To our knowledge, this case series is the first detailed report of pediatric use of fluvoxamine to enhance clozapine. Our four patients clinically improved after the introduction of fluvoxamine as add on to clozapine, enhancing its metabolism and therefore plasma level within clozapine’s concentration therapeutic range. Tolerance was good except for the case 2. However this could be attributed to a drug-drug interaction, since it did not occur when the combination fluvoxamine and clozapine was introduced again. It is important to stress that all our patients presented severe and resistant psychiatric disorders (EOS, ASD and PDD/Tourette syndrome), which justified the initial use of clozapine. It is clear that those patients already represented a therapeutic challenge, as their psychiatric features could not be improved after several therapeutic options. Furthermore, those patients had initially shown no sign of improvement since the introduction of clozapine, with extremely low or negative plasma levels. We would like to discuss three main points: 1) the specifics of the interesting use of clozapine in severe pediatric psychiatric disorders, 2) the importance of exploring CYPs polymorphisms profile in severe psychiatric patients and 3) the encouraging use of fluvoxamine for a customized management of clozapine.
Among the severe psychiatric disorders supporting clozapine, early-onset schizophrenia (EOS) is a rare form of schizophrenia defined by an onset before the age of 18 with a prevalence < 1/10 000 children in the general population (50).Two systematic reviews of the literature (5, 6), have shown the superior efficacy of clozapine in resistant EOS. Clozapine improves all EOS features, including negative symptoms (7, 9) allowing a reduction in the number and period of hospitalizations (5). In our study, clozapine combined with fluvoxamine reduced SAPS score in case report 2 and 3, respectively by 66% and 66,3%, showing therapeutic response. Aggressive behavior is a significant concern with a major burden on the quality of life for patients with ASD and their caregivers. A retrospective review of 135 individuals with ASD demonstrated that 39.5% (n = 53) of individuals met criteria for drug refractory behaviors (defined by trials of risperidone and aripiprazole or three or more psychotropic drugs targeting irritability) (51). Clozapine has received very little attention, despite open label studies suggesting its potential efficacy on aggressive behavior in ASD (21–24, 52) and ID (27, 53). In our study, clozapine improved the ABC-Irritability sub-scores by 32% in case report 1 (severe neurodevelopmental disorder), and 100 % in case report 4 (ASD). Despite its drastic efficacy, clozapine remains rarely used in youth due to its side effects, particularly hematologic toxicity with the risk of agranulocytosis. The risk associated with clozapine can be minimized and better apprehended with a careful and close monitoring in severely ill youth who present psychotic symptoms resistant to conventional treatments (54). Among the four studied patients, clozapine was well tolerated, excepted in case report 2 according to a pre-existing drug-drug interaction involving levomepromazine that was corrected afterwards. None of the four patients had agranulocytosis, myocarditis or seizures, even after the add-on of fluvoxamine (See Additional files 1).
Several CYP450 enzymes are involved in clozapine metabolism. Clozapine pharmacokinetics may vary according to several functional single nucleotide polymorphisms (SNPs). There is a growing interest in the area of pharmacogenomics, but the prevalence of these SNPs is usually studied in adults. To date and to the best of our knowledge, data dealing with pharmacogenetics of antipsychotics in the pediatric population are lacking. Hepatic metabolism of clozapine is complex and involve several enzymes (49, 55, 56). Of these, CYP1A2 isoform plays a major role and the allele CYP1A2*1F (rs762551) is associated with an ultra rapid metabolism of substrates including clozapine and olanzapine (57). Ultra rapid metabolism may increase drug clearance and lead to low plasma levels of various treatments such as clozapine, inciting clinicians to conclude falsy about treatment inefficiency. Some authors found a prevalence around 51.3% for this genotype (58). Clozapine is also metabolized in a same manner by CYP3A4 and CYP2C19. CYP3A4 polymorphism exibits considerable interethnic variability and CYP3A4*22 (rs35599367) appears as the main deleterious SNP in Caucasians(54). CYP2C19 isoform is also highly polymorphic and CYP2C19*2 (rs4244285) leads to a reduced metabolic activity while CYP2C19*17 (rs12248560) is the main variant associated with an increased activity of this enzyme. To a lesser extent, CYP2D6 and CYP2C9 isoforms have been involved in the metabolism of clozapine. An increased activity of CYP2D6 is described in 0.7–5.6% of the caucasians (CYP2D6 copy number variation, i.e CYP2D6*xN) while about 30% are carrying deleterious allele(s) associated to a reduced CYP2D6 activity (59). These deleterious variants mailnly include CYP2D6*3 (rs35742686), CYP2D6*4 (rs3892097), CYP2D6*5 (CNV, gene deletion), CYP2D6*6 (rs5030655), CYP2D6*9 (rs5030656), CYP2D6*10 (rs1065852) and CYP2D6*41 (rs28371725). Regarding the CYP2C9 isoform, CYP2C9*2 (rs1799853), CYP2C9*3 (rs1057910) are the main deleterious variants found in Caucasians. Moreover clozapine and its metabolites have been described as substrates of the highly polymorphic P-glycoprotein (P-gp) (MDR1) drug efflux transporters. However, the effects of P-gp variants on clozapine pharmacokinetics and pharmacodynamics remain unclear (55). In the present case series, two patients exhibited CYP1A2 ultra-rapid metabolizer genotype for clozapine. One other case presented an atypical metabolism profile based on the extremely low clozapine plasma level while none mutations were found. Finally, case n°3 illustrates a limit of such targeted pharmacogenetic approach as no well-characterized polymorphism could explain the low clozapine plasma levels. In this specific situation, the decision of adding fluvoxamine was supported by: i) the absence of all other possible etiology (toxic, diet, drugs) which could have enhance CYP450, ii) the persistence of no clinical response to clozapine, iii) the persistence of undetectable clozapine plasma level, iv) the crucial need to improve psychotic symptoms and v) the fact that the absence of well-known mutations does not exclude other possible mutations not yet identified. This case raised two questions: firstly, for which clinical situation should we proceed CYP genotyping? – and secondly, when should we consider the addition of fluvoxamine to clozapine therapy?
Clozapine plasma level should be weekly monitored and reach a threshold of 350 ng/mL in order to expect efficacy (49). However, as discussed above, CYP450 polymorphisms can result in low plasma level and ineffectiveness of clozapine. Genotyping should be considered when persistent low (under 350 ng/mL) or negative clozapine plasma levels occur in regards of proper dosage (300 to 600 mg/d, from our clinical expertise); and when patients do not exhibit any clinical response and improvement. If both situations are present, fluvoxamine adjunction should seriously be considered (See Fig. 2). The add-on of fluvoxamine to clozapine increases clozapine plasma level and clozapine/norclozapine ratio by inhibiting cytochromes(46). This lead to increase therapeutic effect and reduce adverse effect and with lower dosage(46, 48). Our four patients showed a drastic increase of clozapine plasma level and ratio clozapine/ desmethylclozapine plasma level quickly after the introduction of fluvoxamine 25 mg (See Fig. 1).
Based on our clinical experience and the tolerance issue we faced in case 2, starting fluvoxamine at 50 mg/d following the same pattern than in adult population appeared unsuitable and unsafe for pediatric population with a risk of high increase of clozapine plasma level. Therefore, we reviewed our practice after concertation with the pharmacology department in our institution and proposed a strict protocol for the introduction of fluvoxamine as add on to clozapine in pediatric patients: 1) drugs interactions with fluvoxamine must be carefully assessed in order to identify all possible risk of overdosage of any on-going comedication other than clozapine ; 2) clozapine regimen should was adjusted at 200 mg/d in two separate take (100 mg in the morning and 100 mg in the evening), regardless of the previous dosage ; 3) systematically start fluvoxamine 25 mg/d (half the dosage used for adults) in order to better control the increase of clozapine plasma levels ; 4) therapeutic drug monitoring of clozapine plasma levels was systematic with CBC, and ECG monitoring twice a week; and 5) still targeting clozapine plasma through levels at 350 ng/mL. If this target is not achieved, we recommended to increase fluvoxamine regimen at 50 mg/d. If the clozapine plasma level remained systematically under 350 ng/mL after two weeks, we proposed to increase clozapine dosage with the same pattern described above. This protocol is summarized with an algorithm in Fig. 2.
However, the presented results should be interpreted in the context of several limitations. First, the number of case was low. Second, the absence of double blind use of fluvoxamine doesn’t allow any validation of its safety in youth. The strengths of this case series must be underlined. This is the first report of fluvoxamine’s use in pediatric population. To our knowledge, no other study took interest on the key role of CYP450 polymorphisms and fluvoxamine in order to better target treatment options for severe psychiatric disorders in youth. Despite the interesting results of fluvoxamine, we report a severe issue of tolerance for one patient (case report 2), emphasizing the need for caution regarding possible drugs interactions when fluvoxamine is considered. Hence, we propose a detailed protocol (See Fig. 2). Furthermore, we believe exploring CYP450 polymorphisms associated to a close therapeutic monitoring is an interesting clinical tool to better understand low response to drug or clinical resistance and to propose available therapeutic solutions such as fluvoxamine, besides the standard monitoring of treatment plasma levels. In the end, a close plasma level monitoring and adjunction of fluvoxamine could lead to a better understanding of clozapine efficacy, a targeted treatment approach, a shorter hospitalization, and less relapse in the future.