Pôle Pharmacie, Service Hospitalo-Universitaire de Pharmacie, CHU de Rennes, 35033 Rennes, France; Laboratoire de Biopharmacie et Pharmacie Clinique, Faculté de Pharmacie, Université de Rennes 1, 35043 Rennes, France; Univ Rennes, CHU Rennes, Inserm, EHESP, Irset (Institut de recherche en santé, environnement et travail) - UMR_S 1085, F-35000 Rennes, France.
Univ Rennes, CHU Rennes, INSERM, LTSI - UMR 1099, F-35000 Rennes, France.
Eur J Pharm Biopharm. 2021 Jul;164:36-53. doi: 10.1016/j.ejpb.2021.04.014. Epub 2021 Apr 23.
There are few studies in humans dealing with the relationship between physico-chemical properties of drugs and their systemic bioavailability after administration via oral inhalation route (Fpulm). Getting further insight in the determinants of Fpulm after oral pulmonary inhalation could be of value for drugs considered for a systemic delivery as a result of poor oral bioavailability, as well as for drugs considered for a local delivery to anticipate their undesirable systemic effects. To better delineate the parameters influencing the systemic delivery after oral pulmonary inhalation in humans, we studied the influence of physico-chemical and permeability properties obtained in silico on the rate and extent of Fpulm in a series of 77 compounds with or without marketing approval for pulmonary delivery, and intended either for local or for systemic delivery. Principal component analysis (PCA) showed mainly that Fpulm was positively correlated with Papp and negatively correlated with %TPSA, without a significant influence of solubility and ionization fraction, and no apparent link with lipophilicity and drug size parameters. As a result of the small sample set, the performance of the different models as predictive of Fpulm were quite average with random forest algorithm displaying the best performance. As a whole, the different models captured between 50 and 60% of the variability with a prediction error of less than 20%. Tmax data suggested a significant positive influence of lipophilicity on absorption rate while charge apparently had no influence. A significant linear relationship between Cmax and dose (R = "0.79) highlighted that Cmax was primarily dependent on dose and absorption rate and could be used to estimate Cmax in humans for new inhaled drugs.
目前针对经口服肺吸入给药途径（Fpulm）后药物的理化性质与其全身生物利用度之间的关系，仅有少量人体研究。深入了解经口服肺吸入后 Fpulm 的决定因素，对于那些由于口服生物利用度差而考虑全身给药的药物，以及那些考虑局部给药以预测其不良全身作用的药物，可能具有重要价值。为了更好地阐明影响人体经口服肺吸入后全身递送的参数，我们研究了在一系列具有或不具有肺部给药上市许可、旨在局部或全身给药的 77 种化合物中，通过计算获得的理化性质和渗透性参数对 Fpulm 速率和程度的影响。主成分分析（PCA）表明，Fpulm 主要与 Papp 呈正相关，与 %TPSA 呈负相关，而与溶解度和离解分数无显著相关性，与亲脂性和药物大小参数也无明显相关性。由于样本量较小，不同模型预测 Fpulm 的性能相当平均，随机森林算法显示出最佳性能。总的来说，不同的模型捕捉到了 50%至 60%的变异性，预测误差小于 20%。Tmax 数据表明亲脂性对吸收速率有显著的正影响，而电荷显然没有影响。Cmax 与剂量之间存在显著的线性关系（R = "0.79），这突出表明 Cmax 主要取决于剂量和吸收速率，并可用于估计新吸入药物在人体中的 Cmax。
