[en] Intensity-modulated radiation therapy (IMRT) is commonly delivered using the dynamic or segmental mode of multileaf collimators (DMLC or SMLC). Both methods are designed to deliver intensity-modulated beams as determined by inverse planning software. In this study, we have used the Helios IMRT planning system to generate ideal treatment plans for 10 cases of 2 common treatment sites (prostate and head and neck) and have investigated the actual treatment fluence distributions generated for each of the MLC leaf motion choices. The 2 dose delivery techniques were dosimetrically compared to each other and to the treatment plans. For each technique, point doses were measured in a water phantom using ionization chambers. Also for each technique, 2-dimensional dose distributions at a selected depth in a plastic phantom were obtained, using extended range film. The total delivery time and the number of monitor units (MU) delivered by each method were also compared. Our results indicate that the 2 delivery methods produce comparable results dosimetrically. For the cases reviewed, the delivery time was an average of 15% longer for SMLC deliveries, while the number of MUs (beam-on time) required by SMLC was an average of 15% fewer, than that for the DMLC. In the interest of simplicity, lower beam-on time, and potentially fewer mechanically-related problems, we think that the SMLC delivery technique may be the better choice when Helios is used for planning and Varian linear accelerators are used for delivery

[en] Purpose: Chronic radiation cystitis is an uncommon but debilitating late complication of definitive external beam (EB) and brachytherapy (BT) for cervix cancer. During BT an indwelling catheter is usually placed in the bladder, collapsing it closer to the BT sources. We have devised a method to deliver BT with a full bladder. The difference in bladder dose in the full and empty state were analyzed during definitive EBT and BT for cervix cancer. Methods: The technique of Lyman and Wolbarst (1) were used to evaluate the bladder complication probability for a representative cervix cancer patient undergoing EBT and BT. DVHs were generated from CT scans obtained with a full and empty bladder. Three possible dose prescriptions were analyzed. Results: The DVH for the full and empty situations are shown. With the bladder full, the volume of bladder predicted to receive ≥ 80 Gy was approximately 10% for all dose schemes evaluated, whereas with the bladder empty, up to 50% of the bladder volume received ≥ 80 Gy. Conclusions: A distended bladder improves the DVH. A technique for performing full bladder LDR brachytherapy will be discussed

[en] Purpose: We compare practical conformal treatment approaches to pancreatic cancer using 6 and 18 MV photons and contrast those approaches against standard techniques. Methods and Materials: A four-field conformal technique for treating pancreas cancer has been developed using nonopposed 18 MV photons. This approach has been extended to 6 MV photon application by the addition of one to two fields. These techniques have been optimized to increase sparing of normal liver and bowel, compared with opposed-field methods, to improve patient tolerance of high doses. In this study we compare these techniques in a simulated tumor model in a cylindrical phantom. Dose-volume analysis is used to quantify differences between the conformal, nonopposed techniques with conformal, opposed field methods. This model is also used to evaluate the effect of 1-2 cm setup errors on dose-volume coverage. Results: Dose-volume analysis demonstrates that five-to-six field conformal treatments using 6 MV photons provides similar or better dose coverage and normal tissue sparing characteristics as an optimized 18 MV, four-field approach when 1-2 cm margins are included for setup uncertainty. All approaches using nonopposed beam geometry provide significant reduction in the volume of tissue encompassed by the 30-50% isodose surfaces, as compared with four-field box techniques. Conclusions: Three-dimensional (3D) conformal treatments can be designed that significantly improve dose-volume characteristics over conventional treatment designs without costing unacceptable amounts of machine time. Further, deep intraabdominal sites can be adequately accessed and treated on intermediate energy machines with a relatively moderate increase in machine time

[en] We have studied the effect of target and lung density on block margin for small stereotactic body radiotherapy (SBRT) targets. A phantom (50 × 50 × 50 cm"3) was created in the Pinnacle (V9.2) planning system with a 23-cm diameter lung region of interest insert. Diameter targets of 1.6, 2.0, 3.0, and 4.0 cm were placed in the lung region of interest and centered at a physical depth of 15 cm. Target densities evaluated were 0.1 to 1.0 g/cm"3, whereas the surrounding lung density was varied between 0.05 and 0.6 g/cm"3. A dose of 100 cGy was delivered to the isocenter via a single 6-MV field, and the ratio of the average dose to points defining the lateral edges of the target to the isocenter dose was recorded for each combination. Field margins were varied from none to 1.5 cm in 0.25-cm steps. Data obtained in the phantom study were used to predict planning treatment volume (PTV) margins that would match the clinical PTV and isodose prescription for a clinical set of 39 SBRT cases. The average internal target volume (ITV) density was 0.73 ± 0.17, average local lung density was 0.33 ± 0.16, and average ITV diameter was 2.16 ± 0.8 cm. The phantom results initially underpredicted PTV margins by 0.35 cm. With this offset included in the model, the ratio of predicted-to-clinical PTVs was 1.05 ± 0.32. For a given target and lung density, it was found that treatment margin was insensitive to target diameter, except for the smallest (1.6-cm diameter) target, for which the treatment margin was more sensitive to density changes than the larger targets. We have developed a graphical relationship for block margin as a function of target and lung density, which should save time in the planning phase by shortening the design of PTV margins that can satisfy Radiation Therapy Oncology Group mandated treatment volume ratios

[en] Purpose: To evaluate how changes in the measured small field output factors affect the doses in intensity-modulated treatment planning. Methods: IMRT plans were created using Philips Pinnacle treatment planning system. The plans were optimized to treat a cylindrical target 2 cm in diameter and 2 cm in length. Output factors for 2 × 2 and 3 × 3 cm² field sizes were changed by ±5%, ±10%, and ±20% increments from the baseline measurements and entered into the planning system. The treatment units were recommissioned in the treatment planning system after each modification of the output factors and treatment plans were reoptimized. All plans were delivered to a solid water phantom and dose measurements were made using an ionization chamber. The percentage differences between measured and computed doses were calculated. An Elekta Synergy and a Varian 2300CD linear accelerator were separately evaluated. Results: For the Elekta unit, decreasing the output factors resulted in higher measured than computed doses by 0.8% for −5%, 3.6% for −10%, and 8.7% for −20% steps. Increasing the output factors resulted in lower doses by 2.9% for +5%, 5.4% for +10%, and 8.3% for +20% steps. For the Varian unit no changes were observed for either increased or decreased output factors. Conclusions: The measurement accuracy of small field output factors are of importance especially when the treatment plan consists of small segments as in IMRT. The method proposed here could be used to verify the accuracy of the measured small field output factors for certain linear accelerators as well as to test the beam model. The Pinnacle treatment planning system model uses output factors as a function of jaw setting. Consequently, plans using the Elekta unit, which conforms the jaws to the segments, are sensitive to small field measurement accuracy. On the other hand, for the Varian unit, jaws are fixed and segments are modeled as blocked fields hence, the impact of small field output factors on IMRT monitor unit calculation is not evaluable by this method.

[en] Purpose: To evaluate the acute toxicity of a group of 37 pancreatic cancer patients treated with noncoplanar, nonopposed, conformal radiation therapy with concurrent chemotherapy (5-FU). Materials and Methods: We retrospectively evaluated a group of initially nonadvanced 37 pancreatic cancer patients treated with combined concurrent chemotherapy and 3D radiation therapy treated between 1992 until 1995. During this period we began treating the initially unresectable patients with preoperative chemo-RT (50.4 Gy) after treating an initial group of unresectable patients to a higher dose of 66.6 Gy. We also include a group of patients who received postop chemo-RT after Whipple resection (59.4 Gy). All radiation was delivered at a 1.8 Gy per fraction dose rate. The total group was made up of 37 patients of whom 21 were male (57%) and 16 female (43%). There were 22 (59%) head of pancreas lesions, 10 (27%) body of pancreas lesions, and 5 (14%) head and body of pancreas cancers. Of these 37 patients 7 (19%) were treated with chemo-RT as their only treatment, 10 patients (29%) were treated post Whipple resection, and 20 patients (54%) were treated with preoperative intent. Results: Three patients (8%) required a treatment break, one with a body and 2 with head lesions. Two of these patients stopped RT short of planned dose (32.56 and 46.8 Gy) both suffering from nausea, vomiting, and anorexia with the third, who finished a planned 66.6 Gy dose, after a 4 day rest for leukopenia. One of 20 patients (5%) preop patients underwent the planned post chemo-RT Whipple resection, while 4 of the 20 patients (20%), remained unresectable, but without disease progression and had Iodine 125 interstitial implants at exploration delivering a minimal tumor dose of 120 Gy on top or the 50.4 Gy delivered preoperatively. Four patients (11%) maintained a minimal Karnofsky score of 100, 23 patients (62%) maintained a minimal KPS of 90, 6 patients (16%) maintained a minimal KPS of 80, and 4 patients (11%) had a minimal KPS =/< 70. Seven of the 37 patients (19%) either lost no weight or gained weight during the chemo-RT, while the mean weight lost was 6.9 lb. with a median weight loss of 6.0 lb. for the entire group. Eighteen patients (49%) required narcotic analgesics during chemo-RT, 9 patients (24%) had ECOG grade 1 or 2 thrombocytopenia. Of those 16 patients receiving a 3 day bolus of 5-FU, 6 of 16 (38%) had thrombocytopenia while of the 96 hr infusion group 4 of 19 (21%) patients had thrombocytopenia. Eight patients had leukopenia of whom 3 had ECOG grade 3 and 4 (both on 3 day bolus 5-FU). Of 6 of 37 (16%) patients developing anemia during chemo-RT 5 were treated with 3 day bolus and one with the 96 hr infusion 5-FU. Nine of 37 (24%) patients suffered grade 0 or 1 diarrhea. Of the two patients treated with prolonged venous infusion 5-FU (225 mg/m2/d) during RT, neither developed any hematologic toxicity. Conclusions: Noncoplanar, nonopposed, conformal radiation therapy can be used with acceptable acute toxicity in the treatment of pancreatic cancer, whether the chemo-radiation is delivered preoperatively, postoperatively, or as the only treatment. The relatively acceptable acute toxicity with the 3D RT approach may allow for more effective and maybe even more potentially toxic treatment strategies to be used in combination with 3D RT

[en] Purpose: To evaluate the efficacy of stereotactic radiotherapy (SRT) in patients with recurrent high-grade gliomas by comparing two different treatment regimens, single dose or fractionated radiotherapy. Methods and Materials: Between April 1991 and January 1998, 71 patients with recurrent high-grade gliomas were treated with SRT. Forty-six patients (65%) were treated with single dose radiosurgery (SRS) and 25 patients (35%) with fractionated stereotactic radiotherapy (FSRT). For the SRS group, the median radiosurgical dose of 17 Gy was delivered to the median of 50% isodose surface (IDS) encompassing the target. For the FSRT group, the median dose of 37.5 Gy in 15 fractions was delivered to the median of 85% IDS. Results: Actuarial median survival time was 11 months for the SRS group and 12 months for the FSRT group (p = 0.3, log-rank test). Variables predicting longer survival were younger age (p = 0.006), lower grade (p = 0.0006), higher Karnofsky Performance Scale (KPS) (p = 0.0005), and smaller tumor volume (p 0.02). Patients in the SRS group had more favorable prognostic factors, with median age of 48 years, KPS of 70, and tumor volume of 10 ml versus median age of 53 years, KPS of 60, and tumor volume of 25 ml in the FSRT group. Late complications developed in 14 patients in the SRS group and 2 patients in the FSRT group (p < 0.05). Conclusion: Given that FSRT patients had comparable survival to SRS patients, despite having poorer pretreatment prognostic factors and a lower risk of late complications, FSRT may be a better option for patients with larger tumors or tumors in eloquent structures. Since this is a nonrandomized study, further investigation is needed to confirm this and to determine an optimal dose/fractionation scheme

[en] Purpose: To evaluate the efficacy of stereotactic radiosurgery (SRS) in patients with solitary brain metastasis from extracranial primary cancer and to compare the outcome with that of external whole brain irradiation with or without surgical resection. Materials and Methods: Between September 1970 and November, 1995, 231 patients with solitary brain metastasis were treated at the Department of Radiation Oncology, University of Minnesota Hospital. One hundred twenty six patients (56%) were treated with external whole brain irradiation (WBI) only (Group 1), seventy three (32%) underwent surgical resection prior to WBI (Group 2) and thirty two (14%) underwent stereotactic radiosurgery (SRS) with WBI (Group 3). Lung (38%) was the most common site of primary cancer, followed by breast (15%), unknown primary (12%), gastro-intestinal tract (10%), skin (malignant melanoma: 9%), kidney (renal cell carcinoma: 8%) and others (9%). The median dose to the whole brain was 3750 cGy in 15 fractions (ranges from 2000 cGy to 5000 cGy). The median radiosurgical dose of 17.5 Gy (range, 12-40 Gy) was delivered to the 40%-90% isodose line encompassing the target. Eighteen patients were treated with SRS for recurrent or persistent disease following WBI and 14 patients received SRS as a boost in conjunction with WBI. Actuarial survival was calculated from the date of treatment according to the Kaplan-Meier method and statistical significance was assessed with the log-rank test. Results: The actuarial median survivals were 3.8 months for Group 1 (ranges from 1 to 84 months), 10.5 months for Group 2 (ranges from 1 to 125 months) and 9.8 months for Group 3 (ranges from 1 to 36 months). The survivals at one and two years were 19% and 6% for Group 1, 47% and 19% for Group 2, and 44% and 21% for Group 3, respectively. The survival advantage of Groups 2 or 3 over Group 1 was statistically significant (p < 0.0001 by log-rank test). There was no survival advantage of surgery (Group 2) over SRS (Group 3) (p=0.69 by log-rank test). Conclusion: Although it is difficult to draw a definite conclusion from a retrospective analysis of an unbalanced number of patients among the Groups, our data suggest that SRS improves survival when compared to WBI only and is a reasonable alternative to surgery in the management of solitary brain metastases

[en] Most current electron beam models, as are used in commercial treatment planning systems, combine measured broad beam central axis depth dose data with measured or modeled functions to approximate radial scatter and heterogeneity effects. In this paper, we extend a recently developed pencil beam model to calculate doses outside the field edge and doses in heterogeneous media. We have also explored use of this model as a tool for evaluating commercial electron planning programs. The algorithm we have developed, based on the concept of the lateral buildup ratio (LBR), enables calculation of dose at any point in an irregular electron field, and is capable of generating both on- and off-axis depth dose curves and isodose profiles. This model includes the effects of density and mass-angular scattering power in measured broad beam central axis depth dose data, which when combined with small field reference data, can be used to generate LBR ratios. From these ratios one can infer the depth dependent, effective pencil beam radial spread parameter σ in water or other materials, which can be used to model any arbitrary field. We have used this approach to calculate fractional depth doses for small fields incident on aluminum and cork, which we have then compared against measurements and the calculations of several commercial planning systems

[en] Heterogeneous lung and bone phantoms have been constructed for the purpose of testing monitor unit calculations at or near interfaces for different planning systems. Data have been acquired for 2 linear accelerators: a Varian 2300cd (6 and 25 MV) and an Elekta Synergy (6, 10, and 18 MV). We have reviewed Pinnacle and the correction-based, pre-AAA version of Eclipse planning systems with the intent of exploring the limits of these systems with energy and field size. Data were acquired from 2 x 2 to 10 x 10 cm² field sizes over the available range of energies. Our measurements confirm that Pinnacle predicts doses mostly to within ± 5%, even near lung-tissue interfaces over the full range of energies and field sizes tested. The Eclipse-modified Batho and equivalent TMR algorithms overpredicted doses by 10% or more in the lung and near the lung-tissue interfaces if the field size was less than 10 x 10 cm² when the energy was 18 MV or higher. At lower energies, the field size had to be at least 6 x 6 cm² for calculated doses to be within 10% of measurement. For bone-tissue interfaces, doses were generally underestimated by 5% to 10% or more by all calculation methods over the range of field sizes and energies reviewed. A second goal of this study was to review methods for hand-checking monitor units when heterogeneities are included. We evaluated the range of applicability of 2, one-dimensional (1D) inhomogeneity correction factors: the effective attenuation method and the TMR ratio method. The effective attenuation method for monitor unit checking was within ± 5% to as small as 6 x 6 cm² fields for 6 to 10 MV, useable for 4 x 4 cm² fields (within 7%) for 6 MV and close to ± 5% for 10 x 10 cm² fields in the 18- to 25-MV range. The TMR ratio method was not as good, being within about ± 5% to 7% of measurements only for 6 x 6 to 10 x 10 cm² fields at 6 MV and 10 x 10 cm² fields at the higher energies. Both simple 1D correction methods performed almost as well as Pinnacle for the bone-soft tissue cases. We recommend that if direct measurement of dose for heterogeneous treatment plans is not practiced, then one of these simple cross checks be performed to assure patient safety.