Physiologically Based Pharmacokinetic Model for Drugs Encapsulated into Liposomes

摘要 尽管脂质体在药物输送方面前景广阔，而且批准的脂质体药物数量也很多， 在开发这类药物及其后续版本方面仍然存在挑战。目前，这一领域正面临着 “论文多，药物少”的困境。造成这种情况的因素有很多。一个重要因素 是与脂质体的物理化学性质有关的开发复杂性， 差异可能导致体内性能和治疗功效的根本变化。另一个关键因素是 缺乏有效的战略来利用药物开发中不断增加的研究和知识。 开发的复杂性也给监管决策带来挑战。在这里，我们提出一个 生理学为基础的药代动力学（PBPK）建模策略，以解决这些挑战。这一战略 为脂质体药物开发了一个专门的PBPK平台，有望实现从 物理化学性质，以在体内性能，允许开发人员之间的早期相互作用， 监管机构，最终加快药物开发。这一战略的一个关键特点是将"模型"和 以动态和顺序的方式进行"实验"。一方面，PBPK模型提供了一个先验模拟， 指导实验设计;此外，实验产生新的知识来约束PBPK模型 这将有助于进一步的实验。选择脂质体多柔比星（Doxil®）作为模型药物。通用 模型将逐步发展成为一个专门的PBPK平台，沿着包括更具体的 通过严格的文献研究或额外的实验获得关于脂质体多柔比星的信息。五 具体目标是：目标1：构建基于库的脂质体药物通用PBPK模型以训练文献- 衍生的性能关系。目标是利用 文献，寻求脂质体的广泛性质-性能关系，然后训练这些关系 在一个基于库的通用PBPK模型，以探索潜在的关键质量属性（CQA）。目标二： 一系列聚乙二醇化脂质体阿霉素的制备、表征和体外研究。这些 将根据目标1中预计的CQA修改后的配方进行制备。本节将产生 更具体的知识，以进一步限制和改进通用PBPK模型。目的3：评估PK和 所需制剂的生物分布。制剂性质将与体内性能相匹配， 更新PBPK模型。目的4：定量脂质体在实体瘤中的不均匀分布， 的决定因素。在目标4之后，PBPK模型将完成从"制剂性质→系统"的级联 保留→异质性肿瘤分布→功效"。目标5：模型验证和翻译。模型 将通过基于机制的Monte Carlo模拟进行翻译。一个"虚拟" 将进行生物等效性研究。

ABSTRACT Despite the great promise held for liposomes in drug delivery and the number of liposomal drugs approved, challenges remain in development of such drugs and their follow-on versions. Currently, this field is confronting a dilemma of “so many papers and so few drugs”. A number of factors contribute to this. One significant factor is the development complexity pertaining to the physicochemical properties of liposomes, where a minute difference may result in a radical change in in vivo performance and treatment efficacy. Another critical factor is a lack of effective strategies to leverage the mounting research and knowledge in drug development. Development complexity also poses challenges to regulatory decision-making. Here we propose a physiologically-based pharmacokinetic (PBPK) modeling strategy to address these challenges. This strategy develops a specialized PBPK platform for liposomal drugs that is expected to enable effective translations from physicochemical properties to in vivo performance, allowing early interactions between developers and regulators, ultimately expediting drug development. A key feature of this strategy is to integrate “model” and “experiment” in a dynamic and sequential manner. On one side, PBPK models provide a prior simulation to guide experimental design; additionally, the experiment generates new knowledge to constrain PBPK models that will benefit further experimentation. Liposomal doxorubicin (Doxil®) is chosen as a model drug. A generic model will step-wisely grow into a specialized PBPK platform along with the inclusion of more specific information about liposomal doxorubicin either by rigorous literature research or additional experiments. Five Specific Aims are: Aim 1: Build a library-based generic PBPK model for liposomal drugs to train literature- derived property-performance relationships. The objective is to leverage the vast amount of knowledge in the literature, seeking broad property-performance relationships for liposomes and then training these relationships in a library-based generic PBPK model to explore potential Critical Quality Attributes (CQAs). Aim 2: Preparation, characterization, and in vitro studies of a series of PEGylated liposomal doxorubicin. These formulations will be prepared with modifications of the CQAs that are projected in Aim 1. This section will yield more specific knowledge to further confine and improve the generic PBPK model. Aim 3: Assess the PK and biodistribution of the desired formulations. The formulation properties will match with in vivo performance to update PBPK model. Aim 4: Quantify the heterogeneous distribution of liposomes in solid tumors and explore the determinants. After Aim 4, the PBPK model will complete a cascade from “formulation properties → system retention → heterogeneous tumor disposition → efficacy”. Aim 5: Model validation and translation. The model translation to human will be performed by mechanism-based Monte Carlo Simulation. A “virtual” bioequivalence study will be conducted.