Phase 2Cell culture and production (upstream)
02Fed-batch / perfusion culture in the production bioreactor (production culture)
The central step where cells make the antibody; feeding and environmental control balance titer and quality.
- Standard method
- Fed-batch culture in a stirred-tank bioreactor (PID control of pH, DO, temperature and agitation)
Why it is the standard. Fed-batch is a robust method with an extensive scale-up track record, and with optimization titers reach the g/L range. It is highly platformed, and its regulatory handling is well established.
About this step and the alternatives
This is the “main event,” where the selected cells make the antibody itself. It is the goal of upstream (cell culture), and what accumulates in the culture broth here becomes the raw material that downstream purification works with.
The core of how it runs is “fed-batch”: nutrients are not added all at once but topped up little by little during the run. In the stirred tank, cells are first grown to high density, and in the later phase they produce antibody. Automatic control that holds temperature, pH (how acidic or alkaline the medium is) and dissolved oxygen constant supports stable production.
The art lies in balancing quantity and quality. If the culture is stretched out to chase titer (antibody concentration), aggregates and fragments tend to increase. What is built in here carries over as a load on downstream, so judging when to harvest matters.
This is the central step where the cells make the antibody itself. Cells are grown to high density while nutrients are added in portions (fed-batch), and temperature, pH and dissolved oxygen are tightly controlled to balance yield and quality.
Alternatives in practice
- Perfusion culture (continuous / intensified) · Repligen XCell ATF 6 (alternating tangential flow, cell retention) + Cytiva/Sartorius bioreactor — High cell density, continuous production. Requires an ATF/TFF cell retention device
- N-1 perfusion + high-seed fed-batch · Repligen XCell ATF 4 (N-1 step) → high-density inoculation of the production bioreactor — Shortens the production fed-batch and raises tank productivity
- Batch culture · Sartorius Biostat STR or Thermo HyPerforma S.U.B — Simple process but low titer. For early studies or specific products
Viable cell density and viability monitoring
Basic daily measurement of viable cell count and viability that informs feed volumes and harvest timing.
- Standard method
- Automated viable cell density and viability measurement by trypan blue dye exclusion
- Makers
- Logos Biosystems, Nexcelom, Hamilton, Beckman Coulter
- Reagents and materials
- Beckman Coulter Vi-CELL BLU Reagent Pack (C39291, all-in-one trypan blue, buffer, cleaning and disinfectant solutions, RFID level tracking)
Why it is the standard. The Vi-CELL XR is the de facto reference instrument for automated offline counting by trypan blue, and it serves as the validation reference for alternative monitoring technologies.
About this step and the alternatives
How many cells in the culture tank are alive and growing right now? This step checks that every day. It is a kind of “health checkup” for the culture: it counts only the healthy cells, accurately, to gauge how vigorous the culture is.
The key is a blue dye called trypan blue. Live cells exclude the dye and stay unstained, while only dead cells turn blue. An automated counter reads this difference and calculates the number of viable cells and their share of the total in one go.
These numbers inform when to add nutrients and when to end the culture and harvest. A drop in viability signals declining quality. Another strength is that the method is more objective than manual visual counting, so the daily values scatter less.
This is the basic measurement that checks every day whether the culture is running well. A dye stains only dead cells, and the number and fraction of live cells are tracked. The results inform feed volumes and harvest timing.
Alternatives in practice
- Online viable cell measurement by capacitance · Hamilton Incyte Arc sensor or Aber Futura/Biomass probe — Continuously measures only viable cells. Insensitive to debris; useful for feed control
- Inline multivariate monitoring by Raman spectroscopy · Endress+Hauser (Kaiser) Raman Rxn2 analyzer — Estimates glucose, lactate, VCD, etc. simultaneously. Requires building PAT models
- Quantitative phase imaging / image cytometry · Nexcelom Cellaca MX or Logos Biosystems LUNA-FX7 — Fast, high-throughput counting. Suited to parallel measurement of small samples
Antibody titer measurement
Standard titer assay that quantifies antibody output and drives culture progress assessment and harvest decisions.
- Standard method
- Titer quantitation by Protein A affinity HPLC
Why it is the standard. Because it binds antibodies selectively, it is little affected by impurities and quantifies reproducibly in minutes; it is the standard titer method for the culture process.
Check also. Titer measurement based on Protein A binding is fast and convenient, but what it measures is the “total amount of antibody with an Fc region.” It cannot tell whether the antibody is correctly assembled or free of aggregates, so quality decisions combine it with size-exclusion and charge analyses (later steps).
With Perfusion / process intensification (continuous production)
- Perfusion harvest is a low-cell-density permeate that has passed through the cell retention device, so the clarification load and configuration change
- Method: Perfusion permeate is clarified continuously with depth / particle-removal filters (0.2 µm in some cases). High-g continuous centrifugation becomes less important, and filter-only configurations are also used
- Equipment: Depth filter train + particle-removal and 0.2 µm filters (continuous skid)
- Check also: Fed-batch handles high-density, high-viscosity whole culture broth, so continuous centrifugation + depth filtration is the standard. With perfusion, designs that omit centrifugation are possible
About this step and the alternatives
How much antibody has been made in the bioreactor? This step measures that amount (titer) to judge whether the culture is on track, and reads the signal to move on to harvest.
The key is Protein A, a protein that grabs only the antibody's “tail” region (Fc). When culture broth passes through a dedicated column, impurities flow through, and the captured antibody is released with a mildly acidic solution and quantified.
Its strength is fast, reproducible measurement in minutes. But it only tells you the total amount of antibody that has the “tail.” Whether the antibody is correctly assembled or aggregated is assessed together with the size and charge analyses in later steps.
This step quantifies how much antibody has been produced. Binding to Protein A lets the antibody in the culture broth be measured quickly and selectively. It feeds directly into assessing culture progress and deciding when to harvest.
Alternatives in practice
- Biolayer interferometry (BLI) · Sartorius Octet R8/RH16 + Protein A biosensors — Label-free, multi-sample in parallel. Quick at-line titer of culture supernatant
- Surface plasmon resonance (SPR) · Cytiva Biacore 8K + Protein A sensor chip — High precision and reproducibility. Lower throughput than BLI
- Online Protein A SPR/HPLC · At-line automation with Cytiva Biacore or Agilent 1290 Bio LC — Connected directly to the bioreactor to monitor titer in real time
Phase 4Purification (downstream)
06Capture by Protein A affinity chromatography
The capture step at the heart of antibody purification: Protein A grabs only the Fc and raises purity in one step.
- Standard method
- Protein A affinity chromatography (bind-elute)
Why it is the standard. The alkali-stabilized Protein A ligand for Fc delivers high dynamic binding capacity and selectivity. It withstands hundreds of caustic cleaning cycles, making it the de facto standard capture resin for antibody platform purification.
Check also. Protein A targets the Fc region, so it cannot be used for antibody fragments that lack an Fc. In addition, Protein A leaching in small amounts from the resin remains as an impurity, so removal in later polishing steps and confirmation of the residual level are a given.
About this step and the alternatives
After culture, the broth contains large amounts of host-cell proteins and fragments besides the target antibody. This step picks only the antibody out of that clarified but still crowded harvest; it is both the entrance to purification and its heart.
The key is the catcher, “Protein A.” It grabs only the Fc region, the antibody's “tail,” so when the liquid flows through a column packed with resin, only the antibody binds and impurities flow through. Switching to a mildly acidic solution (acetic acid or glycine) then releases the antibody, like loosening a grip, and it elutes. This single bind-and-elute cycle raises purity in one stroke.
Its strengths set up the next step. The eluted antibody is already acidic, so it leads naturally into the low-pH viral inactivation that follows. Engineered ligands withstand alkaline cleaning, and reusing the resin hundreds of times pays off in commercial production. On the other hand, because it targets the Fc, it cannot be used for antibody fragments that lack an Fc, and Protein A itself leaches from the resin in small amounts and remains as an impurity, so removal in later polishing steps and confirmation of the residual level are a given.
This is the heart of antibody purification. Protein A grabs only the Fc region of the antibody, so a single step raises purity substantially. Engineered ligands that withstand alkaline cleaning make repeated use possible.
Alternatives in practice
- Next-generation high-capacity Protein A resins · Thermo Fisher POROS MabCapture A Select or Purolite Praesto Jetted A50 — High flow rate, high dynamic binding capacity. Shorter cycle times
- Protein A membranes / monoliths · Sartorius Sartobind Protein A membrane or BIA Separations CIMmultus rProtein A monolith — No diffusion limitation, high flow rate. Well suited to continuous / intermittent capture
- Continuous multicolumn capture (PCC/SMB) · Cytiva ÄKTA pcc or Sartorius Resolute BioSC (3–4-column MCC) + MabSelect PrismA — Maximizes resin utilization. For integration into continuous processes
Low-pH viral inactivation
A safety step on an orthogonal principle that uses the acidic Protein A eluate to inactivate enveloped viruses.
- Standard method
- Viral inactivation by low-pH hold (around pH 3.5, 30–60 min or longer)
Why it is the standard. An inactivation step against enveloped viruses, on an orthogonal principle, that takes advantage of the low pH of Protein A elution. It is one part of the clearance by two or more orthogonal principles required by ICH Q5A(R2).
Check also. Low pH inactivates only enveloped viruses, which have a lipid membrane. It is less effective against small non-enveloped viruses, so it is always combined with nanofiltration downstream (removal on a different principle) to provide double assurance of viral safety. Note also that inactivation is insufficient if pH or hold time is not tight enough.
About this step and the alternatives
This step deliberately shifts the antibody solution captured on Protein A to acidic conditions. It acts as a safety device that gently disables, with acid, any “membrane-wrapped viruses” that may have slipped in.
The neat part is that the Protein A eluate is already acidic. The pH (acidity) is lowered slightly to around 3.5 and held quietly for about 30–60 minutes. The envelope breaks down and the viruses lose their infectivity.
The catch is that the antibody itself is sensitive to acid. Going too low or holding too long causes aggregation and denaturation. So the solution is neutralized promptly after the hold, and safety is doubled by the virus filtration later on, which works on a different principle.
This safety step inactivates any enveloped viruses that might have entered. The Protein A eluate is already acidic, so simply lowering the pH and holding for a set time inactivates enveloped viruses efficiently.
Alternatives in practice
- Detergent treatment (Triton X-100 alternatives) · Merck/Sigma-Aldrich reduced Triton X-100 or Ecosurf / lauryl maltoside addition + stirred tank — Alternatives to Triton X-100 are being evaluated because of EU environmental regulations. Inactivates enveloped viruses
- Solvent/detergent (S/D) treatment · TnBP + polysorbate 80/Triton addition + hold tank — Proven in plasma fractionation. Requires a downstream step to remove S/D reagents
- UVC irradiation · Bayer/Sartorius UVivatec or Cytiva UVC flow-through unit — No chemicals required. A complementary option for antibodies that do not tolerate low-pH inactivation
Neutralization and depth filtration
The low-pH inactivated pool cannot be used downstream while strongly acidic, and the precipitate formed on neutralization must be removed.
- Standard method
- Neutralization to pH 5–7 by adding base (Tris/sodium hydroxide), then depth filtration to remove precipitate and turbidity
Why it is the standard. Millistak+ HC Pro is the industry standard for removing turbidity and precipitate from neutralization, delivering low-HCP, low-turbidity pools consistently.
About this step and the alternatives
This step brings the pool, made strongly acidic in the previous step, back to “near neutral” so it can move on. In cooking terms, it tones down an overly sour dish so it is palatable and strains out the cloudiness at the same time; it does both jobs.
A base (Tris or sodium hydroxide) is added gradually to bring the pH back to 5–7. The problem is that some of the protein solidifies at this point, causing precipitate and turbidity. A depth filter (a multilayer filter cloth) catches it and sends on a clear pool.
Why is it needed? While still acidic, the pool cannot be loaded onto the next ion exchange chromatography. Precipitate left from neutralization clogs downstream columns and carries in impurities (host cell proteins). The more turbidity is removed here, the lighter and more stable the next step.
The pool after low-pH viral inactivation cannot go onto the next ion exchange chromatography while it is still strongly acidic. Bringing the pH back to 5–7 by neutralization generates protein precipitate and other solids, so depth filtration removes them and prevents clogging of downstream columns and HCP carryover.
Alternatives in practice
- Switch to Sartorius depth filters · Sartorius Sartoclear PB/PC or Sartopure series depth filters — Dual sourcing and area optimization. Choose capsules to suit the scale
- Pretreatment with Clarisolve · Merck Clarisolve 20MS/40MS — Spreads the load when neutralization forms a lot of precipitate
- 3M/Cuno Zeta Plus depth filters · 3M Zeta Plus VR/EXT series (charged depth media) — Charged media may also provide additional HCP and DNA removal
Impurity removal by ion exchange chromatography (polishing)
Finishing step that removes residual HCP, DNA, aggregates and charge variants and sets final purity.
- Standard method
- Anion exchange
Why it is the standard. AEX flow-through removes HCP/DNA/viruses and also contributes orthogonal viral clearance. CEX reduces aggregates and charge variants. Combining the two is the standard antibody polishing configuration.
Check also. Anion exchange in flow-through mode is effective for removing DNA and endotoxin, but it does not adequately remove aggregates or variants with different charge. The design assumes these are removed separately in a 2nd polishing step on a different principle (such as cation exchange).
About this step and the alternatives
This polishing step carefully removes, at the end, the trace contaminants left after capture. DNA, host cell proteins (HCP) and viruses that a single Protein A step cannot fully remove are still present.
The tool is anion exchange. The resin is a sticky sheet that catches negatively charged contaminants. The antibody mostly passes straight through, while only the negatively charged DNA, HCP and viruses stick and are removed; that is why it is called flow-through mode.
This step sets the final purity of the drug substance and also supports safety by physically removing viruses. However, aggregates and variants with different charge are not fully removed in flow-through mode, so the design assumes they are handed over to a 2nd polishing step on a different principle.
This finishing step shaves off the remaining impurities. Anion exchange in flow-through mode adsorbs and removes DNA, HCP and viruses, and cation exchange separates aggregates and charge variants. This is the stage that sets final purity.
Alternatives in practice
- Multimodal (mixed-mode) chromatography · Cytiva Capto adhere (multimodal strong AEX, flow-through) or Bio-Rad Nuvia aPrime — Less dependent on salt concentration; removes aggregates and HCP at the same time
- Hydrophobic interaction chromatography (HIC) · Cytiva Capto Phenyl ImpRes or Tosoh TOYOPEARL PPG-600M — Binds and removes aggregates under high-salt conditions. High buffer consumption
- Membrane chromatography (flow-through) · Sartorius Sartobind Q or Merck Natrix Q (single-use AEX membranes) — High flow rate, disposable. Flow-through polishing for DNA/viruses/HCP
A single polishing step cannot bring aggregates, charge variants and residual HCP down to specification, so a 2nd polishing step is standard.
- Standard method
- Removal of aggregates, charge variants and residual HCP by cation exchange chromatography (bind-elute)
Why it is the standard. Capto SP ImpRes offers high resolution and excellent separation of aggregates and charge variants, and it has an extensive process track record as the de facto standard for mAb polishing.
About this step and the alternatives
This step is a “2nd polish”: it scrubs off again, by a different method, the impurities that remain after the 1st polish. The targets include clumps of antibody stuck together (aggregates), relatives whose charge is slightly off (charge variants) and Protein A leached from the capture resin.
The standard is cation exchange chromatography. At the working pH, the positively charged antibody binds firmly to the negatively charged resin. Then, as salt or pH is changed gradually, impurities with similar properties come off first, and the desired antibody is recovered afterward at high purity.
The key is to “combine a principle different from the previous polish.” If the 1st step was flow-through, this one is bind-elute. The more the approach differs, the less slips through, keeping aggregates and charge variants within specification and stabilizing later steps and final purity.
The standard mAb platform is a 3-column train of Protein A capture + 2 polishing steps, and anion exchange (flow-through) alone cannot fully remove aggregates, charge variants and leached Protein A. Cation exchange or multimodal/HIC bind-elute polishing is combined as an orthogonal principle to bring CQAs within specification.
Alternatives in practice
- Multimodal mixed-mode chromatography · Cytiva Capto MMC ImpRes or Bio-Rad CHT/Nuvia cPrime — Removes aggregates and HCP by an orthogonal mechanism. High salt tolerance
- Hydrophobic interaction chromatography (HIC) · Tosoh TOYOPEARL Butyl-600M or Cytiva Capto Butyl — Effective for removing aggregates (HMW). Requires high-salt buffers
- Ceramic hydroxyapatite (CHT) · Bio-Rad CHT Ceramic Hydroxyapatite Type I/II — Removes a wide range of impurities by a combined metal affinity + cation exchange mechanism
Physical removal by nanofiltration (virus filtration)
A safety step on a principle different from low-pH inactivation: size exclusion physically removes small non-enveloped viruses.
- Standard method
- Nanofiltration with a 20 nm-class virus removal filter
- Reagents and materials
- — (prefilter Merck Viresolve Shield/Prefilter, compatible buffer)
Why it is the standard. Removes small viruses (parvovirus, etc.) by size exclusion, a physical removal step independent of inactivation. It complements low-pH inactivation on a different principle and is a core step for meeting the viral safety requirements of ICH Q5A(R2).
Check also. Nanofiltration physically removes viruses by size, and because its principle differs from low-pH inactivation, it assures orthogonality for safety. On the other hand, high-concentration or high-viscosity antibody can reduce flux or clog the filter easily, and virus removal performance (log reduction) must be verified under the relevant lot and load conditions.
About this step and the alternatives
This is the second half of the two-stage virus defense. If the previous step attacks with chemistry, this one “strains out” viruses with a fine filter. The principle is completely different.
Pore size decides what passes and what is stopped. A 20 nm-class membrane lets the antibody through and physically captures smaller viruses that lack an envelope, such as parvoviruses.
The key is how it flows. With high-concentration, high-viscosity antibody, flux drops and the membrane tends to clog. Removal performance is verified for each lot and load condition to firmly support safety.
This is a second safety step that physically removes viruses by size. The filter lets the antibody through, while its smaller pores also capture small non-enveloped viruses such as parvoviruses. Because its principle differs from low-pH inactivation, orthogonality is assured.
Alternatives in practice
- Sartorius virus removal filters · Sartorius Virosart HF (hollow fiber) or Virosart HC/CPV (flat sheet) — Dual sourcing. Choose flux/retention by membrane material and structure
- Pall virus removal filters · Pall Pegasus Prime/SV4 (PVDF flat sheet) — High flux. Used with a prefilter for hard-to-filter mAbs
- Asahi next-generation regenerated cellulose filter · Asahi Kasei Planova S20N (new regenerated cellulose 20 nm membrane) — High throughput even for hard-to-filter mAbs. Wide design space
- Additional virus removal by AEX · Sartorius Sartobind Q membrane (flow-through) — Complements nanofiltration. Improves overall viral clearance by an orthogonal mechanism
Concentration and buffer exchange by UF/DF (drug substance preparation)
Drug substance step that concentrates to target and exchanges into formulation buffer, enabling high-concentration subcutaneous products.
- Standard method
- UF/DF by tangential flow filtration (TFF)
- Equipment
- TFF system, TFF cassette, TFF hollow fiber
Why it is the standard. UF/DF is the standard final unit of the manufacturing train, handling concentration, buffer exchange and volume reduction. It enables high-concentration (subcutaneous) formulations.
Check also. Toward the end, concentration polarization at the membrane slows filtration and recovery varies. Reaching the target concentration and completing buffer exchange must be confirmed by measurement, and holdup left in piping and membranes matters more the smaller the batch.
About this step and the alternatives
This is the final finishing of purification, adjusting the purified antibody to the target concentration and composition to make the “drug substance.” It removes excess salt and makes the material ready to hand over to formulation.
The lead role goes to a way of using membranes called tangential flow filtration (TFF). The liquid flows along the membrane surface, water and small salts pass through, and only the antibody stays on the retentate side. It is first concentrated (UF), then its contents are gently exchanged into the formulation buffer (DF).
The key is to concentrate and clean up the antibody while treating it gently. Shear at the membrane surface and overconcentration cause aggregation, so judging flow rate, pressure and the final concentration determines recovery and quality.
This step brings the purified antibody to its final formulation. Ultrafiltration concentrates it to the target concentration, and diafiltration exchanges it into the formulation buffer. For high-concentration products for subcutaneous injection, the challenge is keeping viscosity and aggregation in check.
Alternatives in practice
- Single-pass TFF (SPTFF) · Repligen ProConnex SPTFF or Merck Pellicon Single-Pass module — Continuous concentration without a tank. Well suited to concentration before the final UF
- C/D/C strategy + inline PAT · Cytiva ÄKTA flux + inline UV/refractive index/Raman — Automatically controls concentrate–diafilter–concentrate for high-concentration formulations
- Dialysis / centrifugal ultrafiltration (small scale) · Merck Amicon Ultra (30 kDa) or Thermo Slide-A-Lyzer dialysis cassette — For small lab/preclinical samples. Not scalable
Final filtration of drug substance (BDS) for bioburden reduction, and filling into freeze bags for storage
Final 0.2 µm filtration before frozen storage of drug substance to lower bioburden, then filling into freeze bags.
- Standard method
- Final filtration through a 0.2 µm sterilizing-grade filter (bioburden reduction) → fill into single-use freeze bags and store frozen
- Reagents and materials
- — (stability during freezing depends on the formulation buffer exchanged in by UF/DF; nitrogen overlay as needed)
Why it is the standard. The standard is to give the drug substance (BDS) a 0.2 µm final filtration before frozen storage to reduce the viable count (bioburden). It is separate from the filtration before aseptic filling of the drug product (DP) and handles microbial control at the drug substance stage.
Check also. This is different from the sterile filtration before aseptic filling on the drug product (DP) side; the aim at the drug substance stage is bioburden reduction, not sterility assurance. For high-concentration, high-viscosity products, flow drops and the filter tends to clog in this final filtration, so sufficient filter area, holdup and integrity testing are the key points. In freezing/thawing, container material compatibility and freezing rate affect quality.
About this step and the alternatives
Before the drug substance, adjusted in concentration and composition by UF/DF, is frozen, its microbial level is brought down once. This is the final filtration of the drug substance.
There are two parts. First, the drug substance passes through a 0.2 µm sterilizing-grade filter (pores finer than bacteria) to lower the viable count (bioburden); then it is filled into single-use freeze bags and stored frozen. At this point the “drug substance” is complete for now and is kept frozen until the day of formulation.
Keep in mind that this is different from the “sterile filtration before aseptic filling” of the drug product. The aim at the drug substance stage is bioburden reduction, not sterility assurance. For high-concentration, high-viscosity subcutaneous products, flow slows and the filter tends to clog in this last filtration, so how much filter area to provide and how to reduce holdup are the key points.
The standard is to reduce the bioburden (viable microbial count) of the drug substance finished by UF/DF with a 0.2 µm final filtration before frozen storage, then fill it into single-use freeze bags and freeze it. This is separate from the filtration before aseptic filling of the drug product (DP); it handles microbial control at the drug substance stage.
Alternatives in practice
- 0.2 µm filtration with 2 filters in series (redundant configuration) · Sartorius Sartopore 2, two in series, or redundant Merck Millipore Express SHC installation — Putting 0.2 µm filters 2 in series provides integrity margin. Integrity testing at each stage
- Controlled-rate freezing (plate-based) · Plate-based controlled-rate freezing systems such as Sartorius Celsius FFTp — Freezes large drug substance volumes uniformly and limits quality variability from freeze-concentration
Phase 6Quality attributes and analytics
16Quantitation of aggregates and fragments (size variants)
Key CQA quantitation that measures and tightly controls aggregates and fragments, which pose an immunogenicity risk.
- Standard method
- Size-exclusion chromatography (SEC-HPLC/UHPLC)
- Makers
- Waters, Wyatt, Beckman Coulter, Tosoh
- Reagents and materials
- HPLC mobile phase solvents
Why it is the standard. The long-standing industry standard in biopharmaceuticals for quantifying soluble aggregates and fragments. Aggregates and fragments are CQAs controlled under ICH Q6B.
Check also. Size-exclusion chromatography is the standard for aggregate assessment, but large aggregates can be missed through dilution or interaction with the resin. For important products, cross-checking aggregate levels by orthogonal methods such as analytical ultracentrifugation or light scattering makes the result reliable.
About this step and the alternatives
What does this step look at? Whether antibodies have stuck together into large clumps (aggregates) or, conversely, broken into small fragments, sorted and measured by size.
The procedure is simple. When the sample passes through a column with a porous mesh structure, larger clumps come out first and smaller fragments later. This time difference separates species by molecular weight, and the proportion of aggregates (high-molecular-weight species, HMW) is expressed as a number.
Why is it controlled so tightly? Because the body treats aggregates as foreign, and they risk triggering an immune response. However, clumps can be missed through dilution or interaction with the column, so for important products the result is cross-checked by an orthogonal principle such as light scattering.
This key test checks whether the antibody has clumped into aggregates or broken into fragments. Aggregates pose an immunogenicity risk, so species are separated by size and the proportion of high-molecular-weight species (HMW) is tightly controlled.
Alternatives in practice
- SEC-MALS (multi-angle light scattering) · Wyatt (Waters) DAWN MALS + Optilab dRI detector — Measures absolute molecular weight and eliminates the effect of column interactions
- Sedimentation velocity analytical ultracentrifugation (SV-AUC) · Beckman Coulter Optima AUC (successor to XL-A) — Orthogonal, matrix-independent confirmation of aggregates. Low throughput
- Asymmetric flow field-flow fractionation (AF4) · Wyatt Eclipse AF4 + DAWN MALS — No stationary phase, so little loss of very high-molecular-weight aggregates
Purity and size heterogeneity by CE-SDS
Resolves heavy/light chain assembly and fragmentation by size to assess purity.
- Standard method
- Capillary electrophoresis SDS (CE-SDS, reduced/non-reduced)
- Reagents and materials
- Disulfide reduction alkylation reagents
Why it is the standard. The standard for purity assessment as the automated, quantitative version of SDS-PAGE. Reduced conditions separate heavy/light chains, and non-reduced conditions separate intact IgG from low-molecular-weight species; it is used for purity control under ICH Q6B.
About this step and the alternatives
Is the antibody assembled as designed, with nothing coming apart or broken off? CE-SDS lines the antibody up in order of molecular size and expresses purity as a number.
It works in two modes. When reduced, the antibody comes apart into heavy and light chains, revealing assembly defects. Under non-reducing conditions, the proportion of whole, intact antibody is measured as is.
It automates and quantifies the familiar SDS-PAGE, so purity can be controlled with low-variability numbers. It feeds directly into pass/fail against specifications and release to the next step, and uneven action of the reducing agent can sway the results.
This step resolves heavy- and light-chain assembly and fragmentation by molecular size to determine purity. Reducing conditions split the antibody into heavy and light chains, and non-reducing conditions show the proportion of intact antibody. It is an automated, quantitative version of SDS-PAGE.
Alternatives in practice
- SDS-PAGE (Coomassie / silver staining) · Bio-Rad Criterion/Mini-PROTEAN + Coomassie/silver stain — Low cost but weaker in quantitation and throughput
- Microchip CE · PerkinElmer LabChip GXII Touch (HT Protein Express) — High throughput. Resolution can be lower than capillary CE-SDS
- CE-SDS-MS coupling · SCIEX PA800 Plus + ZipChip/CESI-MS interface — Allows mass identification of size-separated peaks. Advanced technique
Assessment of charge variants (acidic and basic species)
CQA assessment that separates charge variants from deamidation, C-terminal lysine, etc. to control consistency and stability.
- Standard method
- Imaged capillary isoelectric focusing (icIEF)
- Reagents and materials
- Carrier ampholytes pI markers
Why it is the standard. The industry standard for protein separation by isoelectric point. It is used for charge profiling across all stages of mAb manufacturing, from scale-up to formulation, stability and QC, and is also adopted in the charge variant analysis of USP reference standards.
About this step and the alternatives
Even with the same sequence, slight chemical changes alter an antibody's charge distribution. Deamidation or leftover C-terminal lysine shifts the molecule's “electrical fingerprint” bit by bit. This step reads that fingerprint, sorts the antibody into acidic, main and basic species, and reports their ratios.
The deciding factor is the difference in the pH at which each molecule is balanced (isoelectric point, pI). Using this difference, voltage is applied in a narrow capillary to line the antibodies up by position. The standard method, icIEF (imaged capillary isoelectric focusing), images the whole capillary as a single picture and reads the proportion of each species at once. Reference pI markers run together with the sample, so peak positions are also pinned down reproducibly.
Why is it needed? Because charge heterogeneity can affect how well the antibody works, and it is controlled as a critical quality attribute (CQA). It is the yardstick for consistency and stability: does each lot give the same profile, and does it stay put during storage? The key insight is that much of this heterogeneity is built in during culture. Downstream fixes rarely work, so when something looks off, you need to trace it back upstream.
This test separates and measures charge differences arising from deamidation, C-terminal lysine and similar changes. Differences in isoelectric point separate acidic species, the main species and basic species, and the profile serves as an indicator of manufacturing consistency and stability. Because charge variants can affect activity, they are controlled as a CQA.
Alternatives in practice
- Cation exchange chromatography (CEX) · Thermo Fisher MAbPac SCX-10 or Tosoh TSKgel SP-STAT column + HPLC (pH/salt gradient) — Charge variants can be fractionated. Identification is also possible with MS coupling
- Conventional cIEF (UV detection) · SCIEX PA800 Plus (cIEF kit, 280 nm UV detection) — Mobilizes instead of scanning the whole capillary. Less real-time
- Preparative icIEF + MS · Bio-Techne ProteinSimple Maurice Flex (preparative icIEF) + Orbitrap MS — Collects separated charge species for structural assignment by MS
Measures the types and ratios of Fc glycans, which shape effector activity and PK, to control glycoforms.
- Standard method
- PNGase F release → fluorescent labeling → HILIC-UPLC-FLR (with MS coupling as needed)
- Reagents and materials
- N-glycan release and labeling (RapiFluor-MS/2-AB), Analytical antibody enzymes
Why it is the standard. Releasing N-glycans, labeling them fluorescently and analyzing them by HILIC-FLD/MS is the reference method for glycoform quantitation. 2-AB labeling is widely recognized as the reference derivatization before HILIC. Glycan heterogeneity is a CQA that affects safety, efficacy and PK.
Check also. Glycans are critical attributes that shape effector activity and in vivo behavior, and they are readily influenced by manufacturing conditions (culture). Separation with fluorescent labeling suits the overall picture, but confirming individual glycan structures requires mass spectrometry as well.
About this step and the alternatives
The antibody's “tail” region (Fc) carries decorations called glycans, and they influence how strongly immune cells are drawn in. This step reads the types and proportions of those decorations.
First, an enzyme (PNGase F) cuts off the glycans, and a fluorescent tag is attached. Next, a hydrophilic column (HILIC) separates them by type, and the ratio of each glycan is measured. When the structure needs to be confirmed, mass spectrometry (MS) is added.
Glycan forms shift easily with culture conditions and bear directly on efficacy and in vivo behavior. That is why low-fucose forms and high mannose are monitored as critical quality attributes (CQAs). The key is the division of labor: fluorescence for the overall picture, MS for confirming structure.
Glycans attached to the Fc are critical attributes that shape effector activity and pharmacokinetics. Glycans are cleaved off enzymatically and fluorescently labeled to measure their types and ratios. This feeds directly into control of afucosylation, high mannose and similar attributes.
Alternatives in practice
- RapiFluor-MS rapid HILIC-FLR-MS · Waters BioAccord LC-MS + Glycan BEH Amide — 30-minute sample prep, with both fluorescence and MS detection. For routine release testing
- Middle-up HILIC-HRMS (subunits) · Thermo Vanquish + Orbitrap (IdeS/reduced subunits) — Analyzes glycans at the subunit level. Preserves the profile
- Hydrophilic anion exchange / mass spectrometry (sialic acids, monosaccharides) · Monosaccharide/sialic acid quantitation by Thermo Dionex ICS (HPAEC-PAD) or LC-MS — Direct quantitation of terminal sialylation and monosaccharide composition
Identity and structure confirmation by peptide mapping and intact mass
Structure confirmation at the core of identity testing: checks the sequence against the design, down to modification sites.
- Standard method
- Tryptic peptide mapping (LC-MS/MS) + intact/subunit mass analysis
- Reagents and materials
- Peptide mapping protease, Disulfide reduction alkylation reagents, Analytical antibody enzymes
Why it is the standard. ICH Q6B calls for peptide mapping as a basic method for sequence identification, primary structure characterization and QC. Tryptic digestion gives 85–95% sequence coverage, and combining multiple enzymes gives more than 98%. Intact mass confirms the overall mass with ±0.01% accuracy.
About this step and the alternatives
This step checks, at the molecular level, whether the finished antibody is “the molecule in the blueprint.” Both the sequence and the positions of any modifications are checked precisely against the answer key.
It works in two stages. First, enzyme scissors (trypsin) cut the antibody into short peptides, and high-resolution mass spectrometry (LC + MS) matches them one by one, filling in the sequence like a map. The mass of the whole antibody, as 1 intact molecule, is also measured to confirm the overall picture.
Here is where it pays off. Cutting the antibody up reveals exactly “at which amino acid” oxidation or deamidation (signs of degradation) occurred. It is the core of identity testing and the basis for quality specifications, lot-to-lot comparisons and change control decisions.
This step confirms precisely whether the amino acid sequence matches the design and where modifications have occurred. Digesting with an enzyme and running LC-MS confirms the sequence and also pinpoints sites of oxidation and deamidation. It is the core of identity testing.
Alternatives in practice
- Multi-enzyme mapping · Parallel digestion with trypsin + chymotrypsin + Asp-N on the Orbitrap/BioAccord above — Improves sequence coverage and confirms modification localization
- Middle-down / subunit mass (IdeS digestion) · Genovis FabRICATOR (IdeS) + Orbitrap/Xevo — Quick confirmation by Fc/Fab subunit masses
- Native MS / HDX-MS · Thermo Q Exactive UHMR (native MS) or Waters HDX system — Higher-order structure, conformation and epitope analysis
Host cell protein (HCP) quantitation
Quantifies residual CHO proteins, which bear on immunogenicity, and confirms they are below 100 ppm in drug substance.
- Standard method
- Sandwich ELISA (anti-CHO HCP polyclonal)
Why it is the standard. ELISA is the most common method for measuring total HCP from bioreactor harvest through each purification step. The target for drug substance HCP is below 100 ppm (ng/mg), typically about 10 ppm. Kits from Cygnus and others are the industry standard.
Check also. ELISA with polyclonal antibodies is the standard for measuring host cell proteins, but some proteins that the antibodies cannot recognize may go undetected. For important products, mass spectrometry is used as a complementary check for residual proteins that ELISA misses.
About this step and the alternatives
This step counts the “cell fragments” that have slipped into the product. When antibodies are made in CHO cells, trace host-derived proteins (HCP) inevitably remain, and their total amount is measured here.
The measurement is a sandwich immunoassay that captures HCP between antibodies from above and below. Polyclonal antibodies that recognize many kinds of HCP are used, and an industry-standard kit designed to capture about 97% reads the total.
The target is below 100 ppm in drug substance, and in practice around 10 ppm. However, HCPs that the antibodies cannot recognize are missed, so for important products mass spectrometry is used to check for gaps.
Residual CHO-derived proteins are process-related impurities that bear on immunogenicity and product stability. An ELISA using polyclonal antibodies measures the total amount to confirm it is held below 100 ppm in drug substance (often about 10 ppm).
Alternatives in practice
- Identification and quantitation of individual HCPs by LC-MS/MS · Thermo Orbitrap or SCIEX (DIA/PRM) + nanoLC — Identifies, by species, hard-to-detect HCPs that ELISA has trouble capturing
- Residual DNA (qPCR) · Thermo Applied Biosystems QuantStudio + resDNASEQ CHO kit — Sensitive quantitation of residual host-cell DNA (a separate CQA)
- Residual leached Protein A (ELISA) · Cygnus Technologies Protein A ELISA Kit (F400/F050H) — Quantifies residual Protein A from ligand leaching
Injectable biologics must meet endotoxin specifications, and this test is widely used for release.
- Standard method
- Bacterial endotoxin quantitation by kinetic chromogenic/turbidimetric LAL (Limulus amebocyte lysate) method
- Reagents and materials
- Charles River Endosafe LAL cartridges (Limulus lysate + chromogenic substrate), control standard endotoxin (CSE)
Why it is the standard. The Endosafe nexgen-PTS is cartridge-based, rapid, quantitative and GMP compliant, so it is widely adopted as the standard for in-process and release endotoxin control.
About this step and the alternatives
Has bacterial endotoxin, which triggers fever, slipped into an antibody drug that enters the body by injection? The test uses the fact that a component of horseshoe crab blood clots on contact with endotoxin, and this sentinel catches trace amounts.
The standard is the LAL method, which reacts the sample with Limulus reagent. The basic format is kinetic chromogenic/turbidimetric, following the reaction over time, and a dedicated cartridge quantifies in about 15 minutes. A control standard endotoxin (CSE) calibrates the scale to judge whether the result is within specification.
Endotoxin is a mandatory specification that decides whether a lot can ship, and it is monitored both at release and in process. The sample can interfere with the reaction, so the key is to confirm validity by dilution and spike recovery. Recombinant reagents are also an option if animal-derived materials are to be avoided.
mAb drug substance and drug product are injectables, so control of bacterial endotoxin, a pyrogen, is mandatory under the pharmacopoeias. This standard release and in-process test quantifies endotoxin by the kinetic turbidimetric/chromogenic LAL method and confirms it is within specification.
Alternatives in practice
- Recombinant factor C (rFC) method · Lonza PyroGene rFC assay or bioMérieux ENDONEXT/ENDOZYME II GO (fluorescence plate reader) — Sustainable, with no animal-derived horseshoe crab reagent. Fluorescence detection
- Gel-clot LAL (limit test) · FUJIFILM Wako/Charles River gel-clot LAL reagent + reaction tubes — Simple limit test. Less quantitative than kinetic methods
Binding activity and potency
Confirms target binding and function; potency is checked by binding kinetics and biological activity.
- Standard method
- Binding kinetics by SPR/BLI + biological activity by cell-based / reporter gene assays
- Equipment
- SPR surface plasmon resonance, BLI
- Reagents and materials
- SPR / BLI sensor chips, Reporter gene bioassay ADCC CDC
Why it is the standard. SPR measures binding kinetics and affinity label-free with high precision. Confirming potency with a cell/reporter system matched to the mechanism of action is the standard configuration for biological activity assessment under ICH Q6B.
Check also. Measuring binding strength (affinity) alone cannot guarantee actual efficacy (biological activity). Depending on the mechanism of action, cell-based functional assays and effector activity assessment must be combined and controlled as potency.
About this step and the alternatives
Does this antibody really “stick” to the intended partner, and does it really “work”? This quality attribute and analytics step confirms that by measurement.
First, binding is checked by surface plasmon resonance (SPR). The antibody flows over the target antigen fixed on a chip, and how fast it attaches and how tightly it holds are measured without labels. Biacore and Octet are the representative instruments.
But a tight grip does not guarantee that it works. So, depending on the mechanism of action, cell or reporter gene assays confirm biological activity as well, and the result is controlled as potency. This is the foundation of efficacy.
This step confirms that the antibody binds its target correctly and performs its expected function. SPR measures binding kinetics and affinity for the target, and depending on the mechanism of action, cell-based or reporter gene assays confirm biological activity.
Alternatives in practice
- ADCC/CDC bioassays · Promega ADCC/CDC Reporter Bioassay (FcγRIIIa/complement) + luminometer — Assesses effector function in line with the MOA. Cell-based and highly reproducible
- ELISA binding assay · Tecan/Agilent plate reader + antigen-coated ELISA — Simple and low cost. Gives no kinetics or affinity
- Flow cytometry cell binding · Beckman Coulter CytoFLEX or BD FACSymphony — Directly measures binding to cell-surface targets. Highly relevant to the MOA
The reference (“gold standard”) method is the one broadly recognized in published practice and guidelines; your process, product and regulator may justify another. Makers and products are examples, not endorsements. This map is translated from the Japanese edition, where it is updated first (Japanese version).