There Is Life After the UK Clozapine Central Non-Rechallenge Database

Abstract

Background and Aims

In the United Kingdom, patients on clozapine whose hematological parameters fall below certain thresholds are placed on the Central Non-Rechallenge Database (CNRD), meaning that they cannot be prescribed clozapine again except under exceptional circumstances. This practice was discontinued in the United States in 2015 by expanding the hematological monitoring guidelines, allowing more patients to receive clozapine. Our objective was to investigate the implications this policy change would have on clozapine utilization in the United Kingdom.

Methods

This was an observational, retrospective analysis of patients registered on the CNRD in a large mental health trust. The first objective was to compare the number of patients placed on the CNRD under the United Kingdom and the US Food and Drug Administration (FDA) criteria. The second objective was to explore the hematological and clinical outcomes of CNRD patients. The third objective was to investigate the hematological outcomes of patients rechallenged on clozapine after nonrechallengeable status.

Results

One hundred and fifteen patients were placed on CNRD from 2002 to 2019, of whom 7 (6%) met the equivalent criteria for clozapine discontinuation under the FDA guidelines. Clinical outcomes, as measured by the Clinical Global Impression-Severity scale, were worse 3 months after clozapine cessation than on clozapine (t = −7.4862; P < .001). Sixty-two (54%) patients placed on CNRD were rechallenged. Fifty-nine of those (95%) were successfully rechallenged; 3 patients were placed back on CNRD, only one of which would have had to stop clozapine again under FDA criteria.

Conclusion

Implementation of the updated FDA’s monitoring criteria in the United Kingdom would significantly reduce clozapine discontinuation due to hematological reasons. The evidence suggests an urgent need for revising the UK clozapine monitoring guidelines to improve outcomes in treatment-resistant schizophrenia.

Introduction

Clozapine has been regarded as the gold-standard antipsychotic for treatment-resistant schizophrenia (TRS) since the seminal study of Kane et al in the 1980s.¹ Despite this, the underuse of clozapine, including significant delays in the initiation, the use of unjustified antipsychotic polypharmacy, and early discontinuation of treatment, has been well documented.² This can be attributed in part to the safety concerns around the risk of neutropenia and agranulocytosis, which are reported to occur with a prevalence of 3.8% and 0.4%, respectively, in clozapine-treated patients.^3,4

Clozapine associated neutropenia was first identified in the 1970s, following fatalities amongst 8 Finnish patients and it prompted stringent hematological monitoring in most countries when the drug was reintroduced in 1989.⁵ In the United Kingdom, white cell counts (WCC) and absolute neutrophil count (ANC) monitoring is a requirement for continuing clozapine therapy, so that any emergent neutropenia can be detected early and clozapine treatment promptly discontinued.⁶ This system has been effective in preventing deaths from severe neutropenia (also known as agranulocytosis), with only 8 fatalities from clozapine-induced blood dyscrasias to date in the United Kingdom.⁷ However, the intensive nature of this monitoring practice has presented further barriers to treatment initiation and maintenance.⁸

In October 2015, the US Food and Drug Administration (FDA) updated its clozapine regulations, lowering the ANC cutoff for clozapine cessation compared with the United Kingdom⁹ (table 1). In addition, the requirements for monitoring WCC were removed. Patients with a formal diagnosis of benign ethnic neutropenia (BEN), a condition seen mainly in individuals of African descent who have low ANC values less than 1500/µL, without an increase in infections, were able to commence clozapine treatment under even lower thresholds⁸ (table 1). These changes were intended to reduce barriers to the initiation and continuity of clozapine treatment with a negligible effect on patient safety.¹⁰ Similar changes have been made to clozapine monitoring guidelines in other countries,¹¹ replicating the FDA recommendations. However, such changes have yet to be adopted in the United Kingdom.

Table 1.

Comparison of UK Medicines and Healthcare Products Regulatory Agency (MHRA) Guidelines for Clozapine Monitoring Compared With US FDA Guidelines

Current UK MHRA Guidelines				Current US FDA Guidelines
Status	GeneralPopulationCriteria	BEN Criteria	Action	Status	GeneralPopulationCriteria	BEN Criteria	Action
Green				Normal range
WCC	≥3·5 × 109/L	≥3·0 × 109/L	Continue clozapine	WCC	Not mandatory	Not mandatory	Continue clozapine
ANC	≥2·0 × 109/L	≥1·5 × 109/L		ANC	≥1.5 × 109/L	≥1.0 × 109/L
Amber				Mild neutropenia
WCC	≥3·0 and <3·5 × 109/L	≥2·5 and <3·0 × 109/L	Continue clozapine with increased monitoring frequency (twice a week) until green result is obtained.	WCC	Not mandatory	Not mandatory	General criteria: Continue clozapine with increased monitoring frequency (3 times a week) until ANC returns to normal range. BEN: Continue clozapine
ANC	≥1·5 and <2·0 × 109/L	≥1·0 and <1·5 × 109/L		ANC	≥1·0 and <1.5 × 109/L	≥0·5 and <1.0 × 109/L
Red				Moderate-severe neutropenia
WCC	<3·0 × 109/L	<2·5 × 109/L	Stop clozapine (interruption) and sample blood daily, monitoring for infection. Patient not re-exposed to clozapine until there have been 2 separate green results on 2 consecutive days. Two consecutive red results are classified as a “confirmed” red result, where a patient is registered on the CNRD.	WCC	Not mandatory	Not mandatory	Hematology consultation recommended, stop treatment (interruption) for suspected clozapine-induced neutropenia. Moderate neutropenia: Sample blood daily. Resume treatment once ANC normalizes to ≥1.0 × 109/L aSevere neutropenia: Sample blood daily. Do not rechallenge unless prescriber determines benefits outweigh risks.
ANC	<1·5 × 109/L	< 1·0 × 109/L		ANC	<1.0 × 109/L	<0.5 × 109/L

Note: ANC, absolute neutrophil count; BEN, benign ethnic neutropenia; FDA, Food and Drug Administration; MHRA, Medicines and Healthcare Products Regulatory Agency; US: United States; REMS: Risk Evaluation and Mitigation Strategy; WCC, white cell counts.

^aThe prescriber must provide a “treatment rationale” in the national monitoring system REMS but there is no mandatory discontinuation or mandatory nonrechallenge category.

It is not clear how adopting the updated US hematological thresholds would affect the number of patients who meet the cutoff point for clozapine discontinuation in the United Kingdom. The Clozapine Central Non-Rechallenge Database (CNRD) provides data to examine this question, as it includes all patients in the United Kingdom deemed to have a hematological reaction to clozapine, ensuring that they are not inadvertently re-exposed to clozapine treatment.¹² It has been suggested that the adoption of the US parameters in the United Kingdom would prevent frequent, unnecessary, and costly clozapine discontinuations.¹³ In this study, we aim to investigate this hypothesis.

Method

To assess the potential impact that the implementation of the US FDA guidelines would have on clozapine continuation in the United Kingdom, we examined data for patients registered for clozapine use from South London and Maudsley (SLaM) NHS Foundation Trust. SLaM provides mental health services to 1.2 million people across 4 South London boroughs, as well as national specialist services, including the National Psychosis Unit, which specializes in the management of complex, treatment-resistant psychosis. The first objective was to compare the number of patients registered on the CNRD under the current UK Medicines and Healthcare Products Regulatory Agency (MHRA) guidelines with the number who would have been placed on the CNRD under revised US FDA criteria. The second objective was to explore the hematological and clinical outcomes of individuals who were placed on the CNRD. We investigated the time to the first below threshold hematological result, the time to CNRD registration from first clozapine initiation, and the number of severe neutropenia cases in our cohort, as well as potential contributing factors such as concomitant medication(s). Time to CNRD registration was presented using a Kaplan-Meier plot. In addition, we evaluated the clinical impact of clozapine cessation due to low ANC and/or WCC values using the Clinical Global Impression-Severity (CGI-S) scale.¹⁴ The third objective was to investigate the hematological outcomes of patients rechallenged on clozapine after nonrechallengeable status and explore whether subsequent hematological events occurred upon re-exposure to clozapine.

Data Source and Ethical Approval

The study used data from the Clinical Records Interactive Search (CRIS) system; a large, anonymized case register derived from SLaM NHS Foundation Trust electronic case records, fully described elsewhere.¹⁵ The CRIS system allows authorized researchers to search structured and free-text fields of over 300 000 patients following formal approval for the search and is linked to external clinical databases such as the national Zaponex Treatment Access System (ZTAS) clozapine register.¹⁶ Oxfordshire Research Ethics Committee C (19-049) gave ethical approval for the use of CRIS as a research dataset and the CRIS oversight committee granted permission for this study.

Definition of Neutropenia, CNRD, and Clozapine Rechallenge

In the United Kingdom, clozapine monitoring is regulated by the MHRA using the criteria set out in table 1, which includes lower hematological cutoff points for patients with a diagnosis of BEN since 2002. In the United Kingdom, there are 3 manufacturers of clozapine and associated hematological monitoring services: Clozaril (Mylan) monitored by Clozaril Patient Monitoring Service (CPMS),¹⁷ Denzapine (Britannia Pharmaceuticals Limited) monitored by Denzapine Monitoring System (DMS),¹⁸ and Zaponex (Leyden Delta BV) monitored by ZTAS.¹⁹

In this study, hematological events were defined according to the MHRA guidelines for ANC and WCC values that required clozapine treatment interruption or discontinuation (table 1). Under current recommendations, 2 consecutive red results are classified as a “confirmed” red result. In this event, the relevant clozapine monitoring service submits the patient’s details to the CNRD, then the patient is classified as “nonrechallengeable,” where re-exposure to clozapine is not permitted expect under exceptional circumstances. After the establishment of CNRD in 2002, the sole manufacturer of clozapine at the time (Clozaril) added all patients who had previously met CNRD criteria to the register retrospectively. Prior to changes to the MHRA clozapine monitoring requirements in 2013, patients could meet CNRD criteria with a single red result. In exceptional cases, the product manufacturer’s medical director may register patients on the CNRD if treatment continuation is considered unsafe after repeated neutropenic events (including amber results) or in the absence of follow-up hematological monitoring after an initial red result. Clinical practice dictates that some patients were registered on the CNRD many years ago, without formal assessment to determine whether the hematological event was indeed caused by clozapine and whether CNRD registration was only applicable with preceding MHRA criteria.

Clozapine manufacturers allow for the re-exposure of clozapine following CNRD registration under an off-licence agreement.²¹ The decision for clozapine rechallenge is undertaken on an individual basis and must be agreed by a multidisciplinary team, in close liaison with a consultant hematologist. The final decision is driven by a comprehensive assessment, which includes extensive information gathered from various sources such as hematological profiling such as blood films and hematinics.²²

Study Sample

The study sample included all active SLaM patients on 1 October 2019 who had ever been registered on the CNRD. The “nonrechallengeable” status in patient history was used to identify these patients from the CRIS-linked clozapine monitoring system used in SLaM: ZTAS. The initial and confirmatory (if present) ANC and WCC values that resulted in nonrechallengeable status were identified from CRIS ZTAS-linkage or patient notes. The flowchart in figure 1 shows the sample selection.

Figure 1.

Study flowchart.

Demographic and prescription data, including concomitant medication at the point of CNRD registration, were obtained from anonymized electronic medical records and dispensing records at SLaM. Clozapine dosage at CNRD registration and length of treatment from “first clozapine prescription” and “active clozapine treatment” to CNRD registration were recorded. Active clozapine treatment was defined as continuous clozapine treatment without a recorded treatment break longer than 1 week, whereas “first clozapine prescription” refers to a noncontinuous course of treatment, where applicable. For all patients, the number of days of clozapine treatment at the time of the first red result was recorded.

The numbers of patients who developed severe neutropenia were identified. Severe neutropenia was defined as an ANC values <0.5 × 10⁹/L while treated with clozapine.³

To ensure reliability, another author (CC) reviewed the extracted data and checked its accuracy by independently examining anonymized patient electronic notes. Any disagreement was discussed, clarified, and decisions were documented.

Proportion of Patients Who Would Have Discontinued Clozapine Under US FDA Criteria

The US FDA monitoring parameters (table 1) were applied to all SLaM patients who had ever been registered on the CNRD. Of these patients, 6 had a diagnosis of BEN confirmed by a hematologist and therefore US BEN monitoring parameters were applied to these patients.

Hematological Outcomes of Patients Placed on CNRD

The time to the first red result and time to CNRD registration from clozapine initiation and the number of severe neutropenia cases in our cohort were identified from CRIS. Time to CNRD was presented using a Kaplan-Meier plot.

Clinical Outcomes of Patients Placed on CNRD

The CGI-S scale¹⁴ evaluates the severity of psychopathology capturing clinical impressions on a 7-point scale that ranges from 1 (“Normal, not at all ill”) to 7 (“Among the most extremely ill patients”). It was used to retrospectively evaluate mental illness severity for patients who had not been rechallenged on clozapine after CNRD registration. It was retrospectively assessed by an experienced psychiatrist from anonymized clinical notes, derived at 2 time points: at the last clinical review while on a consistent dose of clozapine (for at least 6 weeks) and 12 weeks after discontinuation.⁵

Hematological Outcomes of Clozapine Rechallenge

Patients rechallenged on clozapine after CNRD registration were identified using the “on-treatment” status after a previous “nonrechallengeable” status in ZTAS patient history. This was then confirmed by manual analysis of patient’s anonymized notes in the electronic medical records. For patients with more than one rechallenge attempt, only the first event was included in the primary analysis. The time to clozapine rechallenge and the number of antipsychotic trials from registration on the CNRD to rechallenge were recorded. Where applicable, the date of clozapine discontinuation and reason for discontinuation were recorded from patient notes. Treatment discontinuation was defined as a discontinuation for longer than 7 consecutive days, even if clozapine was later reinitiated. The number of patients who met CNRD criteria again after rechallenge was also documented. For these patients, FDA parameters were applied to the hematological results to determine whether re-registration would have occurred. All-cause mortality was recorded after clozapine rechallenge until the end of the study period or discharge from SLaM, whichever came sooner. Successful rechallenge was defined as no recurrence of CNRD registration.

Statistical Analysis

Statistical analysis was carried out using Stata, version 15.²³ The percentage of patients who would not have met CNRD criteria and clozapine interruption under the revised US guidelines were calculated. The change in mean CGI-S total score was analyzed using a paired sample t-test. Kaplan-Meier survival curves were used to graphically present the time to CNRD registration from active clozapine treatment, that is clozapine initiation with no interruption longer than 7 days.

Results

A total of 115 unique patients had been once registered on the CNRD and were included in this study (figure 1). Clinical and demographic characteristics of the study population are shown in table 2.

Table 2.

Sociodemographic and Clinical Characteristics of Patients Registered on the CNRD

Characteristic	Not Rechallenged (n = 53)	Rechallenged (n = 62)	Total (n = 115)
Male gender (%)	32 (60)	39 (63)	71 (62)
Age at CNRD registration (years ± SD)	44 (14)	36 (12)	40 (14)
Duration of illness (years ± SD)	16.3 (7.0)	21.5 (8.6)	19.3 (8·3)
Mean time to clozapine initiation from illness onset (years ± SD)	6.4 (5.5)	7.3 (7.6)	6.9 (6·7)
Median time to clozapine initiation from illness onset (years, IQR)	6.0 (9.0)	5.0 (8.8)	5.0 (8·0)
BEN diagnosis (%)	4 (8)	20 (32)	24 (21)
Mean time to BEN diagnosis from first clozapine initiation (years ± SD)a	2.0 (1.8)	10.2 (8.2)	8.8 (8·1)
Median time to BEN diagnosis from first clozapine initiation (years, IQR)a	2.0 (2.5)	10.5 (12.0)	7.0 (10·8)
Diagnosis (%)
F20 paranoid schizophrenia	37 (70)	49 (79)	86 (75)
F25 schizoaffective disorder	14 (26)	12 (19)	26 (23)
F31 bipolar disorder	1 (2)	1 (2)	2 (2)
Other	1 (1)	0 (0)	1 (1)
Psychiatric comorbidity	3 (6)	5 (8)	8 (7)
Ethnicity (%)
White	21 (40)	33 (53)	54 (47)
Black	25 (47)	20 (32)	45 (39)
Asian	4 (8)	5 (8)	9 (8)
Other	3 (6)	4 (6)	7 (6)
Concomitant medicationb (%)
Antipsychotic polypharmacy	14 (26)	10 (16)	24 (21)
Valproate	24 (45)	23 (37)	47 (41)
Lithium	7 (13)	3 (5)	10 (9)
Hematological characteristics
Mean time to first red result (years ± SD)	3.2 (4.5)	5.5 (6.8)	4.4 (5.9)
Median time to first red result (years, IQR)	1.0 (6.0)	2.0 (11.0)	2.0 (6.5)
Cases of severe neutropenia	3 (6)	2 (3)	5 (4)
Mean time to CNRD registration from first clozapine prescription (years ± SD)	3.5 (4.5)	5.6 (6.8)	4.6 (5.9)
Median time to CNRD registration from first clozapine prescription (years, IQR)	2.0 (6.0)	2.5 (10.6)	2.0 (7.5)
Mean time to CNRD registration from active clozapine treatment (years ± SD)	2.5 (4.3)	5.6 (6.8)	4.1 (6.0)
Median time to CNRD registration from active clozapine treatment (years, IQR)	0 (3.0)	2.5 (10.6)	1.0 (6.0)

Note: BEN, benign ethnic neutropenia; CNRD, Central Non-Challenge Database; IQR, interquartile range.

^aSix patient started clozapine before BEN clozapine monitoring criteria introduction in the United Kingdom (2002).

^bMissing data: 4 patients.

Comparison of Patients with Hematological Events During Clozapine Treatment Under UK MHRA and US FDA Guidelines

Table 3 outlines and compares the distribution of patients who would have met the criteria for clozapine treatment interruption or discontinuation under current MHRA guidelines and their treatment status if FDA criteria were applied. Of the 115 patients who had hematological events that qualified for nonrechallengeable status and treatment discontinuation under current MHRA guidelines, 7 patients would have qualified for nonrechallengeable status and clozapine discontinuation under FDA guidelines; 67 patients would have required 3 times a week hematological monitoring; and 12 patients would have required clozapine interruption with daily hematological monitoring.

Table 3.

Comparison of Clozapine Treatment Status and Monitoring Requirements for SLaM Patients Ever Registered on the CNRD Under Current UK MHRA Criteria and US FDA Criteria

Guideline	Continue Clozapine With Routine Monitoring	Twice a Week Monitoring	Daily Monitoring and Clozapine Interruption	Clozapine Discontinuation
MHRA (United Kingdom)	0 (0%)	0 (0%)	115 (100%)	115 (100%)
FDA (United States)	29 (26%)	67 (58%)	12 (10%)	7 (6%)

Note: MHRA, Medicines and Healthcare Products Regulatory Agency; FDA, Food and Drug Administration.

Clinical Outcomes in Nonrechallenged CNRD Patients

The median CGI-S score for patients who were not rechallenged was 4 (interquartile range [IQR] 3–4) during clozapine treatment compared with 5 (IQR 4–6) post-3 months after clozapine cessation (t = −7.4862, P < .001).

Time to CNRD Registration

Figure 2 shows the Kaplan-Meier curve of the time to CNRD registration from clozapine initiation. The median time to CNRD registration from “active clozapine treatment” was 518 (IQR 70–2550) days.

Figure 2.

Kaplan-Meier for the time to CNRD registration from first clozapine initiation.

Outcomes of Patients Rechallenged on Clozapine

Table 4 outlines the patients who were rechallenged on clozapine after registration on the CNRD. Of the 62 patients rechallenged, 3 were placed back on the CNRD following rechallenge. Only one of these patients would have had to discontinue clozapine under US FDA criteria. Eighteen patients were given a diagnosis of BEN. Of these patients, 16 (89%) were black and 2 (11%) of other ethnicity. Three Caucasian patients were monitored off-license with BEN criteria on rechallenge. There was one case of severe neutropenia, which resolved without medical treatment after cessation of clozapine. This patient had a previous case of severe neutropenia before initial CNRD registration.

Table 4.

Clozapine Rechallenges

Characteristic (Mean, SD)	Number of Cases (N = 62)	Number of Patients Placed Back on CNRD After Rechallenge (n = 3)
BEN diagnosis after CNRD registration (%)	18 (29)	2 (67)
Time to BEN diagnosis from first clozapine initiation (years)a	10.2 (8.2)	8.5 (7.8)
Number of antipsychotic trials before rechallenge (± SD)	3 (2)	1 (1)
Time to rechallenge from CNRD registration (years)	3.6 (5.7)	1.8 (1.4)
Medication on rechallenge (n, %)
Antipsychotic polypharmacy	15 (24)	0 (0)
Valproate	15 (24)	0 (0)
Lithium	29 (47)	1 (33)
GCSF prescription; administration	9 (15); 3 (5)	0 (0); 0 (0)

Note: CNRD, Central Non-Rechallenge Database; BEN, benign ethnic neutropenia; GCSF, granulocyte-colony stimulating factor.

^aSix patients started clozapine before BEN clozapine monitoring criteria introduction in the United Kingdom (2002) – (average when excluding these patients = 7.70 ± 5.42).

In the cohort of rechallengers, 13 patients (21%) stopped clozapine treatment during the follow-up period (mean follow-up 4.41 years ± 2.74). Three patients stopped clozapine treatment due to hematological reasons, and 10 patients for other reasons, eg patient choice.

Of the 3 patients who met CNRD criteria again, 2 discontinued clozapine, whereas one patient was successfully rechallenged on a second occasion and remained on clozapine at the end of data collection, 3 years later. One death occurred during the study period, from a nonhematological cause.

Discussion

Our Findings

To our knowledge, this is the first study in the United Kingdom that has investigated the potential impact of changes to the UK MHRA clozapine monitoring guidelines on treatment access and maintenance. Of the 115 patients who were placed on the CNRD and therefore had to discontinue clozapine immediately, only 7 (6%) would have had to discontinue clozapine immediately under US FDA criteria. In the United Kingdom, a further 67 (58%) would have had an amber result in the United Kingdom under US FDA criteria.

Comparison With Other Studies

Our study complements the previous retrospective cohort study by Sultan et al. in the United States,²⁴ where a comparison of the proportion of patients who developed or would have developed at least one hematological event under the previous and revised FDA guidelines in the first year following initiation of clozapine was made. The authors reported that 5 (3.1%) of 246 TRS patients would have qualified for treatment interruption during the first year of clozapine treatment under the previous US guidelines, which are equivalent to current UK guidelines, whereas only one patient (0.65%) qualified under the revised FDA recommendations.

Limitations

A key limitation of this study is the fact that we cannot know what would have happened to the 67 patients (58%) who had to discontinue clozapine under UK guidelines but would have registered with an amber result under the US FDA thresholds, if they had continued to take clozapine. An amber result in the United States does not warrant treatment discontinuation, but it does mandate repeated hematological monitoring (3 times a week) until resolved. Because clozapine was discontinued in our patients under the monitoring criteria in the United Kingdom, it is unknown what proportion of them would have developed more severe neutropenia with continued clozapine use. However, several authors have observed that most amber results (mild neutropenia) do not augur a serious hematological reaction to clozapine (ie, severe neutropenia) and there is evidence that clozapine-induced severe neutropenia (CISN) and clozapine-induced neutropenia (CIN) may have different etiologies.^25–27

Another limitation relates to the patients who were rechallenged on clozapine. Patients are typically only rechallenged after a comprehensive risk-benefit assessment in close liaison with a hematologist in an inpatient setting, where any potential deterioration in physical health can be promptly detected.²¹ Our center hosts several specialist services, including the National Psychosis Service, a tertiary referral service with a special interest in clozapine rechallenge in patients in whom this is deemed to be safe.²⁸ This could limit the generalizability of our rechallenge results because patients are thoroughly investigated with hematology input prior to rechallenge, so (a) patients deemed at high risk of developing neutropenia and potential severe neutropenia are not rechallenged and (b) our specialist services may have rechallenged patients who would not be rechallenged in routine settings. Our data revealed that 29 patients (47%) were given treatment designed to elevate ANC,⁸ including lithium (47%) and granulocyte-colony stimulating factor prescription (15%); these interventions are used less frequently outside specialist settings, so the data must be interpreted with caution. Future research with more focus on the clinical impact of CNRD status in routine settings is recommended. Furthermore, assessing clinical outcomes retrospectively is another limitation, as it does not control for any variable except clozapine and was done by a clinician who was not evaluating the patients.

Clozapine Underutilization

There is an increasing urgency to improve access to clozapine for individuals who qualify for this uniquely effective treatment, a position reflected in international priorities.^29,30 Several noteworthy initiatives to address this need have been reported in the literature,^32–34 including the aforementioned decision by the FDA to relax the hematological requirements to enable more patients with TRS to be initiated and maintained on clozapine.²⁴ Despite recent literature, it is unclear whether these changes have contributed to an increase in clozapine prescriptions in the United States.^10,35

There is compelling evidence to suggest that premature clozapine discontinuation in TRS has significant adverse effects on patient outcomes, especially given the existing challenges of treatment initiation and maintenance.⁵ Epidemiological data suggest that the most common reasons for clozapine discontinuation are hematology related,³⁶ which is likely due to clozapine’s widely publicized risk of neutropenia. However, recent literature has suggested that clozapine discontinuation due to hematological reasons is often done unnecessarily.¹³ Frequently, findings of neutropenia during clozapine monitoring are unrelated to drug treatment, resulting instead from causes such as undiagnosed BEN, vitamin deficiencies, and viral infections.⁵ The widely cited 3.8% prevalence rate of CIN is therefore likely to be an overestimation. Furthermore, survey data indicate that limited knowledge of the prevalence and management of CISN an additional barrier to treatment.³⁷

Clozapine-Induced Neutropenia

Despite clozapine being widely available for decades, the precise mechanism of CIN or CISN remains unclear, though it is clear that susceptibility is to a large extent genetically mediated.²⁴ Notwithstanding, the FDA’s decision to change the criteria for clozapine interruption in 2015 reflects known clinical and scientific knowledge of immunology.²⁴ ANC values above 1000/μL are generally sufficient to provide phagocytic defense, while counts below 500/μL increase the risk of opportunistic infections.³⁷ Agranulocytosis is commonly defined as an ANC value below 500/μL; however, serious infections are often most likely to occur at counts fewer than 200/μL.³⁸ Moreover, diurnal variation in circulating neutrophil granulocytes, dependent on blood sampling time, has been reported.³⁹ There are also suggestions that factors such as exercise and smoking can increase circulating neutrophil granulocytes.⁴⁰ Post hoc exploration of our data showed that 14 patients (12%) were registered onto the CNRD solely due to WCC, even when ANC values were within the threshold necessary to continue treatment (results not shown). Interestingly, under revised FDA guidelines, these patients would not have discontinued because WCC is not assessed.²⁰

Neutropenia Timescale and Severe Neutropenia Episodes

The risk of neutropenia is highest between weeks 6 and 18 of the first 6 months of clozapine treatment.⁴¹ Data from the United Kingdom and other countries indicate that 76% of CIN occurs in the first year of treatment.⁴² Although some cases of late-onset CISN or CIN have been reported in the literature.^43,44 In our study, the first neutropenic event was recorded at median of 2 years of treatment. These findings suggest that many of the later cases of neutropenia were unrelated to clozapine use.

Five patients (4%) had an episode of severe neutropenia during clozapine treatment, all of whom recovered. This occurred after a median of 18 weeks from clozapine initiation (results not shown) compared with a median of 2 years for CIN, which supports the view that CIN and CISN are clinically distinct.²⁷ From a clinical standpoint, this is perhaps more relevant because the risk of severe infection is significantly greater in severe neutropenia.⁴⁵

Importance of Concomitant Medication

Previous studies have identified significant and independent factors associated with CISN or CIN such as female sex and advanced age.^42,46 In addition, concomitant medication that interacts with clozapine or directly causes hematological toxicity may account for or contribute to, some cases of neutropenia.⁴⁷ In our cohort, 47 patients (41%) received coadministration of valproic acid at CNRD registration. Because several studies propose an increased incidence of neutropenia with concomitant valproic acid use,^48,49 it is possible that neutropenia could have been avoided in these patients. Nevertheless, despite limited evidence to support the efficacy of valproic acid as an adjunct to clozapine treatment, its common use for clozapine-induced seizure prophylaxis necessitates a comprehensive risk-benefit evaluation.⁵⁰ Overall, 9 patients (8%) in the CNRD cohort were coprescribed lithium. Since lithium is commonly used to boost WCC,⁸ confounding by indication was possible and no firm conclusions can be drawn on its effect.

Clinical Outcomes

Adequate and sustained symptomatic control is often a prerequisite for good functional recovery in the treatment of psychiatric conditions.⁵¹ Several studies have consistently emphasized the importance of continued clozapine treatment, where safely tolerated, given its unique therapeutic benefits in TRS.^52,53 Furthermore, evidence has shown that antipsychotic treatment response is reduced or delayed after relapse due to nonadherence, and a case series suggests that this could also be true for clozapine.⁵⁴ Clinical outcomes in our study, measured by CGI-S, were worse after clozapine discontinuation. Median CGI-S scores were 4 (IQR 3–4) (“moderately ill”) before stopping clozapine compared with 5 (IQR 4–6) (“markedly ill”) 3 months after clozapine discontinuation. Additionally, the mean time scale for clozapine rechallenge was 3.6 (SD 5.7) years, with a mean of 3.2 (SD 2.3) antipsychotic trials during this timeframe. These findings agree with prior studies that emphasize the importance of continued clozapine treatment for seriously and persistently ill patients, especially given the limited clinical benefit of using nonclozapine strategies to address inadequate antipsychotic response.⁵²

Clozapine Rechallenge

Although the results contributed to existing literature, one important question that remains unanswered was whether a change in clozapine blood monitoring parameters would compromise patient safety. We attempted to answer this question by identifying patients who were rechallenged on clozapine following a nonrechallengeable status. Previous research in our care setting demonstrated that 79% of patients were successfully reestablished on clozapine therapy following previous treatment-emergent neutropenia.²¹ In our study, we identified 62 patients who were rechallenged on clozapine following a nonrechallengeable status. Importantly, under the US FDA monitoring criteria and with early identification of BEN, 59 patients (95%) would not have originally attained a nonrechallengeable status. Of the 62 patients rechallenged, 3 patients (5%) were required to stop clozapine treatment due to suspected CISN or CIN. The one fatality recorded in the study period was not due to hematological causes. Although this evidence is positive, the high success rates in the patients rechallenged cannot be extrapolated to all patients as there is likely bias whereby patients at lower risk of neutropenia were selected for rechallenge.

Clinical Implications and Future Research

Our study data show that BEN was recognized in 18 patients at the stage of clozapine rechallenge. Early recognition of BEN in these patients could have prevented initial CNRD registration and avoided unnecessary treatment interruption. This is further supported by the 11-year mean (table 4) taken to recognize BEN from clozapine initiation. Furthermore, it is notable that patients of black ethnicity constituted a smaller proportion of the rechallenged patients (32% vs 47% in non-black ethnicities), especially important in view of unidentified BEN as a plausible cause for initial CNRD registration. This inconsistency may be partially due to the stringency of hematological thresholds in the United Kingdom leading to treatment interruptions even in individuals who are identified with BEN, which further emphasizes how implementing the less-restrictive US FDA thresholds could improve treatment maintenance in the United Kingdom.

Discussions with ZTAS/CPMS revealed that approximately 4000 patients are currently registered on the CNRD in the United Kingdom, of whom 20% are estimated to have been rechallenged on clozapine. Extrapolating the results of our study may suggest that up to 90% of these patients could have continued clozapine treatment if the US FDA guidelines are implemented in the United Kingdom. To put this in perspective, this means up to 2800 patients are potentially deprived, unnecessarily, of the unique benefits of clozapine. However, this may be an overestimate given the likely overrepresentation of unrecognized BEN in our study population.

This article is a call to action for an urgent review of current clozapine monitoring guidelines in the United Kingdom. If changes are made to the guidelines in the United Kingdom, this could significantly improve the continuity of clozapine treatment in practice. Moreover, the summary of product characteristics for all countries regulated by the European Medicines Agency have equivalent hematological cut off points for clozapine discontinuation, suggesting that an evaluation of international guidelines may be necessary. Nevertheless, further work, on a larger scale, may be required to establish this.

Conclusion

Implementation of the US FDA’s clozapine monitoring guidelines in the United Kingdom may reduce the necessity for increased blood monitoring and clozapine treatment interruption and cessation, with the associated deterioration in mental state. Selected patients who developed previous neutropenia have been successfully rechallenged on clozapine. This evidence suggests the urgent need for a revision of the clozapine monitoring guidelines in the United Kingdom to improve clinical outcomes in TRS. It also highlights a need for greater awareness and earlier diagnosis of BEN.

Conflict of interest

The authors have declared that there are no conflicts of interest in relation to the subject of this study.