There is growing interest in relaxing homogeneity constraints in IMRT treatments for solid tumors that do not contain normal tissue. Knowledge-based dose predictions can improve plan quality, but depend on historical data typically incorporating homogeneity constraints. An analytical TCP and NTCP model could be used to estimate optimal target dose heterogeneity prior to plan optimization by assuming basic properties of the dose distribution, thereby improving prospective plan quality. We derived and validated an analytical model of TCP and NTCP based on the properties of co-planar VMAT including target dose heterogeneity, beam penumbra, and dose spillage outside of the target. We derived the TCP/NTCP model with input parameters including tumor size, tumor edge dose (prescription dose), tumor center dose, fractionation, and estimates of standard radio-biological parameters such as tumor clonogen density, radio-sensitivity mean and variance, and Lyman-Kutcher-Burman (LKB) NTCP model parameters. To validate the model, we created a virtual phantom incorporating a spherical tumor with 2 cm diameter and produced a series of VMAT plans in a treatment planning system with varying central target dose between 108% and 262% of the prescription while minimizing dose spillage outside of the target. Differential DVHs were exported and used to estimate TCP and NTCP using the Marsden and LKB models assuming a radio-resistant tumor prescribed 60 Gy in 30 fractions completely surrounded by a single cylindrical parallel organ. The TCP and NTCP were also calculated for these test cases using the analytical model, and compared to the values calculated using the treatment planning system data. The target heterogeneity and corresponding TCP and NTCP values estimated by the analytical model and planning system based approaches are summarized in Table 1. The errors in the TCP and NTCP estimates of the model were within 0.04 and 0.01, respectively, for all plans, despite not requiring a dose calculation. The analysis also demonstrated a steady increase in TCP with increased central target dose, and a transient decrease in NTCP. A central target dose of 111% minimized NTCP, and a central target dose of 124% maximized TCP while providing the same or lower NTCP as a homogeneous target dose. We have derived an analytical model of TCP and NTCP based on simple tumor and dose characteristics that can be applied immediately following contouring to estimate optimal target dose heterogeneity. This tool can be used to improve the dose objectives used in prospective clinical trials. We will present results following model validation in multiple clinical scenarios and demonstrate patient use cases.Abstract 3822; Table 1Central Target Dose (% of Rx)TCPNTCPPinnacleModelPinnacleModel1080.680.640.0500.0481110.720.680.0490.0471240.800.790.0500.0471610.920.930.0530.0522050.960.970.0590.0592340.980.980.0570.0632490.990.980.0580.0642621.000.980.0560.065 Open table in a new tab