Meyer-Lindenberg A. BMBF - Bundesministerium für Bildung und Forschung 01ZX1904A: COMMITMENT - COMorbidity Modeling via Integrative Transfer machine-learning in MENTal illness. 09/2019-08/2022.
Psychotic disorders, including schizophrenia and bipolar disorders, comprise some of the most severe mental illnesses that cause an enormous clinical and healthcare burden, costing close to €100 billion annually in Europe alone. The diagnostic delineation of these conditions is clinically defined and does not index appropriately the underlying biology. Patients are treated largely by a “one-fits-all” approach, despite substantial clinical heterogeneity in course, treatment response and presence of somatic comorbidities that include type 2 diabetes, cardiovascular diseases and neurodegenerative processes. There is a strong need to identify biological means to stratify patients with psychotic disorders and identify the biological basis of somatic comorbidity. This will allow improved clinical delineation of psychotic illnesses and facilitate novel intervention strategies targeted at the minimization of comorbidity risk, reducing mortality and morbidity. To address this, COMMITMENT will develop an innovative computational Framework for stratification of psychotic disorders and identification of biological domains shared with comorbidities. COMMITMENT builds on distributed machine learning that integrates mechanistic information mined from the literature. It extends approaches successfully used for oncological systems-medicine investigations, to optimally extract disease signatures from partially overlapping multi-OMICs data. We will use massive-scale genetic and neuroimaging data to explore the lifespan trajectories of stratification and comorbidity profiles, to identify age periods with pronounced comorbidity risk and to disentangle state- from trait-effects. We will explore the predictivity of comorbidity profiles for illness course, treatment response and occurrence of comorbidity during early illness phases. With this, COMMITMENT will provide the basis for biologically-informed clinical tools for improved personalized care of patients with psychotic disorders.
BMBF - Bundesministerium für Bildung und Forschung 01KU1905A: IMPLEMENT - Improved Personalized medicine through machine Learning in Mental disorders. 05/2019-04/2022.
Psychotic disorders are severe mental illnesses with early onset, frequently chronic course and often lifelong impairment. As a consequence, they cause an enormous healthcare burden, costing close to €100 billion annually in Europe alone. The biology of these illnesses is insufficiently understood and no objective tools exist to aid in diagnosis or treatment selection. This leads to long periods of inadequate and ineffective treatment, significantly limiting the opportunity for achieving more optimal clinical outcomes. To address this, IMPLEMENT will develop a translational research framework that identifies biomarkers for treatment-relevant stratification of the most severe psychotic disorder, schizophrenia. Building on known candidates, IMPLEMENT will use advanced machine learning on high-dimensional multi-OMICS and brain scans to identify illnessassociated profiles indexing patient subgroups. Using big data approaches (n > 60,000), IMPLEMENT will explore the impact of genetic risk and neurodevelopmental processes on the formation of biological subgroups and use clinical studies of conventional antipsychotic treatment and innovative treatment approaches to tune subgroup profiles towards clinical utility. The IMPLEMENT framework will incorporate preclinical validation to leverage neurobiological understanding and optimize biological subgroup profiles. The clinical utility of these profiles will be validated in independent clinical samples and prospectively recruited subjects. IMPLEMENT will integrate these efforts with ICT development, to optimize the use of high-dimensional datasets across diverse repositories, to optimally harmonize data for personalized medicine investigations and safeguard patient privacy. Overall, IMPLEMENT will provide the basis for biologically-informed personalized medicine approaches in schizophrenia, addressing an enormous unmet medical Need in an area of medicine in which currently no robust clinical stratification tools exist.
Schwarz, Ph.D E. DFG - Deutsche Forschungsgemeinschaft SCHW 1768/2-1: Fortführung der Nachwuchsgruppe im Emmy Noether-Programm: Diagnose-übergreifende Rekonstruktion psychotischer Störungen durch multimodale genetisch-neuronale Signaturen. 02/2019-01/2020.
Schwarz, Ph.D E. SCHW 1768/1-1 Emmy Noether-Programm: Diagnose-übergreifende Rekonstruktion psychotischer Störungen durch multimodale genetisch-neuronale Signaturen. 08/2014-07/2018.
Psychotische Störungen gehören zu den führenden Ursachen für den Verlust an gesunder Lebenszeit, Arbeitsunfähigkeit und Frühberentung. Die derzeitige Diagnostik basiert auf Symptom-Verlaufskriterien geringer biologischer Valididät, was die evidenzbasierte Auswahl geeigneter Therapieoptionen erheblich erschwert. Eine Vielzahl von Studien zeigt entsprechend, dass die Gruppe der Psychosen ätiologisch heterogen ist, und zwischen Schizophrenie und der bipolaren Störung erhebliche Überschneidungen genetischer, molekularer oder bildgebender Faktoren aufweist. Da es sich um hochgradig heritable Gehirnerkrankungen handelt, sollte sich aus genom-weiten genetischen Daten in Kombination mit einer Charakterisierung der Hirnfunktion und Struktur eine verbesserte, biologisch basierte und therapierelevante Taxonomie dieser Erkrankungsgruppe entwickeln lassen. Hierzu werden in diesem Vorhaben neue multivariate statistische Verfahren entwickelt und eingesetzt, um diagnose-übergreifende Untergruppen psychotischer Patienten mit Hilfe polygener genetischer Signaturen zu identifizieren und deren neurobiologische Grundlagen auf der Systemebene durch multimodale Bildgebung zu untersuchen.
Meyer-Lindenberg A. EU - Europäische Union HEALTH-F2-2013-602450: IMAGEMEND. 10/2013-09/2017.
Mental disorders are leading causes of disability, absence from work and premature retirement in Europe. While magnetic resonance imaging (MRI) facilities are broadly available and a vast research literature exists, few neuroimaging applications have reached clinical practice in psychiatry. A major problem is that mental illnesses are currently diagnosed as discrete entities defined clinically. Instead, recent results show that mental disorders are best understood as quantitative alterations in neural systems relevant across traditional diagnostic boundaries that reflect individual, genetic and environmental risk factors. In the IMAGEMEND consortium, we aim to discover these systems to identify the patient characteristics most relevant for treatment, derive biomarkers and decision rules from this systems-level dimensional account, and systematically validate biomarker panels in patient, high-risk and epidemiological samples to produce automated imaging-based diagnostic and predictive tests tailored for wide distribution throughout Europe in standard clinical settings. Focusing on schizophrenia, bipolar disorder and attention deficit-hyperactivity disorder, we have assembled one of Europe’s largest datasets combining neuroimaging, genetic, environmental, cognitive and clinical information on approximately 13.000 participants, and have recruited international replication datasets of more than 30.000 people. This unique resource will be processed using a new generation of multivariate statistical analysis tools to optimize existing imaging technology for the benefit of patients. We will also develop new imaging technology to enable the direct imaging-based therapeutic modification of neural circuits through rapid real-time MRI. Our deliverables will promote personalized treatment through more accurate patient stratification, allow diagnoses at the pre-symptomatic stage for early intervention and prevention, and improve prediction of treatment response and disease progression.
Meyer-Lindenberg A. EU - Europäische Union 115008: IMI JU NEWMEDS: Novel Methods Leading to New Medications in Depression and Schizophrenia. 09/2009-08/2014.
Despite remarkable advances in molecular and imaging technologies and nearly 15,000 articles on schizophrenia and depression (S&D) every year, there have been few truly innovative new chemical entities (NCEs) which have made it to the clinic. While there has been a tremendous explosion of new knowledge: dozens of single-nucleotide polymorphisms linked to disease, hundreds of new molecules and pathways identified, numerous imaging findings differentiating patients from controls, yet, it has been hard to take these findings from the bench to the clinic. We think there are three major bottlenecks that are holding the field back: i) a lack of pathophysiologically-accurate animal models guiding the drug discovery of NCEs; ii) a lack of tools and tests in healthy volunteers that can provide early indication of efficacy; and iii) the reliance of clinical trials on symptom-based DSM-categories which inevitably lead to biologically heterogeneous groups of patients. To overcome these limitations, we have brought together a consortium of six leading European and an Israeli academic institution (which bring expertise in animal models, genetics, functional MRI and PET imaging, clinical settings and analysis methods) and two SMEs (which bring expertise in high-throughput genetics, transcriptomics and proteomics) who will partner with the dozen EFPIA partners in the NEWMEDS consortium. To specifically target the challenges identified in Call 10, the NEWMEDS consortium will: a) develop animal models that focus on reliable cross-species endophenotypes (e.g., cognitive function, electrophysiology) and use crossspecies methods (small-animal MRI, EEG and micro-PET) to bring animal models closer to clinical endpoints; b) validate the use of fMRI-based paradigms as early and surrogate markers for efficacy; and to combine this with PET approaches for measuring changes in endogenous transmitters – thus providing new methods that can be implemented in small Phase 1B studies in healthy volunteers to provide guidance for drug development; and c) identify pharmacogenetic biomarkers that can be used to stratify patients within an umbrella DSM-diagnosis, thus allowing for targeted clinical trials, individualized treatment and back-translation of subgroup-specific biomarkers into preclinical drug discovery. To increase the chance of a breakthrough we will implement new analytical approaches – the use of support vector machine learning algorithms for image analyses; the use of Bayesian and growth mixture models for more meaningful analyses of clinical trial data. The project will be delivered through a series of integrated workpackages organized in three clusters – preclinical models, imaging methods, and biomarker development as exemplified in Figure 1. Our consortium has achieved its 1:1 in-kind match, indicative of the involvement and commitment of all EFPIA partners. One of Europe’s leading scientific management SMEs (GABO:mi) will facilitate the management of NEWMEDS and a distinguished international Scientific Advisory Board will provide input and guidance. To ensure that we maximally integrate with other ongoing international initiatives, we have commitments of collaborations from several international consortia and experts (e.g. MATRICS, NIH Biomarkers Consortium). By the end of the 5 year project we expect to provide ready to use new cross-validated animal models, new fMRI methods with dedicated analysis techniques, new PET radioligands, as well as new genetic and proteomic biomarkers for patient-segmentation or individual response prediction. These tools should provide our EFPIA partners with an added competitive advantage in developing new drugs for S&D.