ESTRO 38 Abstract book

S445 ESTRO 38

to the first 5fx) at treatment planning. Interfractional variabilities resulted in an increase of D 50 (0.40±0.38Gy) and a decrease of D 95 (2.03±5.40Gy) to the PTV. The bladder volume showed the biggest interfractional volume variability (286±168ml), effecting a dose deviation of at mean 7.76±6.05Gy of the D 50 . By using a plan database, the applied dose of the bladder could be reduced by 4.55 ±5.81Gy. The D 50 of the PTV was marginally lower (- 0.23±0.31Gy) and the D 95 of the PTV was marginally higher (1.0±0.9Gy) than without using plan database and thus closer to the planning value. Conclusion The observed variability resulted in significant dose increases of the D 50 to the bladder, whereas in the PTV, only small non-significant dose deviations could be detected. By including the bladder volume in the daily choice of the treatment plan - especially for patients with a challenging organ situation - a significantly lower dose to the OAR was achieved, while the target volume coverage was virtually unchanged. Further analyses will reveal a potential correlation of the observed dosimetric benefits with reduced clinical toxicities. PO-0847 Rapid modulation of PSMA expression by ADT: Serial PSMA PET in men commencing androgen blockade. L. Emmett 1 , V. Chalasani 2 , A. Kneebone 3 , G. Hruby 3 , A.M. Joshua 4 1 St Vincent's Hospital- Sydney- Australia, Department of Theranostics, Sydney, Australia ; 2 Royal North Shore Hospital- Sydney, Urology, Sydney, Australia ; 3 Royal North Shore Hospital, Radiation Oncology, Sydney, Australia ; 4 St Vincent's Hospital Sydney, Oncology, Sydney, Australia Purpose or Objective Prostate specific membrane antigen (PSMA) can be targeted for both imaging and therapy purposes in the management of prostate cancer(PCa). In pre-clinical models, androgen blockade appears to increase expression of PSMA in both hormone sensitive and castrate resistant xenotypes. The aim of this study was to prospectively evaluate the effect on PSMA PET in men with measurable metastatic disease commencing initial androgen blockade in the hormone sensitive cohort or those commencing a novel antiandrogen (enzalutamide or abiraterone) in the castrate resistant cohort. Material and Methods Serial PSMA PET scans were performed at baseline, and days 9, 18 and 28 in 8 men with measurable metastatic hormone sensitive PCa commencing ADT (cohort 1) and 7 men with castrate resistant PCa commencing either enzalutamide or abiraterone (cohort 2). Gleason score (GS), age, time since diagnosis and prior treatments were documented. Testosterone and PSA (ng/ml) were measured at baseline and at all imaging time points. PET/CT was quantitatively analysed using MIM® software for number of lesions, SUV max, SUV mean and total tumour volume. Results Cohort 1: A mean 25% (IQR) reduction in intensity (SUV max) of PSMA by day 9 post androgen blockade (AB). This reduction in PSMA SUV max was seen in 88 % (7/8) men by day 9, with an associated PSA response in 100% men. There was more heterogeneity on day 18 and 28 PSMA with an increase in PSMA SUV max in 2/8 (25%) men, although overall tumour volume reduced in all men. Cohort 2: A mean 28% (IQR) increase in intensity of PSMA SUV was recorded by day 9 post AB. All men demonstrated an increase in both the intensity of uptake and volume on PSMA PET compared to baseline. This increase in quantitative parameters occurred by day 9 in all men, and had reduced by day 18 and 28 imaging. PSA responses were more delayed in cohort 2, with 2/7 men demonstrating PSA progression on the study.

(0.52±0.39Gy; 2.97±2.88Gy) to the PTV compared to doing daily fx-CTs for primary and salvage treatments, which lead to a worse target volume coverage. For primary treatments, the applied D 50 to the bladder was reduced (1.22±1.32Gy) by doing weekly fx-CTs instead of daily fx- CTs, while the applied D 50 to the rectum was increased (2.40±2.21Gy). For salvage treatments, the applied D 50 to the bladder was increased (1.13±3.64Gy) by doing weekly fx-CTs instead of daily fx-CTs, while the applied D 50 to the rectum was reduced (0.14±1.47Gy). Doing once-a-week fx- CTs affected a decrease (3%) of the gamma passing rate compared to doing daily fx-CTs. Conclusion The observed variability resulted in significant dose increases and decreases of the D 50 to the bladder and rectum, respectively. It could be shown that the dose decreases of the D 50 and D 95 to the PTV were statistically significant. Doing daily fx-CTs affected a higher gamma passing rate. Adaptive treatment planning may help to minimize the observed dosimetric deviations. PO-0846 Dosimetric effects of a novel concept of adaptive radiotherapy for prostate cancer patients M. Splinter 1,2,3 , T. Bostel 2,4 , C. Lang 1,2,3 , P. Häring 1,2,3 , J. Debus 2,3,5 , N.H. Nicolay 2,6 1 German Cancer Research Center DKFZ, Division of Medical Physics in Radiation Oncology, Heidelberg, Germany ; 2 German Cancer Research Center, Clinical Cooperation Unit "Radiation Oncology", Heidelberg, Germany ; 3 Heidelberg Institute for Radiooncology, National Center for Radiation Research in Oncology, Heidelberg, Germany ; 4 Mainz University Medical Center, Department of Radiation Oncology, Mainz, Germany ; 5 University Hospital Heidelberg, Department of Radiation Oncology, Heidelberg, Germany ; 6 Freiburg University Medical Center, Department of Radiation Oncology, Freiburg, Germany Purpose or Objective Interfractional variations in bladder and rectal anatomy challenge current concepts of high-precision radiotherapy for prostate cancer patients, as the delivered dose may deviate from the planned dose. As the precise effects of these dose deviations are largely unknown and therefore plan adaptation strategies are difficult to evaluate, this analysis was aimed at investigating dosimetric consequences of a novel method for adaptive radiotherapy using an individualized plan database. Material and Methods The data of ten patients with prostate carcinoma (salvage radiotherapy to the prostatic bed, 68Gy in 34 fx, step-and- shoot IMRT) were investigated. Prior to each fraction, a diagnostic position-control-CT (fx-CT) was performed using a diagnostic in-room CT scanner (SOMATOM ® Emotion Syngo, SIEMENS). Based on the daily fx-CT, the target volume and all OARs were contoured and volumetrically compared to the structures of the planning CT. One treatment plan was calculated based on the planning CT, and two additional plans were calculated based on the bladder filling observed using the first five fx-CTs, thereby creating a plan database that contained plans for low, intermediate and high bladder volume for each patient. Using a deformable registration algorithm for each daily fx-CT, applied doses were tracked and analyzed against the planned doses. The dosimetric effects were quantitatively compared. Results The use of a plan database did not result in significant benefits for all cases. In 9 of 10 cases, a better PTV coverage was reached. In 6 of these 9 cases, better sparing of the bladder and in 5 of these 6, better sparing of the rectum was achieved. A plan database was useful for 9, 6 or 5 cases of the cohort, depending on the maintained criteria. A plan database seemed particularly useful for patients with large or medium bladder volume (compared