- Email: [email protected]
Prognostic significance of Gleason score 3+4 versus Gleason score 4+3 tumor at radical prostatectomy
ADULT UROLOGY
PROGNOSTIC SIGNIFICANCE OF GLEASON SCORE 3⫹4 VERSUS GLEASON SCORE 4⫹3 TUMOR AT RADICAL PROSTATECTOMY THERESA Y. CHAN, ALAN W. PARTIN, PATRICK C. WALSH,
AND
JONATHAN I. EPSTEIN
ABSTRACT Objectives. To determine the clinical significance of Gleason score 3⫹4 versus 4⫹3 on radical prostatectomy. Methods. Of 2390 men who underwent radical prostatectomy by a single surgeon, 570 had Gleason score 7 tumors without lymph node metastasis, seminal vesicle invasion, or tertiary Gleason pattern 5. Patients were evaluated for biochemical recurrence (prostate-specific antigen progression) and distant metastases. Results. Eighty percent of patients had Gleason score 3⫹4, 20% had 4⫹3. The rate of established extraprostatic extension at radical prostatectomy for Gleason score 3⫹4 and 4⫹3 tumors was 38.2% and 52.7%, respectively (P ⫽ 0.008). With a mean follow-up of 4.6 years for men without progression, Gleason score 4⫹3 tumors had an increased risk of progression independent of stage and margin status (P ⬍0.0001). The 5-year actuarial risk of progression was 15% and 40% for Gleason score 3⫹4 and 4⫹3 tumors, respectively. The mean time to progression was 4.4 years for Gleason score 3⫹4 tumors and 3.2 years for Gleason score 4⫹3 tumors. We stratified the patients into four prognostic groups on the basis of organ-confined status, margin status, and Gleason score (3⫹4 versus 4⫹3). The 5-year actuarial risk of progression was 10%, 35%, 45%, and 61%, with 10-year progression rates of 29%, 42%, 69%, and 84%, for the four groups. 3.9% of patients with Gleason score 3⫹4 and 10.5% with Gleason score 4⫹3 tumors developed metastatic disease within a mean of 5.7 and 5.6 years, respectively. A Gleason score of 4⫹3 versus 3⫹4 was predictive of metastatic disease (P ⫽ 0.002) but not local recurrence. Conclusions. Gleason score 7 tumors are heterogeneous in their biologic behavior. The differences in prognosis for patients with Gleason scores 3⫹4 and 4⫹3 tumors at radical prostatectomy are significant. Although the assessment of the percentage of pattern 4 at radical prostatectomy is not likely to be reproducible, the distinction between Gleason score 3⫹4 and 4⫹3 should be easier for pathologists to perform. UROLOGY 56: 823–827, 2000. © 2000, Elsevier Science Inc.
T
he Gleason score assigned to prostate cancer in the radical prostatectomy specimen is strongly predictive of postoperative progression. Gleason scores 2-4 and 8-10 account for only a small percentage of the total number of patients who undergo radical prostatectomy, and the prognosis for these groups is dictated largely by the score alone.1 The differences in progression of tumors with This study was supported in part by a National Institutes of Health SPORE grant for prostate cancer. From the Departments of Urology and Pathology, Johns Hopkins University School of Medicine, James Buchanan Brady Urological Institute, Johns Hopkins Hospital, Baltimore, Maryland Reprint requests: Jonathan I. Epstein, M.D., Department of Pathology, Johns Hopkins Hospital, 600 North Wolfe Street, Baltimore, MD 21287 Submitted: May 8, 2000, accepted: June 16, 2000 © 2000, ELSEVIER SCIENCE INC. ALL RIGHTS RESERVED
Gleason score 5-6 have been shown to be based on the margin status and extraprostatic extension.1 Gleason score 7 tumors have been shown to behave significantly worse than Gleason score 5-6 tumors, but have a better prognosis than Gleason score 8-9 tumors.2 However, the prognosis of Gleason score 7 tumors is not uniformly poor. In an earlier study of Gleason score 7 tumors, differences in the risk of progression were found 8 years after surgery, depending on the pathologic stage and margin status.1 A follow-up study that included a larger group of men followed up for a longer period determined that more than 50% of men with Gleason score 7 tumors without extraprostatic extension and without positive margins appeared to be cured.3 In this study, we address the question of whether the dominant pattern in Gleason score 7 0090-4295/00/$20.00 PII S0090-4295(00)00753-6 823
TABLE I. Distribution of patients with Gleason score 3ⴙ4 and 4ⴙ3 tumors stratified by surgical margin and organ-confinement status
Organ confined Focal extraprostatic extension Negative margins Positive margins Established extraprostatic extension Negative margins Positive margins
Total (n ⴝ 570)
3ⴙ4 (n ⴝ 458)
4ⴙ3 (n ⴝ 112)
198 138 117 21 234 172 62
162 121 106 15 175 125 50
36 17 11 6 59 47 12
(34.7) (24.2) (20.5) (3.7) (41.0) (30.2) (10.9)
(35.4) (26.4) (23.1) (3.3) (38.2) (27.3) (10.9)
(32.1) (15.2) (9.8) (5.4) (52.7) (42.0) (10.7)
Numbers in parentheses are percentages.
tumors (score 3⫹4 versus 4⫹3) is an independent predictor of progression after radical prostatectomy. MATERIAL AND METHODS Between September 1982 and May 1998, 2390 consecutive men underwent radical prostatectomy by a single surgeon at Johns Hopkins Hospital. There were 608 patients (25.4%) with Gleason score 7 tumors without lymph node metastasis or seminal vesicle invasion. Patients with capsular incision and no follow-up were excluded. Patients with Gleason score 7 tumors with tertiary Gleason pattern 5 were also excluded, as we have demonstrated that this group has a worse prognosis.4 The remaining 570 patients (23.8%) formed the basis for this study. All men were evaluated preoperatively with digital rectal examination of the primary lesion and bone scan. None of the patients received neoadjuvant hormonal or radiation therapy. Radical prostatectomy was performed by an anatomic approach, as previously described.5 In general, patients were followed up with digital rectal examination and serum prostate-specific antigen (PSA) levels every 3 months for 1 year, twice yearly for the next 2 years, and yearly thereafter. Serum PSA levels were determined by the Hybritech immunoradiometric assay. Postoperative bone scans were performed when patients complained of bone pain or had detectable PSA levels. Patients were considered to have disease progression when they had elevated postoperative serum PSA levels of greater than 0.2 ng/mL, evidence of local recurrence, or radiologic evidence of distant metastases. Patients received hormonal or radiation therapy when progression occurred. The prostates were either serially sectioned and entirely embedded or processed according to a sampling method, which has been shown to result in accurate pathologic staging.6,7 Tumors were determined to be Gleason score 3⫹4 or 4⫹3. Tumors were staged as organ confined or as having focal or established extraprostatic extension. Focal extraprostatic extension refers to only a few neoplastic glands immediately exterior to the prostate in one to two sections; established extraprostatic extension denotes tumor with a greater degree of extraprostatic spread. Statistical analysis was performed using STATA Software (College Station, Tex). The endpoints measured as time to progression were calculated by the Kaplan-Meier method for estimation. The Wilcoxon-Gehan test was used to determine whether the differences between the Kaplan-Meier curves were statistically significant. The assessment of multiple variables to determine which were independent prognostic indicators of disease progression was performed by the Cox proportional hazards analysis. 824
RESULTS The average age of the 570 men at the time of surgery was 59 years (median 60, range 39 to 74). Gleason score 3⫹4 tumor was identified in 458 (80%) of the 570 men, with the remaining men having Gleason score 4⫹3 cancer. The overall mean serum PSA level was 8.6 ng/mL (median 6.9). The average PSA level was 8.4 ng/mL (median 6.8) in men with Gleason score 3⫹4 tumors and 9.8 ng/mL (median 8.0) in men with Gleason score 4⫹3 tumors (P ⫽ 0.04). STAGE AND MARGINS Table I depicts the subclassification of patients with Gleason score 3⫹4 and 4⫹3 on the basis of the presence and extent of extraprostatic spread and surgical margin status. The overall extent of disease was organ confined in 198 patients (34.7%), focal extraprostatic extension in 138 patients (24.2%), and established extraprostatic extension in 234 patients (41.1%). Tumors with Gleason score 4⫹3 had more advanced disease than those with Gleason score 3⫹4 (P ⫽ 0.008). For example, the rate of established extraprostatic extension in radical prostatectomy specimens with Gleason score 3⫹4 and 4⫹3 tumors was 38.2% and 52.7%, respectively. The surgical margins were positive in 83 patients (14.6%); the surgical margin status was not significantly different between patients with Gleason score 3⫹4 tumors and those with Gleason score 4⫹3 tumors. PROGRESSION The mean follow-up for men without progression was 4.6 years (median 4, range 1 to 15). The mean follow-up for men with progression was 4.0 years (median 3, range 1 to 10). Of the 456 men without progression, 369 were followed up for more than 2 years and 114 for more than 7 years. The Kaplan-Meier progression-free curves for patients with Gleason score 3⫹4 and Gleason score 4⫹3 tumors are illustrated in Figure 1. Univariate UROLOGY 56 (5), 2000
FIGURE 1. Actuarial risk of progression in men with Gleason score 3⫹4 and score 4⫹3 tumors without lymph node or seminal vesicle involvement.
analysis showed that Gleason score 3⫹4 versus 4⫹3 was significant in predicting progression (P ⬍0.00001). Gleason score 4⫹3 tumors had an increased risk of progression independent of stage and margin status (P ⬍0.0001). The multivariate analysis also showed that surgical margin status had a greater influence in predicting progression (P ⬍0.0001) than the extent of extraprostatic extension (P ⫽ 0.015). Preoperative PSA levels did not add to the multivariate model. The 5-year actuarial risk of progression was 15% and 40% and the 10year actuarial risk of progression was 37% and 56% for Gleason score 3⫹4 and 4⫹3 tumors, respectively. The mean time to progression was 4.4 years and 3.2 years for Gleason score 3⫹4 and 4⫹3 tumors, respectively. Patients were stratified into four prognostic groups on the basis of similar risks of progression (Fig. 2). The number of men in groups 1 to 4 was 429, 26, 97, and 18, respectively. For the four groups, the 5-year actuarial risk of progression was 10%, 35%, 45%, and 61%, with 10-year progression rates of 29%, 42%, 69%, and 84%, respectively. It is likely that the lack of differences between groups 2 and 3 and between groups 2 and 4 were due in part to the relatively small number of patients in groups 2 and 4 (Table II). Thirty patients developed metastatic disease, 21 patients had local recurrence, and 6 patients had both. The stage of disease in patients with distant metastases was organ confined in 16.7%, focal extraprostatic extension in 23.3%, and established extraprostatic extension in 60%. Metastatic disease developed in 3.9% of patients with Gleason score 3⫹4 tumors and 10.5% with Gleason score 4⫹3 tumors within a mean time of 5.7 and 5.6 years, respectively. Gleason score 4⫹3 tumors had an increased risk of metastatic disease (P ⫽ 0.002), independent of the surgical margin status and extent UROLOGY 56 (5), 2000
FIGURE 2. Actuarial risk of progression in men with Gleason score 3⫹4 and score 4⫹3 tumors without lymph node or seminal vesicle involvement stratified into four prognostic groups. oc ⫽ organ confined; epe ⫽ extraprostatic extension; fepe ⫽ focal extraprostatic extension; eepe ⫽ established extraprostatic extension; mar (⫺) ⫽ margins negative; mar (⫹) ⫽ margins positive.
TABLE II. Statistically significant differences in progression Group
P Value
1. Organ confined or EPE and negative margins and 3⫹4 vs. 2. FEPE and negative margins and 0.0003 4⫹3 or FEPE and positive margins and 3⫹4 3. EEPE and negative margins and ⬍0.00001 4⫹3 or EEPE and positive margins and 3⫹4 4. EPE and positive margins and ⬍0.00001 4⫹3 2. FEPE and negative margins and 4⫹3 or FEPE and positive margins and 3⫹4 vs. 3. EEPE and negative margins and ⬎0.05 4⫹3 or EEPE and positive margins (not significant) and 3⫹4 4. EPE and positive margins and 0.034 4⫹3 3. EEPE and negative margins and 4⫹3 or EEPE and positive margins and 3⫹4 vs. 4. EPE and positive margins and 4⫹3 ⬎0.05 (not significant) EPE ⫽ extraprostatic extension; FEPE ⫽ focal EPE; EEPE ⫽ established EPE.
of extraprostatic extension. A Gleason score 3⫹4 versus 4⫹3 was not predictive of local recurrence. COMMENT The most common grades of the index (dominant) tumor seen at radical prostatectomy are 825
Gleason scores 6 and 7.1 As opposed to Gleason score 6, which is typically composed of only one pattern, Gleason score 7 tumors contain patterns 3 and 4 in various proportions. Given the marked adverse affect of Gleason pattern 4, one would expect that whether a tumor is Gleason score 3⫹4 or 4⫹3 would influence prognosis. Several studies have addressed the question of Gleason score 3⫹4 versus 4⫹3 at radical prostatectomy with somewhat conflicting results. In a study of 116 men with high-grade carcinoma and a median follow-up of 7 years, the group from Northwestern University reported a subset of 77 patients with Gleason score 7 tumors with no significant survival advantage identified for Gleason pattern 3⫹4 compared with 4⫹3.8 The lack of statistically significant differences in their study may be due to the small group of patients and the inclusion of patients with positive lymph nodes and/or seminal vesicle invasion. At Wayne State University, 610 men with Gleason score 7 cancer were analyzed and the distinction between 3⫹4 versus 4⫹3 correlated with both stage and progression. However, the median follow-up was only 25.8 months, and the difference between the two grades was independently predictive only in men with serum PSA values less than 10 ng/mL.9 The group from Baylor reported on 853 men with a 24-month median follow-up and found that 3⫹4 versus 4⫹3 correlated with progression, yet was not independently statistically significant (personal communication).10 Another investigation from the University of Michigan in a study of 361 men with only a 12-month median follow-up for those without progression found that Gleason score 3⫹4 versus 4⫹3 predicted progression, yet it was not reported whether it was an independent prognosticator.11 The question remained, once other routinely assessed parameters such as pathologic stage and margins are factored in, is it important to know whether a tumor is Gleason score 3⫹4 versus 4⫹3? In our study, a significant difference in recurrence-free survival rates was found between patients with Gleason score 3⫹4 and score 4⫹3 tumors, independent of surgical margin status and extraprostatic extension. We were able to stratify men with Gleason score 7 tumors into four very different prognostic groups on the basis of organconfined status, margin status, and Gleason score 3⫹4 versus 4⫹3. The 5-year actuarial progressionfree rate for these four groups was 90%, 65%, 55%, and 39%. Patients with the best prognosis were those who had either Gleason 3⫹4 or 4⫹3 and organ-confined disease or who had extraprostatic extension of any degree with Gleason score 3⫹4 and negative surgical margins. In the next worse prognostic group, men had focal extraprostatic ex826
tension and only one adverse finding (ie, either 3⫹4 with positive margins or 4⫹3 with negative margins). The next worse prognostic group had established extraprostatic extension with again only one adverse finding in terms of grade and surgical margins. The patients with the worst prognosis were those who had focal or established extraprostatic extension and both adverse findings (ie, positive margins and Gleason score 4⫹3). Although the differences between some of these groups did not reach statistical significance, that lack most likely resulted from the smaller number of patients in some of the subgroups, since the prognostic differences among the groups were what one would have predicted. That is, the greater number of adverse findings in terms of the degree of extraprostatic extension, 4⫹3 versus 3⫹4, and margin positivity, the worse the prognosis. In our multivariate analysis, surgical margin status was more influential than the extent of extraprostatic extension in predicting progression after radical prostatectomy. Regardless of the extent of extraprostatic extension, patients with Gleason score 3⫹4 tumors and negative margins fared as well as those with organ-confined disease and better than patients with Gleason score 3⫹4 and positive margins. Likewise, regardless of the extent of extraprostatic extension, patients with Gleason score 4⫹3 tumors and negative margins had a better prognosis than patients with score 4⫹3 tumors and positive margins. Patients with Gleason score 4⫹3 tumors, if they had negative margins, even had progression rates comparable with patients with Gleason score 3⫹4 tumors and positive margins. It is unclear whether the adverse effect of positive margins relates to the intrinsic biology of disease or to the ability to achieve local control. The results in our study showing the significance of the extent of extraprostatic extension and surgical margins on progression differ from various previous reports. Oefelein et al.12 demonstrated that although Gleason score, extraprostatic extension, and margin status were all independent prognostic factors in predicting progression after radical prostatectomy in 238 patients, in a subset of 52 men with Gleason score 7 tumors, extraprostatic extension was not predictive of progression. In contrast to our study, their series included patients with positive lymph nodes and seminal vesicle invasion. Ohori et al.13 found that seminal vesicle invasion, extraprostatic extension, and Gleason score, but not margin status, correlated with progression. They reported that in a subset of 97 patients with Gleason score 7 or more with extraprostatic extension and negative lymph nodes and seminal vesicles, margin status did not correlate with progression.13 The lack of statistical significance for certain pathologic parameters correlating with UROLOGY 56 (5), 2000
progression seen in the above studies may relate to the lack of sufficient sample size for stratification into multiple subgroups. The group from Stanford has been a strong proponent of using the proportion of high-grade tumor (Gleason pattern 4 and 5) as the preferred method for grading prostate cancer.14,15 However, the percentage of pattern 4/5 is only very predictive for progression at the extremes (greater than 70% or less than 20% with pattern 4/5).14 It has not been demonstrated that classifying tumors based on their percentage of pattern 4/5 is more predictive than stratifying patients into Gleason score 3⫹4 versus 4⫹3. Furthermore, assessing the percentage of Gleason pattern 4 is often difficult and not likely to be performed routinely. Often patterns 4 and 3 are intimately admixed such that their relative percentages are not readily calculable. Additional difficulty in asking pathologists to derive a specific percentage of pattern 4/5 stems from studies demonstrating interobserver variability in grading tumors with Gleason scores 5-7.16 In one study of needle biopsy grading, of the 70 cases diagnosed as Gleason score 7 on the outside, on review at Johns Hopkins Hospital, 17 (24%) were diagnosed as having a Gleason score less than 7 and 4 (6%) as having a Gleason score greater than 7. Of 320 cases diagnosed on the outside as Gleason score less than 7 or Gleason score greater than 7, 53 (17%) were diagnosed as Gleason score 7 on review at our institution.17 Even among uropathology experts, there may be difficulty in differentiating between Gleason score 6 and 7. Accurate use of the Gleason grading system also depends on the pathologists’ experience and training. In a study using a webbased tutorial, the most difficult tumors for participants to grade were those with Gleason scores 5-7. However, after a short tutorial, accuracy in grading improved.18 Therefore, although the accurate measurement of the percentage of Gleason pattern 4 may not be practical, distinguishing Gleason score 3⫹4 from 4⫹3 is simpler, more reproducible, and more likely to be performed at routine pathologic examinations. CONCLUSIONS Gleason score 7 tumors are heterogeneous in their biologic behavior. The differences in prognosis in tumors with Gleason scores 3⫹4 and 4⫹3 at radical prostatectomy are significant. Although the assessment of the percentage of pattern 4 at radical prostatectomy is not likely to be reproducible, the distinction between Gleason score 3⫹4 and 4⫹3 should be easier for pathologists to perform.
UROLOGY 56 (5), 2000
REFERENCES 1. Epstein JI, Partin AW, Sauvageot J, et al: Prediction of progression following radical prostatectomy: a multivariate analysis of 721 men with long term follow-up. Am J Surg Pathol 20: 286 –292, 1996. 2. Tefilli MV, Gheiler EL, Tiguert R, et al: Should Gleason score 7 prostate cancer be considered a unique grade category? Urology 53: 372–377, 1999. 3. Epstein JI, Pound CR, Partin AW, et al: Disease progression following radical prostatectomy in men with Gleason score 7 tumor. J Urol 160: 97–101, 1998. 4. Pan C-C, Potter SR, Partin AW, et al: The prognostic significance of tertiary Gleason patterns of higher grade in radical prostatectomy specimens: a proposal to modify the Gleason grading system. Am J Surg Pathol 24: 563–569, 2000. 5. Walsh PC: Anatomic radical retropubic prostatectomy, in Walsh PC, Retik AB, Vaughan ED Jr, et al (Eds): Campbell’s Urology, 7th ed. Philadelphia, WB Saunders, 1998, vol 3, pp 2565–2588. 6. Bova GS, Fox WM, and Epstein JI: Methods of radical prostatectomy specimen processing: a novel technique for harvesting fresh prostate cancer tissue and review of processing techniques. Mod Pathol 6: 201–207, 1993. 7. Hall GS, Kramer CE, and Epstein JI: Evaluation of radical prostatectomy specimens: a comparative analysis of various sampling methods. Am J Surg Pathol 16: 315–324, 1992. 8. Oefelein MG, Smith ND, Grayhack JT, et al: Long-term results of radical retropubic prostatectomy in men with high grade carcinoma of the prostate. J Urol 158: 1460 –1465, 1997. 9. Sakr W, Tiguert R, Gheiler EL, et al: Gleason 7 score prostate cancer: a heterogeneous entity? Correlation with pathologic parameters and disease free survival (abstract). J Urol 161(suppl): 243, 1999. 10. Herman CM, Kattan MW, Scardino PT, et al: Predominant Gleason pattern is a significant predictor of disease progression in Gleason score 7 prostate cancer (abstract). Mod Pathol 12: 97A, 1999. 11. Bassily NH, Pienta KJ, and Wojno JH: The proportion of Gleason pattern 4 prostatic adenocarcinoma significantly determines risk of postoperative PSA recurrence (abstract). Mod Pathol 12: 89A, 1999. 12. Oefelein MG, Grayhack JT, and McVary KT: Survival after radical retropubic prostatectomy of men with clinically localized high grade carcinoma of the prostate. Cancer 76: 2535–2542, 1995. 13. Ohori M, Wheeler TM, Kattan MW, et al: Prognostic significance of positive surgical margins at radical prostatectomy specimens. J Urol 154: 1818 –1824, 1995. 14. Stamey TA, McNeal JE, Yemoto CM, et al: Biological determinants of cancer progression in men with prostate cancer. JAMA 281: 1395–1400, 1999. 15. McNeal JE, Villers AA, Redwine EA, et al: Histologic differentiation, cancer volume and pelvic lymph node metastasis in adenocarcinoma of the prostate. Cancer 66: 1225– 1233, 1990. 16. Allsbrook WC Jr, Mangold KA, Yang X, et al: The Gleason grading system. J Urol Pathol 10: 141–157, 1999. 17. Steinberg DM, Sauvageot J, Piantadosi S, et al: Correlation of prostate needle biopsy and radical prostatectomy Gleason grade in academic and community settings. Am J Surg Pathol 21: 566 –576, 1997. 18. Kronz JD, Silberman MA, Allsbrook WC Jr, et al: Use of a web-based tutorial to improve practicing pathologists’ Gleason grading of prostate cancer on needle biopsies (abstract). Mod Pathol 12: 100A, 1999.
827
PROGNOSTIC SIGNIFICANCE OF GLEASON SCORE 3⫹4 VERSUS GLEASON SCORE 4⫹3 TUMOR AT RADICAL PROSTATECTOMY THERESA Y. CHAN, ALAN W. PARTIN, PATRICK C. WALSH,
AND
JONATHAN I. EPSTEIN
ABSTRACT Objectives. To determine the clinical significance of Gleason score 3⫹4 versus 4⫹3 on radical prostatectomy. Methods. Of 2390 men who underwent radical prostatectomy by a single surgeon, 570 had Gleason score 7 tumors without lymph node metastasis, seminal vesicle invasion, or tertiary Gleason pattern 5. Patients were evaluated for biochemical recurrence (prostate-specific antigen progression) and distant metastases. Results. Eighty percent of patients had Gleason score 3⫹4, 20% had 4⫹3. The rate of established extraprostatic extension at radical prostatectomy for Gleason score 3⫹4 and 4⫹3 tumors was 38.2% and 52.7%, respectively (P ⫽ 0.008). With a mean follow-up of 4.6 years for men without progression, Gleason score 4⫹3 tumors had an increased risk of progression independent of stage and margin status (P ⬍0.0001). The 5-year actuarial risk of progression was 15% and 40% for Gleason score 3⫹4 and 4⫹3 tumors, respectively. The mean time to progression was 4.4 years for Gleason score 3⫹4 tumors and 3.2 years for Gleason score 4⫹3 tumors. We stratified the patients into four prognostic groups on the basis of organ-confined status, margin status, and Gleason score (3⫹4 versus 4⫹3). The 5-year actuarial risk of progression was 10%, 35%, 45%, and 61%, with 10-year progression rates of 29%, 42%, 69%, and 84%, for the four groups. 3.9% of patients with Gleason score 3⫹4 and 10.5% with Gleason score 4⫹3 tumors developed metastatic disease within a mean of 5.7 and 5.6 years, respectively. A Gleason score of 4⫹3 versus 3⫹4 was predictive of metastatic disease (P ⫽ 0.002) but not local recurrence. Conclusions. Gleason score 7 tumors are heterogeneous in their biologic behavior. The differences in prognosis for patients with Gleason scores 3⫹4 and 4⫹3 tumors at radical prostatectomy are significant. Although the assessment of the percentage of pattern 4 at radical prostatectomy is not likely to be reproducible, the distinction between Gleason score 3⫹4 and 4⫹3 should be easier for pathologists to perform. UROLOGY 56: 823–827, 2000. © 2000, Elsevier Science Inc.
T
he Gleason score assigned to prostate cancer in the radical prostatectomy specimen is strongly predictive of postoperative progression. Gleason scores 2-4 and 8-10 account for only a small percentage of the total number of patients who undergo radical prostatectomy, and the prognosis for these groups is dictated largely by the score alone.1 The differences in progression of tumors with This study was supported in part by a National Institutes of Health SPORE grant for prostate cancer. From the Departments of Urology and Pathology, Johns Hopkins University School of Medicine, James Buchanan Brady Urological Institute, Johns Hopkins Hospital, Baltimore, Maryland Reprint requests: Jonathan I. Epstein, M.D., Department of Pathology, Johns Hopkins Hospital, 600 North Wolfe Street, Baltimore, MD 21287 Submitted: May 8, 2000, accepted: June 16, 2000 © 2000, ELSEVIER SCIENCE INC. ALL RIGHTS RESERVED
Gleason score 5-6 have been shown to be based on the margin status and extraprostatic extension.1 Gleason score 7 tumors have been shown to behave significantly worse than Gleason score 5-6 tumors, but have a better prognosis than Gleason score 8-9 tumors.2 However, the prognosis of Gleason score 7 tumors is not uniformly poor. In an earlier study of Gleason score 7 tumors, differences in the risk of progression were found 8 years after surgery, depending on the pathologic stage and margin status.1 A follow-up study that included a larger group of men followed up for a longer period determined that more than 50% of men with Gleason score 7 tumors without extraprostatic extension and without positive margins appeared to be cured.3 In this study, we address the question of whether the dominant pattern in Gleason score 7 0090-4295/00/$20.00 PII S0090-4295(00)00753-6 823
TABLE I. Distribution of patients with Gleason score 3ⴙ4 and 4ⴙ3 tumors stratified by surgical margin and organ-confinement status
Organ confined Focal extraprostatic extension Negative margins Positive margins Established extraprostatic extension Negative margins Positive margins
Total (n ⴝ 570)
3ⴙ4 (n ⴝ 458)
4ⴙ3 (n ⴝ 112)
198 138 117 21 234 172 62
162 121 106 15 175 125 50
36 17 11 6 59 47 12
(34.7) (24.2) (20.5) (3.7) (41.0) (30.2) (10.9)
(35.4) (26.4) (23.1) (3.3) (38.2) (27.3) (10.9)
(32.1) (15.2) (9.8) (5.4) (52.7) (42.0) (10.7)
Numbers in parentheses are percentages.
tumors (score 3⫹4 versus 4⫹3) is an independent predictor of progression after radical prostatectomy. MATERIAL AND METHODS Between September 1982 and May 1998, 2390 consecutive men underwent radical prostatectomy by a single surgeon at Johns Hopkins Hospital. There were 608 patients (25.4%) with Gleason score 7 tumors without lymph node metastasis or seminal vesicle invasion. Patients with capsular incision and no follow-up were excluded. Patients with Gleason score 7 tumors with tertiary Gleason pattern 5 were also excluded, as we have demonstrated that this group has a worse prognosis.4 The remaining 570 patients (23.8%) formed the basis for this study. All men were evaluated preoperatively with digital rectal examination of the primary lesion and bone scan. None of the patients received neoadjuvant hormonal or radiation therapy. Radical prostatectomy was performed by an anatomic approach, as previously described.5 In general, patients were followed up with digital rectal examination and serum prostate-specific antigen (PSA) levels every 3 months for 1 year, twice yearly for the next 2 years, and yearly thereafter. Serum PSA levels were determined by the Hybritech immunoradiometric assay. Postoperative bone scans were performed when patients complained of bone pain or had detectable PSA levels. Patients were considered to have disease progression when they had elevated postoperative serum PSA levels of greater than 0.2 ng/mL, evidence of local recurrence, or radiologic evidence of distant metastases. Patients received hormonal or radiation therapy when progression occurred. The prostates were either serially sectioned and entirely embedded or processed according to a sampling method, which has been shown to result in accurate pathologic staging.6,7 Tumors were determined to be Gleason score 3⫹4 or 4⫹3. Tumors were staged as organ confined or as having focal or established extraprostatic extension. Focal extraprostatic extension refers to only a few neoplastic glands immediately exterior to the prostate in one to two sections; established extraprostatic extension denotes tumor with a greater degree of extraprostatic spread. Statistical analysis was performed using STATA Software (College Station, Tex). The endpoints measured as time to progression were calculated by the Kaplan-Meier method for estimation. The Wilcoxon-Gehan test was used to determine whether the differences between the Kaplan-Meier curves were statistically significant. The assessment of multiple variables to determine which were independent prognostic indicators of disease progression was performed by the Cox proportional hazards analysis. 824
RESULTS The average age of the 570 men at the time of surgery was 59 years (median 60, range 39 to 74). Gleason score 3⫹4 tumor was identified in 458 (80%) of the 570 men, with the remaining men having Gleason score 4⫹3 cancer. The overall mean serum PSA level was 8.6 ng/mL (median 6.9). The average PSA level was 8.4 ng/mL (median 6.8) in men with Gleason score 3⫹4 tumors and 9.8 ng/mL (median 8.0) in men with Gleason score 4⫹3 tumors (P ⫽ 0.04). STAGE AND MARGINS Table I depicts the subclassification of patients with Gleason score 3⫹4 and 4⫹3 on the basis of the presence and extent of extraprostatic spread and surgical margin status. The overall extent of disease was organ confined in 198 patients (34.7%), focal extraprostatic extension in 138 patients (24.2%), and established extraprostatic extension in 234 patients (41.1%). Tumors with Gleason score 4⫹3 had more advanced disease than those with Gleason score 3⫹4 (P ⫽ 0.008). For example, the rate of established extraprostatic extension in radical prostatectomy specimens with Gleason score 3⫹4 and 4⫹3 tumors was 38.2% and 52.7%, respectively. The surgical margins were positive in 83 patients (14.6%); the surgical margin status was not significantly different between patients with Gleason score 3⫹4 tumors and those with Gleason score 4⫹3 tumors. PROGRESSION The mean follow-up for men without progression was 4.6 years (median 4, range 1 to 15). The mean follow-up for men with progression was 4.0 years (median 3, range 1 to 10). Of the 456 men without progression, 369 were followed up for more than 2 years and 114 for more than 7 years. The Kaplan-Meier progression-free curves for patients with Gleason score 3⫹4 and Gleason score 4⫹3 tumors are illustrated in Figure 1. Univariate UROLOGY 56 (5), 2000
FIGURE 1. Actuarial risk of progression in men with Gleason score 3⫹4 and score 4⫹3 tumors without lymph node or seminal vesicle involvement.
analysis showed that Gleason score 3⫹4 versus 4⫹3 was significant in predicting progression (P ⬍0.00001). Gleason score 4⫹3 tumors had an increased risk of progression independent of stage and margin status (P ⬍0.0001). The multivariate analysis also showed that surgical margin status had a greater influence in predicting progression (P ⬍0.0001) than the extent of extraprostatic extension (P ⫽ 0.015). Preoperative PSA levels did not add to the multivariate model. The 5-year actuarial risk of progression was 15% and 40% and the 10year actuarial risk of progression was 37% and 56% for Gleason score 3⫹4 and 4⫹3 tumors, respectively. The mean time to progression was 4.4 years and 3.2 years for Gleason score 3⫹4 and 4⫹3 tumors, respectively. Patients were stratified into four prognostic groups on the basis of similar risks of progression (Fig. 2). The number of men in groups 1 to 4 was 429, 26, 97, and 18, respectively. For the four groups, the 5-year actuarial risk of progression was 10%, 35%, 45%, and 61%, with 10-year progression rates of 29%, 42%, 69%, and 84%, respectively. It is likely that the lack of differences between groups 2 and 3 and between groups 2 and 4 were due in part to the relatively small number of patients in groups 2 and 4 (Table II). Thirty patients developed metastatic disease, 21 patients had local recurrence, and 6 patients had both. The stage of disease in patients with distant metastases was organ confined in 16.7%, focal extraprostatic extension in 23.3%, and established extraprostatic extension in 60%. Metastatic disease developed in 3.9% of patients with Gleason score 3⫹4 tumors and 10.5% with Gleason score 4⫹3 tumors within a mean time of 5.7 and 5.6 years, respectively. Gleason score 4⫹3 tumors had an increased risk of metastatic disease (P ⫽ 0.002), independent of the surgical margin status and extent UROLOGY 56 (5), 2000
FIGURE 2. Actuarial risk of progression in men with Gleason score 3⫹4 and score 4⫹3 tumors without lymph node or seminal vesicle involvement stratified into four prognostic groups. oc ⫽ organ confined; epe ⫽ extraprostatic extension; fepe ⫽ focal extraprostatic extension; eepe ⫽ established extraprostatic extension; mar (⫺) ⫽ margins negative; mar (⫹) ⫽ margins positive.
TABLE II. Statistically significant differences in progression Group
P Value
1. Organ confined or EPE and negative margins and 3⫹4 vs. 2. FEPE and negative margins and 0.0003 4⫹3 or FEPE and positive margins and 3⫹4 3. EEPE and negative margins and ⬍0.00001 4⫹3 or EEPE and positive margins and 3⫹4 4. EPE and positive margins and ⬍0.00001 4⫹3 2. FEPE and negative margins and 4⫹3 or FEPE and positive margins and 3⫹4 vs. 3. EEPE and negative margins and ⬎0.05 4⫹3 or EEPE and positive margins (not significant) and 3⫹4 4. EPE and positive margins and 0.034 4⫹3 3. EEPE and negative margins and 4⫹3 or EEPE and positive margins and 3⫹4 vs. 4. EPE and positive margins and 4⫹3 ⬎0.05 (not significant) EPE ⫽ extraprostatic extension; FEPE ⫽ focal EPE; EEPE ⫽ established EPE.
of extraprostatic extension. A Gleason score 3⫹4 versus 4⫹3 was not predictive of local recurrence. COMMENT The most common grades of the index (dominant) tumor seen at radical prostatectomy are 825
Gleason scores 6 and 7.1 As opposed to Gleason score 6, which is typically composed of only one pattern, Gleason score 7 tumors contain patterns 3 and 4 in various proportions. Given the marked adverse affect of Gleason pattern 4, one would expect that whether a tumor is Gleason score 3⫹4 or 4⫹3 would influence prognosis. Several studies have addressed the question of Gleason score 3⫹4 versus 4⫹3 at radical prostatectomy with somewhat conflicting results. In a study of 116 men with high-grade carcinoma and a median follow-up of 7 years, the group from Northwestern University reported a subset of 77 patients with Gleason score 7 tumors with no significant survival advantage identified for Gleason pattern 3⫹4 compared with 4⫹3.8 The lack of statistically significant differences in their study may be due to the small group of patients and the inclusion of patients with positive lymph nodes and/or seminal vesicle invasion. At Wayne State University, 610 men with Gleason score 7 cancer were analyzed and the distinction between 3⫹4 versus 4⫹3 correlated with both stage and progression. However, the median follow-up was only 25.8 months, and the difference between the two grades was independently predictive only in men with serum PSA values less than 10 ng/mL.9 The group from Baylor reported on 853 men with a 24-month median follow-up and found that 3⫹4 versus 4⫹3 correlated with progression, yet was not independently statistically significant (personal communication).10 Another investigation from the University of Michigan in a study of 361 men with only a 12-month median follow-up for those without progression found that Gleason score 3⫹4 versus 4⫹3 predicted progression, yet it was not reported whether it was an independent prognosticator.11 The question remained, once other routinely assessed parameters such as pathologic stage and margins are factored in, is it important to know whether a tumor is Gleason score 3⫹4 versus 4⫹3? In our study, a significant difference in recurrence-free survival rates was found between patients with Gleason score 3⫹4 and score 4⫹3 tumors, independent of surgical margin status and extraprostatic extension. We were able to stratify men with Gleason score 7 tumors into four very different prognostic groups on the basis of organconfined status, margin status, and Gleason score 3⫹4 versus 4⫹3. The 5-year actuarial progressionfree rate for these four groups was 90%, 65%, 55%, and 39%. Patients with the best prognosis were those who had either Gleason 3⫹4 or 4⫹3 and organ-confined disease or who had extraprostatic extension of any degree with Gleason score 3⫹4 and negative surgical margins. In the next worse prognostic group, men had focal extraprostatic ex826
tension and only one adverse finding (ie, either 3⫹4 with positive margins or 4⫹3 with negative margins). The next worse prognostic group had established extraprostatic extension with again only one adverse finding in terms of grade and surgical margins. The patients with the worst prognosis were those who had focal or established extraprostatic extension and both adverse findings (ie, positive margins and Gleason score 4⫹3). Although the differences between some of these groups did not reach statistical significance, that lack most likely resulted from the smaller number of patients in some of the subgroups, since the prognostic differences among the groups were what one would have predicted. That is, the greater number of adverse findings in terms of the degree of extraprostatic extension, 4⫹3 versus 3⫹4, and margin positivity, the worse the prognosis. In our multivariate analysis, surgical margin status was more influential than the extent of extraprostatic extension in predicting progression after radical prostatectomy. Regardless of the extent of extraprostatic extension, patients with Gleason score 3⫹4 tumors and negative margins fared as well as those with organ-confined disease and better than patients with Gleason score 3⫹4 and positive margins. Likewise, regardless of the extent of extraprostatic extension, patients with Gleason score 4⫹3 tumors and negative margins had a better prognosis than patients with score 4⫹3 tumors and positive margins. Patients with Gleason score 4⫹3 tumors, if they had negative margins, even had progression rates comparable with patients with Gleason score 3⫹4 tumors and positive margins. It is unclear whether the adverse effect of positive margins relates to the intrinsic biology of disease or to the ability to achieve local control. The results in our study showing the significance of the extent of extraprostatic extension and surgical margins on progression differ from various previous reports. Oefelein et al.12 demonstrated that although Gleason score, extraprostatic extension, and margin status were all independent prognostic factors in predicting progression after radical prostatectomy in 238 patients, in a subset of 52 men with Gleason score 7 tumors, extraprostatic extension was not predictive of progression. In contrast to our study, their series included patients with positive lymph nodes and seminal vesicle invasion. Ohori et al.13 found that seminal vesicle invasion, extraprostatic extension, and Gleason score, but not margin status, correlated with progression. They reported that in a subset of 97 patients with Gleason score 7 or more with extraprostatic extension and negative lymph nodes and seminal vesicles, margin status did not correlate with progression.13 The lack of statistical significance for certain pathologic parameters correlating with UROLOGY 56 (5), 2000
progression seen in the above studies may relate to the lack of sufficient sample size for stratification into multiple subgroups. The group from Stanford has been a strong proponent of using the proportion of high-grade tumor (Gleason pattern 4 and 5) as the preferred method for grading prostate cancer.14,15 However, the percentage of pattern 4/5 is only very predictive for progression at the extremes (greater than 70% or less than 20% with pattern 4/5).14 It has not been demonstrated that classifying tumors based on their percentage of pattern 4/5 is more predictive than stratifying patients into Gleason score 3⫹4 versus 4⫹3. Furthermore, assessing the percentage of Gleason pattern 4 is often difficult and not likely to be performed routinely. Often patterns 4 and 3 are intimately admixed such that their relative percentages are not readily calculable. Additional difficulty in asking pathologists to derive a specific percentage of pattern 4/5 stems from studies demonstrating interobserver variability in grading tumors with Gleason scores 5-7.16 In one study of needle biopsy grading, of the 70 cases diagnosed as Gleason score 7 on the outside, on review at Johns Hopkins Hospital, 17 (24%) were diagnosed as having a Gleason score less than 7 and 4 (6%) as having a Gleason score greater than 7. Of 320 cases diagnosed on the outside as Gleason score less than 7 or Gleason score greater than 7, 53 (17%) were diagnosed as Gleason score 7 on review at our institution.17 Even among uropathology experts, there may be difficulty in differentiating between Gleason score 6 and 7. Accurate use of the Gleason grading system also depends on the pathologists’ experience and training. In a study using a webbased tutorial, the most difficult tumors for participants to grade were those with Gleason scores 5-7. However, after a short tutorial, accuracy in grading improved.18 Therefore, although the accurate measurement of the percentage of Gleason pattern 4 may not be practical, distinguishing Gleason score 3⫹4 from 4⫹3 is simpler, more reproducible, and more likely to be performed at routine pathologic examinations. CONCLUSIONS Gleason score 7 tumors are heterogeneous in their biologic behavior. The differences in prognosis in tumors with Gleason scores 3⫹4 and 4⫹3 at radical prostatectomy are significant. Although the assessment of the percentage of pattern 4 at radical prostatectomy is not likely to be reproducible, the distinction between Gleason score 3⫹4 and 4⫹3 should be easier for pathologists to perform.
UROLOGY 56 (5), 2000
REFERENCES 1. Epstein JI, Partin AW, Sauvageot J, et al: Prediction of progression following radical prostatectomy: a multivariate analysis of 721 men with long term follow-up. Am J Surg Pathol 20: 286 –292, 1996. 2. Tefilli MV, Gheiler EL, Tiguert R, et al: Should Gleason score 7 prostate cancer be considered a unique grade category? Urology 53: 372–377, 1999. 3. Epstein JI, Pound CR, Partin AW, et al: Disease progression following radical prostatectomy in men with Gleason score 7 tumor. J Urol 160: 97–101, 1998. 4. Pan C-C, Potter SR, Partin AW, et al: The prognostic significance of tertiary Gleason patterns of higher grade in radical prostatectomy specimens: a proposal to modify the Gleason grading system. Am J Surg Pathol 24: 563–569, 2000. 5. Walsh PC: Anatomic radical retropubic prostatectomy, in Walsh PC, Retik AB, Vaughan ED Jr, et al (Eds): Campbell’s Urology, 7th ed. Philadelphia, WB Saunders, 1998, vol 3, pp 2565–2588. 6. Bova GS, Fox WM, and Epstein JI: Methods of radical prostatectomy specimen processing: a novel technique for harvesting fresh prostate cancer tissue and review of processing techniques. Mod Pathol 6: 201–207, 1993. 7. Hall GS, Kramer CE, and Epstein JI: Evaluation of radical prostatectomy specimens: a comparative analysis of various sampling methods. Am J Surg Pathol 16: 315–324, 1992. 8. Oefelein MG, Smith ND, Grayhack JT, et al: Long-term results of radical retropubic prostatectomy in men with high grade carcinoma of the prostate. J Urol 158: 1460 –1465, 1997. 9. Sakr W, Tiguert R, Gheiler EL, et al: Gleason 7 score prostate cancer: a heterogeneous entity? Correlation with pathologic parameters and disease free survival (abstract). J Urol 161(suppl): 243, 1999. 10. Herman CM, Kattan MW, Scardino PT, et al: Predominant Gleason pattern is a significant predictor of disease progression in Gleason score 7 prostate cancer (abstract). Mod Pathol 12: 97A, 1999. 11. Bassily NH, Pienta KJ, and Wojno JH: The proportion of Gleason pattern 4 prostatic adenocarcinoma significantly determines risk of postoperative PSA recurrence (abstract). Mod Pathol 12: 89A, 1999. 12. Oefelein MG, Grayhack JT, and McVary KT: Survival after radical retropubic prostatectomy of men with clinically localized high grade carcinoma of the prostate. Cancer 76: 2535–2542, 1995. 13. Ohori M, Wheeler TM, Kattan MW, et al: Prognostic significance of positive surgical margins at radical prostatectomy specimens. J Urol 154: 1818 –1824, 1995. 14. Stamey TA, McNeal JE, Yemoto CM, et al: Biological determinants of cancer progression in men with prostate cancer. JAMA 281: 1395–1400, 1999. 15. McNeal JE, Villers AA, Redwine EA, et al: Histologic differentiation, cancer volume and pelvic lymph node metastasis in adenocarcinoma of the prostate. Cancer 66: 1225– 1233, 1990. 16. Allsbrook WC Jr, Mangold KA, Yang X, et al: The Gleason grading system. J Urol Pathol 10: 141–157, 1999. 17. Steinberg DM, Sauvageot J, Piantadosi S, et al: Correlation of prostate needle biopsy and radical prostatectomy Gleason grade in academic and community settings. Am J Surg Pathol 21: 566 –576, 1997. 18. Kronz JD, Silberman MA, Allsbrook WC Jr, et al: Use of a web-based tutorial to improve practicing pathologists’ Gleason grading of prostate cancer on needle biopsies (abstract). Mod Pathol 12: 100A, 1999.
827