S1170
Clinical - Urology
ESTRO 2026
Chaudhuri, G. Pash, D.A. Hormuth II, G. Lorenzo, M. Kapteyn, et al. (2023). Predictive digital twin for optimizing patient-specific radiotherapy regimens under uncertainty in high-grade gliomas. Frontiers in Artificial Intelligence, 6, 1222612. Keywords: tumor forecasting, biomechanistic model, prostate
or the onset of secondary treatment due to PCa recurrence. Model parameters were investigated as potential biomarkers of biochemical relapse. Results: Our patient-specific model fits demonstrated excellent agreement with their corresponding longitudinal PSA data series (see Figure 1), achieving a median (interquartile range) of the root-mean-squared error (RMSE) of 0.24 (0.09, 0.65) ng/mL. The tumor cell proliferation rate and its ratio with respect to EBRT- induced death rate were strongly correlated with biochemical relapse (both p<0.001; Wilcoxon rank-sum test). The model-predicted PSA nadir and the time to reach it after EBRT were also significantly aligned with biochemical relapse (both p=0.0025; Wilcoxon rank- sum test).
Poster Discussion 1142
Safety of Elective Pelvic Nodal Irradiation in High- Risk Post-Prostatectomy Patients: A Subgroup Analysis from POPART (PELVART) Cohort Federica Ferrario 1 , Giulia Rossano 2,1 , Lorenzo De Sanctis 2,1 , Chiara Chissotti 2 , Riccardo Ray Colciago 2 , Valerio Pisoni 1 , Valeria Faccenda 3 , Denis Panizza 3 , Stefano Arcangeli 1,2 1 Radiation Oncology Department, Fondazione IRCCS San Gerardo dei Tintori, Monza, Italy. 2 School of Medicine and Surgey, Università Milano Bicocca, Milano, Italy. 3 Medical Physics Department, Fondazione IRCCS San Gerardo dei Tintori, Monza, Italy Purpose/Objective: Several prospective studies are currently exploring the role of SBRT for BCR post-prostatectomy, with promising preliminary safety and efficacy results. Following an amendment to the POPART protocol (NCT04831970), elective pelvic nodal irradiation (ENI, PELVARTcohort) was introduced for patients with predefined high-risk features. This subanalysis aimed to evaluate the safety of this extended-field SBRT compared with the overall POPART population1. Material/Methods: POPART is a multicenter, prospective study assessing postoperativeSBRT delivered as 32.5Gy in five alternate-day fractions. Patients presenting with at least one high-risk feature (ISUP ≥ 4, postoperativePSA ≥ 0.5 ng/mL, p ≥ T3b, PSA-doubling time <6months) could also receive ENI with 25Gy concurrently, at the treating physician’s discretion. Side effects (SE) were assessed using CTCAEv5.0, and quality-of-life (QoL) with ICIQ-SF, IIEF-5, and EPIC-CP questionnaires at baseline and during follow-up. Minimal clinically important difference (MCID) was defined as a fixed change score from baseline, based on previous POPART analyses (Table 1)1. The incidence and severity of GU and GI SE in the PELVART cohort, along with questionnaire scores, were compared with those of the overall POPART population. Results: In the PELVART cohort, 10 patients were treated and analyzed for acute SE and QoL. Four patients received ADT; the median interval from surgery to SBRT was 23months (range 6–34), and the median pre-RT PSA
Figure 1. Examples of model fits for four representative patients. Conclusion:
Our biomechanistic modeling framework reproduced patient-specific PSA trajectories from diagnosis to salvage therapy after EBRT with/withoutADT, and its parameters showed promising potential for early identification of relapsing patients. Future studies will investigate the model capability to predict response to EBRT with/without ADT as well as its implementation as a digital twin to aid in therapeutic and monitoring decision-making [3]. References: [1] G. Lorenzo, N. di Muzio, C.L. Deantoni, C. Cozzarini, A. Fodor, et al. (2022). Patient-specific forecasting of postradiotherapy prostate-specific antigen kinetics enables early prediction of biochemical relapse. iScience, 25(11), 105430.[2] T. Phan, S.M. Crook, A.H. Bryce, C.C. Maley, E.J. Kostelich, Y. Kuang. (2020). Mathematical modeling of prostate cancer and clinical application. Applied Sciences, 10(8), 2721.[3] A.
Made with FlippingBook - Share PDF online