A mathematical model allows the prediction of schizophrenia at the time of the first psychotic outbreak
Cristina G. Pedraz/DICYT Researchers of the Institute of Biomedicine (Ibiomed) of the University of León (ULE), the Physiology Department of the Universidad del País Vasco (UPV/EHU), the Centre for Biomedical Research on the Mental Health Network (Centro de Investigación Biomédica en Red de Salud Mental, Cibersam), and the Hospital Universitario de Álava have developed a mathematical model for predicting schizophrenia at the time of the first psychotic outbreak by means of determining enzymatic activity during the admission of the patient.
The study is the core of the doctoral thesis of Ainhoa Fernández-Atucha and the collaborators were Enrique Echevarría, Gorka Larrinaga, Javier Gil, Mónica Martínez-Cengotitabengoa, Ana M. González-Pinto, Jon Irazusta, and Jesús Seco. The last named is an Ibiomed researcher and the coordinator of the Interuniversity Master's Degree in Healthy Ageing and Quality of Life of the ULE and the UPV/EHU; he explained to DiCYT that the prognostic markers of mental illnesses are vital for improving the clinical handling of patients.
In the case of schizophrenia, the first manifestations tend to occur in the form of psychotic outbreaks, “a disconnection from reality, a delirious state with hallucinations that incapacitate the subject for living in his/her habitual environment”. The psychotic outbreak shares some symptoms with schizophrenia although not all patients with psychotic outbreaks end up developing the disease.
“The models are designed to predict the clinical situation of patients during the forthcoming months. On occasions the initial diagnosis is difficult and this type of information can serve to orientate the application of a therapeutic protocol”, explains the researcher, who adds that the mathematical treatment of the data from the clinical scales of the assessment of the patients' symptomatology, together with the measuring of their plasmatic or tissular (from tissues) biochemical parameters, is currently easier owing to the power of computer systems and the development of advanced statistical software. “In many cases this allows the drawing up of predictive models”, he emphasises.
Enzymatic activity
In order to develop the schizophrenia prediction model the research team has measured various plasmatic proteolytic activities known as aminopeptidases. The Physiology Department of the UPV/EHU has been working on these enzymes in animal models for several years, studying their physiological involvement in various neurochemical aspects of the brain. This study has concentrated on four aminopeptidases: dipeptidyl-peptidase (DPP-IV), prolyl-oligopeptidase (PEP), aminopeptidase N (APN), and aminopeptidase B (APB).
“We sought a mathematical relationship between the quantitative variables deriving from the clinical scales of the assessment of the patients' symptomatology over a year's evolution and the activity levels of these plasmatic aminopeptidases at the start of the psychotic outbreak. We initially detected a correlation and subsequently we drew up several predictive models based on the statistical technique of linear regression”, Jesús Seco details.
According to the results obtained, by using a simple blood analysis and a easy cheap laboratory technique it is possible to make a prediction using aminopeptidase as a biomarker.
To be precise, “the increased activity of plasmatic APB at the time of the initial diagnosis can act as a long-term clinical biomarker indicating a good prognosis in patients with a first psychotic outbreak; while the high DPPIV activity and the plasmatic activities of APN and PSA at the start of a psychotic outbreak could indicate a poor short-term prognosis in patients with psychosis”.
The connection between basic and clinical research
The model developed represents a relevant step in the attempt to discover reliable prognostic indicators. If these indicators are to be established collaboration between psychiatry hospital services and basic research laboratories is necessary.
The next steps of this team of researchers will involve searching for mathematical relationships between the evolution of patients and their initial biochemical parameters, aiming to shed more light on the subject in order to design more specific therapeutic protocols.