Radiotherapy dosimetry audit: three decades of improving standards and accuracy in UK clinical practice and trials.

Dosimetry audit plays an important role in the development and safety of radiotherapy. National and large scale audits are able to set, maintain and improve standards, as well as having the potential to identify issues which may cause harm to patients. They can support implementation of complex techniques and can facilitate awareness and understanding of any issues which may exist by benchmarking centres with similar equipment. This review examines the development of dosimetry audit in the UK over the past 30 years, including the involvement of the UK in international audits. A summary of audit results is given, with an overview of methodologies employed and lessons learnt. Recent and forthcoming more complex audits are considered, with a focus on future needs including the arrival of proton therapy in the UK and other advanced techniques such as four-dimensional radiotherapy delivery and verification, stereotactic radiotherapy and MR linear accelerators. The work of the main quality assurance and auditing bodies is discussed, including how they are working together to streamline audit and to ensure that all radiotherapy centres are involved. Undertaking regular external audit motivates centres to modernize and develop techniques and provides assurance, not only that radiotherapy is planned and delivered accurately but also that the patient dose delivered is as prescribed.

Dosimetry and field matching for radiotherapy to the breast and supraclavicular fossa.

PURPOSE: Early breast cancer radiotherapy aims for local disease control and reduced recurrence. Treatment is directed to breast or chest wall alone using tangential fields, or includes regional lymph nodes with a separate anterior field. The complex geometry of this region necessitates matching adjacent radiation fields in three-dimensions. Potential exists for overdosage or underdosage and cosmetic results may be compromised if fields are not accurately aligned. METHODS AND MATERIALS: A study of dosimetry across the match line region using different techniques, as reported in the multicentre START Trial Quality Assurance programme, was undertaken. A custom-made anthropomorphic phantom assessed dose distribution in three-dimensions using film dosimetry. RESULTS: Methods with varying degrees of complexity were employed for field matching. Techniques combined half beam blocking and machine rotations to achieve geometric alignment. Asymmetric beam matching allowed use of a single isocentre technique. Where field matching was not undertaken a gap between tangential and nodal fields was employed. Results demonstrated differences between techniques and variations for similar techniques in different centres. Geometric alignment techniques produced more homogenous dose distributions in the match region than gap techniques or those techniques not correcting for field divergence. CONCLUSIONS: Field matching techniques during the START trial varied between centres. Film dosimetry used in conjunction with a breast-shaped phantom provided relative dose information. The study highlighted difficulties in matching treatment fields to achieve homogenous dose distribution through the region of the match plane and the degree of inhomogeneity as a consequence of a gap between treatment fields.

A versatile phantom for quality assurance in the UK Medical Research Council (MRC) RT01 trial (ISRCTN47772397) in conformal radiotherapy for prostate cancer.

Within the UK RT01 trial the MRC also funded a quality assurance (QA) programme. This included a planning and dosimetry audit at participating centres using a purpose built phantom. Geometrical setup was visually assessed via field shaping around the phantom GTV (to within the order of 1 mm). Within the phantom, ion chamber positional uncertainties were estimated as 0.6 mm (95% CL, k=2). This was the basis for ion chamber measurements in a variety of dose gradients around the PTV closely simulating a patient case. The design provides a representative but reproducible system for QA in the prostate radiotherapy process, from CT scan to treatment. Setup errors are not eliminated, but minimised and estimated.

Comparison of two dimensional and three dimensional radiotherapy treatment planning in locally advanced non-small cell lung cancer treated with continuous hyperfractionated accelerated radiotherapy weekend less.

BACKGROUND AND PURPOSE: Patients with inoperable non-small cell lung cancer being treated with continuous hyperfractionated accelerated radiotherapy weekend less (CHARTWEL) were planned and treated with a three dimensional (3D) conformal protocol and comparison made with two dimensional (2D) planning, as used previously, to compare past practice and methods. PATIENTS AND METHODS: Twenty-four patients were planned initially using 3D and then replanned using a 2D system. The 2D plans were transferred onto the 3D system and recalculated. Dose volume histograms could then be constructed of planning target volumes for phases 1 and 2 (PTV 1 and 2, respectively), lung and spinal cord for the 2D plans and compared with the 3D plans. RESULTS: There was a significantly lower absolute dose to the isocentre with 2D compared to 3D planning with dose reductions of 3.9% for phase 1, 4.4% for phase 2 and 4.7% for those treated with a single phase. Maximum dose to spinal cord was greater in 17 of the 24 2D plans with a median dose reduction of 0.82 Gy for 3D (P=0.04). The percentage volume of whole lung receiving > or =20 Gy (V20) was greater in 16 of the 24 2D plans with a median reduction in V20 of 2.4% for 3D (P=0.03). CONCLUSIONS: A lower dose to tumour was obtained using 2D planning due to the method of dose calculation and spinal cord and lung doses were significantly higher.

Questionnaire based quality assurance for the RT01 trial of dose escalation in conformal radiotherapy for prostate cancer (ISRCTN 47772397).

BACKGROUND AND PURPOSE: In order to ensure the validity of the outcome of the Medical Research Council's 'RTO1 trial' of dose escalation in conformal radiotherapy for prostate cancer it was considered important that the quality of treatment delivery should meet an adequate standard across all contributing centres. A questionnaire was therefore devised to ensure that all aspects of the planning and delivery process were adequately covered. PATIENTS AND METHODS: The questionnaire considered each step in the planning and delivery process and drew the attention of the participants to the specific requirements of the trial. Before entering patients into the trial each participating centre had to complete the questionnaire and an outlining exercise (reported elsewhere). RESULTS: It was not practicable to define a detailed universally acceptable protocol for the whole process of delivery of conformal radiotherapy, not least because of the different equipment available for planning and treatment in different centres. The questionnaire identified some areas of difference in practice between centres where there may be a need for the development of a consensus as to best practice, particularly in the area of patient set-up. Occasionally it was necessary to follow up responses to questions that had been misunderstood or inadequately answered, but in most cases these issues proved to be easily resolved. CONCLUSIONS: The questionnaire proved to be a useful self-assessment tool as well as enabling the quality assurance group to ensure that the standards of the trial were being met. Subsequent follow-up visits confirmed the usefulness and validity of this self assessment process.

The use of in vivo thermoluminescent dosimeters in the quality assurance programme for the START breast fractionation trial.

BACKGROUND AND PURPOSE: The use of in vivo dosimetry for patient measurement is recommended in many publications. It provides an additional check to verify that the dose delivered to the patient corresponds to the prescribed dose. In the context of a clinical trial investigating the effects of different fractionation regimens, it is imperative that the dose given is that prescribed to ensure that noise in the data between centres does not mask the results of the trial. The methodology for in vivo measurement in a clinical trial of breast radiotherapy was developed and verified. MATERIALS AND METHODS: A cohort of patients in the STAndardisation of breast RadioTherapy (START) trial was monitored using postal thermoluminescent dosimeters chips (TLD). All TLD were processed and analysed at Mount Vernon Hospital. Patients for in vivo measurements were identified at randomisation as a random 1 in 9 samples for the first 2500 patients randomised (282 TLD) increasing to 1 in 3 thereafter. The TLD were left in place for the duration of the tangential field treatment and thus a composite entrance and exit dose was recorded. RESULTS: TLD measurements were performed on 429 patients from 33 hospitals. The average ratio of dose measured using TLD to that prescribed was 0.99+/-0.04. Eight patients had initial measurements more than 10% different to the prescribed dose. The mean TLD results for a given centre correlated well with dose measurements performed using an ionisation chamber in a breast shaped phantom at that centre as part of the START trial audit. CONCLUSION: Thermoluminescence dosimetry has provided useful quality assurance information on the doses received by patients in centres participating in the START trial.

Validation of active breathing control in patients with non-small-cell lung cancer to be treated with CHARTWEL.

PURPOSE: Active breathing control (ABC) was validated using patients with non-small-cell lung cancer (NSCLC) to be treated with continuous hyperfractionated accelerated radiotherapy weekend-less (CHARTWEL). Effects of breath hold (BH) on accuracy and normal tissue doses were evaluated. METHODS AND MATERIALS: Eleven patients were studied. Immediately after a free breathing (FB) planning scan, two ABC scans (ABC 1 and 2) were performed to assess intrafraction variation. A third ABC scan (ABC 3) was performed some weeks later to assess interfraction variation. Assisted BH was set at 75% of vital capacity and reproducibility assessed using computed tomography (CT) lung volumes. Planning target volumes (PTVs), doses to lung and spinal cord for FB and ABC 1 scans were compared. RESULTS: Results were available for 10 patients. Disease and elective nodal regions were easier to define on ABC scans making PTVs smaller. ABC lung volumes showed no significant variation over several weeks, percentage volume of whole lung receiving > or =20 Gy (V(20)) was reduced in all (median 6.4%, p = 0.005), and spinal cord dose in 80% (median 1.03 Gy, p = 0.02), of the plans. CONCLUSION: ABC allowed reproducible BH, and enabled better delineation of tumor and normal structures, as well as reduction in PTV, V(20), and spinal cord dose.

Three-dimensional distribution of radiation within the breast: an intercomparison of departments participating in the START trial of breast radiotherapy fractionation.

PURPOSE: To examine the ability of computer planning systems to calculate the dose to the breast correctly in three dimensions. Both the absolute dose at the center of the breast and the accuracy of the isodose distributions were investigated. METHODS AND MATERIALS: Measurements were performed in a water-filled breast phantom using an ionization chamber. Thirty-six sets of data obtained during the Standardization of Breast Radiotherapy breast fractionation trial quality assurance program were included in the analysis. The planning systems were grouped according to the algorithms used on the basis of the definitions given in International Commission on Radiation Units and Measurements Report No. 24. RESULTS: Thirty-two of the 36 planning systems overestimated the dose to the center of the breast, with a mean measured/calculated dose ratio of 0.979 (SD 0.013). The relative dose within 2 cm of the lung was also overestimated. CONCLUSION: Only one algorithm (collapsed cone) investigated in this study was able to calculate the dose at the center of the breast correctly in tangential breast radiotherapy. With modern algorithms, it is important to include a correction for the lower density of the lung, because the dose close to the interface between breast and lung tissue will also be lower than anticipated.