172 poster Evaluation of technical failures during treatment in pulsed dose rate brachytherapy

$80

Objective: The aim of the project was to analyze survival, local control, prognostic factors and toxicity of patients with cervical cancer treated with brachytherapy. Material and methods: The analysis included 131 patients diagnosed with-cervical cancer treated in our Hospital between June 1991 to December 2002. The median age was 53 years (30-79). Distribution by stages were as follows: stage Ib 52%, Ila 11%, lib t9%, Ilia 1% and IIIb 17%. Histologically 85% were epidermoid carcinoma. 68% patients underwent surgical resection (total abdominal hysterectomy _+ bilateral salpingooophorectomy _+ lymphadenectomy). 11% patients received concurrent chemoradiation and surgical resection subsequently. 21% remaining were treated with external radiotherapy exclusively. All patients received intracavitary brachytherapy (LDR). Dose between 20-25 Gy was indicated if external radiotherapy was administrated and 50 Gy (44-50 Gy) in two insertions of 25 Gy when external radiotherapy wasn't used. Intracavitary treatment was given with Delouche applicators loaded with cesium 137 (remote after loading technique). External irradiation was received with 60Co or 25 MV photons with a median dose of 50 Gy (44-50). Overall survival, 5-year disease-free survival , local control, prognostic factors and toxicity were evaluated. Degree of differentiation, histology of the tumor, stage of disease, performed treatment, type of surgery, margins and age were also considered. Results: The median duration of follow-up was 4 years (0.4-12). Overall survival rate and 5-year disease-free survival rate were 70% and 74% respectively. 26% patients had global recurrence with a 3.7% local disease recurrence. Stage of disease, performed treatment, type of surgery and age were the prognostic factors which predicted overall survival in the univariate analysis. Multivariate analysis showed that stage of disease and age were independent prognostic factors. About survival free of disease, stage of disease and type of surgery were statistically significant. Overall incidence of complications was found as 10%. The grade 1 complication rate was 7%. Conclusion: LDR brachytherapy and external radiotherapy for cervical cancer providing good results in local control and overall survival with excellent tolerance and low morbidity. Our results are similar to other historical series. 171

Intracavitary brachytherapy for endometrial cancer. Experience in the department of Radiation Oncology. H.U. "Reina Sofia", Cordoba. S. Garcia, A. Bejar, A. Palacios, E. Munoz, L. Tudela, M. Rodriguez, M. Espinosa, M.D. Lopez, J.L. Romeo Radiation Oncology, H.U. "Reina Sofia" Cordoba, Spain Objective: The purpose of this work was to determine survival, local control, prognostic factors and toxicity of patients with cancer of the endometrium treated with brachytherapy. Material and methods: From June 1991 to December 2002 a total of 274 patients with endometrial carcinoma were treated in our department. The median age was 64 years (39-83). Histologically 91% were adenocarcinoma and the degree of differentiation was 76% G1-G2. Distribution by Stages were as follows: stage Ib 23%, Ic 38%, Ila 15%, lib 8%, Ilia 10%, IIIb 2% and IIIc 3%. 93% patients underwent surgical resection and external radiotherapy and/or brachytherapy subsequently. 19 patients received radiotherapy alone. All patients received intracavitary brachytherapy (LDR). A dose between 10-25 Gy was indicated if external radiotherapy was administrated and 50 Gy in two insertions of 25 Gy when external radiotherapy wasn't used. Intracavitary treatment was given with Delouche applicators loaded with cesium 137 (remote after loading technique). 85.8 % received external beam pelvic radiotherapy with 60Co or 25 MV photons with a median dose of 50 Gy (44-

Posters

50). Overall survival, 5-year disease-free survival, local control, prognostic factors and toxicity were evaluated. Degree of differentiation, histology of the tumor, stage of disease, performed treatment, type of surgery, margins and age were also considered.

Results: The median duration of follow-up was 4.1 years (0.512.4). The overall survival rate and 5-year disease-free survival rate were 82.8% and 82.9% respectively. 15% patients had global recurrence with a 3.2% local disease recurrence. Overall incidence of complications was found as 9% with 6% grade 1. Conclusion: Our experience suggests that LDR brachytherapy and external radiotherapy for endometrial cancer are safe and effective in the management of endometrial cancer.

Physics 172 poster Evaluation of technical failures during treatment in pulsed dose rate brachytherapy

K. Koedooder, R. Tiggelaar, J. van der Grient, N. van Wieringen, G. D'Olieslager, L. Blank, B. Pieters Academisch Medisch Centrum, Radiotherapy, Amsterdam, The Netherlands Purpose: To determine the occurrence frequency of different types of failures (error calls, error codes) during treatment in Pulsed Dose Rate brachytherapy and its dependence on treated tumorsite. Introduction: About 700 patients were treated in our hospital with Pulsed Dose Rate (PDR) brachytherapy. The average frequency is 3 patients/week. Initially treatment was given only during office hours. Over the last two years all patients were treated in continuous day and night schemes. Treatments were monitored by an extensive, software driven, security system within the afterloaders. This system continuously controls the source's positional status and checks whether the route to its intended position is not obstructed. Failure conditions correspond with uniquely defined error codes. Materials and methods: Treatments were performed with two first generation Micro Selectron PDR afterloaders (NucletronOldelft, the Netherlands), containing respectively a 37 GBq and a 74 GBq stepping source. Within the patient population 15 different tumor sites were treated, the five most frequently being breast, oesophagus, prostate, vagina and bladder. Planned number of pulses and overall treatment time ranged from 4 pulses in about 30 minutes for oesophagus carcinoma, to 62 pulses in 135 hours for anal carcinoma. To date, about 500 patients were entered in a database. Apart from items regarding the medical and demographic status of the patient, all physicsrelated and technical parameters were entered for each treatment. From this database, queries were extracted to determine for each tumor site the percentage of disturbed treatments and for disturbed treatments the percentage of disturbed pulses. Results: Error frequency was strongly dependent on type of tumor site treated. Bladder and prostate were more error prone than breast or oesophagus. For breast the first error most often occurred in the first pulse. For prostate and bladder however, the occurrence of the first error was shifted towards a further stadium of treatment. A general conclusion is that 98% of the patients received the intended dose correctly. Only in 2% of the treatments errors caused adjustments of implant geometry, administered dose or cancellation of part of the treatment. Due to extensive dummy run checks, errors with the active source occurred only rarely and if occurring, they were adequately solved by the sytem's software driven mechanics.

Posters

$81

Conclusions: Errors do occur and are tumor site dependent. This experience restarted the discussion in our department to return to office hour schemes for certain treatment sites or adjustment of the implantation technique. First errors do not always occur in the first pulse, meaning that it cannot be assumed that an error free start of treatment guarantees a complete error free treatment. 173 poster Phantom size in brachytherapy source dosimetric studies

F. Ballester 1, D. Granero 2, J. Perez-Calatayud 2, E. Casal 2, V. Puchades z, R. Cases ~ ~University of Valencia - CSIC, IFIC, Burjassot, Spain 2University of Valencia, Department of Atomic, Molecular and Nuclear Physics, Burajassot, Spain Background. An important point to consider in a brachytherapy dosimetric study is the phantom size involved in calculations or experimental measurements. As pointed out by Williamson [Med. Phys. 18, 776-786(1991)] this topic has a relevant influence on final dosimetric results. Presently, one-dimensional algorithms and newly-developed 3-D correction algorithms are based on physics data that are obtained under full scatter conditions, i.e., assuming infinite phantom size. One can then assume that reference dose distributions in source dosimetry for photon brachytherapy should obtained using an unbounded phantom size. Presently, most published Monte Carlo (MC) dosimetry studies have been obtained using a spherical phantom with a radius of 15 cm or a cylinder phantom of 40cm'40cm. Purpose. The aim of this paper is to study the effect of phantom size in the dose distributions calculated for Cs-137, Ir192, 1-125 and Pd-103, mainly used radionuclides for clinical purposes in brachytherapy. Material and Methods. The MC code GEANT4 has been used in this study. A point source is assumed. This source is located at the centre of a liquid-water spherical phantom. Dose is approximated by kerma and the linear track-length kerma estimator is used. Due to the spherical symmetry a grid system of spherical shells 0.05 cm thick was used. For Pd-103 and 1125 we have used phantom radii of 5, 10, 15, 20, 25 and 30 cm. For Cs-137 and Ir-192 sources, the phantom radii selected are 10, 15, 20, 25, 30, 35, 40, 45 and 50 cm. Results and Conclusions. We have found that for Cs-137 and Ir-192, a spherical phantom with a 40 cm radius is equivalent to an unbounded phantom up to a distance of 20 cm from the source, as this size ensures full scatter conditions at this distance. For 1-125 and Pd-103, the required radius for the spherical phantom in order to ensure full scatter conditions at 10 cm from the source is 15 cm. Differences for Ir-192 and Cs137 sources, in the case of the most extended phantom size cited in literature (a sphere with R = 15 cm), reach 7% (Ir-192) and 4.5% (Cs-137) at r = 10 cm, and 2.5% (Ir-192) and 1% (Cs137) at r = 5 cm. A simple expression based on fits to the dose distributions for various phantom sizes has been developed for Cs-137 and Ir-192 in order to compare the dose rate distributions published for different phantom sizes. Using these relations it is possible to obtain radial dose functions for unbounded medium from bounded phantom ones. 174 poster Practising the LD dosimetry prostatic seed implants

concept

for

permanent

A. van 't Riet ~, C. Hoekstrd, J. Immerzeef , L. Elders ~, R. Westendorp ~, F. Ypma z

1Radiotherapeutic Institute R/SO, Deventer, The Netherlands 2Deventer Hospitals, Urology, Deventer, The Netherlands Purpose: To demonstrate the usefulness of the LD dosimetry concept as a rigid system for treatment dose prescription in prostate implants to arrive at a well defined reported dose e.g. 145 Gy. Methods and materials: In this study, the dose (LD) at half the peak height in the natural dose volume histogram, as introduced by Anderson, is applied for dose specification. LD represents the lower dose side of a relatively homogeneous dose region and corresponds with the Dref (0.85Db) of the Paris system of dosimetry. In the LD dosimetry concept the source arrangement is chosen in such a way that the dose of 145 Gy corresponds with LD and at the same time covers the target volume. Twenty-four patients previously planned subjectively for dose coverage by a dose of 145 Gy and implanted accordingly were recalculated retrospectively using the LD dosimetry concept. The calculations were carried out for prostate volumes ranging from 21.7 cm 3 to 56.3 cm 3. Results: In the 24 patients included in this study the plans, judged subjectively for a treatment dose of 145 Gy, resulted in a mean LD of 167.3 Gy (s=13.0 Gy), a mean Dgo of 171.2 Gy (s=8.2 Gy) and a mean V~o of 56.3% (s=7.7%). For the recalculated dose distributions applying the LD dosimetry concept, the reported dose of 145 Gy corresponds to LD for all patients involved together with a CTV coverage of at least 95%. A significant lower Dgo averaging 155.7 Gy (s=3.0Gy) and V~o averaging 34.4% (s=9.8%) were found, indicating a more homogeneous dose distribution. Conclusions: The strict rules of a well defined dose specification system like the LD dosimetry concept can easily be applied to clinical situations, thus facilitating the comparison of reported doses. 175 poster The BCCA's HDR breast brachytherapy program - A physicist's view

W. Ansbacher 1, C.C. Popescu I, M. Panades 2, P.T. Truong 2, H.A. Kader ~ IBC Cancer Agency, Vancouver Island Centre, Medical Physics, Victoria, BC, Canada 2BC Cancer Agency, Vancouver Island Centre, Radiation Oncology, Victoria, BC, Canada In May 2003 we began a pilot study to assess HDR brachytherapy as the sole post-operative radiotherapy following breast conserving surgery. The first and main objective of the study was to pilot a standard breast brachytherapy technique that is safe, accurate, and reproducible. The second objective, which is the focus of this presentation, was to evaluate the overall procedure for accuracy and consistency of treatment, and to establish whether the resulting workload from its adoption within the Agency would be practicable. The treatment planning part of the procedure involves a preimplant CT scan some 4-8 weeks after surgery, when a radioopaque grid and corresponding skin marks are used to cover the likely entrance and exit sites of the brachytherapy catheters. The oncologist delineates the planning target volume (PTV), and an idealised pre-plan is then generated to encompass this PTV. The needle entrance and exit points are visualized on a 3D surface reconstruction, and a hard copy is produced to use as a guide for actual needle insertion. No template is used for the insertion. The implant consisted of 9 to 17 flexible catheters, most often in three planes. A post-implant CT is registered with the initial scan to allow accurate contouring of the PTV. The treatment plan is generated using standard contouring methods