EURURO Vol. 71 No. 5

Platinum Priority

[1_TD$DIFF]

– Editorial

Referring to the article published on pp. 760–763 of this issue

Prostate Cancer Grade Groups Correlate with Prostate-specific

Cancer Mortality: SEER Data for Contemporary Graded Specimens

Jonathan I. Epstein

Departments of Pathology, Urology, and Oncology, The Johns Hopkins Medical Institutions, Baltimore, MD, USA

In 2013, Epstein and colleagues at Johns Hopkins University

proposed a novel prognostic risk category grade grouping

for prostate cancer

[1]

. Unlike Gleason scores, the grade

groups use a scale from 1 to 5, and tumors with the most

favorable features (Gleason score 6) are referred to as

grade group 1. Subsequently, the new grade group system

was validated in a multi-institutional study including

20 845 consecutive men treated with radical prostatectomy

at five large academic centers, and 5501 men treated with

radiation therapy at two centers. In the surgery cohort, the

5-yr biochemical recurrence (BCR)–free survival based on

prostatectomy grade was 96% for grade group 1 to 5, 88% for

grade group 2, 63% for grade group 3, 48% for grade group 4,

and 26% for grade group 5

[2] .

[2_TD$DIFF]

Assigning a risk of 1 to

Gleason score 6, the relative risk of progression was 2.6 for

grade group 2, 8.5 for grade group 3, 16.8 for grade group 4,

and 29.3 for grade group 5. Grade groups were also

validated on biopsy and correlated with risk of biochemical

progression after radical prostatectomy and following

radiation therapy. The strengths of this validation study

included multi-institutional and international data, the

largest contemporary follow-up study after radical prosta-

tectomy yet assembled, and similar methods for evaluating

and processing specimens by urologic pathologists.

Following these initial studies, numerous efforts have

validated the new grading system using similar surgery

and radiation cohorts

[3–7] .

The major limitation of all the

above studies was the use of prostate-specific antigen BCR

as an endpoint as opposed to cancer-related death. This

limitation was unavoidable, as the modified Gleason

system was first introduced in 2005 and there was

insufficient follow-up length relative to the number of

cases studied to have enough cancer-related deaths for

meaningful analysis. A unique study to look at the five-

tiered grade groups was carried out by Berney et al

[8]

using

the Transatlantic Prostate Group cohort; the study

demonstrated that the five grade groups also correlated

well with death due to prostate cancer among 988 men

treated with watchful waiting. As noted by the authors,

there were inherent weaknesses in this study, as cases were

included from as far back as 1990. The diagnosis and

treatment of prostate cancer during the time period of this

study were very different to current practice, such as the

use of sextant core biopsy as the most common biopsy

technique. However, the advantage of using these older

cases was that there was sufficient follow-up for the study

to assess death from prostate cancer.

In this issue of

European Urology

, He et al

[9]

also

demonstrate that the new five-tiered grade grouping

correlates with prostate cancer–specific mortality using

population-based Surveillance, Epidemiology, and End

Results (SEER) data for 331 320 men. As with the Berney

study, there are limitations and strengths of SEER data

analysis. He et al correctly chose 2006 as the starting date

for their study, as prostate cancer grades assigned before

the International Society of Urological Pathology (ISUP)

consensus conference in 2005 cannot be assumed to be

accurate. In particular, many cases of Gleason score 6 graded

before 2005 would be currently graded as Gleason score 7,

since the definition of Gleason pattern 4 changed in

2005. The weakness acknowledged by He et al is that their

study is based on pathology reports from multiple

institutions, with the vast majority of these lacking urologic

pathology expertise. This shortcoming is evident from the

E U R O P E A N U R O L O G Y 7 1 ( 2 0 1 7 ) 7 6 4 – 7 6 5

available at

www.scienced irect.com

journal homepage:

www.europeanurology.com

DOI of original article:

http://dx.doi.org/10.1016/j.eururo.2016.11.031

* The Johns Hopkins Hospital, Room 2242, The Weinberg Building, 401 N. Broadway Street, Baltimore, MD 21231, USA. Tel. +1 410 9555043;

Fax: +1 443 2873818.

E-mail address:

jepstein@jhmi.edu . http://dx.doi.org/10.1016/j.eururo.2016.12.014

0302-2838/