Analysis of modulation factor to shorten the delivery time in helical tomotherapy

Abstract A low modulation factor (MF) maintaining a good dose distribution contributes to the shortening of the delivery time and efficiency of the treatment plan in helical tomotherapy. The purpose of this study was to reduce the delivery time using initial values and the upper limit values of MF. First, patients with head and neck cancer (293 cases) or prostate cancer (181 cases) treated between June 2011 and July 2015 were included in the analysis of MF values. The initial MF value (MF initial) was defined as the average MF actual value, and the upper limit of the MF value (MFUL) was defined according the following equation: MFUL = 2 × standard deviation of MF actual value + the average MF actual Next, a treatment plan was designed for patients with head and neck cancer (62 cases) and prostate cancer (13 cases) treated between December 2015 and June 2016. The average MF actual value for the nasopharynx, oropharynx, hypopharynx, and prostate cases decreased from 2.1 to 1.9 (p = 0.0006), 1.9 to 1.6 (p < 0.0001), 2.0 to 1.7 (p < 0.0001), and 1.8 to 1.6 (p = 0.0004) by adapting the MF initial and the MFUL values, respectively. The average delivery time for the nasopharynx, oropharynx, hypopharynx, and prostate cases also decreased from 19.9 s cm−1 to 16.7 s cm−1 (p < 0.0001), 15.0 s cm−1 to 13.9 s cm−1 (p = 0.025), 15.1 s cm−1 to 13.8 s cm−1 (p = 0.015), and 23.6 s cm−1 to 16.9 s cm−1 (p = 0.008) respectively. The delivery time was shortened by the adaptation of MF initial and MFUL values with a reduction in the average MF actual for head and neck cancer and prostate cancer cases.

Helical tomotherapy (HT) is a delivery technique that modulates dose intensity using multileaf collimators (MLCs) of 64 leaves while synchronizing with the gantry rotation. 1 The field width in the superior-inferior direction of a patient is 5.0 cm at maximum; therefore, delivery time increases in cases with long target lengths. We have previously shown that delivery time decreases by adjusting parameters for dose optimization computing of the treatment plan. 2 When a small value is set as the modulation factor (MF), that is one of the parameters, delivery time shortens; however, a small MF value results in poorer dose distribution. [2][3][4] Therefore, it is necessary to set MF with a good balance of the delivery time and dose distribution. Because the proper setting of MF values varies across facilities and treatment sites, 5 it is difficult to maintain a balance. A method has been proposed to search and adopt a lower setting of MF value while maintaining a good dose distribution by repeating the dose optimization computing with a lower setting of MF value for the completed treatment plan. However, some treatment planning system of tomotherapy (Accuray, Inc.) is not equipped with a graphics processing unit (GPU), and without GPU, the system takes more time for the dose optimization computing; thus, it is not effective to use this method for each patient. If a low MF value with good dose distribution maintenance is designed at the beginning of a treatment plan, the delivery time will be shortened and the treatment plan will be more efficient. We determined the optimal initial MF value by retrospective analysis of MF values used in the past. In addition, the upper limit of the MF value was used to avoid a larger setting of MF value than required. The purpose of this study was to reduce the delivery time by the initial value and upper limit value of MF.

2.A | MF
MF is an index that expresses the complexity of the MLC motion.
MF is defined by the following equation with the only beamlet (a radiation that passes an opened leaf) being used in the dose optimization computing: show normal distribution, so the average of MF actual was defined as initial MF value (MF initial ). The treatment plans of half of the overall cases could be statistically approved by the use of MF initial ( Fig. 1(a)).
In addition, the value that added double of the standard deviation of MF actual value to the average MF actual was defined as the upper limit of the MF value (MF UL ). Treatment plans of 97.5% of cases could be statistically approved by the use of MF UL ( Fig. 1   Our method effectively reduced MF actual values, and the average delivery time for nasopharynx, oropharynx, hypopharynx, and prostate cases decreased for 3.2 s cm À1 , 1.1 s cm À1 , 1.3 s cm À1 , and 6.7 s cm À1 respectively.

| CONCLUSIONS
Here, we defined an initial value and an upper limit value using a retrospective analysis of MF. The delivery time was shortened by the adaptation of these values with a reduction in the average MF actual for head and neck cancer and prostate cancer cases.