Clozapine Levels: The Normal Range, High Results, and What Smoking Does
By Ryan S. Sultan, MD
Psychiatrist · Columbia University Irving Medical Center
Published · 6-minute read
A 350 ng/mL clozapine response threshold was reported in a 1991 study of 29 inpatients with treatment-resistant schizophrenia. In that study, 64% of patients with levels above 350 ng/mL responded, compared with 22% of patients below it.1
A 2023 analysis of 294 individual participants from nine studies put the optimal cut-off at 372 ng/mL and recommended a range of 250 to 550 ng/mL. For separating responders from non-responders, the area under the curve was 0.612, with a sensitivity of 57.3% and a specificity of 65.7%.2 A clozapine level is one piece of information the prescriber reads alongside symptoms, side effects and dose.
What range the studies support
| Source | What it found |
|---|---|
| Perry 1991: 29 inpatients, 4 weeks | Threshold for response of 350 ng/mL |
| VanderZwaag 1996: 56 inpatients randomized to three level ranges for 12 weeks | 200 to 300 and 350 to 450 ng/mL both outperformed 50 to 150 ng/mL, with no advantage for 350 to 450 over 200 to 300 |
| Northwood 2023: 294 participants, nine studies | Optimal point 372 ng/mL; recommended range 250 to 550 ng/mL |
| Tralongo 2023: meta-analysis of 15 studies | Threshold of 407 ng/mL; responders averaged 117 ng/mL higher than non-responders |
| de Leon 2022: international titration guideline | Reference range of 350 to 600 ng/mL for trough steady-state levels |
The published thresholds differ, and they describe groups of patients. The 2023 analysis noted that some patients may not respond without levels above 550 ng/mL, and that this benefit should be weighed against a higher risk of adverse reactions.2 A 2025 study suggested the target range may differ by smoking status: 0.35 to 0.45 mg/L in smokers and 0.50 to 0.60 mg/L in nonsmokers, which is 350 to 450 and 500 to 600 ng/mL.6 These are ranges prescribers work from.
When the blood is drawn changes the number
All nine studies in the 2023 analysis reported trough samples, drawn before the next dose, and a 2022 review found that almost all studies relating levels to side effects use them.2, 7 The randomized trial of three level ranges sampled 11 to 13 hours after a dose.3 On 100 mg twice daily, the label reports an average steady-state peak of 319 ng/mL about 2.5 hours after dosing and an average minimum of 122 ng/mL.8
The prescriber or the lab sets the draw time. A result is read against the time of the last dose, so that time is worth reporting.
What a high clozapine level means
The studies don't agree on a toxic level. A 2012 review found that recommended upper limits ranged from 600 to 2,000 µg/L, a unit numerically equal to ng/mL, and concluded that the exact range corresponding to toxicity remains unclear. It reported minimal correlation between levels and side effects apart from sedation, low blood pressure and seizures, with seizure risk increasing above 600 µg/L or with rapid upward titration.9
In a 2026 observational study of 32 patients, levels above 600 ng/mL weren't associated with greater efficacy and were linked to more side effects.10 A 2026 meta-analysis of 34 studies cautioned that its figures shouldn't be read as universal therapeutic or toxic thresholds, or as targets for dose escalation.11
A high result goes to the prescriber, who reviews the dose, the draw time and anything that changed recently before deciding on any dose change. The clozapine side effects page covers sedation and seizures.
Levels run lower in smokers and can rise after quitting
Tobacco smoke is a moderate inducer of CYP1A2, the enzyme that mainly metabolizes clozapine.8, 5 In a meta-analysis of 23 studies, pooled concentration-to-dose ratios from 16 of them were significantly lower in smokers than in non-smokers.12 In a model of 78,431 samples from 23,516 patients, the predicted dose to attain a predose level of 0.35 mg/L (350 ng/mL) was 36% higher in smokers.13
After a hospital smoking ban, the share of patients who had smoked with a level of 1,000 µg/L or higher went from 4.2% to 41.7% within six months, despite dose reductions.14
The sources differ on the size of the dose change. The 23-study meta-analysis recommends a 30% reduction when a patient stops smoking, informed by trough levels and a clinical risk-benefit evaluation.12 An earlier meta-analysis of four studies and 196 patients suggested that non-smokers need a dose 50% lower than smokers to reach an equivalent concentration.15 These are figures prescribers work from, and the prescriber makes the change. Tell the prescriber before quitting, cutting down or resuming smoking, and at any hospital admission where smoking will stop.
What else is associated with a change in level
- Fluvoxamine. The label reports trough concentrations about three-fold higher after 14 days of fluvoxamine, and names ciprofloxacin and enoxacin as other strong CYP1A2 inhibitors.8 In 67 patients, the median increase in dose-adjusted clozapine concentration with fluvoxamine was 2.51-fold.16
- Infection and inflammation. A systematic review identified 40 cases of elevated levels associated with infection, commonly respiratory, with sedation the typical adverse event.17 In 1,106 samples, a rise in C-reactive protein from 5 to 60 mg/L corresponded to an expected increase in clozapine concentration of more than twofold.18
- Caffeine and oral contraceptives. The label lists both as moderate or weak CYP1A2 inhibitors.8 In samples from 8,833 men and 3,433 women with recorded coffee intake, coffee and chocolate had no discernible effect on dose or level once smoking was considered.19
A new medication, an antibiotic or an illness with fever is worth reporting to the prescriber that day.
Limitations of the evidence
This evidence has limitations. First, the 350 ng/mL threshold was reported in 29 inpatients, and a randomized comparison of three level ranges found no advantage for 350 to 450 over 200 to 300 ng/mL.1, 3 Second, the smoking, infection and caffeine findings are observational, and neither smoking-cessation figure comes from a trial of dose-adjustment strategies.12, 15 Nevertheless, a trough level can help the prescriber judge whether a clozapine trial was adequate and catch a rise after a change in smoking, medication or health.6, 9, 17
Frequently asked questions
What is a normal clozapine level?
The sources give different ranges. A 2023 analysis of 294 participants recommended 250 to 550 ng/mL, and an international titration guideline gives a reference range of 350 to 600 ng/mL for trough steady-state levels. The prescriber interprets an individual result and decides on any dose change.2, 5
Is a clozapine level below 350 too low to work?
The 350 ng/mL threshold was reported in a 1991 study of 29 inpatients, in which 22% of patients below that level still responded. A randomized trial of 56 inpatients found no advantage for 350 to 450 ng/mL over 200 to 300 ng/mL. Whether a dose change makes sense is the prescriber's decision.1, 3
What time should blood be drawn for a clozapine level?
All nine studies in a 2023 analysis reported trough samples, drawn before the next dose. One randomized trial sampled 11 to 13 hours after a dose. The prescriber or the lab sets the draw time.2, 3
Do I need a lower clozapine dose if I quit smoking?
Tell the prescriber before quitting, because levels can rise after smoking stops. One meta-analysis recommends a 30% dose reduction informed by trough levels, and another suggested non-smokers need a dose 50% lower than smokers. These are figures prescribers work from, and the prescriber makes the change.12, 15
Does coffee raise clozapine levels?
The label lists caffeine as a moderate or weak CYP1A2 inhibitor. In samples from 8,833 men and 3,433 women with recorded coffee intake, coffee and chocolate had no discernible effect on dose or level once smoking was considered.8, 19
In a crisis: call or text 988, or go to the nearest emergency department. A first episode of psychosis belongs in a hospital-based or coordinated specialty care program.
I start and manage clozapine in an outpatient practice in Chelsea, Manhattan. Details are on the schizophrenia care in NYC page. Clozapine should not be stopped, skipped, or restarted without the prescriber. This page is general information and does not replace an evaluation.
Sources
- Perry PJ, Miller DD, Arndt SV, Cadoret RJ. Clozapine and norclozapine plasma concentrations and clinical response of treatment-refractory schizophrenic patients. Am J Psychiatry. 1991;148(2):231-235.
- Northwood K, Pearson E, Arnautovska U, et al. Optimising plasma clozapine levels to improve treatment response: an individual patient data meta-analysis and receiver operating characteristic curve analysis. Br J Psychiatry. 2023;222(6):241-245.
- VanderZwaag C, McGee M, McEvoy JP, et al. Response of patients with treatment-refractory schizophrenia to clozapine within three serum level ranges. Am J Psychiatry. 1996;153(12):1579-1584.
- Tralongo F, Konecki C, Feliu C, Kaladjian A, Djerada Z. Association Between Clozapine Plasma Concentrations and Treatment Response: A Systematic Review, Meta-analysis and Individual Participant Data Meta-analysis. Clin Pharmacokinet. 2023;62(6):807-818.
- de Leon J, Schoretsanitis G, Smith RL, et al. An International Adult Guideline for Making Clozapine Titration Safer by Using Six Ancestry-Based Personalized Dosing Titrations, CRP, and Clozapine Levels. Pharmacopsychiatry. 2022;55(2):73-86.
- Flanagan RJ, Obee SJ, Kim AHM, et al. Plasma Clozapine in Treatment Refractory Schizophrenia: What Is the Target Range? J Clin Psychopharmacol. 2025;45(4):336-344.
- Skokou M, Karavia EA, Drakou Z, et al. Adverse Drug Reactions in Relation to Clozapine Plasma Levels: A Systematic Review. Pharmaceuticals (Basel). 2022;15(7):817.
- Clozaril (clozapine) prescribing information, revised June 2025. DailyMed.
- Stark A, Scott J. A review of the use of clozapine levels to guide treatment and determine cause of death. Aust N Z J Psychiatry. 2012;46(9):816-825.
- Hart XM, Hirjak D, Eap CB, et al. High Clozapine Serum Levels Do Not Improve Response: An Observational Study in Patients with Schizophrenia Treated with Clozapine in a Naturalistic Setting. Pharmacopsychiatry. 2026 May 20. Online ahead of print.
- Fornaro M, Di Lorenzo C, Siskind D, de Bartolomeis A. Efficacy and Safety Outcomes of Different Plasma Clozapine and Norclozapine Levels in Schizophrenia Spectrum Disorder: A Dose-Response Meta-Analysis Exploring "critical" Levels. Psychiatry Res. 2026;365:117340.
- Wagner E, McMahon L, Falkai P, Hasan A, Siskind D. Impact of smoking behavior on clozapine blood levels - a systematic review and meta-analysis. Acta Psychiatr Scand. 2020;142(6):456-466.
- Flanagan RJ, Hunter S, Obee SJ, Reeves S. Clozapine: Dose, Sex, Ethnicity, Smoking Habit, Age, Body Weight, and Plasma Clozapine and N-Desmethylclozapine (Norclozapine) Concentrations in Clinical Practice. J Clin Psychopharmacol. 2023;43(2):131-138.
- Cormac I, Brown A, Creasey S, Ferriter M, Huckstep B. A retrospective evaluation of the impact of total smoking cessation on psychiatric inpatients taking clozapine. Acta Psychiatr Scand. 2010;121(5):393-397.
- Tsuda Y, Saruwatari J, Yasui-Furukori N. Meta-analysis: the effects of smoking on the disposition of two commonly used antipsychotic agents, olanzapine and clozapine. BMJ Open. 2014;4(3):e004216.
- van Weringh G, de Haas HJ, de Haan L, Touw DJ, de Koning MB. The Impact of Fluvoxamine on Clozapine and Norclozapine Serum Concentrations. Eur J Drug Metab Pharmacokinet. 2026;51(1):61-73.
- Clark SR, Warren NS, Kim G, et al. Elevated clozapine levels associated with infection: A systematic review. Schizophr Res. 2018;192:50-56.
- Espnes KA, Spigset O, Bratt GE, et al. Association Between Levels of the Acute Phase Proteins Alpha-1-Acid Glycoprotein and C-Reactive Protein and Serum Concentrations of Clozapine: A Study of 1106 Therapeutic Drug Monitoring Samples. Basic Clin Pharmacol Toxicol. 2026;138(3):e70208.
- Flanagan RJ, Obee SJ, Kim AHM, Every-Palmer S. Effect of Coffee and Chocolate Ingestion on Clozapine Dose and on Plasma Clozapine and Norclozapine Concentrations in Clinical Practice. J Clin Psychopharmacol. 2024;44(2):161-167.