Jake Van Dyk

Dosimetric considerations for patients with HIP prostheses undergoing pelvic irradiation. Report of the AAPM Radiation Therapy Committee Task Group 63.

This document is the report of a task group of the Radiation Therapy Committee of the AAPM and has been prepared primarily to advise hospital physicists involved in external beam treatment of patients with pelvic malignancies who have high atomic number (Z) hip prostheses. The purpose of the report is to make the radiation oncology community aware of the problems arising from the presence of these devices in the radiation beam, to quantify the dose perturbations they cause, and, finally, to provide recommendations for treatment planning and delivery. Some of the data and recommendations are also applicable to patients having implanted high-Z prosthetic devices such as pins, humeral head replacements. The scientific understanding and methodology of clinical dosimetry for these situations is still incomplete. This report is intended to reflect the current state of scientific understanding and technical methodology in clinical dosimetry for radiation oncology patients with high-Z hip prostheses.

Partial volume rat lung irradiation; assessment of early DNA damage in different lung regions and effect of radical scavengers

PURPOSE These studies were designed to examine radiation-induced in-field and out-of-field DNA damage in rat lung as a function of dose and various volumes of irradiation. They also determined whether superoxide dismutase (SOD) and nitro-L-arginine methyl ester (L-NAME) protected against this damage. METHODS AND MATERIALS The whole lung, or various volumes of the lower or upper lungs of Sprague-Dawley rats were exposed to doses up to 20 Gy of 60Co gamma rays. Radiation-induced DNA damage was quantified in fibroblasts obtained at 18 h after irradiation from both irradiated and shielded lung regions using a micronucleus assay. The radioprotective role of SOD (CuZnSOD: 10 mg/kg body weight; MnSOD: 50-100mg/kg body weight) and L-NAME (0.2 mg/kg body weight.) in vivo was determined by injecting them into rats 30 min before or immediately after a dose of 10 Gy. RESULTS Micronucleus formation was approximately linear with dose up to 15 Gy. When 70% of the lung volume was irradiated with 10 Gy, irradiated lower lung gave similar numbers of micronuclei (MN)/binucleate cell (BN) to that observed following whole lung irradiation (0.91 MN/BN), whereas the irradiated upper lung gave only 0.66 MN/BN. Following lower lung irradiation, the shielded upper lung (30% of lung volume) showed substantial (out-of-field) damage (0.43 MN/BN). When 30% of the lung was given 10 Gy, irradiated upper or lower lung showed similar amounts of in-field damage (0.43 MN/BN) but this was smaller than that seen following irradiation of 70% of the lung volume. For 30% lower lung irradiation, the shielded upper lung showed only a small out-of-field effect (0.1 MN/BN). For both volumes of irradiation there was a similar or smaller effect in the shielded lower lung after upper lung irradiation. Injection of SOD before or L-NAME after 10 Gy to the lower 70% lung volume resulted in a reduction in DNA damage both in-field and out-of-field but the percentage was much greater for out-of-field damage (50-60%) than for in-field damage (10-30%). Following whole lung irradiation (10 Gy) significantly greater DNA damage was observed in fibroblasts from the left lung than from the right lung (0.93 MN/BN vs. 0.82 MN/BN). Following whole lung irradiation there was no significant difference in DNA damage observed in fibroblasts from the lower lung and the upper lung. CONCLUSIONS With partial lung irradiation the lower lung sustains more in-field DNA damage following irradiation than the upper lung, whereas out-of-field effects are observed primarily in the upper lung (i.e. following lower lung irradiation). Following whole lung irradiation the left lung sustains more damage than the right lung but there is no difference between the upper and lower lung. The protective effects of SOD and L-NAME suggest that inflammatory cytokines induced by the irradiation may be involved in the initiation of a reaction resulting in the production of reactive oxyradicals and nitric oxide that cause indirect DNA damage both in and out of the radiation field.

VARIABILITY OF TARGET VOLUME DELINEATION IN CERVICAL ESOPHAGEAL CANCER

PURPOSE Three-dimensional (3D) conformal radiation therapy (CRT) assumes and requires the precise delineation of the target volume. To assess the consistency of target volume delineation by radiation oncologists, who treat esophageal cancers, we have performed a transCanada survey. MATERIALS AND METHODS One of three case presentations, including CT scan images, of different stages of cervical esophageal cancer was randomly chosen and sent by mail. Respondents were asked to fill in questionnaires regarding treatment techniques and to outline boost target volumes for the primary tumor on CT scans, using ICRU-50 definitions. RESULTS Of 58 radiation oncologists who agreed to participate, 48 (83%) responded. The external beam techniques used were mostly anterior-posterior fields, followed by a multifield boost technique. Brachytherapy was employed by 21% of the oncologists, and concurrent chemotherapy by 88%. For a given case, and the three volumes defined by ICRU-50 (i.e., gross tumor volume [GTV], clinical target volume [CTV], and planning target volume [PTV]) we determined: 1. The total length in the cranio-caudal dimension; 2. the mean diameter in the transverse slice that was located in a CT slice that was common to all participants; 3. the total volume for each ICRU volume; and 4. the (5, 95) percentiles for each parameter. The PTV showed a mean length of 14.4 (9.6, 18.0) cm for Case A, 9.4 (5.0, 15.0) cm for Case B, 11.8 (6.0, 16.0) cm for Case C, a mean diameter of 6.4 (5.0, 9.4) cm for Case A, 4.4 (0.0, 7.3) cm for Case B, 5.2 (3.9, 7.3) cm for Case C, and a mean volume of 320 (167, 840) cm3 for Case A and 176 (60, 362) cm3 for Case C. The results indicate variability factors (95 percentile divided by 5 percentile values) in target diameters of 1.5 to 2.6, and in target lengths of 1.9 to 5.0. CONCLUSION There was a substantial inconsistency in defining the planning target volume, both transversely and longitudinally, among radiation oncologists. The potential benefits of 3D treatment planning with high-precision dose delivery could be offset by this inconsistency in target-volume delineation by radiation oncologists. This may be particularly important for multicenter clinical trials, for which quality assurance of this step will be essential to the interpretation of results.

Limitations of a convolution method for modeling geometric uncertainties in radiation therapy. I. The effect of shift invariance

Convolution methods have been used to model the effect of geometric uncertainties on dose delivery in radiation therapy. Convolution assumes shift invariance of the dose distribution. Internal inhomogeneities and surface curvature lead to violations of this assumption. The magnitude of the error resulting from violation of shift invariance is not well documented. This issue is addressed by comparing dose distributions calculated using the Convolution method with dose distributions obtained by Direct Simulation. A comparison of conventional Static dose distributions was also made with Direct Simulation. This analysis was performed for phantom geometries and several clinical tumor sites. A modification to the Convolution method to correct for some of the inherent errors is proposed and tested using example phantoms and patients. We refer to this modified method as the Corrected Convolution. The average maximum dose error in the calculated volume (averaged over different beam arrangements in the various phantom examples) was 21% with the Static dose calculation, 9% with Convolution, and reduced to 5% with the Corrected Convolution. The average maximum dose error in the calculated volume (averaged over four clinical examples) was 9% for the Static method, 13% for Convolution, and 3% for Corrected Convolution. While Convolution can provide a superior estimate of the dose delivered when geometric uncertainties are present, the violation of shift invariance can result in substantial errors near the surface of the patient. The proposed Corrected Convolution modification reduces errors near the surface to 3% or less.

Limitations of a convolution method for modeling geometric uncertainties in radiation therapy. II. The effect of a finite number of fractions

Convolution methods can be used to model the effect of geometric uncertainties on the planned dose distribution in radiation therapy. This requires several assumptions, including that the patient is treated with an infinite number of fractions, each delivering an infinitesimally small dose. The error resulting from this assumption has not been thoroughly quantified. This is investigated by comparing dose distributions calculated using the Convolution method with the result of Stochastic simulations of the treatment. Additionally, the dose calculated using the conventional Static method, a Corrected Convolution method, and a Direct Simulation are compared to the Stochastic result. This analysis is performed for single beam, parallel opposed pair, and four-field box techniques in a cubic water phantom. Treatment plans for a simple and a complex idealized anatomy were similarly analyzed. The average maximum error using the Static method for a 30 fraction simulation for the three techniques in phantoms was 23%, 11% for Convolution, 10% for Corrected Convolution, and 10% for Direct Simulation. In the two anatomical examples, the mean error in tumor control probability for Static and Convolution methods was 7% and 2%, respectively, of the result with no uncertainty, and 35% and 9%, respectively, for normal tissue complication probabilities. Convolution provides superior estimates of the delivered dose when compared to the Static method. In the range of fractions used clinically, considerable dosimetric variations will exist solely because of the random nature of the geometric uncertainties. However, the effect of finite fractionation appears to have a greater impact on the dose distribution than plan evaluation parameters.

Pelvic fractures following irradiation of endometrial and vaginal cancers-a case series and review of literature.

PURPOSE To review the induction of pelvic fractures as a result of radiation therapy and to assess their management. MATERIALS AND METHODS The charts of patients with endometrial and vaginal cancers irradiated between 1991 and 1995 were reviewed. All patients were treated with megavoltage machines, energy ranging from cobalt to 25 MV photons. RESULTS We treated 336 patients, with a median follow-up duration of 28.9 months (range 0-73.3). Sixteen patients had symptomatic pelvic fractures. The 5-year actuarial incidence of symptomatic pelvic fracture was 2.1%. All patients had pain as the first symptom. The median time of onset was 11 months (range 4-46). Imaging studies of 37.5% (6/16) were initially interpreted to be recurrent malignancy. All patients were managed conservatively and nine patients showed radiological evidence of healing over a median time of 13 months (range 2-34). Six patients had specific drug treatment including provera, premarin, calcium supplements, or pamidronate. Of these, five healed. For the ten patients who did not have any specific treatment, only four showed signs of healing at the time of last follow-up. There was a trend toward earlier healing with specific drug treatment (P=0.11). CONCLUSIONS Fractures can easily be mistaken for metastatic lesions (37.5% in this series) which might be treated with further irradiation. Although not statistically significant, there was a trend towards early healing with drug therapy. More studies are required to generate quantitative data for dose-response relationships and to evaluate the effect of drug therapy on the healing of such fractures.

In-field and out-of-field effects in partial volume lung irradiation in rodents: Possible correlation between early dna damage and functional endpoints

PURPOSE Recent observations have shown that there are regional variations in radiation response in mouse lung as measured by functional assays. Furthermore, there are both in-field and out-of-field effects in radiation-induced lung damage as observed by DNA assay in rats. The purpose of this work is: (a) to examine mice lethality data following partial volume lung irradiation to assess the possibility of directional or regional effects, (b) to evaluate the correlation between mice lethality data and DNA damage assayed by micronuclei production in rat lung, and (c) to re-interpret mice lethality considering the existence of directional effects in lung cellular response to partial volume irradiation. METHODS AND MATERIALS The lethality data for mice, generated at the M. D. Anderson Cancer Center, Houston, and micronuclei yield data for rats obtained at Princess Margaret Hospital, Toronto, were used. A radiobiological model that allows for out-of-field and in-field effects for lung cell damage and lung response was developed. This model is based on the observation of DNA damage in shielded parts of rat lung that was assumed relevant to cell lethality and consequently overall lung response. RESULTS While the experimental data indicated directional or regional volume effects, the applicability of dose and volume as sole predictors of lung response to radiation was found to be unreliable for lower lung (base) irradiation in mice. This conforms well to rat lung response where micronuclei were observed in shielded apical parts of lung following base irradiation. The radiobiological model, which was specifically developed to account for the lung response outside of primary irradiated volume, provides a good fit to mice lethality data, using parameters inferred from rat micronuclei data. CONCLUSION Response to lung irradiation in rodents, in particular, elevated sensitivity to base irradiation, can be interpreted with a hypothesis of in-field and out-of-field effects for cellular response. If the existence of these effects for lung is subsequently proven in humans, it will require the incorporation of geometrical and directional information in normal tissue complication probability calculations for lung. These considerations are ignored in present approaches based only on conventional dose-volume histograms.

A systematic study of imaging uncertainties and their impact on 125I prostate brachytherapy dose evaluation.

In order to calculate the dose distribution delivered by a prostate brachytherapy implant, the seed positions and prostate volume are normally identified on post-implant CT images. We have systematically considered the impact of uncertainties in contouring the prostate, seed localization, and visualization of all the seeds on the calculated dose distributions, dose-volume histograms, and predicted radiobiological outcome. This study was done for a collection of 27 clinical 125I prostate brachytherapy implants, performed at the London Regional Cancer Centre during our early adoption of this technique. For these clinical dose distributions, the median D90 was 76% of the prescription dose, or 110 Gy, and the median V90 was 80%. We calculated the changes in these dosimetric indices (D90 and V90) and radiobiological outcome (SF2 TCP) as a function of contouring uncertainty, seed localization uncertainty, inability to localize all of the seeds, and binary combinations of these three. The results are presented for a range of uncertainties, which allows the possible application of these results to a variety of imaging modalities that have differing spatial resolutions. We found that both contouring uncertainties and seed localization uncertainties had a large impact on the predicted radiobiological outcome, but that seed localization uncertainties of 6 mm had the largest impact on the dosimetric indices. We also found that the variability in both the predicted radiobiological and dosimetric outcome was largest for contouring uncertainties of 4-8 mm. We conclude that accounting for contouring uncertainties is crucial in accurately deducing the DVHs for post-implant prostate brachytherapy, and hence enabling valid correlation with ultimate clinical outcome.

A quality assurance phantom for three-dimensional radiation treatment planning

PURPOSE Three-dimensional (3D) radiation treatment planning is facilitated through the use of computerized radiation treatment planning systems (RTPSs) and CT simulators (CT-sims). Quality assurance (QA) of these systems is necessary for ensuring that they fulfill their potential. However, comprehensive tools for the systematic QA of these systems have not been developed. We present a phantom that facilitates the evaluation of a large number of nondosimetric functions. These include CT image acquisition and transfer, graphical displays of 3D radiation beams, multiplanar CT image reconstructions, digitally reconstructed radiographs, the representation and manipulation of contoured patient anatomy, dose volume histograms, and the conversion of CT numbers to relative electron densities. METHODS AND MATERIALS A phantom was constructed which contains materials and geometries that are appropriate for the routine QA of the features described above. The anatomy of the phantom is used as a standard against which the performance of the 3D-RTPS or CT-sim is evaluated. The phantom was used to evaluate three different 3D-RTPSs and a CT-sim at four institutions. RESULTS Using this phantom, clinically significant errors and limitations in commercially available 3D treatment planning software were discovered. No errors were discovered in the beam display or image reconstructions in the systems examined. Problems were found in the anatomy display, automatic tools, and the CT number to relative electron density conversion data used in some of the systems. CONCLUSION This phantom is a unique tool designed explicitly for the QA of 3D treatment planning software. Errors and limitations discovered through its use indicate that the QA of commercial treatment planning software is necessary, and that this phantom is an effective device for this task.

Comparison of dose calculation algorithms with Monte Carlo methods for photon arcs

The objective of this study is to seek an accurate and efficient method to calculate the dose distribution of a photon arc. The algorithms tested include Monte Carlo, pencil beam kernel (PK), and collapsed cone convolution (CCC). For the Monte Carlo dose calculation, EGS4/DOSXYZ was used. The SRCXYZ source code associated with the DOSXYZ was modified so that the gantry angle of a photon beam would be sampled uniformly within the arc range about an isocenter to simulate a photon arc. Specifically, photon beams (6/18 MV, 4 x 4 and 10 x 10 cm2) described by a phase space file generated by BEAM (MCPHS), or by two point sources with different photon energy spectra (MCDIV) were used. These methods were used to calculate three-dimensional (3-D) distributions in a PMMA phantom, a cylindrical water phantom, and a phantom with lung inhomogeneity. A commercial treatment planning system was also used to calculate dose distributions in these phantoms using equivalent tissue air ratio (ETAR), PK and CCC algorithms for inhomogeneity corrections. Dose distributions for a photon arc in these phantoms were measured using a RK ion chamber and radiographic films. For homogeneous phantoms, the measured results agreed well (approximately 2% error) with predictions by the Monte Carlo simulations (MCPHS and MCDIV) and the treatment planning system for the 180 degrees and 360 degrees photon arcs. For the dose distribution in the phantom with lung inhomogeneity with a 90 degrees photon arc, the Monte Carlo calculations agreed with the measurements within 2%, while the treatment planning system using ETAR, PK and CCC underestimated or overestimated the dose inside the lung inhomogeneity from 6% to 12%.