LONG-TERM BIOCHEMICAL DISEASE-FREE AND CANCER-SPECIFIC SURVIVAL FOLLOWING ANATOMIC RADICAL RETROPUBIC PROSTATECTOMY

~~

~~~~

~

~~~~

~~~

0094-0143/01 $15.00

RADICAL PROSTATECTOMY

+ -00

LONG-TERM BIOCHEMICAL DISEASE-FREE AND CANCERSPECIFIC SURVIVAL FOLLOWING ANATOMIC RADICAL RETROPUBIC PROSTATECTOMY The 15-Year Johns Hopkins Experience Misop Han, MD, Alan W. Partin, MD, PhD, Charles R. Pound, MD, Jonathan I. Epstein, MD, and Patrick C. Walsh, MD

The management of prostate cancer has undergone dramatic changes in the past 2 decades. The early detection and staging of prostate cancer have been facilitated with improved methods, including serum prostatespecific antigen (PSA) testing. In addition, as a result of numerous discoveries, surgeons now can perform an anatomic dissection of the prostate, reducing perioperative morbidity during and after radical retropubic prostatectomy (RRP).132l 7 Previously, the authors reported their postoperative experience with PSA values following anatomic RRP in 955 and 1623 men with localized di~ease.~,ll These studies demonstrated that an undetectable PSA level can serve as the gold standard for monitoring the tumor-free status following radical prostatectomy; however, determination of the true efficacy of the treatment modality for prostate cancer requires a large cohort of patients with long-term (15-year)

follow-up to observe cancer-specific mortality.,, Herein, the authors report progressionfree, metastasis-free, and cancer-specific survival in 2404 men undergoing anatomic RRP for clinically localized prostate cancer. A 15year cancer-specific survival analysis was performed on this cohort of men. As shown in previous reports, advancing clinical and pathologic variables, alone and in combination, adversely affect patient survival following RRP. Dividing men with Gleason score 7 disease into subgroups according to the predominance of Gleason pattern in the surgical specimen enhances the stratification of actuarial biochemical progression-free survival analysis. MATERIALS AND METHODS

Patient Population and Inclusion Criteria Between April 1982 and February 1999, 2494 men with clinically localized adenocarci-

Funding provided by SPORE grant CA-58326.

From the James Buchanan Brady Urological Institute, Departments of Urology (MH, AWP, JIE, PCW) and Pathology (JIE), The Johns Hopkins Medical Institutions, Baltimore, Maryland; and Department of Urology, University of Tennessee College of Medicine, Memphis, Tennessee (CW)

UROLOGIC CLINICS OF NORTH AMERICA ~~

~

~

~

-

~

_

_

VOLUME 28 NUMBER 3 AUGUST 2001

_

_

_

~

555

556

HANetal

noma of the prostate (Tl, T2, and T3a) underwent staging pelvic lymphadenectomy and anatomic RRP by a single urologist. The average age of these men at the time of surgery was 58.2 plus or minus 6.6 years (range, 33 to 76 years). Of these men, 2404 men (96.4%) were included in the current analysis to evaluate the ability of RRP as a single therapy to cure localized prostate cancer. This analysis includes 15,138 man-years of follow-up. Table 1 summarizes the disease characteristics of the 2404 men according to the tumor, lymph node, and metastasis (TNM)clinical staging system, the preoperative PSA level, the postoperative Gleason score, and the final pathologic stage. Eighty-three percent of the men (1996 of 2404) had Gleason score 6 or higher tumors. Of the 2494 men, 90 (3.6%) were excluded Table 1. CLINICAL STAGE, PREOPERATIVE PSA LEVEL, GLEASON SCORE, AND PATHOLOGIC STAGE IN 2404 MEN UNDERGOING ANATOMIC RADICAL RETROPUBIC PROSTATECTOMY Variable

TNM stage Tla Tlb Tlc T2a T2b T2c T3 Total Serum PSA (ng/mL) 04.0 4.1-10.0 10.1-20.0 >20.0

Total Gleason score 2-4 5 6 7 8-10 Total Pathologic stage Organ confined EPE + , GS<7, SM EPE +, GS<7, SM + EPE +, GS27, SM EPE + ,GS27, SM + SV+, (LN-) LN + Total

Number of Men

Percent of Total

59 118 861 816 390 105 55 2404

3 5 36 34 16 4 2 100

605 1067 351 100 2123

29 50 17 5

100

58 350 1082 754 160 2404

2 15 45 31 7 100

1224 360 111 326 135 113 135

51 15 5 14 6 5 6 100

2404

PSA = prostate-specific antigen; TNM = tumor, node, metastasis; EPE = extraprostatic extension; GS = Gleason score; SM = surgical margins; (-) = negative; ( + ) = positive; SV+, (LN - ) = involvement of seminal vesicles, negative lymph nodes; LN + = micrometastases to pelvic lymph nodes.

for various reasons from the entire study group. Fifteen men (0.6%) were excluded for insufficient follow-up information; 30 men (1.2%) received immediate adjuvant radiotherapy based on pathologic features; 15 men (0.6%) received immediate adjuvant hormonal therapy; three men (0.1%) received preoperative radiotherapy and 17 men (0.7%) neoadjuvant hormonal therapy; nine men (0.4%) were excluded for clinical stage DO/D2 diseases; and one man died in the immediate postoperative period. Of the 30 men excluded owing to immediate adjuvant radiotherapy, 25 received this therapy because of a positive surgical margin at final pathology examination. Pathologically, three men had organ-confined diseases, whereas 16 men had extraprostatic extension only. Seven men had positive seminal vesicle involvement without pelvic lymph node metastases, and four men had micrometastases to pelvic lymph nodes. In terms of follow-up, 7 of 30 men had undetectable PSA levels at 5 or more years following surgery without additional therapy. Twenty men had evidence of recurrence, and 14 of 20 men received hormonal therapy. Nine men had detectable PSA levels as the only evidence of tumor recurrence. Distant metastases developed in 11 men, and 5 of the 11men died of prostate cancer. Of the 15 men excluded owing to immediate adjuvant hormonal therapy, none had organ-confined disease. Five men had involvement of the seminal vesicles without lymph node involvement, and 10 men had micrometastases to pelvic lymph nodes. In terms of follow-up, only 2 of 15 men had undetectable PSA levels at 3 or more years after surgery without additional therapy. Distant metastases developed in nine men, and eight of nine men died of prostate cancer. Preoperative and Postoperative Evaluation and Pathologic Characteristics

The patients were staged according to the 1992 American Joint Committee on Cancer (AJCC) staging guidelines,' including digital rectal examination by a single surgeon and routine serum PSA studies (Hybritech Tandem-R and EC Beckman Coulter, San Diego, CA and TOSOH, Tosoh Medics, San Francisco, CA). A serum enzymatic acid phosphatase level was measured (thymolphthalein

557

LONG-TERM SURVIVAL FOLLOWING ANATOMIC RADICAL RETROPUBIC PROSTATECTOMY

monophosphate assay), and a nuclear bone scan was performed on all patients until recent years; for the past 5 years, bone scans were mandated only on men with PSA levels greater than 10 ng/mL. Pathologic diagnosis of prostate cancer was based on an examination of prostate tissue obtained from digitally or transrectal ultrasonography-guided prostate biopsy, transurethral resection of the prostate, or fine-needle aspiration. The Gleason scoring system6 was used for histologic grading of the needle biopsy and prostatectomy specimens. In addition, pathologic examination of the surgical specimen was performed to categorize men with Gleason score 7 tumors to Gleason 3 predominant (3 4) or Gleason 4 predominant (4 3) subgroups. Pathologic analysis was performed as described in previous report^.^, Postoperative follow-up consisted of digital rectal examinations and serum PSA measurements at quarterly intervals for the first year, semiannually for the second year, and yearly thereafter. Bone scans were performed at the time of initial biochemical recurrence and on a yearly basis thereafter, unless performed earlier for symptoms suggestive of distant metastatic disease. No patients included in this analysis received neoadjuvant therapy (chemotherapy or hormonal therapy). No patients received immediate adjuvant hormonal or radiotherapy based on pathologic features; therefore, adjuvant therapy had no impact on the time to biochemical progression in this analysis. The minimum follow-up was 1 year, with a mean follow-up of 6.3 plus or minus 4.2 years (range, 1-17 years). A total of 621 men (26%) were followed up for at least 10 years postoperatively. The length of followup for the 2404 men is listed in Table 2. Actuarial analysis was performed using the Kaplan-Meier method comparing freedom from biochemical recurrence after RRP (PSA > 0.2 ng/mL). The Wilcoxon-Gehan test was used to determine statistical differences between actuarial curves. All statistical analyses were performed using the Intercooled Stata 6.0 statistics and graphics data management system (Stata Corporation, College Station, TX).

+

+

RESULTS Overall Progression

Of the 2404 men, 412 (17%) experienced recurrence of the disease after RRP. Two hun-

Table 2. YEARS OF FOLLOW-UP AVAILABLE FOR 2404 MEN AFTER ANATOMIC RADICAL RETROPUBIC PROSTATECTOMY Years of FOIIOW-U~

Number of Patients

Percent of Total

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Total

2404 2093 1846 1612 1396 1186 1038 888 746 621 479 343 227 145 78 29 7 2404

100.0 87.0 77.0 67.0 58.0 49.0 43.0 37.0 31.0 26.0 20.0 14.0 9.0 6.0 3.0 1.0 0.3 100.0

dred thirty-four men (9.7%) had an isolated detectable PSA (> 0.2 ng/mL) as the only evidence of recurrence. Forty men (1.7%) recurred locally without evidence of distant metastases, and 138 men (5.8%) had distant metastases with or without evidence of local recurrence. No patient had a local or distant recurrence without an elevation of PSA at the time of progression. Kaplan-Meier analysis showed overall actuarial 5-, lo-, and 15-year freedom from any type of recurrence of 84%, 74%, and 66%, respectively. Overall actuarial 5-, lo-, and 15-year freedom from isolated biochemical progression were 92%, 85%, and 79%, respectively. Kaplan-Meier analysis showed that overall actuarial 5-, lo-, and 15year local recurrence-free rates were 99%, 969'0, and 94% respectively, whereas distant recurrence-free rates for the same period were 9670, go%, and 82%, respectively. The 5-, lo-, and 15-year actuarial probabilities of remaining free of a PSA, local without distant, or distant with or without local recurrence are demonstrated in Figure 1 and Table 3.

Clinical Stage and Progression

Figure 2 illustrates the Kaplan-Meier survival analysis of the likelihood of having an undetectable PSA after RRP based on TNM clinical stage according to the 1992 AJCC criteria. No patient with Tla disease experi-

558

HANetal

L Local

4

CA

- l.O0; P4

B 0

0.75

2

Q

7)

3

Distant PSA only

Ail recurrence

0.50-

%I

0

20

5

0.25-

2 0I

I

I

I

I

Figure 1. Kaplan-Meier actuarial progression-free likelihood for overall progression, isolated prostate-specific antigen (PSA) elevation only, local recurrence, and distant progression ? local progression for 2404 men who underwent anatomic radical retropubic prostatectomy.

enced an elevation of PSA postoperatively. The likelihood of freedom from biochemical recurrence for patients with Tlb versus Tlc disease was similar ( P = 0.73). The number of patients with Tlc disease with more than 8 years of follow-up was much smaller than the number of patients with other TNM stages. The likelihood of freedom from biochemical recurrence for patients with Tlc versus T2a tumor or for patients with T2a versus T2b tumor after RRP was significantly different (P = 0.01 and < 0.01, respectively); however, stage T2b and T2c tumors behaved similarly ( P = 0.25) in the analysis.

previous studies, these patients were divided into four groups according to the preoperative PSA values: PSA less than 4 ng/mL, PSA of 4.1 to 10 ng/mL, PSA of 10.1 to 20 ng/mL, and PSA greater than 20 ng/mL. Table 4 and Figure 3 demonstrate the actuarial progression-free likelihood based on preoperative PSA value. The likelihood of freedom from biochemical recurrence for patients in different PSA groups was significantly different (P < 0.01).

Preoperative Serum Prostate-Specific Antigen Level and Progression

Patients were divided into five groups based on Gleason scores for the RRP specimens: Gleason score 2 to 4, 5, 6, 7, and 8 to 10. There was no statistically significant difference in the actuarial progression-free

Of the 2404 men, 2123 (88%)had available preoperative PSA levels. As in the authors’

Gleason Score and Progression

Table 3. ACTUARIAL PROGRESSION-FREE RATE FOR OVERALL PROGRESSION, ISOLATED PROSTATESPECIFIC ANTIGEN ELEVATION ONLY, LOCAL RECURRENCE, AND DISTANT PROGRESSION WITH OR WITHOUT LOCAL PROGRESSION IN 2404 MEN AT 5,10, AND 15 YEARS AFTER ANATOMIC RADICAL RETROPUBIC PROSTATECTOMY Actuarial Recurrence-Free Likelihood (95% Cl) Recurrence

Number of Men

5 Years

10 Years

15 Years

Overall Isolated PSA only Local Distant + local

412/2404 234/2404 40/2404 138/2404

84 (83-86) 92 (90-93) 99 (99-100) 96 (95-97)

74 (71-76) 85 (83-87) 96 (95-97) 90 (88-92)

66 (61-70) 79 (75-82) 94 (92-96) 82 (77-87)

PSA = prostate-specificantigen; CI = confidence interval

LONG-TERM SURVIVAL FOLLOWING ANATOMIC RADICAL RETROPUBIC PROSTATECTOMY

559

1.oo

3n

$

-

0.75

b)

3

T2c T2b

0.50

rcl

0 .8

8

4

4

0.25

0

0

5

10 Years Postoperative

15

20

Figure 2. Kaplan-Meier actuarial likelihood of prostate-specific antigen (PSA) recurrence by clinical stage. (Dafa from 1992 AJCC staging)

Table 4. ACTUARIAL 5-, lo-, AND 15-YEAR RECURRENCE-FREE RATES FOLLOWING ANATOMIC RADICAL RETROPUBIC PROSTATECTOMY IN RELATION TO CLINICAL STAGE, PREOPERATIVE PROSTATE-SPECIFIC ANTIGEN LEVEL, AND GLEASON SCORE Actuarial Percentage (95% CI) Variable

TN M Tla Tlb Tlc T2a T2b T2c T3a Serum PSA (ng/mL) 0-4.0 4.1-10.0 10.1-20.0 >20.0 Postoperative Gleason score 2-4 5 6 7 (All) 7 (3+4) 7 (4+3) 8-10 Pathologic stage Organ confined EPE + , GS<7, SM EPE , GS<7, SM + EPE , GS27, SM EPE +, GS27, SM + SV+, (LN-) LN +

+ +

5 Year

10 Year

15 Year

100 90 (8S95) 91 (88-93) 86 (8-8) 75 (70-79) 71 (61-79) 60 (45-72)

100 85 (76-91) 76 (48-90) 75 (71-79) 62 (56-68) 57 (45-68) 49 (34-63)

100 75 (58-86) 76+(4&90) 66 (59-72) 50 (41-58) 57 (45-68) NA

94 (92-96) 89 (86-91) 73 (68-78) 60 (49-69)

91 (87-93) 79 (74-83) 57 (48-64) 48 (36-59)

67 (34-86) 75 (69-80) 54*(44-63) 48*(36-59)

100 98 (9699) 95 (93-97) 73 (6%76) 81 (77%) 53 (44-61) 44 (36-52)

100 94 (90-96) 88 (83-92) 54 (48-59) 60 (53-67) 33 (22-43) 29 (22-37)

100 86 (78-92) 73 (59-82) 48 (41-56) 59 (51-65) 33 (22-43) 15 (5-28)

97 (95-98) 97 (9498) 89 (80-94) 80 (75-85) 58 (4946) 48 (38-58) 26 (19-35)

93 (90-95) 93 (89-96) 73 (61-82) 61 (52-68) 42 (32-52) 30 (19-41) 10 (5-18)

84 (77-90) 84 (70-92) 58 (41-71) 59 (50-67) 33 (23-44) 17 (5-35) 0

'14-Year data. PSA = prostate-specific antigen; CI = confidence interval; TNM = tumor, node, metastasis; EPE = extraprostatic extension; GS = Gleason score; SM = surgical margin; ( - ) = negative; (+) = positive; SV+, (LN-) = involvement of seminal vesicles, negative lymph nodes; LN + = micrometastases to pelvic lymph nodes; NA = not available.

560

HANetal

PSA (ndmL)

c

-

0.75

4.1 10.0

Y' .liL

Y

0

-

Y

5

10.1 <4-020.0

0.50

>20.0

0

0

10

5

20

15

Years Postoperative Figure 3. Kaplan-Meier actuarial likelihood of prostate-specific antigen (PSA) recurrence by preoperative serum PSA levels (ng/mL).

probabilities for patients with Gleason score 2 to 4 disease versus patients with Gleason score 5 disease (P = 0.07); however, the actuarial progression-free likelihood based on Gleason score was statistically significant for patients with other Gleason scores ( P < 0.01). Table 4 and Figure 4 show the Kaplan-Meier analysis of Gleason score with regard to progression-free probability. When patients with Gleason score 7 disease

were subdivided into two groups according to the predominance of Gleason pattern 3 or 4, most (75%) had Gleason 3 + 4 tumors, whereas the rest had Gleason 4 + 3 tumors. The actuarial progression-free probabilities for men with Gleason score 6 versus 3 + 4 tumors and for patients with Gleason 3 + 4 versus 4 + 3 tumors were significantly different ( P < 0.001); however, the actuarial progression-free probabilities for men with Glea-

Gleason Score

8-10

0

5

10

,

1

15

20

Years Postoperative Figure 4. Kaplan-Meier actuarial likelihood of prostate-specific antigen (PSA) recurrence by Gleason score.

LONG-TERM SURVIVAL FOLLOWING ANATOMIC RADICAL RETROPUBIC PROSTATECTOMY

son score 4 + 3 versus 8 to 10 tumors were similar ( P = 0.08). The results are shown in Table 4 and Figure 5. Pathologic Stage, Gleason Score, and Surgical Margins in Relation to Progression

Epstein and co-workers have shown that grouping patients according to the combination of pathologic stage, Gleason score, and surgical margin status provides the best stratification of actuarial progression-free probability." l1 Figure 6 and Table 4 demonstrate the actuarial progression-free probability based on the combination of pathologic stage, Gleason score, and surgical margin status. Organconfined tumors and tumors with extraprostatic extension, a Gleason score of 6 or less, and a negative surgical margin behaved almost identically ( P = 0.89). The actuarial recurrence-free probabilities were similar for men who had extraprostatic extension with a positive surgical margin and a Gleason score of at least 7 and for men with tumor invasion of the seminal vesicles ( P = 0.17). Comparison with Other Reported Series

Several RRP series have reported on the follow-up of men with respect to PSA recurrence?, R, 7-12, Is The demographics and clini-

561

cal stages in these series are summarized in Table 5. The authors' current study of 2404 men comprises the largest series of men undergoing anatomic RRP with the longest follow-up. The proportion of men with T1 tumors was 43%, reflecting the increasing proportion of men with Tlc disease as a result of early detection of prostate cancer using PSA testing. The actuarial PSA progression-free survival, the actuarial cancer-specific survival, and the actuarial metastasis-free survival probabilities from several RRP series are summarized in Table 6. The actuarial 5-year PSA progression-free rates range from 69% (UCLA) to 84% (current series). The actuarial 10-year PSA progression-free rates range from 47% (UCLA) to 74% (current series). Only the Mayo Clinic and the authors' series report 15year PSA progression-free rates of 40% and 66%, respectively. The actuarial 15-year metastasis-free survival rates from the Mayo clinic series and the authors' current series are 76% and 82%, respectively, whereas the actuarial 15-year cancer-specificsurvival rates for the Mayo Clinic series and the current series are 82% and 90%, respectively. The authors' data confirm that higher preoperative PSA levels adversely affect the outcome following radical prostatectomy. Recently, Stamey and co-w~rkers'~ reported that only 15.8% of men with peripheral zone cancers and preoperative PSA greater than 15 Gleason Score

6 (3+4) .........

........." (4+3) I 8-10

0

5

10

15

20

Years Postoperative Figure 5. Kaplan-Meier actuarial likelihood of prostate-specific antigen (PSA) recurrence by Gleason score and subdivisions of Gleason score 7 diseases (3 + 4 versus 4 + 3).

562

HANetal

1.00-

$ r&

oc

.!? 0.75-

D

2*u

4

5

LL,\

0.50-

EPE+,L 7,SM-

%

0

-3

0 0

4

0.25-

i i , 7-

sv+ LN+

0 1

I

0

5

I

I0 Years Postoperative

I

I

15

20

Figure 6. Kaplan-Meier actuarial likelihood of prostate-specific antigen (PSA) recurrence by a combination of pathologic stage, Gleason score, and surgical margin status in 2404 men. OC = organ-confined; EPE+,< 7, SM- = extraprostatic extension, Gleason score less than 7, and negative surgical margins; EPE+, 2 7, SM+ = extraprostatic extension, Gleason score greater than or equal to 7, and positive surgical margins; SV + = positive seminal vesicles with negative lymph nodes; LN + = positive lymph nodes.

ng/mL were biochemically free of disease with 3 or more years of follow-up. In the authors' series, patients were not stratified pathologically according to the zone of cancer involvement. To overcome this limitation, disease progression was studied in the 120 men with palpable disease (T2a to T3a) who had preoperative PSA levels greater than 15 ng/ mL (mean, 26.2; range, 15-151). The mean

follow-up for these 120 men was 7.6 plus or minus 3.4 years (range, 1-14). The KaplanMeier progression-free analysis for the 120 men is illustrated in Figure 7. The overall actuarial PSA-free survival rates at 5 and 10 years for the men were 56% and 40%, respectively; therefore, radical prostatedomy was a reasonable treatment option for selected men with a preoperative PSA level greater than 15 ng/mL.

Table 5. DEMOGRAPHICS OF PATIENTS IN SEVERAL RADICAL RETROPUBIC PROSTATECTOMY SERIES

Institution

Johns Hopkins' Johns Hopkins" Johns Hopkins'O Washington University* Washington University3 Washington University12 Baylor College8 Baylor College7 UCLAlS The Mayo CliniP Current study

Number of Men

Years

955 1623 1997 925 1778' 1620 500 996 601 3170$ 2404

1982-1991 1982-1995 1982-1997 1983-1993 1983-1997 1988-1998 19831993 1983-1997 1972-1992 1966-1991 1982-1999

Mean Age in Years (range)

Percent Tumor Stage Mean Months, Follow-up (range)

59.4k 6 (34-76) 53 (12-120) 59& 6 (34-76) 60.4(12-156) 5926 (34-76) 64(6-180) 63k7 (41-79) 28 (0-123) 48 635 7 (38-79) 62& 7 (38-79) 28 (Tlc),63 (T2a),53 (T2b) 36 (1-114) 63 (4S79) 37t (1-168) 63 (38-81) 34t (12-237) 64(44-80) 60 65 ? 4 (31-81) 75.6(12-204) 58t6 (33-76)

11

12

17 31 38 21 38 39 22

83 66 60 79 60 61 78 70

24

27 7 43

-

73

93 55

*Four percent of patients received postoperative adjuvant radiotherapy. tMedian. STwenty-six percent of patients received postoperative adjuvant radiation, hormonal therapy, or both. Modified f m m Pound CR, Partin AW, Epstein JI, et a1 Prostatespecific antigen after anatomic radical retropubic prostatectorny: Patterns of recurrence and cancer control. Urol Clin North Am 24:395-406, 1997.

LONG-TERM SURVIVAL FOLLOWING ANATOMIC RADICAL RETROPUBIC PROSTATECTOMY

563

Table 6. PROGRESSION PATTERNS IN SEVERAL RADICAL RETROPUBIC PROSTATECTOMY SERIES Actuarial PSA Progression-Free Survival

Actuarial Metastasis-Free Survival

Actuarial Cancer-Specific Survival

Institution

5 Year

10 Year

15Year

5Year

10 Year

15Year

5Year

10Year

15Year

Johns Hopkins'", " Washington Universityz, Baylor College"

80 78 76 69 70 84

68 65 73 47 52 74

NA NA NA NA 40 66

NA NA NA NA NA 96

87 NA NA NA 82 90

82 NA NA NA 76 82

99 NA NA 98 NA 99

94 97* NA 94 90 96

91 NA NA NA 82 90

UCLAli The Mayo Clinic'* Current study '7-Year data. NA = not available; PSA

=

prostatespecific antigen.

DISCUSSION

In this analysis, the authors report the 15year follow-up results for a large cohort of men who underwent anatomic RRP for localized prostate cancer. A gradual change in the patient population with early detection of prostate cancer is well demonstrated in the analysis. The proportion of men with T1 disease increased from 17% in the authors' earlier analysis9to 43%in the current series. This trend is also evident in the RRP series from other centers of excellence in the United States as demonstrated in Table 5. Because surgery benefits patients with early stage disease the most, the number of men seeking and receiving definitive therapy is increased in return. The dramatic increase in the number of men with prostate cancer is explained by the early diagnosis of prostate cancer us-

ing the serum PSA test, which prompts biopsy of these previously unsuspected (nonpalpable) cancers. Zincke and co-workers's previously reported their 15-year follow-up results in 3170 men who underwent radical prostatectomy from 1966 to 1991. The mean follow-up was 5 years. The proportion of men with T1 disease was only 7%, reflecting the small number of men with early stage disease. The 5-, lo-, and 15-year overall actuarial PSA-free survival rates were YO%, 52%, and 42%, respectively. The 10- and 15-year cancer-specific survival rates were 90% and 82%, respectively. The authors' data demonstrated that the actuarial PSA-free survival rates at 5, 10, and 15 years were 84%, 74%, and 66%, respectively. The 10- and 15-year actuarial cancer-specific survival rates were 96% and go%, respectively.

1 . 0 o120 j y

c

2 3 6s

-8 L

0.75-

\

0.50-

4

0

3

25

20

0.25-

244 0

5 0

10

5

15

Years Postoperative Figure 7. Kaplan-Meier actuarial likelihood of prostate-specific antigen (PSA) recurrence for the 120 men who had preoperative PSA levels of at least 15 ng/ rnL and stage T2a to T3a diseases.

564

HANetal

The differences between the two series can be explained by several factors. First, the proportion of men with early stage disease (Tl) in the Mayo Clinic series was significantly smaller in comparison with the authors’ current series because the Mayo Clinic series included men who underwent surgery before the initiation of serum PSA testing. The authors’ series also includes approximately 500 men who underwent surgery before the advent of PSA testing. It is likely that the outcome for these men is worse than the outcome for men who underwent surgery since 1989. This difference must be recognized when the researcher attempts to compare the outcome in the current series with the results of surgery or other treatments administered to patients treated in the PSA era. Second, the difference in pathologic evaluation (underestimation of positive surgical margins or preoperative Gleason score) may explain the difference in progression rates. Differences in surgical technique may explain the differences in biochemical recurrence-free recurrence rates. Anatomic RRP was not developed until the early 1980s, and PSA was not used widely in postoperative monitoring for tumor recurrence until the late 1980s. Consistent with other contemporary radical prostatectomy series? 3, 8, 11, the authors’ data demonstrate that advancing Gleason score results in an adverse outcome following surgery. The actuarial progression-free survival rates were significantly different for men with Gleason score 5 or higher disease. In addition, with a simple modification of stratifying Gleason score 7 disease into 3 + 4 and 4 + 3 tumors according to the predominant Gleason pattern, a statistically significant difference in recurrence-free survival rate is shown for this group of patients with Gleason 7 disease after radical prostatectomy. Men with a predominance of Gleason pattern 4 disease (4 + 3) and men with Gleason score 8 to 10 disease had similar actuarial progression-free survival rates. The subpopulation of tumors with a predominance of Gleason patterns 4 or 5 tends to behave similarly following radical prostatectomy. SUMMARY

In a large series of 2404 men with a mean follow-up of 6.3 plus or minus 4.2 years (range, 1-17) after anatomic RRP for clinically localized prostate cancer, 412 men (17%) have

recurred. A detectable PSA was the only evidence of recurrence in 9.7%, whereas 1.7% and 5.8% had local recurrence and distant metastasis, respectively. The overall actuarial 5-, lo-, and 15-year recurrence-free survival rates for these men were 84%, 74%, and 66%, respectively. As demonstrated in the authors’ previous reports, the actuarial likelihood of a postoperative recurrence increased with advancing clinical stage, Gleason score, preoperative PSA level, and pathologic stage. Subdivision of men with Gleason 7 tumors resulted in better stratification. There was a similar actuarial likelihood of postoperative recurrence for men with Gleason 4 + 3 and Gleason score 8 to 10 disease. The actuarial rate of recurrence of tumor for men with Gleason 3 + 4 disease was statistically different from the rate for men with Gleason score 6 or Gleason 4 + 3 disease. The overall actuarial metastasis-free survival rates at 5, 10, and 15 years were 96%‘ go%, and 82%, respectively. The overall actuarial cancer-specific survival rates at 5, 10, and 15 years were 99%, 96%, and go%, respectively. This study provides long-term outcome of patients with clinically localized cancer who underwent RRP between 1982 and 1999. Recognizing that this long-term study includes many patients with more advanced disease diagnosed before the PSA era, caution must be exercised in comparing these results with the outcomes for cohorts of patients treated since 1989. Anatomic RRP is an effective way to manage clinically localized prostate cancer. Excellent long-term results can be obtained with RRP for early stage disease. The proportion of men with early stage prostate cancer will continue to increase with wide use of serum PSA testing and digital rectal examination.

References 1. Beahrs OH, Henson DE, Hutter RVP, et al: American Joint Committee on Cancer Staging Manual. Philadelphia, JB Lippincott, 1992 2. Catalona WJ, Smith DS: 5-Year tumor recurrence rates after anatomical radical retropubic prostatectomy for prostate cancer [see comments]. J Urol152(5 Pt 2):1837-1842, 1994 3. Catalona WJ, Smith Ds: Cancer recurrence and survival rates after anatomic radical retropubic prostatectomy for prostate cancer: Intermediate-term results. J Urol 160(6 pt 2):2428-2434, 1998 4. Epstein JI, Partin AW, Sauvageot J, et al: Prediction

LONG-TERM SURVIVAL FOLLOWING ANATOMIC RADICAL RETROPUBIC PROSTATECTOMY

5. 6.

of progression following radical prostatectomy: A multivariate analysis of 721 men with long-term follow-up. Am J Surg Pathol 20286292, 1996 Epstein JI, Pizov G, Walsh PC: Correlation of pathologic findings with progression after radical retropubic prostatectomy. Cancer 71:3582-3593, 1993 Gleason D: Histologic grading and clinical staging of carcinoma of the prostate. In Tannenbaum M (ed): Urologic Pathologv. Philadelphia, Lea & Febiger, 1977, pp 171-197 Kattan MW, Wheeler TM. Scardino PT:Postouerative nomogram for disease recurrence after radical prostatectomy for prostate cancer. J Clin Oncol 1714991507, 1999 Ohori M, Goad JR, Wheeler TM, et a1 Can radical prostatectomy alter the progression of poorly differentiated prostate cancer [see comments]? J Urol 152(5 pt 2):1843-1849,1994 Partin AW, Pound CR, Clemens JQ, et al: Serum PSA after anatomic radical prostatectomy: The Johns Hopkins experience after 10 years. Urol Clin North Am 20:71>725, 1993 Pound CR, Partin AW, Eisenberger MA, et a 1 Natural histow of progression after PSA elevation following radical p k t a t e c t o m y [see comments]. JAMA 281~1591-1597,1999 Pound CR, Partin AW, Epstein JI, et al: Prostatespecific antigen after anatomic radical retropubic prostatectomy: Patterns of recurrence and cancer control. Urol C h North Am 24395406,1997 "I

7.

8.

9.

10.

11.

.,

565

12. Ramos CG, Carvalhal GF, Smith DS, et a1 Clinical and pathological characteristics, and recurrence rates of stage Tlc versus T2a or T2b prostate cancer. J Urol 161:1525-1529, 1999 13. Reiner WG, Walsh PC: An anatomical approach to the surgical management of the dorsal vein and Santorini's plexus during radical retropubic surgery. J Urol121:198-200, 1979 14. Stamey TA, Yemoto CM, McNeal JE,et al: Prostate cancer is highly predictable: A prognostic equation based on all morphological variables in radical prostatectomy specimens [in process citation]. J Urol 163:1155-1160, 2000 15. Trapasso JG, deKemion JB, Smith RB, et a1 The incidence and significance of detectable levels of serum prostate specific antigen after radical prostatectomy [see comments]. J Urol 152(5 pt 2):1821-1825, 1994 16. Walsh PC: The natural history of localized prostate cancer: A guide to therapy. In Walsh PC, Retik AB, Vaughn ED, et a1 (eds): Campbell's Urology, ed 7. Philadelphia, WB Sunders, 1998, pp 2539-2546 17. Walsh PC, Donker PJ: Impotence following radical prostatectomy: Insight into etiology and prevention. J Urol 128:492497, 1982 18. Zincke H, Oesterling JE, Blute ML, et al: Long-term (15 years) results after radical prostatectomy for clinically localized (stage T2c or lower) prostate cancer [see comments]. J Urol 152(5 pt 2):1850-1857,1994

Address reprint requests to Alan W. Partin, MD, PhD Marburg 205A Department of Urology Johns Hopkins Hospital 600 North Wolfe Street Baltimore, MD 21287-2101 e-mail: apartin@+mi.edu