Biopsy Detected Gleason Pattern 5 is Associated with Recurrence, Metastasis and Mortality in a Cohort of Men with High Risk Prostate Cancer

Author's Accepted Manuscript Biopsy-Detected Gleason Pattern 5 is Associated with Recurrence, Metastasis, and Mortality in a Cohort of Men with High-Risk Prostate Cancer Sean P. Stroup , Daniel M. Moreira , Zinan Chen , Lauren Howard , Jonathan H. Berger , Martha K. Terris , William J. Aronson , Matthew R. Cooperberg , Christopher L. Amling , Christopher J. Kane , Stephen J. Freedland PII: DOI: Reference:

S0022-5347(17)77098-X 10.1016/j.juro.2017.07.009 JURO 14848

To appear in: The Journal of Urology Accepted Date: 1 July 2017 Please cite this article as: Stroup SP, Moreira DM, Chen Z, Howard L, Berger JH, Terris MK, Aronson WJ, Cooperberg MR, Amling CL, Kane CJ, Freedland SJ, Biopsy-Detected Gleason Pattern 5 is Associated with Recurrence, Metastasis, and Mortality in a Cohort of Men with High-Risk Prostate Cancer, The Journal of Urology® (2017), doi: 10.1016/j.juro.2017.07.009. DISCLAIMER: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our subscribers we are providing this early version of the article. The paper will be copy edited and typeset, and proof will be reviewed before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to The Journal pertain.

Embargo Policy All article content is under embargo until uncorrected proof of the article becomes available online. We will provide journalists and editors with full-text copies of the articles in question prior to the embargo date so that stories can be adequately researched and written. The standard embargo time is 12:01 AM ET on that date. Questions regarding embargo should be directed to [email protected].

ACCEPTED MANUSCRIPT

Biopsy-Detected Gleason Pattern 5 is Associated with Recurrence, Metastasis, and Mortality in a Cohort of Men with High-Risk Prostate Cancer Sean P. Stroup, MD1-2, Daniel M. Moreira, M.D.,

3

Zinan Chen4,5, Lauren Howard4,5,

Cooperberg, MD, MPH

RI PT

Jonathan H. Berger, MD1, Martha K. Terris, MD6, William J. Aronson, MD7,8, Matthew R. 9,10

, Christopher L. Amling, MD11, Christopher J. Kane, MD2,12,

Stephen J. Freedland, MD 5,13

TE D

M AN U

SC

Affiliations: 1. Department of Urology; Naval Medical Center San Diego, San Diego, CA 2. Department of Urology; University of California, San Diego; San Diego, CA 3. Department of Urology, Mayo Clinic, Rochester, MN 4. Duke University Medical Center, Durham, NC 5. Durham Veterans Affairs Medical Center, Durham, NC 6. Augusta Veterans Affairs Medical Center, Augusta, GA 7. West Los Angeles Veterans Affairs Medical Center, West Los Angeles, CA 8. University of California, Los Angeles School of Medicine, Los Angeles, CA 9. San Francisco Veterans Affairs Medical Center, San Francisco, CA 10. University of California, San Francisco, San Francisco, CA 11. Oregon Health & Sciences University, Portland, OR 12. San Diego Veterans Affairs Medical Center, San Diego, CA 13. Cedars Sinai Medical Center, Los Angeles, CA

AC C

EP

E-mail addresses: Sean P. Stroup: [email protected] Daniel M. Moreira: [email protected] Zinan Chen: [email protected] Lauren Howard: [email protected] Jonathan H. Berger: [email protected] Martha K. Terris: [email protected] William J. Aronson: [email protected] Matthew R. Cooperberg: [email protected] Christopher L. Amling: [email protected] Christopher J. Kane: [email protected] Stephen J. Freedland: [email protected] Running head: Importance of Biopsy-Detected Gleason Pattern 5 Word count: 2333 Tables: 2 Figures: 4 color

Abstract word count: 305

ACCEPTED MANUSCRIPT

RI PT

Corresponding Author: Sean P. Stroup, MD Chair, Department of Urology Naval Medical Center San Diego 34800 Bob Wilson Drive San Diego, CA 92134 T: 619-532-7202 F: 619-532-7234 [email protected]

SC

Key Words: Prostate cancer, radical prostatectomy, biochemical recurrence, prostate cancer-specific death, Gleason Disclaimer

M AN U

This work was supported by the Department of Veterans Affairs, the Department of Defense, Prostate Cancer Research Program (SJF), NIH 1K24CA160653 (SJF) and P50CA92131 (WJA), Georgia Cancer Coalition (MKT). Views and opinions of, and endorsements by the author(s) do not reflect those of the US Army, US Navy, or the

AC C

EP

TE D

Department of Defense.

ACCEPTED MANUSCRIPT

ABSTRACT

PURPOSE: To evaluate the relative risk of biochemical recurrence (BCR), metastasis, and death from prostate cancer contributed by biopsy Gleason pattern 5 among high-

RI PT

risk men with Gleason 8–10 disease in the Shared Equal Access Regional Cancer Hospital (SEARCH) cohort.

PATIENTS AND METHODS: Men with biopsy Gleason sum 8–10 prostate cancer

SC

treated with radical prostatectomy were evaluated. The cohort was divided: Gleason 4+4 vs. those with any pattern 5 (i.e., Gleason 3+5, 5+3, 4+5, 5+4, and 5+5). Predictors

M AN U

of BCR, metastases, prostate cancer-specific survival, and overall survival were analyzed using Kaplan-Meier, log-rank test, and Cox proportional hazards models.

RESULTS: We identified 634 high-risk men in the SEARCH database, of these, 394(62%) had Gleason 4+4, and 240(38%) had Gleason pattern 5 on biopsy. Baseline characteristics were not significantly different between groups. On multivariable

TE D

analysis, relative to Gleason 4+4, high-risk men with Gleason pattern 5 had no differences in the risk of BCR (HR=1.26; 95%CI=0.99–1.61; P=0.065), but a significantly greater risk of metastasis (HR=2.55; 95%CI=1.50-4.35, p=0.001), prostate cancerspecific mortality (HR=2.67; 95%CI=0.1.26–5.66; P=0.010), and overall mortality

EP

(HR=1.60; 95%CI=1.09–2.34; P=0.016).

AC C

CONCLUSION: Preoperative subclassification of high-risk prostate cancer by biopsy Gleason grading (4+4 vs. presence of any Gleason pattern 5) identified men at highest risk of progression. Presence of any Gleason 5 on biopsy is associated with a greater risk of metastasis, prostate cancer-specific mortality, and overall mortality. Grouping all Gleason 8-10 tumors together as “high-risk” may fail to fully stratify those at highest risk of poor outcomes.

ACCEPTED MANUSCRIPT

INTRODUCTION

It is estimated that 220,800 men were diagnosed with and 27,540 men died of prostate cancer in 2015.1 Approximately 15% of these patients meet criteria for high-risk

RI PT

prostate cancer, which includes those with prostate-specific antigen (PSA) >20 ng/mL, biopsy Gleason grade 8–10, and >cT2c findings on digital rectal examination.

2, 3

Although men with biopsy Gleason grade 8–10 are at significant risk of clinical progression, this is a heterogeneous population, and individual outcomes vary

SC

significantly. Traditional risk stratification schemes, such as the Kattan nomogram, consider Gleason score and PSA as continuous variables and report the probability of disease-specific outcomes in a continuous fashion, resulting in wide variation in

M AN U

probabilities.4

Biopsy Gleason grading has consistently been one of the strongest predictors of biochemical recurrence (BCR), disease-specific survival, and overall survival.

5-7

However, many of these studies evaluate Gleason sum 8-10 as one group, thereby losing the ability to differentiate cancer-specific outcomes among those with Gleason

TE D

pattern 5. Several observational studies have identified patients with any component of Gleason pattern 5 as being at significantly increased risk of clinical progression and adverse outcomes.8-10 Tertiary Gleason pattern 5 in the setting of intermediate-risk disease based on radical prostatectomy specimens has also been shown to be an

EP

independent risk factor for adverse outcomes.11 These observations suggest Gleason pattern 5 represents uniquely aggressive tumor biology.

AC C

We sought to substratify men diagnosed with Gleason 8–10 on prostate biopsy to assess outcomes in the Shared Equal Access Regional Cancer Hospital (SEARCH) cohort. We hypothesized that presence of any Gleason pattern 5 on initial prostate biopsy, even among men who were all clinically high risk and treated with prostatectomy, was an independent predictor of biochemical recurrence, metastasis, prostate cancer-specific mortality, and overall survival.

PATIENTS AND METHODS

ACCEPTED MANUSCRIPT

Patient Characteristics After approval was obtained from institutional review boards, data from patients treated with radical prostatectomy between 1988 and 2015 at Veterans Affairs medical centers in San Diego, West Los Angeles, and Palo Alto, California; Augusta, Georgia;

RI PT

and Asheville and Durham, North Carolina, were combined into the SEARCH database. We excluded patients treated with preoperative androgen deprivation or radiation therapy. We included men with Gleason sum 8–10 and a primary Gleason score of 4 or 5 on prostate biopsy who were treated with radical prostatectomy. The cohort was

SC

divided into 2 groups: those with Gleason 4+4, and those with any pattern 5 (e.g., Gleason 3+5, 4+5, 5+4, 5+3, 5+4, and 5+5). There were too few men with primary

M AN U

Gleason 5 (n=57) to analyze as a separate group, so men with any pattern 5 were grouped together. Tertiary scores, including tertiary pattern 5, were not routinely reported on prostate needle biopsies and therefore not included in the analysis. Biochemical recurrence was defined as PSA >0.2 ng/mL, 2 values at 0.2 ng/mL, or secondary treatment for an elevated PSA. All imaging tests (bone scan, MRI, CT, Xray) after surgery were assessed by trained personnel to determine development of

TE D

metastases. Any man who had progressive metastatic prostate cancer while on androgen deprivation therapy and died without another obvious cause of death was considered to have died from prostate cancer.

EP

Statistical Analysis

Clinical and pathological characteristics of patients with high-risk prostate cancer

AC C

were compared among groups using the Wilcoxon rank sum test for continuous variables and the chi-square test for categorical variables. PSA (logarithmically transformed), age, year of surgery, and body mass index (BMI, in kg/m2) were examined as continuous variables, while clinical stage (T1, T2–4, or unknown) was examined as a categorical variable. Time to biochemical recurrence, metastasis, death from prostate cancer, and overall mortality was analyzed using Kaplan-Meier plots and log-rank tests. We constructed uni- and multivariable Cox proportional hazards models, adjusting for year of surgery, medical center, and known predictors of poor outcome including: age, PSA, race, and clinical stage. All statistical analyses were performed

ACCEPTED MANUSCRIPT

using STATA 14.1 (Stata Corp., College Station, TX, USA). A P<0.05 was used to indicate statistical significance.

Clinical and Pathological Characteristics

RI PT

RESULTS

Of 5,073 men who underwent prostatectomy in the SEARCH database, 634 men (10.0%) had high-risk prostate cancer defined by biopsy Gleason grade 8–10. Of these,

SC

394 (62%) had Gleason 4+4, and 240 (38%) had any Gleason pattern 5 on biopsy including Gleason 3+5, 4+5, 5+4, 5+3, 5+4, and 5+5. Table 1 details the baseline

M AN U

clinicopathologic characteristics of the cohort. There were no significant differences in baseline characteristics between groups except those with any Gleason pattern 5 were more likely to have been biopsied and undergone surgery in more recent years (P=0.028).

Biochemical recurrence

TE D

The median follow-up of men without recurrence was 42 months, and there were 293 (46%) recurrences. At this point, the actuarial BCR-free survival was 56% in the Gleason 4+4 group vs. 43% in those with any Gleason pattern 5 (Figure 1, log-rank P=0.008). On univariable analysis, relative to Gleason 4+4, men with any Gleason

EP

pattern 5 had a higher rate of biochemical recurrence (HR 1.37; 95%CI 1.09-1.74, P=0.008). On multivariable analysis, men with any Gleason pattern 5 on preoperative

AC C

biopsy had higher rates of biochemical recurrence (HR 1.26; 95%CI 0.99-1.61, P=0.065, Table 2), although this did not reach statistical significance.

Metastasis

The median follow-up of men without metastasis was 61 months, and 68 (11%)

patients developed metastases. At this point, the actuarial metastasis-free survival was 96% in the Gleason 4+4 group vs. 85% in those with any Gleason pattern 5 (Figure 2, log-rank P<0.001). On univariable analysis, relative to Gleason 4+4, men with any Gleason pattern 5 had shorter time to metastasis (HR 2.63; 95%CI 1.63-4.25, P<0.001).

ACCEPTED MANUSCRIPT

Results remained significant on multivariable analysis (HR 2.55; 95%CI 1.50-4.35, P=0.001, Table 2).

Prostate cancer mortality

RI PT

The actuarial disease-specific survival at a median follow-up of 62 months was 98% in the Gleason 4+4 group vs. 91% in those with any Gleason pattern 5 (Figure 3, log-rank P<0.001). Overall, 42 men died from prostate cancer in the cohort, including 19 (3%) and 23 (4%) in the Gleason 4+4 and any pattern 5 groups, respectively. On

SC

univariable analysis, men with any Gleason pattern 5 were more likely to die of prostate cancer than those with Gleason 4+4 (HR 2.67; 95%CI 1.46-4.86, P=0.001). After

M AN U

adjusting for covariates, we found similar results (HR 2.67; 95%CI 1.26-5.66, P=0.010, Table 2).

Overall mortality

There were 147 (23%) deaths in the follow-up period. The actuarial overall survival at a median follow-up of 59 months was 91% in the Gleason 4+4 group vs. 81%

TE D

in those with any Gleason pattern 5 (Figure 4, log-rank P=0.007). Compared to Gleason pattern 4+4, presence of any pattern 5 predicted a worse overall survival (HR 1.58; 95%CI 1.13-2.21, P=0.008). These results remained significant on multivariable

AC C

DISCUSSION

EP

analysis (HR 1.60; 95%CI 1.09-2.34, P=0.016, Table 2).

In 1974, Gleason described a grading system for prostate cancer based on the

Veterans Administration Cooperative Urological Research Group study.12 The Gleason scoring system is based on low-power, glandular architectural characteristics of prostate cancer cells.13, 14 Although interobserver variability introduces certain limitations, there is a well-established body of literature supporting the consistency of this method. Since its introduction, the biopsy Gleason score has been a powerful predictor of pathological stage, disease-free survival, and adverse outcomes following prostatectomy.15 Even during a time when there has been a shift in Gleason grading with a trend towards

ACCEPTED MANUSCRIPT

upgrading of the final pathologic specimen and limited use of grades 1–2, it is remarkable that assessment of aggressive tumor architecture at initial biopsy remains one of the most powerful predictors of metastases and survival.16 Numerous risk factors for PSA recurrence, metastasis, prostate cancer-specific

RI PT

mortality, and overall mortality have been identified. Biopsy and pathologic Gleason grading remain two of the most significant predictors of outcome, in addition to nodal metastases and seminal vesicle invasion.17 In the current study using the SEARCH cohort, we substratified men with high-risk prostate cancer by initial biopsy Gleason

SC

grade (4+4 vs. presence of any Gleason pattern 5 in those with total grade ≥8). Substratification along these lines identified men with Gleason pattern 5 at initial biopsy

M AN U

as a group at significantly higher risk of adverse cancer-related outcomes. After adjusting for known risk factors, we found that even within this group of already high-risk men, those with any pattern 5 had a significant 2.5-fold greater risk of metastasis and death from prostate cancer, and an increased risk of overall mortality. The actuarial disease-specific and overall survival at a median follow-up of approximately 62 months was 98% vs. 91% and 91% vs. 80% in the Gleason 4+4 group vs. those with any

TE D

Gleason pattern 5, respectively.

Biopsy-detected Gleason grade 5 trended towards significance as a predictor for BCR following prostatectomy (HR 1.26, P=0.065). There are conflicting data regarding the effect of Gleason grading on BCR among high risk patients. Our findings are similar

EP

to those from two single institution retrospective reviews of post-radical prostatectomy patients in which Gleason pattern 5 on biopsy was associated with biochemical 19

Other groups have found no association between biopsy Gleason

AC C

recurrence.18,

grade and BCR. Nanda et al compared PSA outcomes in a cohort of men with high-risk prostate cancer.8 They studied 312 men with a Gleason sum of 7 with tertiary grade 5, Gleason sum 8, or Gleason sum 9–10 who underwent primary therapy with prostatectomy or radiotherapy with or without androgen suppression. After a median follow-up of 5.7 years, they found no significant difference in the risk of PSA recurrence in those with any Gleason pattern 5 (P=0.09). Wambi et al also found no significant differences in biochemical recurrence–free survival between biopsy Gleason 8, 9, and 10 cancers among 368 men treated with robotic prostatectomy.20

ACCEPTED MANUSCRIPT

In 2009, Stephenson, Kattan, and colleagues published a multi-institutional model to predict the risk of prostate cancer-specific mortality.4 On multivariable analysis they identified primary and secondary biopsy Gleason grade ≥4 (P<0.001 for both) and increasing PSA (P=0.021) to be associated with increased prostate cancer-specific

RI PT

mortality. For a hypothetical patient using this model with a biopsy Gleason grade of 4+4=8, PSA of 10 ng/mL, and cT1c prostate cancer, the predicted 10-year prostate cancer-specific mortality is between 1% and 2%. This nomogram would agree with our actuarial disease-specific survival, at a median follow-up of 72 months, of 98% in the

SC

Gleason 4+4 group. However, it does not account for the significant excess mortality observed in those patients with any Gleason pattern 5 who, in our paper, had nearly a

M AN U

10% prostate-specific mortality at 72 months (Figure 3) with a multivariable model showing a greater than two-fold increase in prostate cancer mortality (Table 2). Our findings also support a recently presented update to the Gleason scoring system by Epstein and associates. Citing deficiencies in the Gleason scoring system, they sought to refine the system. On reviewing a series of prostate cancer patients (20,845 treated with radical prostatectomy and over 5,551 treated by radiotherapy), they

TE D

determined that a five-grade scoring group (Grades 1-5) allowed for the most appropriate prognostic distinction. Notably, in the recommended grading system, Gleason sums of 9 or 10 are given the highest score (worst prognosis) in the newly proposed system, as presence of Gleason pattern 5 (regardless of 4+5, 5+4, or 5+5)

EP

appeared to be an independent variable for prostate cancer-related outcomes.21 In the current paper we did not evaluate tertiary patterns because of their inconsistent

AC C

reporting across the study sites. Furthermore, according to the 2014 International Society of Urological Pathology conference, tertiary patterns should only be recorded for radical prostatectomy specimens with either 3+4=7 or 4+3=7 with tertiary pattern 5. We also performed a subset analysis of biopsy-detected Gleason Grade Group 4

to evaluate the unique risk for each group. Interestingly, when evaluating 4+4=8 vs. 3+5=8 or 5+3=8 we found no differences within this subgroup for BCR, metastasis-free survival, cancer specific survival, or overall survival. This suggests that when Gleason 5 is associated with Gleason 3 on biopsy, the Gleason 5 component may be an over-read of the true pathology. These findings are similar to those of Harding-Jackson et al.,

ACCEPTED MANUSCRIPT

who, using the most contemporary standards of prostate cancer pathology, and rereview, found that Gleason grade 4+4=8 and 3+5=8 had a similar prognosis in terms of metastases and overall survival.9

RI PT

After adjusting for known risk factors, men with Gleason pattern 5 at initial biopsy were at significantly greater risk of developing metastases (HR=1.93; Table 2) and dying from prostate cancer (HR=2.01; Table 2). Many previous studies evaluating the prognostic value of initial biopsy grade among high-risk men have been limited by

SC

analyzing all Gleason high-grade 8–10 patients together or by limiting the analysis to final Gleason grading of pathological specimens, or by limitation to a single institution.

M AN U

Our findings are similar to those of Tsao et al. who also identified that Gleason 9-10 disease treated with definitive local therapy had worse outcomes than those with biopsy Gleason 8 disease.10 Similar to the current study, these authors also found a trend toward a shorter time to BCR for Gleason 8 vs. 9-10 disease (HR 1.13, 95% CI 0.931.36), but significantly worse overall survival. Although a portion of high-risk men will be cured by surgery alone, our data suggests that further stratification by presence or

TE D

absence of biopsy-detected Gleason grade 5 carries important clinical relevance. Accurate assessment of initial biopsy data is crucial for all men diagnosed with prostate cancer.

EP

This retrospective study has several limitations. Several studies described the lack of consistency between Gleason score in needle biopsies and subsequent

AC C

prostatectomy specimens. In a prior study from the SEARCH Database, Kane et al found that a majority of patients with high-risk disease were characterized as such based on Gleason grade alone, but they noted that 34% to 55% of biopsy-detected Gleason 8–10 cancers are downgraded to Gleason 7 or less at the time of surgery.17 Despite possible variations in Gleason grading, our analysis focused on preoperative evaluation of biopsy data to refine how we estimate risk in these patients. Moreover, even after prostatectomy, biopsy Gleason score remains prognostic, supporting the idea that biopsy Gleason grade is a vital component of risk stratification.5 Second, biopsy schemes and techniques varied both throughout the period of analysis and with respect

ACCEPTED MANUSCRIPT

to each hospital’s location. Third, the study period spans the PSA era and the resulting stage migration would have resulted in many higher clinical stage tumors in the early years. To account for these variations over time, we adjusted for year of surgery to mitigate the effect of stage migration on our cohort. Intra- and inter-observer variability

RI PT

combined with undergrading or overgrading of prostate cancer needle biopsy specimens may have introduced some limitations to the analysis, and we were unable to conduct centralized pathologic review. These limitations are present in many retrospective reviews and likely reflect actual practice variation. Further validation of our

SC

findings with longer follow-up are required. Despite these limitations, SEARCH is a multicenter, equal access cohort with an intermediate duration of follow-up and robust

M AN U

findings on multivariable analysis that should be generalizable to the greater community.

CONCLUSION

Prostate biopsy subclassification of high-risk disease (Gleason 4+4 vs. presence

TE D

of any pattern 5) can identify men at the greatest risk of metastasis and prostate cancer mortality. Presence of any Gleason pattern 5 on biopsy portends a poor prognosis, is associated with a significantly greater risk of metastasis and prostate cancer specific mortality, and negatively impacts overall survival. These findings support the concept of

EP

a continuum of risk, even in these high-risk cases, and preoperative risk assessment

AC C

should consider presence of Gleason pattern 5 as a uniquely aggressive feature.

ACCEPTED MANUSCRIPT

REFERENCES

1.

Altekruse, S., Kosary, C., Krapcho, M. et al.: SEER Cancer Statistics Review, 1975-

RI PT

2007, based on November 2009 SEER data submission, posted to the SEER web site, 2010. ed. Bethesda, MD: National Cancer Institute 2.

Cooperberg, M. R., Lubeck, D. P., Mehta, S. S. et al.: Time trends in clinical risk

SC

stratification for prostate cancer: implications for outcomes (data from CaPSURE). J Urol, 170: S21, 2003

D'Amico, A. V., Cote, K., Loffredo, M. et al.: Pretreatment predictors of time to cancer

M AN U

3.

specific death after prostate specific antigen failure. J Urol, 169: 1320, 2003 4.

Stephenson, A. J., Kattan, M. W., Eastham, J. A. et al.: Prostate cancer-specific mortality after radical prostatectomy for patients treated in the prostate-specific antigen era. J Clin Oncol, 27: 4300, 2009

Fitzsimons, N. J., Presti, J. C., Kane, C. J. et al.: Is biopsy Gleason score independently

TE D

5.

associated with biochemical progression following radical prostatectomy after adjusting for pathological Gleason score? J Urol, 176: 2453, 2006 Pound, C. R., Partin, A. W., Eisenberger, M. A. et al.: Natural history of progression after

EP

6.

PSA elevation following radical prostatectomy. JAMA, 281: 1591, 1999 D'Amico, A. V., Moul, J., Carroll, P. R. et al.: Cancer-specific mortality after surgery or

AC C

7.

radiation for patients with clinically localized prostate cancer managed during the prostate-specific antigen era. J Clin Oncol, 21: 2163, 2003

8.

Nanda, A., Chen, M. H., Renshaw, A. A. et al.: Gleason Pattern 5 prostate cancer:

further stratification of patients with high-risk disease and implications for future randomized trials. Int J Radiat Oncol Biol Phys, 74: 1419, 2009

ACCEPTED MANUSCRIPT

9.

Harding-Jackson N., Kryvenko O. N., Whittington E. E. et al.: Outcome of Gleason 3 + 5 = 8 prostate cancer diagnosed on needle biopsy: prognostic comparison with Gleason 4 + 4 = 8. J Urol, 196: 1076, 2016 Tsao, C.K., Gray K. P., Nakabayashi M. et al.: Patients with biopsy Gleason 9 and 10

RI PT

10.

prostate cancer have significantly worse outcomes compared to patients with Gleason 8 disease. J Urol, 194: 91, 2015.

Whittemore, D. E., Hick, E. J., Carter, M. R. et al.: Significance of tertiary Gleason

SC

11.

pattern 5 in Gleason score 7 radical prostatectomy specimens. J Urol, 179: 516, 2008 Gleason, D. F., Mellinger, G. T.: Prediction of prognosis for prostatic adenocarcinoma by

M AN U

12.

combined histological grading and clinical staging. J Urol, 111: 58, 1974 13.

Bostwick, D. G.: Gleason grading of prostatic needle biopsies. Correlation with grade in 316 matched prostatectomies. Am J Surg Pathol, 18: 796, 1994 Bostwick, D. G.: Grading prostate cancer. Am J Clin Pathol, 102: S38, 1994

15.

Kang, D. E., Fitzsimons, N. J., Presti, J. C. et al.: Risk stratification of men with Gleason

TE D

14.

score 7 to 10 tumors by primary and secondary Gleason score: results from the SEARCH database. Urology, 70: 277, 2007 Miyamoto, H., Hernandez, D. J., Epstein, J. I.: A pathological reassessment of organ-

EP

16.

confined, Gleason score 6 prostatic adenocarcinomas that progress after radical

17.

AC C

prostatectomy. Hum Pathol, 40: 1693, 2009 Kane, C. J., Presti, J. C., Amling, C. L. et al.: Changing nature of high risk patients

undergoing radical prostatectomy. J Urol, 177: 113, 2007

18.

Jo, J. K., Kook, H. R., Byun, S. S. et al.: Stratification of Contemporary Patients

Undergoing Radical Prostatectomy for High-risk Prostate Cancer. Ann Surg Oncol, 22: 2088, 2015 19.

Sundi, D., Wang, V., Pierorazio, P. M. et al.: Identification of men with the highest risk of early disease recurrence after radical prostatectomy. Prostate, 74: 628, 2014

ACCEPTED MANUSCRIPT

20.

Wambi, C. O., Siddiqui, S. A., Krane, L. S. et al.: Early oncological outcomes of robotassisted radical prostatectomy for high-grade prostate cancer. BJU Int, 106: 1739, 2010 Epstein, J. I., Zelefsky, M. J., Sjoberg, D. D. et al.: A Contemporary Prostate Cancer

EP

TE D

M AN U

SC

RI PT

Grading System: A Validated Alternative to the Gleason Score. Eur Urol, 69: 428, 2016

AC C

21.

ACCEPTED MANUSCRIPT

TABLES AND FIGURES

Baseline patient characteristics comparing any biopsy Gleason score 5 and Gleason 4+4

Table 2.

BCR-free, metastasis-free, disease-specific and overall survival

RI PT

Table 1.

comparing any biopsy Gleason score 5 and Gleason 4+4

Kaplan-Meier estimates of freedom from biochemical failure comparing

SC

Figure 1.

patients with Gleason 4+4 (solid line) to those with any Gleason 5 (sum ≥8, dashes). Biochemical recurrence is defined as PSA > 0.2 ng/mL, 2

Figure 2.

M AN U

values at 0.2 ng/mL, or secondary treatment for an elevated PSA.

Kaplan-Meier estimates of freedom from metastasis comparing those with biopsy Gleason 4+4 (solid line) and those with any biopsy Gleason pattern

Figure 3.

TE D

5 (sum ≥8, dashes).

Kaplan-Meier estimates of prostate cancer-specific survival among those with biopsy Gleason 4+4 (solid line) and those with any biopsy Gleason

Figure 4.

EP

pattern 5 (sum ≥8, dashes).

Kaplan-Meier estimates of overall survival among those with biopsy Gleason 4+4 (solid line) and those with any biopsy Gleason 5 (sum ≥8,

AC C

dashes).

ACCEPTED MANUSCRIPT

Table 1: Baseline patient characteristics comparing any biopsy Gleason score 5 and Gleason 4+4 Gleason 4+4

Any Gleason 5

P

N (%) Age (years), median (IQR) PSA (ng/mL), median (IQR)

394 (62) 64 (59–67) 7.2 (5.0–11.5)

240 (38) 65 (60–68) 7.8 (5.2–13.6)

-0.056† 0.070†

BMI (kg/m2), median (IQR)

28.2 (25.2–31.2)

28.3 (26.0–31.7)

0.388†

0.098‡

235 (60) 142 (36) 17 (4)

Biopsy Gleason score, N (%)

0.410‡

123 (51) 109 (46) 8 (3)

M AN U

223 (57) 158 (40) 13 (3)

161 (67) 74 (31) 5 (2)

SC

Race, N (%) White Black Other Clinical stage, N (%) T1 T2 or greater Unknown

--

TE D

394 0 0 50 (21) 0 11 (4) 0 133 (55) 0 30 (13) 0 16 (7) 2008 2010 Year of surgery, median (IQR) 0.028† (2003–2012) (2003–2012) BMI: body-mass index, IQR: interquartile range, PSA: prostate-specific antigen. † Wilcoxon rank sum test ‡Chi-square test *31 BMI missing

AC C

EP

4+4 3+5 5+3 4+5 5+4 5+5

RI PT

Characteristic

ACCEPTED MANUSCRIPT

Table 2: BCR-free, metastasis-free, disease-specific and overall survival of any Gleason Pattern 5 vs. Gleason 4+4 (reference)

P

HR (95%CI)

P

Univariable

1.37 (1.09-1.74)

0.008

2.63 (1.63-4.25)

<0.001

Multivariable†

1.26 (0.99-1.61)

0.065

2.55 (1.50-4.35)

0.001

2.67 (1.46-4.86)

0.001

2.67 (1.26-5.66)

0.010

M AN U

HR (95%CI)

Disease-specific survival HR P (95%CI)

RI PT

Model*

Metastasis-free survival

SC

BCR-free survival

Overall survival HR (95%CI) 1.58 (1.13-2.21) 1.60 (1.09-2.34)

P 0.008

0.016

AC C

EP

TE D

BCR: biochemical-recurrence; CI: confidence interval; HR: hazard ratio. † Model adjusted for age, race, pre-operative PSA, medical center, year of surgery and clinical stage. *BCR-free survival N= 618, 298 recurrences; Metastasis-free survival N=632, 53 metastasis; Disease-specific survival N=633, 42 dead of PC; Overall survival N=633, 149 all cause death.

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

body mass index

BCR

biochemical recurrence

PCSM

prostate cancer specific mortality

DSS

disease specific survival

OS

overall survival

PSA

prostate-specific antigen

SEARCH

Shared Equal Access Regional Cancer Hospital

AC C

EP

TE D

M AN U

SC

BMI

RI PT

Abbreviations