Effectiveness of aripiprazole once-monthly in schizophrenia patients pretreated with oral aripiprazole: a 6-month, real-life non-interventional study

Schizophrenia is one of the most complex psychiatric disorders often leading to considerable disability and affecting the lives of patients and their families profoundly [1, 2]. Non- or partial adherence to psychopharmacological therapy is one of the major risk factors for relapses [3, 4], which severely diminish the psychosocial and occupational functioning of patients and negatively affect their quality of life [3, 5, 6]. Treatment with antipsychotics, as a major component in a framework of social and psychological therapies, can help to overcome these impairments and is highly effective [2]. However, nonadherence to treatment in patients with schizophrenia is frequently observed [4, 7, 8] with a significantly heightened risk of subsequent relapses [9]. Even small gaps in taking medication can have a significant effect, as stopping medication for as little as one to 10 days in a one-year period (partial adherence) was found to be associated with a significantly increased risk of hospitalization, with an odds ratio of 1.98 [10], while intermittent use of medication leads to a 3-fold higher risk of relapse in stable patients [11]. Improving adherence to medication may be achieved by using long-acting injectable antipsychotics (LAIs) which may in turn reduce the risk of relapse and improve patient functioning [12–14].

Aripiprazole once-monthly (AOM) is an atypical LAI with a unique pharmacological profile: aripiprazole shows partial agonist activity at dopamine D2/D3 receptors [15, 16], therefore the risk of adverse effects, such as parkinsonism, hyperprolactinemia and sexual dysfunction [17] is low. Effects on serotonin receptors include a partial agonist activity at the 5-HT_1A receptor [18] and antagonist activity at 5-HT_2A receptor [19], which adds to the antipsychotic profile. The efficacy and tolerability of AOM in patients with schizophrenia has been demonstrated by two randomized, double-blind, controlled studies (RCTs) conducted in the United States [20] and Europe [21]. In the study by Kane et al., 10% of patients in the active group and 39.6% in the placebo group experienced exacerbation of psychotic symptoms or impending relapse (hazard ratio = 5.03) at study endpoint. Altogether, 2.6% vs. 3.7% of patients were hospitalized because of relapse. In the European study, Kaplan–Meier estimated impending relapse rates at week 26 from randomization were 7.1% for AOM 400 mg, 7.8% for oral aripiprazole (10–30 mg/day) and 21.8% for aripiprazole once-monthly at a subtherapeutic dose (50 mg) [21]. AOM 400 mg has a long half-life of 46.5 days [22], which potentially can provide long-term relapse protection for patients.

In the present non-interventional study, the course of schizophrenia was monitored in patients receiving AOM treatment under usual care conditions. The primary endpoint was the assessment of AOM effectiveness on psychopathology by the attending psychiatrist, using the Brief Psychiatric Rating Scale (BPRS). Secondary endpoints included measurement of illness severity and illness improvement using the Clinical Global Impressions- Severity (CGI-S) and - Improvement (CGI-I) scales and the documentation of treatment related adverse events (TRAE). In addition, other instruments measuring functional status and wellbeing of the patients were applied such as the Global Assessment of Functioning (GAF), the WHO-5 scale, and medical resources used, which are part of a separate manuscript. The data collected in this study in a heterogeneous patient sample were intended to confirm the results of the previous efficacy studies [20, 21], which were conducted in a homogenous population of patients and which were intended for regulatory approval of AOM. Thus, we hypothesized that 6 months of AOM treatment would significantly improve psychopathology in patients receiving usual-care based AOM treatment.

The therapeutic effect of AOM treatment in schizophrenia was shown in previous RCTs, which were initiated to determine its efficacy [20, 21, 25]. Furthermore, the effect of oral aripiprazole was studied in short-term as well as long-term therapy for bipolar disorder [26]. AOM has also been approved by the FDA for the treatment of bipolar disorder [27] and a real-world study has been conducted on AOM use for both bipolar disorder and schizophrenia [28]. However, in Europe, AOM is only approved for schizophrenia treatment. Here, we report prospectively collected real-world data in a relatively large cohort in Germany. This prospective non-interventional study was conducted to support and amplify the effectiveness of AOM treatment in a naturalistic setting of usual-care patients being treated with oral aripiprazole before enrollment in the study.

The main findings were that during the 6-month observation period, patients experienced significant improvements in psychopathology and in severity of illness scale scores. The global BPRS values were significantly better already after 12 weeks (p < 0.001), with further significant improvement after 24 weeks (p < 0.001). This effect was numerically more prominent in younger patients ≤35 years.

In line with the results from the BPRS, CGI-S and CGI-I values also improved significantly during the study period. There were fewer patients with high scores and more patients with low scores at the end of the study, with significant differences between start and end of the study. This finding is in agreement with the improvements seen in the CGI-S values of patients in a previous study [25]. The improvements were significantly stronger in younger patients ≤35 years.

The results of numerically or statistically significant greater improvements in the younger age group further highlight the importance of early and consequent therapeutic intervention, which may help protect patients from the potential deteriorating effects of experiencing repeated psychotic episodes [29–31].

Some patients worsened during the study. Altogether, 29 patients (12%) achieved worse ratings on the BPRS, and 13 patients (5.4%) were considered worse on the CGI-S scale. In most cases, the deterioration was minimal and may have been the result of normal variation of the patient’s status. Seven patients worsened by 10 points on the BPRS and four patients worsened by two points on the CGI-S. However, other studies found similar rates of deterioration among their patients. [21] In the registration trial for AOM by Kane et al., 10% of patients in the active group experienced exacerbation of psychotic symptoms or impending relapse despite receiving ongoing AOM treatment [20]. In the European registration trial, Kaplan–Meier estimated impending relapse rates at week 26 from randomization were 7.1% for AOM 400 mg [21]. Finally, the proportion of patients on AOM or oral aripiprazole meeting exacerbation of psychotic symptoms/relapse criteria at the endpoint of a study by Ishigooka et al. was 6.6% [32]. Reasons for breakthrough psychosis may be psychosocial, comorbidity-related, pharmacokinetic/administration-related, or biological in nature [33], yet the exact reasons for the worsening in the patients in this study are unknown to us.

Among other variables, adherence to psychopharmacological treatment is considerably influenced by the occurrence of side effects of the medication [34]. Particularly side effects, such as weight gain, experiencing extrapyramidal symptoms, or side effects related to hyperprolactinemia, such as sexual dysfunction, are common causes for stopping medication [35]. In our study, these side effects were rare. According to our medication monitoring, 84% of the patients were fully adherent to medication, which supports the notion, that patients prefer an effective and well tolerable psychopharmacological treatment.

Interestingly, only few TRAEs were recorded during the study, which were not different in the age subgroup analyses. TRAEs related to extrapyramidal symptoms (2.9%), weight increase (0.4%) or hyperprolactinemia (0%) were very rare. Extrapyramidal symptoms were only found in patients > 35 years who were diagnosed with schizophrenia > 5 years ago.

Although LAIs can lead to improved adherence rates and potentially to improved long-term outcomes, there has been some debate about the effectiveness of LAIs in the long-term treatment of schizophrenia: the advantages of LAIs in comparison to continuous oral medication have been repeatedly questioned. However, comparison of different study types, such as RCTs [36], mirror image studies [12], and retrospective as well as prospective cohort studies [14] revealed that differences regarding the effectiveness of LAIs are likely dependent on how representative and severely ill the studies sample is [37, 38].

Studies suggested that therapeutic alliance is one of the most important factors affecting adherence [34]. Moreover, studies of collaborative care, which can increase the alliance between the treatment team and the patient, and which respects the patients and their autonomy, have led to significant improvements in adherence rates. We believe that focusing on therapeutic alliance via provision of concurrent psychosocial interventions is important to optimize outcomes, even while using long-acting injectable antipsychotics [39, 40].

In contrast to RCTs with a highly selected study population, both naturalistic study types in general showed better effectiveness. In our study, we could broaden and support findings from RCTs with AOM and show that comparable efficacy results could be obtained.

The observed statistically and clinically relevant improvements in psychopathology and severity of illness ratings are particularly noteworthy, as patients had already received an average of 9.7 months of oral aripiprazole, and 87.9% were deemed by the clinicians as “clinically stable” and for an average of 5.9 months. Thus, the additional improvements in a real-world sample are at least in some relevant part attributable to a change in formulation of the antipsychotic and the related more consistent and reliable delivery of the antipsychotic medication [13]. The fact that 89.3% of the patients decided to continue with AOM after the 6-month observational period further underscores the acceptability of this treatment option.

These findings are in line with the study comparing aripiprazole and paliperidone LAI by Naber et al. [25] where the CGI-S scores also improved continuously during treatment with AOM. In the RCT studies of AOM compared to placebo by [20, 32] the PANSS score remained stable for 52 weeks, and in the noninferiority study compared to oral aripiprazole by Fleischhacker et al., [41] the PANSS score was stable for 38 weeks after switching to treatment with AOM. Similarly, a study comparing risperidone LAI with oral antipsychotics for 1 year found that frequent medication switches led to less favorable outcomes [42], suggesting that antipsychotic medication stability might be an important factor for improved or stable outcomes in patients with schizophrenia.

Similar positive results for AOM treatment as in our study were found in another naturalistic study that used the Positive and Negative Syndrome Scale (PANSS) and CGI scales as endpoints [43]. In that study, highly significant improvements were found as early as 4 weeks after study start.

Finally, in the past, observational studies on the use of aripiprazole in patients with bipolar disorder also confirmed the efficacy results of RCTs [44], showing that aripiprazole was effective both in controlled experimental settings and in real world clinical practice settings.

Due to its naturalistic, non-interventional design, this study has several limitations. Importantly, there is a risk of a selection bias (e.g., patients were willing to take LAI medication), and of an expectation bias (knowledge of being on the LAI formulation of aripiprazole and of being in a study), there are no reference points other than baseline (patient’s condition at T0), and due to a lack of any randomization there are possible confounding factors, which cannot be identified or excluded. The dose of AOM varied during the study period, which can also be a confounding factor. Finally, all patients had already received and tolerated oral aripiprazole. Nevertheless, the initiation with oral aripiprazole before starting AOM is according to the package insert and clinical guidelines, and inclusion of patients who were selected under usual care conditions as being clinically eligible for AOM treatment increases the external validity and generalizability of the findings.

In general, most often, observational studies produce results that are similar to those of RCTs [45–47], although in some cases the magnitude of an observed effect is different [48–50]. If the observational studies are well-designed, their results do not seem to systematically overestimate treatment effects from RCTs [51, 52]. Therefore, observational studies are an important complement to RCTs [53] in order to evaluate the feasibility of a given treatment and generalizability of the findings to less restricted patient populations, both regarding the magnitude of efficacy and of safety signals.

Taken together, our results support previous results of the efficacy, tolerability and acceptability of AOM under routine clinical practice conditions. Outpatients who already respond to oral aripiprazole can further benefit from treatment with AOM, which proved to be effective under routine clinical conditions in a “real-life” sample of patients with schizophrenia. TRAEs were rare, a finding that is particularly important for patients who often undergo long-term antipsychotic treatment and who frequently have psychiatric and medical comorbidities and receive comedications, each of which are restricted in RCTs. The positive therapeutic effect was pronounced in younger patients, which underlines the need for early and continuous treatment.