October 30, 2025 | When patients are prescribed drugs with a narrow therapeutic window, where the difference between an effective and toxic dose can be frighteningly small, therapeutic drug monitoring is crucial to their safety. Outside of a hospital setting with expensive lab equipment and specialized staff to do the monitoring, things can get dicey, according to Alexis Vallée-Bélisle, Ph.D., chemistry professor at the University of Montreal and Canada Research Chair in Bioengineering and Bio-nanotechnology.
But many patients needing precision dosing eventually, if not immediately, return home—often on a complex regimen of medications. Very few blood tests currently exist that they can independently use to determine if the concentration of drugs in their body is within the safe and effective range or needs adjusting, he says, warfarin dosing and monitoring being one exception. Physicians can only “wish for the best” without any data to support whether the proposed dosage is optimal for a specific patient.
Vallée-Bélisle, and his startup company Anasens, are on a mission to change all that with an easy-to-use platform for detecting the concentration of molecules in a drop of blood or saliva in less than five minutes. Measurements are done with DNA-based sensors using inexpensive electronics like those in at-home glucose meters.
“We basically borrowed a strategy that was already employed by nature,” says Vallée-Bélisle in describing the technology’s novel, kinetically programmed “signaling cascade” making it suitable for therapeutic monitoring across drug types. The versatility and modularity of the approach was recently validated through a series of experiments and simulations, one in which four different molecules were detected from 5 microliters of blood taken from mice (Journal of the American Chemical Society, DOI: 10.1021/jacs.5c12059).
The commercial ambition is a universal multiplex platform for therapeutic drug monitoring that could be used in the doctor’s office or at home with the results sent directly to the patient’s healthcare provider. For now, the platform is being tested by police for detecting illicit drugs in saliva (a product called DrugAsens), Vallée-Bélisle says.
Next up is a device for therapeutic drug monitoring in blood, and Anasens is also working on a multiplex product for detecting several blood markers simultaneously. The DNA-based assay is enabling the company to develop sensors for multiple blood molecules even if they are at less than 100,000 times the concentration level of glucose.
In addition to therapeutic drug monitoring at home and in the clinic, Vallée-Bélisle, wearing his academic hat, says he also envisioned this platform being employed as a potential companion diagnostic to give clinical trial sponsors information about the pharmacokinetics (PK) of study participants and thereby improve the late-phase success rate of their lead compounds. It could likewise be of value to preclinical researchers doing PK studies in animal models.
In the latest study, Vallée-Bélisle and his team used the device to quickly determine the half-life of a drug (quinine) in mice. This stood in stark contrast to the gold-standard method, high-performance liquid chromatography, which took a couple of hours to produce readouts.
People metabolize drugs at different rates, depending on a host of factors that include genetics, how much they weigh, what they eat, and how much they exercise, says Vallée-Bélisle. This is much less of a concern when prescribing drugs with a wide therapeutic window where there is more room for minor dosing “misalignment.”
For therapeutic drug monitoring of patients taking drugs with a narrow therapeutic range—a list that includes certain cancer therapies, antibiotics, and immunosuppressants—the race is now on to bring DNA-based electrochemical sensors to market, he says. They have the advantage of providing rapid results using simple, portable, and low-cost devices, and, unlike central lab immunoassays and chromatography, can do so in real time no matter where the patient is. DNA aptamers (synthetic DNA strands) can also be designed to bind to a wide range of molecular targets.
What’s unique about the strategy being employed by Vallée-Bélisle is the signaling mechanism that gives his platform the virtues of a glucose meter, but without the necessity of creating enzymes to initiate the electrochemical process for each drug that needs detecting. Many patients need to take two or three drugs, and it would be unrealistic to expect them to use two or three instruments to do the measuring.
Others in the field have also employed DNA aptamers attached to a gold electrode and modified with a molecule that can exchange electrons with the electrode. “It remains, however, challenging to create one sensor for each application using that type of signaling mechanism given that each such specific aptamer-sensor requires careful density, dynamic and labeling optimizations,” he explains.
The signaling cascade technique admittedly involves “another level of complexity,” since the problem is being split into three parts to mimic the strategy of cells having a receptor on their surface and a signaling protein that travel into the nucleus to influence gene expression, says Vallée-Bélisle. His platform features components specializing in detecting a drug (the aptamer), recognizing if the DNA aptamer successfully attaches to its target molecule, and generating a strong electrical signal.
As demonstrated in the experiment with mice, little optimization is needed to adapt the assay for the detection of different molecules, he notes. Users need only put a drop of blood on the dry region of a test strip that serves as an electrode and initiate the app-controlled analysis. A palm-sized potentiostat device, akin to a glucose meter, measures the current generated by the electrochemical reaction.
“The three components are realized in blood and act as a signaling cascade, and we put that on an electrode and get the signal in less than five minutes,” says Vallée-Bélisle. From the standpoint of users, the instrument is simple and intuitive.
He and his team are effectively programming the signaling cascade that then does the rest of the work on its own, he says. They’re tapping into the preprogrammed biochemistry that makes cells divide and allow people to see and smell, which has been “naturally selected and evolved over a million years, and amazingly much more complex than the kind of nanotechnology the best researcher in the world can do.”
Like many researchers in the field, Vallée-Bélisle uses DNA as an electrochemical barcode. Since the DNA sequence will only bind the other specific DNA sequence that participates in the signaling cascade, multiplexing is another big advantage of the approach.
Four or five electrodes could potentially be used, each associated with “a specific DNA sequence that will bind a specific aptamer that will detect a specific drug,” says Vallée-Bélisle. “The idea of being able to detect different drugs in the same analysis, using the same drop of blood, was something very important to us because ... quite a lot of diseases out there, like cancer, need a cocktail of drugs.”
For Anasens, the opportunity is to come up with a chip with multiple electrodes, he continues. The assays then employ a signaling mechanism that allows for multiplexing, so the testing process isn’t cumbersome for patients, or too complicated to use.
Physicians may want the marker information, but they also don’t want to deal with too much data, Vallée-Bélisle says. That’s where artificial intelligence comes in to interpret the data and flag doctors when patients get out of their therapeutic range so they can take the appropriate corrective action.
Anasens was founded in 2020, in the early days of the pandemic, and initially secured funding for development of a COVID serology test, says Vallée-Bélisle. The pivot to therapeutic drug monitoring that began two years ago coincided with significant acceleration of the practice spurred both by the pandemic, which alleviated regulatory and clinical inertia, and advances in technology driven by point-of-care testing and remote patient monitoring.
Anasens is now “knocking on the doors” of large pharma companies about incorporating its therapeutic drug monitoring technology into their interventional clinical trials. Currently, dosing is often based largely on patient weight rather than pharmacokinetics because there has not been a tool for tracking the concentration level of a study drug of individual participants during their treatment, Vallée-Bélisle says.
That “big average” can jeopardize the outcome of pivotal phase 3 trials of drugs with a narrow therapeutic window, he continues. Some published studies reporting the drug concentration level of chemotherapeutics in participants have found that only half of people were in the right range.
With no practical means to control the concentration of drug levels for individual patients, companies are forced to abandon products that can be dangerous and potentially toxic for slow metabolizers. “Companies don’t even know the right concentration of the drug they need to have in blood; they just know the dosage ... on average they’re going to inject into a person based on their size,” says Vallée-Bélisle.
Field studies are now underway with a prototype device for illicit drug screening in saliva, he adds. Next steps include fine-tuning of the engineering to have a product that is “idiot-proof,” since errors are generally caused by users rather than the technology itself, as well as scaling up the manufacturing to enable the production of “hundreds of thousands” of the units each week. Hundreds of sensors per day are already being produced in the lab setting. Field studies for the blood prototype for therapeutic drug monitoring applications are scheduled for next year.
The fascinating and humbling realization he often shares is that “the nanotechnology system that has been designed by nature is still 20 years ahead of current human nanotechnology. We’re very far from designing systems that are as efficient ... as what nature has come up with.”
His company is exploring many potential use cases for the medical diagnostic platform and is inviting physicians everywhere to weigh in on which therapeutic drugs they would most like to be easily detectable in their patient population. “We can’t solve this problem alone,” Vallée-Bélisle says. Making at-home therapeutic drug monitoring more broadly available is what will help enable more personalized treatments, which is widely regarded as “the future of medicine.”