Molecular neuroimaging provides remarkable insights into human brain pathophysiology, yet clinical translation remains limited. For decades, researchers have investigated neurotransmitter alterations across psychiatric syndromes, particularly within the central dopaminergic pathways. However, standard diagnostic applications have faced substantial hurdles due to high inter-individual biological heterogeneity. Fortunately, normative modelling in psychosis offers an innovative paradigm to map individual dopaminergic deviations against healthy reference populations. By establishing expected physiological variation across healthy individuals, this computational framework identifies subtle, patient-specific abnormalities. Consequently, clinicians and neuroscientists can move beyond broad group-level averages toward precise biological phenotyping.

Understanding the Limits of Conventional Molecular Neuroimaging

Positron emission tomography measures biochemical pathways directly in living human brains. In psychiatric research, molecular imaging has primarily evaluated dopaminergic transmission in schizophrenia and first-episode psychosis. Historically, case-control studies pooled patient cohorts to calculate mean regional differences between diagnostic groups. While these comparative methodologies demonstrated elevated presynaptic dopamine synthesis capacity, they often obscured critical individual variations. Furthermore, clinical trials and diagnostic applications struggled with high scanner costs, protocol discrepancies, and small sample cohorts across participating hospitals.

In addition, standard case-control analyses assume that psychiatric disorders represent biologically uniform diagnostic entities. In reality, patients presenting with identical symptoms frequently exhibit entirely different pathophysiological mechanisms. Some individuals demonstrate pronounced dopaminergic hyperactivity, whereas others show intact or reduced neurochemical signaling. Therefore, classical averaging methods often dilute critical pathological signals. Moreover, variations in radiotracer affinity and imaging scanner hardware across multisite collaborations introduce technical noise. To resolve these challenges, modern psychiatry requires analytical models that distinguish genuine neurochemical pathology from technical noise and normal biological variance.

How Normative Modelling in Psychosis Maps Biological Diversity

Normative modelling functions similarly to pediatric growth charts used in general medicine. Instead of comparing arbitrary group averages, this technique maps normal physiological variation across large healthy populations. Researchers can then plot an individual patient's molecular scan against this normative distribution. Consequently, clinicians can identify statistical outliers who deviate significantly from expected biological centiles. This mathematical method accommodates confounding factors such as age, sex, and acquisition site differences across imaging centers.

Recently, investigators applied this computational framework to multisite positron emission tomography datasets. Specifically, they evaluated two prominent radiotracers: carbon-11 labeled PHNO, which targets dopamine D2 and D3 receptors, and fluorine-18 labeled FDOPA, which measures presynaptic dopamine synthesis capacity. By harmonizing imaging data across multiple international research centers, the investigators built robust normative reference cohorts of healthy control participants. Subsequently, the researchers mapped individual scans from distinct patient cohorts onto these normative ranges. Thus, normative modelling in psychosis successfully overcomes sample size limitations by enabling safe pooling of multi-scanner data. Ultimately, this approach personalizes neurochemical assessment for each patient.

Quantifying Extreme Dopaminergic Deviations in Schizophrenia

The multi-cohort analysis revealed striking biological differences between healthy controls and patients with psychosis. Specifically, patients with schizophrenia and first-episode psychosis exhibited approximately three times more extreme dopaminergic deviations than healthy comparison subjects. These extreme deviations occurred across multiple striatal and extrastriatal brain regions. However, the spatial distribution of these neurochemical abnormalities proved remarkably heterogeneous across individuals.

In fact, the spatial overlap of extreme deviations between any two patients reached a maximum of only twenty percent. This pivotal finding demonstrates that patients share the same clinical diagnosis without sharing identical anatomical or neurochemical footprints. While one patient displays marked presynaptic hyperdopaminergia in the associative striatum, another exhibits abnormalities primarily in the sensorimotor striatum or limbic regions. Furthermore, some individuals display normal or reduced dopaminergic activity. Consequently, broad diagnostic categories like schizophrenia encompass diverse biological entities. Normative modelling captures this profound clinical heterogeneity, explaining why uniform therapeutic approaches often fail across patient subsets.

Predicting Antipsychotic Treatment Response with Striatal Biomarkers

A major clinical objective in psychiatric management involves identifying non-responders early in their treatment course. In clinical settings, standard clinical assessments cannot reliably predict which patients will improve on standard dopamine D2 receptor blockers. Fortunately, normative modelling applied to striatal fluorine-18 FDOPA uptake demonstrated impressive prognostic utility. When investigators calibrated striatal signals against normative reference distributions, the model successfully predicted antipsychotic response with receiver operating characteristic area under the curve values between 0.77 and 0.83.

Moreover, this predictive accuracy holds profound implications for clinical decision-making. Patients who show significant presynaptic dopaminergic elevations respond robustly to conventional second-generation antipsychotics. Conversely, patients who present without dopaminergic elevations demonstrate poor clinical response to standard dopamine blockade. Historically, these treatment-resistant patients endure years of failed medication trials before receiving alternative interventions. Therefore, by detecting non-dopaminergic psychosis early, normative PET profiling can accelerate initiation of clozapine or novel glutamatergic agents. Thus, normative modelling establishes a concrete foundation for precision psychiatry.

Clinical Implications for Psychiatric Practice and Trial Design

The clinical implementation of normative molecular neuroimaging could transform psychiatric diagnostics and pharmaceutical research. Currently, clinical trials evaluating novel compounds face high failure rates due to cohort heterogeneity. By utilizing normative modelling, trial investigators can stratify participants based on specific neurochemical signatures rather than subjective symptom scores. Consequently, researchers can enrich patient samples with individuals possessing the targeted molecular pathology, dramatically improving trial efficiency and lowering drug discovery costs.

Furthermore, this methodological advancement addresses practical constraints in everyday practice. High imaging expenses and tracer availability have long restricted positron emission tomography to specialized academic medical centers. However, normative modelling maximizes the utility of existing scans by referencing them to standardized databases. As multisite repositories expand, community centers could upload standardized imaging acquisitions to centralized normative algorithms. Therefore, clinicians could receive individualized deviation heatmaps and response probabilities without requiring dedicated local control cohorts. Ultimately, this scalable framework bridges the gap between sophisticated neuroimaging science and routine patient care.

Frequently Asked Questions

What is normative modelling in molecular neuroimaging?

Normative modelling is an analytical technique that maps normal biological variation across healthy reference populations, similar to pediatric growth charts. In molecular neuroimaging, it plots an individual patient's PET scan against this reference distribution. Consequently, clinicians can identify statistically significant personal neurochemical deviations rather than relying on cohort averages.

How does normative modelling improve antipsychotic selection?

Normative modelling calculates striatal dopamine synthesis capacity relative to healthy standards. Patients displaying marked dopaminergic elevations usually respond favorably to dopamine D2 receptor blockers. In contrast, patients lacking dopaminergic abnormalities rarely benefit from standard antipsychotics. Early recognition allows clinicians to prompt alternative strategies, such as clozapine, without unnecessary delays.

Why is dopamine heterogeneity significant in first-episode psychosis?

Dopamine heterogeneity explains why patients with identical symptoms respond differently to treatments. Although psychosis cohorts exhibit threefold increases in extreme dopaminergic deviations, individual anatomical overlap reaches only twenty percent. Recognizing this biological diversity helps psychiatrists appreciate that psychosis represents multiple underlying neurochemical subtypes, requiring individualized clinical management rather than uniform pharmacology.

Disclaimer: This content is for informational and educational purposes only. It is not intended to be a substitute for professional medical advice, diagnosis, or treatment. Always seek the advice of your physician or other qualified health provider with any questions you may have regarding a medical condition. Never disregard professional medical advice or delay in seeking it because of something you have read here. Refer to the latest local and national guidelines for clinical practice.

References

Giacomel A et al. Investigating dopaminergic abnormalities in schizophrenia and first-episode psychosis with normative modelling and multisite molecular neuroimaging. Mol Psychiatry. 2025 Aug. doi: 10.1038/s41380-025-02938-w. PMID: 40021831.

Howes OD et al. The nature of dopamine dysfunction in schizophrenia and what this means for treatment. Arch Gen Psychiatry. 2012;69(8):776-786.

Jauhar S et al. Elevated dopamine synthesis capacity in first-episode psychosis: a PET study. Lancet Psychiatry. 2018;5(7):e16.