2026 ECNP Congress

PS04-4055

Biomarker-guided stratification in psychiatry randomized trials: a transdiagnostic systematic review

Fabio Catalano — M. La Maida, M. Festa, G. Cantarella, L. Pelizza, M. Menchetti, A.R. Atti, J. Bralten, S. Matura, A. Serretti, M. Colizzi, C. Fabbri, G. Fanelli · University of Bologna, Biomedical and Neuromotor Sciences, Bologna, Italy

2026 ECNP Congress

Background

Current psychiatric diagnostic systems, largely based on symptom-defined categories, fail to adequately explain the high heterogeneity in treatment-response1,2. Biomarkers have been proposed to identify biologically defined subgroups not captured by Diagnostic and Statistical Manual of Mental Disorders (DSM)-5-based classifications3. However, most randomized controlled trials (RCTs) in psychiatry that examined biomarkers–outcome associations did not incorporate biomarkers in trial design, e.g., as selection criteria. The literature on this topic remains fragmented and lacks systematic mapping from a transdiagnostic perspective.

Aims

This systematic review aimed to (i) identify RCTs in which biomarkers were prospectively used to guide study eligibility, stratification, or treatment allocation in psychiatric disorders; (ii) evaluate their methodological characteristics and clinical outcomes across diagnostic categories; and (iii) synthesise evidence on biomarkers performance and utility across diagnoses and interventions.

Methods

A systematic search was conducted across PubMed, Web of Science, Scopus, and ClinicalTrials.gov (up to May 2026), following PRISMA 2020 guidelines (PROSPERO: CRD420251166201). Eligible studies included adults with DSM/ICD-defined psychiatric disorders and prospective use of biomarkers to inform eligibility, stratification, or randomisation. Data extraction covered trial design, biomarker class and threshold, intervention, outcomes, and safety. Risk of bias was assessed using the Cochrane RoB-2 tool. Given clinical, biomarker, and design heterogeneity across studies, results were synthesized qualitatively.

Results

After screening 4,340 records, 20 RCTs met inclusion criteria. Most included trials focused on major depressive disorder, bipolar depression, or schizophrenia-spectrum disorders. Inflammatory biomarkers were the most frequent stratification strategy (13 trials), predominantly C-reactive protein (CRP). Six trials used metabolic markers, one a hormonal biomarker, and one a genetic biomarker. 18 RCTs adopted a biomarker-enriched design, in which biomarker thresholds were used to select the sample; biomarker-stratified design, in which randomisation was balanced across biomarker-defined strata, and biomarker-guided designs, in which treatment allocation depended on biomarker status, were less frequent. Among inflammation-focused trials, CRP thresholds varied substantially: CRP ≥1 mg/L was often used for enrichment but yielded inconsistent treatment signals, whereas higher thresholds, particularly CRP ≥3 mg/L, appeared more consistently associated with detectable clinical benefit. Interventions included pharmacological agents, non-invasive psychosocial and chronobiological therapies, and dietary modifications. Depressive symptoms and anhedonia showed the most robust evidence, whereas findings for cognitive and broader psychotic outcomes were comparatively less consistent. Biomarker change did not systematically parallel clinical improvement, suggesting that biomarkers may be more useful as enrichment or moderation variables than as direct state markers. Risk of bias was low in eight trials, raised some concerns in 11, and was high in one.

Conclusions

Biomarker-informed RCTs in psychiatry are feasible and may improve treatment testing by reducing biological heterogeneity, particularly in inflammatory and immuno-metabolic contexts. However, the current evidence remains limited by variable biomarker thresholds, heterogeneous designs, inconsistent outcome selection, and few trials directly testing biomarker-by-treatment interactions. The results of this review support the transition toward biologically informed, transdiagnostic trial designs, a research approach that has been still scarcely explored in psychiatry.

References

1 Dalgleish T, Black M, Johnston D, Bevan A. Transdiagnostic approaches to mental health problems: Current status and future directions. J Consult Clin Psychol. 2020 Mar;88(3):179-195. doi: 10.1037/ccp0000482. PMID: 32068421; PMCID: PMC7027356. 2 Howes OD, Thase ME, Pillinger T. Treatment resistance in psychiatry: state of the art and new directions. Mol Psychiatry. 2022 Jan;27(1):58-72. doi: 10.1038/s41380-021-01200-3. Epub 2021 Jul 13. PMID: 34257409; PMCID: PMC8960394. 3 Reiter, J.E., Nickels, S., Nelson, B.W. et al. Increasing psychopharmacology clinical trial success rates with digital measures and biomarkers: Future methods. NPP—Digit Psychiatry Neurosci 2, 7 (2024). https://doi.org/10.1038/s44277-024-00008-7.