SCIENTISTS from Europe have introduced a powerful new research platform that may reshape the way antibody-based medicines are tested before reaching patients. The system, created by an international consortium led by Belgium’s VIB and Ghent University (UGent), is designed to overcome long-standing weaknesses in conventional drug testing. Published in Science Immunology, the study demonstrates how differences between animal and human immune systems can hide dangerous side effects or exaggerate benefits, often until clinical trials begin, News Medical reported.
Antibody therapies have become one of the most important tools in modern medicine. They are used to fight cancer, control autoimmune diseases, and combat infections. In laboratories, many experimental antibodies appear highly effective and safe. Yet when they enter human trials, a large proportion fail or cause serious complications. These failures are not only costly for drug makers but also slow the arrival of new treatments and expose patients to unnecessary risks.
According to the researchers, the main reason for these setbacks lies in biology. Standard preclinical models, such as cell cultures, mice, and non-human primates, do not fully reflect how the human immune system responds to antibody drugs. As a result, early tests can give a false sense of security or promise, masking problems that only become visible later.
Why antibodies behave differently across species
Most therapeutic antibodies are based on Immunoglobulin G (IgG). While the front end of the antibody attaches to a specific disease target, the tail section, known as the Fc region, communicates with the immune system. This Fc region interacts with Fc gamma (Fcγ) receptors found on immune cells. These receptors determine whether immune cells destroy cancer cells, remove harmful material, or reduce inflammation.
READ ALSO: Cancer: Triple-drug strategy wipes out pancreatic tumours, blocks resistance in preclinical models
However, Fcγ receptors are not the same across species. Their structure, distribution, and strength of interaction vary widely between humans, mice, and other animals. Because of this, the same antibody can trigger completely different immune responses depending on the test system. An antibody that seems safe and effective in mice may behave unpredictably in people.
Senior author and researcher at the VIB-UGent Center for Inflammation Research, Prof. Bart Lambrecht,said the new platform brings researchers much closer to real human biology. He explained that studying antibodies in a system that mimics human immune behaviour allows scientists to make more reliable decisions at earlier stages of development.
Hidden dangers in conventional testing
One of the most serious findings in the study involves human platelets, which play a central role in blood clotting. Certain antibodies can directly activate human platelets, increasing the risk of dangerous clots. This pathway does not exist in mice, meaning traditional animal models cannot detect this risk.
The consequences of this gap have already been seen. Anti-CD40L antibodies once showed promising results in early tests and were moved into clinical trials. Unfortunately, patients later developed severe blood clots, and some died. The biological mechanism behind these reactions was invisible in standard preclinical studies because it was unique to humans.
In addition, other immune cells respond very differently depending on the species used. Some cell types that react strongly in humans show weak or no responses in laboratory models. In some cases, immune cells even behave in opposite ways, leading researchers to draw incorrect conclusions about safety and effectiveness.
Building a model around human immune behaviour
To solve this problem, the research team first created a comprehensive map of how Fcγ receptors are distributed across immune cells in humans and in commonly used laboratory systems. Using this blueprint, they developed a new genetically engineered mouse model that mirrors the human Fcγ receptor network.
Unlike earlier ‘humanised’ mice, which often rely on partial gene replacements, this model uses a precise genetic knock-in strategy. This ensures that antibodies interact with immune cells in a way that closely resembles what happens in the human body, including during inflammation.
The model has also been made commercially available, ensuring that both academic scientists and pharmaceutical companies can use it. This accessibility is expected to accelerate adoption across the drug development industry.
Earlier clarity, better decisions
The platform has already been tested in multiple disease settings. It allows researchers to rank antibody candidates based on their true biological effects, measure how effectively immune cells are removed, and evaluate how well treatments slow disease progression, particularly in cancer.
The study’s first author from VIB-UGent and UZ Gent, Dr. Karel Van Damme, said modern antibody design often depends on small molecular changes. The new system enables different designs to be compared directly under realistic conditions, reducing trial-and-error approaches.
Benefits beyond the laboratory
The platform offers significant economic advantages. By identifying ineffective or dangerous antibodies earlier, companies can avoid costly late-stage failures and reduce research timelines. This leads to lower development costs and faster delivery of new medicines.
For patients, the impact is equally important. More accurate early testing improves safety and increases the likelihood that only the most promising therapies reach clinical trials. The approach also aligns with regulatory trends, as agencies such as the U.S. Food and Drug Administration increasingly support the use of predictive, human-relevant models.
The research was carried out by a partnership involving VIB-Ghent University, argenx (Belgium), genOway (France), and Innate Pharma (France).
