Supplemental IMRT May Increase the Risk of Rectal Bleeding in Prostate Cancer Patients Treated with Stereotactic Body Radiation Therapy (SBRT)

Abstract

Late rectal bleeding remains an ongoing clinical challenge for patients treated with external beam radiotherapy for prostate cancer. The goal of this study was to identify patient- or treatment-related characteristics associated with an increased risk of late rectal bleeding following SBRT. From May 2007 to August 2014, patients with clinically localized prostate adenocarcinoma were treated with robotic SBRT in 5 fractions (36.25 Gy) or a boost of 3 fractions (19.5 Gy) followed by fiducial-guided Intensity Modulated Radiotherapy (IMRT) to 45-50.4 Gy on an institutional protocol. Patient and treatment characteristics were recorded for patients with at least 2 years of follow-up. Toxicity was prospectively graded using the Common Terminology Criteria for Adverse Events version 4.0 (CTCAEv.4) at the start of and at 1-6 month intervals after therapy. Grade 2 rectal bleeding was defined as bleeding which required a new medication (i.e. steroid suppository) or up to 2 argon plasma coagulations (APCs). Grade 3 bleeding required > 2 APC procedures, a blood transfusion, or use of hyperbaric oxygen. Supplemental IMRT was selected a priori as a classifier for the toxicity. Chi-squared test was conducted to evaluate categorical data including occurrence of late ≥ grade 2 rectal bleeding in SBRT monotherapy and SBRT boost cohorts. Mann-Whitney U test was used to compare nonparametric continuous variables. We devised a standard binary logistic regression model to predict late ≥ grade 2 rectal bleeding controlling for age, Charlson Comorbidity Index (CCI), anticoagulant and/or antiplatelet use (AC), and supplemental IMRT. An ROC was constructed. 620 patients with a median age of 69 (range 42-94) were treated with SBRT. 459 (74%) received SBRT alone, and 161 (26%) received SBRT plus IMRT. At the start of therapy, 299 patients (48.2%) were on AC (4.8%, 42%, and 1.5% on anticoagulants, antiplatelets, or both, respectively), and 191 patients (30.8%) received Androgen Deprivation Therapy. The cumulative incidence of late ≥ grade 2 rectal bleeding was 2.6%, 2.0% for patients treated with SBRT alone and 4.3% for patients treated with SBRT plus IMRT (P<0.05). Controlling for age and AC use, receiving supplemental IMRT after SBRT predicted for more bleeding events (P=0.015; OR=5.94, 95% CI 1.41-25.01). CCI was itself a significant predictor (P=0.045; OR=1.40, 95% CI 1.01-1.94). The AUC of the ROC constructed from the model’s predicted values was 0.785. Utilization of supplemental IMRT may be associated with an increased risk of late ≥ grade 2 rectal bleeding following SBRT. Further studies should examine whether supplemental IMRT benefits select patient groups and whether interventions such as the use of rectal spacers may reduce this bleeding risk.