02. Treatment Resistant Schizophrenia: What Works When Clozapine Is Not an Option?

Text version

For this Quick Take today, I selected a systematic review and meta-analysis examining non-clozapine interventions for treatment-resistant schizophrenia. I was curious to see if the findings in this publication would match up with my own clinical practice which is centered around managing people with psychotic disorders including those who are treatment resistant.

When Clozapine Is Not an Option

Clinically, this review addresses an important issue — what to offer to patients with schizophrenia who are treatment resistant but who are not candidates for clozapine, clozapine of course being the first-line treatment for treatment-resistant schizophrenia.

This scenario happens not infrequently either because patients do not want to commit to a treatment that requires regular blood work and comes with a high side effect burden, which is the price to pay for being on clozapine to be honest, or because they have medical contraindications to taking clozapine.

Study Selection and Design

Let’s skip over the finer technical details of the review which was conducted according to the standards you would expect today from such systematic reviews and meta-analyses. However, let me at least point out the criteria for study selection.

The authors included studies that were randomized controlled trials with a suitable comparator, meaning either placebo for the pharmacological trials or a controlled study intervention for the psychotherapy trials.

Patients were adults with schizophrenia or schizoaffective disorder who had shown treatment resistance to at least one antipsychotic, and the trial had to report baseline and endpoint scores on a psychiatric rating scale for positive, negative, and total symptoms.

They specifically excluded clozapine-augmentation trials. This is an important point. It’s also important to point out that all the interventions in the end were added on to ongoing antipsychotic treatment. So those were not freestanding interventions, which pretty much mimics clinical practice: you would just add other treatments on to your ongoing treatment with an antipsychotic.

Using these criteria, the authors identified 78 studies of which they were able to include 68 studies with 3,241 patients for their meta-analysis.

Low-Quality Evidence Is a Red Flag

Before giving you a bullet point list of their results, let me point out that the median total sample size was around 30, which is very small. Further, the overall quality of the trials was low, and the authors rated their own findings as low or very low using the so-called GRADE (Grading of Recommendations, Assessment, Development, and Evaluation) rating to assess the certainty in their reported outcomes.

So the big red flag here about this review is that their findings are in fact not on very solid ground. Take them with a grain of salt.

Results Across Intervention Types

Still, here are the results based on their grouping of interventions:

• High-dose antipsychotics showed no benefit
• Augmentation with agonists at the glycine modulatory site of the NMDA receptors showed benefit across the board of symptomatology
• Non-invasive stimulation like transcranial direct current stimulation (tDCS) or rTMS had some benefit for positive symptoms
• Psychotherapy showed some benefit for positive symptoms
• Augmentation with antidepressants helped with negative and total symptoms

High-Dose Antipsychotics Not Effective

Let me now put each of their findings into some clinical context. I think it is generally correct that pushing the dose of an antipsychotic is not an effective strategy once you reach sufficient dopamine receptor occupancy of about 80%.

However, there have been meta-analyses that suggest a high-dose approach is worth a try for olanzapine specifically, something that I do suggest trying before giving up on olanzapine. I am also a strong advocate for therapeutic drug monitoring (TDM) to identify those people who are ultra-rapid metabolizers where you would need to use higher than usual doses of a medication.

Glycine Modulators Lack Clinical Traction

Their findings for benefit using modulators at the glycine site of the NMDA receptors — basically glycine itself, D-serine, or D-cycloserine — is based on the work from mostly one research group and it never honestly spread widely clinically.

I myself used to recommend glycine, which you can buy as a supplement, but I admit I have not been impressed over the years so I’ve moved away from using it.

Antidepressants Help Negative Symptoms

What I do recommend routinely for negative symptoms are antidepressants, but also for subsyndromal depression which is a very common problem for many patients with schizophrenia spectrum disorders.

Non-Invasive Stimulation Has Limited Access

The benefits from the various non-invasive stimulations are difficult to interpret given that there are so many protocols out there — the type of stimulation, where to stimulate, and so on.

I suspect in most settings you will not have access to these treatments outside of research, and insurance will also be an obstacle as those treatments are mostly used for depression and not psychosis.

The one treatment to remember is ECT, which is broadly effective for psychosis and also for clozapine augmentation. So I always keep it in mind as an evidence-based option.

CBT for Psychosis Faces Practical Barriers

Finally, psychotherapy: here, the authors basically mean cognitive behavior therapy or CBT for psychosis, which can help candidates who are able to participate in this treatment — which is a big obstacle in my experience.

The other big obstacle is to find competent therapists who will offer CBT for psychosis. I suspect in most settings you will have a very hard time finding psychologists with even some baseline training in this modality.

A Broader Philosophy for Difficult-to-Treat Illness

Let me close with a more general comment about managing patients with treatment-resistant schizophrenia. This review that we just went over together really should temper your enthusiasm to just add pharmacological treatments where the benefit is small at best.

However, difficult-to-treat illness — a term I prefer over treatment-resistant illness — does not mean that accompanying patients in their suffering as a physician is useless. I think supportive psychotherapy combined with good case management plus the competent and judicious use of medications is one strategy that over time will lead to small but meaningful benefits for many patients and their families.

Bottom Line: Time-Limited Trials and Shared Decision Making

I think in this spirit, add-on pharmacological approaches, if offered as time-limited trials and with some realism about what to expect, are important to help patients and their families remain hopeful that small improvements are possible.

Patients may come to you with their own suggestions about what they want to try based on their readings of the literature. Don’t shut them down immediately. Engage in shared decision making. You may actually learn something yourself.

And then, always emphasize the importance of time-limited trials. You stop the intervention if it does not work and you don’t just keep going, particularly if patients must buy the pills out of pocket. It really can add up. But any such trial must be embedded in a much broader psychiatric treatment philosophy beyond medications that would include rehabilitation or peer support, for example.

Abstract

Non-Clozapine interventions in treatment-resistant schizophrenia: a systematic review and meta-analysis

Richard Carr, Alistair Cannon, Valeria Finelli, Bernard Bukala, Yesim Cimen, Patritsiya Filipova, Connor Cummings, Toby Pillinger, Oliver D. Howes & Robert A. McCutcheon

Background and hypothesis

Clozapine is the only licensed pharmacotherapy for treatment resistant schizophrenia (TRS), but in some cases is not a suitable treatment option. A review of the efficacy of non-clozapine interventions in TRS may help inform clinical decision making when clozapine treatment is not feasible.

Study design

A systematic review and meta-analysis was performed investigating the efficacy of non-clozapine augmentation of antipsychotic treatment in TRS on positive, negative, and total symptoms. The review protocol is registered at PROSPERO (ID: CRD42023418053). PsycInfo, PubMed and EMBASE were searched up until July 2023. Cochrane Risk of Bias tool (v2) was used to assess study quality. Data were pooled using a random-effects model for each class of intervention to give an estimate of effect size (Hedges’ g).

Results

78 studies were included, of which 68 were included in the meta-analysis, comprising 3241 patients. High-dose antipsychotics (7 studies, 467 participants) did not improve any symptom domain. Augmentation of antipsychotics with glycine modulatory site agonists (9 studies, 187 participants) improved positive (g = −0.56 [−0.81, −0.31], GRADE rating Low), negative (g = −1.18 [−1.49, −0.87], GRADE rating Low) and total (g = −1.17 [−1.75, −0.59], GRADE rating Very Low) symptoms. Non-invasive stimulation (26 studies, 893 participants) moderately benefited positive symptoms (g = −0.42 [−0.65, −0.18], GRADE rating Low). Psychotherapy (10 studies, 565 participants) moderately improved positive symptoms (g = −0.56 [−1.01, −0.10], GRADE rating Low). Augmentation with antidepressants (3 studies, 187 participants) improved negative (g = −0.74 [−1.46, −0.02], GRADE rating Very Low) and total (g = −0.69 [−1.00, −0.38], GRADE rating Low) symptoms. Sample sizes were small, and publication bias was apparent for non-invasive stimulation studies.

Conclusions

Several augmentation strategies, including pharmacotherapy, non-invasive stimulation, and psychotherapy demonstrated benefit in small studies, however no intervention reached the threshold of evidence to be routinely recommended as a viable alternative to clozapine. High-quality trials are needed for definitive recommendations.

Reference

Carr, R., Cannon, A., Finelli, V., Bukala, B., Cimen, Y., Filipova, P., Cummings, C., Pillinger, T., Howes, O. & McCutcheon, R. (2026). Non-Clozapine interventions in treatment-resistant schizophrenia: a systematic review and meta-analysis. Molecular Psychiatry 31, 526–544.