Variation in Lung Tumour Breathing Motion between Planning Four-dimensional Computed Tomography and Stereotactic Ablative Radiotherapy Delivery and its Dosimetric Implications: Any Role for Four-dimensional Set-up Verification?

AIMS: To investigate variation in tumour breathing motion (TBM) between the planning four-dimensional computed tomograph (4DCT) and treatment itself for primary or secondary lung tumours undergoing stereotactic ablative radiotherapy (SABR). MATERIALS AND METHODS: Sixteen consecutive patients underwent planning 4DCT at least 1 week after implantation of a fiducial marker. The maximal extent of breathing motion of the intra-tumoural fiducial was measured at 4DCT and again at delivery of each SABR fraction on the linac using stereoscopic kilovoltage imaging. Displacements of the fiducial beyond planned limits were measured in three dimensions and represented as vectors. Variation in breathing motion between the planning 4DCT and treatment, and between individual SABR fractions was analysed. RESULTS: Although TBM at treatment exceeded planned tumour motion limits for at least part of the course for all patients, 31% of patients remained consistently within 1 mm, 50% within 2 mm and 69% consistently within 3 mm of planned parameters. However, 19% of patients experienced TBM variation 5 mm or more beyond planned limits for at least one fraction. For all patients, the median displacement vector at treatment beyond the planned motion envelope was 1.0 mm (mean 2.0 mm, range 0-12.7 mm). Variation in TBM at treatment from 4DCT correlated neither with the magnitude of TBM at 4DCT nor with planning target volume size (rs = 0.13, P = 0.62; rs = 0.02, P = 0.94, respectively). Nor was TBM variation related to tumour type or lobar position (P = 0.35, P = 0.06, respectively). Inter-fraction TBM variation was modest, with an average standard deviation of 1.7 mm (0.3-8.7 mm). CONCLUSIONS: TBM variation between 4DCT and treatment and between SABR fractions was modest for most patients. However, 19% of patients experienced significant TBM variation that could be clinically relevant for those most severely affected. It seems prudent to carry out on-couch assessment of TBM at each SABR fraction to identify such patients who might benefit from respiratory gating or adaptive radiotherapy to maintain tumour motion within the planned limits.

Clinical accuracy of ExacTrac intracranial frameless stereotactic system.

PURPOSE: In this paper, the authors assess the accuracy of the Brainlab ExacTrac system for frameless intracranial stereotactic treatments in clinical practice. METHODS: They recorded couch angle and image fusion results (comprising lateral, longitudinal, and vertical shifts, and rotation corrections about these axes) for 109 stereotactic radiosurgery and 166 stereotactic radiotherapy patient treatments. Frameless stereotactic treatments involve iterative 6D image fusion corrections applied until the results conform to customizable pass criteria, theirs being 0.7 mm and 0.5° for each axis. The planning CT slice thickness was 1.25 mm. It has been reported in the literature that the CT slices' thickness impacts the accuracy of localization to bony anatomy. The principle of invariance with respect to patient orientation was used to determine spatial accuracy. RESULTS: The data for radiosurgery comprised 927 image pairs, of which 532 passed (pass ratio of 57.4%). The data for radiotherapy comprised 15983 image pairs, of which 10 050 passed (pass ratio of 62.9%). For stereotactic radiotherapy, the combined uncertainty of ExacTrac calibration, image fusion, and intrafraction motion was (95% confidence interval) 0.290-0.302 and 0.306-0.319 mm in the longitudinal and lateral axes, respectively. The combined uncertainty of image fusion and intrafraction motion in the anterior-posterior coordinates was 0.174-0.182 mm. For stereotactic radiosurgery, the equivalent ranges are 0.323-0.393, 0.337-0.409, and 0.231-0.281 mm. The overall spatial accuracy was 1.24 mm for stereotactic radiotherapy (SRT) and 1.35 mm for stereotactic radiosurgery (SRS). CONCLUSIONS: The ExacTrac intracranial frameless stereotactic system spatial accuracy is adequate for clinical practice, and with the same pass criteria, SRT is more accurate than SRS. They now use frameless stereotaxy exclusively at their center.

Stereotactic fields shaped with a micro-multileaf collimator: systematic characterization of peripheral dose.

Despite the highly localized doses that may be delivered via stereotactic radiotherapy, a small dose is nonetheless delivered to out-of-field regions, which may cause detriment to the patient. In this work, a systematic set of dose measurements have been undertaken up to a distance of 45 cm from the isocentre, for stereotactic fields shaped by a BrainLAB mini-multileaf collimator (MMLC) mounted on a Varian 600C linear accelerator. A range of treatment parameters were varied so as to determine the factors of greatest influence and establish relationships with dose. The commercial treatment planning software (TPS) miscalculates the dose to out-of-field regions. Measured dose decreases consistently out to 45 cm, whereas the TPS decreases out to 10-15 cm, at which point the predicted dose is constant. At the 5-10 cm off-axis distance (OAD), measurements indicate doses of about 5-10% of the dose at the isocentre, 1% at 15 cm OAD and 0.1% at 45 cm OAD. There are several observed trends. Greater MMLC field sizes (with static jaw) result in higher out-of-field dose, as do shallower depths. The source-to-surface distance does not greatly influence peripheral dose. However, the results given in this work do indicate that simple treatment arrangements, such as preferable collimator rotation, would in certain cases reduce out-of-field dose by an order of magnitude. Peripheral dose raises questions of treatment optimization, particularly in cases where patients have a long life expectancy in which secondary effects may become manifest, such as in the treatment of paediatric patients or those with a non-malignant primary. For instance, for a 20 Gy hypo-fractionated treatment, dose to out-of-field regions is of the order of cGy-a substantial dose in radiation protection terms.

Radiotherapy DICOM packet sniffing.

The Digital Imaging and Communications in Medicine (DICOM) standard is meant to allow communication of medical images between equipment provided by different vendors, but when two applications do not interact correctly in a multi-vendor environment it is often first necessary to demonstrate non-compliance of either the sender or the receiver before a resolution to the problem can be progressed. Sometimes the only way to do this is to monitor the network communication between the two applications to find out which one is not complying with the DICOM standard. Packet sniffing is a technique of network traffic analysis by passive observation of all information transiting a point on the network, regardless of the specified sender or receiver. DICOM packet sniffing traps and interprets the network communication between two DICOM applications to determine which is non compliant. This is illustrated with reference to three examples, a radiotherapy planning system unable to receive CT data from a particular CT scanner, a radiotherapy simulator unable to print correctly on a DICOM printer, and a PACS unable to respond when queried about what images it has in its archive by a radiotherapy treatment planning system. Additionally in this work it has been proven that it is feasible to extract DICOM images from the intercepted network data. This process can be applied to determine the cause of a DICOM image being rendered differently by the sender and the receiver.

Stereotactic dose perturbation from an aneurysm clip measured by Gafchromic EBT film.

Aneurysm clips within stereotactic treatment volumes enhance spatial accuracy but perturb the dose distribution. The dose perturbations caused by a standard titanium alloy aneurysm clip (Ti6Al4V) have been measured with Gafchromic EBT film. The maximum dose perturbation was an increase of 6 % within 0.5 mm of the beam entry surface of the clip, and a decrease of 7 % within 0.5 mm of the beam exit surface of the clip. Results also showed perturbations to film readout due to the presence of micro dust particles on the film affecting optical properties at high spatial resolution (21um) scanning. Special procedures should be used when film is immersed in water, dried and then read at high spatial resolution. We recommend that films should be immersed only in distilled water and tools such as canned air puffs should be used to clean films without scratches.

Determination of dosimetric perturbations caused by aneurysm clip in stereotactic radiosurgery using gel phantoms and EBT-Gafchromic films.

Some radiotherapy patients are treated with titanium surgical aneurysm clips in the radiation field. This is of particular importance for stereotactic radiosurgery brain treatments, where the length of the blade of the clip may be comparable to the size of the radiation field. This study seeks to determine the extent of the dosimetric effects caused by surgical clips in stereotactic radiosurgery, using polyacrylamide gel phantoms and EBT type Gafchromic films. Using gel phantoms scanned with magnetic resonance imaging scanner, dose enhancement of around 20% was noted at distances less than 2 mm away from the clip surface. Gafchromic films showed about 6% variations in the dose up to few millimeters from the clip. These experimental results confirmed results predicted by Monte Carlo simulation techniques for higher density material surgical clips such as lead and platinum. Moreover, these experimental measurements clearly indicate dose reduction due to radiation attenuation behind the clip of about 4%.

Comparison of CT and positron emission tomography/CT coregistered images in planning radical radiotherapy in patients with non-small-cell lung cancer.

Imaging with F-18 fluorodeoxyglucose positron emission tomography (PET) significantly improves lung cancer staging, especially when PET and CT information are combined. We describe a method for obtaining CT and PET images at separate acquisitions, which allows coregistration and incorporation of PET information into the radiotherapy (RT) planning process for non-small-cell lung cancer. The influence of PET information on RT planning was analysed for 10 consecutive patients. Computed tomography and PET images were acquired with the patient in an immobilization device, in the treatment position. Using specially written software, PET and CT data were coregistered using fiducial markers and imported into our RT planning system (Cadplan version 6). Treatment plans were prepared with and without access to PET/CT coregistered images and then compared. PET influenced the treatment plan in all cases. In three cases, geographic misses (gross tumour outside planning target volume) would have occurred had PET not been used. In a further three cases, better planning target volume marginal coverage was achieved with PET. In four patients, three with atelectasis, there were significant reductions in V20 (percentage of the total lung volume receiving 20 Gy or more). Use of coregistered PET/CT images significantly altered treatment plans in a majority of cases. This method could be used in routine practice at centres without access to a combined PET/CT scanner .

High-resolution measurements of small field beams using polymer gels.

Small field sizes are increasingly becoming important in radiotherapy particularly since the introduction of intensity-modulated radiation therapy (IMRT) techniques. It is normally a challenging task to reliably measure the delivered dose and to determine its distribution in a medium for such small fields using conventional-type dosimeters such as gas ionisation chambers. Recently, attempts have been made to use films, but they are not tissue equivalent, they measure the dose only in two dimensions and they are not as responsive to radiations. In the present work, polyacrylamide gel (PAG) dosimeters are employed to measure the dose and its distribution in three dimensions for very small field sizes, such as those typically used in stereotactic radiosurgery. Field sizes of 6 x 6 and 18 x 18 mm in width are investigated. The results show an agreement with radiochromic film and ionisation diode measurements, with some variation in measured doses near the edge of the field, where the gel data decreases more rapidly than the other methods.

Stereotactic radiosurgery planning with ictal SPECT images.

This paper is motivated by a clinical requirement to utilise ictal SPECT images for target localisation in stereotactic radiosurgery treatment planning using the xknife system which only supports CT and MRI images. To achieve this, the SPECT images were converted from raw (pixel data only) format into a part 10 compliant DICOM CT fileset. The minimum requirements for the recasting of a raw format image as DICOM CT or MRI data set are described in detail. The method can be applied to the importation of raw format images into any radiotherapy treatment planning system that supports CT or MRI import. It is demonstrated that the combination of the low spatial resolution SPECT images, depicting functional information, with high spatial resolution MRI images, which show the structural information, is suitable for stereotactic radiosurgery treatment planning.

Discrepancies in volume calculations between different radiotherapy treatment planning systems.

It has been determined that, contrary to expectation, there is a clinically significant variation in the volume calculations of different RTPS (Radiotherapy Treatment Planning System) for identical contours. The situation was investigated prior to a multi-centre trial to determine whether tumour volume is an independent prognostic factor in NSCLC (non-small cell lung cancer) and included four of the commercially available RTPS. The four RTPS tested were, Theraplan Plus V3.0, Cadplan V6.2, Focus V2.6 and ADAC V3.0. Five randomly chosen clinical target related volumes (3 GTVs, one PTV and one CTV) from the trial database originally marked on the Cadplan system were transferred to the other four systems and the resulting volumes were calculated. It was found that Cadplan consistently underestimated the volume relative to the other three systems by 6-12%. This systematic underestimation was found to be caused by different assumptions made by the Cadplan system about the axial outer slice extension of the volume. Cadplan truncates the volume, while the other three systems extrapolate it by half the slice thickness at each end. A short program was written to apply the same method of volume extension to the Cadplan volume that is utilised by the other systems. This produced calculated volumes that were within +/- 1.0% of the average of the volumes calculated by the other three planning systems, and the maximum deviation from the average for any planning system was then reduced to 1.5%. This program was implemented at all participating trial centres utilising Cadplan, thus reducing the intersystem variability to a negligible factor in comparison to the estimates of inter-physician variation. This unexpected finding has significant implications for the validity of multi-centre trials using dose volume histograms, and indeed the adoption of any clinical protocol employing dose volume histogram constraints derived from experience at another centre employing a different RTPS.

Display of positron emission tomography with Cadplan.

Recent clinical experience at Peter MacCallum Cancer Institute (PMCI) with the use of unregistered Positron Emission Tomography (PET) images for radiotherapy target marking in the lung suggests that co-registered PET images would be invaluable. PMCI has three radiotherapy treatment planning systems but none of them currently is able to display or co-register PET images with Computed Tomography (CT) images. This paper details the approach taken to display co-registered PET images with the CADPLAN treatment planning system. CT Image files are normally transferred to Cadplan by DICOM transfer, but the Cadplan DICOM server will not receive (has no presentation context for) PET images. The fundamental design of the CADPLAN system envisages display of only a single image dataset, which must be a CT scan for planning reasons. The problem of data transfer is crudely solved by File Transfer Protocol (FTP) over the network. Fortunately the multislice format of the PET image files makes individual transfer manageable. A menu based C program running at the same time as Cadplan is invoked to sample the DICOM PET Image and create multiple Cadplan CART image format files that are co-registered with each existing transverse CT slice. With the Cadplan in contour mode, the program allows the co-registered PET images to be swapped in and out of the image section of the CART files promptly, while keeping the contour information. This allows radiotherapy target volumes to be marked using transverse PET emission images, and effectively circumvents the design constraints prohibiting the display of more than one image set. Contours can be over-laid for review on reconstructed sagittal or coronal views of CT or PET images constructed using the standard Cadplan tools. Co-registration is facilitated by identical positioning with the aid of lasers and FDG loaded fiducial markers on the PET scanner and CT couch. A polyurethane cast fixed with EFFILOCK is used to ensure identical patient orientation on the CT and PET couches. Since both imaging modalities are without significant geometric distortion the co-registration is then simply a translation. PET transmission images can be used for co-registration verification. The practical implementation of display of PET images with CADPLAN has enabled us to begin a trial of 10 patients, the results of which will be reported separately.

Implementation of enhanced dynamic wedge using Theraplan.

The article discusses how to use the standard utility program on Theraplan to create physical wedges which are equivalent to each of the enhanced dynamic wedges. The user must first create segmented treatment tables for each wedge, using a spreadsheet, and then convert these tables to equivalent thicknesses of an arbitrary wedge material. These thicknesses are then supplied to Theraplan. The paper discusses the agreement achieved between Theraplan and measured data, and the quality assurance procedures which should be adopted.

Comparative evaluation of Wellhöfer ion chamber array and Scanditronix diode array for dynamic wedge dosimetry.

Normalised profiles have been measured using the Scanditronix diode array and the Wellhöfer ion chamber array for the Varian dynamic wedge. Agreement was of the order of 0.1% of central axis peak dose for an open beam at depth, 0.3% for a dynamic wedge field at depth, and up to 0.6% at the peak depth. The use of the arrays for data acquisition is discussed, including user interface limitations. Data reproducibility is determined to be of the order of 0.1% for both systems. The issue of beam hardening within dynamic wedges is discussed and resolved in terms of the dose-gradient effect. A method for interpolation between dynamic wedge profiles using open beam data is presented that allows construction of isodoses to an estimated accuracy of 0.7%. Finally a benchmark for comparison of different measuring systems based on quality assurance requirements for the enhanced dynamic wedge is suggested.

Levels of leakage radiation from electron collimators of a linear accelerator.

The leakage radiation from electron applicators used with our linear accelerator has been measured. For the applicators 6 X 6 to 25 X 25 cm size, the leakage was measured in the plane of the patient and on the sides of the applicators with the available electron energies of 6, 9, 12, 15 and 18 MeV. The levels were significant. The highest leakage on the side was for the combination of 6 X 6-cm applicator and 9-MeV electrons (32%) and in the plane of the patient for 25 X 25-cm applicator with 18 MeV (10%) relative to the peak dose. Adding lead 1-2 mm, at appropriate locations inside the applicators has reduced the leakages to acceptable levels without affecting the beam parameters.