Clozapine is the most effective antipsychotic available for treatment-resistant schizophrenia. That much is settled. What is less discussed is what happens to the substantial fraction of patients who improve on clozapine but don't respond fully — they're better than before, but voices remain, paranoia continues, or function stays impaired. This is a clinical situation called "ultra-resistant" or "clozapine-resistant" schizophrenia, and it affects roughly 40–70% of people on clozapine depending on definition.
What works as augmentation? The evidence base is uneven, and a lot of options sit on weaker ground than clinicians or patients sometimes assume.
Optimise clozapine itself. Check the plasma level (target is generally 350 ng/mL or higher), confirm adherence, address smoking and caffeine changes that affect levels, and give it adequate time (often 6 months at therapeutic level). Many "partial responders" turn out to be sub-therapeutic.
Why level matters before augmentation
Studies repeatedly show that clozapine response correlates with plasma level, not just oral dose. A patient on 400 mg/day with a level of 200 ng/mL is essentially undertreated. See plasma-level monitoring for clozapine. Augmentation makes much more sense after the level question is settled.
Augmentation strategies, ranked by evidence
1. ECT augmentation — strongest evidence
The Petrides et al. randomised controlled trial published in The American Journal of Psychiatry (2015) compared bilateral ECT plus clozapine versus clozapine alone in clozapine non-responders. About 50% of the ECT group met response criteria versus essentially zero in the control. Several smaller trials and a Cochrane review support this as the augmentation strategy with the strongest evidence base. See ECT for schizophrenia.
Practical considerations: requires anaesthesia, requires several treatments per week initially, has cognitive side effects in the short term, and access varies hugely by region. Often the strongest first augmentation choice when feasible.
2. Adding a second antipsychotic — modest, mixed evidence
The most-studied combination is clozapine plus aripiprazole. Several controlled trials suggest aripiprazole augmentation may modestly reduce positive symptoms and metabolic side effects (because aripiprazole is metabolically lighter and may counteract clozapine-related weight gain). Effect sizes are smaller than for ECT.
Clozapine plus amisulpride (where available) has supportive evidence in some trials but isn't FDA-approved in the US.
Clozapine plus risperidone has been widely used clinically but most rigorous trials have not shown clear benefit. See polypharmacy in schizophrenia for a broader discussion of combining antipsychotics.
3. Lamotrigine — promising but inconsistent
The mood stabiliser lamotrigine has been added to clozapine in several small trials, with positive results in some meta-analyses and negative results in others. The Cochrane review concluded the evidence is inconclusive. It is one of the more reasonable next-line options when ECT and antipsychotic augmentation are not chosen, but expectations should be modest.
4. Memantine — possible help for negative and cognitive symptoms
Memantine, an NMDA receptor antagonist used for Alzheimer's, has been added to clozapine in small trials with some positive signals on negative and cognitive symptoms. Evidence is preliminary. It is reasonably well tolerated and a fair option to discuss.
5. CBT for psychosis — should always be on the list
Augmentation doesn't have to be a pill. CBTp targeted at residual voices and beliefs has independent evidence in clozapine partial responders. The COMMAND trial in the UK showed reductions in command hallucination compliance with targeted CBTp. Often underused as augmentation.
6. Less-supported options
The following have been tried with weak or inconsistent evidence:
- Mood stabilisers other than lamotrigine (valproate, lithium) — limited support unless mood symptoms are present
- Topiramate — sometimes used to address weight gain rather than psychosis
- Glycine, D-serine, sarcosine — based on the NMDA hypothesis, modest signals in trials
- Omega-3 fatty acids — generally negative trials in chronic schizophrenia
- NAC (N-acetylcysteine) — early-stage research, not established
- Anti-inflammatory drugs (minocycline, NSAIDs) — mixed results
What to be cautious about
Adding multiple antipsychotics
Three-antipsychotic combinations have almost no evidence and significant cumulative side effect burden. If a second antipsychotic plus clozapine isn't working, the answer is rarely a third antipsychotic.
Dosing clozapine above 600 mg without level monitoring
Pushing dose without checking blood levels means more side effects without reliably more efficacy. The level check is the cheaper and more informative move.
Stopping clozapine to "try something else"
Clozapine non-response on a true clozapine trial is a poor prognosis. Discontinuing clozapine usually means worse symptom control, not better. Augmentation, ECT, and CBTp are generally preferable to discontinuation in partial responders.
How to think about sequencing
A reasonable progression in someone partially responding to clozapine:
- Confirm adherence and check plasma level. Adjust dose based on level (not just symptoms).
- Allow adequate time at therapeutic level — usually 3–6 months.
- Add CBTp if it isn't already in place.
- Consider ECT augmentation, especially if response is needed quickly or symptoms are severe.
- If ECT isn't feasible, consider aripiprazole augmentation.
- Lamotrigine, memantine, or other adjuncts as further options with realistic expectations.
What plasma level you are running, how long you've been at therapeutic level, and which augmentation option matches your specific symptom pattern. There is no one-size-fits-all sequence.
What patients should know
Partial response to clozapine is the rule, not the exception. A meaningful fraction of people benefit substantially from one of the augmentation strategies above. The combination of optimised clozapine, targeted CBTp, and (when needed) ECT augmentation is the most evidence-supported pathway. Anything more exotic should be approached with realistic expectations and informed consent.
This article is for educational purposes only and is not medical advice, diagnosis, or treatment. Always consult a qualified mental health professional. If you or someone you know is in crisis, call or text 988 in the US, or your local emergency number.