The PRISM2 study, which features in the EPND Hub, recruited 182 participants across Alzheimer's disease, schizophrenia and major depressive disorder, alongside matched healthy controls. It has generated a multimodal dataset and biosample collection designed to show how brain network integrity relates to social dysfunction across conditions, rather than within a single diagnosis. In this Cohort Profile, Martien Kas, co-Lead of PRISM2 and Professor of Behavioural Neuroscience at the University of Groningen, explains the goals, design and findings of the study.

What are the goals of the PRISM2 study?

Neuropsychiatric disorders are still largely diagnosed and treated on the basis of symptoms, with little reference to quantitative biology. PRISM2 was set up to address that gap. The study builds directly on the first PRISM project, which identified quantitative neurobiological parameters linked to schizophrenia, Alzheimer's disease and to social function irrespective of diagnosis.

PRISM2 has three objectives. The first is to determine whether the transdiagnostic relationship between default mode network (DMN) integrity and social dysfunction, observed in PRISM1 in schizophrenia and Alzheimer's disease, can be reproduced and generalised to major depressive disorder. The second is to test whether quantitative variation in DMN integrity is causally related to social dysfunction. The third is to translate the results into benefits for patients and healthcare providers.

Underlying all of this is a transdiagnostic approach. We group participants by symptom domains that cut across conditions, not by diagnostic label alone. That is what makes the resulting dataset relevant to researchers in both neurodegenerative and psychiatric disease.

What disease areas and stages are included, and how many participants are involved?

The PRISM2 dataset covers 182 participants, spanning Alzheimer's disease, schizophrenia, major depressive disorder and matched healthy controls. The study was conducted in Spain and the Netherlands. Participants attended on two separate days for screening, clinical interviews and assessments.

Social withdrawal is one of the earliest symptoms across all three conditions, and appears to be an early indicator of cognitive difficulties. Recruiting across these disease areas in a single protocol allows that shared symptom domain to be studied directly.

What types of data and samples are being collected?

All participants provide demographic and clinical information. Assessments include task-based and resting-state functional MRI, EEG, behavioural measures and digital measures collected through a smartphone application developed in the first PRISM project. Blood samples were also collected.

The associated biosample collection listed on the EPND Hub comprises samples from 169 donors. Serum, plasma and DNA samples are available, each collected at a single timepoint. For healthy participants, the collection holds 62 serum, 62 plasma and 62 DNA samples. For participants with Alzheimer's disease, it holds 33 serum, 33 plasma and 33 DNA samples, with full counts for the other disease areas available on the Hub listing.

The samples are accompanied by linked data, including clinical information, demographics, imaging data, digital data, functional ratings and neuropsychiatric assessments. Researchers can therefore combine biological measures with detailed clinical and behavioural characterisation.

How can the EPND Hub advance neurodegeneration research?

Data sharing is essential. It allows the scientific community to test new hypotheses, to validate previous findings independently, and to work with unique datasets that would otherwise sit unused. It also ensures more efficient use of the research funding invested in collecting them.

The EPND Hub gives researchers access to a wide range of relevant clinical datasets in one place, which makes cross-cohort and cross-disease-area studies practical rather than aspirational. For a project like PRISM2, listing on the Hub also supports long-term sustainability, keeping the data and samples discoverable and requestable well beyond the end of the funding period.

What I hope researchers do with these data is twofold. First, that they build on the transdiagnostic, biology-informed approach that looks past traditional diagnostic categories. Second, that they use the data to test meaningful new hypotheses about the biology of shared symptom domains. The aim is precision diagnostics and treatment, bringing the right therapy to the right person at the right time.

Want to learn more? Discover the PRISM2 listing in the EPND Hub, here, and the associated biosample collection, here. You can also read our full interview with Martien Kas, here, or visit the PRISM2 project website.

The PRISM2 project has received funding from the Innovative Medicines Initiative 2 Joint Undertaking under grant agreement No. 101034377. The JU receives support from the European Union's Horizon 2020 research and innovation programme, EFPIA, and Cohen Veterans Bioscience.