Definition of parameters for quality assurance of flattening filter free (FFF) photon beams in radiation therapy.

PURPOSE: Flattening filter free (FFF) beams generated by medical linear accelerators have recently started to be used in radiotherapy clinical practice. Such beams present fundamental differences with respect to the standard filter flattened (FF) beams, making the generally used dosimetric parameters and definitions not always viable. The present study will propose possible definitions and suggestions for some dosimetric parameters for use in quality assurance of FFF beams generated by medical linacs in radiotherapy. METHODS: The main characteristics of the photon beams have been analyzed using specific data generated by a Varian TrueBeam linac having both FFF and FF beams of 6 and 10 MV energy, respectively. RESULTS: Definitions for dose profile parameters are suggested starting from the renormalization of the FFF with respect to the corresponding FF beam. From this point the flatness concept has been translated into one of "unflatness" and other definitions have been proposed, maintaining a strict parallelism between FFF and FF parameter concepts. CONCLUSIONS: Ideas for quality controls used in establishing a quality assurance program when introducing FFF beams into the clinical environment are given here, keeping them similar to those used for standard FF beams. By following the suggestions in this report, the authors foresee that the introduction of FFF beams into a clinical radiotherapy environment will be as safe and well controlled as standard beam modalities using the existing guidelines.

Lessons Learnt from Past Incidents and Accidents in Radiation Oncology.

The purpose of this report is to review and compile what have been and can be learnt from incidents and accidents in radiation oncology, especially in external beam and brachytherapy. Some major accidents from the last 20 years will be discussed. The relationship between major events and minor or so-called near misses is mentioned, leading to the next topic of exploring the knowledge hidden among them. The main lessons learnt from the discussion here and elsewhere are that a well-functioning and safe radiotherapy department should help staff to work with awareness and alertness and that documentation and procedures should be in place and known by everyone. It also requires that trained and educated staff with the required competences are in place and, finally, functions and responsibilities are defined and well known.

Be aware of neutrons outside short mazes from 10-MV linear accelerators X-rays in radiotherapy facilities.

During the radiation survey of a reinstalled 10-MV linear accelerator in an old radiation treatment facility, high dose rates of neutrons were observed. The area outside the maze entrance is used as a waiting room where patients, their relatives and staff other than those involved in the actual treatment can freely pass. High fluence rates of neutrons would cause an unnecessary high effective dose to the staff working in the vicinity of such a system, and it can be several orders higher than the doses received due to X-rays at the same location. However, the common knowledge appears to have been that the effect of neutrons at 10-MV X-ray linear accelerator facilities is negligible and shielding calculations models seldom mention neutrons for this operating energy level. Although data are scarce, reports regarding this phenomenon are now emerging. For the future, it is advocated that contributions from neutrons are considered already during the planning stage of new or modified facilities aimed for 10 MV and that estimated dose levels are verified.

A phase I/II evaluation of metoclopramide as a radiosensitiser in patients with inoperable squamous cell carcinoma of the lung.

The feasibility of administering metoclopramide (MCA) as a radiosensitizer has been evaluated in 23 patients with a pathological or cytological diagnosis of a squamous cell carcinoma of the lung, clinically evaluated as inoperable. All patients received 40-60 Gy radiotherapy fractionated into 1.8 Gy fractions 5 times per week (Monday-Friday). Two MCA treatment regimens were used: (i) MCA at 2 mg/kg administered by intravenous-infusion 1-2 h prior to radiotherapy 3 times per week (Monday, Wednesday, Friday); and (ii) MCA at 1 mg/kg administered by intravenous infusion 1-2 h prior to radiotherapy 5 times per week (Monday-Friday). 11 of the 23 patients treated with radiotherapy and MCA had none to mild pneumonitis or fibrosis and another 8 of the 23 had moderate levels. No patient had their therapy interrupted due to radiation-related side-effects. The MCA-related side-effects were as expected, i.e. 78% of the patients experienced sedation/tiredness and 48% expressed restlessness/anxiety symptoms. Both the total dose and serum levels of MCA were significantly associated to the MCA side-effect profile. Tumour response, duration of tumour response and survival were significantly positively correlated to the total and weekly doses of MCA administered to the patients during their radiotherapy treatment. These favourable phase II data have justified the initiation of a phase II/III randomised multicentred trial being carried out in Europe to evaluate MCA as a radiosensitiser.

Volumetric and dosimetric evaluation of radiation treatment plans: radiation conformity index.

PURPOSE: The use of conformal radiation therapy has grown substantially during the last years since three-dimensional (3D) treatment planning systems with beams-eye-view planning has become commercially available. We studied the degree of conformity reached in clinical routines for some common diagnoses treated at our department by calculating a radiation conformity index (RCI). METHODS AND MATERIALS: The radiation conformity index, determined as the ratio between the target volume (PTV) and the irradiated volume, has been evaluated for 57 patients treated with 3D treatment plans. RESULTS AND CONCLUSION: The RCI was found to vary from 0.3 to 0.6 (average 0.4), a surprisingly low figure. The higher RCI is typical for pelvic treatments (e.g., prostate) and stereotactic treatments. The lower RCI is found for extended tumors, such as mammary carcinomas where the adjacent nodes are included. The latter is also valid for most lung cancer patients studied. The RCI gives a consistent method for quantifying the degree of conformity based on isodose surfaces and volumes. Care during interpretation of RCI must always be taken, since small changes in the minimum dose can dramatically change the treated volume.

Design and dosimetry characteristics of a soft-docking system for intraoperative radiation therapy.

PURPOSE: The design concept and the dosimetric characteristics of an applicator system for intraoperative radiation therapy (IORT) with special emphasis on alignment methods, the effect of a plastic scatterer in the beam, radiation leakage, and misalignment dosimetry, are presented in this paper. MATERIALS AND METHODS: A soft-docking system for a linear accelerator, which enables collimation of electron beams (4-22 MeV) for IORT has been developed. The system includes twenty-one circular polymethylmethacrylate (PMMA) treatment cones of different lengths, diameters and end angles. All in-water measurements are made using p-type silicon diode detectors. RESULTS: The effect of introducing a PMMA scatterer in the therapeutic beam includes increased surface dose values (above 83% for all nominal electron energies and for all cones) and improved dose homogeneity within the therapeutic range. Electrons scattered from the inside wall of the cone result in dose profile horns at depth of dose maximum always lower than 109%. The radiation leakage outside the cone is less than 13%. Large changes in the dose profiles occur if the intraoperative cone is misaligned more than 0.5. CONCLUSION: The alignment procedure of the soft-docking system is easy to handle and the applicator design provides adequate collimation of electron beams for IORT.

A method for conversion of Hounsfield number to electron density and prediction of macroscopic pair production cross-sections.

A method for the determination of electron density using a narrow beam attenuation geometry is described. The method does not require that the elemental composition of the phantom materials is known. The Hounsfield numbers for the phantom materials used were determined using five different CT scanners. A relationship between Hounsfield number and electron density can thus be established, which is of considerable value in radiation therapy treatment planning procedures. Measurements of the ratio coherent/incoherent scattering of low energy photons in a certain geometry has proven valuable for determination of atomic number, which in its turn can be used for estimation of macroscopic pair production coefficients for high energy photons. The combination of knowledge of electron density with methods for determination of processes, dependent on atomic number, can form a base for adequate composition of phantom materials for purposes of testing dose calculation algorithms for photons and electrons.

Comparison of measured and calculated absorbed doses from tangential irradiation of the breast.

Calculated absorbed dose distributions from tangential irradiation of the breast have been compared with TLD measurements in a anthropomorfic body-shaped phantom which has gone through all phases of the radiation treatment planning cycle, including mapping of electron densities with a CT scanner, simulation of beam set-up and several treatments with an accelerator. The absorbed doses, measured in the breast are 2-6% lower than those calculated with the clinically used treatment planning system. The deviation is slightly higher when wedges are used. The main source for this deviation is shown to be the limitations in the loss-of-scatter correction in the dose calculation algorithm used. It has also been noted that the variation of the absorbed dose in the target volume is in general smaller than predicted by the calculations.

Limitations of a pencil beam approach to photon dose calculations in the head and neck region.

The inherent limitations of a specific pencil beam model have been studied when applied to a cylindrical geometry simulating the neck region. A comparison is made between measured and calculated absorbed dose in a cylindrical phantom. The goal is to quantify the deviations in the absorbed dose level, i.e., the dose per monitor unit, when photons are used for the treatment of head and neck tumours. Square fields ranging from 5 x 5 up to 30 x 30 cm2 are studied for photon beam energies of 60Co, 4, 6 and 18 MV. Ionisation chamber measurements have been performed in the cylinder as well as in two other configurations in order to trace the origin of possible deviations. For 18 MV no significant deviations are found between measurement and calculation in the cylindrical configuration. For the lower energies, an overestimation of the calculated dose in the cylindrical configuration up to about 6% for a 20 x 20-cm2 60Co field has been found. These deviations have been traced to the basic approximation for the integration volume for phantom scatter calculations inherent in this pencil beam implementation.

Independent checking of the delivered dose for high-energy X-rays using a hand-held PC.

BACKGROUND AND PURPOSE: The requirements on the delivered dose in radical radiation therapy are extremely high. The dose should be within a few percent and also delivered with high accuracy in space. Vendors and users have successfully managed to implement radiation therapy systems, which are able to achieve these demands with high accuracy and reproducibility. These systems include computerized tomography scanners, treatment planning systems, simulators, treatment machines, and record and verify systems. More and more common are also computer networks to assure data integrity when transferring information between the systems. Even if these systems are commissioned and kept under quality assurance programs to maintain their accuracy, errors may be introduced. Especially, the human factor is an uncontrolled parameter that may introduce errors. Thus, unintentional changes or incorrect handling of data may occur during clinical use of the equipment. Having an independent dose calculation system implemented in the daily quality assurance process may assure a high quality of treatments and avoidance of severe errors. MATERIALS AND METHODS: To accomplish this, a system of equations for calculating the absorbed dose to the prescription point from the set-up information, has been compiled into a dose-calculation engine. The model is based on data completely independent of the treatment planning system (TPS). The fundamental parameter in the dose engine is the linear attenuation coefficient for the primary photons. This parameter can readily be determined experimentally. The dose calculation engine has been programmed into a hand-held PC allowing direct calculation of the dose to the prescription point when the first treatment is delivered to the patient. RESULTS AND CONCLUSION: The model is validated with measurements and is shown to be within +/-1.0% (1 SD). Comparison against a state-of-the-art TPS shows an average difference of 0.3% with a standard deviation of +/-2.1%. An action level covering 95% of the cases has been chosen, i.e. +/-4.0%. Deviations larger than this are with a high probability due to erroneous handling of the patient set-up data. This system has been implemented into the daily clinical quality control program.

Application of the convolution method for calculation of output factors for therapy photon beams.

The output factor for a therapy photon beam is defined as the dose per monitor unit relative to the dose per monitor unit in a reference field. Convolution models for photon dose calculations yield the dose in units normalized to the incident energy fluence with phantom scatter intrinsically modeled. Output factors calculated with the convolution method as the dose per unit energy fluence relative to the calculated dose per unit energy fluence in a reference field could deviate as much as 5% if corrections are not made for perturbations due to treatment head scatter. Significant perturbations are particles backscattered from the collimators to the monitor and photons forward scattered from the filter and collimators in the treatment head. The forward scatter adds an "unmonitored" contribution to the total energy fluence of the beam. A model is developed that describes the field size dependence of these perturbations for conversion of output factors, calculated with the convolution method, to machine output factors as an integrated part in treatment planning. The necessary machine characteristics are derived from measurements of the output in air for a limited set of field sizes. The method has been tested using five different multileaf collimated irregular fields at 6 MV and for a large set of rectangular fields at 5, 6, and 18 MV and found to predict output factors with an accuracy better than 1%.

Comparative dosimetry of diode and diamond detectors in electron beams for intraoperative radiation therapy.

The aim of the present study is to examine the validity of using silicon semiconductor detectors in degraded electron beams with a broad energy spectrum and a wide angular distribution. A comparison is made with diamond detector measurements, which is the dosimeter considered to give the best results provided that dose rate effects are corrected for. Two-dimensional relative absorbed dose distributions in electron beams (6-20 MeV) for intraoperative radiation therapy (IORT) are measured in a water phantom. To quantify deviations between the detectors, a dose comparison tool that simultaneously examines the dose difference and distance to agreement (DTA) is used to evaluate the results in low- and high-dose gradient regions, respectively. Uncertainties of the experimental measurement setup (+/- 1% and +/- 0.5 mm) are taken into account by calculating a composite distribution that fails this dose-difference and DTA acceptance limit. Thus, the resulting area of disagreement should be related to differences in detector performance. The dose distributions obtained with the diode are generally in very good agreement with diamond detector measurements. The buildup region and the dose falloff region show good agreement with increasing electron energy, while the region outside the radiation field close to the water surface shows an increased difference with energy. The small discrepancies in the composite distributions are due to several factors: (a) variation of the silicon-to-water collision stopping-power ratio with electron energy, (b) a more pronounced directional dependence for diodes than for diamonds, and (c) variation of the electron fluence perturbation correction factor with depth. For all investigated treatment cones and energies, the deviation is within dose-difference and DTA acceptance criteria of +/- 3% and +/- 1 mm, respectively. Therefore, p-type silicon diodes are well suited, in the sense that they give results in close agreement with diamond detectors, for practical measurements of relative absorbed dose distributions in degraded electron beams used for IORT.

Verification and implementation of dynamic wedge calculations in a treatment planning system based on a dose-to-energy-fluence formalism.

The use of dynamic movements on linear accelerators during irradiation has found a revised interest lately due to the integration of computers to control the accelerator. In this paper, dynamic wedge fields that are produced by moving one of the collimator blocks during irradiation are studied. Since these wedge fields differ from those of mechanical wedges, certain requirements are to be met on the treatment planning system. A pencil-beam-based treatment planning system that uses the resultant energy fluence distribution from the dynamic collimator movement has been extensively reviewed. In calculations, the system treats the dynamic collimated field as a single, modulated field that yields calculation times close to those for open beams. Details are given on the theoretical model used for the calculation of dynamically generated dose distributions. Measurements of depth doses, profiles, and output factors in dynamic wedge fields indicate that calculations accurately predict the outcome from dynamic wedges without any additional measurements other than those used for characterization of static open beams.

Which depth dose data should be used for dose planning when wedge filters are used to modify the photon beam?

The use of beam data for open photon fields when calculating absorbed dose distributions for beams with wedge filters has been studied. The depth doses for beams with wedge filters are changed through beam hardening and the dose maximum can be shifted; both these changes result in errors in the final dose calculations of several per cent if open beam data are used. The errors are larger for 6 MV than for 18 MV x-rays. The depth of measurement for determining the wedge factor and the influence of other beam modifying devices are discussed. It is recommended that the reference depth be used instead of the dose maximum for these kinds of measurements since the influence of contaminating electrons in the beam will then be avoided and the wedge factor will be correct at a clinically relevant depth.

A virtual linear accelerator for verification of treatment planning systems.

A virtual linear accelerator is implemented into a commercial pencil-beam-based treatment planning system (TPS) with the purpose of investigating the possibility of verifying the system using a Monte Carlo method. The characterization set for the TPS includes depth doses, profiles and output factors, which is generated by Monte Carlo simulations. The advantage of this method over conventional measurements is that variations in accelerator output are eliminated and more complicated geometries can be used to study the performance of a TPS. The difference between Monte Carlo simulated and TPS calculated profiles and depth doses in the characterization geometry is less than +/-2% except for the build up region. This is of the same order as previously reported results based on measurements. In an inhomogeneous, mediastinum-like case, the deviations between TPS and simulations are small in the unit-density regions. In low-density regions, the TPS overestimates the dose, and the overestimation increases with increasing energy from 3.5% for 6 MV to 9.5% for 18 MV. This result points out the widely known fact that the pencil beam concept does not handle changes in lateral electron transport, nor changes in scatter due to lateral inhomogeneitics. It is concluded that verification of a pencil-beam-based TPS with a Monte Carlo based virtual accelerator is possible, which facilitates the verification procedure.

Dosimetric verification of open asymmetric photon fields calculated with a treatment planning system based on dose-to-energy-fluence concepts.

Output normalized dose profiles for asymmetric open photon fields has been calculated using a commercial treatment planning system (TPS) based on a dose-to-energy-fluence concept. The model does not require any additional measurements for off-axis fields. Calculations are compared with measurements for quadratic fields of 5 cm x 5 cm up to 20 cm x 20 cm, with their geometric field centre positioned 10 cm off-axis in the in-plane direction. The measurements include depth doses and profiles in-plane as well as cross-plane for nominal photon energies of 4, 6 and 18 MV x-rays. Both calculated and measured doses are normalized with respect to a 10 cm x 10 cm reference field, therefore making it possible to compare not only the relative distributions but also the absolute dose levels; that is, calculation of monitor units is included. The calculated depth-dose curves are generally in good agreement with measured data with an accuracy at the absolute dose level of 2% at depths beyond the dose maximum. The cross-plane profiles are calculated with an accuracy better than 3% within the field. The 'tilt' towards the collimator central axis of the in-plane profiles is predicted by the model, but is somewhat overestimated at large depths. The system provides the possibility to separate the primary and scattered parts of the dose and the cause of this tilting was studied by comparing calculated phantom-scattering and head-scattering dose profiles for a symmetric 40 cm x 20 cm field to dose profiles for an asymmetric 20 cm x 20 cm field. The tilting is shown to originate from a change both in phantom scattering and in head scattering compared to the case of symmetrical fields. The results indicate that the investigated TPS can calculate dose distributions in open asymmetric fields with a high degree of accuracy, typically better than 2-3%.

The influence of air humidity on an unsealed ionization chamber in a linear accelerator.

The safe and accurate delivery of the prescribed absorbed dose is the central function of the dose monitoring and beam stabilization system in a medical linear accelerator. The absorbed dose delivered to the patient during radiotherapy is often monitored by a transmission ionization chamber. Therefore it is of utmost importance that the chamber behaves correctly. We have noticed that the sensitivity of an unsealed chamber in a Philips SL linear accelerator changes significantly, especially during and after the summer season. The reason for this is probably a corrosion effect of the conductive plates in the chamber due to the increased relative humidity during hot periods. We have found that the responses of the different ion chamber plates change with variations in air humidity and that they do not return to their original values when the air humidity is returned to ambient conditions.

The dosimetric verification of a pencil beam based treatment planning system.

A new three-dimensional treatment planning system (TPS) based on convolution/superposition algorithms (TMS-Radix from HELAX AB, Uppsala, Sweden) was recently installed at the University Hospital in Lund. The purpose of the present study was to design a quality assurance and acceptance testing programme to meet the specific characteristics of this convolution model. The model is based on parametrization of a non-measurable quantity-the polyenergetic pencil beam. However, the verification of the treatment planning model is still dependent on numerous comparisons of measured depth-doses and dose profiles. The test programme was divided in two basic parts: (i) model implementation and beam data consistency and (ii) model performance and limitations in special situations. The first part was scheduled for all photon beam qualities available before they could be used for clinical treatment planning. The second part was performed for selected energies only. The results indicate clearly that the model is well suited for clinical three-dimensional dose planning and that the TPS handles data as expected. For example, calculated depth-doses for open and wedge beams at depths larger than the depth of dose maximum and profiles for open beams shows a very good agreement with measurements. However, depth-dose deviations at shallow depths, especially for high energies, were found. Monitor units calculated by the system were accurate for most fields except for very large fields, where deviations of several per cent were found.

Limitations of a pencil beam approach to photon dose calculations in lung tissue.

A common limitation in treatment planning systems for photon dose calculation is to ignore the impact on electron transport and photon scatter from patient heterogeneities. The heterogeneity correlation is often based on scaling operations along beam rays as for the method according to Batho or the more novel approach of 1D convolutions along beam paths applied in pencil-beam-based systems. The effects of the limitation have been studied in a mediastinum geometry for a wide range of beam qualities by comparing the results from a pencil-beam-based treatment planning system with the results from Monte Carlo calculations. As expected, the deviations within unit-density volumes are small while deviations in low-density volumes increase with increasing beam energy from approximately 3% for 4 MV to 14% for 18 MV x-rays as a result of increased electron disequilibrium.

Verification of a pencil beam based treatment planning system: output factors for open photon beams shaped with MLC or blocks.

The accuracy of monitor unit calculations from a pencil beam based, three-dimensional treatment planning system (3D TPS) has been evaluated for open irregularly shaped photon fields. The dose per monitor unit was measured in water and in air for x-ray beam qualities from 6 to 15 MV. The fields were shaped either with a multileaf collimator (MLC) or with customized alloy blocks. Calculations from the 3D TPS were compared with measurements. The agreement between calculated and measured dose per monitor unit depended on field size and the amount of blocking and was within 3% for the MLC-shaped fields. The deviation could be traced to limitations in head scatter modelling for the MLC. For fields shaped with alloy blocks, the dose per monitor unit was calculated to be within 1.6% of measured values for all fields studied. The measured and calculated relative phantom scatter for fields with the same equivalent field size were identical for MLC and alloy shaped fields. These results indicate that the accuracy in the TPS calculations for open irregular fields, shaped with MLC or blocks, is satisfactory for clinical situations.