- Email: [email protected]
Role of prostate-specific antigen velocity in prediction of final pathologic stage in men with localized prostate cancer
ELSEVIER
ROLE OF PROSTATE-SPECIFIC ANTIGEN VELOCITY IN PREDICTION OF FINAL PATHOLOGIC STAGE IN MEN WITH LOCALIZED PROSTATE CANCER RALF THIEL, JAY D. PEARSON, JONATHAN PATRICK C. WALSH, AND H. BALLENTINE
I. EPSTEIN, CARTER
ABSTRACT Objectives. Prostate-specific antigen (PSA) velocity prior to radical retropubic prostatectomy was evaluated to determine if men with a faster rate of rise in PSA have locally more extensive prostate cancer. Methods. Of 368 men who underwent radical retropubic prostatectomy, 82 had two to seven PSA measurements between 1.3 and 6.7 years before prostate biopsy for evaluation of PSA velocity. PSA velocity and the pretreatment parameters of PSA, Gleason score, and T stage were evaluated as predictive parameters of pathologic stage. Results. In men with pathologically organ-confined disease, PSA velocity was 1.12 ng/mUyr; in non-organconfined cases, it was 1.88 ng/mUyr. There was a statistically significant relationship between a Gleason score of 7 and above and the pathologic extent of disease. There was no statistically significant relationship between T stage, the PSA at diagnosis, and PSA velocity with respect to final pathologic stage. Conclusions. PSA velocity is not a strong predictor of pathologic stage in men with localized prostate cancer. UROLOGY 49: 7 16-720, 1997. 0 1997, Elsevier Science Inc. All rights reserved.
P
rostate-specific antigen (PSA) has become one of the most useful tumor markers in oncology; it is now widely used for the early detection of prostate cancer, as well as in staging and monitoring men with the disease. The pretreatment PSA level provides additional information, when used together with T stage and pretreatment Gleason score, for assessing the extent of prostate cancer prior to recommending treatment.’ Despite progress in the pretreatment prediction of disease extent, contemporary radical prostatectomy series reveal that 40% to 60% of men have extraprostatic disease at the time of surgery.2,3 The long-term risk
From the Departments of Urology and Pathology, The Johns Hopkins University School of Medicine, James Buchanan Brady Urological Institute, The Johns Hopkins Medical Institutions, Baltimore, Maryland Dr. Thiel is currently at the Urological Clinic, HeinrichHeine-University, Diisseldorf, Germany. Dr. Pearson is currently at the Department of Epidemiology, Merck Research Laboratories, Blue Bell, Pennsylvania. Reprint requests: H. Ballentine Carter, M.D., Brady Urological Institute, TheJohns Hopkins Medical Institutions, 600 North Wolfe
Street,
Baltimore,
MD
21287-2101
Submitted: September 23,1996, accepted (with revisions): November 30, 1996 0 1997,
7
16
ALL
RIGHTS
ELSEVIER SCIENCE RESERVED
INC.
of residual disease is increased in the setting of extraprostatic disease. Thus, improvements in the pretreatment prediction of disease extent are needed to provide patients with realistic expectations of the likelihood of total eradication of disease. PSA velocityor rate of change in PSA levelhas been shown to be a specific marker for the presence of prostate cancer because most men without prostate cancer (less than 5%) have a low rate of change in PSA (less than 0.75 ng/mUyr).4*5 In combination with other predictive markers, PSA velocity may also play a role in the differentiation of local recurrence from distant metastases following radical prostatectomy.6 Because a persistently rising PSA suggests disease progression in men with a diagnosis of prostate cancer, we retrospectively evaluated the ability of PSA velocity prior to prostatic biopsy to predict disease extent among patients who underwent radical prostatectomy. MATERIAL
AND
METHODS
PATIENT P~PUUTI~N Between November 1993 and October 1994,82 of 368 consecutive men (mean age 59 years; range 40 to 71) who unOOYO-4295/97/$17.00 PII SOO90-4295(97)00068-X
derwent pelvic lymph node dissection and radical retropubic prostatectomy for clinically localized prostate cancer by two surgeons (H.B.C. and P.C.W.) had serial PSA measurements available prior to prostate biopsy. Men with a history of simple prostatectomy or antiandrogenic treatment were excluded from this retrospective study. PSA velocity was determined from at least two PSA measurements before biopsy (mean 3.7, range 2 to 7) with more than 12 months between the first and the last measurements (mean 2.4 years, range 1.3 to 6.7) (Fig. 1). PSA measurements were not all performed in the same laboratory, and no attempt was made to control for the use of different assays within individuals. Initial PSA measurements were less than 4.0 ng/mL in 11 of 82 men (13%). All patients had negative preoperative bone scans. Local clinical stage by digital rectal examination was assigned according to the TNM system by the attending surgeon.
PATHOLOGICEVALUATION Preoperatively, the slides from the prostate needle biopsy were reviewed and graded according to the Gleason histologic grading system. All patients had grossly negative lymph nodes at the time of surgery. The removed nodes were examined microscopically for the presence of metastases. The prostate and seminal vesicles were weighed, the surface inked, and the specimen fixed in formalin. The specimen was step-sectioned at 3-mm intervals in the transverse plane, and each section was examined. Pathologic stage was determined as organ-confined (disease confined to the prostatic capsule), focal capsular penetration (a few neoplastic glands exterior to the prostate), established capsular penetration (cancer beyond the prostatic capsule, greater than focal capsular penetration), seminal vesical involvement (tumor penetrating the muscular coat of the seminal vesicles), and lymph node involvement. Cases with focal capsular penetration and a Gleason score of 6 or less were categorized in this study as organ-confined, because these patients have a similar long-term prognosis as patients with organ-confined disease.2 A comparison of the study group (n = 82) to the total 368 men in the surgical series revealed that a similar percentage of men had organ-confined disease (52% versus 54%), but that the study group had a lower percentage of men with Gleason scores of 7 or higher (26% versus 43%). STATISTICAL ANALYSIS PSA velocity was calculated as the regression slope of the PSA levels before prostatic biopsy. Logistic regression analysis was used to examine the relationship between clinical stage (palpable versus nonpalpable), Gleason score (less than 7 versus 7 or higher), total PSA level at time of diagnosis (last measurement before biopsy), and PSA velocity prior to biopsy with respect to final pathologic stage. For these analyses, the pathologic stage was classified as either organ-confined or non-organ-confined. All of the stepwise logistic regressions came to the same conclusions as the univariate logistic regressions. A power analysis revealed that the research design had a power of 0.86 to detect a relative risk of 2.0 or greater (alpha level 0.05), whereas a sample size of 160 would be needed to detect a relative risk of 1.5 or greater with a power of 0.8 (alpha level 0.05).
RESULTS AND PATHOLOGIC STAGE Of 82 patients in the study group, 81 had clinically organ-confined prostate cancer (Stage Tl/T2) (Table I). Of these 81 cases, final pathologic stage
CLINICAL
UROLOGY
49 (I), 1997
n
7
6
1
0
FIGURE 1. Number of PSA measurements as a function of time before radical retropubic prostatectomy (RRP) in 82 men with prostate cancer.
was organ-confined in 43 (53%); in the remaining 38, the cancer was non-organ-confined (47%) with established capsular penetration in all 38 cases, seminal vesicle involvement in 6 cases (7%)) and pelvic lymph node involvement in 3 cases (4%). There were no positive margins in this series of patients. In 40 patients with nonpalpable cancer (39 cases with Stage Tic and 1 case with Stage TlA), final pathologic examination revealed nonorgan-confined disease in 20 cases (Table I). Thus, clinical stage alone was not predictive of final pathologic stage. GLEASONSCOREANDPATHOLOGICSTAGE Figure 2 shows the pretreatment Gleason scores determined from prostatic biopsy in comparison to the final pathologic stage. There was a statistically significant relationship between a Gleason score of 7 or more and the presence of non-organconfined disease (P = 0.05). PSAANDPATHOLOGICSTAGE The mean PSA value (last level before biopsy) for patients with organ-confined disease was 7.27 ng/mL (SD = 4.20); for patients with non-organconfined disease, it was 8.73 ng/mL (SD = 4.84). There was no statistically significant difference in the PSA level at diagnosis for men with organ-confined versus non-organ-confined cancer. PSA VELOCITYANDPATHOLOGZC
STAGE
Mean PSA velocity in patients with organ-confined cancer was 1.12 ng/mIJyr (SD = 1.59); in non-organ-confined cases, it was 1.88 ng/mIfyr (SD = 2.75). Figure 3 shows the regression slopes for men with organ-confined and non-organ-confined prostate cancer. The number and percentage of men with organ-confined and non-organ-confined cancer who had PSA velocities of less than 717
TABLE I. Comparison of clinical and pathologic with serial PSA measurements prior to radical Clinical
Stage
No. of Men
Tla Tic T2a T2b T2c T3a All
n
Organ-Confined No. (%)
1 39 26 12 3 1 82
1 19 17 5 1 0 43
Non-Organ-Confined No. (%)
(100) (49) (65) (42) (33) WI (521
0 UN 20(51) 9 (35) 7 (58) 2 (67) 1 (100) 39 (48)
30 25 20
FIGURE 2. Relationship of pretreatment Gleason score to final pathologic stage (open bars, organ-confined disease; hatched bars, non-organ-confined disease).
15 10 5 0 2
3
4
5 Gleason
6
7
6
Score
1.0 ng/mI+r, 1.0 to 2.0 ng/mYyr, and more than 2.0 ng/muyr are shown in Table II. Univariate logistic regression analysis did not reveal a statistically significant relationship between PSA velocity and pathologic stage. However, a statistical power analysis revealed that the study did not have sufficient statistical power to detect small effects (see Material and Methods section). COMMENT After a diagnosis of adenocarcinoma of the prostate has been histologically confirmed, an accurate assessment of the stage or extent of the disease is necessary to recommend treatment rationally. The extent of disease correlates directly with prognosis in men with newly diagnosed prostate cancer.’ Treatment directed at eradication of the primary tumor is not likely to affect prognosis when the disease is no longer confined, because adjuvant therapy capable of eradicating extraprostatic disease is unavailable. Thus, an assessment of disease extent or progression is a pivotal step when counseling the patient with newly diagnosed prostate cancer regarding curative versus palliative treatment options. We evaluated the hypothesis that men with locally more advanced prostate cancer (that is, established capsular penetration, seminal vesical infiltration, or positive pelvic lymph nodes) may have a higher rate of change in PSA than do men 718
stage in 82 men prostatectomy
with organ-confined tumors. The value of PSA velocity was compared with T stage (clinical,stage), pretreatment Gleason score from prostatic biopsy, and the PSA level closest to biopsy of the prostate. CLINICAL STAGE Digital rectal examination generally tends to understage the local extent of prostate cancer among men thought to have organ-confined disease. Of the 81 men in this study with Tl or T2 tumors, only half actually had pathologically organ-confined cancer (Table I). Even among the 40 patients with nonpalpable cancer (Stage Tl), 20 had established capsular penetration on final pathologic examination. Thus in this series, T stage was not predictive of disease extent. GLEASONSCORE The role of pretreatment Gleason score as a predictor of disease extent has been well described.’ A Gleason score of 7 or more on a pretreatment biopsy is strongly associated with pathologically non-organ-confined disease. In the present study, the correlation between a Gleason score of 7 or above and non-organ-confined disease was the only statistically significant finding. PSA LEVELATDIAGNOSIS
It is well recognized that for groups of men with prostate cancer, serum PSA correlates directly with UROLOGY
49 (51, 1997
Organ-Confined
PSA wmu
Cases
Non-Organ
3o
30
25
25
20
20
15
15
10
IO
5
5
0
0 4
3
2
1
Years Before RRP
0
4
3
Confined
2
Cases
1
0
Years Before RRP
Observed (dotted lines) and average (solid lines) PSA levels in 43 patients with organ-confined disease, and 38 patients with non-organ-confined disease on final pathologic examination as a function of years before radical retropubic prostatectomy (RRP). FIGURE
3.
II. Number and percentage with organ-confined and non-OFgan-COnfined cancer, by PSA velocity*
TABLE
Pathologic OrganConfined
PSA Velocity =c 1 .O ng/mL/yr 1 .O-2.0 ng/mL/yr >2.0 ng/mUyr * Values
in parentheses
23 (55)*
Q (21) 10 (24) are percentage
of men
Stage Non-OrganConfined 13 (33) 16 (41) 10 (26)
of men in each category.
advancing clincal and pathologic stage. However, in most cases, PSA alone does not provide accurate staging information for the individual patient because of overlap in PSA levels between stages.’ The preoperative interpretation of serum PSA as it relates to tumor extent is confounded by both the volume of benign prostatic hyperplasia tissue present and the tumor grade, both of which influence serum PSA levels in men with prostate cancer.’ In the present study, the single PSA value closest to prostate biopsy was not predictive of disease extent. PSA VELOCITY The relationship between PSA velocity and local disease extent determined pathologically has not been previously described. The mean PSA velocity UROLOGY
49 (II), 1997
among men with non-organ-confined cancer (1.88 ng/mIJyr> was higher than for men with organconfined cancer (1.12 ng/mIfyr). Logistic regression analysis did not reveal a statistically significant relationship between PSA velocity and final pathologic stage in the current study. However, a statistical power analysis showed that this study did not have the power to detect small effects. An additional limitation of the current study is the inability to account for the use of different assays for measurement of PSA within and between individuals which could have affected results. Cadeddu et aI.* previously demonstrated that changes in PSA prior to the diagnosis of prostate cancer were not predictive of later outcome in terms of development of metastatic disease and death from prostate cancer. Thus, although a persistently rising PSA among men with a diagnosis of prostate cancer may reflect disease progression in some cases, the rate of rise in PSA prior to a diagnosis of prostate cancer would not appear to reflect disease extent using pathologic stage as a surrogate. CONCLUSIONS Accurate preoperative determination of disease extent in patients with prostatic cancer remains a challenging dilemma. As in previous studies, Gleason score obtained from pretreatment prostate biopsy was a powerful predictor of disease extent in 719
the current study. Assessment of the PSA velocity prior to the diagnosis of prostate cancer would not appear to be predictive of disease extent among men with clinically localized prostate cancer. REFERENCES 1. Partin AW, Yoo J, Carter HB, Pearson JD, Chan DW, Epstein JI, and Walsh PC: The use of prostate specific antigen, clinical stage and Gleason score to predict pathological stage in men with localized prostate cancer. J Urol 150: 1 lo- 114, 1993. 2. Partin AW, Pound CR, Clemens JQ, Epstein JI, and Walsh PC: Prostate-specific antigen after anatomical radical prostatectomy: the Johns Hopkins experience after ten years. Urol Clin North Am 20: 713-725, 1993. 3. Catalona WJ, and Smith DJ: Five-year tumor recurrence rates after anatomical radical retropubic prostatectomy for prostate cancer. J Urol 152: 1837- 1842, 1994. 4. Carter HB, Pearson JD, Waclawic Z, Metter EJ, Chan DW, Guess HA, and Walsh PC: Prostate-specific antigen vari-
720
ability in men without prostate cancer: effect of sampling interval on prostate specific antigen velocity. Urology 45: 591596, 1995. 5. Kadmon D, Weinberg AD, Williams RH, Pavlik VN, Cooper P, and Migliore PJ: Pitfalls in interpreting prostate specific antigen velocity. J Urol 155: 1655-1657, 1996. 6. Partin AW, Pearson JD, Landis PK, Carter HB, Pound CR, Clemens JQ, Epstein JI, and Walsh PC: Evaluation of serum prostate-specific antigen velocity after radical prostatectomy to distinguish local recurrence from distant metastasis. Urology 43: 649-659, 1994. 7. Partin AW, Carter HB, Chan DW, Epstein JI, Oesterling JE, Rock RC, Weber JP, and Walsh PC: Prostate specific antigen in the staging of localized prostate cancer: influence of tumor differentiation, tumor volume and benign hyperplasia. J Urol 143: 747-752,199O. 8. Cadeddu JA, Pearson JD, Partin AW, Epstein JI, and Carter HB: Relationship between changes in prostate-specific antigen and prognosis of prostate cancer. Urology 42: 383389, 1993.
UROLOGY
49 61, 1997
ROLE OF PROSTATE-SPECIFIC ANTIGEN VELOCITY IN PREDICTION OF FINAL PATHOLOGIC STAGE IN MEN WITH LOCALIZED PROSTATE CANCER RALF THIEL, JAY D. PEARSON, JONATHAN PATRICK C. WALSH, AND H. BALLENTINE
I. EPSTEIN, CARTER
ABSTRACT Objectives. Prostate-specific antigen (PSA) velocity prior to radical retropubic prostatectomy was evaluated to determine if men with a faster rate of rise in PSA have locally more extensive prostate cancer. Methods. Of 368 men who underwent radical retropubic prostatectomy, 82 had two to seven PSA measurements between 1.3 and 6.7 years before prostate biopsy for evaluation of PSA velocity. PSA velocity and the pretreatment parameters of PSA, Gleason score, and T stage were evaluated as predictive parameters of pathologic stage. Results. In men with pathologically organ-confined disease, PSA velocity was 1.12 ng/mUyr; in non-organconfined cases, it was 1.88 ng/mUyr. There was a statistically significant relationship between a Gleason score of 7 and above and the pathologic extent of disease. There was no statistically significant relationship between T stage, the PSA at diagnosis, and PSA velocity with respect to final pathologic stage. Conclusions. PSA velocity is not a strong predictor of pathologic stage in men with localized prostate cancer. UROLOGY 49: 7 16-720, 1997. 0 1997, Elsevier Science Inc. All rights reserved.
P
rostate-specific antigen (PSA) has become one of the most useful tumor markers in oncology; it is now widely used for the early detection of prostate cancer, as well as in staging and monitoring men with the disease. The pretreatment PSA level provides additional information, when used together with T stage and pretreatment Gleason score, for assessing the extent of prostate cancer prior to recommending treatment.’ Despite progress in the pretreatment prediction of disease extent, contemporary radical prostatectomy series reveal that 40% to 60% of men have extraprostatic disease at the time of surgery.2,3 The long-term risk
From the Departments of Urology and Pathology, The Johns Hopkins University School of Medicine, James Buchanan Brady Urological Institute, The Johns Hopkins Medical Institutions, Baltimore, Maryland Dr. Thiel is currently at the Urological Clinic, HeinrichHeine-University, Diisseldorf, Germany. Dr. Pearson is currently at the Department of Epidemiology, Merck Research Laboratories, Blue Bell, Pennsylvania. Reprint requests: H. Ballentine Carter, M.D., Brady Urological Institute, TheJohns Hopkins Medical Institutions, 600 North Wolfe
Street,
Baltimore,
MD
21287-2101
Submitted: September 23,1996, accepted (with revisions): November 30, 1996 0 1997,
7
16
ALL
RIGHTS
ELSEVIER SCIENCE RESERVED
INC.
of residual disease is increased in the setting of extraprostatic disease. Thus, improvements in the pretreatment prediction of disease extent are needed to provide patients with realistic expectations of the likelihood of total eradication of disease. PSA velocityor rate of change in PSA levelhas been shown to be a specific marker for the presence of prostate cancer because most men without prostate cancer (less than 5%) have a low rate of change in PSA (less than 0.75 ng/mUyr).4*5 In combination with other predictive markers, PSA velocity may also play a role in the differentiation of local recurrence from distant metastases following radical prostatectomy.6 Because a persistently rising PSA suggests disease progression in men with a diagnosis of prostate cancer, we retrospectively evaluated the ability of PSA velocity prior to prostatic biopsy to predict disease extent among patients who underwent radical prostatectomy. MATERIAL
AND
METHODS
PATIENT P~PUUTI~N Between November 1993 and October 1994,82 of 368 consecutive men (mean age 59 years; range 40 to 71) who unOOYO-4295/97/$17.00 PII SOO90-4295(97)00068-X
derwent pelvic lymph node dissection and radical retropubic prostatectomy for clinically localized prostate cancer by two surgeons (H.B.C. and P.C.W.) had serial PSA measurements available prior to prostate biopsy. Men with a history of simple prostatectomy or antiandrogenic treatment were excluded from this retrospective study. PSA velocity was determined from at least two PSA measurements before biopsy (mean 3.7, range 2 to 7) with more than 12 months between the first and the last measurements (mean 2.4 years, range 1.3 to 6.7) (Fig. 1). PSA measurements were not all performed in the same laboratory, and no attempt was made to control for the use of different assays within individuals. Initial PSA measurements were less than 4.0 ng/mL in 11 of 82 men (13%). All patients had negative preoperative bone scans. Local clinical stage by digital rectal examination was assigned according to the TNM system by the attending surgeon.
PATHOLOGICEVALUATION Preoperatively, the slides from the prostate needle biopsy were reviewed and graded according to the Gleason histologic grading system. All patients had grossly negative lymph nodes at the time of surgery. The removed nodes were examined microscopically for the presence of metastases. The prostate and seminal vesicles were weighed, the surface inked, and the specimen fixed in formalin. The specimen was step-sectioned at 3-mm intervals in the transverse plane, and each section was examined. Pathologic stage was determined as organ-confined (disease confined to the prostatic capsule), focal capsular penetration (a few neoplastic glands exterior to the prostate), established capsular penetration (cancer beyond the prostatic capsule, greater than focal capsular penetration), seminal vesical involvement (tumor penetrating the muscular coat of the seminal vesicles), and lymph node involvement. Cases with focal capsular penetration and a Gleason score of 6 or less were categorized in this study as organ-confined, because these patients have a similar long-term prognosis as patients with organ-confined disease.2 A comparison of the study group (n = 82) to the total 368 men in the surgical series revealed that a similar percentage of men had organ-confined disease (52% versus 54%), but that the study group had a lower percentage of men with Gleason scores of 7 or higher (26% versus 43%). STATISTICAL ANALYSIS PSA velocity was calculated as the regression slope of the PSA levels before prostatic biopsy. Logistic regression analysis was used to examine the relationship between clinical stage (palpable versus nonpalpable), Gleason score (less than 7 versus 7 or higher), total PSA level at time of diagnosis (last measurement before biopsy), and PSA velocity prior to biopsy with respect to final pathologic stage. For these analyses, the pathologic stage was classified as either organ-confined or non-organ-confined. All of the stepwise logistic regressions came to the same conclusions as the univariate logistic regressions. A power analysis revealed that the research design had a power of 0.86 to detect a relative risk of 2.0 or greater (alpha level 0.05), whereas a sample size of 160 would be needed to detect a relative risk of 1.5 or greater with a power of 0.8 (alpha level 0.05).
RESULTS AND PATHOLOGIC STAGE Of 82 patients in the study group, 81 had clinically organ-confined prostate cancer (Stage Tl/T2) (Table I). Of these 81 cases, final pathologic stage
CLINICAL
UROLOGY
49 (I), 1997
n
7
6
1
0
FIGURE 1. Number of PSA measurements as a function of time before radical retropubic prostatectomy (RRP) in 82 men with prostate cancer.
was organ-confined in 43 (53%); in the remaining 38, the cancer was non-organ-confined (47%) with established capsular penetration in all 38 cases, seminal vesicle involvement in 6 cases (7%)) and pelvic lymph node involvement in 3 cases (4%). There were no positive margins in this series of patients. In 40 patients with nonpalpable cancer (39 cases with Stage Tic and 1 case with Stage TlA), final pathologic examination revealed nonorgan-confined disease in 20 cases (Table I). Thus, clinical stage alone was not predictive of final pathologic stage. GLEASONSCOREANDPATHOLOGICSTAGE Figure 2 shows the pretreatment Gleason scores determined from prostatic biopsy in comparison to the final pathologic stage. There was a statistically significant relationship between a Gleason score of 7 or more and the presence of non-organconfined disease (P = 0.05). PSAANDPATHOLOGICSTAGE The mean PSA value (last level before biopsy) for patients with organ-confined disease was 7.27 ng/mL (SD = 4.20); for patients with non-organconfined disease, it was 8.73 ng/mL (SD = 4.84). There was no statistically significant difference in the PSA level at diagnosis for men with organ-confined versus non-organ-confined cancer. PSA VELOCITYANDPATHOLOGZC
STAGE
Mean PSA velocity in patients with organ-confined cancer was 1.12 ng/mIJyr (SD = 1.59); in non-organ-confined cases, it was 1.88 ng/mIfyr (SD = 2.75). Figure 3 shows the regression slopes for men with organ-confined and non-organ-confined prostate cancer. The number and percentage of men with organ-confined and non-organ-confined cancer who had PSA velocities of less than 717
TABLE I. Comparison of clinical and pathologic with serial PSA measurements prior to radical Clinical
Stage
No. of Men
Tla Tic T2a T2b T2c T3a All
n
Organ-Confined No. (%)
1 39 26 12 3 1 82
1 19 17 5 1 0 43
Non-Organ-Confined No. (%)
(100) (49) (65) (42) (33) WI (521
0 UN 20(51) 9 (35) 7 (58) 2 (67) 1 (100) 39 (48)
30 25 20
FIGURE 2. Relationship of pretreatment Gleason score to final pathologic stage (open bars, organ-confined disease; hatched bars, non-organ-confined disease).
15 10 5 0 2
3
4
5 Gleason
6
7
6
Score
1.0 ng/mI+r, 1.0 to 2.0 ng/mYyr, and more than 2.0 ng/muyr are shown in Table II. Univariate logistic regression analysis did not reveal a statistically significant relationship between PSA velocity and pathologic stage. However, a statistical power analysis revealed that the study did not have sufficient statistical power to detect small effects (see Material and Methods section). COMMENT After a diagnosis of adenocarcinoma of the prostate has been histologically confirmed, an accurate assessment of the stage or extent of the disease is necessary to recommend treatment rationally. The extent of disease correlates directly with prognosis in men with newly diagnosed prostate cancer.’ Treatment directed at eradication of the primary tumor is not likely to affect prognosis when the disease is no longer confined, because adjuvant therapy capable of eradicating extraprostatic disease is unavailable. Thus, an assessment of disease extent or progression is a pivotal step when counseling the patient with newly diagnosed prostate cancer regarding curative versus palliative treatment options. We evaluated the hypothesis that men with locally more advanced prostate cancer (that is, established capsular penetration, seminal vesical infiltration, or positive pelvic lymph nodes) may have a higher rate of change in PSA than do men 718
stage in 82 men prostatectomy
with organ-confined tumors. The value of PSA velocity was compared with T stage (clinical,stage), pretreatment Gleason score from prostatic biopsy, and the PSA level closest to biopsy of the prostate. CLINICAL STAGE Digital rectal examination generally tends to understage the local extent of prostate cancer among men thought to have organ-confined disease. Of the 81 men in this study with Tl or T2 tumors, only half actually had pathologically organ-confined cancer (Table I). Even among the 40 patients with nonpalpable cancer (Stage Tl), 20 had established capsular penetration on final pathologic examination. Thus in this series, T stage was not predictive of disease extent. GLEASONSCORE The role of pretreatment Gleason score as a predictor of disease extent has been well described.’ A Gleason score of 7 or more on a pretreatment biopsy is strongly associated with pathologically non-organ-confined disease. In the present study, the correlation between a Gleason score of 7 or above and non-organ-confined disease was the only statistically significant finding. PSA LEVELATDIAGNOSIS
It is well recognized that for groups of men with prostate cancer, serum PSA correlates directly with UROLOGY
49 (51, 1997
Organ-Confined
PSA wmu
Cases
Non-Organ
3o
30
25
25
20
20
15
15
10
IO
5
5
0
0 4
3
2
1
Years Before RRP
0
4
3
Confined
2
Cases
1
0
Years Before RRP
Observed (dotted lines) and average (solid lines) PSA levels in 43 patients with organ-confined disease, and 38 patients with non-organ-confined disease on final pathologic examination as a function of years before radical retropubic prostatectomy (RRP). FIGURE
3.
II. Number and percentage with organ-confined and non-OFgan-COnfined cancer, by PSA velocity*
TABLE
Pathologic OrganConfined
PSA Velocity =c 1 .O ng/mL/yr 1 .O-2.0 ng/mL/yr >2.0 ng/mUyr * Values
in parentheses
23 (55)*
Q (21) 10 (24) are percentage
of men
Stage Non-OrganConfined 13 (33) 16 (41) 10 (26)
of men in each category.
advancing clincal and pathologic stage. However, in most cases, PSA alone does not provide accurate staging information for the individual patient because of overlap in PSA levels between stages.’ The preoperative interpretation of serum PSA as it relates to tumor extent is confounded by both the volume of benign prostatic hyperplasia tissue present and the tumor grade, both of which influence serum PSA levels in men with prostate cancer.’ In the present study, the single PSA value closest to prostate biopsy was not predictive of disease extent. PSA VELOCITY The relationship between PSA velocity and local disease extent determined pathologically has not been previously described. The mean PSA velocity UROLOGY
49 (II), 1997
among men with non-organ-confined cancer (1.88 ng/mIJyr> was higher than for men with organconfined cancer (1.12 ng/mIfyr). Logistic regression analysis did not reveal a statistically significant relationship between PSA velocity and final pathologic stage in the current study. However, a statistical power analysis showed that this study did not have the power to detect small effects. An additional limitation of the current study is the inability to account for the use of different assays for measurement of PSA within and between individuals which could have affected results. Cadeddu et aI.* previously demonstrated that changes in PSA prior to the diagnosis of prostate cancer were not predictive of later outcome in terms of development of metastatic disease and death from prostate cancer. Thus, although a persistently rising PSA among men with a diagnosis of prostate cancer may reflect disease progression in some cases, the rate of rise in PSA prior to a diagnosis of prostate cancer would not appear to reflect disease extent using pathologic stage as a surrogate. CONCLUSIONS Accurate preoperative determination of disease extent in patients with prostatic cancer remains a challenging dilemma. As in previous studies, Gleason score obtained from pretreatment prostate biopsy was a powerful predictor of disease extent in 719
the current study. Assessment of the PSA velocity prior to the diagnosis of prostate cancer would not appear to be predictive of disease extent among men with clinically localized prostate cancer. REFERENCES 1. Partin AW, Yoo J, Carter HB, Pearson JD, Chan DW, Epstein JI, and Walsh PC: The use of prostate specific antigen, clinical stage and Gleason score to predict pathological stage in men with localized prostate cancer. J Urol 150: 1 lo- 114, 1993. 2. Partin AW, Pound CR, Clemens JQ, Epstein JI, and Walsh PC: Prostate-specific antigen after anatomical radical prostatectomy: the Johns Hopkins experience after ten years. Urol Clin North Am 20: 713-725, 1993. 3. Catalona WJ, and Smith DJ: Five-year tumor recurrence rates after anatomical radical retropubic prostatectomy for prostate cancer. J Urol 152: 1837- 1842, 1994. 4. Carter HB, Pearson JD, Waclawic Z, Metter EJ, Chan DW, Guess HA, and Walsh PC: Prostate-specific antigen vari-
720
ability in men without prostate cancer: effect of sampling interval on prostate specific antigen velocity. Urology 45: 591596, 1995. 5. Kadmon D, Weinberg AD, Williams RH, Pavlik VN, Cooper P, and Migliore PJ: Pitfalls in interpreting prostate specific antigen velocity. J Urol 155: 1655-1657, 1996. 6. Partin AW, Pearson JD, Landis PK, Carter HB, Pound CR, Clemens JQ, Epstein JI, and Walsh PC: Evaluation of serum prostate-specific antigen velocity after radical prostatectomy to distinguish local recurrence from distant metastasis. Urology 43: 649-659, 1994. 7. Partin AW, Carter HB, Chan DW, Epstein JI, Oesterling JE, Rock RC, Weber JP, and Walsh PC: Prostate specific antigen in the staging of localized prostate cancer: influence of tumor differentiation, tumor volume and benign hyperplasia. J Urol 143: 747-752,199O. 8. Cadeddu JA, Pearson JD, Partin AW, Epstein JI, and Carter HB: Relationship between changes in prostate-specific antigen and prognosis of prostate cancer. Urology 42: 383389, 1993.
UROLOGY
49 61, 1997