A Comparison of IMRT and VMAT Treatment Planning for Head and Neck Cancer With NTCP/TCP Analysis

Abstract

Intensity-modulated radiation therapy (IMRT) and volumetric intensity-modulated arc therapy (VMAT) for HN cancers have been used with comparable coverage of planning target volumes (PTV). However, there is controversy over superiority of these techniques in management of HN cancers due to proximity to organs at risk (OAR), beam time, monitor units (MU) and integral dose. Using conventional dosimetric parameters conformity and homogeneity index (CI, HI), and biological indices NTCP/TCP, we evaluated VMAT and IMRT in their ability to provide superior coverage of PTV while sparing OAR. Fourteen consecutive HN cancer patients treated to the bilateral neck were planned with both IMRT and VMAT techniques in a commercial treatment planning system with 100% coverage to 95% PTV volume to a dose of 7000cGy. If OAR dose constraints were met for both VMAT and IMRT plans, a VMAT plan was preferred for treatment. The CI, HI, MU, gradient, as well as NTCP/TCP values were calculated. OAR studied include spinal cord, brachial plexus, bilateral parotids, constrictors, and larynx. Maximum doses for serial organs and mean doses for parallel organs were calculated for each OAR. NTCP model parameters were taken from QUANTEC and calculated for larynx, parotid, and spinal cord. TCP was modeled with published parameters based on a Poisson model with repopulation and sublethal damage. Statistical comparisons between VMAT and IMRT parameters were made using the paired t-test. The TCP values ranged 70-85% and were not significantly different (P>0.05) in IMRT compared to VMAT. NTCPs for the larynx and left/right parotid did not differ significantly between IMRT and VMAT plans. The PTV HI and CI were not significantly different between plans. Based on OAR coverage, six plans were treated with IMRT instead of VMAT. The OARs that were better spared in these six IMRT plans were constrictors (N=3), oral cavity (N=1), parotid (N=1) and brachial plexus (N=1). Significantly more MUs were used in IMRT plans than VMAT plans (average difference 733 MU, p<0.001). Maximum doses in VMAT plans were significantly greater than IMRT plans (mean max-dose difference 98cGy, p = 0.001). VMAT plans had an average maximum cord dose 1800cGy lower than IMRT plans (p < 0.001). IMRT delivered significantly lower mean dose to the oral cavity (mean dose difference 288cGy, p = 0.003) and middle/inferior constrictors (mean dose difference 121cGy, p = 0.02; 220cGy, p = 0.03, respectively). There was no significant difference between techniques in mean dose delivered to larynx or parotids or maximum dose delivered to the brachial plexus. IMRT and VMAT techniques generated comparable coverage of PTV and resulted in similar TCP. VMAT showed a significant reduction of MU and maximum dose delivered to spinal cord. Selection of techniques should be based on a more nuanced clinical decision rather than purely physical parameters.