Development and Application of a Simplified Mathematical Model for The Estimation of Tumor Dose of Non-Small Cell Lung Cancer in BNCT.

Abstract

Lung cancer has exhibited a sustained increase in both incidence and mortality rates in recent years. This study aims to develop and apply a simplified mathematical model for estimating tumor dose in Boron Neutron Capture Therapy (BNCT) for non-small cell lung cancer (NSCLC), in order to facilitate clinical treatment planning. We analyzed the effects of six key parameters on dose distribution based on experimental data and established a simplified mathematical model to estimate tumor dose. The model was validated in screened lung cancer patients under both fixed and random irradiation conditions. Patients were categorized according to whether the dose limits of critical organs were exceeded, and the model's accuracy was further assessed in these subgroups. In addition, the Surface-to-Tumor center Distance (STcD) was defined to evaluate its correlation with the minimum prescribed dose for a single SBRT fraction. The study also investigated the effects of respiratory motion and setup errors on dose accuracy. The simplified mathematical model achieved estimated dose errors of 4.0% ± 2.8% under fixed irradiation and 4.4% ± 3.2% under random irradiation. For patients categorized based on critical organ dose limits, estimated errors were 9.3% ± 12.2% and 4.6% ± 9.8%, respectively. These results demonstrated the model's high accuracy. Analysis of STcD values revealed that 6.66 cm could serve as a reference threshold for patient selection to ensure the minimum prescribed dose. Respiratory motion introduced a dose estimation error within ±10%, and each 5 mm setup error caused less than 1% deviation, indicating that these effects are negligible. The proposed simplified mathematical model offered accurate and efficient tumor dose estimation for NSCLC patients undergoing BNCT. It served as a valuable reference in clinical practice, enabling physicians to rapidly assess tumor dose using medical imaging data, thus improving the efficiency and precision of treatment planning.