Relationship between antipsychotic blood levels and treatment failure during the Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE) study

Abstract Objective Antipsychotic blood levels (ABLs) may help identify patients at risk for treatment failure. Reference ranges (RR) for plasma concentrations of ABLs that account for between-patient variability were developed for risperidone and olanzapine based on population pharmacokinetic models. The Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE) collected clinical outcomes and ABLs, allowing testing of the relationship of ABLs with outcomes. Methods ABLs from 694 patients who were randomized to olanzapine or risperidone were compared to the 80% RRs and were assessed as below or within/above the RR. Treatment failure was defined per any of these criteria: (1) emergency room visit for psychiatric reasons, (2) hospitalization for psychiatric reasons, (3) adverse event of completed suicide, suicidal ideation, or suicide attempt, (4) assaultive behavior, (5) arrested or jailed, (6) 2-point increase from baseline in Clinical Global Impression-Severity score, (7) 25% increase in Positive and Negative Syndrome Scale total score. Patients assessed with treatment failure within 100 days of drug concentration measurement were analyzed. Results Treatment failure occurred in 126 of 323 patients. The proportion of patients with ABLs below RR was 18.3% (59/323) compared to 10% expected in a fully adherent population. Among the 59 with ABLs below RR, 50.8% had treatment failure (compared to 36.4% for the 264 with ABLs within/above RR). The difference between groups was significant (odds ratio = 1.810; 95% CI = 1.025, 3.197; p  = 0.0408). Conclusions Analysis of CATIE data showed that ABLs within the context of RRs may identify patients with higher risk of relapse.