ESTRO 2021 Abstract Book

S1111

ESTRO 2021

reported, the majority of which were already present at the baseline.

PO-1354 Time Dependence of PSADT prognostic threshold in men treated with post-prostatectomy salvage RT C. Cozzarini 1 , A. Magli 2 , D. Cante 3 , L. Andreoli 4 , A. Pastorino 5 , A. Faiella 6 , A. Fodor 1 , C. Deantoni 1 , F. Zerbetto 7 , F. Tonetto 2 , B. Noris Chiorda 4 , C. Piva 3 , E. Olivetta 5 , B. Avuzzi 4 , G. Gandaglia 8 , N. Fossati 8 , C. Fiorino 9 , G. Sanguineti 6 , R. Valdagni 10 , A. Briganti 11 , F. Montorsi 11 , N. Di Muzio 12 1 San Raffaele Scientific Institute, Radiotherapy, Milan, Italy; 2 Azienda Ospedaliero Universitaria S. Maria della Misericordia, Radiotherapy, Udine, Italy; 3 Ospedale di Ivrea, Radiotherapy, Ivrea, Italy; 4 Fondazione IRCCS Istituto Nazionale dei Tumori, Radiotherapy, Milan, Italy; 5 A.O. SS. Antonio e Biagio, Radiotherapy, Alessandria, Italy; 6 IRCCS Istituto Nazionale dei Tumori “Regina Elena", , Radiotherapy, Rome, Italy; 7 San Raffaele Scientific Institute, Radiotherapy, MIlan, Italy; 8 San Raffaele Scientific Institute, Urology, Milan, Italy; 9 San Raffaele Scientific Institute, Medical Physiscs, Milan, Italy; 10 Fondazione IRCCS Istituto Nazionale dei Tumori, Programma Prostata - Università degli Studi Milano, Radiotherapy, Milan, Italy; 11 San Raffaele Scientific Institute - Università Vita Salute San Raffaele, Urology, Milan, Italy; 12 San Raffaele Scientific Institute - Università Vita Salute San Raffaele, Radiotherapy, Milan, Italy Purpose or Objective To date, PSADT thresholds, in the range of 6-12 months, discriminating between local and systemic relapse in the case of biochemical recurrence (BCR) after radical prostatectomy (RP) have been assessed at different time intervals from the first post-RP detectable PSA. The aim of this retrospective investigation was to assess whether a time–dependence of the prognostic PSADT threshold may be postulated, and to investigate the predictive accuracy of PSADTs evaluated at the first evidence of detectable post-RP PSA values. Materials and Methods A retrospective series of 440 pts treated in six Institutes with early salvage RT (SRT) with at least two PSA ≥0.10 ng/mL after RP was analyzed. The median interval from RP to SRT was 35.8 months, median PSA@SRT 0.44 ng/mL (IQR 0.25-0.80), median follow-up 113 months (IQR 79-150), median 2-Gy equivalent (α/β 3 Gy) SRT dose to the prostatic bed 72 Gy. Adjuvant ADT was given to 37% of patients for a median of 14.2 months. A total of 1775 PSA values ≥0.10 ng/mL were available. For each patient, the PSADT at 3 (±1) months after the first post-RP value ≥0.10 ng/mL was calculated (PSADT3). The same procedure was repeated for the PSADT at 6 (±1), 9 (±1) and 12 (±1) months (PSADT6, PSADT9 and PSADT12, respectively). Finally, the PSADT on all the PSA values available (median 4, IQR 3-5) for each patient was calculated (PSADTALL). The predictive value of PSADT3-12 as compared to that of PSADTALL with respect to the risk of post-SRT biochemical and clinical relapse (bREL and cREL, respectively) was investigated by means of Receiver Operating Characteristics (ROC) curve comparisons. Results The median PSADTALL was 7.90 months; median PSADT3 (n=240), PSADT6 (n=226), PSADT9 (n=159) and PSADT 12 (n=175) were 5.18, 6.62, 8.96 and 10.97 ng/mL, respectively. In the Figure, the median values of PSADT3-12 for pts with no evidence of disease (NED), post-SRT BCR only or evident clinical relapse are plotted.

With the sole exception of lower accuracy of PSADT12 vs PSADTALL in predicting the risk of cREL (AUC 60% vs 66%, p=0.02), no significant differences in terms of predictive value between PSADT3-12 and PSADTALL emerged with respect to the risk of post-SRT bREL or cREL (p ≥0.13). Nevertheless, PSADT6-9 and PSADT3-6-9 performed better with respect to the prediction of post-SRT bREL and cREL, respectively, (AUC ≥60). Conclusion The PSADTs calculated at 6 or 9 months after the first post-surgical PSA ≥0.10 mg/mL showed a predictive