Extracellular vesicles and outer membrane vesicles FAQs

Yes-NorthX describes development and GMP manufacturing support for bacterial OMVs/EVs and mammalian EVs, including microbial and mammalian processing and product-specific analytics. The exact scope and route to GMP are assessed for each source, payload and development stage.

NorthX’s has microbial experience relevant to OMV/EV strains, fermentation, purification and GMP work. Its mammalian capabilities include adherent systems such as fixed-bed and hollow-fibre approaches that may support selected EV-producing cells. NorthX also reports active investment in OMV/EV analytical platform development and an intention to extend relevant learning to mammalian EVs. Because the field lacks one universal manufacturing and release platform, the project plan must remain product-specific. An initial technical review should confirm biosafety, equipment fit, analytical gaps, material needs and the evidence required for the next regulatory milestone.

Manufacturing-scale purification normally combines clarification, filtration, concentration or diafiltration and one or more chromatographic or size-based operations. The sequence is product-specific because recovery, impurity clearance and vesicle integrity trade off against one another.

Low-speed clarification and depth or membrane filtration first remove cells and larger debris. Tangential-flow filtration can concentrate vesicles and exchange buffer, but membrane choice and operating conditions affect recovery and aggregation. Chromatography or other orthogonal separations may reduce soluble proteins, nucleic acids, free payload and unwanted particles. Research ultracentrifugation can be useful for development but is often difficult to scale, automate and validate for routine manufacture. Each step should be evaluated using mass balance, particle recovery, impurity clearance and retained biological activity.

OMV lipopolysaccharide is controlled through source-strain design, culture and harvest conditions, purification, compositional testing and relevant biological assays. Because lipopolysaccharide may contribute to both activity and reactogenicity, the goal is a justified and consistent profile rather than indiscriminate removal.

Genetic changes can reduce or alter activity, but they may also affect strain growth, vesicle yield and product composition. Process conditions influence the amount and type of bacterial material packaged into or associated with the vesicles. Standard endotoxin assays may not fully predict the biological behaviour of a complex OMV product, so additional pyrogenicity or cell-based testing may be needed. Dose and route of administration determine the clinical relevance of the measured activity. The control strategy should connect strain genetics, process parameters, composition, potency and safety data.

Sterility risk is managed through controlled source materials, closed or contamination-controlled processing where possible, bioburden monitoring, validated sterile filtration when the product allows it and aseptic filling. If vesicle size or formulation prevents a final sterilising filter, upstream aseptic controls become even more important.

The process design should identify how contamination is prevented or detected. Mammalian cell substrates also require an adventitious-agent strategy, while bacterial products need controls that distinguish intended organisms or components from contamination. Filter compatibility must be demonstrated because vesicles can be retained, damaged or change size distribution. Methods are selected according to sample constraints and regulatory expectations. Hold times, storage, transport and container closure are part of the same sterility-assurance strategy.

Choose a mammalian EV platform when cell-derived delivery or signalling biology is central to the product, and evaluate an OMV platform when immune stimulation, antigen display or bacterial scalability is advantageous. The choice must match the intended mechanism, safety profile, payload and route of administration. Note that bacterial EVs can have properties similar to bacterial OMVs and/or Mammalian EVs.

Mammalian EVs can carry proteins, lipids and nucleic acids in a biologically derived membrane, but their heterogeneity and often modest productivity make scale-up challenging. OMVs can provide natural pathogen-associated signals and dense antigen presentation, which is useful for vaccines and immunotherapy but requires deliberate control of reactogenic components such as lipopolysaccharide. Source-cell or source-strain engineering can alter yield, cargo and targeting for either platform. The purification process must preserve function while removing product- and process-related impurities. A feasibility package should compare potency per dose, recovery, consistency, safety and cost at the intended scale.

EV or OMV potency should measure a biological function related to the proposed mechanism of action, such as target-cell uptake, cargo delivery, immune activation or a downstream functional response. Particle count alone is not a potency assay.

An early programme may use a matrix of complementary assays while the mechanism is still being refined. A cell-based assay often provides biological relevance but can be variable and slow, so supporting binding, enzymatic or molecular assays may strengthen the control strategy. The assay must distinguish active product from degraded, empty or compositionally altered particles. Reference material and system-suitability controls are especially important for heterogeneous vesicle products. Starting potency development early reduces the risk that an otherwise manufacturable process reaches GMP without a release-ready functional method.

EV and OMV formulation must preserve particle integrity, cargo and biological activity through storage, thawing, transport and administration. The optimal buffer, concentration, temperature and freeze-thaw limit must be established experimentally for each product.

Vesicles can aggregate, fuse, leak cargo or adsorb to container surfaces during handling. Buffer composition, ionic strength, cryoprotectants and fill concentration can change both recovery and function. Stability-indicating methods should measure physical properties, composition and potency rather than particle count alone. Short-term hold studies support manufacturing, while real-time and accelerated studies support retest periods and shelf life.

EVs and OMVs are being developed for vaccines, immunotherapy, regenerative medicine and the delivery of proteins, drugs or nucleic acids. The clinical maturity and risk profile vary substantially by source, cargo and indication.

OMVs can present native or engineered antigens and provide intrinsic immune stimulation, making them relevant to vaccine and cancer-immunotherapy strategies. Mammalian EVs are being studied as cell-free mediators of biological effects and as delivery vehicles that can carry complex cargo. Bacterial EVs can have properties similar to bacterial OMVs and/or mammalian EVs. All platforms can be engineered at the producer-cell level or modified after harvest. A broad platform claim is not enough to define a medicine, the intended mechanism of action and dose must be connected to measurable product attributes. Potency, biodistribution, off-target effects and repeat-dose safety require product-specific evidence.

An EV or OMV project should start with the producer cell or strain, construct and engineering history, intended cargo, mechanism of action, target dose and available analytical data. Existing production, purification, formulation and non-clinical results help define the first scalable work package.

The source material and banking status determine traceability, biosafety and adventitious-agent requirements. The intended product definition should specify whether the active substance is the vesicle, a surface antigen, an encapsulated payload or a combination of these. Early samples and methods help assess particle concentration, size distribution, identity markers, impurities and biological activity. Route of administration and dose influence sterility, endotoxin or pyrogen expectations, formulation and container choice. The initial assessment should convert unknowns into decision points for strain or cell-line work, process development, analytics and GMP readiness.

Extracellular vesicle (EV) is the broad term for lipid-bilayer particles released by cells, exosomes are a proposed EV subtype linked to an endosomal origin, outer membrane vesicles (OMVs) are bacterial vesicles released from the outer membrane of Gram-negative bacteria. The terms should not be used interchangeably.

Mammalian EV preparations can contain vesicles formed through different cellular pathways that overlap in size and composition. Under MISEV2023 guidance, the term exosome is best reserved for material whose endosomal origin is demonstrated, rather than inferred from size alone. OMVs contain bacterial membrane components, including proteins, lipids and lipopolysaccharide, and can carry engineered antigens or payloads. Their biological and manufacturing risks therefore differ from those of mammalian EVs. Clear source, production and characterisation language is essential for both scientific communication and regulatory documentation.

EV and OMV control usually covers particle quantity and size, identity, composition, purity, cargo, potency, safety and stability. No single analytical method is sufficient, so orthogonal measurements are needed.

Particle counting and size-distribution methods describe physical populations but do not prove vesicle identity or biological activity. Protein, lipid, nucleic-acid and source-specific markers help define composition, while imaging can provide structural evidence. Product-related aggregates, free cargo, non-vesicular particles, host-cell proteins and residual DNA may require separate assays. OMV programmes can also need a strategy for lipopolysaccharide composition and activity. The final assay panel should be linked to the product’s mechanism of action, administration route and the process steps intended to control each risk.

EV and OMV scale-up is difficult because productivity, vesicle composition and impurity profiles can change with culture conditions and equipment. Methods that isolate research-scale material, especially high-speed centrifugation workflows, may not be practical or reproducible at manufacturing scale.

Producer-cell density, media, stress, growth phase and harvest timing can alter both the number and biological properties of vesicles. Larger vessels introduce gradients, different shear conditions and longer processing times. Purification must separate particles from cells, debris, soluble proteins, nucleic acids and similarly sized contaminants without damaging the vesicles. Concentration and diafiltration can create membrane fouling or particle loss if not designed for the specific product. Scale-down models and representative engineering runs are therefore important before a GMP campaign.