For people living with schizophrenia whose symptoms refuse to respond to standard medication, the situation can feel hopeless. Clozapine has long been the go-to treatment when other antipsychotics fail, but it carries a significant side effect burden and is unsuitable for many patients. A major new systematic review and meta-analysis, published in Molecular Psychiatry, has examined what the evidence says about the alternatives.
The research, led by a team at King’s College London and the University of Oxford, analysed 78 studies involving more than 3,200 patients. All of the participants had been diagnosed with treatment-resistant schizophrenia, meaning their symptoms had not responded adequately to at least one prior course of antipsychotic medication. The goal was to assess whether non-clozapine interventions could meaningfully reduce positive symptoms such as hallucinations and delusions, negative symptoms such as emotional withdrawal and reduced motivation, or overall symptom burden.
One of the clearest findings was that simply increasing the dose of an antipsychotic does not appear to help. High-dose antipsychotic treatment showed no significant benefit across any symptom domain, a result consistent with broader Cochrane evidence. Despite this, around a third of people with treatment-resistant schizophrenia are given above-licensed doses before clozapine is considered, suggesting clinical practice may not yet be aligned with what the research shows.
Glycine modulatory site agonists, a class of compounds that act on the NMDA receptor in the brain, produced the most striking results in the analysis. Full agonists such as glycine and d-serine showed large effect sizes for both positive and negative symptoms. The partial agonist d-cycloserine, by contrast, produced no significant benefit, which the researchers linked to its tendency to compete with naturally occurring compounds in the brain at higher doses. The certainty of the evidence remains low, partly because most of these studies came from a single research group, but the findings are considered sufficiently promising to justify larger trials.
Non-invasive brain stimulation techniques, including repetitive transcranial magnetic stimulation and transcranial direct current stimulation, showed a moderate benefit for positive symptoms, particularly auditory hallucinations. However, publication bias was evident in this area of research, and when statistical adjustments were applied to account for missing studies, the effect became non-significant. Psychotherapy, particularly cognitive behavioural therapy, also showed a moderate benefit for positive symptoms in people with treatment-resistant schizophrenia, with consistent effect sizes and no evidence of publication bias.
Antidepressants showed meaningful improvements in negative symptoms and overall symptom scores across three well-designed studies, an encouraging finding given how difficult negative symptoms are to treat. Famotidine and Ginkgo biloba extract each showed promise in limited studies, but with only two trials each, firm conclusions cannot be drawn.
The researchers are clear that none of the interventions reviewed met a sufficient standard of evidence to be recommended as a routine alternative to clozapine. What the review does offer is a roadmap for where future research should focus, particularly larger, well-designed trials of glycine modulators and non-invasive stimulation in patients who genuinely cannot tolerate clozapine.