Clinical Pharmacokinetics最新文献

Background and objective: In pediatric trial design, it is particularly essential to maximize data utilization and ensure robust designs while minimizing sample collection. A common criterion to justify pediatric pharmacokinetic study designs is based on parameter precision (PP) evaluation, recommended by the US Food and Drug Administration. Here, we propose an alternative approach to design evaluation based on accuracy for dose selection (ADS).

Methods: This work was conducted using a simulation-and-reestimation framework, based on a real-case scenario of designing the single-dose pharmacokinetic study of the anti-tuberculosis (TB) drug pretomanid, with the aim of selecting doses for the next multidose long-term study. The study powers were computed using the ADS approach under scenarios with (1) real-case conditions, (2) high variabilities, (3) available options of tablet doses for selections. The study power using a PP approach was computed to compare with the ADS approach.

Results: The ADS approach suggested that the design selected accurate doses with study power >80% in almost all dosing weight groups, whereas the PP approach found the design underpowered for clearance. The ADS-based power was decreased to ~65% in the smallest weight groups given high variability. Varying the options of dose levels affected the ADS-based power non-monotonically, although fewer levels generally yielded higher power.

Conclusion: The ADS approach practically evaluates the precision in dose selection, providing a directly relevant decision criterion for designing pediatric pharmacokinetic studies and could be an alternative for power evaluation when the study is focused on determining doses using discrete tablet sizes.

Background and objective: Epidural sufentanil is widely used for intraoperative analgesia and is recognized as a safe and effective route of administration when carefully monitored and administered at the lowest effective doses. Optimizing drug dosing requires a precise understanding of its pharmacokinetics. While it is known that sufentanil concentrations in the blood decline more slowly after epidural administration compared with intravenous administration, and the flip-flop phenomenon has been observed in this context, a pharmacokinetic model accounting for this phenomenon has not yet been described.

Methods: This study aimed to develop a pharmacokinetic model of sufentanil following epidural administration, with a focus on its absorption from the epidural space. The study was performed in 23 patients, aged 30-83 years with an American Society of Anesthesiologists (ASA) classification of 2-3, undergoing major abdominal (oncologic and non-oncologic) surgery under general anesthesia and epidural analgesia. Blood samples were collected, and sufentanil concentrations were measured at time points ensuring coverage of the absorption phase. Population analysis was performed using a full Bayesian approach implemented in Stan/Torsten programs with the "cmdstanr" and "bbr.bayes" packages in RStudio.

Results: The Bayesian approach helped incorporate prior information about sufentanil disposition. A two-compartment disposition model with two-compartmental absorption best described the data, featuring a fast absorption rate constant into plasma and the peripheral compartment, and a slow redistribution process from the epidural fat compartment.

Conclusions: This study mechanistically describes the flip-flop pharmacokinetics of sufentanil and allows for more precise dosing of epidural sufentanil (NCT06069219).

Malnutrition significantly alters the pharmacokinetics of medications, particularly in vulnerable populations such as children, pregnant women, elderly individuals, and individuals in low- and middle-income countries. These populations are often more vulnerable to the effects of malnutrition because of physiological, metabolic and socioeconomic factors. Changes in body composition, organ function and plasma protein levels associated with malnutrition can impact drug absorption, distribution, metabolism and excretion. In malnourished individuals, decreased serum albumin levels may increase the free (unbound) fraction of highly protein-bound acidic drugs, potentially elevating the risk of toxicity. However, this relationship is not universally straightforward, as it depends on the drug's protein-binding characteristics, hepatic and renal function, volume of distribution and compensatory changes in drug clearance. In addition, malnutrition's effects on liver enzymes, such as cytochrome P450 isoforms, and kidney function can result in unpredictable drug clearance, particularly for narrow-therapeutic-index medications. Emerging evidence also highlights the interplay between malnutrition and pharmacogenomics, with genetic variations further modulating drug metabolism and response. Addressing these complexities requires the development of tailored dosing regimens and adaptive therapeutic strategies to optimise treatment outcomes in these at-risk groups. This review accentuates the critical need for more robust research to inform clinical guidelines and improve health equity in managing malnourished populations globally.

Background and objectives: Nedosiran (Rivfloza^®) is an RNA interference (RNAi) therapy approved for individuals aged ≥ 2 years with primary hyperoxaluria type 1 (PH1), a rare autosomal-recessive disorder causing renal failure and systemic oxalosis. Nedosiran silences lactate dehydrogenase (LDH) mRNA in hepatocytes, reducing oxalate levels. This study evaluated the model-informed clinical development of nedosiran to support proposed doses in children aged 2 to < 12 years with PH1.

Methods: A population pharmacokinetic/pharmacodynamic (Pop-PK/PD) model characterizing the plasma concentration-time profile of nedosiran and its effect on the spot urine oxalate-to-creatinine ratio (Uox/Cr) was developed using data from six trials. Simulations assessed spot Uox/Cr reduction in children aged 2 to < 12 years for the proposed dosing regimen versus those aged ≥ 12 years weighing ≥ 50 kg with similar renal function.

Results: The datasets included 2087 PK (N = 148) and 668 spot Uox/Cr (N = 41, with PH1) observations. Body weight, estimated glomerular filtration rate (eGFR), and PH type were covariates in the PK model, with body weight in low and high percentiles affecting nedosiran exposures. Moderate renal impairment (eGFR 30-59 mL/min/1.73 m²) increased exposure, while only age was significant for baseline Uox/Cr in the PD model. Simulations showed similar Uox/Cr reduction and times to maximum effect in children aged 2 to < 12 years, treated once-monthly (Q1M) with 3.5 mg/kg, compared to those aged ≥ 12 years treated Q1M with 170 mg.

Conclusions: Simulations based on the final Pop-PK/PD model support the 3.5 mg/kg Q1M dosing regimen in children aged 2 to < 12 years with PH1 and relatively intact kidney function (eGFR ≥30 mL/min/1.73 m²).

Trial registration: Trials are registered at ClinicalTrials.gov with study numbers NCT03392896 (PHYOX1), NCT03847909 (PHYOX2), NCT04042402 (PHYOX3), and NCT05001269 (PHYOX8) and at EudraCT with study numbers 2018-003098-91 (PHYOX2) and 2018-003099-10 (PHYOX3).

Background and objective: Individualized drug dosing is a highly effective strategy for optimizing therapeutic outcomes, especially for drugs with high inter-individual variability. Population pharmacokinetic modeling is a widely used approach to characterize inter-individual variability in therapeutic drug monitoring. However, the development of population pharmacokinetic models is labor intensive and requires significant technical expertise. Machine learning (ML) represents a promising alternative for personalized drug dosing strategies. Despite numerous studies applying ML in this context, no previous work has comprehensively reviewed and compared their methodologies and predictive performance. This scoping review addresses this gap in the existing literature with the aim to examine the methodological approaches used in ML-based pharmacokinetic modeling for dose optimization.

Methods: Five databases were systematically searched from their inception to May 2025. Studies comparing predictions of drug concentrations or pharmacokinetic parameters between ML and population pharmacokinetic models were included. Studies published in non-English language, reviews, protocols, or studies that did not employ ML models for individualized dose regimens or treatment plans were excluded.

Results: Fifty-eight studies were included. We found that boosting-based models, tree-based models, instance-based, and regression-based models were the most commonly used ML approaches. Approximately 31% of the studies integrated ML with population pharmacokinetic models, while the remainder developed stand-alone ML models. Inconsistencies in reporting were evident, as only 60% of the studies detailed their feature selection methods. Model evaluation approaches also varied: 47% of ML models used internal test sets, while the remainder employed external datasets or mixed approaches. In terms of predictive accuracy, ML models performed comparably to or better than population pharmacokinetic models, especially for drugs with significant pharmacokinetic variability.

Conclusions: This review identifies substantial heterogeneity in ML modeling approaches, feature selection, and model evaluation. To enhance the reproducibility and clinical applicability of ML models in individualized drug dosing, standardization in reporting and methodological practices is essential.

Background and objectives: Emerging evidence suggests that comorbidities like human immunodeficiency virus (HIV) infection, diabetes mellitus (DM), and malnutrition in tuberculosis (TB) patients can alter drug concentrations, thereby affecting the treatment outcomes. For these populations, personalised strategies such as therapeutic drug monitoring (TDM) may be essential. We investigated the variations of drug levels within comorbid populations and analysed the differences in patterns observed between sub-Saharan Africa (SSA) and non-SSA regions.

Methods: We performed a systematic review and meta-analysis of rifampicin drug pharmacokinetics (PK) through searches of major databases from 1980 to December 2023. A random-effects meta-analysis model using R-studio version 4.3.2 was conducted to estimate pooled serum rifampicin exposure (area under the concentration-time curve [AUC], and peak maximum concentration [C_max]) between patients with TB-HIV infection, and TB-DM.

Results: From 3300 articles screened, 24 studies met inclusion criteria, contributing 33 comorbidity subgroups for meta-analysis. In SSA, 14 subgroups assessed rifampicin PK in TB-HIV, 1 in TB-DM, and none in TB-malnutrition. The pooled mean C_max was below the recommended range (8-24 mg/L) for all subgroups. For TB-HIV, the pooled C_max was 5.59 mg/L, 95% CI (4.59-6.59), I² = 97% for SSA populations and 5.59 mg/L, 95% CI (3.65; 6.59) for non-SSA populations. The C_max for TB-DM in SSA (9.60 ± 4.4 mg/L) exceeded non-SSA (4.27 mg/L, 95% CI [2.77-5.76]). The lowest AUC was in TB-HIV (SSA, 29.09 mg/L h, 95% CI [21.06; 37.13, I² = 91%]). High variability and heterogeneity (I² >90%) were observed, with most studies (20/23) showing low bias.

Conclusion: Our results emphasise the need for individualised dosing and targeted TDM implementation among TB-HIV and TB-DM populations on rifampicin in SSA. Although all populations exhibited low C_max levels, TB-HIV populations may be prioritised as AUC levels were lowest. In clinical settings in SSA, C_max-based TDM is more practical, but AUC can be used in treatment where feasible.

Background and objective: Binimetinib is approved for multiple indications at a therapeutic dose of 45 mg twice a day (BID), in combination with encorafenib. A clinical hepatic impairment (HI) study was designed to evaluate the pharmacokinetics (PK), safety, and tolerability of a single oral dose of binimetinib in participants with mild, moderate, and severe HI compared with demographically matched healthy participants with respect to age, gender, and body weight.

Methods: Participants were enrolled according to National Cancer Institute (NCI) classification criteria for hepatic function based on their total bilirubin and aspartate aminotransferase levels at screening. Participants enrolled into Group 1 (normal hepatic function) were matched to participants enrolled into Groups 2, 3, and 4 (mild, moderate, and severe HI, respectively) with respect to age, gender, and body weight. Dose-normalized PK parameters were evaluated because of a difference in doses for the severe HI group compared to the other groups, with the dose reduction due to the increased exposures observed in the moderate HI group.

Results: Among 27 PK evaluable participants, changes in binimetinib dose-normalized PK parameters C_max/D and AUC_inf/D were minimal in participants with mild HI compared to the normal hepatic function group. Both the moderate and severe HI groups had significant changes as AUC_inf/D increased by 81% and 111%, respectively, compared to the normal hepatic function group. Unbound AUC_last/D for the moderate and severe HI groups increased by 280% and 248% compared to the normal hepatic function group, respectively.

Conclusion: Based on these findings on total and unbound exposures, dose reductions are recommended for binimetinib in cancer patients with moderate and severe HI.

Clinical trial registration: ClinicalTrials.gov NCT02050815, registered 29 January 2014.

Introduction: Benralizumab is approved as add-on subcutaneous therapy in patients aged ≥ 12 years with severe eosinophilic asthma in > 80 countries, including mainland China.

Objective: The study objective was to update benralizumab population pharmacokinetic (popPK) and exposure-response (ER) models in Chinese, Asian (including Chinese), and non-Asian participants.

Methods: Benralizumab popPK/ER models for asthma exacerbation rate and pre-bronchodilator forced expiratory volume in 1 second (FEV₁) were updated for three benralizumab trials involving Chinese, Asian (including Chinese), and non-Asian participants. The ER analysis examined correlations between pharmacokinetic quartiles and annual asthma exacerbation rate (AAER) ratios with simulations comparing predicted clinical outcomes.

Results: Updated data included 17,465 benralizumab concentrations (n = 2855). The updated model predicted a slight, and not clinically relevant, increase (< 14%) in benralizumab exposure for Chinese versus non-Asian adults. Median exposure increased in Chinese adolescents versus adults owing to body weight differences, but no dose adjustment was needed. Chinese children weighing < 35 kg receiving a 10 mg dose had similar exposure to those weighing ≥ 35 kg receiving a 30 mg dose. In Chinese versus non-Chinese participants, there was no trend concerning AAER ratios across different trough concentration quartiles; the maximal treatment effect significantly increased (+127%; p < 0.001), and there was no statistically significant effect on pre-bronchodilator FEV₁. Steady-state simulations showed lower predicted AAER ratios in Chinese (0.38; 95% confidence interval [CI] 0.32-0.45) than in non-Chinese adults (0.64; 95% CI 0.60-0.71), and no relevant differences between Chinese adults (0.46; 95% CI 0.38-0.54) and adolescents (0.46; 95% CI 0.37-0.55).

Conclusion: The benralizumab popPK/ER models showed good predictive performance across Chinese demographics.

Trial registration number: NCT03186209.

Trial registration date: 6 July 2017.

Background and objective: Obesity can alter the physiological profile of individuals, potentially impacting the pharmacokinetics of anesthetic agents. This study compared the pharmacokinetic profiles of lidocaine and its metabolites between obese patients and normal-weight patients following a single intravenous bolus during surgical operation, to inform dosing strategies for the obese Chinese population.

Methods: Twenty-nine obese patients scheduled for laparoscopic sleeve gastrectomy and 29 normal-weight patients for laparoscopic cholecystectomy were enrolled. Lidocaine (2%, 1.5 mg/kg) was administered intravenously to obese patients and normal-weight patients on the basis of adjusted body weight (ABW) and total body weight, respectively. Plasma samples were collected to analyze the pharmacokinetic profiles of lidocaine and its metabolites. Adverse events (AEs) were recorded throughout the study.

Results: Obese patients had a significantly longer half-life for lidocaine (2.27 ± 0.69 h vs 0.94 ± 0.16 h, p < 0.0001), a higher volume of distribution (105 ± 27.3 L vs 54.9 ± 14.0 L, p < 0.0001), and a lower clearance (33.6 ± 9.08 L/h vs 40.5 ± 8.67 L/h, p = 0.008) compared to normal-weight patients. Although exposure to lidocaine was similar between groups within 2 hours, obese patients had lower metabolite concentrations due to decreased metabolic capacity. The plasma concentrations in all patients remained below the toxic concentration of 5 μg/mL, and no serious lidocaine-related AEs were reported.

Conclusions: Obesity significantly affects the pharmacokinetics of lidocaine and its active metabolites, and administering lidocaine intravenously via ABW is safe and reasonable for obese patients.

Clinical trial registration: ChiCTR2200064980, 25 October 2022.