In early antibody discovery, the price per construct often drives procurement decisions. HEK293 cells are widely used at the High-Throughput Antibody Production stage because they are easier to transfect and cheaper per construct. Selecting a host system on upfront cost alone is a false economy because early savings frequently generate a far larger liability downstream.
A reliable High-Throughput Antibody Production Service built on the manufacturing-relevant host avoids that downstream liability from the first experiment.
This guide is written by Julia Pizzolato, PhD, Commercial Scientific Liaison at evitria. With a background spanning antibody engineering, emerging modalities, and translational science, she provides technical consultation on antibody design and development, connecting deep scientific understanding with strategic decision-making.
Where the Real Cost Accumulates
HEK293 is not the host used for clinical manufacturing. CHO is, accounting for over 70% of approved biologics. When discovery uses HEK and manufacturing uses CHO, the biological foundation of the project shifts mid-pipeline.
HEK and CHO cells produce divergent glycan profiles, with different sialylation patterns that affect antibody half-life, pharmacokinetics, and ADCC activity. Their chaperone environments differ, meaning a sequence that folds correctly in HEK may misfold or aggregate in CHO. Affinity and specificity data generated in HEK requires re-verification in CHO before any manufacturing decision can be made with confidence. The HEK293 vs CHO in HTP comparison details these glycosylation and folding differences.
The Risk of Resetting the Discovery Timeline
When host-switch liabilities surface late, they force a complete reset of the discovery timeline. The average capitalized cost of bringing a new drug to market exceeds 2.5 billion dollars[1]. Within that context, any failure that resets a discovery timeline represents a compounding loss at one of the most expensive phases of development.
Preventing host-switch liabilities at the screening stage is categorically cheaper than resolving them later. Maintaining CHO-Native expression from the first transfection eliminates this source of attrition. It also removes the need for any re-verification step at the transition between phases and preserves project momentum.
These same dynamics are examined in detail in why HTP screening results sometimes fail to translate to manufacturing:
The Glycosylation Problem Is Not Correctable in Post
A common assumption is that data generated in HEK can be adjusted or normalized when moving to CHO, but it cannot. Glycosylation differences between HEK and CHO are protein-specific and non-linear. No model reliably predicts how a HEK-produced sequence will behave in a CHO environment, and physical re-expression and re-characterization are always required.[2]
The functional consequences extend directly to therapeutic activity. Identical antibody constructs expressed in HEK and CHO can produce functionally distinct molecules, with differences in Fc-mediated biological activity that are directly attributable to host-specific glycosylation.[3] Host cell choice is not a secondary consideration: it determines the functional identity of the molecule.
This means that HEK-based HTP screening does not just carry a risk of rework. It structurally guarantees an additional validation step before any lead can advance with confidence into CHO-Native development.
Additional Costs That Are Often Overlooked
HEK293 cells are of human origin, which introduces a higher inherent risk of human viral contamination. This requires more stringent biosafety testing than CHO-derived material, adding cost and regulatory complexity that is rarely factored into the initial price comparison.
On sequence selection, screening in HEK effectively selects for sequences optimized for a HEK environment. Leads that rank highly in HEK may never reach commercial viability in CHO, meaning the screening campaign itself can point the program in the wrong direction.
The Practical Solution: CHO-Native HTP
The cost problem disappears when the screening environment matches the manufacturing environment. CHO-Native HTP screening generates data that is directly predictive of manufacturing behavior. There is no host-switch liability to accumulate and no mandatory re-verification step before development can proceed. Lead selection is made on biologically valid ground truth from the first experiment.
The additional cost per construct of CHO-Native screening is modest. Relative to the expense of a discovery timeline reset at a later development stage, it is straightforward prevention.
Strategic Guidance: Partnering with evitria
Our High-Throughput Antibody Production Service is built on a proprietary CHO transient platform, the same expression system used across all downstream scales. Candidates can progress from screening to larger-scale expression without host-system changes or re-optimization.
The workflow supports campaigns from 24 to several hundred constructs, with a standardized QC package covering titer measurement, HPLC-SEC, CE-SDS, and endotoxin measurement. Standard turnaround is approximately four weeks.
By combining Swiss-quality execution with 15 years of exclusive CHO transient expertise, evitria enables reliable data generation, efficient candidate selection, and confident progression toward downstream studies.
Frequently Asked Questions
Transient transfection in HEK293 uses simple, low-cost protocols. CHO transient expression requires more specialized platforms. The unit cost difference is real but small relative to the downstream risk it introduces.
Aggregation and changes in N-glycosylation are the two most frequently observed failure modes. Both can fundamentally alter the immunogenicity and pharmacokinetics of the antibody.
No. Glycosylation differences between HEK and CHO are protein-specific and non-linear. Physical re-expression in CHO is always required.
It is important for standard monoclonal antibodies and critical for complex formats such as bispecifics, fusion proteins, and ADCs, where folding and glycosylation are primary drivers of stability.
evitria’s HTP service delivers a standard turnaround of approximately four weeks, within a platform optimized over 15 years of exclusive CHO transient expertise.
Sources
[1] Joseph A. DiMasi, Henry G. Grabowski, Ronald W. Hansen, Innovation in the pharmaceutical industry: New estimates of R&D costs, Journal of Health Economics, Volume 47, 2016, Pages 20-33, ISSN 0167-6296, https://doi.org/10.1016/j.jhealeco.2016.01.012. Croset, A., Delafosse, L., Gaudry, J. P., Arod, C., Glez, L., Losberger, C., Begue, D., Krstanovic, A., Robert, F., Vilbois, F., Chevalet, L., & Antonsson, B. (2012). Differences in the glycosylation of recombinant proteins expressed in HEK and CHO cells. *Journal of biotechnology*, *161*(3), 336–348. https://doi.org/10.1016/j.jbiotec.2012.06.038 Patricia A Blundell, Dongli Lu, Anne Dell, Stuart Haslam, Richard J Pleass, Choice of Host Cell Line Is Essential for the Functional Glycosylation of the Fc Region of Human IgG1 Inhibitors of Influenza B Viruses, The Journal of Immunology, Volume 204, Issue 4, February 2020, Pages 1022–1034, https://doi.org/10.4049/jimmunol.1901145