MiMédi: democratizing advanced therapy drugs

Many diseases affecting human beings remain difficult to treat: cancers, autoimmune diseases, inflammatory syndromes... To combat these pathologies more effectively, new treatments have been developed: advanced therapy drugs. The principle behind them is to take cells from a patient or a healthy donor, apply selection or enhancement processes to enable them to acquire new physiological properties, and then reinject these cells - or a product derived from them - into the patient, who can then fight the disease more effectively.

For example, French start-up MED'INN'Pharma aims to combat inflammatory diseases by exploiting a natural phenomenon: apoptosis. This process corresponds to a cell self-destruction mechanism, which leads to the emission of chemical signals that reduce inflammation. The idea here is to reproduce this reaction, by placing T lymphocytes in programmed death, thus creating an innovative drug.

While the principle may seem clear-cut and the benefits obvious, advanced therapy drugs come up against a major obstacle: their production cost. " Their manufacture calls on complex technologies, requires numerous skills and involves a huge number of steps," notes Olivier Lehmann, an engineer specializing in robotics at FEMTO Engineering. The environment in which they are manufactured must also meet the strict criteria of a cleanroom, with multiple controls to minimize the risk of contamination. Added to these difficulties are the new health constraints, which represent an additional burden. " At the end of the day, these drugs can cost 300 to 500,000 euros per dose ", sums up the engineer. A cost that would be prohibitive for marketing to the general public.

Reducing production costs for innovative medicines

The MiMédi (Microtechniques for Innovative Medicines) project was created with the aim of making these therapies more accessible. This collaboration brings together biologists and engineering specialists, as well as players from the academic and industrial worlds. The project brings together the Établissement Français du Sang, the University of Franche-Comté through two laboratories (UMR RIGHT and FEMTO-ST), the Besançon CHRU and FEMTO Engineering, part of Carnot TSN. In addition, there are six industrial partners: iLsa, Smaltis, AUREA Technology, Diaclone, MED'INN'Pharma and Bioexigence. The aim of the project is to rationalize the manufacture of advanced therapy drugs, in order to significantly reduce their production costs.

The first step was to analyze and understand the processes involved. " Biologists and engineering specialists don't really speak the same language," notes Olivier Lehmann. " So first we had to familiarize ourselves with each other's vocabulary and decipher the problems to be solved. This was an essential task, which enabled us to identify the main sources of cost and to uncover relevant avenues for rationalization. The MiMédi team then proposed two main areas for improvement, which led to the filing of several patents.

1) Efficient selection of target cells

In order to produce innovative therapeutic drugs, it is first necessary to isolate the cells which will then be modified (T lymphocytes, for example). " Today, the most common method involves centrifugation, followed by the addition of complementary products to separate the different layers," explains the FEMTO Engineering engineer.

Researchers have explored several avenues for improving this stage. The first is based on a different approach to centrifugation, limiting the use of separation products, which can be toxic if not properly removed.

Another solution we have developed enables cells to be separated efficiently without centrifugation. It uses the response of different biological structures to acoustic waves or electrostatic fields. In the latter case (dielectrophoresis), microfluidic chips are manufactured by assembling two electrodes one on top of the other. " This step requires extreme precision," stresses Olivier Lehmann. " That ' s why we use robots with nanometric resolution. We use them to position each half-chip in relation to the other, thanks to a vision system and periodic patterns previously applied to each one. The result is measurements accurate to hundredths or even thousandths of a pixel. These devices are then used to remotely control the position and speed of the cells.

2) Producing innovative drugs in closed systems

The research team has also focused on minimizing contamination risks, by developing a closed manufacturing system. Indeed, current conventional methods generally require the culture medium to be opened up, necessitating multiple sterility tests as production progresses. Here, on the contrary, all components operate in a closed system and are linked by sterile connections.

Even so, it is essential to be able to check the state of the product, without having to open it or take a sample. Here again, several solutions have been developed. " Firstly, we used spectroscopy," explains the engineer. " In concrete terms, this involves passing white light through the solution and studying the spectrum at the output. This information, in comparison with numerous reference samples, tells us about the concentration of cells or possible contamination by other structures. "

Another solution developed by the MiMédi teams is based on computer vision and machine learning algorithms. Using a neural network, these are able to identify the various cells present in the product circulating in the closed system, observed through microscopes. However, the artificial intelligence still needed to be trained. " There was no reference corpus, so we had to create one," says Olivier Lehmann. " We had to identify tens of thousands of cells by hand from a large bank of images. This labeled data then served as a training set for the machine, which adapted its neural network accordingly. "

This closed-system approach could even make it possible to dispense with a cleanroom during production. " Initial tests have shown a total absence of contamination via this manufacturing process ", says the engineer. At present, however, it is impossible to dispense with a cleanroom, as current standards require the use of one.

New methods already under study

The MiMédi project, completed in December 2022, led to the creation of a start-up to exploit the results obtained. CellQuest has developed a machine for producing "CAR-T cells", T lymphocytes treated to detect cancer cells more effectively, via the addition of markers. Although the process already existed, the company wanted to make it much more affordable, by dividing the manufacturing cost by ten, thanks to its automated solution. And initial results show that the CAR-T cells thus obtained meet requirements: the modified cells are present in the expected concentration and there is no trace of contamination.

Unfortunately, no patient has yet been able to benefit from the innovative drugs produced in this way. Before they can even enter clinical trials, they have to receive authorizations, which take several years to obtain.

However, this does not discourage the MiMédi team, which intends to continue its research to optimize the production of innovative therapy drugs. " We are currently working on a new project, with the same partners, which could start at the end of 2023," says the FEMTO Engineering engineer. " In particular, we want to develop tests, at a reasonable cost, to verify that patients are receptive to these drugs or to ensure donor compatibility. " New avenues could be explored, such as measuring the impedance of a complex liquid medium, i.e. its electrical response to a low-intensity current. This could be a decisive new marker for the democratization of innovative drugs.