Utility of clozapine trough level reporting in clinical practice: A four-year retrospective evaluation from India

ABSTRACT

Background:

Clozapine therapeutic drug monitoring (TDM) supports individualized dosing in treatment-resistant schizophrenia.

Aim

This study evaluated the time trend of TDM reporting in routine practice.

Methods

A retrospective audit of 100 consecutive clozapine TDM reports over 4 years was performed. Service utilization was assessed. Associations between daily dose, trough concentration, concentration–dose (C/D) ratio, therapeutic impressions (low, matching the dose, high, very high clozapine concentrations), and clinical advice accordingly were analyzed statistically.

Results

The mean turnaround time was 20.9 ± 12 days, with a stable quarterly demand (~6 requests/quarter) for clozapine TDM. Dose correlated significantly with trough concentration (ρ = 0.453, P < 0.001). Forty-seven percent of concentrations were subtherapeutic. C/D ratio differed significantly between therapeutic impressions (β = 13.12, P < 0.001). Therapeutic impressions significantly influenced clinical advice (χ² = 54.03, P < 0.001). Nonadherence was identified in 8% of cases.

Conclusions:

Pharmacology TDM service utilization for clozapine was steady and continuing in treatment-resistant schizophrenia patient care.

INTRODUCTION

Clozapine remains gold standard treatment for treatment-resistant schizophrenia (TRS), offering superior efficacy in reducing psychotic symptoms compared with other antipsychotics.[1] However, its clinical use is complicated by substantial pharmacokinetic variability, dose-dependent adverse effects, and narrow therapeutic window. As a result, therapeutic drug monitoring (TDM) has become important for optimizing clozapine therapy.[2]

International guidelines have attempted to standardize interpretation of clozapine plasma concentrations. In 2017, German-led Arbeitsgemeinschaft für Neuropsychopharmakologie und Pharmakopsychiatrie (AGNP) consensus for TDM in neuropsychopharmacology recommended a therapeutic reference range of approximately 350–600 ng/mL for clozapine trough concentrations, suggesting that concentrations ≥350 ng/mL are associated with improved clinical response in TRS.[3] These recommendations are largely derived from Western studies, where average prescribed doses and metabolic profiles differ significantly from those observed in Asians.[4,5,6]

International practice frequently supports higher doses (often 300–900 mg/day), whereas Asian countries tend to prescribe comparatively lower doses to achieve clinical response while limiting adverse effects.[5,7] Emerging pharmacokinetic evidence suggests that Asian patients may achieve comparable plasma concentrations at lower oral doses.[8] This diversity raises an important clinical question: Do optimal clozapine trough concentration targets differ regionally, and should therapeutic reference ranges be contextualized to local populations?[9]

While several international studies have examined drug level–response relationship, relatively few have assessed laboratory reporting practices and clinical decision-making in routine psychiatric care.[10,11,12,13] Therefore, evaluating laboratory tests, reporting, and their utility is essential to determine whether TDM is being used optimally.[14] The study planned to analyze the past 4-year time trend of clozapine TDM test utilization, trough concentrations, and kind of opinions/advice that were shared to the treating psychiatry team.

METHODS

This retrospective study assessed TDM reports of clozapine trough levels from February 2022 to 2026 at a tertiary care research hospital in North India. Informed consent waiver and Institutional Ethics Committee approvals were obtained. Patient care Clozapine TDM lab test was rolled out in year 2022 in our institute.[9,15] The test required patients to be on stable clozapine dose for minimum 5 days before blood sample collection (Supplementary File Figure SF1 ^{(1,022.1KB, tif)} and 2 ^{(502.5KB, tif)} ). The psychiatry team adhered to lab procedures (mentioned on the requisition form) before scheduling any patient for sampling. TRS patients from the psychiatry department provided a 2-ml EDTA blood sample collected 12 hours after the last dose. For example, the patient was scheduled by the psychiatry team for 8.45–9.15AM blood collection time, if he/she was receiving clozapine at 9 PM daily. The requisition form asked for patient identifiers, concurrent medications, indication for TDM, and any extra details of the psychiatry consultant if required. The pharmacology team examined each requisition form. No other patient record files or outpatient card information was sought. In summary, two things from the psychiatry team were received – blood sample and a filled requisition form. Four things were generated by the pharmacology team – plasma concentration, C/D (concentration–dose) ratio, concentration classification (very low, low, matching, and high), and therapeutic impression (clinical advice).

The high-performance liquid chromatography (HPLC) method was used (Supplementary File Figure SF3 ^{(303KB, tif)} ). As per the method, the EDTA whole-blood sample required immediate plasma separation and storage at -80°C. Concentrations were calculated using regular calibration. Dose–concentration relationship was established a priori based on the observed proportionality between daily clozapine dose (mg/day) and the corresponding trough plasma concentration (ng/mL).[9,15] In the foundational experiments of method establishment, a near-linear relationship was confirmed, with approximate equivalence between dose and concentration (i.e. 1 mg/day ≈ 1 ng/mL at steady-state trough levels). Based on this proportionality, expected therapeutic trough concentrations were defined relative to the prescribed daily dose. A tolerance window of ± 12.5% around the expected concentration (e.g., approximately ± 25 mg/day for 200 mg dose) was considered acceptable to account for biological and analytical variability, which may arise from a 30-minute window period of sample collection.[9,15] The framework for evaluating the concentrations is mentioned in supplementary file Table ST1.

Table ST1.

Framework for evaluating TDM reports

TDM results were categorized as follows:

“Very Low”: Concentration<75% of expected

“Low”: Concentration between 75–87.5% of expected

“Matching”: Observed concentration within ±12.5% of expected dose-equivalent concentration

“High”: Concentration >112.5% of expected For example, in patients receiving 200mg/day of clozapine, expected trough concentration was 200ng/mL. Concentrations between 175–225ng/mL were classified as “Matching,” 150–174ng/mL as “Low,” <150ng/mL as “Very Low,” and >225ng/mL as “High.” This classification framework was used to support clinical interpretation, including assessment of adherence and potential dose optimization advice.

Scenario -1

Treating psychiatrist was advised to re-verify compliance if lab measurement showed “very low” plasma clozapine concentration. Telephonic re-verification after the lab report generation confirmed non-compliance/sampling issue.

Scenario -2

Similarly, advice to increase dose or verify smoking history was given if measured clozapine concentration was “low” and requisition form had a mention compliance confirmed but remission not achieved.

Scenario -3

If clozapine concentration was more than 12.5% expected (“High”) with respect to daily dose, advice of reduce the dose was provided.

Scenario-4

Clozapine concentration well within range (“matching”) was used to confirm compliance.

Additionally, therapeutic impressions/clinical advice included comments like- technical/sampling issue, drug–drug interaction.

The study used the Medscape drug interaction checker tool (https://reference.medscape.com/drug-interactionchecker) for assessment and giving advice on drug–drug interactions between concurrent medicines prescribed to the patient along with clozapine.[16]

Statistical analysis

IBM Statistical Package for Social Sciences (SPSS) version 28 was used for analysis. Temporal trends were assessed using linear regression, Durbin–Watson statistics, and Holt’s forecasting. Associations between daily dose, trough concentration (ng/mL), C/D (concentration–dose) ratio, therapeutic impression (low, matching, high, very high), and clinical advice were analyzed using correlation, Kruskal–Wallis testing, and Chi-square analysis.

RESULTS

One hundred clozapine TDM reports were analyzed. The median patient age was 33.3 (IQR range 18–74) years, and 62% were male. The mean TDM turnaround time was 20.9 ± 12.0 days. No significant temporal trend in service utilization was observed over 4 years (β = –0.023, P = 0.93; R² = 0.0006). The Durbin–Watson statistic (2.05) excluded autocorrelation. Forecasting predicted a stable demand (~6 requests/quarter; 95% CI 0–13). Clozapine TDM requisition requests were received once for 56 patients, twice for 16 patients (32 samples), and three times for 4 patients (12 samples). The mean coefficient of variation was 33.6% among patients who underwent two TDM measurements and 54.0% among those who underwent three measurements. However, a statistically significant difference was not observed in the mean clozapine concentrations of patients requesting once, twice, and thrice (Figure 1, ANOVA, P = 0.39).

Figure 1.

Spaghetti plot showing clozapine trough concentration variations in patients opting TDM test once, twice, and thrice

The average clozapine dose observed was 297.13 ± 118.24 mg/day, and the median clozapine trough concentration was 198.60 (104.99–289.57) ng/mL, while the C/D ratio of clozapine was 0.79 (0.40–1.04). Daily dose correlated significantly with trough concentration (Spearman ρ = 0.453, P < 0.001; Kendall τ = 0.332, P < 0.001).

Forty-two percent concentrations were concordant with dose, 41% were low, 6% were very low, and 11% were high. In total, 47% were in subtherapeutic range [Figure 2a].

Figure 2.

Correlation analysis between Clozapine daily dose versus trough concentration (a) Association check between Clozapine level impression versus concentration/dose ratio (b)

C/D ratio differed significantly across different concentration categories (very low, low, matching, high) (Kruskal–Wallis H = 79.05, P < 0.001). Ordinal logistic regression identified C/D ratio as a strong independent predictor of concentration-specific impression classification (β = 13.12, P < 0.001), with higher C/D values associated with higher concentration categories [Figure 2b].

Clinical advice shared in TDM reports included: dose increase (42%), questionable blood sample collection timing (13%), technical/sampling issue (12%), compliance reverification advised (11%), verify smoking history (10%), possible drug–drug interaction (9%), noncompliance confirmation (8%), maintain/no escalation (4%), dose reduction (1%) and hypothyroid influence (1%), ongoing inflammation in body (1%), and sedation-efficacy balance advice (1%). Concentration-specific impression classification significantly associated with clinical advice provided to psychiatry consultants (P < 0.001; Cramer’s V = 0.42).

Drug–drug interactions were consistent with CYP1A2 inhibition (fluvoxamine), induction (rifampicin, modafinil), and serotonergic antidepressant coadministration in 8% of patients.

DISCUSSION

International experience suggests that shorter turnaround times (e.g., reduction from 3 days to 1 day) in the United States increased the demand for clozapine level requisitions; our mean reporting time was 21 days. This longer turnaround time reflects contextual constraints typical of a developing healthcare system, including limited procurement, restricted and shared instrument availability, interruptions due to routine maintenance, and more. We also performed time-series analysis indicating stable quarterly demand of approximately six requisitions, with nonsignificant upward or downward trends. Although the annual requisition rate was lower than that reported in Bye et al. and Jogi et al. studies, clozapine TDM remains an evolving clinical service in the Indian setting.[2,17]

In the current study, the mean daily clozapine dose was 297 mg/day, with a mean trough concentration of 196 ng/mL and a mean C/D ratio of 0.77. These values are comparable considering our 50 patients’ pharmacometabolomics data showing a mean concentration of 229 ng/mL trough concentration with an average dose of 284 mg, reinforcing the observation that Indian TRS patients may achieve clinical response at plasma concentrations lower than those suggested by the German consensus range of 350–600 ng/mL.[5,9,15] The latter range typically corresponds to substantially higher prescribed doses in western countries.[15] Another observation was that 47% samples reported to be lying in the subtherapeutic concentration category considering the Asian ethnicity specific low dosing style and it aligns with observations of the Mach et al.[18] group showing that only 37% polish patients are able to achieve the therapeutic concentration range.

We observed significant correlation between daily dose and trough concentration across dosing ranges. Importantly, the C/D ratio demonstrated strong association with concentration categories made (very low, low, matching, high) generated after paying attention to daily dose as well, supporting its role as a superior interpretative parameter compared with absolute trough levels alone. Variability in C/D ratios reflects differences in hepatic metabolic capacity rather than dose alone. Clozapine metabolism is primarily mediated by CYP1A2, with contributions from CYP2D6, CYP2C19, and CYP3A4. Different metabolic patterns provide mechanistic explanation for observed C/D variability and support individualized dose adjustment strategies rather than blanket adherence to Western suggested ranges.[19]

Significant association between concentration impressions (very low, low, matching, high) and subsequent clinical advice (increase, maintain, or reduce dose advice, compliance advice, etc.) was demonstrated. Compliance was confirmed in 19% cases. Comparable findings have been reported internationally, including Chilean data identifying approximately 12/26 patient noncompliance through TDM.[20] Large UK audits of more than 3,500 reports have similarly demonstrated that very low or absent trough concentrations reliably indicate nonadherence.[21] Dose escalation advice predominated in the very low and low concentration categories. Similar strategies have been described by Rostami-Hodjegan et al.,[22] recommending dose increases when trough concentrations fall below expected therapeutic ranges. Meta analysis from Australia also supports dose uptitration advice to achieve concentrations near 350 ng/mL in addition to suggesting 250 to 500 ng/mL reference range suggestion that closely matches with Asian ethnicity observations.[12]

The novelty of this study lies in integration of pharmacokinetic interpretation with real-world clinical advice, similar to clinical pharmacology consultations tailored for medical specialities.[23]

Limitations

This is a retrospective laboratory audit study, and authors abided by the study protocol approved by the Ethics committee. The study relied on information documented on the requisition forms, and independent verification from clinical records was not performed. The psychiatry team did not consider reporting smoking and caffeine intake history on the requisition forms.

CONCLUSION

Clozapine TDM service utilization is steady and continuing in TRS over the past 4 years. Dose escalation advice predominated in lab reports assessment. Multicenter validation with the same instrument is required on utility of such biomarker-driven clozapine dosing.

Data availability

The data supporting the findings of this study are available from the corresponding author upon request.

Declaration regarding the use of generative AI

The authors did not use generative artificial intelligence (AI) technology in the generation of text, figures, or other informational content of this manuscript. Grammarly was used solely for grammar checking, and Turnitin was used for plagiarism detection.

Ethics approval

Institutional Ethics Committee approval was secured via communication number INT/IEC/2022-SPL-599.

Supplementary material

This article has supplementary material and can be accessed at this link. Supplementary Material at https://links.lww.com/IPSY/A24.

Conflicts of interest

There are no conflicts of interest.

Figure SF1

Requisition form used for clozapine TDM patient care service. Web address - https://pgimer.edu.in/PGIMER_PORTAL/AbstractFilePath?FileType=E&FileName=version%201.1%20Clozapine%20level%20lab%20test%20requisition%20form.pdf&PathKey=EDITORREPOSITORY_PATH. Under heading- https://pgimer.edu.in/PGIMER_PORTAL/PGIMERPORTAL/home.jsp. https://pgimer.edu.in > Download tab > Various forms >65. PLASMA CLOZAPINE LEVEL LAB TEST REQUISITION FORM

Figure SF2

Sample report used for clozapine TDM patient care service

Method for the measurement of trough clozapine concentration

The clozapine reference standard was prepared as a 1 mg/mL stock solution in methanol. High performance liquid chromatography (HPLC)-grade acetonitrile and methanol were used as organic solvents. Orthophosphoric acid, formic acid, ammonium hydroxide, cetrimide (hexadecyltrimethylammonium bromide), and disodium hydrogen sulfate were of analytical grade. Drug-free human plasma was used to prepare calibration standards and quality control (QC) samples. Imipramine was used as an internal standard. Chromatographic analysis was performed using a high-performance liquid chromatography (HPLC) system equipped with a UV Ultra-violet (UV) detector. Separation was achieved using a reversed-phase C18 column (250 × 4.6 mm, 5 μm).

Figure SF3

HPLC method used for clozapine TDM patient care service

Funding Statement

Nil.