Adaptive radiation therapy of nasopharyngeal carcinoma using megavoltage computed tomography in helical tomotherapy

Abstract

Adaptive radiotherapy (ART) has been introduced in treating head and neck cancer recently. It refers to the restoration of the planned dose distribution by modifying the treatment plan according to the radiation-induced anatomic changes during a treatment course. ART in interval basis is feasible in head and neck cases because the organ changes in this region are shown to follow a progressive trend rather than in a random manner. Nevertheless, few studies have been carried out specifically on nasopharyngeal carcinoma (NPC), and the issue of when and how often the ART should be performed is still controversial. This study evaluated the potential dosimetric benefits of applying a three-phase adaptive radiotherapy protocol in NPC patients treated with helical tomotherapy. Comparison was made between the three-phase protocol and the non-adaptive single phase treatment protocol. Using daily megavoltage computed tomographic (MVCT) images provided by the tomotherapy system, this study also quantified and characterized the volumetric and geometric changes of tumors and organs at risk (OARs). An optimum ART strategy for NPC was therefore established by defining various thresholds that indicated significant anatomic changes during the treatment course and hence the needs for replanning. A retrospective study involving 30 NPC patients who had a total of 37-38 fractions of radiotherapy was conducted. Among them, 24 patients that followed the routine three-phase radiotherapy protocol were included in the first part of the study. Two new plans, denoted as PII-ART and PIII-ART, were generated based on the up-to-date CT images and contours and were used for treatment in phase two (PII; after 25th fraction) and phase three (PIII; after 35th fraction) respectively. To simulate the situation of no replanning, two hybrid plans denoted as PII-NART and PIII-NART were generated using the original contours pasted on the PII-and PIII-CT sets by CT-CT fusion. Dosimetric comparisons were made between the NART plans and the corresponding ART plans. For all 30 patients on their alternate day MVCT images, the posterolateral wall of nasopharynx (P-NP) which represented the nasopharyngeal gross tumor, neck volume covering the whole cervical spine level, bilateral parotid glands and vertebral canals at the level of 3 cm (VC3), 6 cm (VC6) and 9 cm (VC9) inferior to the base of skull were manually contoured. Pattern of their anatomic changes throughout the treatment course were being assessed. Statistical analysis was applied to each recorded parameter to define threshold(s), which indicated a significant change that replanning was suggested at any time point of the treatment course. This study showed that due to the radiation-induced tumor shrinkage took place within the initial planned high-dose region, the dose homogeneity to the targets was better even if no replanning was applied. Nevertheless, the significantly low degree of dose conformity of all target volumes without replanning revealed that there were many abutting normal tissues being unnecessarily irradiated with high dose due to the tumor shrinkage. In particular, some patients would have their brainstem, spinal cord and optic chiasm dose exceeding tolerance, inducing disastrous complications and hence degrading their quality of life after treatment. The present study demonstrated that replannings in between the treatment course restored the target dose conformity, hence allowed significantly better sparing of the surrounding OARs. In addition, ART strategy with multiple replannings was deem necessary to account the continuous anatomic changes over the radiotherapy course. The study also showed that the volumes of the P-NP, the parotid glands and the neck region demonstrated a progressive regression trend over time. For parotid glands, they tended to displace towards the medial and superior directions throughout the treatment course. Due to the decreased neck volume, the cervical cord was displaced backwardly during the treatment course, leading to the significant posterior shifting of the VC6 and VC9 in this study. Based on the found characteristic trends and the threshold occurrences in targets and OARs, an optimum ART strategy specifically for NPC cases was established. Three replans at 9th, 19th and 29th treatment fractions were proposed as most of the thresholds appeared in these time points of the treatment course. The present study emphasized the importance of implementing ART strategy in NPC cases by quantifying the dosimetric benefits of applying the three-phase adaptive radiotherapy protocol. The benefits mainly lied on limiting the dose to OARs than keeping the tumor coverage. This study also emphasized the necessity of multiple replanning strategy, in which an optimum ART strategy for NPC cases involving three replans was proposed. This proposed strategy can accommodate the dosimetric consequences due to anatomic deviation over the treatment course. It also enables prompt reaction to the regressed tumor by conforming the target dose, hence allowing safe target dose escalation without risking the surrounding OARs. This proposed strategy is clinical feasible with justified increase of departmental workload and should be promoted especially in centres where adaptive planning system is not yet available.