The Gleason grading system of prostatic adenocarcinoma does not account for the existence of a tertiary (third most prevalent) pattern, and there are no studies concerning the latter's prognostic influence. The authors analyzed 114 radical prostatectomies with small tertiary components, which mostly occupied less than 5% of whole tumors. These specimens were compared with a prostatectomy database comprised of 2,276 cases without a tertiary component. The pathologic stages of "typical" Gleason score 5 to 6 tumors (Gleason scores 2 + 3 = 5, 3 + 2 = 5, 3 + 3 = 6), which contained tertiary patterns 4 or 5, were significantly higher than those of "typical" Gleason score 5 to 6 tumors without pattern 4 (p = 0.018) but lower than those of "typical" Gleason score 7 tumors (p = 0.021; Gleason scores, 3 + 4 = 7, 4 + 3 = 7). Typical Gleason score 7 tumors with a tertiary pattern 5 showed significantly worse pathologic stages than typical Gleason score 7 tumors (p = 0.008) without pattern 5 and were not different statistically from typical Gleason score 8 (Gleason score, 4 + 4 = 8) tumors. Both typical Gleason score 5 to 6 and 7 tumors with tertiary components revealed significantly higher progression rates than typical Gleason score 5 to 6 tumors (p <0.0001) and Gleason score 7 tumors (p = 0.003) without tertiary components, and progressed like typical Gleason score 7 and 8 tumors respectively. Tertiary high-grade components have an adverse impact on biologic behavior. The authors propose that the Gleason system for radical prostatectomy specimens be modified to take into account small volumes of patterns 4 and 5, which are important prognostically.