2% and 14% and at 24 and 48 months, respectively (Fig. 1). Table 2 describes the number of procedures and site of stricture. Ten patients required more than one intervention, 7 had two procedures, 2 had three procedures, and 1 patient required five urethrotomies. Strictures were generally extraprostatic: 33.3% (15/45) had an apex/external sphincter stricture, 35.6% (16) had a bulbar urethral stricture, and 13.3% (6) had a membranous stricture. Only 1 patient had a prostatic urethral stricture and 1 patient had a late meatal stricture. The

risk of stricture development was strikingly different between the dose groups (Fig. 2). The estimated cumulative risk of stricture at 2 years was 0%, 2.3%, 3.4%, and 31.6% for 16 Gy/2 (n = 2), 20 Gy/4, 18 Gy/3, and 19 Gy/2 patients, respectively (p < 0.00001, log rank). In a Gefitinib cell line univariate analysis, the 19 Gy/2 group, urologist, radiation oncologist, failed trial of void, implant year, and biologic equivalent dose (BED) all predicted for increased risk of stricture (Table 3). No significant association was seen for IPSS, order of treatment, acute urinary retention,

or previous TURP. In a multivariable analysis, including all factors, the 19 Gy/2 group and implant year were two factors that remained predictive of an increased risk of stricture formation (Table 4). The D10 (defined as the minimum dose received by the “hottest” 10% of the urethral volume) was calculated as an estimated BED for 2 Gy fractions Smoothened antagonist (BED2Gy). This was done with an assumed α/β DOK2 of 3 Gy for prostate cancer and late effects. This dose included the external beam prescribed dose (It was assumed that the urethra received the total prescribed EBRT dose). The mean urethral D10 (BED2Gy) was 91.4 Gy in patients with a stricture compared with 87.0 Gy in those with no stricture (p < 0.0017,

t test). However, the D10 (BED2Gy) was significantly higher in the 19 Gy/2 dose group compared with all others ( Table 5). No correlation was seen within dose groups between D10 and stricture risk. A urethral stricture is a recognized late effect of any prostate cancer therapy (10). It appears that stricture rates are higher in HDRB compared with low-dose-rate brachytherapy (LDRB) and EBRT (11), and this may imply a BED response. For example, Mohammed et al. (11) analyzed 1903 patients who received EBRT, LDRB, or HDRB. The stricture risk was significantly higher in HDRB patients compared with EBRT and LDRB, 11%, 2%, and 4% respectively. We have reported a large patient database, with prospective gathering of stricture occurrence as well as other toxicity in the followup for HDRB used as a boost to EBRT. In our patients, the overall crude stricture incidence was 12.7% and is comparable with other series [12] and [13]. A concerning predictive factor seen in this study was the fractionation schedule and the BED delivered to the urethra, measured by the D10.