Abstract: The dose-averaged linear energy transfer (LET_d) in proton therapy is primarily computed using either analytical methods or Monte Carlo (MC) simulations. Analytical approaches are computationally efficient but limited in accuracy, whereas MC methods provide high precision at the cost of substantial computation time. In this study, we developed a hybrid analytical LET_d calculation method based on precomputed MC lookup tables, aiming to achieve a balance between accuracy and efficiency. A three-dimensional LET_d dataset in water for all discrete proton energies used in the treatment planning system was generated using the MCsquare Monte Carlo simulation platform and employed as the foundational lookup table for the proposed method. During LET_d calculation, the patient CT images were first converted into water equivalent ratio (WER) maps. For each voxel, the LET_d value was obtained by referencing the precomputed dataset according to its lateral distance from the beam’s central axis and the water equivalent length traversed along the beam incidence direction. The final three-dimensional LET_d distribution was then generated through dose-weighted accμmulation across all contributing spots. The proposed method was evaluated using five clinical lung cancer cases and compared against LET_d calculations from OpenTPS. Agreement between the D·LET_d distributions was assessed using 3%/3 mm three-dimensional Gamma analysis, yielding a passing rate of 96.68±1.65%. On a workstation equipped with an AMD EPYC 9374F CPU and an NVIDIA RTX A5000 GPU, the average computation time was 36 s, significantly shorter than that of OpenTPS (P=0.0313). These results demonstrate that the proposed fast LET_d calculation method can substantially reduce computation time while maintaining high accuracy. This provides a practical foundation for assessing LET_d distributions in targets and organs at risk (OARs), and supports future biologically informed treatment plan optimization in proton therapy.