Comparing 3DCRT and inversely optimized IMRT planning for head and neck cancer: equivalence between step-and-shoot and sliding window techniques

Abstract

Background and purpose: To investigate the feasibility and the advantages of using Intensity-Modulated Radiotherapy (IMRT) for the treatment of head-and-neck cancer. Comparing different methods to deliver IMRT in this clinical setting.

Materials and methods: Seven patients (four radical; three post-operative), treated on a 6MV Varian Linac (equipped with an 80 leaves MLC) in accordance with a routine 3DCRT plan, were replanned. Original treatment plans were computed to irradiate a primary Planning Target Volume (PTV1, 54 Gy) and then to perform a boost on a PTV2 (radical: 70.2 Gy; post-operative: 64.8 Gy). IMRT dose plans were inversely-optimized using appropriate constraints with the Helios tool on a Varian Eclipse system. Once the optimal fluences were calculated, different modalities to deliver IMRT were considered: Sliding Window (SW) and Step and Shoot (SS) techniques using a different number of intensity levels to approximate the optimal fluences (e.g. 5, 10 and 20). Mean dose, maximum dose and a number of dose-volume parameters regarding CTV1, CTV2, PTV1, PTV2, OARs (spinal and planning spinal cord, parotids, optical structures, brain and temporal mandibular joint) were considered to compare the five modalities (3DCRT, SW, SS5, SS10, SS20); the Conformity Index (CI), the Irradiated Volume (IV) and the Treated Volume (TV) were also considered in the comparison.

Results: A more uniform coverage of the PTV in the IMRT dose plans with respect to the 3DCRT plan was found (for PTV2: V90% = 94.3 for 3DCRT, 97.6 for SS5, 98 for SS10 and 98.1 for SW; V107% = 20.7 for 3DCRT, 5.9 for SS5, 2 for SS10 and 1.3 for SW). Concerning OARs, they all present a significant reduction of mean and/or maximum dose and dose-volume patterns assessed from DVHs: in particular the mean dose of parotids decrease on average of about 13.5Gy passing from 3DCRT to IMRT with an average reduction of NTCP ranging from about 20% to more than 40% for radically treated patients, depending on the chosen end-point. IV and TV are also slightly smaller with IMRT. The results obtained with SS techniques employing 10 or more intensity levels are comparable with those obtained with SW; no differences between SS10 and SW may be appreciated when considering the DVHs of PTV, CTV and OARs. On the other hand, in some cases SS5 may be slightly sub-effective with respect to SS10-SW when considering PTV coverage and Dmax of the spinal cord.

Conclusions: With the Varian planning and delivery system, Step-and-shoot approximations of inversely optimised fluences in head-neck IMRT compare well with SW delivery, even with only five intensity levels. With a number of intensity level of 10 or more, no differences can be appreciated in PTV coverage/OAR sparing with respect to SW.