Three open competition grants for health research

Prof Dr Erik Koffijberg — Faculty of BMS

New medical tests can detect dozens of types of cancer from a single blood draw, or flag hundreds of abnormalities simultaneously on a single X-ray. The promise is great. But how significant is the downside? That question cannot be answered by the medical world right now: the methods to weigh the benefits of these tests against the harms simply don’t exist yet. Researchers from the TechMed Centre at the University of Twente and Amsterdam UMC will develop those methods for the first time.

You have a blood test done that checks for fifty types of cancer at once. A week later, your GP calls: the result is abnormal and further testing is needed. What follows are months of uncertainty, sleepless nights, scans, specialist appointments, and more tests. In the end, nothing is wrong. The test got it wrong.

For you, it’s an emotionally difficult period you won’t soon forget. For the healthcare system, it’s costs and valuable time from caregivers that led nowhere. Multiply this by the hundreds of thousands of people who would take such a test each year, and the bill adds up quickly. This scenario is not an exception. The more conditions a test tries to detect, the greater the chance it flags something that isn’t there. And the medical world has no good way yet to determine, across all the conditions being screened for, what the real consequences are.

Finding more is not the same as helping more

False positives are only one of the problems. Even when a test correctly identifies something, treatment is not always the right choice. Some conditions would never lead to symptoms or death even without treatment. In early-stage prostate cancer, for example, treatment is not always necessary. Yet there is a risk that detecting it more often leads to more treatment — without any health benefit. The Dutch Health Council warned about this in its report ‘Everyone Almost Ill’.

Testing too broadly can therefore have major downsides. “This makes people sick, even when they’re not,” says Prof. Erik Koffijberg. The problem is that the medical world cannot map this well. For tests aimed at a single disease, the trade-off between benefits and harms is already complex. For tests covering fifty diseases at once, the methods to make that assessment are entirely absent.

New methods for an invisible problem

That is precisely what the MERIT project aims to change. Prof. Erik Koffijberg from the TechMed Centre at the University of Twente and Prof. Mariska Leeflang from Amsterdam UMC are developing new methods together with Dr. Michiel Pegtel (Amsterdam UMC) and Dr. Merel Huisman (Radboudumc) to systematically map the benefits and harms of multi-disease tests. How reliable are these tests? When do the benefits outweigh the harms? And how do you determine for which patient groups deployment is worthwhile at all?

To answer those questions, the researchers are working with two concrete case studies: a blood test that simultaneously screens for more than fifty types of cancer, and an AI system that can detect more than a hundred different abnormalities on a single X-ray. Two extremes, with the aim of making the new methods more broadly applicable afterwards.

About the researchers

MERIT (Multi-disease tEsting: fRom Information to impacT) is led by Prof Dr Erik Koffijberg, professor of Technology Assessment of Digital Health Innovations at the University of Twente. He is affiliated with the Health Technology & Services Research (HTSR) group within the Faculty of Behavioural, Management and Social Sciences (BMS) and the TechMed Centre.

Koffijberg has more than twenty years of experience in evaluating diagnostic innovations, including tests, AI, and medical devices. He collaborates with Prof. Mariska Leeflang and Dr Michiel Pegtel from Amsterdam UMC, and Dr Merel Huisman from Radboudumc. 

Prof Dr Jurriaan Huskens — Faculty of S&T

Flu viruses mutate just enough each year to stay one step ahead of vaccines. To keep up, scientists need to understand how the virus binds to cells — but the very enzyme responsible for that process erases its own traces. Researchers from the University of Twente and Utrecht University will develop methods to catch the virus while it’s still in the act.

Flu viruses bind to cells via sugar structures on the cell surface. Two proteins on the virus play a key role: hemagglutinin (HA) handles binding, while neuraminidase (NA) cuts the connection so the virus can move on. To get a complete picture of how a virus mutates, you need both. But NA makes this difficult: it destroys its own binding sites while it’s active. You therefore have to catch the virus red-handed, so to speak.

From laboratory to flu season

Huskens and De Vries are developing new methods to measure both the binding via HA and the activity of NA simultaneously, something that is not currently possible. To do this, they combine chemical labelling with so-called glycan arrays: platforms on which sugar structures are arranged at controlled densities, allowing viral behaviour to be measured and compared with precision.

The methods Huskens and De Vries develop could open several doors. Better testing approaches for antiviral drugs, enabling more effective development in the future. Faster surveillance to track new virus strains — including animal variants that could become dangerous to humans. And better input for the annual selection of strains for the influenza vaccine.

About the researchers

The project is led by Prof Dr Jurriaan Huskens, professor at the University of Twente. He is affiliated with the Molecular Nanofabrication (MnF) group within the Faculty of Science and Technology (S&T) and with both the MESA+ Institute and the TechMed Centre. He collaborates with Dr Robert de Vries from Utrecht University.

Prof Dr Robert Passier — Faculty of S&T

Heart disease is the world's number one killer, claiming more than 17 million lives every year. A major reason is that the human heart cannot repair itself after injury. When heart muscle cells die, they are replaced by stiff scar tissue that weakens the heart and often leads to heart failure.

Remarkably, spiny mice (Acomys) are able to repair their hearts after damage without forming harmful scars. Instead, they create a flexible tissue environment that supports recovery. Our research has shown that fibroblasts  (support cells in the heart) from Acomys send protective signals to heart muscle cells, while fibroblasts from ordinary mice (Mus) send damaging signals.

The project is led by Prof Dr E. Van Rooij (Hubrecht Institute), who collaborates with Prof Dr Robert Passier (Faculty of S&T/ TechMed Centre) to uncover the key signals that make the difference between harmful scarring and healthy repair. Using advanced human heart models grown from stem cells, they aim to translate these insights into new therapies to protect patients from heart failure.