Quality assurance for intraoperative electron radiotherapy clinical trials: ionization chamber and mailable thermoluminescent dosimeter results.

Ionization chamber and thermoluminescent dosimeter measurements were made to verify the dosimetry data provided to the Radiation Therapy Chart Review office of interinstitutional electron intraoperative radiotherapy clinical trials. The ionization measurements included intraoperative radiotherapy applicator output and depth dose measurements made at the stated 80% and 50% depths at 14 different radiotherapy facilities. Mailable thermoluminescent dosimeter measurements, including output and depth dose measurements were made at 16 institutions. The mailable thermoluminescent dosimeter results show similar inter-institutional agreement, both for output and depth dose comparison, with the corresponding ion chamber results for intraoperative radiotherapy applicators. These results are compared to similar measurements made for conventional electron applicators.

Determination of depth for electron breast boosts.

A technique has been developed to determine the depth for the electron boost treatments for patients undergoing definitive irradiation for early stage breast cancer. A series of parallel link chains are placed on the breast over the clinically determined site of the boost. Using fluoroscopy, the physician confirms that the chains overlie the tumor bed which is outlined by radiopaque surgical clips placed at the time of the breast biopsy. A pair of orthogonal films and/or rotational stereo shift films are obtained with a standard simulator unit. Using the image of the chains to define the surface contour, the depth of each surgical clip is measured directly from the orthogonal films or calculated from the rotational stereo shift films. With this information, the physician can determine the appropriate electron energy to cover the target volume. This method was tested by comparison with depths measured from CT scan, and close agreement was demonstrated.

Patterns of change in the physics and technical support of radiation therapy in the USA 1975-1986.

Information on the patterns of personnel and related equipment support and availability at various types of radiation oncology facilities are included in the Facilities Master List surveys conducted by the American College of Radiology. This paper summarizes the surveyed data obtained during 1975-1986. The data presented include the use of equipment and the degree of personnel support at government owned, hospital or university based, and freestanding facilities. There is increasing use of linear accelerators, simulators, and treatment planning computers among all types of facilities. The use of 60Co units has been progressively decreasing. Almost all types of facilities show inadequate, but slowly improving, numbers of physicians, physicists, dosimetrists, and technologists when compared with the level recommended by the Blue Book.

Conformal static field therapy for low volume low grade prostate cancer with rigid immobilization.

The ability to improve existing standards of treating prostate cancer was investigated. To deliver a homogenous dose to the prostate with as little normal tissue margin as practical, seven patients with low volume carcinoma of the prostate were immobilized with alpha cradle body casts prior to using a CT-based 3D treatment planning system and beam's eye view (BEV) template. All patients had clinical Stage B-1 prostate cancer of favorable histologic differentiation (Gleason Score 2-5). A four field box technique was used, each beam having a single customized cerrobend block cut-out conforming to the exact contour of the prostate. To assess the accuracy of this process, daily port films were taken for 5 consecutive days and compared to a matched control group who were treated in a similar fashion, but were not casted. Dose volume histograms illustrate an average of 14% of bladder dose and 14% of rectal dose that can be eliminated using this technique when compared to field sizes and block placement in our previous technique. Daily setup variation was markedly improved using the cast, with a median daily variation of 1 mm as compared to 3 mm without the cast. The average range of movement for each of the seven casted patients was 3.3 mm as compared to 8 mm for the seven uncasted patients. Immobilization eliminated the worst 10% of all daily positioning errors. Using CT treatment planning with the patient casted and BEV allows for precise block placement with the prostate gland in its proper orientation during daily treatment. With improved immobilization and precise localization of the prostate gland, margins around the target can be made significantly smaller, and this may translate into a decrease in acute and/or late complications.

Using simulation data to predict lung geometry for inhomogeneity corrections in breast cancer treatments.

PURPOSE: To develop a statistical model based only on simulation measurement data, to predict the lung geometry in the central slice of the tangential radiation treatment fields for breast cancer. METHODS AND MATERIALS: A linear regression analysis was performed on 22 patients to determine the shape of lung in the central axis plane of the tangential radiation fields. Data collected include the greatest perpendicular distance (GPD) measured from the chest wall to the field border on computed tomography (CT) images, the central lung distance (CLD) measured from the posterior field border to the chest wall on the simulation portal images, and the lung contours digitized at 1 cm intervals. The lung contours of these patients were fitted to a parabolic curve through a polynomial regression model. A lung template based on the regression model is used to construct a "generic lung" contour on patients' external body contours for treatment planning. The accuracy of this technique was tested on another group of 15 patients for its ability to predict the shape of lung on the central axis plane and the accuracy of dose to the prescription point. RESULTS: The polynomial regression indicates that all the patients' lung contours in the tangential fields follow a parabolic curve: Y = -0.0808 X2 + 0.0096 X + 0.0326. The maximum lung involvement (GPD) can be determined from the value of CLD measured on the simulation film by the linear regression model with a determination coefficient of 0.712. The 15-patient test results indicate that our model predicts the lung separation on the central axis with an average deviation of 1.35 cm, and the average absolute dose deviation to the dose prescription point is 1.46%. CONCLUSION: The model presented in this article provides an efficient method to estimate the lung geometry for breast cancer treatment planning without the requirement of CT data. The lung contour predicted by our model is useful for calculating dose distributions with inhomogeneity correction and may potentially benefit patients at higher risk of pulmonary toxicity.

Dose to contralateral breast: a comparison of four primary breast irradiation techniques.

PURPOSE: Contralateral breast dose from primary breast irradiation has been implicated in the risk of second breast malignancies. It has been previously shown that the use of half-beam blocking can increase the opposite breast dose by a factor of five. This study evaluates four different breast treatment techniques to compare the radiation dose to the contralateral breast. METHODS AND MATERIALS: Dose measurements were made using thermoluminescent dosimeters (TLD) capsules, which were placed in the Rando phantom in the following locations in the contralateral breast: seven along the central axis plane, on at 5 cm superior to, and one 5 cm inferior to the central axis plane. One TLD capsule was placed in the midcenter of the treated breast. The following radiation techniques were used: (a) half-beam with a custom block (HB+CB), (b) half-beam using asymmetric collimator jaw (HB/AJ), (c) half-beam using asymmetric collimator jaw with custom block (H/AJ+CB), and (d) isocentric technique with nondivergent posterior borders [Joint Center for Radiation Therapy (JCRT) techique]. For each technique, isodose distributions for the Rando phantom were optimized using (a) 15 degree medial and lateral compensating wedges, and (b) a single 30 degree lateral compensating wedge. The phantom was treated with 6 MV photons. Each technique was repeated six times, and the TLD readings were averaged. RESULTS: The custom cerrobend half-beam block technique gives the highest contralateral breast dose, regardless of wedge. The remaining techniques give results in a similar range, with the asymmetric jaw with no medial wedge technique giving the lowest total dose (p = not significant). The use of a medial wedge increases the opposite breast dose for all techniques. The asymmetric half-beam technique gives significantly less dose than the cerrobend half-beam technique, due to decreased transmission through the asymmetric collimators. The asymmetric jaw vs JCRT technique results in similar contralateral breast dose. CONCLUSIONS: As expected, dose to the contralateral breast increases with the use of a medial wedge. Cerrobend half-beam blocking gives the highest opposite breast dose. The lowest contralateral breast dose is with the asymmetric jaw with no medial wedge and no block. The asymmetric jaw technique with block yields equivalent contralateral breast doses to the JCRT technique.

A nondivergent three field matching technique for breast irradiation.

Effective radiation therapy to intact breasts requires the delivery of adequate dose to a large target volume using complex beam arrangements. A semi-empirical method is described to determine the correct gantry, collimator, and couch positions for a geometrically accurate field match among adjoining radiation beams. The technique uses a metal rod and chain combination to aid determination of the proper couch setting under remote fluoroscopy control. A couch position error of more than half a degree is easily detectable by this technique.

Significance of prone positioning in planning treatment for esophageal cancer.

The treatment of esophageal cancer is made difficult by the close proximity of the esophagus to the spinal cord and the requirement to treat the esophageal target volume to doses greater than or equal to 60 Gy while limiting the spinal cord dose to less than or equal to 46 Gy. By placing the patient in the prone position, the esophagus can be displaced away from the spinal cord. We explored the results of this commonly used technique on 16 patients who have undergone simulation in both supine and prone positions. Both AP and lateral orthogonal radiographs were obtained in both positions. The distance between contrast material in the esophagus and spinal cord was noted in at least four transverse planes through the thoracic esophagus on each of the 16 patients. These four transverse planes were located at 3 cm above the carina, at the carina, 3 cm below the carina and 6 cm below the carina. The mean displacement (+/- 1 SD) of the esophagus away from the spinal cord when the patient was in the prone position compared to supine at each of these levels was 1.3 (+/- 0.8) cm, 1.8 (+/- 0.9) cm, 1.8 (+/- 1.0) cm, and 1.9 (+/- 1.1) cm. The range of displacement for all 64 displacement determinations was 0 to 4.2 cm with a mean of 1.7 cm. To evaluate further the consequences of prone positioning on treatment planning and doses received to target volumes and critical structures, we performed 3-dimensional treatment planning with a patient in both prone and supine positions. The requirements were to achieve a tumor volume dose of 60 Gy while keeping the spinal cord dose below 46 Gy. Two types of conventional treatment plans were examined in prone and supine positions. A 6-field plan consisted of delivery of 40 Gy through a large 3-field beam arrangement followed by delivery of 20 Gy through a similar 3-field cone down. An 8-field plan involved the delivery of 30 Gy through AP/PA beams followed by a 3-field beam arrangement to 40 Gy and a subsequent 3-field cone-down for the final 20 Gy. Comparison of dose volume histograms revealed that the 6-field plan spared relatively more heart whereas the 8-field plan spared relatively more lung. Regarding the primary consideration of coverage of target volume with avoidance of spinal cord, prone positioning was superior to supine positioning whether 6- or 8-field arrangements were used.

Three-dimensional photon treatment planning of the intact breast.

Three-dimensional treatment planning for the intact breast was performed on two patients who had undergone CT scanning. A total of 38 treatment plans were evaluated. Multiple plans were evaluated for each patient including plans with and without inhomogeneity corrections, plans using varying photon energies of 60Co, 4 MV, 6 MV, 10 MV, and 15 MV, and three-dimensionally unconstrained plans. Increased hot spots were appreciated in the central axis plane when lung inhomogeneity corrections were used. Additional hot spots were appreciated in off-axis planes towards the cephalad and caudad aspects of the target volume because of lung inhomogeneity corrections and changes in the breast contour. The use of 60Co was associated with an increase in the magnitude and volume of hot spots, whereas the use of higher energy photons such as 10 MV and 15 MV was associated with an unacceptable target coverage at shallow depths. Therefore, for the two patients studied, the use of a medium energy photon beam (such as from a 6 MV linear accelerator) appeared to be the energy of choice for treatment of the intact breast. The three-dimensionally unconstrained plans were able to improve slightly upon the standard plans, particularly with relationship of dose to normal tissue structures. Areas for future research were identified, including the use of tissue compensators.

Three-dimensional treatment planning for postoperative treatment of rectal carcinoma.

The role of three-dimensional (3-D) treatment planning for postoperative radiation therapy was evaluated for rectal carcinoma as part of an NCI contract awarded to four institutions. It was found that the most important contribution of 3-D planning for this site was the ability to plan and localize target and normal tissues at all levels of the treatment volume, rather than using the traditional method of planning with only a single central transverse slice and simulation films. There was also a slight additional improvement when there were no constraints on the types of plans (i.e., when noncoplanar beams were used). Inhomogeneity considerations were not important at this site under the conditions of planning, i.e., with energies greater than 4 MV and multiple fields. Higher beam energies (15-25 MV) were preferred by a small margin over lower energies (down to 4 MV). The beam's eye view and dose-volume histograms were found quite useful as planning tools, but it was clear that work should continue on better 3-D displays and improved means of translating such plans to the treatment area.

CT and PET lung image registration and fusion in radiotherapy treatment planning using the chamfer-matching method.

PURPOSE: We present a validation study of CT and PET lung image registration and fusion based on the chamfer-matching method. METHODS AND MATERIALS: The contours of the lung surfaces from CT and PET transmission images were automatically segmented by the thresholding technique. The chamfer-matching technique was then used to register the extracted lung surfaces. Arithmetic means of distance between the two data sets of the pleural surfaces were used as the cost function. Matching was then achieved by iteratively minimizing the cost function through three-dimensional (3D) translation and rotation with an optimization method. RESULTS: Both anatomic thoracic phantom images and clinical patient images were used to evaluate the performance of our registration system. Quantitative analysis from five patients indicates that the registration error in translation was 2-3 mm in the transverse plane, 3-4 mm in the longitudinal direction, and about 1.5 degree in rotation. Typical computing time for chamfer matching is about 1 min. The total time required to register a set of CT and PET lung images, including contour extraction, was generally less than 30 min. CONCLUSION: We have implemented and validated the chamfer-matching method for CT and PET lung image registration and fusion. Our preliminary results show that the chamfer-matching method for CT and PET images in the lung area is feasible. The described registration system has been used to facilitate target definition and treatment planning in radiotherapy.

Numerical scoring of treatment plans.

This is a report on numerical scoring techniques developed for the evaluation of treatment plans as part of a four-institution study of the role of 3-D planning in high energy external beam photon therapy. A formal evaluation process was developed in which plans were assessed by a clinician who displayed dose distributions in transverse, sagittal, coronal, and arbitrary oblique planes, viewed dose-volume histograms which summarized dose distributions to target volumes and the normal tissues of interest, and reviewed dose statistics which characterized the volume dose distribution for each plan. In addition, tumor control probabilities were calculated for each biological target volume and normal tissue complication probabilities were calculated for each normal tissue defined in the agreed-upon protocols. To score a plan, the physician assigned a score for each normal tissue to reflect possible complications; for each target volume two separate scores were assigned, one representing the adequacy of tumor coverage, the second the likelihood of a complication. After scoring each target and normal tissue individually, two summary scores were given, one for target coverage, the second reflecting the impact on all normal tissues. Finally, each plan was given an overall rating (which could include a downgrading of the plan if the treatment was judged to be overly complex).

Dose to superficial node for patients with head and neck cancer treated with 6 MV and 60Co photons.

The superficial neck nodes in only 1 out of 7 patients with head and neck cancer studied received more than 90% of the prescribed dose when treated with opposed 6 MV photons. Beam spoilers placed upstream from the patient enhanced the dose to the superficial node at the expense of higher dose to the skin.

Applications of simulator computed tomography number for photon dose calculations during radiotherapy treatment planning.

PURPOSE: To study the variation of computed tomography (CT) number from a simulator-based scanner and the effect of this variation on photon-dose calculations. METHOD AND MATERIALS: CT images of a cylindrical phantom with multiple inserts were obtained using a commercially-available simulator-CT (Ximatron: Varian, Palo Alto, CA). The linear correlation coefficient and Chi-square methods were used to determine the X-ray effective energy in a phantom. CT numbers in Hounsfield units (HU) were measured as a function of phantom size, orientation, field of view (FOV), distance from the center, and time for various inserts. The change of dose calculations due to the CT number variations was then determined using the equivalent path-length (EPL) and collapsed cone convolution methods. RESULTS AND DISCUSSION: A significant beam-hardening effect was observed for the simulator-CT. Consequently, the CT number from the sim-CT was more sensitive to the size of the phantom than those from a conventional CT. The sim-CT number is not sensitive to the locations within the phantom and is stable over a 6-week period. It is important to use the proper FOV for sim-CT studies; scanning a small polystyrene phantom using a large FOV may result in an increase of l20 HU in CT number at the center of the field. However, the dose-calculation variations, due to the CT number uncertainty, do not exceed 2-3% for 6-18 MV photon beams. CONCLUSION: The simulator CT images were acquired with patients in the treatment position, and these CT numbers are useful for CT-based dose calculations.

Practice of 3-dimensional treatment planning at the Fox Chase Cancer Center, University of Pennsylvania.

The use of 3-dimensional (3-D) dose distributions and dose-volume histograms in radiation therapy treatment planning is illustrated on a patient with a head and neck tumor. The patient was immobilized in a rectangular tissue compensation bolus box. The treatment was planned with a 14 MeV D-T derived fast neutron therapy beam. The isodose distributions and the dose-volume histograms at multiple adjacent levels are used to evaluate the adequacy of coverage of target volumes and the doses to the normal tissues. Such dose-volume histograms are useful and practical in summarizing the dose distribution throughout the irradiated volume, assessing the degree of uniformity of the dose distribution within the target volume, quantifying the amount of normal tissue irradiated, and evaluating rival treatment plans for both particle and nonparticle beams.

Static multileaf collimator for fast-neutron therapy.

This paper describes a multileaf collimator, designed to obtain irregularly shaped neutron fields, that fits into an existing fixed field-size collimator system. The shape of the desired portal field is obtained by appropriately setting the position of each leaf of the collimator. The collimator consists of 40 polyethylene leaves to produce fields of irregular shape of up to 400 cm2. The leaves are interlaced with grooves to prevent neutron leakage. The measured penumbra defined by the multileaf collimator is similar to that defined by the same length fixed field-size water extended polyester (WEP) collimator. The multileaf collimator has a lower neutron transmission due to its higher hydrogen concentration.

Dynamic wedge versus physical wedge: a Monte Carlo study.

The purpose of this study is to analyze the characteristics of dynamic wedges (DW) and to compare DW to physical wedges (PW) in terms of their differences in affecting beam spectra, energy fluence, angular distribution, contaminated electrons, and dose distributions. The EGS4/BEAM Monte Carlo codes were used to simulate the exact geometry of a 6 MV beam and to calculate 3-D dose distributions in phantom. The DW was simulated in accordance with the segmented treatment tables (STT). The percentage depth dose curves and beam profiles for PW, DW, and open fields were measured and used to verify the Monte Carlo simulations. The Monte Carlo results were found to agree within 2% with the measurements performed using film and ionizing chambers in a water phantom. The present EGS4 calculation reveals that the effects of a DW on beam spectral and angular distributions, as well as electron contamination, are much less significant than those for a PW. For the 6 MV photon beam, a 45 degrees PW can result in a 30% increase in mean photon energy due to the effect of beam hardening. It can also introduce a 5% dose reduction in the build-up region due to the reduction of contaminated electrons by the PW. Neither this mean-energy increase nor such dose reduction is found for a DW. Compared to a DW, a PW alters the photon-beam spectrum significantly. The dosimetric differences between a DW and a PW are significant and clearly affect the clinical use of these beams. The data presented may be useful for DW commissioning.

Calculation of monitor units for a linear accelerator with asymmetric jaws.

A simple approach was developed that calculates the output factors and tissue maximum ratio of an asymmetric field at any point within the open field, and specifically both at the central axis (when the jaws do not shadow it) and at the effective center of the open field, using the existing tables for symmetric fields and the multidepth profile information for the largest available field size (either open or with a wedge present). Day's method was adapted to calculate the effective values of the usual field-size-dependent parameters. This approach makes these parameters also dependent on the location of the calculation point relative to the field edges in an asymmetric field. This algorithm was tested by comparing its predictions with measurements of asymmetric and half blocked fields, with and without wedges, in a water phantom at different depths and off-axis distances. The agreement between calculated and measured dose rate is within 1%-3% even in highly asymmetric fields for both 6- and 18-MV photons.

Evaluating the effectiveness of a longitudinal magnetic field in reducing underdosing of the regions around upper respiratory cavities irradiated with photon beams--a Monte Carlo study.

The problem of underdosing lesions adjacent to upper respiratory cavities and a proposal to correct it are presented in this work. The EGS4 Monte Carlo code was used to simulate a 6 MV x-ray beam passing through a block of tissues with air cavities 2, 4, and 6 cm wide. The geometry used approximates the tracheal geometry used by previous researchers who investigated the underdosing phenomenon. A uniform longitudinal magnetic field of 0.5 T strength is used to reduce secondary electron outscatter caused by the presence of an air gap, and thus improving the dose at the distal surface of air cavities. We introduce the term "percent dose reduction" (PDR), which is defined as the difference between the dose after the air cavity and the dose at the same depth in soft-tissue phantom normalized to the dose in the tissue phantom, to quantify the reduction in dose after an air gap. We also introduce the term dose improvement ratio (DIR), which is defined as the dose ratio with magnetic field to the dose, at the same point, without magnetic field, to quantify the improvement in dose when the magnetic field is applied. For 2 x 2 x 20 cm3 and 4 x 4 x 20 cm3 air cavities irradiated by 2 x 2 cm2 beams, we found PDRs of 38% and 52%, respectively. This means that for these cavities, there is a 38% and a 52% reduction in dose at the cavity edge compared to the same dose in tissue at the same depth for each cavity. The dose improved by 30% (DIR= 1.3) and 87% (DIR= 1.87), respectively, when applying the magnetic field. The worst effect on dose at the distal side came from larger cavities irradiated with small fields. In these situations, the improvement in dose due to the presence of magnetic field was the largest. This article deals with "ideal" head and neck geometries with a uniform magnetic field. In a paper to follow we will use a CT-based phantom to study the effect in realistic geometries with the presence of a magnetic field from a Helmholtz coil pair.

A method for more efficient source localization of interstitial implants with biplane radiographs.

The conventional method for source localization in an interstitial ribbon implant by means of biplane radiographs can be difficult, especially when a large number of seeds are involved. We present a new algorithm for more efficient source localization with the same conventional biplane radiographs. The method does not require a one-to-one source correspondence between two radiographs. The user needs only to digitize several points, following the shape of each ribbon from both films. The points that are digitized do not need to be the location of the seeds, and they do not have to correspond to the same points on both films. The algorithm uses the multidimensional minimization method to reconstruct the three-dimensional locus of the ribbon. The location of each seed is then determined by its pathlength relative to the corresponding starting point. We have used phantom experiments and clinical cases to test the reliability of the algorithm. The results show that the errors in the determination of seed locations are less than 2 mm, and the efficiency in source digitization for data entry can be increased by a factor up to 5.