Please use this identifier to cite or link to this item:
Population Pharmacokinetics of Weekly Isoniazid and Rifapentine for Latent Tuberculosis Infection
isoniazid,rifapentine,latent tuberculosis infection,population pharmacokinetics,Monolix,
|Publication Year :||2020|
潛伏結核感染是一種結核菌長期存在於宿主體內，使宿主持續產生免疫反應的狀態。潛伏結核感染者沒有症狀，但終生都有可能發展為活動性結核。為了達成世界衛生組織（World Health Organization, WHO）2035年結核盛行率降低至每十萬人口10例以下的目標，針對尚未發病的個案給予潛伏結核感染治療是重要的疾病防治策略。速克伏（3HP）為每週一次isoniazid合併rifapentine總計12週之短程處方（依體重調整劑量，成人劑量約600-900 mg），其優勢在於短期療程之服藥順從性高，肝毒性發生率較傳統處方低，但近年研究發現病人容易發生全身性藥物反應副作用，影響藥物血中濃度或分布代謝個體間變異的共變項是否扮演重要角色有待釐清，且目前缺乏針對臺灣潛伏結核族群使用3HP的族群藥動學模型。因此本研究旨在探討臺灣潛伏結核族群高劑量isoniazid與rifapentine之藥動參數及影響藥動參數個體間差異的共變項，做為未來臨床分析之基礎。
本研究為前瞻、觀察性研究，研究對象為2017年10月至2019年10月使用3HP做為潛伏結核感染預防性治療之成年人。Isoniazid與rifapentine抽血時間為服藥後3, 6, 24, 48小時，受試者血中藥物濃度採用超高效液相層析串聯質譜分析，並使用nonlinear mixed effects model (Monolix software, 2019R2)估算受試者藥動學參數。由於本研究藥物於吸收相濃度資料不足，因此採用文獻之吸收速率常數與吸收延遲情形並從中選出最適合者固定（fixed）於本研究模型；isoniazid與rifapentine皆採用一級線性吸收與排除建立模型，尋找最適當的個體間、個體內、場合間變異模型，並分析臨床共變項、共病症、肝腎功能、併用藥物、藥物代謝酵素或運輸蛋白基因多型性與藥動參數的相關性。
Isoniazid模型為一級吸收與一級排除速率之二室模型，個體內變異以常數型（constant），擬似分布體積與擬似清除率個體間變異以指數型（exponential）最能夠描述本研究之濃度資料。群體中央室擬似分布體積為141 L，年齡為顯著影響的共變項，加入模型後中央室擬似分布體積之個體間變異從56.5%下降至15.9%，年齡越大則中央室擬似分布體積越大，呈正相關。群體清除率為95.6 L/h，NAT2表現型與是否洗腎是影響isoniazid清除率的2個重要共變項，清除率個體間差異從94%下降至45.5%。NAT2慢速乙醯代謝者之isoniazid清除率約為快速與中等乙醯代謝者的26.4%；洗腎患者（服藥、抽血日與洗腎日間隔）之isoniazid清除率為非洗腎者的35.7%。
Rifapentine模型為一級吸收與一級排除速率之一室模型，個體內變異以常數型（constant），擬似分布體積個體間變異以指數型（exponential），清除率個體間變異以疊加型（additive）描述最佳。群體擬似分布體積為29.5 L，真實體重以及是否為洗腎病人是顯著影響的共變項，其中體重與擬似分布體積呈正相關，洗腎者擬似分布體積為非洗腎者之1.69倍，將此二共變項加入模型後rifapentine擬似分布體積個體間變異從55.7%下降至33.3%。群體清除率為0.929 L/h，於藥動模型中沒有發現顯著影響的共變項。進行多變項迴歸分析發現代謝酵素AADAC rs1803155 GG genotype與清除率較高相關，服用steroids與清除率較低相關。
Latent tuberculosis infection (LTBI) is a state of positive immune response to stimulation by Mycobacterium tuberculosis antigens without evidence of clinically manifested active TB. The World Health Organization (WHO) has established the goal of tuberculosis (TB) elimination (incidence rate < 10 per 100,000 populations) by 2035. Therefore, providing preventive therapy for LTBI individuals is important. 3HP regimen containing 12 doses of isoniazid (INH) and rifapentine (RPT) is recommended to prevent the development of active TB. It demonstrated lower incidence of hepatotoxicity but higher incidence of systemic drug reactions (SDR). Whether the drug concentrations or the factors affecting the variability in pharmacokinetics (PK) among individuals play an important role remain uncertain. In addition, PK of 3HP are not fully described in Taiwanese. Therefore, this study aimed to characterize the population PK of 3HP in Taiwanese LTBI participants.
Adult participants who diagnosed with LTBI and received 3HP were enrolled prospectively from October 2017 to October 2019. Plasma INH and RPT concentrations were drawn at 3, 6, 24 and 48 hours after dose administration, and analyzed by liquid chromatography with tandem mass spectrometry. Pharmacokinetic data of INH and RPT were analyzed using nonlinear mixed effects model in Monolix software (version 2019R2) to estimate population pharmacokinetics. The absorption rate constant (Ka) and absorption lag time were fixed according to the values from literature due to the lack of absorption information in our data. One- and two-compartment disposition models with first-order absorption and elimination were used to describe the PK of INH and RPT, respectively. Suitable models for inter-individual variability, intra-individual variability, and inter-occasion variability were tested to describe the residual errors. The effects of potential covariates such as demographics and clinical characteristics, concomitant diseases, liver and kidney laboratory tests, co-medication, genetic polymorphisms of metabolic enzymes and transporters were evaluated.
A total of 35 subjects were enrolled and contributed 77 plasma concentrations. The median age and weight of the subjects were 52 years (range, 24 to 75 years) and 64.4 kg (range, 42 to 93 kg), respectively; 17 of the 35 subjects were female.
The population PK of INH were best described by a two-compartment model with first-order elimination and first-order absorption. Residual and inter-individual variabilities were described as constant and exponential error models, respectively. The value of apparent central volume of distribution (Vc) was 141 L. Age for allometric scaling on Vc was identified as a significant covariate, reducing the interindividual variability from 56.5% to 15.9%. Subjects who were older were associated with larger Vc. The value of clearance (CL) was 95.6 L/h. NAT2 phenotype and dialysis were confirmed as significant covariates, reducing the interindividual variability from 94% to 45.5%. Slow acetylators were associated with a 73.6% lower CL compared with rapid and intermediate acetylators, and patients who were under dialysis were found to have a 64.3% lower CL.
The population PK of RPT were well described by a one-compartment model with first-order elimination and first-order absorption with a lag time. Residual was described as constant error model. Inter-individual variabilities of apparent volume of distribution (V) and CL were described as exponential and additive error models, respectively. The value of V was 29.5 L. Total body weight (via allometric scaling) and dialysis significantly affected RPT V, reducing the interindividual variability from 55.7% to 33.3%. Patients with higher weight were associated with larger V, and patients who were under dialysis were found to had a 1.69-fold higher V. The value of CL was 0.929 L/h. Factors responsible for the variability in RPT CL was not found in the study. In stepwise multivariate regression analysis, factors independently associated with CL included AADAC rs1803155 GG genotype (coefficient 0.27; 95% confidence interval [CI], 0.13, 0.42; p=0.0003) and receipt of steroids (coefficient -0.21; 95% CI, -0.39, -0.03; p=0.02).
We developed a population pharmacokinetic model for high dose INH and RPT in Taiwan. In the INH model, age explained the variability in INH Vc, while NAT2 phenotype status and dialysis significantly affect the variability in INH clearance. In the RPT model, weight and dialysis had a major impact on V. This study pointed out that severe renal insufficiency had significant impact on drug distribution and metabolism from the population pharmacokinetics perspective.
Isoniazid, rifapentine, latent tuberculosis infection, population pharmacokinetics, Monolix
|Appears in Collections:||臨床藥學研究所|
Files in This Item:
|7.28 MB||Adobe PDF|
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.