Immune diseases and machine learning feature in new IMI Call for proposals

The Innovative Medicines Initiative (IMI) is launching a new EUR 82 million Call for proposals focusing on immune diseases, imaging, machine learning, and digital clinical trials.

Better control of immune-mediated diseases

Many patients with immune-mediated diseases do not respond at all to current treatments. A major challenge for researchers is to understand which patients are most likely to respond to which treatments. This topic aims to add to our understanding of these diseases and identify biomarkers (clues) that could predict different patients’ responses to treatments. They will also carry out early-stage clinical trials with a view to identifying new patient-centric treatment approaches. The project will focus on the following diseases: lupus, rheumatoid arthritis, multiple sclerosis, ulcerative colitis, Crohn’s disease, asthma, and chronic obstructive pulmonary diseases (COPD). The project will ultimately make it easier to provide the right treatment to the right patient, thereby increasing the proportion of patients taking a treatment that works for them.

Non-invasive imaging of immune cells

Immune cells play a key role in a wide range of disease areas affecting many parts of the body. Today, assessments of immune cells involve taking blood samples or biopsies, which involves invasive procedures. This topic will pave the way for the wider use of (non-invasive) imaging technologies for the quantitative and qualitative assessment of immune cells. They will do this by developing and characterising imaging agents for different kinds of immune cell, as well as investigating how different imaging modalities (e.g. magnetic resonance imaging, positron emission tomography) can deliver quantitative data. Ultimately, molecular imaging of immune cells will make it easier to assess which patients are most like to benefit from certain treatments. It will also make it easier to assess the efficacy of drugs during clinical trials.

Machine learning for drug discovery

The vast amounts of data generated during drug discovery are not always used optimally, meaning scientists miss out on opportunities to make new discoveries using old data. Big data analysis and machine learning approaches could uncover valuable information for researchers. The goal of this topic is to establish a machine-learning platform for use in drug discovery. Crucially, this platform would respect organisations’ confidential data and assets, which would never leave the control of the respective data owners. The platform would also take a federated machine learning approach, meaning that the learning effort is not centralised but spread over different, physically separated partners. The project will test the platform in an industry setting and publish guidelines and advice on how to address various challenges related to machine learning in drug discovery.

Towards better, more patient-friendly clinical trials

Recruiting patients to take part in a clinical trial can be very difficult; many patients are understandably put off by how far they would have to travel to the clinical site, and how often they would be expected to make the trip. Digital technologies and wearable devices mean it is now possible to assess patients remotely – while they are at home, or going about their daily lives. If used during clinical trials, they could dramatically reduce the number of times patients are expected to visit the clinic. This topic will assess the feasibility of running so-called ‘remote, decentralised’ clinical trials, using these kinds of devices, in Europe. If successful, remote, decentralised trials would make it easier for patients to participate in trials, and this in turn would result in a more diverse trial population. They would also increase the frequency and quality of data collection.



Share this...