A CDMO strategy that works for an antibody can fail with a viral vector. A plasmid process cannot be judged like a recombinant protein process. A live microbial product does not behave like a sterile injectable. A nanoparticle delivery system creates a different control problem from a purified enzyme.

A cell therapy adds chain of identity, viability, phenotype, potency, and clinical-site handling pressure that most conventional biologics do not carry.

CDMO Network supports biologic and advanced therapy programs by routing each modality to the manufacturing, testing, formulation, quality, and regulatory capabilities built for that specific product class.

The objective is not to place every program into the same CDMO model. The objective is to define the product’s technical architecture, identify the right capability set, and preserve continuity from early development through manufacturing.

Each modality determines how the product is made, what must be controlled, what data matter, and where the program can fail.

A modality is more than the name of the product type. It defines the production system, impurity profile, process risk, analytical burden, storage condition, release strategy, and regulatory pathway. When the modality changes, the CDMO model changes with it.

An antibody program starts with cell line development, upstream mammalian culture, downstream purification, potency assays, glycan analysis, formulation, and sterile fill-finish. A viral vector program starts with plasmids, producer cells, transfection or infection, vector purification, functional titre, residual impurity testing, frozen storage, and delivery-function control.

A cell therapy program starts with cells, modification, expansion, phenotype, viability, potency, cryopreservation, chain of identity, and clinical logistics.

The wrong modality assumptions create expensive rework. A team can select a CDMO too early, build analytics around the wrong readout, overinvest in a process that cannot scale, or underbuild quality systems for a product that will face regulatory scrutiny later.

The Network uses these answers to build the CDMO route.

The product category comes first. The vendor search comes second.

Antibodies and engineered binders remain one of the most established biologic modality groups, but they are not simple products.

This category includes monoclonal antibodies, bispecific antibodies, multispecific antibodies, antibody fragments, Fab fragments, scFv formats, VHH and nanobodies, Fc-fusion proteins, recombinant antibodies, antibody mimetics, engineered scaffold proteins, immunocytokines, immunotoxins, radioimmunoconjugates, Fc-silenced antibodies, afucosylated antibodies, glycoengineered antibodies, high-concentration antibodies, and other engineered binding systems.

The central manufacturing question is clear: can the expression system produce the correct binder with the correct structure, function, purity, and stability?

Antibody programs usually require mammalian expression, stable cell line development, clone screening, upstream process development, Protein A or affinity capture, polishing chromatography, viral clearance, aggregation control, residual host-cell protein control, residual DNA control, potency testing, glycan analysis, charge variant analysis, formulation, and sterile drug product support.

Engineered formats add complexity. Bispecifics create chain-pairing risk, product-related impurities, fragments, mispaired species, and dual-function potency requirements. Nanobodies and antibody fragments can use microbial expression in some programs, but solubility, folding, disulfide formation, endotoxin, and activity must remain controlled.

Fc-fusions can introduce glycosylation, aggregation, clipping, or potency risks that differ from standard monoclonal antibodies.

CDMO Network supports antibody and engineered binder programs with the manufacturing and analytical systems required to move from sequence to controlled product.

The binder must not only bind. It must bind consistently, survive manufacturing, and remain stable in the intended presentation.

Recombinant proteins are broad, technically uneven, and often underestimated.

This modality group includes therapeutic proteins, complex biologics, enzymes, cytokines, growth factors, hormones, blood factors, plasma proteins, fusion proteins, peptides, long-acting proteins, modified proteins, difficult-to-express proteins, secreted proteins, membrane proteins, disulfide-rich proteins, aggregation-prone proteins, protease-sensitive proteins, lysosomal enzymes, diagnostic proteins, recombinant antigens, viral antigens, multimeric proteins, and protein complexes.

A recombinant protein program starts with expression-system fit.

CHO, HEK293, E. coli, yeast, insect cells, Bacillus, Pichia, Saccharomyces, cell-free systems, and engineered cell substrates each solve different problems and create different constraints. A glycoprotein often needs mammalian expression. A small non-glycosylated enzyme can fit microbial fermentation. A difficult multimeric protein can require insect cells or specialized mammalian systems.

A disulfide-rich protein can need oxidative folding control. A membrane protein can need fusion partners, detergent compatibility, or lipid environment strategy.

The process must protect structure and function.

For enzymes, activity matters more than mass. For cytokines, potency and impurity control carry high importance. For growth factors, folding, dimerization, purity, and bioactivity matter. For diagnostic proteins, epitope presentation and lot consistency matter. For lysosomal enzymes, glycosylation and activity can define the product.

A recombinant protein CDMO route includes host selection, construct optimization, expression screening, upstream process development, fermentation or cell culture, purification, refolding where required, activity or potency assays, impurity testing, formulation, stability, and scale-up.

A protein program does not succeed because protein appears on a gel. It succeeds when the correct protein form can be produced, purified, measured, stabilized, and supplied.

Genetic medicines and nucleic-acid products convert sequence into therapeutic or functional output.

This category includes mRNA, self-amplifying RNA, circular RNA, plasmid DNA, linear DNA, oligonucleotides, siRNA, ASO, miRNA, guide RNA, sgRNA, CRISPR reagents, donor DNA templates, minicircle DNA, aptamers, IVT RNA, capped RNA, modified mRNA, and other nucleic-acid systems.

The core control problem is information integrity.

The product carries sequence information. Manufacturing must preserve identity, purity, molecular form, modification status, and functional usability.

Plasmid programs require microbial strain control, fermentation, lysis, purification, topology preservation, supercoiled percentage, endotoxin reduction, residual RNA removal, residual host-cell DNA control, sequence confirmation, and storage strategy.

Plasmids used for viral vector production require additional planning because they become critical inputs for another modality.

mRNA programs require DNA template control, in vitro transcription, capping, poly(A) strategy, modified nucleotide strategy where relevant, impurity reduction, dsRNA control, purification, concentration, formulation interface, and delivery-system compatibility.

Guide RNA and CRISPR materials require sequence accuracy, purity, nuclease control, delivery compatibility, and functional assay support.

Oligonucleotide programs require chemistry, purification, impurity profiling, identity, modification control, stability, and formulation alignment.

The analytics must match the molecule. Concentration alone does not define quality. Sequence, purity, topology, capping, fragment profile, residual template, residual enzymes, dsRNA, endotoxin, and storage behavior all matter depending on product type.

Nucleic-acid manufacturing must protect the message. The process fails if it changes the information or delivers it in the wrong form.

Viral vectors create one of the highest-complexity CDMO categories.

This group includes AAV, lentiviral vectors, adenoviral vectors, retroviral vectors, HSV vectors, oncolytic viruses, baculovirus systems, viral vaccine vectors, helper virus systems, replication-defective vectors, replication-competent platforms, capsid-engineered vectors, dual AAV systems, integration-deficient lentiviral vectors, hybrid viral vectors, and conditionally replicating viral platforms.

A vector is not just a particle. It is a delivery system.

Manufacturing has to control producer cells, plasmid or helper inputs, transfection or infection, upstream production, harvest timing, nuclease treatment, purification, residual impurities, concentration, formulation, fill-finish, and storage.

Analytics must show physical quantity and biological function. Genome titre, particle titre, capsid or envelope quality, infectivity, functional titre, potency, residual plasmid, residual host-cell DNA, host-cell protein, aggregation, empty/full ratio, safety-related testing, and stability all matter.

AAV programs often require capsid-specific process control, full-empty analysis, genome integrity, potency, and residual impurity reduction. Lentiviral programs require envelope stability, infectious titre, vector copy number, transduction efficiency, and frozen handling.

Adenoviral and HSV platforms require platform-specific safety, potency, and purification logic. Oncolytic viruses require replication behavior, potency, identity, and stability control.

Vector CDMO services need manufacturing partners with deep modality experience, not generic biologics capacity.

A high physical titre means little if the vector loses function. The vector must arrive, enter, express, and perform.

Cell therapy products make the cell itself part of the therapeutic system.

This category includes autologous cell therapy, allogeneic cell therapy, CAR-T, TCR-T, NK cell therapy, CAR-NK, iPSC-derived therapies, stem cell therapies, MSC products, dendritic cell therapy, macrophage therapy, hematopoietic cell therapy, tumor-infiltrating lymphocytes, Treg therapy, gamma delta T cells, engineered cell therapies, ex vivo cell engineering, closed-system manufacturing, and point-of-care manufacturing models.

Cell therapy manufacturing changes the meaning of product control.

The process handles living cells. Product quality includes identity, purity, viability, phenotype, potency, expansion behavior, vector copy number where relevant, editing outcome where relevant, sterility, mycoplasma, endotoxin, and post-thaw performance.

Autologous programs require chain of identity, chain of custody, patient-specific scheduling, rapid release, and clinical-site coordination. Allogeneic programs require banking strategy, scale, consistency, comparability, storage, and broader distribution planning.

Engineered cell programs require vector supply, editing materials, transduction or editing efficiency, phenotype, function, and release testing.

The manufacturing route can include cell sourcing, selection, activation, gene modification, editing, expansion, washing, formulation, cryopreservation, storage, shipment, thawing, and administration preparation.

A cell therapy program has little room for loose coordination. The product is biological, operational, and logistical at the same time.

Vaccines require manufacturing systems that preserve immune-relevant quality.

This modality group includes mRNA vaccines, DNA vaccines, protein subunit vaccines, viral vector vaccines, live attenuated vaccines, inactivated vaccines, conjugate vaccines, VLP vaccines, recombinant vaccines, pandemic vaccines, therapeutic vaccines, cancer vaccines, adjuvanted vaccines, multivalent vaccines, bioconjugate vaccines, and protein nanoparticle vaccines.

The core question is not only whether antigen exists. The question is whether the product presents the right immune signal.

Protein subunit vaccines require antigen expression, purification, conformation, potency, antigenicity, adjuvant compatibility, formulation, and stability. Viral vector vaccines require vector production, infectivity or functional potency, residual impurity control, and storage strategy.

mRNA vaccines require RNA production, LNP or delivery formulation, potency, purity, stability, and fill-finish alignment. VLP vaccines require particle assembly, morphology, antigen display, and purification control. Conjugate and bioconjugate vaccines require conjugation control, ratio, consistency, potency, and characterization.

Pandemic and outbreak response programs add speed, tech transfer, regulatory coordination, and supply pressure. Commercial vaccine programs add consistency, validation, stability, adjuvant supply, and global distribution.

Vaccine manufacturing must protect biological signal through process, formulation, storage, and administration. The immune system responds to what the product presents. Manufacturing must preserve that presentation.

Live microbial products require a different manufacturing mindset.

This category includes microbiome therapeutics, live biotherapeutic products, probiotic therapeutics, engineered microbes, synthetic biology therapeutics, microbial consortia, spore therapeutics, phage therapy, bacteriophage products, engineered probiotics, anaerobes, live bacterial therapeutics, engineered commensals, oral live biotherapeutics, microbiome consortia, postbiotics, microbial metabolite therapeutics, synthetic microbial communities, spore-based products, and live biotherapeutic fill-finish.

The product often contains living or biologically derived systems that need viability, identity, composition, stability, and controlled delivery.

Live biotherapeutic manufacturing requires strain banking, fermentation, anaerobic handling where needed, biomass recovery, concentration, formulation, cryoprotection, lyophilization, encapsulation, fill-finish, storage, and stability. Strict anaerobes require oxygen-control infrastructure. Microbial consortia require ratio control and strain-specific identity.

Phage products require host systems, amplification, purification, endotoxin control, potency, and host-cell impurity removal.

Analytics must confirm identity, purity, potency, viability, strain composition, residual impurities, microbial contaminants, and stability. For postbiotics or microbial metabolites, the focus shifts toward composition, bioactivity, impurity control, and consistency.

These products do not fit conventional sterile biologics templates. A live product must remain alive when required, inactive when intended, and controlled in either state.

Delivery systems increasingly define whether a therapy can function.

This group includes lipid nanoparticles, LNP-mRNA systems, liposomes, polymer nanoparticles, exosomes, extracellular vesicles, RNA delivery systems, gene delivery systems, LNP-DNA, targeted nanoparticles, polyplex systems, bioconjugates, protein-LNP hybrids, virus-like particles, nonviral gene delivery systems, and other hybrid delivery platforms.

The delivery system has its own CMC requirements.

Particle size, polydispersity, encapsulation efficiency, surface charge, payload integrity, release profile, lipid or polymer composition, residual solvents, sterility, stability, and potency can all determine product performance.

LNP programs require lipid sourcing, mixing process control, RNA integrity, encapsulation, particle-size control, buffer exchange, sterile filtration where applicable, fill-finish, frozen storage, and stability.

Exosome and extracellular vesicle programs require cell substrate control, harvest, purification, identity, cargo profile, potency, and impurity reduction. Liposomes and polymer nanoparticles require composition, size distribution, loading, release behavior, and stability.

Hybrid modalities create CDMO routing complexity because they cross boundaries. A protein-LNP hybrid needs protein control and particle control. A virus-like particle program needs assembly control and vaccine-style analytics. A nonviral gene delivery program needs nucleic-acid quality and delivery-system performance.

The payload matters. The delivery system decides whether the payload reaches the biology.

Expression systems are not just internal process choices. They are modality-defining platforms.

This group includes CHO systems, HEK293 cells, E. coli, yeast expression, Pichia pastoris, Saccharomyces, Bacillus, Corynebacterium, Streptomyces, microbial fermentation, insect cells, Sf9 cells, High Five cells, cell-free expression, engineered cell substrates, microbial cell factories, and precision fermentation platforms.

Each production platform creates its own advantages and constraints.

CHO supports many complex glycoproteins and antibodies. HEK293 supports transient expression, viral vectors, and human-like processing. E. coli supports fast microbial production of non-glycosylated proteins, enzymes, plasmids, and fragments, but adds endotoxin and folding constraints.

Yeast supports scalable eukaryotic expression with glycoengineering options. Insect cells support baculovirus systems, VLPs, and complex assemblies. Cell-free systems support rapid expression and specialized proteins. Precision fermentation supports enzymes, food proteins, industrial biologics, and specialty materials.

The platform affects product quality, cost, speed, impurity profile, scalability, and regulatory documentation.

A product should not enter a host system because that system is available. It should enter because the system fits the product’s biology and destination.

Emerging modalities sit between established categories.

They include advanced therapies, precision medicine manufacturing, personalized medicine, rare disease programs, oncology biologics, immunotherapies, regenerative medicine, phage display reagents, synthetic biology products, programmable therapeutics, CRISPR therapeutics, gene circuit therapeutics, living medicines, hybrid biologics, immunomodulators, alternative protein biomanufacturing, cultivated meat media and reagents, and research-grade to GMP reagent transitions.

These programs require early structure because no standard CDMO category fully contains them.

A programmable therapeutic can involve nucleic-acid design, protein engineering, delivery systems, cell-based testing, and functional assays. A living medicine can require microbial manufacturing, formulation, stability, release testing, and biological containment. A rare disease program can require small-batch GMP manufacturing with high documentation discipline.

A cultivated meat reagent program can require food-oriented production economics with biologics-level protein quality.

Emerging modalities often fail because teams use the closest old template.

The better strategy defines the product system from scratch. What is the active entity? What makes it function? What system produces it? What must be removed? What assay proves activity? What quality system fits intended use? What scale is realistic?

CDMO Network supports these programs by building capability routes around the actual product instead of forcing the product into legacy categories.

Emerging modalities need structure before they need scale.

Many modern programs are not single-modality programs.

A gene therapy uses plasmids and viral vectors. An mRNA product uses nucleic acid manufacturing and lipid nanoparticles. A cell therapy uses cells, vectors, media, cytokines, release assays, and cryopreservation. A vaccine can combine protein antigens, adjuvants, nanoparticles, and sterile fill-finish. A diagnostic platform can use antibodies, enzymes, nucleic acids, controls, and stabilized reagents.

These programs need more than one CDMO capability. They need interfaces.

The plasmid has to fit the vector process. The vector has to fit the cell process. The RNA has to fit the LNP process. The antigen has to fit the adjuvant and final presentation. The analytical package has to follow the product across those handoffs.

The main risk in cross-modality programs is not that one vendor cannot execute its task. The main risk is that the handoff between tasks fails.

A strong modality strategy defines the interfaces before execution begins.

That is where multi-CDMO programs either become controlled systems or disconnected vendor chains.

Programs use this modality layer when they need to:

The modality page should act as the central routing hub for all product-type pages.

Modality-based CDMO support uses capabilities across:

The right capability set depends on the product’s biological form, development stage, and intended use.

Modality is the first technical filter in CDMO routing. It determines the product’s manufacturing system, analytical model, formulation constraints, quality risks, and regulatory logic. Antibodies require one control strategy. Viral vectors require another.

Cell therapies, plasmids, vaccines, LNPs, live biotherapeutics, and recombinant proteins each define different process and release requirements. A strong modality strategy keeps the product’s biology connected to manufacturing, testing, stability, documentation, and supply.

A strong modality CDMO strategy starts with the product class.

It defines the product’s biological form, production platform, critical quality attributes, release logic, impurity risks, stability needs, formulation requirements, regulatory pathway, and supply model before choosing the CDMO route.

Strong support includes modality-specific manufacturing, analytics, formulation, stability, quality, regulatory CMC, tech transfer, scale-up, and logistics.

Antibody programs need binder-specific production and characterization. Recombinant proteins need structure-function control. Nucleic acids need sequence and purity control. Viral vectors need functional delivery control. Cell therapies need viable cell control and chain of identity. Vaccines need immune-relevant potency. Live biotherapeutics need viability and strain control. Nanoparticles need payload and particle control. Emerging modalities need custom architecture.

The wrong CDMO path usually starts with the wrong category assumption.

The right path starts by naming what the product is and what the product must remain through manufacturing.