Science-Driven Analytics Are the Foundation of Robust Biologics Development
Biologics have moved from plasma proteins and antibodies to nucleic acids, viral vectors, antibody conjugates and cell therapies. The modalities have grown more complex, yet the core analytical questions remain strikingly consistent. Biologics characterisation that is built around those questions, and around the right method for each one, is what lets development teams move quickly without sacrificing the depth regulators now expect.
The Questions Stay the Same as Modalities Multiply
Whatever the molecule, drug product development rests on the same foundations. Teams need to characterise the product itself, from primary structure and size to post-translational modifications. They need to understand how it degrades, whether chemically or physically through aggregation and adsorption, what impurities are present, and under which formulation, packaging and stress conditions it stays stable. Bringing that together with clinical experience is what defines a genuine critical quality attribute. Some of those questions are modality independent: excipient analytics, pH, osmolality and particles apply to almost everything. Others are specific. For proteins, aggregation, chemical change and glycosylation dominate. For AAVs, the focus shifts to full and empty capsids, genome integrity and infectivity. Nucleic acid therapeutics bring their own concerns. A sound analytical strategy balances both kinds, and the balance changes with stage. Early discovery works with limited material and little prior knowledge, so developability is the priority. As programmes move towards the clinic and commercial supply, the emphasis shifts to thorough characterisation, CQA definition and, finally, formal release and stability testing. This consistency is an opportunity. Deep experience in one modality translates into insight for the next, and organisations that understand the science behind their methods can work efficiently without cutting corners.
Surfactant Analytics Must Match the Question Being Asked
Analytical readouts are only as useful as the question behind them, which is why method selection starts with what needs to be understood.
Most proteins are formulated with polysorbate 20, polysorbate 80 or poloxamer, and viral vector products increasingly use surfactants too. These excipients degrade, through metal-catalysed oxidation or hydrolysis by residual host cell proteins, and that degradation can compromise product stability and drive particle formation. Monitoring surfactant content and integrity therefore belongs in the analytical package, and the method has to fit the purpose.
The single peak method, a mixed-mode separation in which the surfactant elutes as one peak, is the industry standard for quick quantification and suits GMP release testing and in-process controls where degradation is not expected. In a stability setting, however, it can mislead. In an enzymatic degradation study of polysorbate 20, single peak and gradient methods gave different kinetics, because degradants still eluted within the single peak. Gradient separation with charged aerosol detection resolves the underlying species, shows how the profile changes, and can be paired with mass spectrometry to follow oxidation or hydrolysis markers. A two-dimensional LC approach adds speed. Online protein removal followed by gradient separation analyses polysorbates in around eight minutes and poloxamers in around ten, including sample preparation. Coriolis has applied it across marketed antibody products, including high-concentration formulations, recovering the claimed surfactant content and detecting enzymatic degradation during stability studies. It has also shown, in surfactant removal work, that polysorbate 80 is far harder to strip than polysorbate 20, and that removal is selective, with polyester species persisting. Only the chromatograms reveal that.
The method you choose must align with the question you are asking. Neither approach is superior in absolute terms. Both are correct when applied to their proper purpose.
Dr Andrea Hawe, Chief Scientific Officer, Coriolis Pharma
Particles Can Come From Anywhere, So Identification Matters
Particles are a critical quality attribute with documented links to product quality, safety, efficacy and immunogenicity. Visible particles still trigger clinical holds and product recalls, and FDA recall listings from the past few years show the problem is far from solved. They can come from the product itself through aggregation, from degrading excipients, from container closure systems and administration sets, or from the process. Regulators expect sponsors to characterise them across the nanometre, subvisible and visible ranges, and to build that profile through batch comparisons, forced degradation, stability, shipment and in-use compatibility studies. For visible particles, Coriolis isolates the particle under controlled conditions, then applies FTIR microscopy and, where needed, scanning electron microscopy for elemental analysis, scanning several points because many particles are mixtures. Identification is supported by an AI-assisted spectral library trained on thousands of pharmaceutically relevant reference spectra collected on the same systems, which avoids the irrelevant matches common in generic databases. In one antibody investigation, the library identified a combination of protein and PTFE, confirmed by fluorine in the elemental scan. The source was traced to a manufacturing line, and once the line was replaced, the problem disappeared.
Orthogonal Methods Reveal What Single Techniques Miss
In the subvisible range, methods differ sharply in what they reveal. Light obscuration and electrical sensing zone methods give size and count. Membrane microscopy adds images but loses particles such as silicone oil that pass through the filter. Flow imaging microscopy is more sensitive to transparent particles and captures up to a million images per run, and AI-based image analysis now turns that data into genuine classification.
No single technique covers the full particle size range, so a characterisation programme combines orthogonal methods from nanometre to visible.
In one study, four polysorbates subjected to the same enzymatic stress showed identical degradation by LC-CAD. After 14 days, AI-based flow imaging analysis showed that one produced markedly different, spherical particles. That difference would have gone undetected by chromatography alone. Below about five micrometres, where images lack resolution, total holographic characterisation adds size, refractive index and concentration, distinguishing silicone oil from protein particles and even air bubbles in syringe-expelled samples. For AAVs, the same thinking applies. Empty capsids raise toxicity concerns, so full, empty and partial species must be quantified alongside aggregation, titre, free DNA and capsid integrity. AUC, anion exchange, UV, TEM and mass photometry largely agree on empty versus full ratios, but AUC also resolves partial species and, with multi-wavelength analysis, DNA and capsid titres, revealing cases where an apparently full peak is actually overpacked.
Early strategic investment in the right analytical panel accelerates development by preventing late-stage surprises. That is where genuine competitive advantage lies.
Dr Amber Raines, Senior Director of Client Project Management, Coriolis Pharma
Tight clinical schedules push teams towards narrow, expedient analytics, yet experience points the other way. Running the same method on every project regardless of context creates false efficiency. Standardising the thinking means building strategy around molecular behaviour, likely degradation pathways, clinically relevant CQAs and a package matched to phase and modality. Coriolis has built that experience, from 18 years and more than 3,000 client projects, into Fast Track Analytics: standardised, platform methods that need no development, work across modalities and can be delivered in as little as a week. Surfactant monitoring is a good example of why timing matters. Starting early, even with a simple method, flags problems such as residual host cell proteins before they become costly, and regulators increasingly expect surfactant data from first-in-human onwards. Coriolis Pharma is a Munich-based CRDO specialising in biologics drug product development, with expertise spanning formulation, analytical characterisation and lyophilisation. It supports proteins, antibodies, viral vectors, nucleic acids and emerging modalities from early developability through to GMP testing.
This article was produced in partnership with Pharma D-mand and originally published on Biologics Analytical Development: Surfactants and Particles
More Articles
Articles
October 30, 2026
In Silico Formulation Development: What CMC Teams Can Decide Before the Lab
Publications
October 13, 2026
INFOGRAPHIC
Driving Success from Preclinical to Commercial Approval & Beyond
A guide for early and late-stage development, commercialization, and lifecycle management
Learn More
