Assessment of Intrafraction Motion of the Urinary Bladder Using Magnetic Resonance Imaging (cineMRI).

AIM: To assess the intrafraction motion of the urinary bladder and delineate the appropriate margin size for radiotherapy planning, for both the full and empty bladder. MATERIALS AND METHODS: This was a single-site, single-arm study of 20 patients planned to undergo radical cystectomy for histologically confirmed muscle-invasive bladder cancer. Patients underwent magnetic resonance imaging (cineMRI) of the entire pelvis using a 3-Tesla system, prior to cystectomy. Patients first underwent a cineMRI with a full bladder, then voided and underwent a second MRI with an empty bladder. All MRI sequences were acquired over 18 min. We assessed the differences in bladder filling and subsequent bladder wall displacement, between the empty and full bladder, during a time period consistent with radiotherapy treatment delivery. RESULTS: Twenty patients underwent cineMRI of the entire pelvis. The maximum mean directional displacements of the bladder walls over the 18 min duration of the scan for the empty bladders were 9.8 mm superiorly, 1.1 mm inferiorly, 2.39 mm anteriorly, 3.73 mm posteriorly, 2.74 mm to the left and 2.48 mm to the right. The maximal mean displacements for the full bladders were 9.2 mm superiorly, 1.1 mm inferiorly, 2.28 mm anteriorly, 1.08 mm posteriorly, 1.85 mm to the left and 1.73 mm to the right. Statistically significant differences were seen in the posterior, left and right displacements but were quantitatively small. CONCLUSIONS: Intrafractional motion secondary to bladder filling showed minimal variation between the full and empty bladder. Similar clinical target volume to planning target volume margins can be applied for the delivery of radiotherapy for a full and empty bladder.

DOSE AND GAMMA-RAY SPECTRA FROM NEUTRON-INDUCED RADIOACTIVITY IN MEDICAL LINEAR ACCELERATORS FOLLOWING HIGH-ENERGY TOTAL BODY IRRADIATION.

Production of radioisotopes in medical linear accelerators (linacs) is of concern when the beam energy exceeds the threshold for the photonuclear interaction. Staff and patients may receive a radiation dose as a result of the induced radioactivity in the linac. Gamma-ray spectroscopy was used to identify the isotopes produced following the delivery of 18 MV photon beams from a Varian 21EX and an Elekta Synergy. The prominent radioisotopes produced include 187W, 63Zn, 56Mn, 24Na and 28Al in both linac models. The dose rate was measured at the beam exit window (12.6 µSv in the first 10 min) following 18 MV total body irradiation (TBI) beams. For a throughput of 24 TBI patients per year, staff members are estimated to receive an annual dose of up to 750 µSv at the patient location. This can be further reduced to 65 µSv by closing the jaws before re-entering the treatment bunker.

Calculating geometrical margins for hypofractionated radiotherapy.

Formulae to calculate margins for external beam radiotherapy traditionally treat the radiotherapy regimen for analysis purposes as consisting of a very large number of treatment sessions, each delivering a very small dose. It is assumed that the difference in the calculated margin for an infinitely fractionated regimen, and the margin for real world treatment scenarios with finite fractions per patient, is negligible given the usual large number of fractions used to treat most malignancies. However hypofractionated radiotherapy using five fractions or even fewer is becoming increasingly common, causing the accuracy of traditional margin formulae to break down. This work introduces an algorithm that accurately calculates margins for hypofractionated treatment regimens. A method for estimating an upper limit for the required margin is introduced, and an adjustment to the van Herk formula (van Herk et al (2000 Int. J. Radiat. Oncol. Biol. Phys. 47 1121-35)) is used to provide a lower limit. A single-term mathematical model is then used to interpolate between these limits. Results are compared to those calculated by Monte Carlo simulation, demonstrating that the proposed method yields accurate estimates of the actual margin required (consistently within ~1%) for wide ranges of values of the systematic error, random error and penumbral width, even when the number of fractions per patient is as low as 2.

Cost minimisation analysis: kilovoltage imaging with automated repositioning versus electronic portal imaging in image-guided radiotherapy for prostate cancer.

AIMS: To compare the treatment time and cost of prostate cancer fiducial marker image-guided radiotherapy (IGRT) using orthogonal kilovoltage imaging (KVI) and automated couch shifts and orthogonal electronic portal imaging (EPI) and manual couch shifts. MATERIALS AND METHODS: IGRT treatment delivery times were recorded automatically on either unit. Costing was calculated from real costs derived from the implementation of a new radiotherapy centre. To derive cost per minute for EPI and KVI units the total annual setting up and running costs were divided by the total annual working time. The cost per IGRT fraction was calculated by multiplying the cost per minute by the duration of treatment. A sensitivity analysis was conducted to test the robustness of our analysis. Treatment times without couch shift were compared. RESULTS: Time data were analysed for 8648 fractions, 6057 from KVI treatment and 2591 from EPI treatment from a total of 294 patients. The median time for KVI treatment was 6.0 min (interquartile range 5.1-7.4 min) and for EPI treatment it was 10.0 min (interquartile range 8.3-11.8 min) (P value < 0.0001). The cost per fraction for KVI was A$258.79 and for EPI was A$345.50. The cost saving per fraction for KVI varied between A$66.09 and A$101.64 by sensitivity analysis. In patients where no couch shift was made, the median treatment delivery time for EPI was 8.8 min and for KVI was 5.1 min. CONCLUSIONS: Treatment time is less on KVI units compared with EPI units. This is probably due to automation of couch shift and faster evaluation of imaging on KVI units. Annual running costs greatly outweigh initial setting up costs and therefore the cost per fraction was less with KVI, despite higher initial costs. The selection of appropriate IGRT equipment can make IGRT practical within radiotherapy departments.

Assessment of leakage doses around the treatment heads of different linear accelerators.

Out-of-field doses to untargeted organs may have long-term detrimental health effects for patients treated with radiotherapy. It has been observed that equivalent treatments delivered to patients with different accelerators may result in significant differences in the out-of-field dose. In this work, the points of leakage dose are identified about the gantry of several treatment units. The origin of the observed higher doses is investigated. LiF:Mg,Cu,P thermoluminescent dosimetry has been employed to quantify the dose at a several points around the linac head of various linear accelerators (linacs): a Varian 600C, Varian 21-iX, Siemens Primus and Elekta Synergy-II. Comparisons are also made between different energy modes, collimator rotations and field sizes. Significant differences in leaked photon doses were identified when comparing the various linac models. The isocentric-waveguide 600C generally exhibits the lowest leakage directed towards the patient. The Siemens and Elekta models generally produce a greater leakage than the Varian models. The leakage 'hotspots' are evident on the gantry section housing the waveguide on the 21-iX. For all machines, there are significant differences in the x and y directions. Larger field sizes result in a greater leakage at the interface plate. There is a greater leakage around the waveguide when operating in a low-energy mode, but a greater leakage for the high-energy mode at the linac face. Of the vendors investigated, the Varian 600C showed the lowest average leakage dose. The Varian 21-iX showed double the dose of the 600C. The Elekta Synergy-II had on average four times the dose leakage than the 600C, and the Siemens Primus showed an average of five times that of the 600C. All vendors show strong differences in the x and y directions. The results offer the potential for patient-positioning strategies, linac choice and shielding strategies to reduce the leakage dose to patients.

Finding the optimal statistical model to describe target motion during radiotherapy delivery--a Bayesian approach.

Early approaches to characterizing errors in target displacement during a fractionated course of radiotherapy assumed that the underlying fraction-to-fraction variability in target displacement, known as the 'treatment error' or 'random error', could be regarded as constant across patients. More recent approaches have modelled target displacement allowing for differences in random error between patients. However, until recently it has not been feasible to compare the goodness of fit of alternate models of random error rigorously. This is because the large volumes of real patient data necessary to distinguish between alternative models have only very recently become available. This work uses real-world displacement data collected from 365 patients undergoing radical radiotherapy for prostate cancer to compare five candidate models for target displacement. The simplest model assumes constant random errors across patients, while other models allow for random errors that vary according to one of several candidate distributions. Bayesian statistics and Markov Chain Monte Carlo simulation of the model parameters are used to compare model goodness of fit. We conclude that modelling the random error as inverse gamma distributed provides a clearly superior fit over all alternatives considered. This finding can facilitate more accurate margin recipes and correction strategies.

Thermoluminescence dosimetry for skin dose assessment during intraoperative radiotherapy for early breast cancer.

Dosimetry for intraoperative radiotherapy (IORT) after wide local excision for breast cancer using a 50 kV X-ray needle (Intrabeam) was performed in vivo using thermoluminescence dosimetry. Eight LiF:Mg,Ti chips were placed on the skin around the incision site after wide local excision while the tumour bed was irradiated to a prescribed dose of 5 Gy 10 mm from the applicator surface. The maximum and mean measured skin dose for 57 patients ranged from 0.64 to 7.1 Gy and 0.56 to 4.78 Gy, respectively, reflecting different tissue thicknesses overlying the applicator. The average maximum dose of 2.93+/-1.46 Gy was below the threshold for severe radiation skin toxicity.

Continuing professional development needs of Australian radiation oncology medical physicists--an analysis of applications for CPD funding.

In November 2004, the Australian federal government allocated $775,000 to individual Australian radiation oncology medical physicists (ROMPs) to access continuing professional development (CPD) activities. The funding was administered by the Australasian College of Physical Scientists and Engineers in Medicine (ACPSEM). In order to receive funding, individuals had to submit an application to ACPSEM, which assessed each application and distributed funds to successful applicants. 248 separate applications were received from 143 individuals in two rounds of applications. Information from the applications was collated and analysed, with the aim of identifying patterns that will be of use in future planning for CPD. This paper presents a summary of the information extracted from the analysis.

Radiation quality of a tomotherapy photon fan beam.

Tomotherapy, a novel radiotherapy technique, uses narrow fan beams for cancer patient treatment. Photon energy spectra for a rectangular 10 x 1 cm2 photon beam were analyzed in central axis and penumbra regions at depths of 3 to 10 cm in a water phantom. A 6 MV beam of a Varian 2100C/D Linear Accelerator was modeled using BEAM99 Monte Carlo calculations to simulate energy transport in a water phantom. Arrays of 4 x 2 mm2 scoring regions were arranged to cover the central axis and penumbra areas. Radiation quality factors were calculated based on dose-mean linear energy transfer. Although there appears to be a trend towards higher quality factor values in the penumbra area, this change is fairly small, at most 3% in penumbra region. We conclude that change in radiation quality is not likely to be an issue in a tomotherapeutic approach when 6 MV x rays are used.

Junctioning of lateral and anterior fields in head and neck cancer: a dosimetric assessment of the monoisocentric technique (including reproducibility).

PURPOSE: The matching or junctioning of two lateral fields with an anterior field is commonly performed in the treatment of head and neck cancer. A monoisocentric technique utilising asymmetric collimation is potentially associated with improved dosimetry in the plane of the junction due to decreased reliance on operator skill and the avoidance of couch movement. The aim of this study was not only to assess the average dose delivered in the plane of the junction, but also the reproducibility of this dose for the monoisocentric technique and two other commonly used techniques. METHODS AND MATERIALS: An anthropomorphic head and neck wax phantom was fashioned to allow the placement of 22 TLD chips in a 2-mm thick transverse plane positioned superior to the potential site of the larynx. Three different treatment techniques were used with the phantom being treated by four different operators a minimum of 20 times for each technique: (1) "straight fields"--using isocentric laterals with an anterior field junctioned in the midline. This technique makes no allowance for divergence; (2) "angled fields"--couch and gantry rotation are used to account for divergence; (3) "monoisocentric"--using asymmetric collimators to create a single isocenter. RESULTS: For an applied dose of 1 Gy the monoisocentric technique produced a mean dose measured of 1.01 Gy compared with 1.23 and 0.92 Gy for techniques 1 and 2. The reproducibility of the mean dose measured was better for the monoisocentric technique by a factor of 2. The superior reproducibility of the monoisocentric technique was not found to be operator dependent. CONCLUSIONS: A monoisocentric technique for the treatment of two laterals and an anterior field in head and neck cancer is likely to be associated with more accurate and reproducible dosimetry in the plane of the junction. Our center has subsequently adopted this technique for matching such fields.

Assessment of mucosal underdosing in larynx irradiation.

PURPOSE: Mucosal underdosing as a result of electron disequilibrium at the air cavity may affect local recurrence rates for T1 and T2 larynx cancers. Secondary build-up properties of high-energy beams have been demonstrated in a slab phantom. It was the aim of this investigation to determine whether significant surface underdosing exists for the mucosa under clinical conditions. METHODS AND MATERIALS: Measurements were made using a thermoluminescent dosimetry (TLD) extrapolation technique in an anatomic larynx phantom. The larynx phantom was constructed using tissue and cartilage equivalent material, based on patient cross-sectional anatomy. Three different thicknesses of LiF ribbons, 0.14, 0.39, and 0.89 mm, were placed reproducibly at 12 different positions at the anterior, posterior, and lateral walls on the endolarynx surface. Measured doses were plotted and an extrapolation was made back to the mucosal depth to obtain the dose received at each of the positions. Results were obtained for two different field configurations, opposed laterals and oblique fields, for 6-MV X rays and opposed lateral fields from a telecesium unit. In addition, the larynx surface doses of field sizes from 4 x 6 cm2 to 7 x 6 cm2 were investigated. RESULTS: Surface underdosing was observed owing to the secondary build-up and build-down effect of the air cavity, and the dose measured for the three extrapolation TLDs at any position varied by up to 18%. An average variation of 6% was observed. The surface underdosing was most apparent for the 6-MV opposed lateral beam technique, where mucosa doses down to 76% of the prescribed dose were observed. Mucosal underdosing at the measurement positions was less marked with oblique techniques, telecesium treatment, and increasing field size. CONCLUSION: Because of underdosing, some surface positions receive < 80% of the prescribed dose. This may contribute to the potential for higher recurrence rates observed with high-energy photons.

A Monte Carlo technique to establish the water/tissue equivalence of phantom materials.

The quality of phantom materials is crucial for accurate dosimetry in radiotherapy. A wide range of factors such as density, electron density and elemental composition can influence the radiation properties, and hence the absorbed dose, of materials. New materials can be tested by direct measurements which requires considerable time and the availability of relatively large amounts of the material. Alternatively, the dosimetric properties of a proposed phantom material can be compared to those of water or tissue using Monte Carlo calculations. The aim of this study was to evaluate the use of a Monte Carlo technique for the investigation of the water/tissue equivalence of phantom materials. The material used for this investigation was Standard Dosimetry Agarose (SDA) gel, which is useful in MRI dosimetry. Depth doses in gel and water were calculated for mono-energetic electron beams of 6, 12 and 20 MeV, and photon beams of 60 keV and 6 MV. For each radiation quality the depth dose distributions are in close agreement.

Dosimetry of 6-MV x-ray beam penumbra.

The measurement of x-ray beam dose profiles in the penumbral region, using silicon diode, ionization chamber, TLD, and film dosimetry, has been investigated for a 6-MV beam defined by independent collimators. Penumbral width (80%-20%) at dmax, as measured by diode, film, and TLD was found to be 3.6, 3.6, and 3.4 mm, respectively. These results reflect the relative sensitive widths of each of the measurement systems (2.5, 2.0, and 1.0 mm, respectively). An empirical forming function was used to relate the penumbral shape measured with a finite-sized detector to that which would be measured with a point detector, the width of the point detector penumbra calculated from the diode penumbra is 3.4 mm, indicating that the TLD rods are a good approximation to a point detector. An alternative method of determining the width of a point detector penumbra is to extrapolate the penumbral widths obtained using two or more detectors of sensitive width. With this method, using Farmer and RK ionization chambers, a point detector penumbra width of 3.1 mm is obtained. An EGS4 Monte Carlo simulation, where a point source was assumed, gave a penumbral width of 2.8 mm. Negligible differences between the penumbra of beams defined by symmetric and asymmetric collimators was observed.