoxic ultraviolet wavelengths for assessing NADH and FAD autofluorescence paves the way to incorporate autofluorescence measurements into closed and automated systems for in-process monitoring of CAR T-cell manufacturing." (Cheung et al., 2025)
Capacitance sensing achieved real-time viable cell concentration monitoring with an R² correlation of 0.98 to offline measurements in a 2-litre stirred-tank bioreactor (Springuel et al., 2025). An automated aseptic micro-sampling system (Auto-CeSS) demonstrated insignificant differences from manual sampling for all metabolites across 30 μL minimum volumes (Chan et al., 2025). A microfluidic capacitive sensing platform detected as few as ~120 cells — far below the 10⁷–10⁸ cells per kilogram required therapeutically — enabling non-destructive, real-time quality checks (Foscarini et al., 2024). Reinforcement learning-based control strategies have been demonstrated in simulation, with PPO-tabular algorithms generating cell-type-specific expansion control strategies (Ferdous et al., 2023).
While real-time QC tools are technically mature at the bench level, their clinical integration into fully closed, adaptive manufacturing loops remains largely aspirational. This represents the most forward-looking automation frontier with the least current clinical validation.
| Claim | Position A | Papers Supporting A | Position B | Papers Supporting B | Evidence Strength |
|---|---|---|---|---|---|
| Lentiviral vs. non-viral transduction for clinical-scale manufacturing | Non-viral (hydroporation) achieves 1.7–2x higher yields with superior viability | Sytsma et al. (2025); Sytsma et al. (2024a); Sytsma et al. (2024b); Sytsma et al. (2025) | Lentiviral integration remains the GMP gold standard; non-viral methods remain investigational at clinical scale | Lock et al. (2017); Glienke et al. (2022); Keyer et al. (2025) | Medium — non-viral evidence is preclinical/early translational (n=4) vs. lentiviral clinical evidence (n=7+) |
| Centralised GMP vs. decentralised point-of-care economics | Centralised automation achieves 20–45% cost reduction; economies of scale | Woods et al. (2025); Weltin et al. (2025) | Point-of-care avoids logistics/cold-chain costs; enables timely treatment | Palani et al. (2023); Luanpitpong et al. (2024); Salazar-Riojas et al. (2025); Castella et al. (2020) | Medium — only 2 papers provide detailed cost metrics |
| Bioreactor type (stirred-tank vs. perfusion vs. wave) | Stirred-tank achieves highest cell densities (>5×10⁶/mL) with controlled agitation | Costariol et al. (2019); Costariol et al. (2020); Gatla et al. (2022) | Wave and perfusion systems achieve comparable viability with reduced shear stress concern | Zhao et al. (2026); Springuel et al. (2025) | Low — no direct head-to-head comparison in identical experimental context |
| Can microfluidic transduction platforms replace conventional viral transduction for clinical manufacture? | Microfluidic platforms show 27% efficiency (MOI 3), 2-fold functional improvement, 20-fold cost reduction | "Scalable and ultrafast CAR-T..." (2025); Little et al. (2025); Sin et al. (2024) | Clinical implementation remains unclear; approved processes use lentiviral or established electroporation | Lock et al. (2017); Glienke et al. (2022) | High uncertainty — microfluidic transduction evidence mostly pre-clinical |
The most consequential unresolved disagreement concerns non-viral versus lentiviral transduction at clinical scale. The quantitative advantages of non-viral methods — particularly hydroporation — are consistently demonstrated in bench-scale studies, but the absence of Phase I or GMP validation data means the field cannot yet conclude whether these gains translate to regulatory-compliant clinical manufacturing.
| Method | Associated Finding | Paper Count | Confidence |
|---|---|---|---|
| Automated stirred-tank bioreactors | Viable cell densities >5×10⁶/mL; 5–15-fold higher than static culture | 4 | High |
| Semi-automated CliniMACS Prodigy | 53-fold median CD3+ expansion; >90% batch success rate; GMP-compliant | 7 | High |
| Inertial microfluidic enrichment | 80–91% T-cell purity; ≥80% recovery; >85% contaminant removal | 6 | High |
| Hollow-fiber bioreactor (Quantum Flex) | 150–200-fold expansion over 7 days; >93% viability at harvest | 1 | Moderate |
| Non-viral hydroporation | 1.7–2.0x higher CAR-T yields; superior viability; faster day-5 proliferation | 4 | Moderate |
| Perfusion/continuous media exchange | Viability >96%; sustained expansion; reduced nutrient stress | 3 | Moderate |
| Automated closed-system (end-to-end) | 3-fold fewer batch failures; 20–45% cost reduction | 3 | Moderate |
| Label-free rapid QC (CTM, autofluorescence, capacitive) | <10 min results; <10,000 cells required; real-time feasibility | 3 | Low–Moderate |
| Optimised bioreactor parameters (agitation, pH, dissolved oxygen) | 15-fold expansion at 75 RPM vs. 10-fold at variable; pH improves transduction | 3 | Moderate |
| Point-of-care decentralised platforms | Feasible at academic centres; 9/9 product success rate in single site | 4 | Moderate |
| Automated CD4/CD8 enrichment | T-cell proliferation 1.97- to 24.2-fold improvement; purity 90–98% | 4 | High |
The strong correlation between automated CD4/CD8 enrichment and downstream manufacturing performance — including proliferation fold-change improvements of up to 12-fold — suggests that investment in upstream T-cell selection quality yields multiplicative benefits across subsequent process steps.
| Paper | Tier | Key Strengths | Key Weaknesses |
|---|---|---|---|
| Costariol et al. (2019); Costariol et al. (2020) | Medium | Multiple donors; controlled comparisons; quantitative metrics | Limited clinical translation data |
| Lock et al. (2017) | Medium | Multi-operator, multi-device reproducibility; clinical context | Small n per condition |
| Castella et al. (2020) | Medium | Phase I clinical trial embedding; 28-product dataset | Single-centre; no control arm |
| Sytsma et al. (2025); Sytsma et al. (2024a); Sytsma et al. (2024b) | Medium | Quantitative yield comparisons; scalability demonstration | Preprint/early publication stage; limited clinical validation |
| Jeon et al. (2024) | Medium | Head-to-head MDDS vs. conventional; quantitative purity/recovery | Single study; small donor n |
| Elsemary et al. (2024); Elsemary et al. (2026) | Medium | Disease-specific validation (ALL, ovarian); dual-stage process | Healthy donor optimisation; limited patient sample size |
| Woods et al. (2025) | Medium | Broad cost and throughput claims; closed-system focus | Aggregate reporting; limited primary data transparency |
| Weltin et al. (2025) | Medium | Facility-level cost modelling; automation gradient analysis | Modelling assumptions not fully disclosed |
| Song et al. (2024) | Medium | Multi-platform comparison (bag, Prodigy, G-Rex, Xuri); gene expression analysis | Single institution; n not stated per platform |
| Shah et al. (2017) | Medium | Phase I clinical data; pre/post paired comparison | Very low n post-TCS (n=7); single failure case cited |
| Gupta & Shaz (2025) | Medium | Global survey; 45 institutions; multi-regional | 35% response rate; self-reported data |
| Remaining medium-tier papers (n≈34) | Medium | Varied; clinical or process quantitative data present | Small batch numbers; limited reproducibility documentation |
| Low-tier papers (n=24) | Low | Provide supporting context and benchmarks | Inadequate process parameter reporting; low sample sizes; missing SOPs |
The common credibility limitations across the corpus include: insufficient manufacturing run numbers for platform validation (minimum 5 independent runs rarely met), absent standard operating procedure documentation, incomplete process parameter reporting, and reliance on single-arm outcome designs without cross-batch statistical comparison.
Across the 74 papers with grounding data, the mean claim coverage was 61% (535 of 873 total claims confirmed as verbatim substrings of source text). Only 2 papers fell below the 30% coverage threshold, indicating strong overall text-level verifiability. The cell expansion and viability data clusters (n=28 and n=26 papers respectively) represented the most robustly evidenced domains. Cost-per-dose data were the sparsest, with detailed metrics available in only 2 papers.
One high-severity consistency issue was identified: papers pubmed_38979201 and pubmed_39962112 share 100% of claim tokens, indicating likely corpus duplication (corresponding to Sytsma et al. 2024b and Sytsma et al. 2025 respectively — a preprint and its subsequent journal publication). Findings attributable to this work should be interpreted as deriving from a single study group rather than independent replication.
An additional 16 medium-severity issues were identified across the corpus, primarily relating to incomplete process parameter reporting, absence of variability metrics, and single-case manufacturing failure citations without reproducibility context. None of these rise to the level of data fabrication or irreconcilable contradiction.
This meta-analysis is based on 74 papers, of which the majority are small-cohort feasibility or process development studies rather than large randomised comparisons. The following specific limitations apply.
Sparse data domains: Cost-per-dose analysis rests on only 2 papers with detailed metrics; the reported 20–45% cost reductions should be treated as preliminary estimates pending broader economic modelling. Batch consistency and variability data were available in only 9 papers, most of which reported single-arm outcomes.
Clinical translation gap: Microfluidic transduction platforms are represented by 3 clinical-relevant reports, all pre-clinical or early translational. Lentiviral vector GMP manufacturing characterisation is represented by more papers (n=7+ clinical-use papers) but with heterogeneous process definitions. No paper directly compared all major platform types (Prodigy vs. wave bioreactor vs. stirred-tank vs. custom microfluidic) under identical experimental conditions.
Population gaps: The evidence base is dominated by CD19-targeted CAR-T products. Regulatory T-cell (Treg) manufacturing, γδ T-cell products, and non-haematological malignancy targets are represented by only 2–3 papers each, limiting generalisability of automation conclusions to these contexts. Patient-derived starting material data are systematically underrepresented relative to healthy donor data.
Publication and language biases: All papers are in English; non-English language studies from major CAR-T manufacturing centres (particularly in East Asia) may have been missed. Publication bias likely inflates the proportion of positive manufacturing outcomes in the dataset. Grey literature and proprietary internal process validations from commercial manufacturers are not represented.
Temporal limitations: The field is evolving rapidly; several 2025–2026 papers describe platforms not yet in clinical use, and the regulatory landscape for automated GMP manufacturing is not systematically addressed in the included papers.
What this analysis cannot conclude: Whether any specific automated platform is superior to others in a controlled head-to-head clinical study; whether non-viral transduction achieves equivalent long-term in vivo persistence to lentiviral integration at scale; and the precise cost structure of decentralised versus centralised manufacturing under real-world operational conditions.
| Paper | Claim | Verified Quote | Section |
|---|---|---|---|
| Shah et al. (2017) | CD4/CD8 selection increases CD3+ content | "Following CD4/CD8 TCS, the CD3 T-cell content in the starting product increased from 47% to 90%" | Abstract |
| Shah et al. (2017) | Proliferation improvement post-selection | "T cell proliferation increased from 1.97-fold to 24.2-fold" | Abstract |
| Shah et al. (2017) | Transduction efficiency improvement | "transduction increased from 2.57% to 33.4%" | Abstract |
| Kekre et al. (2022) | Semi-automated closed-process manufacturing | "Using a GMP compliant semi-automated, closed process on the Miltenyi Prodigy, T cells were transduced with lentiviral vector bearing a 4-1BB anti-CD19 CAR transgene and expanded." | Abstract |
| Kekre et al. (2022) | Clinical response rate | "Overall response rate at day 28 was 76.7%." | Abstract |
| Elsemary et al. (2026) | Two-stage microfluidic enrichment | "the dual-stage process showed an efficient 87% (SD ± 6%) enrichment and 80% (SD ± 30%) recovery of T-cells." | Abstract |
| Elsemary et al. (2026) | Ovarian cancer sample validation | "Validation of the process with ovarian cancer samples resulted in a T-cell purity 70% (SD ± 10%) from a starting purity of 48% (SD ± 6%) at a 64% (SD ± 4%) T-cell recovery." | Abstract |
| Elsemary et al. (2026) | No proliferation impact | "The two-stage inertial microfluidic process was also shown to have no detectable effect on the proliferation of the cells." | Abstract |
| Elsemary et al. (2024) | Lymphocyte purity enrichment | "the lymphocyte purity was enriched from 65% (SD ± 0.2) to 91% (SD ± 0.06)" | Abstract |
| Elsemary et al. (2024) | T-cell purity enrichment | "T-cell purity was enriched from 45% (SD ± 0.1) to 73% (SD ± 0.02)" | Abstract |
| Elsemary et al. (2024) | ALL sample contaminant depletion | "B-cells, monocytes and leukemic blasts depleted by 80% (SD ± 0.09), 89% (SD ± 0.1) and 74% (SD ± 0.09), respectively" | Abstract |
| Zhu et al. (2018) | CD4/CD8 enrichment purity | "Enrichment (N = 7) resulted in CD4/CD8 purity of 98 ± 4.0%, 55 ± 6% recovery and CD3+ T-cell purity of 89 ± 10%." | Abstract |
| Zhu et al. (2018) | Transduction efficiency | "Vectors at multiplicity of infection 5-10 resulted in transduction averaging 37%." | Abstract |
| Zhu et al. (2018) | Clinical-scale expansion | "An average 30-fold expansion of 10⁸ CD4/CD8-enriched cells resulted in sufficient transduced T cells for clinical use." | Abstract |
| Palani et al. (2023) | Developing-country feasibility | "a decentralized manufacturing process for anti-CD19-CAR-T cells using a fully automated closed system (Miltenyi CliniMACS Prodigy®) is feasible in a developing country setting." | Abstract |
| Palani et al. (2023) | Expansion range | "T cell expansion of 25 to 47-fold." | Abstract |
| Palani et al. (2023) | Transduction and viability | "The median transduction efficiency was 48.8%, with a median viability of 98%" | Abstract |
| Lock et al. (2022) | Transposon-based automated expansion | "2E8 T cells were enriched from leukapheresis product, activated, gene-engineered and expanded to yield up to 3.5E9 total T cells/1.4E9 CAR-modified T cells within 12 days" | Abstract |
| Lock et al. (2022) | CAR+ frequency | "yield up to 3.5E9 total T cells/1.4E9 CAR-modified T cells within 12 days (CAR-modified T cells: 28.8%±12.3%)" | Abstract |
| Malakhova et al. (2024) | Long vs. short process yield | "The CliniMACS Prodigy platform can produce a product of CD19 CAR-T cells with sufficient cell expansion (4.6 × 10⁹ cells-median for long process [LP] and 1.6 × 10⁹-for short process [SP])" | Abstract |
| Malakhova et al. (2024) | Equivalent transduction across cycle lengths | "transduction efficiency (43.5%-median for LP and 41.0%-for SP)" | Abstract |
| Malakhova et al. (2024) | 7-day equivalence | "We did not find significant differences in the properties of the products obtained during the 7- and 11-day manufacturing cycles, which is in favor of reducing the duration of production to 7 days" | Abstract |
| "Scalable and ultrafast CAR-T..." (2025) | Microfluidic transduction superiority | "The MFD geometry allowed reaching a transduction rate of 27% (for MOI 3) compared to 17% and 8% transduction (MOI 3) in 48- and 6-well plates, respectively" | Abstract |
| "Scalable and ultrafast CAR-T..." (2025) | Naïve T-cell preservation | "in the ultrafast protocol in our MFD, the amount of CD3+ Tnlp is approximately six times higher than that remaining after the standard 9 day protocol (18.07 ± 6.03% vs. 3.97 ± 2.37%)." | Abstract |
| Jeon et al. (2024) | Unactivated T-cell removal | "with up to 85% removal of unactivated T cells" | Abstract |
| Jeon et al. (2024) | Activated T-cell recovery | "approximately 80% recovery of activated T cells" | Abstract |
| Jeon et al. (2024) | Transduction efficiency doubling | "we found approximately 2-fold increase in CAR transduction efficiency for a specific sample, escalating from ∼10% to ∼20%" | Abstract |
| Sytsma et al. (2025) | Hydroporation yield advantage | "hydroporation yielded 1.7x and 2.0x more CAR-Ts, on average, than electroporation and nucleofection" | Results |
| Sytsma et al. (2025) | High-throughput scalability | "scaled-up hydroporation can process 5 × 10⁸ cells in less than 10 s" | Abstract |
| Marshall et al. (2025) | Hollow-fiber bioreactor expansion | "Expansion kinetics for all 4 starting material amounts were remarkedly similar, resulting in a 150- to 200-fold increase in cell numbers." | Abstract |
| Marshall et al. (2025) | Maximum cell yield | "This allowed for a study maximum of 2.6 billion cells from loading 15 million cells." | Abstract |
| Marshall et al. (2025) | Harvest viability | "Viability remained high throughout the expansion process with >93% for all 4 donors at harvest." | Abstract |
| Costariol et al. (2019) | Bioreactor vs. static culture | "not only T-cells can be cultivated in an automated stirred-tank bioreactor system (ambr® 250), but that their growth is consistently and significantly better than that in T-flask static culture" | Abstract |
| Costariol et al. (2019) | Shear tolerance | "there is no indication that T-cells prefer being grown under static conditions or are sensitive to fluid dynamic stresses within a stirred-tank bioreactor system at the agitation speeds investigated. Indeed, the opposite has proven to be the case" | Abstract |
| Tumaini et al. (2013) | Semi-closed system yields | "a mean expansion of 10.6-fold and a mean transduction efficiency of 68%" | Abstract |
| Remley et al. (2021) | Automated spinoculation performance | "Sepax spinoculation 1-h results being significantly better than the static transduction (83.4% vs. 35.7%)" | Results |
| Campos-González et al. (2018) | DLD cell recovery and platelet depletion | "This study demonstrates key metrics of cell recovery (80%) and platelet depletion (87%)" | Abstract |
| Campos-González et al. (2018) | Central memory phenotype advantage | "resulting in twofold greater central memory cells compared to Ficoll-Hypaque (Ficoll) and direct magnetic approaches" | Abstract |
| Skelley et al. (2025) | DLD vs. Ficoll leukocyte recovery | "DLD processing yielded significantly higher leukocyte recovery (88% vs. 58%)" | Abstract |
| Aleksandrova et al. (2019) | Automated Prodigy T-cell purity | "Starting with a CD4/CD8-positive selection, a high purity of a median of 97% T cells with a median 65-fold cell expansion was achieved." | Abstract |
| Sa & Yu (2024) | Naïve T-cell preservation in rapid process | "The product T cells have twice more naïve population (CD45RA+CCR7+) than the T cells harvested after 7-day process." | Abstract |
| Sa & Yu (2024) | Rapid transduction efficiency | "Transduction efficiency was 60%-70% (CAR expression was stable after day 2)." | Abstract |
| Sytsma et al. (2024b) | Hydroporation vs. electroporation viability | "Compared to electroporation and nucleofection, this method of membrane poration is less detrimental to cell health, which is reflected in the improved recovery and viability in hydroporated cells" | Results |
| Alzubi et al. (2021) | Automated gene-edited CAR T cells | "we established an automated, good manufacturing practice (GMP)-compliant process that integrates lentiviral transduction with electroporation of TALEN mRNA to produce functional TCRα/β-free CAR19 T cells at clinical scale" | Abstract |
| Alzubi et al. (2021) | TRAC disruption and viability | "the TRAC locus was disrupted in >35% of cells with high cell viability (>90%)" | Abstract |
| Dreyzin et al. (2026) | Clinical equivalence of Prodigy vs. bag culture | "No significant differences in response rates or incidence of CAR-associated toxicities were observed between cohorts, although the BC cohort had slightly higher rates of severe CRS and IEC-HS." | Abstract |
| Mucha et al. (2024) | Feeder cell yield enhancement | "Expansion in the presence of the feeder enhanced CAR-T production yield (4.5-fold in CAR19 and 9.3-fold in CAR123)." | Abstract |
| Little et al. (2025) | Digital microfluidic platform functionality | "demonstrating how this novel system can lead to a 2-fold improvement in immunotherapeutic functionality compared to gold standard methods" | Abstract |
| Little et al. (2025) | Cost reduction | "while providing up to a 20-fold reduction in cost" | Abstract |
| Blaeschke et al. (2018) | Automated GMP CAR-T in ALL patients | "The clinical situation of these critically ill and refractory patients with leukemia leads to inconsistent cellular compositions at start of the procedure including high blast counts and low T-cell numbers with exhausted phenotype. Nevertheless, a robust T-cell product was achieved (mean CD4+ = 50%, CD8+ = 39%, transduction = 27%, Tcm = 50%, Tscm = 46%)." | Abstract |
| Costariol et al. (2020) | Stirred-tank cell density | "At agitation speeds of 200 rpm and greater (up to 500 rpm), the CAR-T cells are able to proliferate effectively, reaching viable cell densities of >5 × 10⁶ cells ml⁻¹ over 7 days." | Abstract |
| Hood et al. (2024) | Single activation protocol advantage | "the combination of a single activation step and a shorter, 3-day, seed train resulted in significant CAR-T yield and quality improvements, specifically a 3-fold increase in cell yield, a 30% reduction in exhaustion marker expression and more efficient metabolism" | Abstract |
| Luostarinen et al. (2025) | Consistent clinical-grade yield | "The 19-FiCART manufacturing process consistently yielded more than 2 × 10⁹ highly viable CAR+ T cells, which is considered sufficient for a clinical product." | Abstract |
| Springuel et al. (2025) | Scale-down model validation | "Parallel runs in 250 mL Ambr® 250 STRs conducted at equivalent volumetric power input (P/V) of ∼8.78 W/m³ demonstrated comparable process performance and final product quality" | Abstract |
| Springuel et al. (2025) | Capacitance sensing correlation | "Integrating capacitance sensing in the 2 L STR enabled robust monitoring of viable cell concentrations in real-time, with strong correlation to offline measurements (R² = 0.98)." | Abstract |
| Springuel et al. (2025) | Automated harvesting recovery | "the Ksep® 400 was used to automate CAR-T cell harvesting, concentration, and washing at the 2 L scale, achieving >90% product recovery and nine-fold volume reduction without impacting product quality attributes" | Abstract |
| Castella et al. (2020) | Point-of-care batch success rate | "Twenty-seven out of 28 CAR T-cell products manufactured met the full list of specifications and were considered valid products." | Abstract |
| Castella et al. (2020) | Transduction and expansion time | "Ex vivo cell expansion lasted an average of 8.5 days and had a mean transduction rate of 30.6 ± 13.44%." | Abstract |
| Gatla et al. (2022) | Agitation rate and expansion | "agitation at 75 RPM resulted in a 15-fold T cell expansion, compared to 10-fold T cell expansion observed when agitation is initiated at 50 RPM and increased to 100 RPM" | Results |
| Gatla et al. (2022) | ATF perfusion viability | "we did not observe a notable decrease in the VCD of the T cells, 24 h after perfusion. This was accompanied by the absence of T cells in the waste bag" | Results |
| Salazar-Riojas et al. (2025) | Decentralised manufacturing metrics | "with a median transduction percentage of 44.7% (range, 39.2%-60.5%)" | Abstract |
| Salazar-Riojas et al. (2025) | Viability in decentralised setting | "a median CD3+ cell viability of 97.7% (range, 90.4%-98.7%). Sterility was maintained throughout the manufacturing process." | Abstract |
| Noaks et al. (2021) | Monocyte depletion efficacy | "In monocyte-depleted cultures, CD14+ cells underwent at least a 10-fold reduction, from 9.3%–32% to 0.1%–1.1%" | Results |
| Vedvyas et al. (2019) | Automated Prodigy performance | "our manufacturing protocol produced a 55 ± 9% transduction rate, 96.5 ± 1.5% viability, 2.9 ± 0.7 × 10⁹ final cell number, corresponding to ~100-fold expansion over ~10-day manufacturing" | Results |
| Song et al. (2024) | Platform differentiation state comparison | "bag cultures had the highest capacity for expansion while the Prodigy had the lowest (481-fold versus 84-fold, respectively)" | Abstract |
| Song et al. (2024) | Prodigy naïve/SCM enrichment | "with the Prodigy significantly enriched for CCR7+CD45RA+ naïve/stem central memory (Tn/scm)-like cells at 46% compared to bag and G-Rex with 16% and 13%, respectively" | Abstract |
| Zeming et al. (2025) | CTM assay rapid analysis | "the CTM assay requires fewer than 10,000 cells and delivers results within 10 minutes" | Abstract |
| Cheung et al. (2025) | Label-free autofluorescence for closed systems | "The use of 405-nm visible light to substitute for the genotoxic ultraviolet wavelengths for assessing NADH and FAD autofluorescence paves the way to incorporate autofluorescence measurements into closed and automated systems" | Abstract |
| Woods et al. (2025) | Closed-system batch failure reduction | "Transitioning to closed-system manufacturing and cryostorage has been shown to reduce batch failures threefold" | Abstract |
| Woods et al. (2025) | Cost reduction | "manufacturing costs by up to 45%" | Abstract |
| Weltin et al. (2025) | Automation and facility efficiency | "a higher degree of automation can reduce manufacturing costs by lowering personnel costs, cleanroom grade requirements and spatial footprint." | Abstract |
| Lu et al. (2016) | 6-day manufacturing timeline | "This study reports a procedure for generating CD19 CAR-T cells in 6 days, using a functionally closed manufacturing process" | Abstract |
| Jackson et al. (2020) | Manufacturing time reduction | "shortened duration of production from 12 to 8 days followed by fresh infusion into patients" | Abstract |
| Lock et al. (2017) | Platform-independent reproducibility | "Independent of the starting material, operator, or device, the process consistently yielded a therapeutic dose of highly viable CAR T cells." | Abstract |
| Glienke et al. (2022) | Automated TRUCK manufacturing yield | "Transduction efficiencies of 77.7% and 55.1% was obtained, and yields of 4.5 × 10⁹ and 3.6 × 10⁹ engineered T cells from the two donors, respectively, within 12 days." | Abstract |
| Zhao et al. (2026) | Wave bioreactor performance | "CAR-T cells expanded by the wave bioreactor exhibited faster proliferation and stronger cytotoxicity during culture." | Abstract |
| Lamas et al. (2022) | pH effect on transduction | "higher pH can significantly improve CAR T-cell transduction and proliferation" | Abstract |
| Luanpitpong et al. (2024) | POC academic centre feasibility | "Nine out of the nine products manufactured were used in a pilot study (ISRCTN17901467)." | Abstract |
| Pisani et al. (2025) | G-Rex bioreactor time reduction | "Transitioning from T-flasks to G-Rex bioreactors reduced operator hands-on time from 21 to 28 days to 14-17 days" | Abstract |
| Gupta & Shaz (2025) | Barriers to local manufacturing | "Variability in product quality was cited by 73% (29/40) of institutions." | Abstract |
| Arcangeli et al. (2020) | Protocol compatibility with automation | "we exploited our expertise with paramagnetic beads and IL-7/IL-15 to develop an optimized protocol for CAR T cell production based on reagents, including a polymeric nanomatrix, which are compatible with automated manufacturing via the CliniMACS Prodigy" | Abstract |
| Aleksandrova et al. (2024) | Cross-site manufacturing consistency | "We demonstrated consistency in cell composition and functionality of the products manufactured at two different production sites." | Abstract |
| Foscarini et al. (2024) | Microfluidic QC sensitivity | "The device showed good sensitivity even with a low number of cells-around 120, compared with the 10⁷ to 10⁸ needed per kilogram of body weight" | Abstract |
| Sin et al. (2024) | Microfluidic bioreactor patient-derived yield | "more than 60 million CAR T cells from donor cells derived from patients with lymphoma" | Abstract |
| Chan et al. (2025) | Automated sampling equivalence | "Comparison of the metabolic profiles of the samples collected via Auto-CeSS versus manual sampling revealed insignificant differences in metabolite levels" | Abstract |
| Ren et al. (2025) | Cryopreserved leukapheresis viability | "cryopreserved leukapheresis achieved ≥ 90% post-thaw viability, with recovery and phenotypic profiles comparable to peripheral blood mononuclear cells (PBMCs)." | Abstract |
Abdo, L. de M., Barros, L. R. C., Viegas, M. S., Marques, L. V. C., Ferreira, P. de S., & Chicaybam, L. (2019). Development of CAR-T Cell Therapy for B-ALL Using a Point-of-Care Approach. bioRxiv. https://doi.org/10.1101/770297
Aleksandrova, K., Leise, J., Priesner, C., Melk, A., Kubaink, F., & Abken, H. (2019). Functionality and Cell Senescence of CD4/CD8-Selected CD20 CAR T Cells Manufactured Using the Automated CliniMACS Prodigy® Platform. Transfusion Medicine and Hemotherapy, 46(3), 142–151. https://doi.org/10.1159/000495772
Aleksandrova, K., Leise, J., Priesner, C., Aktas, M., Apel, M., & Assenmacher, M. (2024). Automated manufacturing and characterization of clinical grade autologous CD20 CAR T cells for the treatment of patients with stage III/IV melanoma. Frontiers in Immunology. https://doi.org/10.3389/fimmu.2024.1328368
Alzubi, J., Lock, D., Rhiel, M., Schmitz, S., Wild, S., & Mussolino, C. (2021). Automated generation of gene-edited CAR T cells at clinical scale. Molecular Therapy: Methods & Clinical Development, 20, 379–388. https://doi.org/10.1016/j.omtm.2020.12.008
Arcangeli, S., Falcone, L., Camisa, B., De Girardi, F., Biondi, M., & Giglio, F. (2020). Next-Generation Manufacturing Protocols Enriching TSCM CAR T Cells Can Overcome Disease-Specific T Cell Defects in Cancer Patients. Frontiers in Immunology, 11, 1217. https://doi.org/10.3389/fimmu.2020.01217
Auw, N. von, Serfling, R., Kitte, R., Hilger, N., Zhang, C., & Gebhardt, C. (2023). Comparison of two lab-scale protocols for enhanced mRNA-based CAR-T cell generation and functionality. Scientific Reports, 13, 18539. https://doi.org/10.1038/s41598-023-45197-x
Blaeschke, F., Stenger, D., Kaeuferle, T., Willier, S., Lotfi, R., & Kaiser, A. D. (2018). Induction of a central memory and stem cell memory phenotype in functionally active CD4+ and CD8+ CAR T cells produced in an automated good manufacturing practice system for the treatment of CD19+ acute lymphoblastic leukemia. Cancer Immunology, Immunotherapy, 67(7), 1053–1066. https://doi.org/10.1007/s00262-018-2155-7
Campos-González, R., Skelley, A. M., Gandhi, K., Inglis, D. W., Sturm, J. C., & Civin, C. I. (2018). Deterministic Lateral Displacement: The Next-Generation CAR T-Cell Processing? SLAS Technology, 23(4), 338–351. https://doi.org/10.1177/2472630317751214
Castella, M., Caballero-Baños, M., Ortiz-Maldonado, V., González-Navarro, E. A., Suñé, G., & Antoñana-Vidósola, A. (2020). Point-Of-Care CAR T-Cell Production (ARI-0001) Using a Closed Semi-automatic Bioreactor: Experience From an Academic Phase I Clinical Trial. Frontiers in Immunology, 11, 482. https://doi.org/10.3389/fimmu.2020.00482
Chan, Z. X., Pandi Chelvam, S., Sin, W.-X., Teo, D. B. L., Abdul Rahim, A. A. B., & Wu, Y. Y. (2025). Automated, aseptic sampling with small-volume capacity from microbioreactors for cell therapy process analysis. Frontiers in Bioengineering and Biotechnology. https://doi.org/10.3389/fbioe.2025.1612648
Chen, J., & Xia, X. (2019). Abstract A026: Novel electroporation method for quick CAR-T-cell manufacture. Cancer Immunology Research. https://doi.org/10.1158/2326-6074.cricimteatiaacr18-a026
Cheung, K.-W., Kairi, F., Teo, D. B. L., Sin, W.-X., Luah, Y. H., & Chen, Y. (2025). Use of cellular FAD autofluorescence as a label-free cellular attribute for the production of chimeric antigen receptor T cells. Cytotherapy. https://doi.org/10.1016/j.jcyt.2025.05.002
Costariol, E., Rotondi, M., Amini, A., Hewitt, C. J., Nienow, A. W., & Heathman, T. R. J. (2019). Establishing the scalable manufacture of primary human T-cells in an automated stirred-tank bioreactor. Biotechnology and Bioengineering, 116(9), 2252–2266. https://doi.org/10.1002/bit.27088
Costariol, E., Rotondi, M. C., Amini, A., Hewitt, C. J., Nienow, A. W., & Heathman, T. R. J. (2020). Demonstrating the Manufacture of Human CAR-T Cells in an Automated Stirred-Tank Bioreactor. Biotechnology Journal, 15(10), e2000177. https://doi.org/10.1002/biot.202000177
Dolinski, B., Chaturvedi, V., Humes, D., Proano, C., Ott, M., & Moreno, J. (2023). Abstract 2734: Modulation of resting T cell status to enhance transduction and CAR T expansion following exposure to CD8-targeted fusosomes. Cancer Research. https://doi.org/10.1158/1538-7445.am2023-2734
Dreyzin, A., Kramer, A. M., Yates, B., Wang, H.-W., Sahaf, B., & Yuan, C. (2026). Outcomes following CD22 CAR T-cells in B-ALL: a tale of two manufacturing strategies. Cytotherapy. https://doi.org/10.1016/j.jcyt.2025.09.013
Edwards, W., Sun, N., Wang, Y., Lu, Y., Wang, C., & Mastronicola, D. (2026). Perfusion microbioreactor for CAR-Treg manufacturing. iScience. https://doi.org/10.1016/j.isci.2026.115246
Elsemary, M. T., Maritz, M. F., Smith, L. E., Warkiani, M. E., & Thierry, B. (2024). Enrichment of T-lymphocytes from leukemic blood using inertial microfluidics toward improved chimeric antigen receptor-T cell manufacturing. Cytotherapy. https://doi.org/10.1016/j.jcyt.2024.05.005
Elsemary, M. T., Maritz, M. F., Smith, L. E., Warkiani, M. E., & Thierry, B. (2026). Two-stage inertial microfluidics enrichment of activated T-cells towards a bead-less chimeric antigen receptor manufacturing protocol. Medical Oncology. https://doi.org/10.1007/s12032-026-03276-9
Ferdous, S., Shihab, I. F., Chowdhury, R., & Reuel, N. F. (2023). Reinforcement learning-guided control strategies for CAR T-cell activation and expansion. bioRxiv. https://doi.org/10.1101/2023.07.14.548968
Foscarini, A., Romano, F., Garzarelli, V., Turco, A., Bramanti, A. P., & Tarantini, I. (2024). A Microfluidic-Based Sensing Platform for Rapid Quality Control on Target Cells from Bioreactors. Sensors, 24(22), 7329. https://doi.org/10.3390/s24227329
Gatla, H., Uth, N., Levinson, Y., Navaei, A., Sargent, A., & Ramaswamy, S. (2022). Enabling Allogeneic T Cell-Based Therapies: Scalable Stirred-Tank Bioreactor Mediated Manufacturing. Frontiers in Medical Technology, 4, 850565. https://doi.org/10.3389/fmedt.2022.850565
Glienke, W., Dragon, A. C., Zimmermann, K., Martyniszyn-Eiben, A., Mertens, M., & Abken, H. (2022). GMP-Compliant Manufacturing of TRUCKs: CAR T Cells targeting GD2 and Releasing Inducible IL-18. Frontiers in Immunology, 13, 839783. https://doi.org/10.3389/fimmu.2022.839783
Gupta, D., & Shaz, B. (2025). Surveying local CAR T-cell manufacturing processes to facilitate standardization and expand accessibility. Journal of Translational Medicine. https://doi.org/10.1186/s12967-025-06400-x
Hood, T., Slingsby, F., Sandner, V., Geis, W., Schmidberger, T., & Bevan, N. (2024). A quality-by-design approach to improve process understanding and optimise the production and quality of CAR-T cells in automated stirred-tank bioreactors. Frontiers in Immunology, 15, 1335932. https://doi.org/10.3389/fimmu.2024.1335932
Jackson, Z., Roe, A., Sharma, A. A., Tavares Pereira Lopes, F. B., Talla, A., & Kleinsorge-Block, S. (2020). Automated Manufacture of Autologous CD19 CAR-T Cells for Treatment of Non-Hodgkin Lymphoma. Frontiers in Immunology, 11, 1941. https://doi.org/10.3389/fimmu.2020.01941
Jeon, H., Perez, C. R., Kyung, T., Birnbaum, M. E., & Han, J. (2024). Separation of Activated T Cells Using Multidimensional Double Spiral (MDDS) Inertial Microfluidics for High-Efficiency CAR T Cell Manufacturing. Analytical Chemistry, 96(22), 8913–8922. https://doi.org/10.1021/acs.analchem.4c01981
Kekre, N., Hay, K. A., Webb, J. R., Mallick, R., Balasundaram, M., & Sigrist, M. K. (2022). CLIC-01: Manufacture and distribution of non-cryopreserved CAR-T cells for patients with CD19 positive hematologic malignancies. Frontiers in Immunology, 13, 1074740. https://doi.org/10.3389/fimmu.2022.1074740
Keyer, V., Kydyrbayeva, A., Kulatay, T., Zauatbayeva, G., Bazhenov, D., & Ingirbay, B. (2025). Introducing CAR-T Therapy in Kazakhstan: Establishing Academic-Scale Lentiviral Vector and CAR-T Cell Production. Biomolecules, 15(8), 1166. https://doi.org/10.3390/biom15081166
Kim, H., Kim, S., Lim, H., & Chung, A. J. (2024). Expanding CAR-T cell immunotherapy horizons through microfluidics. Lab on a Chip, 24(5), 1088–1114. https://doi.org/10.1039/d3lc00622k
Lakshmi, K., von Jutrzenka-Trzebiatowski, A., Loureiro, L., González Soto, K. E., Peter, K., & Marín Morales, J. M. (2025). Automated GMP-compatible production of universal CAR Tregs for organ-targeted tolerance induction. Journal of Translational Medicine. https://doi.org/10.1186/s12967-025-07431-0
Lamas, R., Ulrey, R., Ahuja, S., & Sargent, A. (2022). Changes to culture pH and dissolved oxygen can enhance chimeric antigen receptor T-cell generation and differentiation. Biotechnology Progress, 38(4), e3275. https://doi.org/10.1002/btpr.3275
Lim, J., Oh, D., Cheng, M., Chintapula, U., Liu, S., & Reynolds, D. (2025). Enhancing Chimeric Antigen Receptor T-Cell Generation via Microfluidic Mechanoporation and Lipid Nanoparticles. Small. https://doi.org/10.1002/smll.202410975
Little, S. R., Rahbari, N., Hajiaghayi, M., Gholizadeh, F., Cloarec-Ung, F.-M., & Phillips, J. (2025). A Digital Microfluidic Platform for the Microscale Production of Functional Immune Cell Therapies. Analytical Chemistry. https://doi.org/10.1021/acs.analchem.4c06911
Lock, D., Mockel-Tenbrinck, N., Drechsel, K., Barth, C., Mauer, D., & Schaser, T. (2017). Automated Manufacturing of Potent CD20-Directed Chimeric Antigen Receptor T Cells for Clinical Use. Human Gene Therapy, 28(10), 914–925. https://doi.org/10.1089/hum.2017.111
Lock, D., Monjezi, R., Brandes, C., Bates, S., Lennartz, S., & Teppert, K. (2022). Automated, scaled, transposon-based production of CAR T cells. Journal for Immunotherapy of Cancer, 10(9), e005189. https://doi.org/10.1136/jitc-2022-005189
Lu, T. L., Pugach, O., Somerville, R., Rosenberg, S. A., Kochenderfer, J. N., & Better, M. (2016). A Rapid Cell Expansion Process for Production of Engineered Autologous CAR-T Cell Therapies. Human Gene Therapy Methods, 27(6), 209–218. https://doi.org/10.1089/hgtb