Radiobiological Impact of Calculation Slice Thickness on Head and Neck IMRT Plans Using MATLAB

In radiotherapy there are many evaluation tools available to achieve the best treatment plan. One of them is equivalent uniform dose (EUD) based model. It use to estimate the tumor control probability (TCP) and normal tissue complication probability (NTCP). While the current generation of radiobiological models has low predictive power that prevents it from being used as a primary evaluation tool, projections from radiobiological model may still be a helpful supplement to clinical experience. The perfect treatment plan provides the highest tumor control and lowest normal tissue complications. The purpose of this study is to use different grid sizes (2, 3, 4, 5) and different algorithms (Monte Carlo and Pencil Beam) when calculating TCP and NTCP. Eleven patients with head and neck (H&N) cancer cases were included in this study. Comparison achieved for each patient with the variation of grid size and algorithm. A total of 88 plans were generated in MONACO treatment planning system (TPS). Treatment plans were designed using Intensity modulated radiation therapy (IMRT) technique. Dose and volume parameters were derived from the dose volume histograms (DVHs) for target and critical structures. The average value of TCP was 94.13 ±12.80% for the 2 mm grid size and 95.16 ±10.05% for 5 mm with Monte Carlo (MC) algorithm. Statistically there was significance difference between two plans (p < 0.05). For Pencil Beam (PB) algorithm, the average TCP value was 91.78 ±19.54% and 93.04 ±17.13% for 2 mm and 5 mm respectively with p < 0.05. In comparison between MC and PB plans, the NTCP of PB algorithm plans were greater for brainstem, spinal cord, and chiasm compared to MC algorithm plans. It could be concluded that the smallest available grid size (2 or 3 mm) is the favorable. The MC algorithm is recommended for improved plan accuracy.