ESTRO 2022 - Abstract Book

S355

Abstract book

ESTRO 2022

PD-0412 Impact of adjuvant radiotherapy on biochemical recurrence rates for pn1 prostate cancer patients

G. Corrao 1 , G. Marvaso 1 , F.A. Mistretta 2 , S. Luzzago 2 , I. Sabatini 2 , E. Di Trapani 3 , G. Cozzi 3 , R. Bianchi 2 , M. Ferro 3 , D.V. Matei 3 , G. Musi 2 , M. Pepa 4 , M. Zaffaroni 4 , B.A. Jereczek-Fossa 1 , O. De Cobelli 5 1 IEO, European Institute of Oncology IRCCS; University of Milan, Division of Radiation Oncology; Department of Oncology and Hematoncology, Milan, Italy; 2 IEO, European Institute of Oncology IRCCS, Department of Urology, Milan, Italy; 3 IEO, European Institute of Oncology IRCCS, Department of Urology , Milan, Italy; 4 IEO, European Institute of Oncology IRCCS, Division of Radiation Oncology, Milan, Italy; 5 IEO, European Institute of Oncology IRCCS; University of Milan, Department of Urology; Department of Oncology and Hematoncology, Milan, Italy Purpose or Objective The optimal management strategy for pN1 prostate cancer (PCa) patients after primary surgery is still debated. To address these voids, we compared long term biochemical recurrence rates (BCR) in pN1 patients that underwent adjuvant radiotherapy (aRT) vs. observation +/- early salvage RT (esRT) after radical prostatectomy (RP). Materials and Methods Inclusion criteria were the following: patients treated between 2010 and 202, pN1 after RP, informed written Consent to research purpose. Exclusion criteria were: < 10 lymph nodes removed at surgery, > 10 positive lymph nodes, persistently detectable PSA after RP, distant metastases at diagnosis. First, Kaplan-Meier plots depicted BCR rates and univariable and multivariable Cox regression models focused on predictors of BCR. Second, univariable and multivariable Cox regression models were refitted after propensity score (PS) matching. Results Two hundred and twenty patients met inclusion criteria, 133 (60%) vs. 87 (40%) patients were treated with aRT vs. noRT/esRT respectively. Specifically, 26 (12%) patients initially managed with observation after RP developed BCR and were subsequently treated with esRT. Median time from RP to esRT was 40 months (IQR: 17-62). The aRT patients were older (67 vs. 63 yrs, p<0.001). Higher rates of postoperative pathological ISUP grade group 4-5 pCa were observed in aRT patients (51 vs. 25%; p<0.001). A statistically significant difference was recorded in aRT and noRT/esRT regarding pT stage (5 vs. 14 patients in stage pT2; 43 vs. 40 in stage pT3a and 85 vs. 33 in stage pT3b, p <0.001). Median time to BCR was 62 vs. 38 months in aRT vs. noRT/esRT patients (p=0.001) (Figure 1). In multivariable Cox regression models, noRT/esRT patients were associated with higher BCR rates (hazard ratio [HR]: 3.27, p<0.001), relative to aRT group. BCR were comparable independently of the number of positive lymph nodes (<1 vs ≥ 2). After PS matching (ratio 1:1; aRT = 57 vs. noRT/ esRT= 57) a 5-year BCR rate significant difference was observed (respectively, 40.4 (aRT) vs. 76.4% (noRT/sRT); p<0.01) (Figure 2).