Slower clozapine titration than the official Japanese protocol led to fewer inflammatory adverse effects: A retrospective chart review of seven hospitals

Background: Higher frequencies of inflammatory adverse effects of clozapine have been reported in Japan. As the international titration protocol for Asians has set slower dose titration than the Japanese package insert, we hypothesized that a dose titration speed slower than the recommendation of the guideline would be associated with fewer inflammatory-related adverse events. Methods: The medical records of all 272 patients who were first started on clozapine at seven hospitals between 2009 and 2023 were studied retrospectively. Of those, 241 were included in the analysis. The patients were divided into two groups regarding whether the titration speed was faster or slower than the guideline for Asians. The incidence of inflammatory adverse events with clozapine was compared between the groups. Results: The frequency of inflammatory adverse events was 34 % (37/110) in the faster titration group and 13 % (17/131) in the slower titration group, and a significant difference was observed by Fisher exact test (odds ratio 3.38; 95 % confidence interval 1.71 – 6.91; p < 0.001). Serious adverse effects, fever for more than five days, and clozapine discontinuation were significantly more frequent in the faster titration group. Logistic regression analysis indicated significantly more inflammatory adverse events in the faster titration group (adjusted odds ratio 4.01; 95 % confidence interval 2.02 – 7.87; p < 0.001) considering age, sex, body mass index, concomitant valproic acid, and smoking as confounding factors. Conclusion: Clozapine-induced inflammatory adverse events were less frequent in Japanese individuals when a titration rate was more gradual than the protocol recommended in the Japanese package insert.


Inflammatory adverse effects with clozapine
Clozapine is underused worldwide owing to its wide array of adverse effects (Bogers et al., 2016).Agranulocytosis, among the adverse effects of clozapine, has received the most attention since the Food and Drug Administration (FDA) approval in 1989 with the introduction of frequent blood testing systems (Crilly, 2007).Stringent neutrophil monitoring criteria can lead to the discontinuation of clozapine in some patients.In 2015, the FDA changed the standard for neutrophil monitoring to ≥1500/μl, and countries seek to relax the relevant requirements (Oloyede et al., 2022a(Oloyede et al., , 2022b)).On the other hand, the inflammatory adverse effects of clozapine have received increasing attention in recent years (de Leon et al., 2020a).According to a study using the World Health Organization (WHO) pharmacovigilance database, there have been 34,931 reported cases of agranulocytosis (or severe neutropenia) since inception through April 2019, with only 550 deaths providing a relative fatality rate of 2 % within reports.In contrast, there were 6983 and 4586 reported cases of pneumonia and myocarditis, respectively, with relative fatality rates of 30 % (2077/ 6983) and 12 % (539/4586), respectively (De Las Cuevas et al., 2021;de Leon et al., 2020b).Inflammatory adverse effects of clozapine range from transient, self-limited ones called clozapine-induced fever to serious ones such as myocarditis, pneumonia, enteritis, nephritis, and pancreatitis (Verdoux et al., 2019).A serious condition called drug reaction with eosinophilia and systemic symptoms (DRESS) syndrome, characterized by hypereosinophilia, rash, and organ damage, has also been observed (de Filippis et al., 2022c).The clozapine-related DRESS syndrome is mainly characterized by hypereosinophilia, fever, and internal organ implication more than other signs (de Filippis et al., 2022a).These inflammatory adverse effects, often accompanied by eosinophilia and elevated C-reactive protein (CRP), often occur within 4 weeks of starting clozapine.Based on these findings, it has been pointed out that a series of spectra may be formed through immunological mechanisms (de Filippis et al., 2022a(de Filippis et al., , 2022b)).

Proposal of the guideline for clozapine titration by ethnicity
Jose de Leon and colleagues recently published the international guideline for clozapine titration by ethnicity (de Leon et al., 2022a(de Leon et al., , 2022b)).Blood levels of clozapine tend to increase more rapidly in Asians than in Caucasians with the same doses.Therefore, the guideline suggests that psychiatrists should increase clozapine more slowly in Asians, as rapid dose escalations can lead to increased inflammatory adverse effects.

Reports of inflammatory adverse effects with clozapine in Japanese patients
Yada et al. reported that the mean concentration/dose ratio of clozapine in Japanese individuals was 1.80 (ng/ml per mg/day), which was higher than in Caucasian samples (Yada et al., 2021).Two previous studies investigating the adverse events of clozapine in Japanese individuals reported that the frequency of clozapine-induced fever was 29 % (11/38) and 38 % (57/152), respectively (Kishi et al., 2013;Tsukahara et al., 2021).This is more frequent than earlier Western reports (Gaertner et al., 1989;Naber et al., 1989;Safferman et al., 1991).Several severe inflammatory adverse effects have also been reported (Tsukahara et al., 2021): 5 % (7/152) myocarditis, 13 % (20/152) pleuritis, and 1 % interstitial nephritis (2/152).This 5 % myocarditis frequency is even higher than the 3 % reported in Australia (Ronaldson et al., 2015), which is thought to have the highest myocarditis frequency in the world.Furthermore, Japan accounts for 73 % (30/41) of reports of clozapine-induced myocarditis in Asian countries in VigiBase, the WHO database, until 2021.The risk of myocarditis severity was higher in Asian countries than non-Asian countries (De Las Cuevas et al., 2021).Another pharmacovigilance database study identified 51 clozapinerelated drug reaction with eosinophilia and systemic symptoms (DRESS) syndromes, of which Japan was reported to be first with 24 % (12/51) (de Filippis et al., 2022c).These reports suggest higher frequencies of clozapine-induced inflammatory adverse events in Japan.
Clozapine use in Japan has been extremely low compared to the rest of the world (Hata et al., 2020;Oloyede et al., 2022a), probably due to the late approval of clozapine and the stringent system of neutrophil monitoring (Oloyede et al., 2022a;Shimazawa et al., 2012).Additionally, the frequent inflammatory adverse effects in the Japanese population may be a reason for refraining from using clozapine.

Aim of this study
Based on past clinical experience (Kikuchi et al., 2022;Yasui-Furukori et al., 2023), in this study, we aimed to investigate the relationship between the rate of clozapine dose titration and inflammatory adverse events in Japanese individuals.We hypothesized that 1) clozapineinduced inflammatory adverse events would occur more often in Japanese people if the dose were increased according to the Japanese package insert; 2) slower clozapine dose titration would be associated with fewer clozapine-related inflammatory adverse events among Japanese individuals.To test this hypothesis, we retrospectively analyzed whether there was an association between clozapine dose titration and the risk of inflammatory adverse events in Japanese individuals.

Study design and study population
The medical records of all 272 patients with treatment-resistant schizophrenia who were first started on clozapine at seven hospitals between July 2009 and February 2023 were studied retrospectively (Tohoku University Hospital, Kodama Hospital, Miyagi Psychiatric Center, Hanamaki Hospital, Kunimidai Hospital, Aoba Hospital, and Asahi General Hospital).In Japan, patients are registered with Clozaril Patient Monitoring Service (CPMS) by a CPMS-certified psychiatrist at a CPMS-certified hospital according to the criteria for treatment-resistant schizophrenia as written in the package insert.All patients were inpatients because clozapine must be started during hospitalization in Japan.Considering that the study design was a retrospective study using anonymous data, an opt-out form was displayed on each hospital's bulletin boards or websites before collecting the data, and subjects who did not express the intent for exclusion were included in the study.The Tohoku University Hospital Ethics Review Board approved this study (Approval ID: 2022-1-1136).
The following data were collected from medical records: age at initiation of clozapine, sex, body mass index (BMI), smoking status, concomitant use of valproic acid, concomitant use of cytochrome P450 (CYP) 1A2 inhibitors (fluvoxamine, amiodarone, ciprofloxacin, oral contraceptives), clozapine start date, a daily dose of clozapine up to day 21 from the clozapine start date as day 1.In addition, the following information of inflammatory indicators was collected for 12 weeks after starting clozapine: presence or absence of fever, increased CRP, eosinophilia, along with the date of occurrence and duration of fever, and the maximum body temperature, the date of occurrence of increased CRP and the maximum CRP values, the date of occurrence of eosinophilia and the maximum eosinophil count.The reference values for CRP were almost identical in each hospital, with most hospitals using the common reference range [0-0.14 mg/dl] standardized by the academic societies.Detailed information on clozapine-related adverse events (symptoms, blood test data, diagnosis, etc.) was also collected.Fever was defined as an armpit temperature of 38 • C or higher.Low-grade fever was defined as ≥37 • C and <38 • C. Elevated CRP was defined as ≥0.3 mg/dl, and eosinophilia was defined as ≥500/μl.Inflammatory adverse events were defined as 1) fever and 2) low-grade fever and clinical symptoms requiring discontinuation or dose reduction of clozapine.

Abbreviations
According to the Japanese package insert, the study protocol is as follows (Fig. 1): 12.5 mg on day 1, 25 mg on days 2-4, 50 mg on days 5-7, 75 mg on days 8-9, 100 mg on days 10-11, 125 mg on days 12-14, 150 mg on days 15-17, 175 mg on days 18-20, and 200 mg on day 21.This titration rate was defined as the reference protocol rate for the following analyses (clozapine titration rate, CTR = 1).Relative to this, we calculated individual clozapine titration rates as follows: CTR = the cumulative clozapine dose within the first X days/the cumulative clozapine dose within the first X days according to the reference protocol.X was the day on which any inflammatory adverse event with clozapine occurred.X was 21 if no inflammatory adverse events occurred within 21 days of starting clozapine.
The recommended protocol for normal-metabolizing Asians in the international guideline issued by Jose de Leon's group (de Leon et al., 2022b) is approximately (Fig. 1): 12.5 mg on day 1, 25 mg on days 2-3, 37.5 mg on days 4-5, 50 mg on days 6-7, 62.5 mg on days 8-9, 75 mg on days 10-11, 87.5 mg on days 12-13, 100 mg on days 14-17, 125 mg on days 18-20, and 150 mg on day 21.This protocol yields a CTR of approximately 0.75.In addition, the guideline recommends not increasing the dose in the presence of poor tolerance or increased CRP, which results in slower dose titration.CTR calculated from the protocol recommended for Asians is about 0.4 (Fig. 1).The slower titration was set for Asians because of the prevalence of people with low drugmetabolizing enzyme activities among the ethnicities.The prevalence of low metabolizers among the Japanese population has been unknown but estimated to be <10 % (Ruan et al., 2019a).
The subjects were divided into two groups based on whether the dose titration was faster or slower than a CTR of 0.75, recommended for Asians in the de Leon et al.'s international guideline.We defined the former as the faster titration group (FG), the majority of which were the subjects treated according to the Japanese package insert.The latter was defined as the slower titration group (SG), including subjects treated according to the de Leon et al.'s guideline.All cases were scored using the criteria for the possible DRESS cases proposed by Kardaun (Kardaun et al., 2007).

Statistical analysis
All statistical analyses were performed using EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan) (Kanda, 2013).Demographic data and risk of adverse inflammatory events between FG and SG were analyzed using the chi-square test or Fisher exact test for categorical variables depending on the number of subjects.The titration group (FG or SG), age, sex, BMI, smoking status, and concomitant use of valproic acid were selected as variables influencing the risk of inflammatory adverse events because of clozapine and were evaluated by logistic regression analysis.These variables were selected based on the previous studies (de Leon et al., 2022a).Differences were considered statistically significant level if p ≤ 0.05.

Descriptive findings
A total of 272 all clozapine-initiated cases were identified at seven hospitals.The number of cases at each hospital was as follows: Kodama Hospital, 36 cases; Hanamaki Hospital, 80 cases; Asahi General Hospital, 105 cases; Kunimidai Hospital, 2 cases; Aoba Hospital, 2 cases; Tohoku University Hospital, 23 cases; and Miyagi Psychiatric Center, 24 cases (Supplementary Table 1).
At Kodama Hospital, Hanamaki Hospital, and Asahi General Hospital, titrations close to CTR = 1.0 were applied in the first 20-30 patients who started clozapine at each hospital, whereas titrations with a CTR of 0.5 or less became dominant in recent cases.At Kunimidai Hospital, Aoba Hospital, Tohoku University Hospital, and Miyagi Psychiatric Center, titrations with a CTR ≈ 1.0 were applied for most cases.In sum, approximately equal proportions of cases with CTR ≈ 1.0 (n = 110) and cases with CTR ≈ 0.5 or less (n = 131) were included from the participating hospitals.
There was a bias regarding smoking.While there were no smokers at Kodama Hospital, Kunimidai Hospital, Aoba Hospital, and Tohoku University Hospital, smokers were more prevalent at Hanamaki Hospital, where 57 % (16/28) of the FG patients were smokers (Supplementary Table 1).
There was no concomitant use of CYP1A2 inhibitors (fluvoxamine, amiodarone, ciprofloxacin, or oral contraceptives) in any case.
Of the 272 cases, 31 were excluded for the following reasons: eight refused drug intake or withdrew their consent, four discontinued due to physical complications thought not to be caused by clozapine-induced Y. Kikuchi et al. inflammation (one auto-sensitization dermatitis [a dermatologist ruled out drug-induced dermatitis], one coronary artery disease [severe threevessel disease was present prior to initiation of clozapine], one aspiration pneumonia [abnormal behavior of gargling continuously with his saliva at the time of clozapine initiation due to marked worsening of psychiatric symptoms], and one gouty arthritis [diagnosed 1.5 years prior to initiation of clozapine and continuous CRP evaluation]), two discontinued due to abnormal laboratory values occurred immediately after starting clozapine (one leukopenia and one increased creatine kinase), and 17 lacked essential information.The remaining 241 were subjected to the following analyses.

Comparison of demographic data between the groups
Of the 241 patients newly started on clozapine, there were 110 FG and 131 SG patients (Table 1).The two groups had significant differences in age, CTR, and smoking status.The mean age (SD) was 38.3 (11.8) years for FG and 43.3 (13.1) years for SG, with an older age for SG (p < 0.005).Furthermore, the mean CTR (SD) was 0.97 (0.10) in FG and 0.43 (0.15) in SG (p < 0.001), and the mean CTR in SG was less than half that in FG.The number of smokers was 21 (18 %) in FG and 11 (8.8 %) in SG, with a significantly higher number of smokers in FG (p < 0.05).

Comparison of frequency of inflammatory adverse effects between the groups
The frequency of inflammatory adverse events was 34 % (37/110) in FG and 13 % (17/131) in SG, and a significant difference was observed by Fisher exact test (OR 3.38; 95%CI 1.71-6.91;p < 0.001).In FG, 16 patients (14 %) had more severe adverse effects such as pneumonia, myocarditis, enteritis, renal damage, liver damage, rash, and ileus.However, in SG, 4 patients (3.2 %) exhibited more severe adverse effects such as myocarditis, pneumonia, and liver damage.When comparing the frequency of severe adverse effects, Fisher exact test showed a significant difference (OR 5.37; 95%CI 1.68-22.79;p < 0.005).Clozapine was discontinued in 11 patients (9.6 %) owing to inflammatory adverse events in FG and 4 (3.2 %) in SG.Fisher exact test found a significant difference (OR 3.51; 95%CI 1.00-15.6;p < 0.05) when the clozapine discontinuation rate was compared between the two groups.When comparing the cases with fever for five days or more, 21 (19 %) in FG and 7 (5.3 %) in SG.The Fisher exact test revealed a significant difference (OR 4.15; 95%CI 1.61-12.1;p < 0.005).There were no deaths in either group.When looking at the frequency of inflammatory effects by the hospital, contrary to the overall trend, inflammatory adverse effects were less frequent in the FG patients at Hanamaki Hospital (11 %, 3/28; Supplementary Table 1).
Next, a logistic regression analysis was performed on the risk of inflammatory adverse events (Table 2).A significant increase in adverse events was found in the FG (adjusted OR 4.01; 95 % CI 2.02-7.97;p < 0.001).Factors such as age, sex, BMI, concomitant use of valproate, and smoking were not significantly associated with the risk of inflammatory adverse events.Although not significant, there was a trend toward more inflammatory adverse effects with the valproic acid concomitant (OR 1.60; 95%CI 0.785-3.28;p = 0.195) and less inflammatory adverse effects with smoking (OR 0.45; 95%CI 0.156-1.29;p = 0.137).

CRP levels and eosinophil counts in patients with adverse inflammatory events
We analyzed peak CRP and eosinophil counts and their occurrence dates in the 54 patients with inflammatory adverse events (Fig. 2).CRP was measured in 53 patients and was found to be elevated in all the patients.The eosinophil count was measured in all 54 cases and was elevated in 27 of them (50 %).Incidentally, the eosinophil count was 1 (0.9) 0 (0) Enteritis and renal failure, n (%) 1 (0.9) 0 (0) Pneumonia and skin rash, n (%) 1 (0.9) 0 (0) Liver damage, n (%) 1 (0.9) 1 (0.8) Ileus, n (%) 1 (0.9) 0 (0)  Y. Kikuchi et al. measured in all 241 cases, and 26 cases showed eosinophilia, even in cases where no inflammatory adverse events were observed.Thus, eosinophilia was observed in 53 of 241 patients (22 %).Among patients with inflammatory adverse events, the mean peak CRP was 8.31 mg/dl, and the mean peak eosinophil count was 1431/μl.The mean day of peak CRP was 20.4 days, and the mean day of peak eosinophil count was 29.7 days.

Possible DRESS cases
Eight patients (3.3 %, 8/241) met the criteria for the possible DRESS cases proposed by Kardaun (Kardaun et al., 2007), as summarized in Supplementary Table 2.All eight cases were in the FG.Five of eight patients received concomitant valproic acid, which was considered a risk factor for clozapine-induced inflammation.Six of the eight cases had a fever.The maximum temperature was 39 • C or higher in all cases.Two cases without fever had a generalized skin rash and hypereosinophilia.

Seventeen cases of inflammatory adverse effects in the SG
Seventeen patients (7.1 %, 17/241) developed inflammatory adverse effects despite slow dose titration, as summarized in Supplementary Table 3.Of the 17 cases, 7 had concomitant valproic acid.No obvious risk factors were found in the remaining 10 cases.In focusing on the three cases of myocarditis and pneumonia, all had a maximum body temperature of 39 • C or higher, and their peak CRP values were 10 mg/ dl or higher.

Discussion
To the best of our knowledge, this study is the first to demonstrate the relationship between the rate of clozapine dose titration and inflammatory adverse events in Japanese.A slower dose increase than the one recommended in the Japanese package insert was associated with a lower risk of inflammatory adverse effects with clozapine.

Differences in prescribing characteristics among participating hospitals
The titration with CTR = 1.0 was initially applied in all seven hospitals because it is recommended in the Japanese package insert.However, in the three hospitals where clozapine was prescribed for more patients, the titration protocol was switched from CTR = 1.0 to CTR of 0.5 or less, based on the more occasion experiencing inflammatory adverse effects.In the remaining four hospitals with fewer occasions to prescribe clozapine, i.e., fewer occasions experiencing the adverse effect, the titration with CTR = 1.0 was applied throughout the observation period.

Differences in demographic data between the groups
Considering the demographic data for the FG and SG, both groups observed significant differences in age and smoking.On average, individuals in the SG were about 5 years older than individuals in the FG.As the dose increase was up to the physician's discretion, they were probably more cautious about increasing the dose in older patients.In addition, there were more smokers in the FG, probably because most SG subjects were recruited in recent years after smoking was banned in the hospitals, according to the enactment of the Health Promotion Law featuring measures against passive smoking.
In particular, a large proportion of the FG patients at Hanamaki Hospital were smokers and had few inflammatory adverse effects.Given that smoking increases the metabolism of clozapine, the slightly higher number of smokers in FG may have made inflammatory adverse events less likely to occur.

Inflammatory adverse effects with clozapine were less frequent with slower dose escalation
Inflammatory adverse events occurred in 34 % (37/110) of cases in the FG.The frequency of fever because of clozapine reported in two previous Japanese studies was 29 % (11/38) and 38 % (57/152), which were almost the same (Kishi et al., 2013;Tsukahara et al., 2021).On the other hand, inflammatory adverse events occurred in 13 % (17/131) of cases in the SG, which was less frequent than these reports.
More severe adverse effects such as pneumonia, renal failure, and enteritis were also more common in FG patients.A fever duration of 5 days or longer was more common in the cases of FG.Clozapine discontinuation rates were also higher in those in the FG.A more gradual clozapine dose increase may improve the frequency and duration of fever, the severity of inflammatory adverse events, and the clozapine continuation rate.
Logistic regression analysis adjusted for confounding factors such as age, sex, BMI, concomitant use of valproic acid, and smoking revealed that the odds ratio for the risk of inflammatory adverse events was approximately 4 times higher in the FG than in the SG.

Reports of the relationship between clozapine titration rate and inflammatory adverse effects
Three previous studies have examined the rate of clozapine dose titration and its association with inflammatory adverse effects (Table 3).First, Pui-yin Chung et al. retrospectively analyzed the medical records of 227 hospitalized patients who started clozapine in Hong Kong and revealed that 14 % (31/227) had a fever (Pui-yin Chung et al., 2008).They compared 31 febrile patients with 196 controls.Multivariate analysis showed a significant association with fever: OR 18.9 (95 % CI 5.3-66.7)for a dose increase rate of 50 mg/week or more, OR 3.6 (95 % CI 1.5-8.9)for concomitant use of valproic acid, and OR 3.2 (95 % CI 1.2-8.3)for physical illness.This study's results agree with those Pui-yin Chung reported regarding the dose escalation rate.In this study, a dose increase of 50 mg/week or more corresponded to the FG, and an average dose increase rate of 25 mg/week corresponded to the SG.Concerning concomitant valproic acid, this study did not find a significant association with the risk of inflammatory adverse events, which was inconsistent with the report by Pui-yin Chung.
Second, Ronaldson et al. reported a significant association between clozapine-induced myocarditis in Australia and rapid dose escalation (Ronaldson et al., 2012).The 250 mg cumulative clozapine dose increase during the first 9 days was significantly associated with the risk of myocarditis (OR 1.26; 95 % CI 1.02-1.55).Furthermore, concomitant Fig. 2. Relationship between peak CRP/eosinophil counts and number of days after clozapine initiation in patients with inflammatory adverse effects caused by clozapine.use of valproic acid was significantly associated with the risk of myocarditis (OR 2.59;.The protocol recommended in the Japanese package insert results in a cumulative clozapine dose of 387.5 mg over the first 9 days.The Australian protocol places this at 612.5-812.5 mg, with 920 mg and above in 15 % of cases.Thus, the official Japanese protocol sets a slower titration than the protocols for other ethnicities.However, the titration may still be faster than the one appropriate for the Japanese concerning risks for inflammatory adverse events. Third, Tsukahara et al. reported no significant association between the risk of fever and clozapine titration rate; they revealed the following results: 38 % (57/152) fever, 5 % (7/152) myocarditis, 13 % (20/152) pleuritis, and 1 % interstitial nephritis (2/152) (Tsukahara et al., 2021).Two-thirds of the patients in this study had a clozapine titration rate of 0.8 or higher.Thus, significant differences could not be detected due to the small number of cases corresponding to the SG in this study.Moreover, the calculation method for the clozapine titration rate was unclear-the cumulative clozapine dose up to day X (up to 21 days) was used in the calculation, but the determination of day X was not described.In our study, days with inflammatory adverse events were defined as X days, and X was defined as 21 if no inflammatory adverse events occurred.It is clinically typical for clozapine doses to be reduced or discontinued rather than increased when inflammatory adverse events occur.Therefore, if the day after the day on which the adverse event occurred, such as the 21st day, is set to Day X, the clozapine titration rate will decrease.In other words, patients with inflammatory adverse events tended to have their dose increased slowly.The primary endpoint of this study was the clozapine continuation rate after 18 weeks, and the incidence of fever was a secondary endpoint.The incidence of fever being a secondary endpoint is likely why no significant association between the risk of fever and clozapine titration rate was observed.▪ No significant association between the rate of clozapine dose titration and clozapine continuation for the first 18 weeks (the primary outcome).▪ No significant association between the rate of clozapine dose titration and fever occurrence within the first 4 weeks (the secondary outcome).
▪ The high frequency of inflammatory adverse effects with clozapine in Japanese patients was shown.
▪ Few cases of slow titration based on the international guideline.▪ Unclear how to calculate the rate of titration.

The present study
Japan 241 patients who were first started on clozapine were divided into two groups regarding whether the titration speed was faster or slower than the international guideline for Asians.The incidence of inflammatory adverse events with clozapine was compared between the groups.
▪ Clozapine-induced inflammatory adverse events were less frequent in Japanese individuals when a titration rate was more gradual than the protocol recommended in the Japanese package insert ▪ This study includes many cases of slow titration based on the international guideline.▪ The method of calculating the clozapine titration rate is clearly defined.
▪ Because of the inclusion of a wide range of cases with all fevers and associated clinical symptoms, there may be some cases that are not strictly related to clozapine.
Y. Kikuchi et al. 4.5.Are there any mechanisms by which inflammatory adverse effects are more likely to occur with clozapine in the Japanese population?
Although the mechanism by which clozapine causes inflammatory adverse effects is unknown, an allergic reaction or hypersensitivity reaction to clozapine may be involved, as the onset occurs within 4 weeks in most cases and is accompanied by eosinophilia ( de Leon, 2022).Like Stevens-Johnson syndrome with lamotrigine, rapid dose escalation can lead to inflammatory adverse effects, and slow dose escalation can promote desensitization (de Leon et al., 2015;Freudenreich, 2015).
Although the protocol recommended in the Japanese package insert is a gradual dose increase from a global perspective, the mechanism by which Japanese people are more likely to experience inflammatory adverse effects remains unknown.However, as highlighted in the international guideline by Jose de Leon et al., Asians, including Japanese, have a low ability to metabolize clozapine.Several studies have shown that East Asians have a lower capacity to metabolize clozapine than Caucasians (Pardinas et al., 2023;Reeves et al., 2023;Ruan et al., 2019b).Moreover, inflammatory adverse events may be more likely to occur because of the higher blood levels of clozapine (de Leon et al., 2022b).Assuming an immunological mechanism for allergic reactions, there is a possibility, although only hypothetical, that genetic factors (for example, human leukocyte antigen (HLA) polymorphisms) in the Japanese may cause hypersensitivity reactions to the drug.It has been noted that there is a possibility that the HLA alleles of the Japanese may differ from those of other East Asians ( de Leon, 2023).
The mean CTR (SD) in SG was 0.43 (0.15), corresponding to less than half the rate of increase in the official Japanese protocol, equivalent to a dose increase of approximately 25 mg/week.Inflammatory adverse events were less frequent than in the FG with this dose escalation rate.However, 13 % (17/131) of inflammatory adverse events still occurred in the SG, and 2 individuals developed myocarditis.The number of myocarditis cases in the FG was also 2, and there was no difference between the FG and SG.Further investigation is needed to determine whether a more gradual increase in dose would prevent inflammatory adverse events.Given allergenic mechanisms, even slow dose escalation may be insufficient to prevent inflammatory adverse events from occurring in some cases.However, 66 % (73/110) of individuals in the FG exhibited no inflammatory adverse events.This result suggests that more than half of the Japanese population will have no problems with the dosage escalation method recommended in the Japanese package insert.However, at this time, it is not possible to identify patients who will not experience inflammatory adverse events in advance.Therefore, slow dose escalation appears to be the best approach to administer clozapine to all Japanese patients safely.

Recommendations for preventing clozapine-induced inflammation in Japanese
Based on the results of this study and our recent clinical experience, we have summarized our recommendations in Box 1.

Future study
As future issues, it is desirable to elucidate the mechanism of inflammatory adverse events caused by clozapine and to develop a method to identify patients prone to adverse events in advance.It is necessary to find out how much pre-measurement of CRP can prevent the occurrence of inflammatory adverse events and when and how often it is appropriate to measure CRP.Moreover, it is essential to measure blood levels of clozapine in patients who have started clozapine to determine the relationship to inflammatory adverse effects.In addition, it may be necessary to conduct genetic analysis in patients with and without inflammatory adverse effects to determine whether the Japanese have a genetic background predisposing them to inflammatory adverse effects.

Limitation
This study had some limitations.First, the number of cases was small.Since the number of cases varied among the hospitals and the frequency of inflammatory adverse effects was also diverse, those biases may have affected the outcome of this study.The findings need to be validated with more extensive studies.Second, it was also a retrospective study, which may have biased physicians' choice of titration rate.For example, when the patients were deemed at risk for inflammatory adverse effects, their physicians may have opted for gradual titration, which may have increased the proportion of at-risk patients among the SG.Third, because the study included a wide range of cases with all fevers and Box 1 Recommendations for preventing clozapine-induced inflammation in Japanese.
1. Clozapine titration at 25 mg/week or less is safer for Japanese.
According to the Japanese package insert, the titration is 66 mg/week, equivalent to the FG in this study.In contrast, a 25 mg/week titration corresponds to the SG in this study.It is also equivalent to the titration recommended for Asians with low metabolism in the international guideline.In this study, inflammatory adverse effects were less frequent in the SG than in the FG; serious adverse effects including possible DRESS cases, fever for more than five days, and discontinuation of clozapine, were all less frequent in the SG. 2. Weekly measurement of CRP helps detect clozapine-induced inflammation before the onset of fever.
Only peak CRP data were available in this study because clinicians measured CRP after fever.However, according to the guideline, our current clinical practice acknowledges that there are cases where elevated CRP can be detected prior to fever by measuring CRP weekly along with neutrophil monitoring.As shown in Fig. 2, the eosinophil count rises later than CRP, so it is not helpful for early detection.3. Clozapine dosage should not be increased if CRP is elevated.
When CRP is elevated, increasing the dose of clozapine should be avoided because it increases the risk of inflammatory adverse effects.In this study, CRP tended to be higher in severe inflammatory adverse events.4. If the fever develops, clozapine should be discontinued until the fever resolves.
As shown in Supplementary Table 3, the SG patients also had 13 % inflammatory adverse events.These include severe cases of myocarditis and pneumonia.It is difficult to distinguish in advance between cases of fever that resolve spontaneously and severe cases.If a fever of 39 • C or higher or an elevated CRP of 10 mg/dl or higher is observed, it may often have a severe inflammatory adverse effect and should be treated cautiously.Discontinuation of the drug often results in rapid resolution of the fever, thus preventing the severity of inflammatory adverse effects.The safe approach is to slowly increase the dose again after the fever resolves.associated clinical symptoms, some cases may not be strictly related to the drug.Fourth, this study did not examine the relationship between blood levels of clozapine and adverse inflammatory events.Fifth, the four cases excluded due to physical complications were excluded because there was reason to believe they were unrelated to clozapine.However, a relationship may not be completely ruled out.Sixth, there were 17 cases excluded due to insufficient information, which we could not evaluate.Seventh, the method of administration of clozapine, such as once or multiple daily doses, was not investigated in this study.

Conclusion
Clozapine-induced inflammatory adverse events were less frequent in Japanese individuals when a titration rate was more gradual than the protocol recommended in the Japanese package insert.This research provides evidence to support the creation of the international clozapine dosage titration guideline specifically for Japanese patients, considering ethnic differences in response to the drug.As the slower titration protocol would widely be accepted, clozapine may be applied to more Japanese patients with schizophrenia without inflammatory adverse events.

Role of the funding source
None.

Declaration of competing interest
None.

Table 1
Comparison of faster titration group and slower titration group.

Table 2
Logistic regression analysis for the risk of inflammatory adverse events with clozapine.

Table 3
Reports of the relationship between clozapine titration rate and inflammatory adverse effects.