Model outputs were used to estimate threshold steady state blood concentration and occupancy required to elicit a clinically relevant response (>25% reduction in scores on delusions, hallucinations and agitation domains of the Neuropsychiatric Inventory) and extrapyramidal side-effects (Simpson Angus Scale scores >3). Average steady state blood levels were low (71+-30 ng/ml), and associated with high D2/3 occupancies (65+-8%, caudate; 67+-11%, thalamus; 52+-11%, putamen). Antipsychotic clinical response occurred at a threshold concentration of 20ng/ml and D2/3 occupancies of 43% (caudate), 25%(putamen), 43% (thalamus). Extrapyramidal side effects (n=7) emerged at a threshold 4 concentration of 60ng/ml, and D2/3 occupancies of 61% (caudate), 49% (putamen) and 69%(thalamus). This study has established that, as in schizophrenia, there is a therapeutic window of D2/3 receptor occupancy for optimal treatment of psychosis in Alzheimer's disease. We have also shown that occupancies within and beyond this window are achieved at very low amisulpride doses in Alzheimer's disease due to higher than anticipated occupancies for a given blood drug concentration. Our findings support a central pharmacokinetic contribution to antipsychotic sensitivity in Alzheimer's disease and implicate the blood brain barrier, which controls central drug access. Whether high D2/3 receptor occupancies are primarily accounted for by age-or disease-specific blood brain barrier disruption is unclear, and this is an important future area of future investigation, as it has implications beyond antipsychotic prescribing.5

These findings implicate the ventral visual stream in psychosis in AD, consistent with integrative theories regarding origins of psychosis, and provide further evidence for a role in the misidentification subtype. Specifically, reduced volume in the parahippocampal gyrus is implicated in misidentification delusion formation, which we hypothesise is due to its role in context attribution. Copyright © 2017 John Wiley & Sons, Ltd.

Introduction Current prescribing guidelines for the antipsychotic amisulpride are based largely on pharmacokinetic (PK) studies in young adults, and there is a relative absence of data on older patients, who are at greatest risk of developing adverse events.Methods This study aimed to develop a population PK model for amisulpride specifically in older people, by combining data from a richly sampled phase 1, single (50 mg) dose study in healthy older people (n = 20, 65–79 years), with a clinical dataset obtained during off label, low-dose (25–75 mg daily) amisulpride prescribing in older people with Alzheimer’s disease (AD) (n = 25, 69–92 years), as part of an observational study.Results After introducing a scaling factor based on body weight, age accounted for 20 % of the inter-individual variability in drug clearance (CL), resulting in a 54 % difference in CL between those aged 65 and those aged 85 years, and higher blood concentrations in older patients.Discussion These findings argue for the consideration of age and weight-based dose stratification to optimise amisulpride prescribing in older people, particularly in those aged 85 years and above.Electronic supplementary materialThe online version of this article (doi:10.1007/s00213-016-4379-6) contains supplementary material, which is available to authorized users.

Despite the small sample size, our findings are highly relevant as they suggest that, as in AD, 50 mg/day amisulpride is associated with >40% occupancy and clinically relevant responses in VLOSLP. It was not possible to fully characterise concentration-occupancy relationships in VLOSLP, and it is thus unclear whether the greater susceptibility of those with AD to emergent EPS was accounted for by increased central drug access. Further investigation of age- and diagnosis-specific threshold sensitivities is warranted, to guide amisulpride prescribing in older people, and therapeutic drug monitoring studies offer a potentially informative future approach. Copyright © 2017 John Wiley & Sons, Ltd.