Neurocognition and NMDAR co-agonists pathways in individuals with treatment resistant first-episode psychosis: a 3-year follow-up longitudinal study

Sara Camporesi,Lijing Xin,Philippe Golay,Chin Bin Eap,Martine Cleusix,Michel Cuenod,Margot Fournier,Kenji Hashimoto,Raoul Jenni,Julie Ramain,Romeo Restellini,Alessandra Solida,Philippe Conus,Kim Q. Do,Ines Khadimallah
DOI: https://doi.org/10.1038/s41380-024-02631-4
IF: 11
2024-06-08
Molecular Psychiatry
Abstract:This study aims to determine whether 1) individuals with treatment-resistant schizophrenia display early cognitive impairment compared to treatment-responders and healthy controls and 2) N-methyl-D-aspartate-receptor hypofunction is an underlying mechanism of cognitive deficits in treatment-resistance. In this case‒control 3-year-follow-up longitudinal study, n = 697 patients with first-episode psychosis, aged 18 to 35, were screened for Treatment Response and Resistance in Psychosis criteria through an algorithm that assigns patients to responder, limited-response or treatment-resistant category (respectively resistant to 0, 1 or 2 antipsychotics). Assessments at baseline: MATRICS Consensus Cognitive Battery; N-methyl-D-aspartate-receptor co-agonists biomarkers in brain by MRS (prefrontal glutamate levels) and plasma (D-serine and glutamate pathways key markers). Patients were compared to age- and sex-matched healthy controls (n = 114). Results: patient mean age 23, 27% female. Treatment-resistant (n = 51) showed lower scores than responders (n = 183) in processing speed, attention/vigilance, working memory, verbal learning and visual learning. Limited responders (n = 59) displayed an intermediary phenotype. Treatment-resistant and limited responders were merged in one group for the subsequent D-serine and glutamate pathway analyses. This group showed D-serine pathway dysregulation, with lower levels of the enzymes serine racemase and serine-hydroxymethyltransferase 1, and higher levels of the glutamate-cysteine transporter 3 than in responders. Better cognition was associated with higher D-serine and lower glutamate-cysteine transporter 3 levels only in responders; this association was disrupted in the treatment resistant group. Treatment resistant patients and limited responders displayed early cognitive and persistent functioning impairment. The dysregulation of NMDAR co-agonist pathways provides underlying molecular mechanisms for cognitive deficits in treatment-resistant first-episode psychosis. If replicated, our findings would open ways to mechanistic biomarkers guiding response-based patient stratification and targeting cognitive improvement in clinical trials.
biochemistry & molecular biology,neurosciences,psychiatry
What problem does this paper attempt to address?
This paper attempts to solve two main problems: 1. **Whether patients with treatment - resistant schizophrenia (TRS) show cognitive impairment at an early stage**: The study aims to determine whether patients with treatment - resistant schizophrenia have cognitive function impairment in the early stage of the disease compared with those with good treatment response and healthy control groups. Specifically, the study focuses on whether patients with treatment - resistant schizophrenia show lower cognitive scores in aspects such as processing speed, attention/vigilance, working memory, verbal learning, and visual learning. 2. **Whether hypofunction of the N - methyl - D - aspartate receptor (NMDAR) co - agonist pathway is a potential mechanism of treatment resistance**: The study explores whether the dysfunction of the NMDAR co - agonist pathway (such as D - serine and glutamate pathways) is the underlying mechanism leading to cognitive deficits in patients with treatment - resistant schizophrenia. In particular, the study evaluates key biomarkers in these pathways, including changes in D - serine, glutamate, and their related enzymes in plasma and the brain. ### Overview of research methods To answer these questions, researchers conducted a three - year longitudinal follow - up study, recruiting 697 patients with first - episode psychosis and classifying them into three categories according to treatment response: - **Treatment - resistant (TRS)**: No response to at least two antipsychotic drugs. - **Limited - responsive (LRS)**: No response to one antipsychotic drug. - **Good treatment - responsive (RESP)**: Good response to at least one antipsychotic drug. In addition, 114 age - and gender - matched healthy control groups were also recruited. The cognitive function of patients was evaluated by the MATRICS Consensus Cognitive Battery, and key biomarkers of the NMDAR co - agonist pathway were evaluated by magnetic resonance spectroscopy (MRS) and plasma biochemical tests. ### Main findings 1. **Differences in cognitive function**: Treatment - resistant patients performed significantly lower in multiple cognitive domains (such as processing speed, attention/vigilance, working memory, verbal learning, and visual learning) than those with good treatment response, while limited - responsive patients showed an intermediate phenotype between the two. 2. **Dysregulation of the NMDAR co - agonist pathway**: In treatment - resistant and limited - responsive patients, the D - serine pathway was dysregulated, manifested as decreased levels of serine racemase (SRR) and serine hydroxymethyltransferase 1 (SHMT1), and increased levels of glutamate - cysteine transporter 3 (EAAT3). These changes were related to the deterioration of cognitive function, and in particular, this association was disrupted in treatment - resistant patients. ### Conclusion This study reveals the existence of early - stage cognitive dysfunction in treatment - resistant patients with first - episode psychosis and provides evidence that dysregulation of the NMDAR co - agonist pathway is its potential mechanism. If these findings can be further verified, it will be helpful to develop mechanism - based biomarkers to guide patient stratification in clinical trials and cognitive improvement strategies. ### Formula summary - **NMDAR activation condition**: \[ \text{NMDAR}=\text{Glutamate}+\text{Co - agonist (D - serine or Glycine)} \] - **D - serine synthesis path**: \[ \text{L - serine}\xrightarrow{\text{SHMT1}}\text{Glycine} \] - **D - serine degradation path**: \[ \text{L - serine}\xrightarrow{\text{SRR}}\text{D - serine}\xrightarrow{\text{DAAO}}\text{Alanine} \] These formulas show the relationships of key metabolic steps in the NMDAR co - agonist pathway.