A Comprehensive Analysis of the Association between Gleason Score at a Positive Surgical Margin and the Risk of Biochemical Recurrence after Radical Prostatectomy

Viacheslav Iremashvili, Liset Pelaez, Merce Jorda, Dipen J Parekh, Sanoj Punnen

Research output: Contribution to journalArticle


Our objective was to identify the best of the existing definitions of Gleason score (GS) at a positive surgical margin (PSM) by validating them in our radical prostatectomy cohort. We analyzed 251 patients who had mixed (3+4, 3+5, 4+3 or 5+3) pathologic GS and PSM. We used 5 definitions to record GS at a PSM. Univariate and multivariate analyses were used to study the association between each definition and the risk of biochemical recurrence (BCR). We also tested the prognostic value of multivariate models including established predictors and each of the studied definitions of GS at a PSM. GS 3+3 was seen at a PSM in 57.4% of the cases and was more common in patients with lower overall GS. Over a median follow-up of 4.0 years 89 patients (35.5%) developed BCR. All of the definitions of GS at a PSM were independent predictors of the BCR-free survival. Most of them also improved the prognostic value of the multivariate models when added to the established parameters. The degree of improvement was similar for the most complex definition (full GS at a PSM) and the easiest to record binary definition (presence of Gleason 4/5 pattern at a PSM). We conclude that compared with the other possible options of reporting GS at a PSM, the presence of Gleason 4/5 pattern may be the most practical definition. It is at least as predictive as other definitions, may be the easiest to record and is the best studied of the existing alternatives.

Original languageEnglish (US)
JournalAmerican Journal of Surgical Pathology
StateAccepted/In press - Jan 1 2018



  • Gleason score
  • Key Words:
  • prostate cancer
  • radical prostatectomy
  • surgical margin

ASJC Scopus subject areas

  • Anatomy
  • Surgery
  • Pathology and Forensic Medicine

Cite this