3283 - Intravesical Prostatic Protrusion: Exploring a Novel Imaging Biomarker for Late Genitourinary Outcomes following Prostate SBRT
Presenter(s)
M. Ito1, H. Takahashi2, J. Suzuki3, T. Mukohara2, S. Abe1, Y. Kubota4, Y. Yoshioka2,5, T. Okuda2, and K. Suzuki1; 1Department of Radiology, Aichi Medical University Hospital, Nagakute, Japan, 2Department of Radiation Oncology, Toyota Memorial Hospital, Toyota, Japan, 3Department of Radiation Oncology Quality Control Group, Toyota Memorial Hospital, Toyota, Japan, 4Department of Urology, Toyota Memorial Hospital, Toyota, Japan, 5Radiation Oncology Department, Cancer Institute Hospital of Japanese Foundation for Cancer Research, Tokyo, Japan
Purpose/Objective(s): Intravesical prostatic protrusion (IPP) is a simple, non-invasive quantitative metric for median lobe hypertrophy; however, its significance in predicting toxicity after prostate SBRT remains elusive. We hypothesized that IPP, measured on planning CT, serves as a robust predictor for both physician-assessed late genitourinary (GU) toxicity and patient-reported outcome (PRO) deterioration.
Materials/Methods: This single-center retrospective study evaluated 151 patients treated with definitive SBRT from 2017 to 2022. Patients received 36.25 Gy in 5 fractions via CyberKnife (without focal boost), and androgen deprivation therapy was administered for intermediate-risk or higher disease. The primary endpoint was the cumulative incidence of late CTCAE v5.0 Grade =2 GU toxicity. Secondary endpoints included gastrointestinal (GI) toxicity, longitudinal PROs (IPSS), and survival. IPP was determined using the maximum length on sagittal or coronal planning CT. To rigorously adjust for confounding biases, time-dependent ROC analysis was used to determine the optimal IPP cut-off, and a multivariate Cox proportional hazards model was applied for late toxicity and survival. Furthermore, a linear mixed-effects model (LMM) was utilized for longitudinal IPSS to appropriately handle missing values and baseline symptom heterogeneity.
Results: Median follow-up was 5.0 years. The incidence of acute Grade =2 GU/GI toxicity was 31.8%/6.0%, and late toxicity was 27.8%/6.0%. The 5-year cumulative incidence of late Grade =2 GU toxicity was 29.8%. Multivariate Cox regression identified IPP (HR=2.13; 95% CI, 1.21–3.77; p=0.009), age (HR=1.07; 95% CI, 1.01–1.13; p=0.02), and baseline IPSS (HR=1.49; 95% CI, 1.21–1.84; p<0.001) as independent predictors. Time-dependent ROC analysis established an optimal IPP cut-off of 0.8 cm (AUC=0.70). LMM analysis revealed no significant baseline IPSS difference between the high (=0.8 cm) and low IPP cohorts (11.7 vs. 10.2, p=0.155); however, the high IPP group experienced significant IPSS deterioration 1-2 years post-treatment (+1.96 points; 95% CI, 0.23–3.70; p=0.027). Five-year overall and biochemical relapse-free survival rates were 93.9% and 88.5%, respectively, with 8 (5.3%) biochemical recurrences and 7 (4.6%) intercurrent deaths. All toxicities were Grade 2 or lower, except for one case of Grade 3 late hematuria.
Conclusion: IPP is a simple and robust imaging biomarker that consistently predicts both objective late GU toxicity and subjective mid-term PRO deterioration following SBRT. Risk stratification utilizing the 0.8 cm cut-off holds promise for personalized treatment strategies, although prospective external validation is warranted.