Short- and long-term studies of long-acting risperidone have suggested an optimal population benefit/risk ratio at doses of 25 mg up to 50 mg administered every 2 weeks, the dosing interval for which this product has been approved [4
]; however, there is clinical interest in the possibility of a longer injection schedule, and models derived from single-dose pharmacokinetic data at the 50-mg dose suggested that 50 mg every 4 weeks would result in average plasma concentrations similar to those with 25 mg given every 2 weeks, albeit with higher peak-to-trough fluctuations and lower trough levels (Janssen Pharmaceutica, Inc., data on file). To date, there are no published reports supporting alternative dosing intervals with long-acting injectable risperidone, and important clinical questions on the associated tolerability, safety, and efficacy associated with a longer injection interval remain unanswered. We explored these issues as well as the pharmacokinetics of once-monthly administration of long-acting risperidone 50 mg in a small pilot study of stable patients with schizophrenia or schizoaffective disorder. Since this is the first clinical study assessing a monthly dosing regimen, a pilot approach was chosen to minimize potential risk to a larger number of patients.
The relapse rate was relatively low (17.9%) in this 1-year study of patients clinically stable at study entry receiving once-monthly long-acting risperidone 50 mg; the Kaplan-Meier estimate of the risk of relapse at 1 year was 22.4%. No meaningful difference was observed in the rate of relapse between subjects with a recent hospitalization and those with a longer time since last hospitalization. While it is difficult to make comparisons across studies for many reasons, including differences in relapse criteria, this relapse incidence is similar to those found in US-based clinical trials examining long-acting injectable antipsychotics [16
]. A head-to-head comparison of risperidone and haloperidol found 1-year relapse rates were 25.4% for patients receiving risperidone 2–8 mg/d vs 39.9% for haloperidol 5–20 mg/d [1
]. One-year rates of relapse for ziprasidone were 43%, 35% and 36% for doses of 40, 80, and 160 mg/d, respectively [20
], whereas treatment with fluphenazine decanoate 12.5–100 mg/3 weeks was associated with a 1-year relapse rate of 28% [19
]. Other relapse data for long-acting risperidone are available from a randomized, double-blind, 52-week study of 2 doses of long-acting risperidone (25 or 50 mg) administered every 2 weeks. One-year relapse incidence was 21.6% and 14.9% for patients receiving the 25 mg and 50 mg doses, respectively [7
Although patients were symptomatically stable at baseline, additional efficacy analyses showed that significant improvements were noted on the PANSS and the CGI-S scores throughout the study, except at endpoint, demonstrating maintenance or improvement of symptom control. In this study, the mean change from baseline in the PANSS total score at endpoint was -3.7 (17.7); P
= .087. Significant improvements in PANSS total scores were observed at endpoint in 2 other long-term studies of long-acting risperidone [4
]. In a 1-year, open-label study that included 561 patients, the mean (± SE) change in PANSS total score at endpoint was -6.1 (0.7), P
< 0.01 vs baseline [4
]. In a randomized, double-blind, 52-week study that included 323 patients, the mean (± SD) change in PANSS total score was -4.9 (16.8), P
≤ .001 vs baseline [7
]. Baseline total PANSS scores in this study and the other 1-year studies [4
] were similar. The smaller numbers of patients in the current study may explain the differences in significance in PANSS total scores at endpoint. As relapse prevention and symptom control are key goals in the treatment of chronic schizophrenia, these results suggest that some symptomatically stable patients with schizophrenia or schizoaffective disorder might be safely managed with long-acting risperidone 50 mg once monthly; however, these data do not allow us to distinguish those patients for whom this dosing regimen may be sufficient versus those for whom it may be suboptimal, nor does this pilot design permit firm conclusions concerning comparative efficacy or tolerability of the alternative dosing regimen.
In addition to a relatively low relapse rate and maintenance of effect in most patients, other results of this pilot study suggested that long-acting risperidone 50 mg once monthly was well tolerated, with a safety profile similar to that reported for long-acting risperidone administered on a biweekly schedule [4
]. Movement disorder ratings were all low at baseline in this population and were unchanged or improved during the study. There were no clinically meaningful changes in mean laboratory values, vital signs, or ECG parameters. The mean body weight decreased modestly over the course of this study. In contrast, small mean body weight increases were noted in two other 1-year studies of long-acting risperidone in which risperidone was administered biweekly [4
]. Overall, this pilot data suggests that the monthly dosing regimen may not have greater liability regarding safety or tolerability compared with biweekly dosing.
Single-dose pharmacokinetic modeling with long-acting risperidone 50 mg once monthly predicted average plasma concentrations similar to those seen with 25 mg given every 2 weeks, but with higher peak-to-trough fluctuations and lower troughs, resulting in intermittent periods of low exposure (Janssen, L.P., data on file). Consistent with those data, the pharmacokinetic results showed more variability with once-monthly compared with biweekly dosing. Indeed, Cmin
values associated with the 50-mg monthly injections were approximately half of the steady-state values measured in another study of long-acting risperidone injection 25 mg administered biweekly [2
]. Further, the percentage fluctuation and the Cmax
ratios were approximately 3 times higher after monthly injections than after biweekly injections;[2
] however, the average exposure to risperidone plus 9-hydroxyrisperidone was comparable between once-monthly dosing and historical data from biweekly dosing. While previous reports have suggested that greater plasma level fluctuations may be associated with poorer tolerability [2
], this was not supported in the overall results from this pilot study.