On board cone beam CT for treatment planning in image guided radiotherapy

摘要

Background: Movement of tumours between or during radiotherapy treatment fractions poses a risk to surrounding healthy tissues and potentially lowers the treatment dose to the intended area. To increase the efficacy of radiotherapy, radiation oncologists utilise image-guided radiotherapy (IGRT) to enhance the delivery of radiation to cancerous tumours. Concern about concomitant radiation doses and poor quality images have previously limited the use of such technology when developing treatment plans for adaptive radiotherapy. Recent improvements to the On-board Imager (OBI; Varian version 1.4) including expansion of the number of acquiring modes from four to six, have rejuvenated efforts to use Cone Beam Computed Tomography (CBCT) with OBI as a radiotherapy treatment planning tool. Aim: This research aimed to investigate the possibility of using the new version of the Varian On-Board CBCT imager Vl.4, for adaptive radiotherapy. This work has led to the development of a methodology on how to initiate and implement CBCT scans for - - - - - -- -. -- - -- .- -- the purpose of increasing the accuracy of radiotherapy treatments using adaptive radiotherapy. Methods: The adaptation of radiotherapy plans using CBCT scan images involved three stages. CBCT concommitant doses were determined in the first stage by measuring the dose received by three types of phantom; the RANDO anthropomorphic phantom, the computer-imaging reference system phantom (CIRS) and cylindrical water phantoms of varying diameter. Two- and three-dimensional simulations were also obtained for CBCT using EXCEL, and Monte Carlo codes (BEAMnrc and DOSXYZnrc). The manufacturer's schematic diagram of the head was used to simulate a detailed CBCT dose simulation with the effect of beam output and bow-tie filter included as dose-modifiers. Based on these dose measurements, relationships between CBCT concomitant dose and patient size were found. In addition, estimations of secondary induced cancer were modelled based on these doses. In the second stage, CBCT scan calibrations were conducted. The relationship IV Abstract between the Hounsfield Unit (HU) and electron density (ED) of CBCT scans were described mathematically for each CIRS-062A phantom configuration. Later, these CBCT HU-to-ED calibrations were benchmarked against the CT RU-to-ED relationship of GE lightspeed CT employed in treatment planning. Finally, in the third stage, the obtained HU-to-ED calibrations were applied to treatment plans calculated on CIRS and RANDO phantoms using single-beam and IMRT configurations. Dose calculations derived from the OBI CBCT were compared with those from the GE Lightspeed CT. Results:Using a female RANDO phantom, doses were lowered by factors of36, 8,22 and 16, at the eyes, oesophagus, thyroid and brain, respectively, when using the new version ofVarian CBCT vl.4. In both the standard dose head mode and pelvis mode, the concomitant dose at all positions decreases as the phantom size increases. The concomitant dose measured on the smallest cylindrical water phantoms (10cm in diameter) resulted in a theoretical risk of secondary skin cancer of 0.005% in the standard dose mode and 0.05% in the pelvis mode, assuming a 30-fraction course of ---- -treatmentwith CBCT images acquired on a daily basis. Importantly, these-doses are - approximately 10 times greater than those measured for the largest phantom. The risk of secondary cancer for this phantom size at the oesophagus, thyroid, and brain sites are 0.0443, 0.0106 and 0.0439 % respectively for 30 daily images of head and neck treatment. Dose calculations on both the CIRS and RANDO phantoms showed that for the single beam treatment, only 1 % difference in the mean dose values are delivered to the majority of insertions when using the original CT or CBCT images and respective calibration curves. The only exception was for dense bone, which exhibited a 2% difference. For the IMRT treatment plan results showed that when the CT scan image is used the mean doses were less than 1.1 %. Conclusion: CBCT doses from the OBI version 1.4 are significantly lower than doses from version 1.3, making it possible to use CBCT to assist with adaptive radiotherapy on a daily basis, without a significantly increased secondary cancer risk. This technology is a useful tool to aid patient positioning for radiotherapy and to allow v Abstract VI daily adaptive IGRT. Radiation dose varies significantly with both patient size and tumour position in relation to scanning mode. It is therefore recommended that patient-specific imaging protocols be considered, especially with regard to paediatric patients who can be expected to receive a higher dose. The single beam and the WRT comparisons showed that the CBCT images and calibration curves can be used in treatment planning.