Retrospective analysis of dose delivery in intra-operative high dose rate brachytherapy Moonseong Oh, Jaiteerth S. Avadhani, Harish K. Malhotra, Barbara Cunningham, Patrick Tripp, Wainwright Jaggernauth, Matthew B. Podgorsak Department of Radiation Medicine, Roswell Park Cancer Institute, Buffalo, New York, USA Background. This study was performed to quantify the inaccuracy in clinical dose delivery due to the incomplete scatter conditions inherent in intra-operative high dose rate (IOHDR) brachytherapy. Methods. Treatment plans of 10 patients previously treated in our facility, which had irregular shapes of treated areas, were used. Treatment geometries reflecting each clinical case were simulated using a phantom assembly with no added build-up on top of the applicator. The treatment planning geometry (full scatter surrounding the applicator) was subsequently simulated for each case by adding bolus on top of the applicator. Results. For geometries representing the clinical IOHDR incomplete scatter environment, measured doses at the 5 mm and 10 mm prescription depths were lower than the corresponding prescribed doses by about 7.7% and 11.1%, respectively. Also, for the two prescription methods, an analysis of the measured dose distributions and their corresponding treatment plans showed average decreases of 1.2 mm and 2.2 mm in depth of prescription dose, respectively. Conclusions. Dosimetric calculations with the assumption of an infinite scatter environment around the applicator and target volume have shown to result in dose delivery errors that significantly decrease the prescription depth for IOHDR treatment. Key words: intraoperative period; brachytherapy; radiotherapy dosage Introduction Intra-operative radiation therapy (IORT) is the delivery of a relatively high dose of radiation to the tumour bed or residual disease at the time of surgical resection. The benefit of this technique is the po- Received 12 October 2007 Accepted 19 October 2007 Correspondence to: Matthew B. Podgorsak, Department of Radiation Medicine, Roswell Park Cancer Institute, Elm & Carlton Streets, Buffalo, NY 14263, USA; Phone: +1 716 845 1536; Fax: +1 716 845 7616; E-mail: moonseong.oh@roswellpark.org tential to shield or displace normal tissues thus minimally exposing them to radiation. Clinically, IORT has been used as an adjuvant to surgery and/or fractionated external beam radiation therapy for locally advanced cancers of the abdomen, pelvis, head and neck, brain, thorax, and extremities.1-7 Historically, linear accelerators employing electron beams were used for IORT.5, 8-10 However, there has been an interest in applying high dose rate (HDR) brachytherapy for this purpose.11-14 In intra-operative high dose rate (IOHDR) brachytherapy, applicators are Oh M et al. / Dose delivery in intra-operative high dose rate brachytherapy Table 1. Foreshortening of the depth of prescription dose ients Intended prescription depth (mm) Prescription dose (cGy) d (Prescription distance from the plan (mm)) d' (Actual distance (mm)) d-d' (mm) A 10 1500 9.8 7.8 2.0 B 10 1000 10 7.6 2.4 C 10 500 10 7.7 2.3 D 10 1000 9.9 8.0 1.9 E 10 1250 10 7.9 2.1 F 10 750 10 7.8 2.2 G 10 1250 10 7.8 2.2 H 10 750 10 7.7 2.3 I 5 1250 5 3.8 1.2 J 5 1500 5 3.7 1.3 secured directly to the residual tumour or tumour bed. The region anterior to the applicators is mostly air with significantly less scattering properties than tissue. The dose computation algorithm in commercial treatment planning system assumes that the applicators are surrounded by an infinite scatter medium.15 This assumption, however, is strictly valid only in cases of interstitial and intracavitary brachytherapy and may lead to an over-estimation of the dose in the case of IOHDR brachytherapy. In a recent publication,16 we have shown that this lack of scatter from one side of the applicator has the potential of leading to significant underdosage during treatments. Our measurements showed that underdos-ages at two planned prescription depths (5 mm and 10 mm) were 8.5% and 12.5% for each of the conventional treatment geometries studied (applicators with surface areas of 4, 7, and 12 cm2). In a clinical environment, IOHDR brachytherapy treatments typically involve irregular surface areas and there has been concern whether the previous published results with standard irradiation geometries can be ported to these clinical situations as well. In the present study, we have used an experimental approach to quantify the magnitude of underdosage in clinical cases with irregularly shaped applicators. Materials and methods In this retrospective study, the treatment plans of 10 consecutive patients previously treated at our facility were analyzed. Eight patients had a prescription depth of 10 mm where the therapy was delivered out using applicators (Freiburg Flap Applicator, Freiburg, Germany) consisting of a contiguous array of 5 mm radius plastic spherical beads, which have a provision to insert multiple nylon catheters separated by 10 mm. The remaining two patients were treated without an applicator, resulting in a prescription depth of 5 mm. In our practice, the prescription depth is defined as the distance between the center of the source dwell positions and the treatment plane. The clinical set up of a representative IOHDR brachytherapy treatment is shown in Figure 1. In this study, the prescribed dose varied from 5 Gy to 15 Gy (Table 1). The treated areas were irregular in shape, covering surface areas of 8 cm2 to 180 cm2 Figure 1. Clinical setup of a representative IOHDR case. and consisted of 3 to 16 catheters, depending on the size of the target volumes. The clinical treatment plans for each patient were restored to the planning system (Plato, v. 14.2, Nucletron, Columbia, MD) and were renormalized to deliver a dose of 200 cGy to the original prescription depth. The measurement setup is shown in Figure 2. For patients with a 10 mm prescription depth, the treatment delivery was simulated Figure 2. The measurement setup: (a) Full scatter environment, (b) No scatter environment. Radiol Oncol 2007; 41(4): 188-95. by inserting 5 mm bolus material posterior to the applicator to achieve the prescribed distance from the center of the source to a piece of radiographic film. EDR2 films used to get dose profiles were placed on top of a solid water phantom (30 cm x 30 cm x 15 cm) and full scatter conditions were obtained by putting 15 cm of bolus material on the top of the applicator. The H & D curve for EDR2 film for Ir-192 was generated using a reference applicator having a treatment area of 7 x 7 cm2 and a prescription depth of 10 mm with the prescribed dose varying from 0 to 400 cGy (Figure 3). Fifteen centimeters bolus on top of the applicator was used to simulate the H & D Curve ^r- (/> e