ESTRO 2025 - Abstract Book

S2045

Clinical - Urology

ESTRO 2025

SS Antonio e Biagio e Cesare Arrigo, Alessandria, Italy. 16 Unit of Data Science, Fondazione IRCCS Istituto Nazionale dei Tumori, Milano, Italy. 17 Radiation Oncology, IRCCS San Raffaele Scientific Institute, Milano, Italy

Purpose/Objective: Exploring the association between lymphopenia and late (2-year) patient-reported intestinal toxicity after intensity modulated Whole-Pelvis Radiotherapy (WPRT) for prostate cancer. Intestinal toxicity was assessed using the Inflammatory-Bowel-Disease-Questionnaire (IBDQ). The current investigation focused on urgency/fecal incontinence, given their clinical relevance (urgency to defecate, IBDQ24 and accidental soiling, IBDQ26). Material/Methods: Between 2012 and 2021, 884 patients treated with conventional/moderate hypo-fractionation were enrolled in 14 Institutes and followed within a prospective study including the collection of patient-reported toxicities and blood samples data (baseline, mid-RT, end-RT, 6 months and every 6 months up to 5 years). Patients with complete data (both IBDQ and ALC counts) at baseline, mid or end-RT and 2- year and with baseline IBDQ≥5 (on a 1 -7 scale) were considered for the analysis (n=555). The median/interquartile dose to PTV-LN was 51.1Gy/49.9-52.0Gy/1.8-2.0Gy/fr). Patients with CTCAE-v4.03 acute- Grade≥3 (G3+,ALC<500/μL) and 2 -year- Grade≥2 (G2+,ALC<800/μL) were first identified. Logistic regression was used to assess the impact of acute/2- year lymphopenia on ≤ -2 2-year changes from baseline, considered to be clinically relevant. The impact of many available covariates was tested, including baseline IBDQ24/26 values, age, PTV-LN cranial extension, PTV-LN volume, daily dose to PTV-LN, dose to PTV-LN, BMI, diabetes, smoking, hypertension, prostatectomy and hormonal therapy. Given the number of events, a bootstrap-based selection of the best combination of the three major predictors was followed to assess the most robust model. Time patterns of ALC according to 2-year urgency/incontinence was also quantified and differences assessed by Mann-Whitney test. Results: 211/47 patients experienced acute-G3+/2-year-G2+ lymphopenia respectively. The patients with changes of IBDQ24/IBDQ26≤ -2 were 35/38. Both Acute/2-year Lymphopenia were significantly associated with IBDQ24/IBDQ26≤ -2, with OR between 2.2 and 3.5, p<0.007). The best multivariable models for IBDQ24 retained lymphopenia (Acute or 2-year, tested separately) as major predictor, as summarized in Table 1 (OR=2.4,p=0.017, and 3.9,p=0.004, for acute or 2-year lymphopenia respectively). Concerning IBDQ26, acute lymphopenia was the most predictive variable at univariate analysis, with OR=2.4 [p=0.021, 95%CI:1.1 – 4.9] while, in multivariable analysis, no additional covariates added predictive value. Looking at ALC values at baseline, acute (worst between mid and end-RT) and 2- year for patients with or without IBDQ24≤ -2 at 2-year, they were different at all three times, reaching statistical significance in the acute phase (p=0.003).