Secondary Logo

Journal Logo

Institutional members access full text with Ovid®

Histologic Grading of Prostatic Adenocarcinoma Can Be Further Optimized

Analysis of the Relative Prognostic Strength of Individual Architectural Patterns in 1275 Patients From the Canary Retrospective Cohort

McKenney, Jesse K. MD; Wei, Wei MS; Hawley, Sarah MS; Auman, Heidi PhD; Newcomb, Lisa F. PhD; Boyer, Hilary D. BSc; Fazli, Ladan MD; Simko, Jeff MD, PhD; Hurtado-Coll, Antonio MD; Troyer, Dean A. MD, PhD; Tretiakova, Maria S. MD, PhD; Vakar-Lopez, Funda MD; Carroll, Peter R. MD, MPH; Cooperberg, Matthew R. MD, MPH; Gleave, Martin E. MD; Lance, Raymond S. MD; Lin, Dan W. MD; Nelson, Peter S. MD; Thompson, Ian M. MD; True, Lawrence D. MD; Feng, Ziding PhD; Brooks, James D. MD

The American Journal of Surgical Pathology: November 2016 - Volume 40 - Issue 11 - p 1439–1456
doi: 10.1097/PAS.0000000000000736
Original Articles

Histologic grading remains the gold standard for prognosis in prostate cancer, and assessment of Gleason score plays a critical role in active surveillance management. We sought to optimize the prognostic stratification of grading and developed a method of recording and studying individual architectural patterns by light microscopic evaluation that is independent of standard Gleason grade. Some of the evaluated patterns are not assessed by current Gleason grading (eg, reactive stromal response). Individual histologic patterns were correlated with recurrence-free survival in a retrospective postradical prostatectomy cohort of 1275 patients represented by the highest-grade foci of carcinoma in tissue microarrays. In univariable analysis, fibromucinous rupture with varied epithelial complexity had a significantly lower relative risk of recurrence-free survival in cases graded as 3+4=7. Cases having focal “poorly formed glands,” which could be designated as pattern 3+4=7, had lower risk than cribriform patterns with either small cribriform glands or expansile cribriform growth. In separate multivariable Cox proportional hazard analyses of both Gleason score 3+3=6 and 3+4=7 carcinomas, reactive stromal patterns were associated with worse recurrence-free survival. Decision tree models demonstrate potential regrouping of architectural patterns into categories with similar risk. In summary, we argue that Gleason score assignment by current consensus guidelines are not entirely optimized for clinical use, including active surveillance. Our data suggest that focal poorly formed gland and cribriform patterns, currently classified as Gleason pattern 4, should be in separate prognostic groups, as the latter is associated with worse outcome. Patterns with extravasated mucin are likely overgraded in a subset of cases with more complex epithelial bridges, whereas stromogenic cancers have a worse outcome than conveyed by Gleason grade alone. These findings serve as a foundation to facilitate optimization of histologic grading and strongly support incorporating reactive stroma into routine assessment.

*Cleveland Clinic, Cleveland, OH

University of Texas MD Anderson Cancer Center, Houston

††University of Texas Health Sciences Center San Antonio, San Antonio, TX

Canary Foundation, Palo Alto

#University of California- San Francisco, San Francisco

‡‡Stanford University Medical Center, Stanford, CA

§Fred Hutchinson Cancer Research Center

University of Washington Medical Center, Seattle, WA

**Eastern Virginia Medical School, Norfolk, VA

University of British Columbia, Vancouver, BC, Canada

Conflicts of Interest and Source of Funding: Supported by the Canary Foundation and the Department of Defense (W81XWH-11-1-038). The authors have disclosed that they have no significant relationships with, or financial interest in, any commercial companies pertaining to this article.

Correspondence: Jesse K. McKenney, MD, Section Head, Surgical Pathology, Cleveland Clinic, Robert J. Tomsich-Pathology and Laboratory Medicine Institute, Anatomic Pathology, 9500 Euclid Ave., L25, Cleveland, OH 44195 (e-mail:

Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.