Scholarlink

Objective <p>This study aimed to establish a population pharmacokinetic (PPK) model of trimethoprim-sulfamethoxazole (TMP/SMZ) in critically ill patients from Hainan, to identify key determinants of drug exposure, and to provide a basis for individualized dosing optimization.</p> Methods <p>Twenty-eight ICU patients receiving TMP/SMZ therapy were enrolled. Plasma concentrations of TMP, SMZ, and the metabolite <i>N-acetyl-sulfamethoxazole</i> (NSMZ) at steady state were quantified using LC-MS/MS. Demographic characteristics, clinical laboratory indices, and genotypes of <i>NAT2</i>, <i>CYP2C9*</i>3, and <i>GCLC</i> were collected as key covariates. PPK modeling was performed using NONMEM 7.4. We compared one-compartment and multi-compartment structural models and assessed the influence of covariates on pharmacokinetic parameters. The final model was validated through multiple approaches, including goodness-of-fit (GOF) plots, normalized prediction distribution errors (NPDE), prediction-corrected visual predictive checks (pcVPC), and nonparametric bootstrap analysis. Monte Carlo simulations were then conducted to evaluate target attainment rates and toxicity risks of TMP, SMZ, and NSMZ under different dosing regimens.</p> Results <p>TMP pharmacokinetics were best described by a one-compartment model with first-order absorption, and no significant covariates were identified for interindividual variability. SMZ and its metabolite NSMZ were also fitted with one-compartment models. Covariate analysis indicated that body weight significantly influenced SMZ clearance (CL, <i>P</i> &lt; 0.01), while the <i>CYP2C9*3</i> genotype significantly affected NSMZ CL (<i>P</i> &lt; 0.001). Model diagnostics showed good fit in GOF plots, with NPDE and pcVPC indicating stable predictive performance, and bootstrap success rate reached 100%. Monte Carlo simulations revealed that conventional dosing could result in TMP and NSMZ concentrations exceeding toxicity thresholds in some patients, whereas SMZ concentrations often remained below therapeutic levels. Considering both efficacy and safety, simulations suggested that 960&#xa0;mg TID or 1440&#xa0;mg BID regimens provide an optimal balance for TMP, SMZ, and NSMZ concentrations, supporting subsequent individualized dosing strategies.</p> Conclusion <p>This study is the first to establish a PPK model of TMP/SMZ, including the metabolite of SMZ, in critically ill patients from Hainan, China. Body weight and <i>CYP2C9*</i>3 genotype were identified as key covariates influencing drug exposure, indicating that genetic information should be considered in individualized dosing. The findings provide a theoretical basis for optimized TMP/SMZ use in special populations and offer methodological and data support for future model-informed precision dosing (MIPD) applications.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Population pharmacokinetic modeling and Monte Carlo simulation–based dosing optimization of trimethoprim–sulfamethoxazole in critically ill patients

  • 期刊论文

  • 发表日期：

    2026年3月31日

查看全文
  • Min Wang,
  • Chen Li,
  • Xiangxiang Fu,
  • Fangling Wang,
  • Xiaojun Zhou,
  • Bingli Zhang,
  • Wenya Xu,
  • Liqiu Liang,
  • Zhenlin Lei,
  • Hengxing Qu,
  • Ling Xie,
  • Lili Zhong,
  • Yuanyuan Zheng,
  • Tao Liu

摘要

Objective

This study aimed to establish a population pharmacokinetic (PPK) model of trimethoprim-sulfamethoxazole (TMP/SMZ) in critically ill patients from Hainan, to identify key determinants of drug exposure, and to provide a basis for individualized dosing optimization.

Methods

Twenty-eight ICU patients receiving TMP/SMZ therapy were enrolled. Plasma concentrations of TMP, SMZ, and the metabolite N-acetyl-sulfamethoxazole (NSMZ) at steady state were quantified using LC-MS/MS. Demographic characteristics, clinical laboratory indices, and genotypes of NAT2, CYP2C9*3, and GCLC were collected as key covariates. PPK modeling was performed using NONMEM 7.4. We compared one-compartment and multi-compartment structural models and assessed the influence of covariates on pharmacokinetic parameters. The final model was validated through multiple approaches, including goodness-of-fit (GOF) plots, normalized prediction distribution errors (NPDE), prediction-corrected visual predictive checks (pcVPC), and nonparametric bootstrap analysis. Monte Carlo simulations were then conducted to evaluate target attainment rates and toxicity risks of TMP, SMZ, and NSMZ under different dosing regimens.

Results

TMP pharmacokinetics were best described by a one-compartment model with first-order absorption, and no significant covariates were identified for interindividual variability. SMZ and its metabolite NSMZ were also fitted with one-compartment models. Covariate analysis indicated that body weight significantly influenced SMZ clearance (CL, P < 0.01), while the CYP2C9*3 genotype significantly affected NSMZ CL (P < 0.001). Model diagnostics showed good fit in GOF plots, with NPDE and pcVPC indicating stable predictive performance, and bootstrap success rate reached 100%. Monte Carlo simulations revealed that conventional dosing could result in TMP and NSMZ concentrations exceeding toxicity thresholds in some patients, whereas SMZ concentrations often remained below therapeutic levels. Considering both efficacy and safety, simulations suggested that 960 mg TID or 1440 mg BID regimens provide an optimal balance for TMP, SMZ, and NSMZ concentrations, supporting subsequent individualized dosing strategies.

Conclusion

This study is the first to establish a PPK model of TMP/SMZ, including the metabolite of SMZ, in critically ill patients from Hainan, China. Body weight and CYP2C9*3 genotype were identified as key covariates influencing drug exposure, indicating that genetic information should be considered in individualized dosing. The findings provide a theoretical basis for optimized TMP/SMZ use in special populations and offer methodological and data support for future model-informed precision dosing (MIPD) applications.