PD72-09 EARLY OUTCOMES OF FOCAL BRACHYTHERAPY FOR LOCALIZED PROSTATE CANCER: COMPARISON WITH WHOLE GLAND BRACHYTHERAPY.

radical prostatectomy (RP) but does not include multiparametric pelvic MRI (mpMRI).Here, we evaluate the utility of mpMRI in predicting sXRT failure after RP.METHODS: Men undergoing RP at Mayo Clinic from 2003-2015 who had biochemical recurrence and received sXRT were included in a retrospective chart review.Men who underwent prostate cancer treatment prior to RP, received adjuvant XRT, or who were hormone refractory at sXRT were excluded.Patients with mpMRI within 12 months of sXRT were retained.mpMRI lesions were grouped according to location: vesicourethral, seminal vesical (SV) bed / prostate fossa, pelvic nodes, pelvic bones.If no lesion was present, the mpMRI was categorized as negative.Standard descriptive statistics and multivariable cox regression analyses were performed to assess the impact of mpMRI on PSA recurrence after sXRT.Models were adjusted for the variables in the Stephenson Nomogram: PSA at RP, PSA prior to sXRT, PSA doubling time, hormone therapy with sXRT, sXRT dosage, extracapsular extension, SV invasion, pathologic Gleason score, margin status, and pN stage.RESULTS: Overall, 473 men had mpMRI prior to sXRT (median PSA at sXRT 0.45ng/ml).Of these, 56.9% (204) had an indeterminate or suspicious lesion on MRI: 25.6% (124) vesicourethral, 27.8% (135) SV bed / prostatic fossa, 7.0% (34) pelvic node, and 0.6% (3) pelvic bone.Median PSA with a visible lesion on mpMRI was 0.45ng/ml.At a median follow up of 42 months after sXRT, 29.3% (142) had PSA recurrence and 14.0% (68) had distant metastasis.Patients without a mpMRI lesion or with a suspicious nodal/bone lesion had higher rates of PSA recurrence at 4 years compared to those with vesicourethral/SV/prostate fossa lesions alone: 39.5% vs 21.9% (p<0.001), Figure 1.On multivariable analysis, patients without a mpMRI lesion or a suspicious nodal/bone lesion were significantly more likely to have PSA recurrence (HR 2.41, 95% CI 1.52-3.83,p<0.001) after adjusting for variables in the Stephenson Nomogram.CONCLUSIONS: Pre-sXRT mpMRI is a valuable tool in risk stratifying men undergoing sXRT.The median PSA with a visible lesion on mpMRI was 0.45ng/ml, supporting the use of mpMRI in the early sXRT setting.