8/97.8% vs. 93.1/93.1%, p = 0.006) ( Fig. 1). The Gleason pattern 3 patients also trended toward a higher 10- and 14-year CSS (99.3/99.3% vs. 96.9/96.9%, p = 0.058) ( Fig. 2). OS was not statistically different between the two Gleason 7 cohorts (78.2/70.7% vs. 76.0/56.9%, p = 0.198) ( Fig. 3). Subset analyses were performed to control for imbalances in PSA and PPC between the two study groups. In the subset of patients with PSA ≤10, primary Gleason pattern 3 patients maintained a significantly higher 10-

and 14-year bPFS (98.7/98.7% vs. 94.8/94.8%, p = 0.009) and CSS (100/100% vs. 97.0/97.0%, p = 0.013). ZD1839 nmr In those patients with PSA >10, the bPFS (93.0/93.0% vs. 90.0/90.0%, p = 0.52) and CSS (96.2/96.2% vs. 96.2/96.2%, p = 0.95) did not differ according to primary Gleason pattern. In the subset of patients with PPC ≤50%, there was a trend toward improved bPFS (97.5/97.5% vs. 94.3/94.3%, p = 0.14) and CSS (99.8/99.8% vs. 97.5/97.5%, p = 0.066) for Gleason pattern 3, but this did not reach statistical significance. In those patients with PPC >50%, there was a superior bPFS among

primary Gleason pattern 3 patients (97.7/97.7% vs. 90.5/90.5%, p = 0.018), but this did not translate into an improved CSS (97.9/97.9% vs. 96.4/96.4%, p = 0.69). Univariate and multivariate analyses were performed to identify the strongest predictors of bPFS, CSS, and OS (Table 2). Primary Gleason pattern was predictive of bPFS on both univariate (relative risk, 2.73; p = 0.005) and multivariate (relative risk, 2.265; p = 0.024) analyses. Primary Gleason pattern also trended toward predicting CSS (p = 0.081) on univariate analysis although STAT inhibitor this did not reach statistical significance. Gleason score is an important prognostic factor having been shown to predict for bPFS and CSS after definitive treatment of prostate cancer [1], [2], [3], [4] and [5]. Gleason 7 prostate cancer represents one of the most common histologic patterns. Some studies indicate that within the Gleason 7 stratum, a primary pattern 4 carries a less

favorable prognosis than a primary pattern 3, although conflicting results have been reported [5], [6], [7], [8], [14], [15], [16] and [17]. In a prior publication, we reported our outcome data for Gleason Thalidomide 7 patients treated with LDR interstitial brachytherapy. At that time, there were no statistically significant differences observed between primary Gleason pattern 3 and 4 (8). In this updated analysis, which includes a larger study population and longer median followup, we are now seeing a trend in outcome that favors primary Gleason pattern 3. The primary Gleason 3 cohort exhibited a superior bPFS and a nonsignificant trend toward improved CSS. One notable limitation of the present study is an imbalance in prognostic factors between the two study arms. The primary Gleason 4 population had a statistically higher PSA and PPC, which in itself would portend a less favorable outcome.