The end of animal use for antibody generation in Europe. Really ?


On 2021 September 16, and based on the recommendation of the EURL ECVAM report, the European Parliament strongly voted (667 votes to 4) in favor of the transition to a research system that no longer uses animals. And of course nobody is in favor of unnecessary animal suffering.


But when we review the ECVAM report (https://ec.europa.eu/jrc/en/science-update/better-antibodies-without-using-animals), several sentences can only leave us a bit puzzled.


Let's look at an anthology of these statements.


Building a display library for monoclonals


« Non-animal methods for generating and producing antibodies have been available for years. One such method, based on so-called ‘phage display’ technology, is extremely versatile…»


Yes, phage display is available for years now. John McCafferty and Sir. Gregory Winter developed the concept of displaying exogenous proteins on the surface of the filamentous bacteriophage M13, showing the possibilities of building phage libraries.


And how do you build a phage display library?


By extracting RNA from spleen cells or from circulating lymphocytes before (naive library) or after immunization with the antigen of interest (immune library). So from an animal. For a so called non-animal method. We don't deny that synthetic biology is becoming more popular, but let's be honest : how many companies are able to generate completely synthetic librairies today ? 


Technical limits of display approaches


« … ‘phage display’ technology, is extremely versatile and can be used to generate a near-infinite range of high quality antibodies in a very efficient way »


Near-infinite range of high quality antibodies?


No, not exactly. And that's where the problem lies.


Working from B-cells of non immunized donors library, the host did not triggered a specific immune response against the antigen. So the generated antibodies have lower affinities compared to those generated during in vivo affinity maturation. Consequently, scientists are still trying to create larger libraries to circumvent this issue and increase the diversity. But with the limitations of transformation efficiency in the case of phage approach, the volume to expand the library, the quality of gene synthesis, the potential insertions or deletions that change the open reading frame and many more.


Noting the limitations of the naive library approach, you decide to go for an immune library generation to get access to in-vivo maturation. A solution is to start from biological material from sick people recovering from diseases for instance but with a biased antibody repertoire toward specific targets. This is typically what have been done for COVID and scientist eventually stumbled upon nice neutralizing antibodies.


But what if the patients are not in remission? What if the infection does not allow the collection of biological material?


You’re then compelled to inject animals. Back to square one.


Batch-to-batch reproductibility of antibodies manufacturing


«  Animal-derived antibodies typically suffer from batch-to-batch variability and many show low specificity towards the target molecule ».


Polyclonals, yes. Monoclonals, no.



In parallel to phage display, the technology of hybridomas generation was developed by George Köhler and César Milstein for mouse species, and by Hervé Bazin for rat species (the latter gave birth to the company SYnAbs). The goal was to allow an infinite and repeatable production of a monoclonal antibody of interest thanks to the progress of cell culture. One the clone is generated, the manufacturing becomes animal-free. And single cell B approach permits to go directly for a recombinant expression.


Why animal-derived monoclonal antibodies systematically outperform any equivalent



« Non-animal-derived antibodies typically exhibit highly desirable characteristics such as being able to form a strong bond with targets (binding affinity), having a long shelf life (stability) and the ability to be very selective (specificity), usually outperforming animal-derived equivalents »


In fact, it is the exact opposite.


One of the well-known limitation of non-animal derived antibodies is that the VH-VL natural pairing found in antibody-producing B cells are not retained in combinatorial libraries. This can potentially lead to suboptimal solubility or other biophysical characteristics.


Morevoer, in vitro libraries can express unnatural sequences that may conduct to developability issues such as aggregation, V-region glycosylation sites, poor expression yield in expression cells, weak thermal and long-term stability.


The diversity is not defined by the number of antibody variants but by the number of functional antibodies that can recognize the epitopes. And that makes all the difference.




If it is certain that progress must be made to limit the number of animals, this could first be done by a harmonization of European laws, concerning first the use of ascites (forbidden in Belgium, but not in France) or a progressive replacement of polyclonals (which are the largest consumer of animals) by monoclonals.


As for the generation of monoclonals against non-immunogenic small molecules or complex antigens such as transmembrane proteins, we are deeply convinced that nothing is better than to start with what nature has already made available for the creation of a complete and functional repertoire.


For SYnAbs, we have chosen to work with the rat species, a species considered as a nuisance in the sewers of cities, and eradicated via anti-coagulants generating a death in suffering.


Stopping the use of animals altogether seems to be a nonsense in the Pharmaceutical industry, and beyond the generation of very high value-added antibodies, we are not even talking about the imperative need for access to transgenic animal models for proof-of-concept and validation studies.