Blood Plasma Biomarkers May Predict Psychosis and Guide Personalised Treatment

Two new studies from Singapore and Ireland are harnessing the power of large-scale proteomic analysis to identify blood plasma protein biomarkers that could help predict the onset of psychosis in at-risk individuals and guide treatment decisions.

Study 1: Predicting Psychosis Onset in an Asian Cohort

A collaborative study led by the Institute of Mental Health (IMH), NHG Health, and Nanyang Technological University's Lee Kong Chian School of Medicine (NTU LKCMedicine) has identified protein biomarkers that may predict the development of psychosis in people at ultra-high risk.

Background

Psychosis affects approximately 1 in 43 adults in Singapore over their lifetime, including schizophrenia.

Early intervention currently relies on clinical assessments alone. The study leveraged data from IMH's Longitudinal Youth at Risk Study (LYRIKS), which followed 173 young people aged 14–29. Of these, 65 were identified as being at ultra-high risk for psychosis, and 13 developed psychosis within two years.

Key Findings

Researchers examined 1,757 proteins from blood plasma samples using mass spectrometry. Machine learning models were then developed to test prediction accuracy:

• Two models based on protein patterns identified in Caucasian populations achieved 75–81% accuracy on the LYRIKS dataset.
• Three models developed specifically from the LYRIKS dataset achieved up to 96% accuracy.

Although the specific proteins differed between populations, similar biological processes—such as immune function—were involved in both.

What the Researchers Say

Associate Professor Jimmy Lee (NHG Health, IMH): "The objective is to identify the 20% of ultra-high-risk youths who develop psychosis within two years to enable closer follow-up or earlier intervention."

Dr. Chan Wei Xin (NTU LKCMedicine, first author): "The findings show protein signatures from Caucasian populations can be applied broadly, but Asian-specific signatures performed better."

Assistant Professor Wilson Goh (NTU LKCMedicine, co-lead): "This study moves toward molecular-based profiling to support personalized care."

Next Steps

The researchers plan to validate their findings in larger, independent studies. They noted limitations in detecting low-abundance proteins. Future research may combine proteomic data with genomics, metabolites, and social factors using explainable AI.

Study 2: Proteomics Study for Personalised First-Episode Psychosis Treatment

FutureNeuro, the Research Ireland Centre for Translational Brain Science hosted by RCSI University of Medicine and Health Sciences, has launched a proteomics study aimed at identifying biological markers for personalised treatment of first-episode psychosis.

Methodology

The study will analyze nearly 11,000 proteins in plasma samples from over 500 participants, including individuals with first-episode psychosis and healthy controls. The research team includes Professor David Cotter, Dr. Melanie Föcking, Dr. Subash Susai, and Dr. David Mongan.

Background

Previous work by Professor Cotter's team identified complement cascade proteins associated with symptom recovery and functional improvement in first-episode psychosis. Large-scale proteomics studies, including analyses of UK Biobank data, have demonstrated how blood protein signatures can provide insights into disease risk, prognosis, and treatment response.

What the Researcher Says

Professor David Cotter: "By using large-scale proteomic approaches in people experiencing first-episode psychosis, we aim to identify biological markers that reflect what is happening at a molecular level in the earliest stages of illness. These insights could help move clinical care beyond a trial-and-error approach and support more personalised, evidence-based treatment decisions."

Outlook

Successful findings could support evidence-based treatment decisions from the earliest stages of illness, reducing reliance on trial and error. The study reflects FutureNeuro's commitment to translational psychiatric research and RCSI's focus on patient benefit.