ABSTRACT Animal models are the backbone of many areas of biomedical and preclinical research and the mouse is the most used model organism in human disease research. Understanding and characterizing mouse models is considered key to improving the reproducibility of preclinical results in human subjects. To this end it is important to use preclinical techniques that are analogous to modern clinical techniques to increase the success of results that translate well to clinical implementation. Unfortunately, the standardization of preclinical radiation therapy (RT) studies using clinically analogous methodologies has not yet been achieved. In the last few decades clinical RT technology has developed dramatically, with intensity modulated RT (IMRT) representing one of the most significant developments. However, these advancements have not translated to preclinical small animal RT, which largely resembles the state of clinical RT in the 1970s in terms of its use of simple circular/rectangular beam geometries and uniform beam intensities, and lack of high-quality 3D target dose conformality. Consequently, small animal RT studies do not simulate the radiobiological, radioimmunological, hypoxic and toxicity environment of human therapies. To fully standardize preclinical irradiators to clinically analogous methodologies it is necessary to fully implement IMRT. However, reverse translation of IMRT from a clinical to preclinical scale (1-2 orders of magnitude smaller) has been challenging due to technical limitations in engineering appropriately small multi-leaf collimators (MLC). Our group has recently developed a 3D printed compensator (3DPC) approach that provides a simple, reliable, and cost-effective solution in implementing mouse based IMRT. In terms of simplicity and cost- effectiveness, 3DPC-IMRT requires no additional specialized equipment besides an inexpensive commercially available 3D printer and printing material. Unlike MLC systems, 3DPC-IMRT has no segments/steps during delivery leading to shorter treatment times and no mechanical/electrical parts leading to high reliability, lower maintenance, and reduced QA efforts. Our central hypothesis is that we can develop 3DPC-IMRT into a high- throughput preclinically useful small animal IMRT system using AI assisted contouring, novel high speed dose calculation and optimization algorithms, and a mechanism that will allow rapid switching of compensators for different fields. We propose the following specific aims: (1) Design and development of 3D printed compensator IMRT (3DPC-IMRT), (2) Development of VMAT like 3DPC-IMRT and AI mouse segmentation methods, and (3) Dosimetric accuracy evaluation and pancreatic ductal adenocarcinoma (PDAC) mouse model study. The success of the proposed project will help radiation biology research better simulate the clinical RT environment and enable future studies where accurate complex dose distributions are critical.