- Email: [email protected]
LIMITATIONS OF BIOPSY GLEASON GRADE: IMPLICATIONS FOR COUNSELING PATIENTS WITH BIOPSY GLEASON SCORE 6 PROSTATE CANCER
0022-5347/04/1721-0098/0 THE JOURNAL OF UROLOGY® Copyright © 2004 by AMERICAN UROLOGICAL ASSOCIATION
Vol. 172, 98 –102, July 2004 Printed in U.S.A.
DOI: 10.1097/01.ju.0000132135.18093.d6
LIMITATIONS OF BIOPSY GLEASON GRADE: IMPLICATIONS FOR COUNSELING PATIENTS WITH BIOPSY GLEASON SCORE 6 PROSTATE CANCER PAUL D. SVED, PABLO GOMEZ, M. MANOHARAN, SANDY S. KIM
AND
MARK S. SOLOWAY*
From the Department of Urology, University of Miami School of Medicine, Miami, Florida
ABSTRACT
Purpose: We examined the implications of underestimating Gleason score by prostate biopsy in patients with biopsy Gleason 6 prostate cancer with respect to adverse pathological findings and biochemical recurrence after radical prostatectomy. Materials and Methods: We retrospectively reviewed clinical and pathological data on a cohort of 531 patients with Gleason score 6 on prostate biopsy who underwent radical retropubic prostatectomy between June 1992 and January 2002. Patients were excluded if they received neoadjuvant androgen deprivation. Concordance between biopsy and radical prostatectomy Gleason score was examined. A comparison was made with respect to final radical prostatectomy specimen pathology and the risk of biochemical recurrence between cases that remained Gleason 6 and those with a final grade of 7 or greater. Results: A total of 451 patients were included in the analysis. Mean followup was 55.1 months (range 12 to 123.4). Of the patients 184 (41%) had a Gleason score of 7 or greater after a review of the entire prostate, while 37 (8%) had a score of less than 6 and 230 remained with Gleason 6. Patients who were under graded were more likely to have extraprostatic extension (22% vs 4%, p ⬍0.01), seminal vesicle invasion (9% vs 2%, p ⬍0.01) and biochemical recurrence (10% vs 3%, p ⬍0.01) compared to those who remained with Gleason score 6. Conclusions: Gleason grade on needle biopsy is an inexact predictor of the final grade following radical prostatectomy. Patients with biopsy Gleason score 6 who are under graded are at significantly higher risk for adverse pathological features and biochemical recurrence than patients who remain with Gleason score 6 or less on final pathology findings. KEY WORDS: prostate, prostatic neoplasms, prostatectomy, biopsy, prognosis
Tumor grade on needle biopsy, serum prostate specific antigen (PSA) and findings on digital rectal examination are important components in planning treatment in patients with clinically localized prostate cancer. Multivariate models, nomograms and expert panel guidelines incorporating these factors have been widely used to advise patients of the risk of recurrence following surgery and radiation therapy.1 Therefore, accurate preoperative assessment of these variables is important to enable informed choice regarding treatment options. Several studies have highlighted the potential inaccuracy of one of these key variables, namely biopsy Gleason grade and sum.2– 6 Not surprisingly Gleason score following radical prostatectomy (RP) is a more accurate indication of patient prognosis. Tumor Gleason sum on biopsy is in agreement with the radical prostatectomy Gleason sum in as few as 30% of cases.6 The most common difference is the detection of a higher Gleason grade, eg Gleason grade 4 and, thus, a higher Gleason sum. Concentrating on the most common biopsy Gleason score of 6, 30% to 70% of cases are under graded with majority having a Gleason 7 tumor on analysis of the entire prostate.5, 6 There is little specific information regarding the clinical and histopathological characteristics, and the risk of recurrence following radical prostatectomy of Gleason score 6 prostate cancers that are under graded on pretreatment biopsy.
MATERIALS AND METHODS
We retrospectively reviewed the records of 1,177 consecutive patients who underwent radical retropubic prostatectomy performed by a single surgeon (MSS) between 1992 and 2002, and identified those with a pretreatment biopsy Gleason score of 6. Due to the inherent difficulties in pathological interpretation of biopsy specimens after androgen deprivation patients who received neoadjuvant androgen deprivation therapy were excluded from analysis. Preoperative assessment consisted of a history and physical examination (including digital rectal examination), PSA and prostate biopsy. Preoperative imaging, such as computerized tomography of the pelvis and bone scan, was performed depending on the referring physician. Radical retropubic prostatectomy was performed using modifications of the technique described by Walsh.7 One such modification relates to apical dissection of the prostate. Following ligation of the dorsal venous complex, it is divided and dissected off of the antero-apical aspect of the prostate using sharp dissection and diathermy. As the dorsal venous complex is dissected off the prostate, the prostate is gradually rolled back until the urethra is observed. Sharp dissection of periurethral fibers permits clear definition of the urethra prior to its division. The subsequent procedure is similar to that described by Walsh. Bilateral lymph node dissection was usually performed, although it was omitted more frequently in the last 3 years when PSA was less than 10 ng/ml. A nerve sparing technique and bladder neck preservation8 were performed in select patients. All biopsy specimens, including those obtained elsewhere, and all radical prostatectomy specimens were reviewed by pathologists at our institution. Step sections (2 to 3 mm) were made of the entire specimen.
Accepted for publication February 20, 2004. Supported by a grant from the Jackson Memorial Health Foundation and Mr. Harvey Chaplin. * Correspondence: Department of Urology, P. O. Box 016960 (M814), Miami, Florida 33101 (telephone: 305-243-6596; FAX: 305243-4653; e-mail: [email protected]). 98
99
LIMITATIONS OF BIOPSY GLEASON GRADE
Extraprostatic extension was defined as neoplastic cells in contact with periprostatic fat. A positive surgical margin was defined as any cancer cells in contact with ink. Gleason score was calculated by the sum of primary and secondary grades. Tumor volume in the prostatectomy specimen is expressed as a visual estimate of the percent of carcinoma (VEPC).9 Clinical and pathological data included patient age, PSA, clinical stage, biopsy Gleason score, radical prostatectomy Gleason score, VEPC, adjuvant therapy, extraprostatic extension, seminal vesicle involvement, surgical margin status and lymph node status. Postoperatively PSA values were obtained at 3 and 6-month intervals. Biochemical recurrence was indicated if postoperative PSA was 0.4 ng/ml or greater10 and time to recurrence was calculated. Statistical methods. Patients characteristics are presented as frequencies and means ⫾ SD. The chi-square and Student t tests were used to assess differences between groups with respect to clinical and pathological parameters, eg clinical stage, PSA, surgical margin status, seminal vesicle involvement and VEPC. For disease-free survival data we calculated survival function estimates based on time to recurrence using the Kaplan-Meier method. Univariate log rank tests were used to test the homogeneity of survival functions.
TABLE 2. Study patient characteristics Group 1
Group 2
RP Gleason 2–5 6 No. pts (%) 37 (8) 230 (51) Mean age ⫾ SD 60.3 ⫾ 8.2 58.5 ⫾ 7.2 PSA (ng/ml): Mean ⫾ SD 6.2 ⫾ 2.7 6.2 ⫾ 3.6 No. 0–10 (%) 35 (95) 210 (91) No. 10.1–20 (%) 2 (5) 17 (7) No. greater than 20 (%) 0 3 (1) No. clinical stage (%): T1 29 (78) 176 (76.5) T2 8 (22) 53 (23) T3 0 1 (0.4) No. extraprostatic extension (%) 2 (5) 10 (4) No. pos margin (%) 13 (35) 76 (33) No. seminal vesicle involvement (%) 0 5 (2) No. lymph node (%) 0 0 VEPC: Mean ⫾ SD 11.2 ⫾ 13.3 8.5 ⫾ 9.2 No. 0–9 (%) 20 (54) 139 (60) No. 9.1–20 (%) 8 (22) 58 (25) No. greater than 20 9 (24) 33 (14) No. adjuvant therapy (%) 0 3 (1) No. biochemical recurrence (%) 1 (3) 6 (3) All groups had Gleason 6 on biopsy.
Group 3 7–10 184 (41) 61.4 ⫾ 7.5 7.7 ⫾ 3.9 155 (84) 26 (14) 3 (2) 135 (73.4) 48 (26) 1 (0.5) 40 (22) 74 (40) 16 (9) 1 (0.5) 14.5 ⫾ 12.0 72 (39) 65 (35) 47 (26) 5 (3) 18 (10)
TABLE 3. Differences detected by chi-square and t tests by groups
RESULTS
Of the 1,177 patients who underwent radical prostatectomy during this time 531 had a biopsy Gleason score of 6. However, 80 patients who received neoadjuvant androgen deprivation were excluded. Thus, 451 men met inclusion criteria. Mean followup was 55.1 ⫾ 28.4 months (range 12 to 123.4, median 49). Patient age was 40 to 75 years (mean 62.3 ⫾ 7.4, median 63). Mean initial PSA was 8.3 ⫾ 7.2 ng/ml. There was concordance between the pretreatment biopsy Gleason score and radical prostatectomy Gleason score in 230 patients (51%) (group 2). The radical prostatectomy Gleason score was higher than the biopsy Gleason score in 184 patients (41%) (group 3) and lower in 37 (8%) (group 1). The majority of those who were under graded had Gleason score 7 on final pathological study (table 1). Table 2 lists the characteristics of patients in the 3 groups. Significant differences were identified between cases that remained Gleason 6 on the final pathological study and those that were up graded (groups 2 and 3) with regard to several variables (table 3). They were extraprostatic tumor extension (4% vs 22%, p ⬍0.01), seminal vesicle involvement (2% vs 9%, p ⬍0.01) and VEPC (p ⬍0.01). The incidence of positive surgical margins and lymph node metastases did not differ between the 2 groups. Mean preoperative PSA was higher in those with Gleason score greater than 6 in the radical prostatectomy specimen (p ⬍0.01) but there was no significant difference in clinical stage. The incidence of positive surgical margins, extraprostatic extension, seminal vesicle or lymph node involvement and VEPC did not differ between patients under graded at biopsy (group 1) and those in whom biopsy and specimen Gleason scores were the same (group 2). In each group the most common site of positive margins was the apex (table 4). Positive margin rates were consistently higher in cases associated with extraprostatic extension (pathological stages T3a and T3b). The overall biochemical recurrence rate was 25 of 451 cases (6%). The recurrence rate for under graded patients
TABLE 1. RP Gleason score in 451 patients with biopsy Gleason 6 Score
No. Pts (%)
5 6 7 8 9
37 (8) 230 (51) 161 (36) 18 (4) 5 (1)
p Value
PSA Clinical stage Extraprostatic extension Pos margin Seminal vesicle involvement Lymph node VEPC Adjuvant therapy Biochemical recurrence
Group 1 vs 2
Group 2 vs 3
0.97 0.9 0.77 0.8 0.37 Not applicable 0.26 0.48 0.97
⬍0.01 0.76 ⬍0.01 0.13 ⬍0.01 0.26 ⬍0.01 0.3 ⬍0.01
TABLE 4. Positive margin rate by location and pathological stage No./Total No. (%) Group 1 RP Gleason 2–5 Margin location:* Apex 4 (31) Anterior 2 (15) Posterior 1 (8) Posterolat 0 Bladder neck 0 Lat 0 Multiple 6 (46) Pathological stage: T2a, T2b 12/35 (34) T3a, T3b 1/2 (50) T3c 0 All groups had Gleason 6 on biopsy. * Unavailable for 1 group 3 patient.
Group 2 6
Group 3 7–10
28 (37) 10 (13) 9 (12) 4 (5) 5 (7) 2 (3) 18 (24)
18 (24) 4 (5) 9 (12) 7 (10) 4 (5) 2 (3) 29 (39)
68/215 (32) 6/10 (60) 2/5 (40)
47/148 (32) 15/40 (38) 11/16 (69)
was significantly higher than in those remaining with Gleason score 6 on the final pathological assessment (log rank 6.88, p ⫽ 0.014, fig. 1). There was no difference in the incidence of biochemical recurrence when comparing patients with a final Gleason score of 6 and those with a score of 5 or less (log rank 0.01, p ⫽ 0.94, fig. 2). The biochemical recurrence rate in those with positive margins and no extraprostatic extension (pT2a-b) was 0%, 2 of 68 (3%) and 6 of 51 (12%) in groups 1 to 3, respectively. In patients with pT3a-b disease and positive margins biochemical recurrence rates were higher at 0%, 1 of 8 (13%) and 8 of 23 (35%) in groups 1 to 3, respectively. DISCUSSION
In addition to patient age and comorbidity, clinicians use information on clinical stage, Gleason score and serum PSA
100
LIMITATIONS OF BIOPSY GLEASON GRADE
FIG. 1. PSA failure-free survival in patients with biopsy Gleason score 6 who were under graded vs accurately graded by biopsy.
FIG. 2. PSA failure-free survival in patients with biopsy Gleason score 6 who were over graded vs accurately graded by biopsy.
to form the basis for treatment planning in patients with newly diagnosed prostate cancer. For those with clinically localized prostate cancer who are candidates for radical prostatectomy or radiation therapy techniques may be modified according to tumor grade and stage to optimize treatment success and minimize side effects. In select patients with low grade tumors a strategy of watchful waiting may be appropriate.11 In the current study more than 40% of patients with Gleason 6 cancer on biopsy who underwent radical prostatectomy had a final Gleason score of 7 or greater. These results are consistent with those in other series (table 5). Patients with a final Gleason score of 7 or greater following a biopsy score of 6 are at significantly higher risk for advanced pathological findings, such as extracapsular extension and seminal vesicle invasion. The risk of biochemical recurrence was 3 times higher in patients up graded following radical prostatectomy (6% vs 18%). Steinberg et al identified possible reasons for the discrepancy between biopsy and radical prostatectomy Gleason scores.3 They were sampling error (biopsy misses an area of grade 4 or 5 cancer), biopsy samples a minor component of the cancer that is primarily Gleason score 6 or greater, and error in the pathologist interpretation of biopsy and/or radical prostatectomy specimens. Strategies aimed to improve agreement between biopsy and radical prostatectomy Gleason scores have been investigated. Intuitively the most obvious way to improve accuracy is to perform more biopsies. San Francisco et al reported improved concordance with extended (10 or more) biopsies.12 In those who underwent sextant biopsies the overall accuracy rate was 63% compared to 76% in those undergoing extended biopsies. However, using no fewer than 18 directed cores King and Long reported concordance in only 57% of cases.13 A further possible strategy is to eliminate interobserver variability. Even when the same pathologist reports biopsy and radical prostatectomy specimens agreement was only 60%.5, 9 Carlson et al reported a higher correlation rate (68%) between biopsy and radical prostatectomy Gleason score when all specimens were analyzed at a large, specialized uropathology laboratory.5 However, this is not always attainable in community practice. We observed improvement in the concordance between biopsy and final pathology Gleason score with time in patients with a biopsy Gleason 6 tumor. From 1992 to 1995 the concordance rate was 31% compared to 52% in 1996 to 2001. Several biopsy parameters in addition to Gleason score have been shown to predict the likelihood of adverse pathological findings and biochemical failure after radical prostatectomy. They are the maximal percent of a core involved with cancer, the percent of cores with cancer, the percent of biopsy tissue with cancer, and perineural invasion or extracapsular extension.14 Carlson et al found that neither the percent of cores with cancer nor the maximum percent of tumor on biopsy could predict a correlation between biopsy and radical prostatectomy Gleason scores. In the current study 90% of under graded tumors were Gleason 7. Several groups have emphasized the importance of differentiating between Gleason 6 and 7 prostate cancer.1, 5 Gleason 7 tumors are more likely to be associated with extraprostatic extension, seminal vesicle involvement and biochemical progression.5, 15 We found that Gleason 6 tumors up graded to 7 were 4 times more likely to extend beyond the
TABLE 5. Biopsy Gleason 6 correlated with RP grade References
No. Pts
% Exact Correlation
% Under Graded
% Over Graded
Cookson et al2 Steinberg et al3 Thickman et al4 Carlson et al5 Lattouf and Saad6 Present series
74 308 24 79 164 451
31 59 16 67 28 51
50 30 50 29 30 41
19 11 34 4 42 8
101
LIMITATIONS OF BIOPSY GLEASON GRADE
prostatic capsule and invade the seminal vesicle. This has important implications for patients considering treatment modalities such as brachytherapy. D’Amico et al identified Gleason score 6 as a crucial cutoff point for the patient response to brachytherapy.16 They found that the risk of biochemical relapse was 4 times greater in patients with biopsy Gleason score 7 compared with Gleason 6. Similar results have been reported previously. The American Brachytherapy Society recommends restricting brachytherapy as monotherapy to patients with Gleason score 2 to 6 due to the significant risk of extraprostatic extension and a poorer outcome in those with a higher grade tumor.17 For those with a higher grade tumor the society recommends combining brachytherapy with external beam radiation therapy. Similar recommendations were recently made by Scherr et al for the National Comprehensive Cancer Network.1 External beam radiation therapy alone may be suboptimal treatment in patients with clinically localized prostate cancer and Gleason score 7 or higher. In a retrospective review of 1,586 patients D’Amico et al reported significantly improved 5-year disease-free survival when 3-dimensional conformal radiation therapy was combined with 6 months of androgen suppression compared to radiation alone in these patients.18 These results imply that many patients with biopsy Gleason score 6 who actually have a higher grade tumor may benefit from the addition of androgen suppression. Men with clinically localized, small volume prostate cancer and a life expectancy of greater than 10 years may be candidates for expectant management and delayed therapy if disease progresses. Patients with a Gleason score of greater than 6 are not considered suitable candidates for expectant management due to the risk of nonorgan confined disease. In the series of Carter et al 31% of patients with biopsy Gleason score 6 who were treated expectantly had disease progression at a median followup of 23 months.11 Five of the 13 patients (35%) with disease progression who underwent radical prostatectomy had a tumor with Gleason score 7 or greater and, therefore, they were at higher risk for biochemical failure. The possibility of under grading by biopsy must be considered before offering expectant management to patients with biopsy Gleason score 6. The incidence of a positive surgical margin in this series was 36%. Patients who were under graded were not at higher risk for positive margins than those who remained with Gleason 6. Despite this somewhat high incidence the overall risk of biochemical recurrence was only 5.5%. We described in detail the location of positive margins (table 4). We believe that this incidence is related to the technique of apical dissection. The anterior capsule is opened, which may allow tumor cells to touch the ink, thereby, producing an iatrogenic positive margin in the anterior or apical prostate. As previously documented, patients with a positive margin located only in the anterior or apical prostate rarely have biochemical recurrence.19 Patients who remained with Gleason score 6 and who had a positive margin in the absence of extraprostatic extension had a low biochemical failure rate of 3%. As expected, extraprostatic extension was associated with a higher incidence of positive surgical margins in all groups. These patients also had a higher rate of biochemical recurrence than patients with a positive margin without extraprostatic extension. Also of note is the low risk of lymph node metastasis in our series. Only 1 of 451 patients with biopsy Gleason 6 had node metastases. Narayan et al proposed eliminating lymphadenectomy in patients with biopsy Gleason score 6 or less and a PSA of less than 10 ng/ml.20 Our findings support this recommendation. Even with a final Gleason score of 7 only 1% of patients had metastases to lymph nodes. Our study has limitations. As a retrospective analysis, it is subject to inherent selection bias. Moreover, because a large proportion of the patients treated at our institution are re-
ferred from a number of community and international centers, several urologists using different biopsy techniques performed prostate biopsies. For the same reasons several different pathology laboratories were involved in biopsy processing. However, instead of representing a limitation, this may broaden the relevance of the study, reflecting a range of community practices. CONCLUSIONS
Patients with clinically localized prostate cancer who have a Gleason 6 tumor on pretreatment biopsy have a high chance of being under graded. Patients who are under graded are at significantly higher risk for adverse pathological features and biochemical failure after radical prostatectomy. Those who are considering management options such as brachytherapy and watchful waiting must be advised of the implications of harboring a higher grade tumor. To date there is no consensus on how under grading may be predicted or avoided. REFERENCES
1. Scherr, D., Swindle, P. W., Scardino, P. T. and National Comprehensive Cancer Network: National Comprehensive Cancer Network guidelines for the management of prostate cancer. Urology, suppl., 61: 14, 2003 2. Cookson, M. S., Fleshner, N. E., Soloway, S. M. and Fair, W. R.: Correlation between Gleason score of needle biopsy and radical prostatectomy specimen: accuracy and clinical implications. J Urol, 157: 559, 1997 3. Steinberg, D. M., Sauvageot, J., Piantadosi, S. and Epstein, J. I.: Correlation of prostate cancer biopsy and radical prostatectomy Gleason grade in academic and community settings. Am J Surg Pathol, 157: 566, 1997 4. Thickman, G. D., Spears, W. C., Philpott, P. J. and Shapiro, H.: Effect of the number of core biopsies of the prostate on predicting Gleason score on prostate cancer. J Urol, 156: 110, 1996 5. Carlson, G. D., Calvanese, C. B., Kahane, H. and Epstein, J. I.: Accuracy of biopsy Gleason scores from a large uropathology laboratory: use of a diagnostic protocol to minimize observer variability. Urology, 51: 525, 1998 6. Lattouf, J. B. and Saad, F.: Gleason score on biopsy: is it reliable for predicting the final grade on pathology? BJU Int, 90: 694, 2002 7. Walsh, P. C.: Anatomic radical prostatectomy: evolution of the surgical technique. J Urol, 160: 2418, 1998 8. Soloway, M. S. and Neulander, E.: Bladder-neck preservation during radical retropubic prostatectomy. Semin Urol Oncol, 18: 51, 2000 9. Manoharan, M., Civantos, F., Kim, S. S., Gomez, P. and Soloway, M. S.: Visual estimate of percent of carcinoma predicts recurrence after radical prostatectomy. J Urol, 170: 1194, 2004 10. Amling, C. L., Bergstrahl, E. J., Blute, M. L., Sleezak, J. M. and Zincke, H.: Defining prostate specific antigen progression after radical prostatectomy: what is the most appropriate cut point? J Urol, 165: 1146, 2001 11. Carter, H. B., Walsh, P. C., Landis, P. and Epstein, J. I.: Expectant management of nonpalpable prostate cancer with curative intent: preliminary results. J Urol, 167: 1231, 2002 12. San Francisco, I. F., DeWolf, W. C., Rosen, S., Upton, M. and Olumi, A. F.: Extended prostate needle biopsy improves concordance of Gleason grading between prostate needle biopsy and radical prostatectomy. J Urol, 169: 136, 2003 13. King, C. R. and Long, J. P.: Prostate biopsy grading errors: a sampling problem? Int J Cancer, 90: 326, 2000 14. Freedland, S. J., Aronson, W. J., Csathy, G. S., Kane, C. J., Amling, C. L., Presti, J. C., Jr. et al: Comparison of percentage of total prostate needle biopsy tissue with cancer to percentage of cores with cancer for predicting PSA recurrence after radical prostatectomy: results from the SEARCH database. Urology, 61: 742, 2003 15. Epstein, J. I., Pizov, G. and Walsh, P. C.: Correlation of pathologic findings with progression after radical retropubic prostatectomy. Cancer, 71: 3582, 1993 16. D’Amico, A. V., Whittington, R., Malkowicz, S. B., Schultz, D.,
102
LIMITATIONS OF BIOPSY GLEASON GRADE
Blank, K., Broderick, G. A. et al: Biochemical outcome after radical prostatectomy, external beam radiation therapy, or interstitial radiation therapy for clinically localized prostate cancer. JAMA, 280: 969, 1998 17. Nag, S., Beyer, D., Friedland, J., Grimm, P. and Nath, R.: American brachytherapy society (ABS) recommendations for transperineal permanent brachytherapy of prostate cancer. Int J Radiat Oncol Biol Phys, 44: 789, 1999 18. D’Amico, A. V., Schultz, D., Loffredo, M., Dugal, R., Hurwitz, M., Kaplan, I. et al: Biochemical outcome following external beam
radiation therapy with or without androgen suppression therapy for clinically localized prostate cancer. JAMA, 284: 1280, 2000 19. Sofer, M., Hamilton-Nelson, K. L., Civantos, F. and Soloway, M. S.: Positive surgical margins after radical retropubic prostatectomy: the influence of site and number on progression. J Urol, 167: 2453, 2002 20. Narayan, P., Jajodia, P., Stein, R. and Tanagho, E. A.: A comparison of fine needle aspiration and core biopsy in diagnosis and preoperative grading of prostate cancer. J Urol, 141: 560, 1989
Vol. 172, 98 –102, July 2004 Printed in U.S.A.
DOI: 10.1097/01.ju.0000132135.18093.d6
LIMITATIONS OF BIOPSY GLEASON GRADE: IMPLICATIONS FOR COUNSELING PATIENTS WITH BIOPSY GLEASON SCORE 6 PROSTATE CANCER PAUL D. SVED, PABLO GOMEZ, M. MANOHARAN, SANDY S. KIM
AND
MARK S. SOLOWAY*
From the Department of Urology, University of Miami School of Medicine, Miami, Florida
ABSTRACT
Purpose: We examined the implications of underestimating Gleason score by prostate biopsy in patients with biopsy Gleason 6 prostate cancer with respect to adverse pathological findings and biochemical recurrence after radical prostatectomy. Materials and Methods: We retrospectively reviewed clinical and pathological data on a cohort of 531 patients with Gleason score 6 on prostate biopsy who underwent radical retropubic prostatectomy between June 1992 and January 2002. Patients were excluded if they received neoadjuvant androgen deprivation. Concordance between biopsy and radical prostatectomy Gleason score was examined. A comparison was made with respect to final radical prostatectomy specimen pathology and the risk of biochemical recurrence between cases that remained Gleason 6 and those with a final grade of 7 or greater. Results: A total of 451 patients were included in the analysis. Mean followup was 55.1 months (range 12 to 123.4). Of the patients 184 (41%) had a Gleason score of 7 or greater after a review of the entire prostate, while 37 (8%) had a score of less than 6 and 230 remained with Gleason 6. Patients who were under graded were more likely to have extraprostatic extension (22% vs 4%, p ⬍0.01), seminal vesicle invasion (9% vs 2%, p ⬍0.01) and biochemical recurrence (10% vs 3%, p ⬍0.01) compared to those who remained with Gleason score 6. Conclusions: Gleason grade on needle biopsy is an inexact predictor of the final grade following radical prostatectomy. Patients with biopsy Gleason score 6 who are under graded are at significantly higher risk for adverse pathological features and biochemical recurrence than patients who remain with Gleason score 6 or less on final pathology findings. KEY WORDS: prostate, prostatic neoplasms, prostatectomy, biopsy, prognosis
Tumor grade on needle biopsy, serum prostate specific antigen (PSA) and findings on digital rectal examination are important components in planning treatment in patients with clinically localized prostate cancer. Multivariate models, nomograms and expert panel guidelines incorporating these factors have been widely used to advise patients of the risk of recurrence following surgery and radiation therapy.1 Therefore, accurate preoperative assessment of these variables is important to enable informed choice regarding treatment options. Several studies have highlighted the potential inaccuracy of one of these key variables, namely biopsy Gleason grade and sum.2– 6 Not surprisingly Gleason score following radical prostatectomy (RP) is a more accurate indication of patient prognosis. Tumor Gleason sum on biopsy is in agreement with the radical prostatectomy Gleason sum in as few as 30% of cases.6 The most common difference is the detection of a higher Gleason grade, eg Gleason grade 4 and, thus, a higher Gleason sum. Concentrating on the most common biopsy Gleason score of 6, 30% to 70% of cases are under graded with majority having a Gleason 7 tumor on analysis of the entire prostate.5, 6 There is little specific information regarding the clinical and histopathological characteristics, and the risk of recurrence following radical prostatectomy of Gleason score 6 prostate cancers that are under graded on pretreatment biopsy.
MATERIALS AND METHODS
We retrospectively reviewed the records of 1,177 consecutive patients who underwent radical retropubic prostatectomy performed by a single surgeon (MSS) between 1992 and 2002, and identified those with a pretreatment biopsy Gleason score of 6. Due to the inherent difficulties in pathological interpretation of biopsy specimens after androgen deprivation patients who received neoadjuvant androgen deprivation therapy were excluded from analysis. Preoperative assessment consisted of a history and physical examination (including digital rectal examination), PSA and prostate biopsy. Preoperative imaging, such as computerized tomography of the pelvis and bone scan, was performed depending on the referring physician. Radical retropubic prostatectomy was performed using modifications of the technique described by Walsh.7 One such modification relates to apical dissection of the prostate. Following ligation of the dorsal venous complex, it is divided and dissected off of the antero-apical aspect of the prostate using sharp dissection and diathermy. As the dorsal venous complex is dissected off the prostate, the prostate is gradually rolled back until the urethra is observed. Sharp dissection of periurethral fibers permits clear definition of the urethra prior to its division. The subsequent procedure is similar to that described by Walsh. Bilateral lymph node dissection was usually performed, although it was omitted more frequently in the last 3 years when PSA was less than 10 ng/ml. A nerve sparing technique and bladder neck preservation8 were performed in select patients. All biopsy specimens, including those obtained elsewhere, and all radical prostatectomy specimens were reviewed by pathologists at our institution. Step sections (2 to 3 mm) were made of the entire specimen.
Accepted for publication February 20, 2004. Supported by a grant from the Jackson Memorial Health Foundation and Mr. Harvey Chaplin. * Correspondence: Department of Urology, P. O. Box 016960 (M814), Miami, Florida 33101 (telephone: 305-243-6596; FAX: 305243-4653; e-mail: [email protected]). 98
99
LIMITATIONS OF BIOPSY GLEASON GRADE
Extraprostatic extension was defined as neoplastic cells in contact with periprostatic fat. A positive surgical margin was defined as any cancer cells in contact with ink. Gleason score was calculated by the sum of primary and secondary grades. Tumor volume in the prostatectomy specimen is expressed as a visual estimate of the percent of carcinoma (VEPC).9 Clinical and pathological data included patient age, PSA, clinical stage, biopsy Gleason score, radical prostatectomy Gleason score, VEPC, adjuvant therapy, extraprostatic extension, seminal vesicle involvement, surgical margin status and lymph node status. Postoperatively PSA values were obtained at 3 and 6-month intervals. Biochemical recurrence was indicated if postoperative PSA was 0.4 ng/ml or greater10 and time to recurrence was calculated. Statistical methods. Patients characteristics are presented as frequencies and means ⫾ SD. The chi-square and Student t tests were used to assess differences between groups with respect to clinical and pathological parameters, eg clinical stage, PSA, surgical margin status, seminal vesicle involvement and VEPC. For disease-free survival data we calculated survival function estimates based on time to recurrence using the Kaplan-Meier method. Univariate log rank tests were used to test the homogeneity of survival functions.
TABLE 2. Study patient characteristics Group 1
Group 2
RP Gleason 2–5 6 No. pts (%) 37 (8) 230 (51) Mean age ⫾ SD 60.3 ⫾ 8.2 58.5 ⫾ 7.2 PSA (ng/ml): Mean ⫾ SD 6.2 ⫾ 2.7 6.2 ⫾ 3.6 No. 0–10 (%) 35 (95) 210 (91) No. 10.1–20 (%) 2 (5) 17 (7) No. greater than 20 (%) 0 3 (1) No. clinical stage (%): T1 29 (78) 176 (76.5) T2 8 (22) 53 (23) T3 0 1 (0.4) No. extraprostatic extension (%) 2 (5) 10 (4) No. pos margin (%) 13 (35) 76 (33) No. seminal vesicle involvement (%) 0 5 (2) No. lymph node (%) 0 0 VEPC: Mean ⫾ SD 11.2 ⫾ 13.3 8.5 ⫾ 9.2 No. 0–9 (%) 20 (54) 139 (60) No. 9.1–20 (%) 8 (22) 58 (25) No. greater than 20 9 (24) 33 (14) No. adjuvant therapy (%) 0 3 (1) No. biochemical recurrence (%) 1 (3) 6 (3) All groups had Gleason 6 on biopsy.
Group 3 7–10 184 (41) 61.4 ⫾ 7.5 7.7 ⫾ 3.9 155 (84) 26 (14) 3 (2) 135 (73.4) 48 (26) 1 (0.5) 40 (22) 74 (40) 16 (9) 1 (0.5) 14.5 ⫾ 12.0 72 (39) 65 (35) 47 (26) 5 (3) 18 (10)
TABLE 3. Differences detected by chi-square and t tests by groups
RESULTS
Of the 1,177 patients who underwent radical prostatectomy during this time 531 had a biopsy Gleason score of 6. However, 80 patients who received neoadjuvant androgen deprivation were excluded. Thus, 451 men met inclusion criteria. Mean followup was 55.1 ⫾ 28.4 months (range 12 to 123.4, median 49). Patient age was 40 to 75 years (mean 62.3 ⫾ 7.4, median 63). Mean initial PSA was 8.3 ⫾ 7.2 ng/ml. There was concordance between the pretreatment biopsy Gleason score and radical prostatectomy Gleason score in 230 patients (51%) (group 2). The radical prostatectomy Gleason score was higher than the biopsy Gleason score in 184 patients (41%) (group 3) and lower in 37 (8%) (group 1). The majority of those who were under graded had Gleason score 7 on final pathological study (table 1). Table 2 lists the characteristics of patients in the 3 groups. Significant differences were identified between cases that remained Gleason 6 on the final pathological study and those that were up graded (groups 2 and 3) with regard to several variables (table 3). They were extraprostatic tumor extension (4% vs 22%, p ⬍0.01), seminal vesicle involvement (2% vs 9%, p ⬍0.01) and VEPC (p ⬍0.01). The incidence of positive surgical margins and lymph node metastases did not differ between the 2 groups. Mean preoperative PSA was higher in those with Gleason score greater than 6 in the radical prostatectomy specimen (p ⬍0.01) but there was no significant difference in clinical stage. The incidence of positive surgical margins, extraprostatic extension, seminal vesicle or lymph node involvement and VEPC did not differ between patients under graded at biopsy (group 1) and those in whom biopsy and specimen Gleason scores were the same (group 2). In each group the most common site of positive margins was the apex (table 4). Positive margin rates were consistently higher in cases associated with extraprostatic extension (pathological stages T3a and T3b). The overall biochemical recurrence rate was 25 of 451 cases (6%). The recurrence rate for under graded patients
TABLE 1. RP Gleason score in 451 patients with biopsy Gleason 6 Score
No. Pts (%)
5 6 7 8 9
37 (8) 230 (51) 161 (36) 18 (4) 5 (1)
p Value
PSA Clinical stage Extraprostatic extension Pos margin Seminal vesicle involvement Lymph node VEPC Adjuvant therapy Biochemical recurrence
Group 1 vs 2
Group 2 vs 3
0.97 0.9 0.77 0.8 0.37 Not applicable 0.26 0.48 0.97
⬍0.01 0.76 ⬍0.01 0.13 ⬍0.01 0.26 ⬍0.01 0.3 ⬍0.01
TABLE 4. Positive margin rate by location and pathological stage No./Total No. (%) Group 1 RP Gleason 2–5 Margin location:* Apex 4 (31) Anterior 2 (15) Posterior 1 (8) Posterolat 0 Bladder neck 0 Lat 0 Multiple 6 (46) Pathological stage: T2a, T2b 12/35 (34) T3a, T3b 1/2 (50) T3c 0 All groups had Gleason 6 on biopsy. * Unavailable for 1 group 3 patient.
Group 2 6
Group 3 7–10
28 (37) 10 (13) 9 (12) 4 (5) 5 (7) 2 (3) 18 (24)
18 (24) 4 (5) 9 (12) 7 (10) 4 (5) 2 (3) 29 (39)
68/215 (32) 6/10 (60) 2/5 (40)
47/148 (32) 15/40 (38) 11/16 (69)
was significantly higher than in those remaining with Gleason score 6 on the final pathological assessment (log rank 6.88, p ⫽ 0.014, fig. 1). There was no difference in the incidence of biochemical recurrence when comparing patients with a final Gleason score of 6 and those with a score of 5 or less (log rank 0.01, p ⫽ 0.94, fig. 2). The biochemical recurrence rate in those with positive margins and no extraprostatic extension (pT2a-b) was 0%, 2 of 68 (3%) and 6 of 51 (12%) in groups 1 to 3, respectively. In patients with pT3a-b disease and positive margins biochemical recurrence rates were higher at 0%, 1 of 8 (13%) and 8 of 23 (35%) in groups 1 to 3, respectively. DISCUSSION
In addition to patient age and comorbidity, clinicians use information on clinical stage, Gleason score and serum PSA
100
LIMITATIONS OF BIOPSY GLEASON GRADE
FIG. 1. PSA failure-free survival in patients with biopsy Gleason score 6 who were under graded vs accurately graded by biopsy.
FIG. 2. PSA failure-free survival in patients with biopsy Gleason score 6 who were over graded vs accurately graded by biopsy.
to form the basis for treatment planning in patients with newly diagnosed prostate cancer. For those with clinically localized prostate cancer who are candidates for radical prostatectomy or radiation therapy techniques may be modified according to tumor grade and stage to optimize treatment success and minimize side effects. In select patients with low grade tumors a strategy of watchful waiting may be appropriate.11 In the current study more than 40% of patients with Gleason 6 cancer on biopsy who underwent radical prostatectomy had a final Gleason score of 7 or greater. These results are consistent with those in other series (table 5). Patients with a final Gleason score of 7 or greater following a biopsy score of 6 are at significantly higher risk for advanced pathological findings, such as extracapsular extension and seminal vesicle invasion. The risk of biochemical recurrence was 3 times higher in patients up graded following radical prostatectomy (6% vs 18%). Steinberg et al identified possible reasons for the discrepancy between biopsy and radical prostatectomy Gleason scores.3 They were sampling error (biopsy misses an area of grade 4 or 5 cancer), biopsy samples a minor component of the cancer that is primarily Gleason score 6 or greater, and error in the pathologist interpretation of biopsy and/or radical prostatectomy specimens. Strategies aimed to improve agreement between biopsy and radical prostatectomy Gleason scores have been investigated. Intuitively the most obvious way to improve accuracy is to perform more biopsies. San Francisco et al reported improved concordance with extended (10 or more) biopsies.12 In those who underwent sextant biopsies the overall accuracy rate was 63% compared to 76% in those undergoing extended biopsies. However, using no fewer than 18 directed cores King and Long reported concordance in only 57% of cases.13 A further possible strategy is to eliminate interobserver variability. Even when the same pathologist reports biopsy and radical prostatectomy specimens agreement was only 60%.5, 9 Carlson et al reported a higher correlation rate (68%) between biopsy and radical prostatectomy Gleason score when all specimens were analyzed at a large, specialized uropathology laboratory.5 However, this is not always attainable in community practice. We observed improvement in the concordance between biopsy and final pathology Gleason score with time in patients with a biopsy Gleason 6 tumor. From 1992 to 1995 the concordance rate was 31% compared to 52% in 1996 to 2001. Several biopsy parameters in addition to Gleason score have been shown to predict the likelihood of adverse pathological findings and biochemical failure after radical prostatectomy. They are the maximal percent of a core involved with cancer, the percent of cores with cancer, the percent of biopsy tissue with cancer, and perineural invasion or extracapsular extension.14 Carlson et al found that neither the percent of cores with cancer nor the maximum percent of tumor on biopsy could predict a correlation between biopsy and radical prostatectomy Gleason scores. In the current study 90% of under graded tumors were Gleason 7. Several groups have emphasized the importance of differentiating between Gleason 6 and 7 prostate cancer.1, 5 Gleason 7 tumors are more likely to be associated with extraprostatic extension, seminal vesicle involvement and biochemical progression.5, 15 We found that Gleason 6 tumors up graded to 7 were 4 times more likely to extend beyond the
TABLE 5. Biopsy Gleason 6 correlated with RP grade References
No. Pts
% Exact Correlation
% Under Graded
% Over Graded
Cookson et al2 Steinberg et al3 Thickman et al4 Carlson et al5 Lattouf and Saad6 Present series
74 308 24 79 164 451
31 59 16 67 28 51
50 30 50 29 30 41
19 11 34 4 42 8
101
LIMITATIONS OF BIOPSY GLEASON GRADE
prostatic capsule and invade the seminal vesicle. This has important implications for patients considering treatment modalities such as brachytherapy. D’Amico et al identified Gleason score 6 as a crucial cutoff point for the patient response to brachytherapy.16 They found that the risk of biochemical relapse was 4 times greater in patients with biopsy Gleason score 7 compared with Gleason 6. Similar results have been reported previously. The American Brachytherapy Society recommends restricting brachytherapy as monotherapy to patients with Gleason score 2 to 6 due to the significant risk of extraprostatic extension and a poorer outcome in those with a higher grade tumor.17 For those with a higher grade tumor the society recommends combining brachytherapy with external beam radiation therapy. Similar recommendations were recently made by Scherr et al for the National Comprehensive Cancer Network.1 External beam radiation therapy alone may be suboptimal treatment in patients with clinically localized prostate cancer and Gleason score 7 or higher. In a retrospective review of 1,586 patients D’Amico et al reported significantly improved 5-year disease-free survival when 3-dimensional conformal radiation therapy was combined with 6 months of androgen suppression compared to radiation alone in these patients.18 These results imply that many patients with biopsy Gleason score 6 who actually have a higher grade tumor may benefit from the addition of androgen suppression. Men with clinically localized, small volume prostate cancer and a life expectancy of greater than 10 years may be candidates for expectant management and delayed therapy if disease progresses. Patients with a Gleason score of greater than 6 are not considered suitable candidates for expectant management due to the risk of nonorgan confined disease. In the series of Carter et al 31% of patients with biopsy Gleason score 6 who were treated expectantly had disease progression at a median followup of 23 months.11 Five of the 13 patients (35%) with disease progression who underwent radical prostatectomy had a tumor with Gleason score 7 or greater and, therefore, they were at higher risk for biochemical failure. The possibility of under grading by biopsy must be considered before offering expectant management to patients with biopsy Gleason score 6. The incidence of a positive surgical margin in this series was 36%. Patients who were under graded were not at higher risk for positive margins than those who remained with Gleason 6. Despite this somewhat high incidence the overall risk of biochemical recurrence was only 5.5%. We described in detail the location of positive margins (table 4). We believe that this incidence is related to the technique of apical dissection. The anterior capsule is opened, which may allow tumor cells to touch the ink, thereby, producing an iatrogenic positive margin in the anterior or apical prostate. As previously documented, patients with a positive margin located only in the anterior or apical prostate rarely have biochemical recurrence.19 Patients who remained with Gleason score 6 and who had a positive margin in the absence of extraprostatic extension had a low biochemical failure rate of 3%. As expected, extraprostatic extension was associated with a higher incidence of positive surgical margins in all groups. These patients also had a higher rate of biochemical recurrence than patients with a positive margin without extraprostatic extension. Also of note is the low risk of lymph node metastasis in our series. Only 1 of 451 patients with biopsy Gleason 6 had node metastases. Narayan et al proposed eliminating lymphadenectomy in patients with biopsy Gleason score 6 or less and a PSA of less than 10 ng/ml.20 Our findings support this recommendation. Even with a final Gleason score of 7 only 1% of patients had metastases to lymph nodes. Our study has limitations. As a retrospective analysis, it is subject to inherent selection bias. Moreover, because a large proportion of the patients treated at our institution are re-
ferred from a number of community and international centers, several urologists using different biopsy techniques performed prostate biopsies. For the same reasons several different pathology laboratories were involved in biopsy processing. However, instead of representing a limitation, this may broaden the relevance of the study, reflecting a range of community practices. CONCLUSIONS
Patients with clinically localized prostate cancer who have a Gleason 6 tumor on pretreatment biopsy have a high chance of being under graded. Patients who are under graded are at significantly higher risk for adverse pathological features and biochemical failure after radical prostatectomy. Those who are considering management options such as brachytherapy and watchful waiting must be advised of the implications of harboring a higher grade tumor. To date there is no consensus on how under grading may be predicted or avoided. REFERENCES
1. Scherr, D., Swindle, P. W., Scardino, P. T. and National Comprehensive Cancer Network: National Comprehensive Cancer Network guidelines for the management of prostate cancer. Urology, suppl., 61: 14, 2003 2. Cookson, M. S., Fleshner, N. E., Soloway, S. M. and Fair, W. R.: Correlation between Gleason score of needle biopsy and radical prostatectomy specimen: accuracy and clinical implications. J Urol, 157: 559, 1997 3. Steinberg, D. M., Sauvageot, J., Piantadosi, S. and Epstein, J. I.: Correlation of prostate cancer biopsy and radical prostatectomy Gleason grade in academic and community settings. Am J Surg Pathol, 157: 566, 1997 4. Thickman, G. D., Spears, W. C., Philpott, P. J. and Shapiro, H.: Effect of the number of core biopsies of the prostate on predicting Gleason score on prostate cancer. J Urol, 156: 110, 1996 5. Carlson, G. D., Calvanese, C. B., Kahane, H. and Epstein, J. I.: Accuracy of biopsy Gleason scores from a large uropathology laboratory: use of a diagnostic protocol to minimize observer variability. Urology, 51: 525, 1998 6. Lattouf, J. B. and Saad, F.: Gleason score on biopsy: is it reliable for predicting the final grade on pathology? BJU Int, 90: 694, 2002 7. Walsh, P. C.: Anatomic radical prostatectomy: evolution of the surgical technique. J Urol, 160: 2418, 1998 8. Soloway, M. S. and Neulander, E.: Bladder-neck preservation during radical retropubic prostatectomy. Semin Urol Oncol, 18: 51, 2000 9. Manoharan, M., Civantos, F., Kim, S. S., Gomez, P. and Soloway, M. S.: Visual estimate of percent of carcinoma predicts recurrence after radical prostatectomy. J Urol, 170: 1194, 2004 10. Amling, C. L., Bergstrahl, E. J., Blute, M. L., Sleezak, J. M. and Zincke, H.: Defining prostate specific antigen progression after radical prostatectomy: what is the most appropriate cut point? J Urol, 165: 1146, 2001 11. Carter, H. B., Walsh, P. C., Landis, P. and Epstein, J. I.: Expectant management of nonpalpable prostate cancer with curative intent: preliminary results. J Urol, 167: 1231, 2002 12. San Francisco, I. F., DeWolf, W. C., Rosen, S., Upton, M. and Olumi, A. F.: Extended prostate needle biopsy improves concordance of Gleason grading between prostate needle biopsy and radical prostatectomy. J Urol, 169: 136, 2003 13. King, C. R. and Long, J. P.: Prostate biopsy grading errors: a sampling problem? Int J Cancer, 90: 326, 2000 14. Freedland, S. J., Aronson, W. J., Csathy, G. S., Kane, C. J., Amling, C. L., Presti, J. C., Jr. et al: Comparison of percentage of total prostate needle biopsy tissue with cancer to percentage of cores with cancer for predicting PSA recurrence after radical prostatectomy: results from the SEARCH database. Urology, 61: 742, 2003 15. Epstein, J. I., Pizov, G. and Walsh, P. C.: Correlation of pathologic findings with progression after radical retropubic prostatectomy. Cancer, 71: 3582, 1993 16. D’Amico, A. V., Whittington, R., Malkowicz, S. B., Schultz, D.,
102
LIMITATIONS OF BIOPSY GLEASON GRADE
Blank, K., Broderick, G. A. et al: Biochemical outcome after radical prostatectomy, external beam radiation therapy, or interstitial radiation therapy for clinically localized prostate cancer. JAMA, 280: 969, 1998 17. Nag, S., Beyer, D., Friedland, J., Grimm, P. and Nath, R.: American brachytherapy society (ABS) recommendations for transperineal permanent brachytherapy of prostate cancer. Int J Radiat Oncol Biol Phys, 44: 789, 1999 18. D’Amico, A. V., Schultz, D., Loffredo, M., Dugal, R., Hurwitz, M., Kaplan, I. et al: Biochemical outcome following external beam
radiation therapy with or without androgen suppression therapy for clinically localized prostate cancer. JAMA, 284: 1280, 2000 19. Sofer, M., Hamilton-Nelson, K. L., Civantos, F. and Soloway, M. S.: Positive surgical margins after radical retropubic prostatectomy: the influence of site and number on progression. J Urol, 167: 2453, 2002 20. Narayan, P., Jajodia, P., Stein, R. and Tanagho, E. A.: A comparison of fine needle aspiration and core biopsy in diagnosis and preoperative grading of prostate cancer. J Urol, 141: 560, 1989