ESTRO 36 Abstract Book

S726 ESTRO 36 _______________________________________________________________________________________________

Purpose or Objective Salvage radiotherapy (SRT) is the only potentially curative therapy available for patients experiencing biochemical recurrence after radical prostatectomy (RP), and likely more effective when offered early at low PSA levels. This work aims to retrospectively assess clinical outcome and toxicity of patients treated with delayed SRT to radiologically and/or histologically proven macroscopic local recurrence. Material and Methods We report on a cohort of 56 patients with radiologically detected isolated macroscopic local recurrence on MRI and/or CT scan and treated with SRT between 2001 and 2015. Histological confirmation was available in 35 (63%) patients. A dose of 64-66 Gy (2 Gy/fr) was delivered to the prostatic bed followed by a dose escalation to 72-74 Gy (2 Gy/fr) to the site of macroscopic disease using image- guided IMRT (IG-IMRT). Patients were treated with concomitant short-course (6 months) of androgen deprivation therapy. Biochemical relapse-free survival (PSA nadir + 0.2 ng/ml; bRFS) and clinical relapse-free- survival (cRFS) were calculated using Kaplan-Meier method. Baseline, acute and late urinary and gastrointestinal (GI) toxicity rates were reported using CTCAE v4. Results Median age at SRT was 71 years (57-81). Out of 56 patients, 9 (16%) had pT3b disease and 19 (34%) had positive surgical margins. Sixteen patients (29%) had Gleason score ≥ 8 at RP. The median time from RP to SRT was 58 months (5-172). Median pre-RT PSA was 2.8 ng/ml (0.2-29). At a median follow-up of 39 months (8-153) post- SRT, 20 patients (36%) had biochemical failure and 8 (14%) developed distant metastasis. Median time to BF after SRT was 30 months (13-116). The 3- and 5-year bRFS were 70.5% and 53,5%, respectively. The 3- and 5-year cRFS was 89.2% and 80%, respectively. High-risk patients (pT3b and/or Gleason score ≥ 8) had 3- and 5-year bRFS of 54% and 27%, while 3- and 5-year cRFS was 66.7% and 44.4%, respectively. Univariate and multivariate analyses showed that Gleason score ≥ 8 and perineural invasion were associated with lower bRFS (p=0.005 and 0.046, respectively). High-risk patients presented lower bRFS and cRFS when compared to lower risk patients (p= 0.03 and 0.001, respectively). At baseline, 4 patients (7%) had grade 3 urinary toxicity. Twelve patients (21%) developed grade ≥ 2 acute urinary toxicities. Three patients (5%) had grade 2 acute GI toxicities. No grade ≥ 3 acute GI toxicity occurred. Nine patients (16%) had grade ≥ 2 late urinary toxicities. No grade ≥ 2 late GI toxicity was reported. Conclusion Delayed SRT using dose escalated IG-IMRT to isolated macroscopic local recurrence is associated with poor oncological outcomes particularly in high risk patients (i.e. pT3b and/or Gleason score ≥ 8). Toxicity profile seems to be acceptable at a medium term follow up. Patients with high risk features should be strongly considered for earlier and intensified treatment approaches. EP-1355 Comparing toxicity in IMRT and particle therapy of prostate cancer in a ROCOCO in silico trial Y. Van Wijk 1 , E. Roelofs 2 , B. Vanneste 2 , S. Walsh 2 , P. Lambin 2 1 Maastricht university, School for Oncology and Developmental Biology, Maastricht, The Netherlands 2 MAASTRO clinic, Radio Oncology, Maastricht, The Netherlands Purpose or Objective Studies have shown that dose escalation has positive effect on tumour control probability when treating high

risk prostate cancer. However, due to the higher dose in the rectum, the normal tissue complication probability (NTCP) for gastral intestinal (GI) toxicity is increased. The goal of this study is to investigate whether radiotherapy of high-risk prostate cancer with light-ion particles results in reduced NTCP when compared to IMRT. A comparison between doses and NTCP was made for three modalities: IMRT, proton therapy (IMPT) and carbon-ion therapy (IMIT) (figure 1).

Material and Methods Twenty-five consecutive patients with T3-T4 prostatic cancer were included for high-dose radiotherapy. Proton and Carbon-ion treatment planning was performed by the ROCOCO trial partners following a strict clinical protocol, prescribing 78 Gy (biologically equivalent dose) to the target and compared pair-wise to generated IMRT treatment plans. All 75 plans were analysed centrally to compare the dose in the rectum (the main organ at risk) and the NTCP for late rectal bleeding. A validated multi-factorial NTCP model used the mean dose (Gy) in the rectum and the percentage of the rectum receiving more than 75 Gy (V75) as dosimetric predictors. It also included a set of clinical predictors such as haemorrhoids and hormonal therapy. Results An overview of the results is shown in table 1.

The mean dose in the rectum resulting from the IMRT, proton and carbon-ion plans was 42.9 Gy [range: 30.8-47.2 Gy], 30.8 Gy [range: 23-45.2 Gy] and 18.9 Gy [range: 11.8- 33.3 Gy] respectively. The V75 for the IMRT, proton and carbon-ion was 3.2% [range: 0.3-10.8 %], 2.1% [range: 0.5- 4.3 %] and 1.9% [range: 0.6-3.9 %] respectively. Both proton and carbon-ion plans showed improvement with respect to the IMRT plans in both dosimetric parameters, and for carbon-ions it showed an improvement when compared to protons.