This is the first study to identify predictors of continuation of treatment with asenapine by examining the patients’ background factors before the treatment. This study demonstrated that CP equivalent dose of ≥ 600 mg/day and disease duration of ≥ 25 years were predictors of continuation of treatment with asenapine. In addition, in the evaluation of the 52-week continuation rate with a combination of the two predictors, the continuation rate was the highest (52.5%) in patients with both predictors (CP equivalent dose of ≥ 600 mg/day and disease duration of ≥ 25 years), and low (34.7%) in patients with none of the predictors, indicating a significant difference between the two groups (Fig. 1).
Continuation of medication is one of the most important issues to prevent relapse and maintain remission. However, the 52-week asenapine continuation rate varies widely among studies. In 2010, the 52-week asenapine continuation rate was reported as 38.0%16). In a 52-week long-term administration study (P06125 study) conducted as an extension study of a placebo-controlled study of asenapine, the continuation rate was reported as 42.8%17). In contrast, a retrospective study reported it to be 19.0%18). In a phase III study of asenapine (P06238 study) conducted in patients with residual schizophrenia, polypharmacy, overdose, treatment-resistance, or elderly schizophrenia, the 52-week continuation rate was 50.3%, which was the highest value so far reported19). Interestingly, in the study, 71.3% of patients had a disease duration of at least 20 years, and 68.0% of patients received antipsychotics at a CP equivalent dose of ≥ 600 mg/day19). However, in the aforementioned P06125 study17), the percentage of patients whose duration of disease was 20 years or longer was 25.9%. No medication was administered before the start of the treatment with asenapine because a washout period was conducted according to the study design. From these previous reports, it was considered that the continuation rate of asenapine varied among studies depending on the study design and characteristics of the patient population enrolled.
Several antipsychotics have been investigated on the predictors of treatment continuation in patients with schizophrenia. The CP equivalent dose, which is a predictor found in the present study, was reported as a predictor of the continuation of brexpiprazole and clozapine. However, contrary to asenapine, prior treatment with high-dose antipsychotics was associated with treatment discontinuation20,21). Regarding the relationship between disease duration and continuation rate of antipsychotic drugs, studies of aripiprazole and brexpiprazole have reported that the risk of treatment discontinuation is higher in patients with a longer disease duration20,22), again showing conflicting results with the findings on asenapine.
Although it is difficult to explain biologically why these two factors contribute to the continuation of asenapine treatment, there could be two possible reasons: the uniqueness of the pharmacological action of asenapine and the impact of unique oral adverse drug reactions deriving from the route of administration. First, with regard to the unique pharmacological action, the binding affinity of asenapine for dopamine D2 receptors is similar to or greater than that of endogenous dopamine. However, unlike other drugs, asenapine has the unique pharmacological property of having a greater affinity for a variety of receptors involved in the pathology of schizophrenia23). Recently, accumulating evidence suggests that asenapine is highly efficacious for patients with treatment-resistant schizophrenia24,25) or dopamine supersensitivity psychosis26). The potential reason for this is the full antagonist activity against dopamine D2 and serotonin 2A receptors. In addition, pharmacological effects of asenapine such as calming effect associated with high affinity for α1A and histamine receptors, improvement effect on cognitive function and anxiety symptoms due to partial agonist effect on serotonin 1A receptors, and high safety derived from low affinity for muscarinic M1 receptors are listed. However, the effects of the combination of these pharmacological effects are unknown, and future studies are warranted. Second, aging is known to be associated with decreased oral sensory function and taste, especially bitter taste27,28). For example, it is reported that the sensory threshold of the tongue is greatly affected by aging, and the sensory function declines with aging29). It is also reported that taste and smell are strongly affected by polypharmacy30), suggesting that patients receiving multiple antipsychotics and other concomitant drugs for a long period could have a certain negative impact. In the future, it is necessary to accumulate evidence regarding the relationship between the pathological conditions and the changes in oral sensation and taste in patients with a long disease duration receiving high-dose antipsychotics.
In this study, asenapine-related adverse drug reactions were observed in 33.9% of patients, and the medication was generally well tolerated. In contrast, the proportion of patients who continued treatment for 52 weeks was 23.0% for those with adverse drug reactions and 49.7% for those without adverse drug reactions, reflecting the presence of patients who discontinued treatment due to tolerability issues (Table 2). The stratified analysis by the presence or absence of predictors demonstrated no marked difference in the type and incidence of adverse drug reactions between the groups. However, among adverse drug reactions with an incidence of 3% or higher, a significant between-group difference was noted in akathisia and somnolence using the chi-square test (Table 3).
A recent study reported that the risk of akathisia varies depending on the type and dose of antipsychotics31). The risk of akathisia with asenapine is classified as moderate, and it changes gradually and monotonically within the dose range of 5 to 20 mg/day, without significant dose effects31). In a study comparing the risk of somnolence among antipsychotics, asenapine was classified as low somnolence32). However, in this study, 76.9% of patients received concomitant antipsychotics. Therefore, the risk factors could include not only the pharmacological action of asenapine but also certain patient background factors. A cluster analysis of the phase III clinical study suggests that the pharmacological effects of asenapine, such as akathisia and somnolence, are more likely to cause adverse drug reactions in mild patients.33). In this study, there was a possibility that the proportion of patients with less severe schizophrenia was high in patients with none of the two predictors identified in this study (i.e., patients with lower doses of antipsychotics and shorter disease duration). Because the severity assessment was not performed in this study, future study is required on this issue.
Contrary to expectations, no significant difference was observed in oral hypoesthesia among groups (Table 3), which is an asenapine-related adverse drug reaction. The reason for this is unknown; however, it is suggested that background factors such as aging and high-dose antipsychotic therapy in the patients included in this study could have affected the results.
Limitation
Because this study used data from the post-marketing survey for safety evaluation, information on the assessment of the severity of schizophrenia is not included. Therefore, the relationship between the improvement of symptoms of schizophrenia and the continuation of medication is unknown. In addition, adherence is unknown because it was not studied. Despite limitations in controlling confounding factors such as drugs used to treat complications, the findings of this study using real-world data led to the identification of predictors of continuation of treatment with asenapine in clinical practice.