| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Commentary |
The Bristol Haematology and Oncology Centre, United Bristol NHS Trust, Horfield Road, Bristol BS2 8ED, UK
A great deal has been written about portal imaging in radiotherapy in recent years. With the constant drive to improve survival rates/local control rates by the application of conformal techniques, intensity-modulated radiotherapy (IMRT) and dose escalation, there needs to be a greater emphasis placed on the reproducibility and verification of radiotherapy treatments. With tighter tumour margins and the introduction of IMRT comes the potential for geographical misses of part of the tumour volume when random and systematic set-up errors occur. Portal imaging to determine the accuracy of the placement of radiotherapy treatment fields has therefore become an essential part of practice.
As portal imaging technology evolves, with electronic portal imaging devices (EPIDs) more commonly used, the practice of treatment verification needs to change to exploit to the full the advantages that this technology provides. Many radiotherapy departments may still be following portal imaging protocols more suited to the use of film. However, most EPIDs, and particularly those using amorphous silicon panels, can acquire images with far fewer monitor units than would be required to produce a similar quality image on film. The Ionising Radiation (Medical Exposure) Regulation 2000 (IR(ME)R) [1] requires the justification of all exposure to ionising radiation, and therefore it is important that this advantage of EPIDs is maximized so that the number of monitor units per image is reduced, and thus the dose to the patient. This is particularly relevant when double exposures are used, as there is a small chance of inducing a cancer in the tissue encompassed by the large exposure outside of the tumour volume. In their study, Waddington and McKenzie [2] sought to quantify the effective dose per monitor unit to tissues outside of the target volume, as a result of concomitant exposures from planning and double exposure portal images, in order to determine an acceptable number of images that could be taken during the course of a patient's treatment. This information should be considered when implementing a new portal imaging regimen, as there is perhaps a tendency to increase the number of images acquired with little consideration given to the risks incurred as a consequence.
The impact on resources must also be considered when devising a portal imaging protocol. This is even more of an issue when images are assessed and interventions made on-line, such as in the study by Henry et al [3], in this journal. On-line corrections allow the adjustment of daily inconsistencies in patient position, or in this case organ motion. With appropriate patient immobilization random daily set-up errors due to patient movement should be small, thus lessening the need for daily imaging. However, prostate motion has been found to be significant and the use of bony landmarks to template match may give false confidence in reproducibility [4]. Prostate treatments form a large percentage of the work of most radiotherapy departments, and therefore the impact of on-line assessment and correction for organ motion is vast. With new cancer treatment targets upon us and the constant pressure on already stretched resources it is unfortunately likely that many radiotherapy departments will not be able to implement such protocols in the near future, particularly if extra time is needed for on-line assessment.
Henry et al [3] conclude that new portal imaging technology allows for speedier and more accurate assessment of images. EPIDs generate and display an image instantly, but the drawing of a template and matching to a reference image is not fast, although software providing automatic matching should speed up this process. However, template matching or measuring tools have been found to be more accurate than visual inspection, particularly when assessing small field placement errors [5].
The purpose of the study by Truong et al [6], also in this journal, was to determine the random and systematic deviations for the breast technique in their department. McKenzie et al [7] advocate the estimation of site-specific geometric uncertainties by each radiotherapy centre in order to minimize them and to inform protocols. These uncertainties will vary depending on "immobilization, treatment techniques, precision of radiographers, application of correction strategies" in each department. The magnitude of tolerances for tumour margins and portal imaging, and the frequency of imaging should be decided after considering these geometric uncertainties. There is little published evidence on setup deviations for locoregional radiotherapy, encompassing breast/chest wall and regional lymphatics. This paper should provide data for other radiotherapy centres to compare their own deviations and geometric uncertainties. With such a large proportion of a radiotherapy centre's workload taken up with treating women with breast cancer any changes to protocols will have a significant impact. Decisions regarding the use of three-dimensional planning, MRI and IMRT need to be related to improvement in cosmesis or local control and still feasible in terms of time and cost.
In both of the studies published in this journal radiographers performed the assessment and correction of displacements. Radiographers are ideally placed to extend their role in this area. Online corrections require instant assessments to be made at the time of treatment delivery, and therefore it is practical for radiographers to make these decisions. Perera et al [5] showed that when interpreting portal films, experienced radiographers made as good, if not better, decisions than oncologists. Radiographers working with EPIDs see a great many images and are perhaps more familiar with the accompanying software. Offline assessments also need to be carried out by staff with appropriate training and time allocated, bearing in mind that the more images taken the longer time needs to be spent assessing them. Therefore, provided clear protocols are used, appropriate decisions should be made.
Protocols developed in research centres may produce state of the art treatments with small volumes and margins that necessitate daily imaging and on-line correction, but in the real world of waiting lists, staff shortages and government targets this may not yet be achievable.
Received for publication June 16, 2005. Accepted for publication June 21, 2005.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| BJR | DMFR | IMAGING | ALL BIR JOURNALS |
