A unified strategy in selection of the best allometric scaling methods to predict human clearance based on drug disposition pathway

1. It is critical to develop a unified strategy to select the best allometric scaling (AS) method for a given group of drugs. 2. A total of 446 drugs with known human CLiv, clear disposition pathway and animal (rat, dog, monkey) CLiv were analyzed. All drugs were stratified based on their disposition pathway, liver extraction ratio (ERH) and ratios of unbound clearance to renal glomerular filtration rate (RGFR). Up to 22 AS methods were applied and compared in prediction of human CLiv to each group of drugs. 3. AS methods that give the best prediction of human CLiv, were identified for drugs primarily eliminated through liver with a fraction of renal elimination (frenal) within 0.3–0.5 or ERH > 0.3, where human CLiv of more than 80% or 90% drugs could be accurately (within 2- or 3-fold error) predicted. For drugs with ERH TSR,D resulting more than 60% or 75% drugs were predicted within 2- or 3-fold error. 4. By stratified analysis of drugs, according to their disposition pathway and organ extraction ratio, a unified strategy was developed to select the best AS method in prediction of human CLiv.

1. 针对特定药物群组筛选最优异速标度法（allometric scaling, AS）的统一策略，其开发工作至关重要。
2. 本研究共纳入446种已知人体肝脏清除率（CLiv）、明确药物处置通路，以及大鼠、犬、猴等动物肝脏清除率（CLiv）的药物进行分析。所有药物均依据其处置通路、肝脏提取率（ERH），以及游离清除率与肾小球滤过率（RGFR）的比值完成分层。针对每一组药物，共应用至多22种异速标度法开展人体肝脏清除率（CLiv）的预测，并对各方法的预测性能进行对比。
3. 针对主要经肝脏消除且肾消除分数（frenal）处于0.3~0.5区间，或肝脏提取率（ERH）大于0.3的药物，本研究明确了预测人体肝脏清除率（CLiv）效果最优的异速标度法；在此类药物中，超过80%或90%的药物的人体肝脏清除率（CLiv）可被精准预测，预测误差控制在2倍或3倍范围内。针对肝脏提取率相关指标（ERH TSR,D）符合要求的药物，其中60%或75%的药物的人体肝脏清除率（CLiv）预测误差可控制在2倍或3倍范围内。
4. 通过按照药物的处置通路与器官提取率开展分层分析，本研究构建了一套用于筛选人体肝脏清除率（CLiv）预测最优异速标度法的统一策略。