Socio-economic position as a moderator of cardiometabolic outcomes in patients receiving psychotropic treatment associated with weight gain: results from a prospective 12-month inception cohort study and a large population-based cohort.

Weight gain and metabolic complications are major adverse effects of many psychotropic drugs. We aimed to understand how socio-economic status (SES), defined as the Swiss socio-economic position (SSEP), is associated with cardiometabolic parameters after initiation of psychotropic medications known to induce weight gain. Cardiometabolic parameters were collected in two Swiss cohorts following the prescription of psychotropic medications. The SSEP integrated neighborhood-based income, education, occupation, and housing condition. The results were then validated in an independent replication sample (UKBiobank), using educational attainment (EA) as a proxy for SES. Adult patients with a low SSEP had a higher risk of developing metabolic syndrome over one year versus patients with a high SSEP (Hazard ratio (95% CI) = 3.1 (1.5-6.5), n = 366). During the first 6 months of follow-up, a significant negative association between SSEP and body mass index (BMI), weight change, and waist circumference change was observed (25 ≤ age < 65, n = 526), which was particularly important in adults receiving medications with the highest risk of weight gain, with a BMI difference of 0.86 kg/m2 between patients with low versus high SSEP (95% CI: 0.03-1.70, n = 99). Eventually, a causal effect of EA on BMI was revealed using Mendelian randomization in the UKBiobank, which was notably strong in high-risk medication users (beta: -0.47 SD EA per 1 SD BMI; 95% CI: -0.46 to -0.27, n = 11,314). An additional aspect of personalized medicine was highlighted, suggesting the patients' SES represents a significant risk factor. Particular attention should be paid to patients with low SES when initiating high cardiometabolic risk psychotropic medications.

Valproate is associated with early decrease of high-density lipoprotein cholesterol levels in the psychiatric population.

Few studies have evaluated the influence of valproate on the deterioration of the lipid profile in psychiatric patients. This observational study aimed to compare the evolution of metabolic parameters in a sample of adult patients starting valproate (n = 39) with a control group (n = 39) of patients starting aripiprazole, a drug associated with a low risk of metabolic deterioration. Data were obtained from a prospective study including psychiatric patients with metabolic parameters monitored during the first year of treatment. During the first month of treatment with valproate (median: 31 days [IQR: 25-36]), mean body mass index increased significantly (from 24.8 kg/m2 at baseline to 25.2 kg/m2 after one month; P = .03) and mean HDL-C levels decreased significantly (from 1.39 mmol/L to 1.27 mmol/L; P = .02). In comparison, these metabolic variables remained stable during the first month of treatment with aripiprazole. The proportion of patients with early (ie during the first month of treatment) HDL-C decrease of ≥ 5% was significantly higher under valproate (54%) than aripiprazole (15%) treatment (P < .001). These findings remind the importance of a prospective metabolic monitoring in patients who initiate valproate treatment. Further research should be conducted on larger samples and should focus on finding effective interventions to prevent such metabolic adverse effects.

Psychological trauma occurring during adolescence is associated with an increased risk of greater waist circumference in Early Psychosis patients treated with psychotropic medication.

BACKGROUND: It has been suggested that exposure to Childhood Trauma [CT] may play a role in the risk of obesity in Early Psychosis [EP] patients; however, whether this is independently of age at exposure to CT and the medication profile has yet to be investigated. METHODS: 113 EP-patients aged 18-35 were recruited from the Treatment and Early Intervention in Psychosis Program [TIPP-Lausanne]. Body Mass Index [BMI], Weight Gain [WG] and Waist Circumference [WC] were measured prospectively at baseline and after 1, 2, 3, 6 and 12 months of weight gain inducing psychotropic treatment. Patients were classified as Early-Trauma and Late-Trauma if the exposure had occurred before age 12 or between ages 12 and 16 respectively. Generalized Linear Mixed-Models were adjusted for age, sex, socioeconomic status, baseline BMI, medication and for diagnosis of depression. RESULTS: Late-Trauma patients, when compared to Non-Trauma patients showed greater WCs during the follow-up [p = 0.013]. No differences were found in any of the other follow-up measures. CONCLUSIONS: Exposition to CT during adolescence in EP-patients treated with psychotropic medication is associated with greater WC during the early phase of the disease. Further investigation exploring mechanisms underlying the interactions between peripubertal stress, corticoids responsiveness and a subsequent increase of abdominal adiposity is warranted.

Evaluation of Cardiometabolic Risk in a Large Psychiatric Cohort and Comparison With a Population-Based Sample in Switzerland.

BACKGROUND: Psychiatric patients are known to be at high risk of developing cardiovascular diseases (CVDs), leading to an increased mortality rate. OBJECTIVE: To assess the CVD risk (presence of metabolic syndrome [MetS] and calculated 10-year CVD risk) in a Swiss psychiatric cohort taking weight gain-inducing psychotropic drugs, compare the findings to a Swiss population-based cohort, and evaluate the prevalence of participants treated for metabolic disruptions in both cohorts. METHODS: Data for 1,216 psychiatric patients (of whom 634 were aged 35-75 years) were obtained between 2007 and 2017 from a study with metabolic parameters monitored during psychotropic treatment and between 2003 and 2006 for 6,733 participants from the population-based CoLaus|PsyCoLaus study. RESULTS: MetS as defined by the International Diabetes Federation (IDF) was identified in 33% of the psychiatric participants and 24.7% of the population-based subjects. Specifically, prevalence per the IDF definition was more than 3 times higher in the psychiatric cohort among women aged 35 to 49 years (25.6% vs 8.0%; P < 10-4). The psychiatric and population-based cohorts, respectively, had comparable predicted CVD risk (10-year risk of CVD event > 20%: 0% vs 0.1% in women and 0.3% vs 1.8% [P = .01] in men; 10-year risk of CVD death > 5%: 8.5% vs 8.4% [P = .58] in women and 13.4% vs 16.6% [P = .42] in men). No difference was observed among the proportion of participants with MetS treated for metabolic disturbances in the two cohorts, with the exception of women aged 35-49 years, for whom those in the psychiatric cohort were half as likely to receive treatment compared to participants in CoLaus|PsyCoLaus (17.8% vs 38.8% per the IDF definition; P = .0004). CONCLUSIONS: These findings emphasize the concern that psychiatric patients present an altered metabolic profile and that they do not receive adequate treatment for metabolic disruptions. Presence of metabolic disturbances should be routinely assessed, and adequate follow-up is needed to intervene early after illness onset.

Amisulpride: Real-World Evidence of Dose Adaptation and Effect on Prolactin Concentrations and Body Weight Gain by Pharmacokinetic/Pharmacodynamic Analyses.

BACKGROUND: Amisulpride is an antipsychotic used in a wide range of doses. One of the major adverse events of amisulpride is hyperprolactinemia, and the drug might also induce body weight gain. OBJECTIVE: The aims of this work were to characterize the pharmacokinetics of amisulpride in order to suggest optimal dosage regimens to achieve the reference range of trough concentrations at steady-state (Cmin,ss) and to describe the relationship between drug pharmacokinetics and prolactin and body weight data. METHODS: The influence of clinical and genetic characteristics on amisulpride pharmacokinetics was quantified using a population approach. The final model was used to simulate Cmin,ss under several dosage regimens, and was combined with a direct Emax model to describe the prolactin data. The effect of model-based average amisulpride concentrations over 24 h (Cav) on weight was estimated using a linear model. RESULTS: A one-compartment model with first-order absorption and elimination best fitted the 513 concentrations provided by 242 patients. Amisulpride clearance significantly decreased with age and increased with lean body weight (LBW). Cmin,ss was higher than the reference range in 65% of the patients aged 60 years receiving 400 mg twice daily, and in 82% of the patients aged > 75 years with a LBW of 30 kg receiving 200 mg twice daily. The pharmacokinetic/pharmacodynamic model included 101 prolactin measurements from 68 patients. The Emax parameter was 53% lower in males compared with females. Model-predicted prolactin levels were above the normal values for Cmin,ss within the reference range. Weight gain did not depend on Cav. CONCLUSIONS: Amisulpride treatment might be optimized when considering age and body weight. Hyperprolactinemia and weight gain do not depend on amisulpride concentrations. Modification of the amisulpride dosage regimen is not appropriate to reduce prolactin concentrations and alternative treatment should be considered.

Psychotropic drug-induced genetic-epigenetic modulation of CRTC1 gene is associated with early weight gain in a prospective study of psychiatric patients.

BACKGROUND: Metabolic side effects induced by psychotropic drugs represent a major health issue in psychiatry. CREB-regulated transcription coactivator 1 (CRTC1) gene plays a major role in the regulation of energy homeostasis and epigenetic mechanisms may explain its association with obesity features previously described in psychiatric patients. This prospective study included 78 patients receiving psychotropic drugs that induce metabolic disturbances, with weight and other metabolic parameters monitored regularly. Methylation levels in 76 CRTC1 probes were assessed before and after 1 month of psychotropic treatment in blood samples. RESULTS: Significant methylation changes were observed in three CRTC1 CpG sites (i.e., cg07015183, cg12034943, and cg 17006757) in patients with early and important weight gain (i.e., equal or higher than 5% after 1 month; FDR p value = 0.02). Multivariable models showed that methylation decrease in cg12034943 was more important in patients with early weight gain (≥ 5%) than in those who did not gain weight (p = 0.01). Further analyses combining genetic and methylation data showed that cg12034943 was significantly associated with early weight gain in patients carrying the G allele of rs4808844A>G (p = 0.03), a SNP associated with this methylation site (p = 0.03). CONCLUSIONS: These findings give new insights on psychotropic-induced weight gain and underline the need of future larger prospective epigenetic studies to better understand the complex pathways involved in psychotropic-induced metabolic side effects.

Escitalopram population pharmacokinetics in people living with human immunodeficiency virus and in the psychiatric population: Drug-drug interactions and probability of target attainment.

AIMS: The aims of this study were to characterize escitalopram pharmacokinetic profile, to identify factors influencing drug exposure, notably drug-drug interactions with antiretrovirals, and to simulate expected exposure under standard dosage regimen. METHODS: A population pharmacokinetic analysis was performed using NONMEM. A total of 159 plasma concentration measurements were obtained from 39 human immunodeficiency virus (HIV)-infected and 71 uninfected psychiatric patients. The influence of age, weight, sex, HIV and psychiatric cohorts, racemic citalopram treatment, and comedications on oral clearance was examined. Simulations served to calculate the percentage of patients expected to be under- or over-exposed, considering established therapeutic targets (15-80 ng/mL). RESULTS: A 1-compartment model with first-order absorption and elimination described the data adequately. The average escitalopram clearance and volume of distribution were 23.1 L/h (interindividual variability 51%), and 920 L, respectively. Escitalopram disposition did not differ between HIV-infected and uninfected patients, and was not affected by antiretroviral treatments. Coadministration of at least 1 proton-pump inhibitor (CYP2C19 inhibitor) modestly influenced escitalopram elimination (clearance decreased by 19%), with limited clinical relevance. Model-based simulations showed that, under a standard regimen of 10 mg once daily, a significant proportion of patients (56%) might be under-exposed. CONCLUSION: The variability in escitalopram disposition is large and poorly explained by demographic, clinical and environmental covariates, thus suggesting a role for dosage individualization based on therapeutic drug monitoring in case of poor clinical response. Escitalopram disposition is modestly impacted by comedications and therefore no a priori dosage adjustments are needed in patients receiving antiretroviral treatments, including boosted regimens.

Lipid Disturbances in Adolescents Treated With Second-Generation Antipsychotics: Clinical Determinants of Plasma Lipid Worsening and New-Onset Hypercholesterolemia.

OBJECTIVE: Lipid disturbances following treatment with second-generation antipsychotics (SGAs) represent a major health concern. A previous study determined that early changes of plasma lipid levels ≥ 5% during the first month of treatment with SGAs predicts further lipid worsening and development of dyslipidemia. This current study aimed to determine the proportion of adolescents with early lipid changes ≥ 5% and who develop dyslipidemia during SGA treatment. METHODS: Data were obtained from a 1-year longitudinal study ongoing since 2007 including 53 adolescent psychiatric (ICD-10) patients (median age 16.5 years; interquartile range [IQR], 14.8-17.5 years) whose metabolic parameters were monitored prospectively during treatment. Plasma lipid levels (total, low-density lipoprotein, high-density lipoprotein [HDL-C], and non-high-density lipoprotein cholesterol and fasting triglycerides ) were measured at baseline and after 1, 3, and/or 12 months of SGA treatment. RESULTS: Half (n = 26; 49%) the adolescents had an early increase of total cholesterol levels by 5% or more during the first month of treatment, and one-third (n = 8/24; 33%) developed new-onset hypercholesterolemia during the first year of treatment. Hypercholesterolemia developed more frequently in female patients (P = .01) and in patients with an early increase of total cholesterol ≥ 5% (P = .02). Finally, patients whose HDL-C levels decreased by ≥ 5% during the first month of treatment had a larger HDL-C worsening after 3 months of treatment as compared with patients with early decrease of HDL-C by < 5% (P = .02). CONCLUSIONS: This study underlines the importance of prospectively monitoring metabolic parameters in adolescents after the introduction of SGAs.

Influence of body weight and UGT2B7 polymorphism on varenicline exposure in a cohort of smokers from the general population.

PURPOSE: The abstinence rate to tobacco after varenicline treatment is moderate and might be partially affected by variability in varenicline concentrations. This study aimed at characterizing the sources of variability in varenicline pharmacokinetics and to relate varenicline exposure to abstinence. METHODS: The population pharmacokinetic analysis (NONMEM®) included 121 varenicline concentrations from 82 individuals and tested the influence of genetic and non-genetic characteristics on apparent clearance (CL/F) and volume of distribution (V/F). Model-based average concentrations over 24 h (Cav) were used to test the impact of varenicline exposure on the input rate (Kin) expressed as a function of the number of cigarettes per day in a turnover model of 373 expired carbon monoxide levels. RESULTS: A one-compartment model with first-order absorption and elimination appropriately described varenicline concentrations. CL/F was 8.5 L/h (coefficient of variation, 26%), V/F was 228 L, and the absorption rate (ka) was fixed to 0.98 h-1. CL/F increased by 46% in 100-kg individuals compared to 60-kg individuals and was found to be 21% higher in UGT2B7 rs7439366 TT individuals. These covariates explained 14% and 9% of the interindividual variability in CL/F, respectively. No influence of varenicline Cav was found on Kin in addition to the number of cigarettes. CONCLUSIONS: Body weight mostly and to a smaller extent genetic polymorphisms of UGT2B7 can influence varenicline exposure. Dose adjustment based on body weight and, if available, on UGT2B7 genotype might be useful to improve clinical efficacy and tolerability of varenicline.

Association Between Plasma Caffeine and Other Methylxanthines and Metabolic Parameters in a Psychiatric Population Treated With Psychotropic Drugs Inducing Metabolic Disturbances.

Importance: Multiple studies conducted in the general population identified an association between self-reported coffee consumption and plasma lipid levels. To date, no study assessed whether and which plasma methylxanthines (caffeine and/or its metabolites, i.e., paraxanthine, theophylline, and theobromine) are associated with plasma lipids. In psychiatric patients, an important coffee consumption is often reported and many psychotropic drugs can induce a rapid and substantial increase of plasma lipid levels. Objective: To determine whether plasma methylxanthines are associated with metabolic parameters in psychiatric patients receiving treatments known to induce metabolic disturbances. Design, Setting, and Participants: Data were obtained from a prospective study including 630 patients with metabolic parameters [i.e., body mass index (BMI), total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), non-high-density lipoprotein cholesterol (non-HDL-C), and fasting triglycerides (TG)] monitored routinely during psychotropic treatment. Exposures: Plasma methylxanthines levels. Main Outcomes and Measures: Metabolic variables including BMI and plasma lipid levels. Results: Multivariate analyses indicated that BMI, TC, HDL-C, and non-HDL-C increased significantly with increasing total methylxanthines (p corrected ≤ 0.05). In addition, compared to patients with plasma caffeine concentration in the lowest quartile, those with caffeine concentration in the highest quartile were twice more prone to suffer from non-HDL hypercholesterolemia (p corrected = 0.05), five times more likely to suffer from hypertriglyceridemia (p corrected = 0.01) and four times more susceptible to be overweight (p corrected = 0.01). Conclusions and Relevance: This study showed that plasma caffeine and other methylxanthines are associated with worsening of metabolic parameters in patients receiving psychotropic treatments known to induce metabolic disturbances. It emphasizes that important caffeine consumption could be considered as an additional environmental risk factor for metabolic worsening in patients receiving such treatments.

Early changes of blood lipid levels during psychotropic drug treatment as predictors of long-term lipid changes and of new onset dyslipidemia.

BACKGROUND: Cardiovascular diseases and dyslipidemia represent a major health issue in psychiatry. Many psychotropic drugs can induce a rapid and substantial increase of blood lipid levels. OBJECTIVE: This study aimed to determine the potential predictive power of an early change of blood lipid levels during psychotropic treatment on long-term change and on dyslipidemia development. METHODS: Data were obtained from a prospective study including 181 psychiatric patients with metabolic parameters monitored during the first year of treatment and with adherence ascertained. Blood lipid levels (ie, total cholesterol [TC], low-density lipoprotein cholesterol [LDL-C], high-density lipoprotein cholesterol [HDL-C], non-high-density lipoprotein cholesterol [non-HDL-C], and fasting triglycerides [TGs]) were measured at baseline and after 1, 3, and/or 12 months of treatment. RESULTS: Receiver-operating characteristic analyses indicated that early (ie, after 1 month of psychotropic treatment) increases (≥5%) for TC, LDL-C, TG, and non-HDL-C and decrease (≥5%) for HDL-C were the best predictors for clinically relevant modifications of blood lipid levels after 3 months of treatment (≥30% TC, ≥40% LDL-C, ≥45% TG, ≥55% non-HDL-C increase, and ≥20% HDL-C decrease; sensitivity 70%-100%, specificity 53%-72%). Predictive powers of these models were confirmed by fitting longitudinal multivariate models in the same cohort (P ≤ .03) as well as in a replication cohort (n = 79; P ≤ .003). Survival models showed significantly higher incidences of new onset dyslipidemia (TC, LDL-C, and non-HDL-C hypercholesterolemia, HDL-C hypocholesterolemia, and hypertriglyceridemia) for patients with early changes of blood lipid levels compared to others (P ≤ .01). CONCLUSION: Early modifications of blood lipid levels following prescription of psychotropic drugs inducing dyslipidemia should therefore raise questions on clinical strategies to control long-term dyslipidemia.

Association of nicotine metabolism and sex with relapse following varenicline and nicotine replacement therapy.

Nicotine is metabolized into cotinine and then into trans-3'-hydroxycotinine, mainly by cytochrome P450 2A6. Recent studies reported better effectiveness of varenicline in women and in nicotine normal metabolizers phenotypically determined by nicotine-metabolite ratio. Our objective was to study the influence of nicotine-metabolite ratio, CYP2A6 genotype and sex on the response to nicotine replacement therapy and varenicline. Data were extracted from a longitudinal study which included smokers participating in a smoking cessation program. Response to treatment was defined by the absence of relapse when a set threshold of reduction in cigarettes per day relative to the week before the study was no more reached. The analysis considered total and partial reduction defined by a diminution of 100% and of 90% in cigarettes per day, respectively. The hazard ratio of relapsing was estimated in multivariate Cox regression models including the sex and the nicotine metabolism determined by the phenotype or by CYP2A6 genotyping (rs1801272 and rs28399433). In the normal metabolizers determined by phenotyping and in women, the hazard ratio for relapsing was significantly lower with varenicline for a partial decrease (HR = 0.33, 95% CI [0.12, 0.89] and HR = 0.20, 95% CI [0.04, 0.91], respectively) and nonsignificantly lower for a total cessation (HR = 0.45, 95% CI [0.20, 1.0] and HR = 0.38, 95% CI [0.14, 1.0]). When compared with the normal metabolizers determined by phenotyping, the hazard ratio for a partial decrease was similar in the normal metabolizers determined by genotyping (HR = 0.42, 95% CI [0.18, 0.94]) while it was significantly lower with varenicline for a total cessation (HR = 0.50, 95% CI [0.26, 0.98]). Women and normal nicotine metabolizers may benefit more from varenicline over nicotine replacement therapy. (PsycINFO Database Record

Influence of polygenic risk scores on lipid levels and dyslipidemia in a psychiatric population receiving weight gain-inducing psychotropic drugs.

OBJECTIVES: Dyslipidemia represents a major health issue in psychiatry. We determined whether weighted polygenic risk scores (wPRSs) combining multiple single-nucleotide polymorphisms (SNPs) associated with lipid levels in the general population are associated with lipid levels [high-density lipoprotein (HDL), low-density lipoprotein (LDL), total cholesterol (TC), and triglycerides] and/or dyslipidemia in patients receiving weight gain-inducing psychotropic drugs. We also determined whether genetics improve the predictive power of dyslipidemia. PATIENTS AND METHODS: The influence of wPRS on lipid levels was firstly assessed in a discovery psychiatric sample (n=332) and was then tested for replication in an independent psychiatric sample (n=140). The contribution of genetic markers to predict dyslipidemia was evaluated in the combined psychiatric sample. RESULTS: wPRSs were significantly associated with the four lipid traits in the discovery (P≤0.02) and in the replication sample (P≤0.03). Patients whose wPRS was higher than the median wPRS had significantly higher LDL, TC, and triglyceride levels (0.20, 0.32 and 0.26 mmol/l, respectively; P≤0.004) and significantly lower HDL levels (0.13 mmol/l; P<0.0001) compared with others. Adding wPRS to clinical data significantly improved dyslipidemia prediction of HDL (P=0.03) and a trend for improvement was observed for the prediction of TC dyslipidemia (P=0.08). CONCLUSION: Population-based wPRSs have thus significant effects on lipid levels in the psychiatric population. As genetics improved the predictive power of dyslipidemia development, only 24 patients need to be genotyped to prevent the development of one case of HDL hypocholesterolemia. If confirmed by further prospective investigations, the present results could be used for individualizing psychotropic treatment.

Association of variants in SH2B1 and RABEP1 with worsening of low-density lipoprotein and glucose parameters in patients treated with psychotropic drugs.

Genetic factors associated with Body Mass Index (BMI) have been widely studied over the last decade. We examined whether genetic variants previously associated with BMI in the general population are associated with cardiometabolic parameter worsening in the psychiatric population receiving psychotropic drugs, a high-risk group for metabolic disturbances. Classification And Regression Trees (CARTs) were used as a tool capable of describing hierarchical associations, to pinpoint genetic variants best predicting worsening of cardiometabolic parameters (i.e total, HDL and LDL-cholesterol, triglycerides, body mass index, waist circumference, fasting glucose, and blood pressure) following prescription of psychotropic drugs inducing weight gain in a discovery sample of 357 Caucasian patients. Significant findings were tested for replication in a second Caucasian psychiatric sample (n=140). SH2B1 rs3888190C>A was significantly associated with LDL levels in the discovery and in the replication sample, with A-allele carriers having 0.2mmol/l (p=0.005) and 0.36mmol/l (p=0.007) higher LDL levels compared to others, respectively. G-allele carriers of RABEP1 rs1000940A>G had lower fasting glucose levels compared to others in both samples (-0.16mmol/l; p<0.001 and -0.77mmol/l; p=0.03 respectively). The present study is the first to observe such associations in human subjects, which may in part be explained by a high risk towards dyslipidemia and diabetes in psychiatric patients receiving psychotropic treatments compared to population-based individuals. These results may therefore give new insight into the etiology of LDL-cholesterol and glucose regulation in psychiatric patients under psychotropic drug therapy.

Comparing probabilistic and descriptive analyses of time-dose-toxicity relationship for determining no-observed-adverse-effect level in drug development.

The no-observed-adverse-effect level (NOAEL) of a drug defined from animal studies is important for inferring a maximal safe dose in human. However, several issues are associated with its concept, determination and application. It is confined to the actual doses used in the study; becomes lower with increasing sample size or dose levels; and reflects the risk level seen in the experiment rather than what may be relevant for human. We explored a pharmacometric approach in an attempt to address these issues. We first used simulation to examine the behaviour of the NOAEL values as determined by current common practice; and then fitted the probability of toxicity as a function of treatment duration and dose to data collected from all applicable toxicology studies of a test compound. Our investigation was in the context of an irreversible toxicity that is detected at the end of the study. Simulations illustrated NOAEL's dependency on experimental factors such as dose and sample size, as well as the underlying uncertainty. Modelling the probability as a continuous function of treatment duration and dose simultaneously to data from multiple studies allowed the estimation of the dose, along with its confidence interval, for a maximal risk level that might be deemed as acceptable for human. The model-based data integration also reconciled between-study inconsistency and explicitly provided maximised estimation confidence. Such alternative NOAEL determination method should be explored for its more efficient data use, more quantifiable insight to toxic doses, and the potential for more relevant animal-to-human translation.

Influence of renal function estimation on pharmacokinetic modeling of vancomycin in elderly patients.

Vancomycin is a renally excreted drug, and its body clearance correlates with creatinine clearance. However, the renal function estimation equation that best predicts vancomycin clearance has not been established yet. The objective of this study was to compare the abilities of different renal function estimation equations to describe vancomycin pharmacokinetics in elderly patients. The NPAG algorithm was used to perform population pharmacokinetic analysis of vancomycin concentrations in 78 elderly patients. Six pharmacokinetic models of vancomycin clearance were built, based on the following equations: Cockcroft-Gault (CG), Jelliffe (JEL), Modification of Diet in Renal Disease (MDRD), Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) (both in milliliters per minute per 1.73 m(2)), and modified MDRD and CKD-EPI equations (both in milliliters per minute). Goodness-of-fit and predictive performances of the six PK models were compared in a learning set (58 subjects) and a validation set (20 patients). Final analysis was performed to estimate population parameters in the entire population. In the learning step, the MDRD-based model best described the data, but the CG- and JEL-based models were the least biased. The mean weighted errors of prediction were significantly different between the six models (P = 0.0071). In the validation group, predictive performances were not significantly different. However, the use of a renal function estimation equation different from that used in the model building could significantly alter predictive performance. The final analysis showed important differences in parameter distributions and AUC estimation across the six models. This study shows that methods used to estimate renal function should not be considered interchangeable for pharmacokinetic modeling and model-based estimation of vancomycin concentrations in elderly patients.