Abstract Book

ESTRO 37

S579

2 St George Hospital, St George Cancer Care Centre, Kogarah, Australia 3 Advanced Computer System Pty Ltd, Research, Rome, Italy 4 Czech Technical University of Prague, Institute of Experimental and Applied Physics, Prague, Czech Republic Purpose or Objective BrachyView is a novel in-body imaging system developed with the objective to provide real-time intraoperative dosimetry for LDR prostate brachytherapy treatments. The system can reconstruct seed positions after implantation by the means of a high-resolution pinhole gamma camera 1 . The BrachyView dataset combined with conventional TRUS imaging, provides the possibility to localise the effective position of the implanted seeds inside the prostate volume providing a unique tool for intra-operative verification of the quality of the implantation. An imaging algorithm for baseline subtraction has been implemented to allow for the reconstruction of a full treatment plan with 98 seeds implanted. The overall accuracy of the seed reconstruction algorithm is tested against a CT scan of the phantom after the completion of the implantation. Material and Methods The probe consists of a 1mm thick tungsten cylindrical collimator, containing three single cone pinholes of 500µm in diameter, situated above three Timepix pixelated silicon detectors. To assess the capabilities of the BrachyView system an LDR treatment plan consisting of 98 seeds (I-125 with average activity of 0.248mCi) were implanted into a prostate gel phantom. TRUS images, automatic segmentation and rendering were performed to reconstruct the 3D shape and position of the prostate, utilising transversal 2.5mm thick ultrasound slicing. BrachyView data was acquired for three minutes after each needle implantation to compensate for the low activity of the seeds. A CT study with O-MAR (orthopaedic metal artefact reduction) was completed post implantation to provide a reference for comparison of the BrachyView reconstructed seed positions. Results When compared to the seed positions obtained by the CT scan the BrachyView seed reconstruction showed that the average discrepancies were 2.4 mm, 1.35 mm and 1.2 mm in the x, y and z coordinate directions, respectively. Where 74% and 95% of the seeds are reconstructed within 1.5 mm and 2.0 mm of discrepancy, respectively. Seeds are also co-registered with the TRUS 3D image and displayed within the prostate volume.

Figure 2: Reconstructed prostate volume utilising an in- house automatic segmentation and rendering software Conclusion BrachyView has demonstrated the reconstruction of 100% of the seeds for the LDR patient plan. Co-registration of seed positions and a 3D prostate volume has also been successful. 74% of the seed positions have been reconstructed within 1.5 mm of discrepancy and 95% within 2 mm. The larger discrepancies noted in the x direction can possibly be attributed to a slight rotation of the pinhole collimator. With the implementation of a system that can mechanically fix the TRUS and BrachyView probes to the same stager, it is expected current uncertainties will be minimised. PO-1031 High-dose-rate contact brachytherapy to treat skin carcinomas on flat surfaces J.L. Guinot 1 , M. Tamarit 1 , V. Gonzalez-Perez 2 , M. Peña 1 , L. Oliver 2 , M. Santos 1 , M. Tortajada 1 , L. Arribas 1 1 Fundación Instituto Valenciano de Oncologia, Department of Radiation Oncology, Valencia, Spain 2 Fundación Instituto Valenciano de Oncologia, Department of Radiation Physics, Valencia, Spain Purpose or Objective We present our results with Ir-192 high-dose-rate (HDR) contact brachytherapy (BT) to treat skin carcinomas on flat surfaces. Material and Methods From 2000 to 2016 forty two skin carcinomas in 39 patients were treated with HDR contact brachytherapy employing Freiburg flaps, or customised flaps with plastic tubes on bolus or wax moulds for the skull (Fig 1-2). Mean age was 80 years (range 64-103 years). BT was radical in 20 cases with a maximum tumour thickness of 3 mm; BT was adjuvant in 22 cases after surgery, due to a positive surgical margin in 11 cases, close margin ≤ 1mm in 6, perineural involvement in 4 cases and one with free margins but histology with areas of sarcomatoid carcinoma. Treatment included 26 cases of squamous cell carcinomas (SCC) 15 basal cell carcinoma (BCC) and one Merkel carcinoma. A fractionation of three times a week was chosen, at 3-4Gy per fraction to a total dose of 48-57 Gy in most cases. Large areas of the skull were treated at 2-2.5Gy daily. Prescription was at 3-4mm under the skin Two local recurrences appeared after a mean follow-up of 43 months. The 2 and 3-year actuarial local control rate was 96.6% and 92% (100% for adjuvant cases, and 92.2 and 85.7% for radical BT). No difference was found between SCC and BCC. Two cases of SCC presented regional failure, with 2 and 3-year actuarial regional control rate of 96.2% and 89.7% for SCC and 100% for BCC. Cause-specific survival was 96.6% and 91.7% at 2 surface. Results

Figure 1: CT and BrachyView reconstructed seed discrepancies after co-registration, with average x, y and z errors