Population Pharmacokinetic-Pharmacodynamic Modeling of Haloperidol in Patients With Schizophrenia Using Positive and Negative Syndrome Rating Scale

The aim of this study was to develop a
pharmacokinetic-pharmacodynamic (PKPD) model that quantifies the
efficacy of haloperidol, accounting for the placebo effect, the
variability in exposure-response, and the dropouts. Subsequently,
the developed model was utilized to characterize an effective
dosing strategy for using haloperidol as a comparator drug in
future antipsychotic drug trials. The time course of plasma
haloperidol concentrations from 122 subjects and the Positive and
Negative Syndrome Scale (PANSS) scores from 473 subjects were used
in this analysis. A nonlinear mixed-effects modeling approach was
utilized to describe the time course of PK and PANSS scores.
Bootstrapping and simulation-based methods were used for the model
evaluation. A 2-compartment model adequately described the
haloperidol PK profiles. The Weibull and E-max models were able to
describe the time course of the placebo and the drug effects,
respectively. An exponential model was used to account for
dropouts. Joint modeling of the PKPD model with dropout model
indicated that the probability of patients dropping out is
associated with the observed high PANSS score. The model evaluation
results confirmed that the precision and accuracy of parameter
estimates are acceptable. Based on the PKPD analysis, the
recommended oral dose of haloperidol to achieve a 30% reduction in
PANSS score from baseline is 5.6 mg/d, and the corresponding
steady-state effective plasma haloperidol exposure is 2.7 ng/mL. In
conclusion, the developed model describes the time course of PANSS
scores adequately, and a recommendation of haloperidol dose was
derived for future antipsychotic drug trials.