ESTRO 36 Abstract Book

S384 ESTRO 36 _______________________________________________________________________________________________

Conclusion Despite its retrospective nature, this study lends support to the use of both WPRT and SRT doses ≥75.6 Gy in the salvage setting in node-negative patients with ≥1 risk factor among PSA at SRT >0.50 ng/mL, Gleason score ≥7 and pathologic stage ≥T3a. PO-0731 Comparison of two fractionation schemes in prostate cancer patients treated with robotic SBRT F. Akyol 1 , P. Hurmuz 1 , G. Ozyigit 1 1 Hacettepe University- Faculty of Medicine, Department of Radiation Oncology, Ankara, Turkey Purpose or Objective The aim of this study is to evaluate the long term treatment results of two different fractionation schemes and PSA bounce phenomenon in our patients treated with robotic stereotactic body radiotherapy (rSBRT) for prostate cancer. Material and Methods We evaluated the medical records of 106 patients who were treated between June 2007 and June 2015 for prostate cancer in our department with CyberKnife™. D’Amico risk classification system (1998) was used to group patients. In the low risk (LR) group (n=54) 20 patients received androgen blockade (AB) (for volume reduction or else purposes). In the intermediate risk (IR) group (n=52) 42 patients received neoadjuvant/adjuvant AB. According to our institution’s treatment protocol rSBRT was delivered in 5 fractions to a total dose of 36.5 Gy either sequentially (n=58) or every other day (n=48). ‘Cavanagh definition’ (≥0.2 ng/mL) was used to define ‘PSA bounce phenomenon’. Results Median follow up time was 56 months. Only one patient died due to pulmonary thromboembolism. Two patients in the LR group and 3 patients in the IR group had PSA relapses. In the whole group 5 year biochemical relapse free survival (BRFS) rate was 93.4% (LR; 97.1% vs. IR; 89.2%; p=0.6). Five year BFRS rate in patients treated with sequential scheme was 92% compared to 100% for every other day scheme (p=0.3). PSA bounce phenomenon was observed in 17 patients (16%) and among them only one patient had PSA relapse. Five year BRFS rate was 89% in patients with PSA bounce phenomenon compared to 94.2% in patients without bounce (p=0.9). There was no difference in the incidence of PSA bounce for different treatment schemes. All of the patients received the planned rSBRT dose without any breaks in the treatment schedule due to toxicity. Late grade III gastrointestinal system toxicity was observed in 4 patients (3 sequential, 1 every other day scheme; p=0.6). No patients reported late grade III genitourinary toxicity. Conclusion Our rSBRT treatment protocol was very well tolerated with high rates of 5 year BRFS rates. PSA bounce phenomenon may also be observed after rSBRT for prostate cancer. We did not find a relation between PSA bounce and biochemical relapse. The fractionation scheme of the rSBRT in our protocol did not affect the treatment results and toxicity. PO-0732 Toxicity and outcome in moderately hypofractionated radiotherapy for 590 prostate cancer patients A. Maucieri 1 , B.A. Jereczek-Fossa 2 , D. Ciardo 1 , C. Fodor 1 , P. Maisonneuve 3 , A. Surgo 1 , S. Volpe 2 , G. Marvaso 1 , A. Vavassori 1 , A. Viola 2 , G. Musi 4 , O. De Cobelli 5 , R. Orecchia 6 1 European Institute of Oncology, Department of Radiation Oncology, Milan, Italy 2 European Institute of Oncology - University of Milan,

73.8 Gy (all with 3DCRT), while 86 received WPRT at a median dose of 50.4 Gy (45% by standard 3D box and 55% by static field IMRT), plus a consequential 3D-CRT boost to PB up to 75.6 Gy. The indication of both WPRT and of adjuvant androgen deprivation (AAD), prescribed to 158 pts for a median of 14 months, was at the discretion of each treating physician Results Median overall FU was 111 months. Owing to the gradual introduction of WPRT over time, the cohort receiving WPRT had a significantly shorter follow-up (90 vs 124 months, p<0.0001) and received significantly higher RT dose (median 75.6 vs 73.8 Gy). All the remaining prognostic variables were evenly distributed. At univariable analysis SRT dose ≥75.6 Gy was the only variable predictive of significantly increased 7-year bRFS in the overall population and in the two subsets of pts with PSA at SRT (PSA@SRT) ≤0.50 or >0.50 ng/mL. WPRT led to a significant improvement in 7-year bRFS in the overall population (84 vs 72%, p=0.02) and in the 144 pts with a PSA at SRT >0.50 ng/mL (83% vs 64%, p=0.01). At multivariable analysis (MVA) SRT dose ≤73.8 Gy, pathologic stage ≥T3a, GS ≥7 and higher PSA@SRT, but not WPRT, emerged as covariates independently predictive of reduced post-SRT bRFS in the overall population. The independent benefit deriving from higher SRT doses and WPRT was confirmed in the 2 subsets of pts with PSA@SRT ≤ and >0.50 ng/mL, respectively. No benefit whatsoever emerged from AAD (Fig 1). In order to identify the subset benefiting most from WPRT (Fig. 2) and dose-escalation, their roles were investigated in the subset of 184 pts with at least 1 risk factor among those individuated at MVA (GS ≥7, pT≥3a, and PSA at SRT >0.50 ng/mL). In this subset MVA confirmed the independent role of WPRT (HR 0.47, p=0.04) and SRT doses ≥75.6 Gy (HR 0.46, p=0.03). Neither WPRT nor higher SRT doses led to any significant increase of either acute or late toxicities.

Department of Radiation Oncology, Milan, Italy 3 European Institute of Oncology, Division of Epidemiology and Biostatistics, Milan, Italy