A comparison between volumetric modulated arc therapy and 3D conformal radiotherapy of stage 3 and 4 Carcinoma of the larynx

Abstract

The introduction of advanced radiation therapy equipment that has the ability to deliver modulated radiation therapy that can replace conventional three-dimensional conformal radiation therapy (3DCRT) methods means that comparisons are needed to understand the clinical impact for the patient. The radiation therapy treatment planning of stage 3 and 4 cancer of the larynx is challenging owing to the proximity of organs at risk (OAR) to the large tumour volume. The International Commission of Radiation Units and Measurements (ICRU) recommend that the dose to the Planning Target Volume (PTV) and the Planning Organ at Risk Volume (PRV) must be reported. The purpose of this research study is to determine the accuracy and reproducibility of the treatment set-up to enable the calculation of the Clinical Target Volume (CTV) to PTV and OAR to PRV margins that should be used for patients treated in the head and neck area. The study also compared the 3DCRT plans to the VMAT plans where the evaluation was based on the doses the OAR received and the conformity and homogeneity of the dose to the PTV. Method: In order to determine reproducibility of the treatment setup all patients, those who were treated over a period of 1 year with a thermoplastic mask, and who had a minimum of 5 image sets were studied. Thirty-three (33) patients were treated in both the head and neck area, and 36 patients were treated in the head area only. In each case the Digitally Reconstructed Radiography (DRR) created from the planning Computer Tomography (CT) images were compared to the weekly Electronic Portal Imager Devices’ (EPID) images. The van Herk’s equation was used to determine the expansion margins in 3 directions: superior to inferior, left to right and anterior to posterior. As these expansion margins were unknown at time of the study, the current departmental practice of a 5mm margin was applied for the treatment plan comparisons. The plan data sets of 10 patients with stage 3 and 4 cancer of the larynx was used to re-create comparative plans. As some OAR were omitted at the original planning process for the treated Volumetric Modulated Arc Therapy (VMAT) plans (labelled “RA treat”), a secondary RA plan was created where all the OAR was present (labelled “RA study”), and these plans were compared to the 3DCRT plans (labelled 3DCRT). Each of the 10 patient-data sets had 3 plan groups.

OAR criteria were recorded for the brainstem, spinal cord, parotids, cochlea, temporomandibular joints, oral cavity, and mandible. The homogeneity index (HI), conformity index (CI) and lesion coverage factor measurements were compared for all plan data sets to determine the impact of each planning technique to the dose to the PTV. Results: In terms of treatment accuracy and reproducibility 227 image sets were used to calculate the expansion margins needed of patients treated in both the head and neck area. The use of van Herk’s equation indicated that the CTV to PTV expansion was 5.6 mm in the anterior to posterior direction, 6.9 mm in the superior to inferior direction and 6.5 mm in the left to right direction. For those treated in only the head area, 273 image sets were evaluated and resulted in 5.1 mm in the anterior to posterior direction, 6.1 mm in the superior to inferior direction and 4.9 mm in the left to right direction. For the comparison between VMAT and 3DCRT, a custom plan score system was developed to record results between the planning techniques. The score for the doses to the OAR when comparing the 3DCRT technique to the RA treat group indicated equal scores. The 3DCRT compared to the RA study group, resulted in 8 (of the 10) patients in the RA study group achieving better OAR sparing. In the comparison of the RA treat and the RA study group, 7 of the 10 patients achieved better OAR sparing with the RA study group with 1 of the 10 patients an equal score. The score results for the PTV dose (coverage, homogeneity, and conformity) indicated no statistical significance between the RA study and RA treat groups. Comparing the 3DCRT group to the RA treat group, 9 of the 10 patients had worse dose results for the 3DCRT plans. Comparing the 3DCRT group to the RA study group, all 3DCRT plans scored worse for PTV dose quality. Conclusion: This study indicates that current PTV and PRV expansion margins are too small. As this information was previously unclear, the results of this study could be used as a baseline, and a tool to implement more rigorous checks and imaging protocols to lessen these margins. The results of the dose to normal tissue when comparing 3DCRT to VMAT plans did not indicate large differences, however the score results of the PTV dose, indicated that VMAT offers large improvements compared to the 3DCRT technique.