Back to search

BEHANDLING-God og treffsikker diagnostikk, behandling og rehabilitering

Prediction of antipsychotic drug response; a personalised medicine approach in schizophrenia

Alternative title: Prediksjon av antipsykotika effekt - en persontilpasset tilnærming for schizofreni.

Awarded: NOK 12.0 mill.

Mental disorders constitute a great burden for those affected, in addition to large costs for the health care system, both here at home and in Europe. Today's treatment options do not work for everyone, and many people struggle with side effects. We will use biomarker profiles to optimize the use of existing drugs. Then we can reduce side effects and increase the treatment effect for patients with mental disorders. In order to achieve such precision psychiatry, we will use large amounts of data from the health service and registers, as well as new methods such as machine learning and artificial intelligence. The development of medicines has made great strides for several diseases in recent years. But this development has not happened in mental disorders. The drugs that are used today to treat schizophrenia, bipolar disorder and depression are old, give great variation in effect, while they can often cause major side effects. Then we need to make sure that the drugs that are already available are used better. We must give the right doses and ensure the least possible side effects. Instead of waiting for new medicines sometime in the future, we would rather adapt existing medicines better based on data we already have access to.

-

Mental disorders, including psychotic disorders, are leading global causes of morbidity and are among the most costly human disorders. The recent gene discoveries in schizophrenia can lead to major health benefits, through better treatment stratification strategies - unleashing the large potential for personalised medicine in schizophrenia. We will identify genetic factors associated with schizophrenia that overlap with antipsychotic drug targets in protein-protein interaction networks, building on findings from the large international genetic consortia, protein interaction repositories, and novel Bayesian statistical tools. We will take advantage of Norwegian biobanks and a wealth of data in health registries, health surveys and medical records, as well as large international cohorts of more than 65,000 genotyped patients, of which 25% are non-responders to antipsychotics. We will use these large samples to build the stratification model (training), by applying frontline big data analytical tools and infrastructure. The model will be tested in a sample from a Nordic collaboration integrating genotyped biobank samples with registry diagnoses and prescription registries, applying secure trans-Nordic protocols for large-scale data analyses. We will also test the prediction tool in a large sample from antipsychotic treatment trials (n=10,000) and validate in a clinical setting. Predict-AP will generate new knowledge that will pave the way for personalised medicine approaches in psychiatry, by providing a tool for selection of patients for antipsychotic drug treatment. This will enable personalized treatment avoiding the current trial and error treatment. Predict-AP has a program for dissemination of findings, and the project is planned together with user groups who will be integrated as advisers in the research. The findings will lead to new knowledge that can form the basis for implementation of a personalised medicine approach in psychiatry.

Publications from Cristin

No publications found

No publications found

No publications found

No publications found

Funding scheme:

BEHANDLING-God og treffsikker diagnostikk, behandling og rehabilitering