One of the challenges in schizophrenia long-term trials is that clinical outcomes are often confounded by covert nonadherence to prescribed oral antipsychotics. This is a post hoc analysis (>2 years) of the symptoms and illness trajectory of patients treated with the long-acting injectable (LAI) antipsychotic aripiprazole lauroxil (AL). As adherence to LAIs can be monitored, these data could assess outcome trajectories unaffected by medication discontinuations that may occur with oral antipsychotics.

The efficacy and safety of once-monthly AL (441 or 882mg) for the treatment of schizophrenia were previously demonstrated in a phase 3 trial, followed by a 52-week, long-term safety study of two AL doses (441 or 882mg once monthly; patients continuing from the phase 3 study remained on their fixed AL dose [NCT01626456]), after which patients could enroll in a second long-term extension study. Patients entering the second long-term study continued on their fixed AL dose, with a variable follow-up period of up to 128 additional weeks (NCT01895452). In this post hoc analysis, the extension studies were combined to provide continuous outcome data over 2 years’ follow-up. The 12-week assessment visit (rather than the first visit) in the first extension study was chosen as the baseline to account for patients entering this study with variable AL exposure histories (with/without prior AL exposure). We report on the trajectory of symptoms and illness severity for >2 years (up to 112weeks) after the 12-week visit using the Positive and Negative Syndrome Scale (PANSS) total and Clinical Global Impression–Severity (CGI-S) scale scores. Course of illness was measured as the difference in PANSS and CGI-S scale scores within dose groups from baseline to end of follow-up, analyzed using MMRM.

Overall, 432/478 patients entering the initial 52-week study were included in the post hoc analysis. For the AL 441 and 882mg groups, respectively, baseline scores (mean±SD) were 59.91±16.25 and 56.27±12.89 (PANSS), and 2.99±0.97 and 2.79±0.79 (CGI-S scale). Approximately 49% of patients (211/432) remained for the entire 112-week follow-up. Over this period, the trajectory of PANSS scores improved significantly compared with baseline for both the 441 and 882mg groups, with changes from baseline (least squares mean±SE) of −5.46±0.92 (P<.0001) and −4.99±0.53 (P<.0001), respectively. CGI-S scale scores had similar improvement: changes from baseline of −0.32±0.07 (P<.0001) and −0.28±0.04 (P<.0001) for the AL 441 and 882mg groups, respectively. Overall, AL was well tolerated, with a safety profile over a 2-year follow-up that was consistent with the initial 52-week safety results.

This post hoc analysis demonstrates the safety and continued therapeutic efficacy of long-term treatment with AL in patients with schizophrenia. There were no apparent dose differences in the trajectory of symptom changes over the course of a 2-year follow-up.

Funding Acknowledgements: This study was funded by Alkermes, Inc.