Over the past decade, extracellular vesicles (EVs) have emerged as one of the most promising frontiers in regenerative medicine. By serving as natural mediators of intercellular communication and transporting diverse bioactive molecules—including proteins, lipids, and nucleic acids—EVs have demonstrated considerable therapeutic potential across a broad range of diseases. As more candidates progress from laboratory research into clinical development, the industry's focus is rapidly shifting from proving therapeutic feasibility to establishing manufacturing processes capable of delivering high-quality EVs consistently and at commercial scale.

Among the various EV sources, mesenchymal stem/stromal cell-derived extracellular vesicles (MSC-EVs) have attracted particular attention because of their immunomodulatory, anti-inflammatory, and regenerative properties. Derived from mesenchymal stem/stromal cells (MSCs), these vesicles carry diverse arrays of bioactive proteins, lipids, and small RNAs that regulate immune responses, control inflammation, and promote tissue regeneration (Riazifar et al., 2017; Na et al., 2026). With more than 160 EV-related clinical trials registered worldwide—most of which investigate MSC-derived EVs—the field has entered a new phase in which manufacturing capability is becoming just as important as biological efficacy.

This evolution presents a fundamental challenge. Success is no longer defined simply by the ability to produce EVs, but by the ability to manufacture them consistently, reproducibly, and in compliance with increasingly stringent regulatory expectations. Robust upstream cell expansion, controlled culture environments, efficient downstream purification, and end-to-end process traceability have become essential requirements for translating promising laboratory discoveries into clinically and commercially viable products.

In response to evolving guidance from the International Society for Extracellular Vesicles (MISEV2023) and regulatory agencies such as China's National Medical Products Administration (NMPA), manufacturers must address three critical questions:

1. How can sufficient quantities of high-quality MSCs be obtained without compromising cellular potency?
2. How can conditioned media be collected efficiently while preventing contamination from exogenous EVs?
3. How can the entire manufacturing workflow remain closed, monitorable, scalable, and GMP-compliant?

Addressing these questions requires more than increasing production capacity—it demands manufacturing strategies that integrate process control, quality assurance, and scalability from the earliest stages of process development. Before designing such a workflow, however, it is essential to understand why existing upstream culture platforms often become bottlenecks during industrialisation. Each platform offers distinct advantages, but differences in process transparency, scalability, nutrient distribution, and quality control can significantly influence both manufacturing performance and the quality attributes of the final EV product.

Upstream Process Comparison

As extracellular vesicle (EV) therapeutics progress from laboratory research towards clinical translation and commercial manufacturing, the demands placed on upstream bioprocessing continue to evolve. Manufacturing platforms are no longer evaluated solely on their ability to expand cells at scale; they must also provide robust process control, enable real-time monitoring, support reproducible product quality, and comply with increasingly stringent regulatory expectations. Within this context, selecting the appropriate upstream culture strategy has become a critical consideration in the development of scalable and GMP-compliant EV manufacturing processes.

The four primary upstream culture options for mesenchymal stem cell-derived extracellular vesicle (MSC-EV) production each serve distinct roles depending on development stage and manufacturing requirements:

• 2D Planar Culture (Flasks, CellSTACK): Offers operational simplicity, low entry barriers, and well-established workflows, making it the preferred choice for early-stage research and process development. However, limited culture surface area, labour-intensive handling, and poor batch-to-batch consistency restrict its suitability for large-scale manufacturing.
• Hollow Fibre Systems: Enable high cell densities and continuous perfusion harvesting within a compact footprint. However, their enclosed architecture limits direct observation and routine sampling, creating a "black box" environment that complicates process monitoring, validation, and quality assurance throughout production.
• Fixed-Bed Systems: Provide a large surface area for adherent cell growth within a closed configuration while reducing manual intervention. Nevertheless, uneven nutrient and oxygen distribution across the matrix may result in spatial heterogeneity, presenting challenges for process optimisation, scale-up, and manufacturing consistency.
• 3D Microcarrier Suspension Culture (Stirred-Tank Bioreactors): Combines scalable cell expansion with homogeneous culture conditions, real-time process monitoring, and straightforward sampling. Although optimisation of agitation conditions, fluid dynamics, and harvesting strategies remains essential, this approach provides the process transparency, operational flexibility, and scalability required for GMP-oriented commercial manufacturing.

Although each culture strategy addresses specific manufacturing requirements, their suitability changes considerably as production scales from research to clinical and commercial applications. Platforms that offer limited process visibility or sampling accessibility may become increasingly difficult to validate and control under GMP conditions. By comparison, stirred-tank microcarrier suspension culture integrates scalability with process transparency, enabling more effective monitoring, quality control, and manufacturing consistency. These characteristics have established it as a leading upstream strategy for industrial-scale MSC-EV production.

Platform Architecture: An Integrated Manufacturing Strategy

The comparison of upstream culture platforms demonstrates that successful EV manufacturing requires more than efficient cell expansion. Commercial production depends on the seamless integration of upstream culture, media management, downstream purification, formulation, and final fill-finish within a controlled manufacturing workflow. Weaknesses at any stage can compromise product quality, increase process variability, or introduce unnecessary manufacturing risks. Consequently, the industry is moving towards integrated process platforms that minimise manual intervention, improve process traceability, and support end-to-end GMP compliance.

Against this backdrop, CytoNiche Biotech developed the 3D FloTrix™ End-to-End Manufacturing Platform—an integrated manufacturing solution designed to support the complete MSC-EV production workflow, from upstream cell expansion to final product packaging. Built around closed-system operation, single-use technologies, and animal-component-free processing, the platform is designed to enhance manufacturing consistency while simplifying scale-up and technology transfer across different production scales.

The 3D FloTrix™ manufacturing workflow consists of the following stages:

• Upstream Seed Expansion: Human umbilical cord-derived MSCs (P4) are seeded onto 3D RecomTrix™ recombinant collagen microcarriers and expanded in 3D FloTrix™ vivaSPIN-SU single-use bioreactors. The stirred-tank suspension environment promotes homogeneous cell growth while allowing critical process parameters—including agitation, dissolved oxygen, and pH—to be monitored and controlled throughout culture.
• Medium Exchange: The culture is transitioned to a chemically defined, exogenous-EV-free medium (RCD288) to eliminate background contamination from serum- or platelet-derived extracellular vesicles, thereby improving the purity and reproducibility of the harvested product.
• Continuous Perfusion Harvest: At passage P6, the bioreactor operates under a multi-harvest perfusion strategy, continuously collecting conditioned medium over an extended production window. This approach maximises medium utilisation while maintaining cell viability and extending the productive lifespan of the culture.
• Downstream Processing: Collected conditioned medium is transferred directly to the 3D FloTrix™ megaEXO Exosome Harvesting System for concentration and purification using tangential flow filtration (TFF). The closed downstream workflow reduces manual handling while enabling efficient removal of impurities and concentration of EVs for subsequent formulation.
• Formulation & Lyophilisation: Purified EVs are combined with appropriate lyoprotectants to support either liquid formulation or freeze-dried products, allowing flexibility for different storage conditions and therapeutic applications.
• Final Product Packaging: The formulated bulk drug substance is aseptically filled into final dosage units using the automated 3D FloTrix™ vivaVIAL Filling System. Automated filling minimises operator intervention, improves batch consistency, and supports GMP-compliant final product manufacture.

Rather than functioning as a collection of independent unit operations, the 3D FloTrix™ platform is designed as an integrated manufacturing ecosystem in which each stage supports the next. By combining scalable upstream expansion with controlled downstream purification and automated fill-finish, the platform establishes a continuous workflow that enhances process reproducibility, facilitates technology transfer, and strengthens overall manufacturing robustness. 

Case Study: Experimental Validation of the 3D FloTrix™ Manufacturing Workflow

To evaluate the feasibility of the proposed manufacturing strategy under representative production conditions, a proof-of-concept study was conducted using the complete 3D FloTrix™ workflow. The objective was not only to assess cell expansion performance, but also to demonstrate that the integrated process could maintain critical cell quality attributes while supporting continuous EV production and downstream processing. Particular emphasis was placed on cell viability, phenotypic stability, production efficiency, and the biological activity of the resulting extracellular vesicles, all of which are essential indicators of manufacturing robustness. 

Cell Expansion and Working Seed Bank Preparation

Human umbilical cord Wharton's jelly-derived MSCs (UC-MSCs) from donor-screened P4 banks were expanded in 3D FloTrix™ vivaSPIN-SU bioreactors using 3D RecomTrix™ microcarriers:

• Expansion Factor: 14-fold expansion achieved over 4 days of continuous culture.
• Viability: Maintained at ≥85% during growth, reaching 98.04% post-harvest.
• Phenotypic Integrity: Flow cytometry confirmed ≥95% positivity for surface markers CD73, CD90, and CD105, maintaining ISEV compliance.

High expansion efficiency alone is insufficient if accompanied by declining cell quality. Throughout the culture process, cell viability remained high while the characteristic MSC phenotype was preserved after harvest, demonstrating that large-scale expansion can be achieved without compromising the biological properties required for reliable EV production.

With a robust working cell bank established, the next step was to evaluate whether the production process could sustain continuous EV harvesting while maintaining manufacturing performance.

Continuous Perfusion Harvesting (2D vs. 3D Platform)

P5 working cells were transitioned to the P6 production stage in a 3 L working-volume 3D FloTrix™ vivaSPIN-SU bioreactor using chemically defined medium (RCD288). Perfusion was performed by harvesting 1 L every 48 hours (Days 2–6) followed by 1 L every 24 hours thereafter, resulting in a cumulative harvest volume of 13 L. Performance was compared with a conventional 2D planar culture control (T75 flasks with 12 mL medium replacement every 72 hours).

Performance Comparison: 2D vs. 3D Perfusion Scheme