neoadjuvant hormone therapy either for volume reduction or prevention of disease progression during waiting time. The entire 10-year and 14-year PSA relapse free survival rates were 97.6% and 97.2%, respectively. Of those with the PPC ! 17%, the 10-year and 14-year PSA relapse free survival rates were both 99.0%, while of those with the PPC $ 17% the rates were 96.2% and 95.6%, respectively (p ! 0.01). On the multivariate Cox regression analysis, the PPC $ 17% was the only independent predictor of PSA relapse (p 5 0.01, hazard ratio; 3.53, 95% CI; 1.33-9.39), while the other factors listed above did not show statistical significance. Conclusion: These results indicate that PPC $ 17% is an independent risk factor for the PSA relapse in patients with low risk prostate cancer who underwent PI as monotherapy. It was suggested that more caution should be made for application of PI to low risk prostate cancer patients with the higher PPC value, and other additional therapeutic modalities may also be taken into account. PO13 The Effect of Bilateral Treatment Plan Symmetry on Postoperative Dosimetric Outcomes in Prostate Low-Dose-Rate Brachytherapy: A Single-Institution Study Alexandru M. Nicolae, MSc 1 , Ahamed Badusha Mohamed Yoosuf, MSc 2 , Sergio Esteve, MSc 2 , Geraldine Workman, MSc 2 , Darren M. Mitchell, FRCR 3 , Suneil Jain, MB BCh, MRCP, FRCR, PhD 4 , Ananth Ravi, PhD 1 . 1 Medical Physics, Sunnybrook Health Sciences Centre, Toronto, ON, Canada; 2 Radiotherapy Medical Physics, Northern Ireland Cancer Centre, Belfast, United Kingdom; 3 Clinical Oncology, Northern Ireland Cancer Centre, Belfast, United Kingdom; 4 Centre for Cancer Research and Cell Biology, Queen’s University Belfast, Belfast, United Kingdom. Purpose: Favorable postoperative dosimetric outcomes in prostate low- dose-rate (LDR) brachytherapy are impacted by a multitude of preoperative treatment factors. The aim of this study was to evaluate the long-standing hypothesis that preoperative bilateral plan symmetry has a profound impact on postoperative dosimetric quality. Materials and Methods: Retrospective analysis of 247 patients with localized prostate cancer treated consecutively with either 145 Gy or 110 Gy preoperatively planned LDR brachytherapy were obtained from the Northern Ireland Cancer Centre (Belfast, Northern Ireland, UK). Unique to the data set was the introduction of bilateral treatment plan symmetry as a critical criterion during planning. Preoperative and 1-month postoperative treatment plans were available for all patients. Bilateral plan symmetry (SYMM v150 ) was measured by creating a contour from the preoperative prostate V 150 isodose, reflecting the volumes about the prostate centroid, and computing a Dice coefficient between the left and right isodose volumes. Principle Component Analysis (PCA) feature elimination was used to identify additional preoperative variables with the largest contribution to postoperative dosimetric variance. A normalized multivariate logistic regression was performed to compare preoperative plan variables impacting postoperative dosimetric outcomes. P- values # 0.05 were considered significant. Results: PCA feature elimination showed that SYMM v150 , case number (the center-specific learning curve), and preoperative D 90 and V 100 were the most significant variables contributing to postoperative dosimetry. In terms of target coverage postoperative prostate V 100 had a significant positive correlation with increasing case number (OR 5 5.20, p 5 0.006), and increased SYMM v150 (OR 5 2.47, p 5 0.05). Furthermore, postoperative prostate D 90 had a positive correlation with increased SYMM v150 (OR 5 4.12, p 5 0.005), but not case number. Increased postoperative rectum D 1cc only had a positive correlation with increased preoperative prostate V 100 (OR 5 1.84, p 5 0.033). Conclusions: Incorporating preoperative bilateral treatment plan symmetry is likely to improve postoperative target coverage - with no significant effect on rectal tissue sparing - independent of the center-specific learning curve. This may be more readily achievable if the geometry of the prostate gland on preoperative imaging is also bilaterally symmetric. PO14 MRI-Based Treatment Planning for Prostate High Dose Rate Brachytherapy Leads to Decreased Target Size and Rectal Dose Gregory Arthur Jordan, BS 1 , Kyle Stang, MD 1 , Alexander Harris, MD 1 , Courtney Hentz, MD 1 , Brendan Martin, PHD 1 , Rakesh Patel, PHD 1 , Michael Mysz, MS 1 , Hyejoo Kang, PHD 1 , Ari Goldberg, MD 2 , Joseph Yacoub, MD 2 , Steven Shea, PHD 2 , Matthew Harkenrider, MD 1 , Abhishek A. Solanki, MD 1 . 1 Radiation Oncology, Loyola University Stritch School of Medicine, Maywood, IL, USA; 2 Radiology, Loyola University Stritch School of Medicine, Maywood, IL, USA. Purpose: We previously performed completely CT-based treatment planning in our High Dose Rate (HDR) prostate brachytherapy program. However, given the difficulty in defining the prostate on CT, we began performing MRI imaging after catheter placement for treatment planning. Our objective was to compare the planning target volume (PTV) and critical structure doses between our CT-only and MRI-based plans. Materials and Methods: We performed a retrospective cohort study of patients treated on a prospectively maintained database of prostate HDR patients treated from 2015-2017 by the same two brachytherapists. Monotherapy patients typically received 13.5 Gy x 2 fractions in 2 implants, and boost patients received 15 Gy in one fraction. Treatment planning was performed with Varian BrachyVision (Palo Alto, Ca). The PTV typically consisted of the prostate with 3-5 mm margin. For patients receiving MR imaging, 2D T2-weighted images were acquired after placement of the catheters and CT simulation. The MR images were registered to the CT images. Then, the PTV and normal tissues were contoured based on the MRI. The same planning goals were used for CT- only and MRI-based planning. Pearson chi-square and independent samples t-tests were used to compare baseline demographic and clinical characteristics between patients treated with CT only and MRI plans. The PTV volume and values for each of the planning goals were compared between both groups using a linear mixed effects model with random intercepts for each patient due to multiple observations for monotherapy patients. Analyses were repeated adjusting for baseline prostate size. A p- value of !0.05 was considered statistically significant. Results: 121 patients (208 treatment plans) were included in the study. 85 patients had CT-only and 36 patients had MRI-based planning. 89 patients (74%) received HDR monotherapy, 32 (26%) received it as a boost, with no difference in the proportion between the CT and MRI groups (p5.39). There was also no difference in T-stage (p5.87), Gleason sum (p5.80), median PSA (p5.28), or risk group (p5.86). There was no difference in the mean baseline prostate volume between groups (39.6 cc [SD 14.3], p5.70). The mean PTV was smaller in the MRI-based planning group compared to the CT group when adjusting for baseline prostate volume (76.53 cc [SD 3.19] vs. 87.98 cc [SD 2.09], p5.003). MRI patients had higher mean PTV D90 (111% vs. 108%, p5.02), V200 (9% vs. 8%, p !.001), V150 (32% vs. 30%, p5.002), and V100 (97.2% vs. 96.5%, p5.03). Comparing the doses to critical structures, MRI patients had similar mean bladder neck D0.1 cc (66% vs. 68%, p5.15), bladder V75% (0.78 cc vs. 0.93 cc, p5.15) and bladder D1 cc (72% vs. 72%, p5.73), but lower bladder D0.1 cc (81% vs. 85%, p5.02). Mean urethral max dose (119% vs. 117%, p !.001) and D0.1 cc (115% vs. 114%, p5.03) were higher in the MRI group, but Urethra V100 was lower (54% vs. 64%, p ! .001. Mean rectum V75% (0.53 cc vs. 0.75 cc, p5.01), D5 cc (53% vs. 56%, p5.01), D2 cc (63% vs. 66%, p5.003), and D1 cc (69% vs. 72%, p5.01) were lower in the MRI group, while D0.1 cc was not (81% vs. 81%, p5.49). These differences in dosimetry were maintained when adjusting for the patient’s prostate size. Conclusions: Integrating MRI at the time of treatment planning for prostate HDR brachytherapy decreases the size of the treatment target compared to CT-only planning, and may reduce the dose to the rectum. The impact on bladder and urethral dosimetry is less consistent. Further investigation into the clinical impact of differences in dosimetry on toxicity outcomes is needed and underway at our institution. S83 Abstracts / Brachytherapy 17 (2018) S15eS142