Patient and Dosimetric Predictors of Genitourinary and Gastrointestinal Toxicity After Prostate SBRT: A Secondary Analysis of a Prospective, Multi-Institutional Phase II Trial

Abstract

Stereotactic body radiation therapy (SBRT) is an emerging treatment option for localized prostate cancer. However, optimal patient selection & normal tissue tolerances for prostate SBRT are not well defined. We therefore analyzed patient factors & dosimetric correlates for genitourinary (GU) and gastrointestinal (GI) toxicity on a multi-institutional prospective trial.Patients (pts) with low- and intermediate-risk prostate cancer were enrolled on a phase II trial from 2011-2020. A dose of 36.25 Gy in 5 fractions to the PTV and 40 Gy to the prostate gland was administered on non-consecutive days. No androgen deprivation therapy or rectal spacer were used. Pts received a CT simulation with Foley catheter for urethral delineation. The clinical target volume was prostate (low-risk) or prostate plus 2cm of proximal seminal vesicles (intermediate-risk). Multiple dosimetric measures for bladder, urethra, & rectum were prospectively recorded. The maximum acute & late GU and GI toxicities using CTCAE v.3.0 were assessed at protocol-specific time points. Linear logistic regression was used to assess factors associated with Grade 1+ GI & GU toxicities. A multivariate (MV) model was constructed using all covariates with P-values < 0.2 on univariate (UV) analysis.A total of 160 evaluable men were enrolled. Median age was 67.9 years (IQR 62.6-73.2), 60 men (37.5%) were low risk, and 100 (62.5%) were intermediate risk. Median follow-up was 2.2 years. No patient factor or dosimetric correlate predicted acute Grade 1+ GU toxicity (all tested covariates P > 0.2). Treatment isodose line (P = 0.065) & age (P = 0.098) reached borderline significance for acute Grade 1+ GI toxicity on UV analysis, though neither were significant on MV analysis (P = 0.057 and P = 0.103, respectively). For late Grade 1+ GU toxicity, prior TURP, bladder V37(cc), bladder D10%, baseline AUA, & treatment isodose line reached borderline significance on UV analysis. On MV analysis, treatment isodose line trended towards significance at P = 0.055. For late Grade 1+ GI toxicity, rectum D20%, rectum D10%, rectum D5%, & history of coronary artery disease reached borderline significance on UV analysis, & a history of coronary artery disease predicted Grade 1+ GI toxicity on MV analysis (odds ratio 4.7, 95% CI 1.37-15.9, P = 0.009).No patient factor or dosimetric correlate predicted acute GI or GU toxicity after prostate SBRT. Late GU toxicity is possibly attributable to dose inhomogeneity. A history of coronary artery disease predicts for late GI toxicity, possibly due to anticoagulation.