Wu Hulin, Huang Yangxin, Acosta Edward P, Park Jeong-Gun, Yu Song, Rosenkranz Susan L, Kuritzkes Daniel R, Eron Joseph J, Perelson Alan S, Gerber John G
Department of Biostatistics and Computational Biology, University of Rochester School of Medicine and Dentistry, NY 14642, USA.
J Pharmacokinet Pharmacodyn. 2006 Aug;33(4):399-419. doi: 10.1007/s10928-006-9006-4. Epub 2006 Apr 1.
We developed a novel HIV-1 dynamic model with consideration of pharmacokinetics, drug adherence and drug susceptibility to link plasma drug concentration to the long-term changes in HIV-1 RNA observation after initiation of therapy. A Bayesian approach is proposed to fit this model to clinical data from ACTG A5055, a study of two dosage regimens of indinavir (IDV) with ritonavir (RTV) in subjects failing their first protease inhibitor treatment. The HIV RNA testing was completed at days 0, 7, 14, 28, 56, 84, 112, 140, and 168. An intensive pharmacokinetic (PK) evaluation was performed on day 14 and multiple trough concentrations were subsequently collected. Pill counts were used to monitor adherence. IC(50) for IDV and RTV were determined at baseline and at virologic failure. Viral dynamic model fitting residuals were used to assess the significance of covariate effects on long-term virologic response. As univariate predictors, none of the four PK parameters C(trough), C(12 hour), C(max), and AUC was significantly related to virologic response (p > 0.05). By including drug susceptibility (IC(50)), or IC(50) and adherence measured by pill counts together, C(trough), C(12 hour), C(max) and AUC were each significantly correlated to long-term virologic response (p = 0.0055,0.0002,0.0136,0.0002 with IC(50) and adherence measured by pill counts considered). The IC(50) and adherence measured by pill counts alone were not related to the virologic response. In predicting virologic response adherence measured by pill counts did not provide any additional information to PK parameters (p = 0.064), to drug susceptibility IC(50) (p = 0.086), and to their combination (p = 0.22). Simple regression approaches did not detect any significant pharmacodynamic (PD) relationships. Any single factor of PK, adherence measured by pill counts and drug susceptibility did not contribute to long-term virologic response. But their combinations in viral dynamic modeling significantly predicted virologic response. The HIV dynamic modeling can appropriately capture complicated nonlinear relationships and interactions among multiple covariates.
我们开发了一种新型的HIV-1动态模型，该模型考虑了药代动力学、药物依从性和药物敏感性，以将血浆药物浓度与治疗开始后HIV-1 RNA观察值的长期变化联系起来。提出了一种贝叶斯方法，将该模型与ACTG A5055的临床数据进行拟合，ACTG A5055是一项关于茚地那韦（IDV）与利托那韦（RTV）两种剂量方案在首次蛋白酶抑制剂治疗失败的受试者中的研究。HIV RNA检测在第0、7、14、28、56、84、112、140和168天完成。在第14天进行了强化药代动力学（PK）评估，随后收集了多个谷浓度。通过药丸计数来监测依从性。在基线和病毒学失败时测定IDV和RTV的IC(50)。使用病毒动态模型拟合残差来评估协变量对长期病毒学反应的影响的显著性。作为单变量预测因子，四个PK参数C(谷值)、C(12小时)、C(最大值)和AUC均与病毒学反应无显著相关性（p>0.05）。通过纳入药物敏感性（IC(50)），或同时纳入IC(50)和通过药丸计数测量的依从性，C(谷值)、C(12小时)、C(最大值)和AUC均与长期病毒学反应显著相关（考虑IC(50)和通过药丸计数测量的依从性时，p分别为0.0055、0.0002、0.0136、0.0002）。单独通过药丸计数测量的IC(50)和依从性与病毒学反应无关。在预测病毒学反应时，通过药丸计数测量的依从性未为PK参数（p = 0.064）、药物敏感性IC(50)（p = 0.086）及其组合（p = 0.22）提供任何额外信息。简单回归方法未检测到任何显著的药效学（PD）关系。PK、通过药丸计数测量的依从性和药物敏感性的任何单个因素均对长期病毒学反应无贡献。但它们在病毒动态建模中的组合显著预测了病毒学反应。HIV动态建模可以适当地捕捉多个协变量之间复杂的非线性关系和相互作用。
