Biologics are therapeutic products derived from or produced in living systems, spanning monoclonal antibodies, antibody-drug conjugates, therapeutic proteins, engineered cells, viral vector gene therapies, and RNA therapeutics. What separates them from small molecules is not simply size but the fact that the manufacturing process itself defines the product: a change in cell line, bioreactor conditions, or purification train can alter the molecule in ways no specification fully captures.
That single principle, often summarised as “the process is the product”, explains why biologics have no straightforward generic pathway, why comparability studies dominate lifecycle management, and why capacity constraints in a handful of specialised facilities can gate an entire therapeutic class.
How to organise the modality landscape
The useful axis is not molecular weight but what the therapeutic entity actually is at the point of administration. Three broad tiers follow, each with a distinct manufacturing logic.
Tier 1: Protein therapeutics. The product is a purified protein made in a bioreactor. Mature platform technology, well-understood analytics.
Tier 2: Conjugates and engineered formats. A protein backbone plus chemistry or additional binding domains. Bioprocessing plus small molecule synthesis and conjugation.
Tier 3: Advanced therapies. The product is a living cell or a genetic instruction. Immature platforms, unresolved potency assays, and the steepest cost of goods.
Monoclonal antibodies
A monoclonal antibody is a single-specificity immunoglobulin, usually IgG1, of roughly 150 kilodaltons, against a small molecule at around 500 daltons. They remain the commercial centre of gravity in biologics.
Mechanistically, mAbs work by neutralising a soluble target or receptor, by recruiting immune effector function through antibody-dependent cellular cytotoxicity and complement-dependent cytotoxicity, or by blocking a checkpoint interaction.
Manufacturing is the most standardised in biologics. Chinese hamster ovary (CHO) cells expressing the antibody are expanded in fed-batch or perfusion bioreactors, with modern titres commonly reaching 3 to 8 grams per litre. Downstream follows a well-established sequence: Protein A affinity capture, low pH viral inactivation, polishing chromatography, viral filtration, and ultrafiltration and diafiltration into the final formulation.
The critical quality attributes are glycosylation profile (which drives effector function and clearance), charge variants, aggregation, and host cell protein residuals. Because glycosylation is exquisitely sensitive to cell culture conditions, it is the attribute that most often complicates process changes and biosimilar development.
Bispecifics and engineered antibody formats
Engineering the antibody scaffold produces formats that no natural immunoglobulin can achieve.
Bispecific antibodies bind two targets simultaneously. The dominant oncology application is T-cell engagement: one arm binds a tumour antigen, the other binds CD3 on a T cell, forcing an immune synapse. This delivers a CAR-T-like mechanism as an off-the-shelf product, at the cost of continuous or repeated dosing and cytokine release risk.
Antibody fragments (Fab, scFv, and single-domain VHH nanobodies) trade half-life and effector function for better tissue penetration, easier microbial expression, and formats compatible with local or inhaled delivery.
Fc-fusion proteins graft a receptor or peptide onto an Fc domain, borrowing the antibody’s long serum half-life through FcRn recycling.
The recurring CMC problem across these formats is chain mispairing during assembly, which creates product-related impurities that are structurally similar to the intended molecule and correspondingly difficult to separate and quantify.
Antibody-drug conjugates
An ADC uses an antibody as a delivery vehicle for a cytotoxic payload too potent to administer systemically. Three design variables govern behaviour:
- Payload: typically a microtubule inhibitor (auristatin, maytansinoid) or a DNA-damaging agent (calicheamicin, camptothecin derivative), with potency in the sub-nanomolar range.
- Linker: cleavable linkers (protease-sensitive, acid-labile, disulfide) release free payload and can produce a bystander effect on neighbouring tumour cells. Non-cleavable linkers require full antibody catabolism and offer a narrower but more controlled release.
- Drug-to-antibody ratio (DAR): usually 2 to 8. Too low and potency suffers, too high and clearance accelerates while aggregation risk climbs.
ADC manufacturing is genuinely hybrid: a full mAb process, a high-potency API synthesis under stringent containment (typically OEB 5), and a conjugation and purification step that must control DAR distribution. Few facilities can do all three, which is why ADC capacity remains a strategic bottleneck.
Therapeutic proteins and enzymes
The oldest biologics category, covering insulin, erythropoietin, clotting factors, growth hormone, interferons, and enzyme replacement therapies for lysosomal storage disorders. Expression is split between microbial systems (E. coli or yeast for non-glycosylated proteins, giving fast, low-cost production) and mammalian systems where glycosylation is required for activity or clearance. Enzyme replacement therapies often need specific glycan structures, such as mannose-6-phosphate for lysosomal targeting, making glycoengineering central rather than incidental.
Cell therapy
Here the product is a living cell population. The dominant approved class is CAR-T, where a patient’s T cells are engineered to express a chimeric antigen receptor, most commonly against CD19 in B-cell malignancies or BCMA in multiple myeloma.
Autologous therapy uses the patient’s own cells. Each batch is one patient, so leukapheresis, transduction (usually lentiviral or gammaretroviral), expansion, cryopreservation, and release testing all sit on the clinical critical path. Vein-to-vein times of three to five weeks are typical, and there is no room for batch failure.
Allogeneic therapy uses healthy donor cells to create many doses from one starting material, with gene editing to remove the T-cell receptor and HLA molecules that would otherwise cause graft-versus-host disease and rejection. It solves the economics and eliminates wait time, but durable persistence has been harder to achieve.
Adjacent approaches include TCR-engineered T cells for intracellular targets, tumour-infiltrating lymphocytes, NK cell therapies with a gentler cytokine profile, and mesenchymal stromal cells for immunomodulation.
The defining CMC problem is potency assay design. Regulators expect a measurement that reflects mechanism of action, but a heterogeneous, living, patient-variable product resists the kind of specification that a purified protein accepts easily.
Gene therapy
Gene therapy delivers genetic material to correct, replace, or silence a gene.
In vivo AAV therapy administers a recombinant adeno-associated virus vector directly. Serotype determines tissue tropism. The practical ceiling is a packaging capacity of roughly 4.7 kilobases, which excludes large genes, and pre-existing neutralising antibodies exclude a meaningful share of patients. Manufacturing yields, full-to-empty capsid ratio, and immunogenicity at high systemic doses remain the central challenges.
Ex vivo lentiviral therapy modifies haematopoietic stem cells outside the body and returns them. Larger cargo capacity (roughly 8 to 9 kilobases) and a controlled modification environment, at the cost of a conditioning regimen for the patient.
In vivo gene editing using CRISPR-Cas systems, base editors, or prime editors makes a permanent sequence change rather than adding an episomal copy. Off-target editing assessment and delivery beyond the liver are the open problems.
RNA therapeutics
mRNA instructs the patient’s own cells to produce a protein, delivered in lipid nanoparticles. Manufacturing is cell-free in vitro transcription, so the platform changes only its sequence between products, which is what enabled pandemic-speed development. Effects are transient and cold chain requirements are demanding.
siRNA and antisense oligonucleotides silence a target transcript. GalNAc conjugation achieves highly efficient hepatocyte delivery through the asialoglycoprotein receptor, enabling quarterly or twice-yearly dosing. These are chemically synthesised, so manufacturing looks closer to a small molecule process than a biologic one.
Modality comparison
| Modality | Product form | Production platform | Typical dosing | Principal CMC challenge |
| Monoclonal antibody | Purified IgG | CHO fed-batch or perfusion | Every 2 to 4 weeks | Glycosylation, aggregation |
| Bispecific | Engineered protein | CHO | Frequent or continuous | Chain mispairing |
| ADC | Conjugated protein | CHO plus HPAPI synthesis | Every 3 weeks | DAR control, containment |
| Therapeutic protein | Purified protein | Microbial or mammalian | Daily to weekly | Glycan structure, immunogenicity |
| CAR-T (autologous) | Living cells | Patient-specific, vector dependent | Single infusion | Potency assay, batch of one |
| AAV gene therapy | Viral vector | HEK293 or insect cell | Single dose | Full-to-empty ratio, yield |
| mRNA | Nucleic acid in LNP | Cell-free IVT | Prime and boost | Cold chain, LNP consistency |
| siRNA | Synthetic oligo | Chemical synthesis | Quarterly or biannual | Delivery beyond liver |
Frequently asked questions
Why are there no generic biologics? Because living systems cannot produce an identical copy. The regulatory route is biosimilarity, requiring extensive analytical, non-clinical, and clinical comparability rather than the bioequivalence study sufficient for a generic small molecule. Cell and gene therapies currently have no equivalent follow-on pathway at all.
What does “the process is the product” mean in practice? Any manufacturing change, including a new facility, bioreactor scale, or raw material source, requires a formal comparability exercise to demonstrate the product is unchanged in quality, safety, and efficacy.
Which modality is growing fastest? Measured by pipeline entries, ADCs, bispecifics, and cell and gene therapies are expanding fastest. Monoclonal antibodies still dominate approved product revenue by a wide margin.
Why do cell and gene therapies cost so much? Batch sizes of one or a few patients, specialised facilities, viral vector and plasmid supply constraints, complex release testing, and cryogenic logistics. Allogeneic and in vivo approaches are the main routes being pursued to change that structure.
Disclaimer:
This article is for informational and educational purposes for life sciences professionals. It is not medical advice, formulation guidance, or a regulatory reference. Technical values are indicative and must be verified against current pharmacopoeial and regulatory requirements for your product and jurisdiction. Patients should consult a qualified healthcare professional regarding any medication.