If a tumour has succeeded in nesting in a living being's brain, it has, from the tumour’s perspective, been particularly clever. It has hidden behind one of the body's strongest barriers, protecting its most crucial organs: the blood-brain barrier, a very selective filter that only allows certain substances through. Most medications are not included. Therefore, finding an effective chemotherapy against brain tumours poses a significant challenge for medicine.
In recent years, medical research has found a promising ally: nanotechnology. Materials on the nanoscale can, figuratively speaking, take on the role of postmen delivering active substances to the desired address. Since nanoparticles are unimaginably small—about 500 times smaller than the diameter of a human hair— some of them manage to pass the body’s protective barriers without damaging them. To stay with the brain tumour example: nanoparticles could transport chemotherapeutic agents through the blood- brain barrier into the brain, where they can then combat the brain tumour.
Searching for
the right nanomaterial
However, nanoparticles must meet very specific properties depending on the task they are to fulfil:
depending on their shape, material composition, and size, they distribute themselves differently in the
body and accumulate in different organs. It is therefore important to find out which particles perform
their task best without causing harm. So far, researchers have used animal models, typically mice, to
investigate these questions: they administered various nanomaterials to mice and then examined how
these spread in the mice's bodies and what side effects they had. These animal studies, however, are
not only laborious, lengthy, and costly but also problematic from an ethical perspective. It's no wonder
that the Swiss Animal Welfare Act demands that the number of animal experiments used be kept to
the necessary minimum.
AI mouse with a decisive advantage
Empa researcher Jimeng Wu, a doctoral student in the 'Nanomaterials in Health' and 'Technology and
Society' departments, has therefore developed a virtual mouse that allows these tests to be
conducted much more quickly using AI. For this so-called physiologically based pharmacokinetic
model (PBPK model), Wu used 18 mouse studies as a basis, data from various research team's
experiments on 'real' mice. Additionally, she integrated a statistical method, Bayesian analysis with
Markov chain Monte Carlo simulations, into her model.
The result is a virtual mouse that can be administered—also virtual—nanoparticles. The model then calculates their distribution in the mouse body based on their properties such as size, coating, and surface charge. Compared to a traditional PBPK model, which is calibrated for only one substance at a time, Wu's AI mouse has a decisive advantage: "The model can adjust its parameters to the measurable properties of the respective nanoparticle," explains Jimeng Wu. This ability is thanks to the tool's 'multivariate linear regression model,' a machine learning approach.
Contribution to 'Safe and Sustainable by
Design'
"This AI-supported screening tool allows researchers to virtually test which type of nanoparticles are
best suited for a specific task before they even manufacture these particles," continues Jimeng Wu.
This not only saves time but also costs, as it provides decision-making aid before starting an
expensive clinical trial.
"Thus, the model contributes to the concept of 'Safe and Sustainable by Design' (SSbD)," adds Peter Wick, who, along with his colleague Bernd Nowack, supervises Jimeng Wu in her doctorate. Because the virtual mouse increases the safety of new materials or therapies even before their development. However, the Empa researcher cautions that the dataset the model has been trained on is still very small: so far, only 18 peer-reviewed papers have been found whose data quality sufficed. "In many studies, the properties of the nanoparticles used are not sufficiently described," he notes. It is now necessary to feed the virtual mouse with additional study data and verify it to further increase the reliability of its predictions. "Our long-term goal is to shorten the process of developing nanomedical materials until their use as a medication in patients, while ideally being able to do without animal testing," he emphasizes.
Adapting the model for
human research
Jimeng Wu's future research work will also focus on a so-called 'bridge strategy' to transfer the
principle of her in silico model to human research. For this, she plans to embed the principles of the
virtual mouse into a human PBPK model. Unlike her AI mouse, which only calculates the distribution
of nanoparticles in the liver, kidneys, lungs, and spleen, a human in silico model could also be used to
examine sensitive target organs—for example, to research the extent to which certain nanoparticles
can overcome the blood-brain barrier. Even the initially mentioned brain tumour should no longer feel
safe behind this barrier—nanoparticles could bring it a parcel with a tailored dose of chemotherapy in
the role of 'postmen.'
Media Contact:
Mirjam Schwaller
Communication
Tel. +41 58 765 4386
redaktion@empa.ch