Novel protein therapeutics (biologics) represent an emerging and important class of anticancer drugs, as they can modulate a diverse array of oncology targets, and assume a plethora of sizes and functionality. Since each `biologic-target' combination is unique, the choice of the biologic for development against the desired target is crucial, and can determine the fate of the development effort. However, this decision is mostly made empirically without knowing if the biologic is capable of achieving the desired exposure at the site-of-action, and differentially target tumor over host tissues. Consequently, most `target-biologic' combinations encounter efficacy or toxicity related failures during in vitro-to-in vivo o preclinical-to-clinical translation. Our long term goal is to alleviate this translational failuresby developing physiologically based pharmacokinetic (PBPK) models that can a priori predict the systemic, tissue, and tumor pharmacokinetics (PK) of a diverse array of biologics in several preclinical species and human. It is hypothesized that once a quantitative relationship between various properties of the biologics and physiological determinants for their absorption, distribution, metabolism, and elimination (ADME) is established, it will lead to the development of integrated mathematical models capable of a priori predicting the PK of biologics. In this proposal we will test this hypothesis by developing PBPK models for: (i) different size biologics, (ii) antibody-drug conjugates (ADCs), and (iii) T-cell engaging bi-specific molecules.
In Aim#1, published data on the PK and biodistribution of different size targeted or non-targeted biologics in various animals and human will be collected, and similar data will be generated in-house in tumor-bearing mice. A quantitative relationship between the size of biologics and their ADME determinants will be established, and integrated into an established PBPK model for monoclonal antibodies (mAbs). The augmented PBPK model will be used to characterize collected data, and will be evaluated for its a priori PK prediction and human translation ability.
In Aim#2 published data on the PK and biodistribution of different ADCs and their components (i.e. mAb and drug) will be collected, and similar in-house data will be generated in tumor-bearing mice. The datasets will be used to build a single PBPK model for ADCs and their components, whose PK prediction and human translation capabilities will be evaluated.
In Aim#3 a tumor antigen and human T-cell targeting bi-specific molecule will be generated, and its biodistribution studies will be conducted in tumor and human T-cells bearing mice. Published PBPK model for T-cells will be implemented in-house, and combined with the PBPK model for biologics from Aim#1, to develop a PBPK model for bi-specific molecules in mice. The model will be tested for human translatability using published clinical data from a similar bi-specific molecule. Thus, this proposal will generate broadly applicable novel systems PK models for anticancer biologics, which would have a significant impact on the discovery, development, and translation of these molecules, across academia and pharmaceutical industries.
Many protein based anticancer agents, either do not reach the tumor cells at the sufficient levels or reach to the non-tumor tissues in access. Hence, they either fail to kill the tumor cells or cause unacceptable toxicity, leading to their failure in the clinic. This project will develop mathematics based computational tools that can beforehand predict the levels of anticancer protein therapeutics in animal or human tumors and non-tumor tissues, thereby allowing scientists to make informed go/no-go decisions regarding the development of these agents. Thus, the resultant tools are expected to reduce clinical failure rates of anticancer protein therapeutics.
Showing the most recent 10 out of 13 publications
