PP11 Presentation Time: 11:10 AM: Assessment of Brachytherapy-Induced Prostate Edema on Post-Implant Dosimetric Analysis in MRI-Assisted Radiosurgery

Abstract

<h3>Purpose</h3> Prior studies have shown that brachytherapy-induced prostate edema (PE) results in altered target dose coverage on post-implant CT dosimetry in prostate cancer patients treated with low dose rate brachytherapy (LDR). The pre-plan target volume (PTV) accounts for PE by adding 1-5mm margins to the prostate for calculation of the goal dose coverage on the day of the implant (day zero, D0). However, visualization of PE on CT is limited, diminishing PE measurement accuracy. MRI-based LDR (MRI-assisted radiosurgery, MARS) improves accuracy through better delineation of target, normal structures, and seeds compared to CT. Thus, MARS may improve the accuracy of PE measurements. We investigated PE using MARS to analyze its effect on D0 dosimetry and determine if current pre-plan PTV margins are adequate in all isotopes. <h3>Materials and Methods</h3> We identified 346 patients with low to intermediate-risk prostate cancer treated with Cesium-131 (Cs-131, 15 [4.3%] patients, prescription dose (PD)=115 Gy), Iodine-125 (I-125, 83 [23.9%] patients, PD=144 Gy), or Palladium-103 (Pd-103, 248 [71.7%] patients, PD=125 Gy) MARS as definitive LDR from 2016-2021. Patients treated with hormone therapy, as a boost with external-beam RT, or for salvage were excluded. Implants were pre-planned using MRI (MIM Symphony treatment planning system) and our published nomograms verified the activity per volume within 5%. PE was measured as the change in prostate volume from the pre-plan to D0 based on physician contours. Post-implant dosimetry was performed using MRI on D0. Treatment characteristics, including pre-plan and D0 prostate volumes, prostate volumes receiving 100%, 150%, and 200% of the PD (V₁₀₀, V_150, V₂₀₀), percent of the PD going to 90% of the prostate volume (D₉₀), rectum volume receiving 100% of the PD (V₁₀₀), and external urethral sphincter (EUS) volume receiving 200% of the PD (V₂₀₀), were extracted from MIM. Simple linear regression with Pearson's correlation analysis was used to assess the effect of PE on post-implant D0 prostate V₁₀₀, V₁₅₀, V₂₀₀, and D₉₀, rectum V₁₀₀, and EUS V₂₀₀. Ordinary one-way ANOVA analysis was used to compare PE between categorical variables. <h3>Results</h3> Median age at diagnosis was 63 years. Median PSA at diagnosis was 6. 24 (6.9%) patients had GS 6 and 322 (93.1%) had GS 7. On the pre-plan, 55 (15.9%) patients had a prostate size of 0-20 ccs, 222 (64.2%) had 20-40 ccs, 63 (18.2%) had 40-60 ccs, 6 (1.7%) had 60+ ccs. On D0 analysis, the median [IQR] change in prostate volume was +28% [15-42%] as compared to the pre-plan. There was no significant difference in magnitude of PE by isotope (median change in prostate volume: +22.6% for Cs-131, +27.3% for I-125, +29.5% for Pd-103, p=0.33). There was no difference in PE by increasing number of needles (R²=0, p=0.70). PE decreased with increasing pre-plan prostate size (R²=0.05, p<0.001). Greater PE was associated with decreased post-implant D0 V₁₀₀ (<b>Graphic</b>, R²=0.08, p<0.001), V₁₅₀ (R²=0.03, p<0.001), and D₉₀(R²=0.12, p<0.001); there was no association with D0 V_{200 (}R²=0.01, p=0.06). Greater PE was associated with decreased D0 rectum V₁₀₀ (R²=0.02, p=0.02) and trended toward decreased EUS V₂₀₀ (R²=0.01, p=0.05). <h3>Conclusions</h3> MARS confirmed that patients undergoing LDR experienced PE. PE was uniform between isotopes, supporting the use of the same pre-plan PTV margins among isotopes. There was decreased D0 prostate V₁₀₀, V₁₅₀, and D₉₀ in patients with greater PE. We suggest pre-plan PTV margins as follows: 3mm laterally, 2mm anteriorly/cranially/caudally, and 0mm posteriorly. These PTV margins can help ensure adequate D0 dosimetry. <b>Graphic:</b> Correlation of post-implant D0 prostate V₁₀₀ with the magnitude of prostate edema (PE). PE was calculated as the percent change in prostate volume from the pre-plan to post-implant D0.