- Email: [email protected]
The Volume and Anatomical Location of Residual Tumor in Radical Prostatectomy Specimens Removed for Stage A1 Prostate Cancer
0022-534 7/88/1395-0975$2.00/0 Vol. 139, May
THE JOURNAL OF UROLOGY
Printed in U.S.A.
Copyright © 1988 by The Williams & Wilkins Co.
THE VOLUME AND ANATOMICAL LOCATION OF RESIDUAL TUMOR IN RADICAL PROSTATECTOMY SPECIMENS REMOVED FOR STAGE Al PROSTATE CANCER JONATHAN I. EPSTEIN,* JOSEPH E. OESTERLING
AND
PATRICK C. WALSH
From the Departments of Pathology and Urology, The Johns Hopkins University School of Medicine and the James Buchanan Brady Urological Institute, The Johns Hopkins Hospital, Baltimore, Maryland
ABSTRACT
A detailed histological analysis of 21 prostatectomy specimens for stage Al disease was performed. Three cases (14 per cent) demonstrated no tumor, although all had foci of severe dysplasia. Of the prostates 18 (86 per cent) contained residual tumor: 13 (62 per cent) had minimal residual tumor and 5 (24 per cent) had substantial tumor. The residual carcinoma in the prostates with minimal tumor was predominantly in the apex of the gland or peripherally adjacent to the capsule such that complete removal by repeat transurethral resection would have been unlikely. Of the 5 specimens with substantial tumor, although some would have been upstaged by repeat transurethral resection, others would have remained stage Al because of the peripheral location of the residual tumor. Within the definition of stage Al, using either the percentage of tumor involvement, weight of tumor or number of tumor foci in the transurethral resection specimen, one could not predict whether the prostatectomy specimen would have no, minimal or substantial residual tumor. Postoperatively, all patients are continent and 93 per cent are potent. Because some men with stage Al carcinoma of the prostate will have substantial residual tumor not necessarily detected by repeat transurethral resection and the majority will have minimal residual tumor, which also appears to have an increased long-term risk of progression, radical prostatectomy should remain an option in the management of relatively young men with stage Al disease. (J. Ural., 139: 975-979, 1988) Previously, the discovery of small amounts of adenocarcinoma of the prostate after transurethral resection or enucleation was done for benign prostatic hyperplasia was considered an insignificant, incidental finding not requiring further treatment. In a study of men with stage Al disease Cantrell and associates in 1981 demonstrated that of 49 untreated incidentally found prostatic adenocarcinomas that occupied 5 per cent of the specimen or less and were not high grade (Gleason grades 8 to 10) only 1 (2 per cent) progressed to clinically significant disease within 4 years. 1 Although these data tended to support the concept of stage Al disease as being relatively innocuous, what remained unanswered was whether stage Al cancer patients who were followed for longer intervals still would have such a low risk of progression as to require no further treatment. We recently reported our expanded series of 94 men with untreated stage Al prostate cancer. 2 While 26 men died of other causes less than 4 years after diagnosis, of the 50 men who remained at risk for 8 years or longer after diagnosis 8 (16 per cent) had progression of disease, 6 of whom died of prostatic cancer. Our rationale for excluding the remaining 18 stage Al cancer patients with 4 to 8 years of followup from the calculation of the long-term risk of progression is that these 18 men had an average of only 6 years of followup. Since the mean interval to progression in our study was 7 years, these men had not been followed for a sufficient period to be at full risk for progression. Recognizing that patients with stage Al disease are not entirely free of the risk of progression, we have been offering radical prostatectomy as an option in the management of relatively young men with stage Al cancer. We present the histological findings in 21 radical retropubic prostatectomy specimens done for stage Al disease. By serially sectioning and totally embedding the specimen, we have been able to depict Accepted for publication September 8, 1987. * Requests for reprints: Department of Pathology, The Johns Hopkins Hospital, 600 N. Wolfe St., Baltimore, Maryland 21205.
accurately the location and volume of residual tumor within the gland. MATERIALS AND METHODS
The findings of 21 radical retropubic prostatectomies done for stage Al adenocarcinoma of the prostate were reviewed. As defined previously, stage Al prostatic cancer was defined as clinically unsuspected carcinoma that occupied 5 per cent or less of the transurethral resection specimen and that was not of high grade (combined Gleason grade 8 to 10). When the amount of the specimen involved by tumor was calculated to be at 5 per cent, the cases were considered and treated as stage A2 to avoid understaging the disease. The percentage of the specimen involved by tumor was based on naked eye examination of the glass slides after the pathologist had circled all identifiable foci of carcinoma with a marking pen. Any mention of nodularity or induration of the prostate in the preoperative evaluation excluded patients from the study. Serum prostatic acid phosphatase levels and bone scans were normal in all patients. Patient age ranged from 46 to 65 years, with a mean age of 55.6 years. All diagnostic material was obtained by transurethral resection. The interval between transurethral resection and radical prostatectomy averaged 7.1 months (range 2 to 30 months). Since most of the transurethral resections in our study were performed at outside institutions, there was no uniformity in the amount of tissue processed. At our institution at least 12 gm. (10 cassettes) of prostatic chips were processed, which will identify 91 per cent of all stage Al tumors. 3 Once an incidentally discovered prostatic cancer was identified in cases from our institution and if the entire specimen had not been processed the remainder of the tissue was submitted to arrive at an accurate assessment of the percentage of tissue involved by tumor. The average weight of the prostatic tissue removed by transurethral resection was 28.5 gm. (range 3.7 to 75 gm.) and the average number of slides prepared per case was 8.3 (range
975
976
EPSTEIN, OESTERLING AND WALSH
3 to 32). Although in cases from other institutions the entire specimen was not always embedded, the percentage of the specimen involved by tumor does not vary significantly by processing more tissue, since the chips are selected randomly for processing. Of the 21 radical prostatectomies 14 were done in 1986, 5 in 1985 and 2 in 1983. Radical prostatectomy for stage Al disease accounted for 7.8 per cent (14 of 180) of the total number of radical prostatectomies done in 1986, 3.6 per cent (5 of 140) of those in 1985 and 3.8 per cent of those (2 of 52) in 1983. After surgical removal, the intact prostate was weighed, coated over its entire external surface with india ink and fixed in 10 per cent buffered formalin for 18 to 24 hours. After fixation, the distal (apical) and proximal (basal) urethral margins were removed for histological examination. The gland was step-sectioned at 2 mm. intervals perpendicular to the long (apical-basal) axis of the gland, and each section was divided in half. These sections were designated in a manner that permitted localization of each section within the prostate and they all were examined histologically. An average of 34 slides of the prostate and seminal vesicles per case was examined (range 18 to 48 slides). RESULTS
Three specimens (14 per cent) showed no residual tumor. However, all 3 cases had foci of severe dysplasia as defined by McNeal and Bostwick.4 In these foci, although the normal glandular architecture was preserved, there was severe cytological atypia with markedly increased nuclear size and numerous prominent nucleoli. Severe dysplasia has been identified only rarely in glands without carcinoma and it has been implicated as a precursor lesion to carcinoma of the prostate. A total of 18 prostates (86 per cent) contained residual tumor. In 13 prostatectomy specimens (62 per cent) the quantity of the residual tumor was less than that which usually is seen in prostates removed for early stage, clinically detected disease. The distribution and volume of tumor are represented in parts A to M of the figure. Of these cases 12 had tumor at the apex (parts A to L of figure), 7 of which also had tumor in the mid portion of the gland. The remaining case had tumor only in the mid and basal sections of the prostate (part M of figure). In 9 of the 13 cases tumor was predominantly located in the anterior regions of the gland. All but 1 (part E of figure) of the 13 cases had some tumor situated peripherally near the prostatic capsule. The remaining 5 cases (24 per cent) had tumor indistinguishable in quantity from those glands removed for clinically evident disease (parts N to R of figure). In 2 of these specimens the carcinoma was found predominantly posterolaterally towards the capsule (parts Q and R of figure) and in another case residual tumor was located predominantly anteriorly adjacent to the capsule (part N of figure). The remaining 2 cases with substantial tumor had more centrally located tumor: 1 in the anterior and posterior regions of the gland (part O of figure) and the other anteriorly (part P of figure). In 2 of these cases there was focal penetration of the capsule at the apex with focal positive margins at these sites. The volume of tumor in the transurethral resection specimens was evaluated by the number of tumor foci, percentage of tumor involvement and estimated weight of tumor. The rationale for calculating the weight of tumor is that some investigators have suggested that 1 per cent tumor involvement of a 50 gm. specimen, which is equivalent to 0.5 gm. tumor, may be worse than 1 per cent tumor involvement of a 5 gm. specimen, which contains 0.05 gm. tumor. Consequently, we calculated the estimated weight of tumor by multiplying the weight of the transurethral resection specimen by the percentage of tumor involvement. There was a tendency for the transurethral resection specimens with lower tumor volumes to have less tumor in the prostatectomy specimens (see table). In those
cases with no, minimal or substantial tumor in the radical prostatectomy specimen the transurethral resection material had a mean of 4.33, 3.8 and 5 tumor foci, respectively, a mean tumor involvement of 1.8, 1.4 and 2.3 per cent, respectively, and mean tumor weights of 0.38, 0.2 and 0.32 gm., respectively. However, regardless of the method used to quantify the tumor, there were cases with a low volume of tumor in the transurethral resection material that had substantial tumor in the prostatectomy specimen (see table). Conversely, there were transurethral resection specimens with higher volume stage Al tumor that had no or minimal residual tumor in the radical prostatectomy specimen. Therefore, we were unable to use the volume of tumor within the transurethral resection specimens to predict the amount of residual tumor within the gland. The grade of the tumor from the transurethral resection material as determined by the Gleason system also showed overlap among the 3 groups (see table). 5 The combined Gleason grade of the tumor was 3 in 4 cases, 4 in 8, 5 in 7 and 6 in 2. In only 1 case was there a difference by more than 1 in the combined Gleason grade between the grade assigned to the tumor in the transurethral resection material and that given to the cancer in the corresponding radical prostatectomy specimen. The highest combined Gleason grade seen within the radical prostatectomy specimens was 5, which was present in 6 cases. In only 3 patients was more than 1 transurethral resection performed. In 1 patient there were 4 foci of tumor in the original resection and 1 focus in the repeat transurethral resection material. The radical prostatectomy specimen obtained from this patient showed a single focus of residual tumor. One patient had 3 foci of cancer in the first transurethral resection specimen and 2 in the second, with the prostatectomy specimen showing only foci of severe dysplasia. One patient had 1 focus in the initial and 6 in the subsequent transurethral resection specimen. The prostatectomy specimen in this patient showed substantial tumor posterolaterally and anteriorly in a subcapsular distribution (part R of figure). Seventeen patients were followed for 6 months or longer. All men are continent. Fifteen patients were potent preoperatively and 14 (93 per cent) are potent postoperatively. The only impotent patient was 1 of the earliest patients to undergo surgery with the nerve-sparing technique. DISCUSSION
Recently, we reported our series of 94 men with untreated stage Al disease with extended followup. 2 While 26 men (mean age 75 years) died of other causes less than 4 years after diagnosis, of the 50 men who remained at risk for progression 8 years or longer after diagnosis 8 (16 per cent) had progression of the disease, 6 of whom died of the cancer. In a smaller series from the Mayo Clinic Blute and associates also demonstrated that men with stage Al disease (less than 1 cc and well differentiated) are at increased risk for progression. 6 Of 15 men less than 60 years old who were followed for a minimum of 10 years 4 (27 per cent) experienced progression of disease. In our series neither volume nor grade predicted progression, since of our 8 tumors that progressed 4 involved less than 1 per cent of the tissue, 7 had 2 tumor foci or less and 6 were low grade (Gleason combined grades 2 to 4). Similarly, in the Mayo Clinic series all tumors that progressed had 2 tumor foci or less and were low grade. The average interval to progression in our series was 7 years and in the Mayo Clinic series it was 10 years. Therefore, patients with stage Al prostate cancer appear to have an increased long-term risk for progression, if they do not die of intercurrent disease. Consequently, we have been offering radical prostatectomy as a treatment option for relatively young men with stage Al disease. Our study has evaluated the pathological findings of 21 radical retropubic prostatectomy specimens done for stage Al adenocarcinoma of the prostate.
R. \a.ten3.\
L. \aleY-al
C
H
R.\atera.l
R..1atera1
L.1ater'al
R.
N
Q 0 L. la/tez'al
p
Schematic diagram of prostate as seen from right and left lateral views. Prostatic urethra is represented by hatched lines. Vertical lines labeled 1 10 correspond to different regions of prostate, that is section 1 represents most apical portion of gland, section 3 shows prostate at level of vernmontanum and so forth. UR, urethra at apex. URbl, urethra at bladder neck. Veru, verumontanum. Ejd, ejaculatory duct. Residual tumor in radical prostatectomy specimen done for stage Al disease has been accurately illustrated and corresponds to schematic diagram to demonstrate location of tumor within gland. On cross-sectional view tumor is depicted by dark black areas with light gray shading used for In lateral views areas of dark shading contiguous with black areas on cross-section demonstrate tumor extension. For example, left lateral vi,;w of N shows 2 tumor foci at most apical portion of gland (section 1), which extends proximally (section 2) to form 3 discrete areas of tumor. Dark shading extending from 3 tumor foci illustrate tumor extension proximally to mid portion of gland (section 5) where tumm merges to form area of residual tumor. Diameter of prostatic urethra at all levels of gland also has been depicted accurately in each case to show ucu,n,rn,rnJJ of residual tumor to prior area of transurethral resection.
978
EPSTEIN, OESTERLING AND WALSH
Parfitt and associates, in the only other large series ofradical prostatectomy for stage Al cancer, reported that 52 per cent of the prostates contained no residual tumor. 7 However, in their study the entire gland was not embedded and examined histologically, such that cases with only a few residual foci would not have been identified. Even with their definition of stage Al (5 foci or less), only 13 per cent (2 of 15) of the radical prostatectomies in our series had no residual tumor. In our study 18 of the glands (86 per cent) contained residual tumor: 13 (62 per cent) had less tumor than what usually is seen in radical prostatectomy specimens done for early stage, clinically detected disease and 5 (24 per cent) had tumor similar to what is found in patients undergoing radical prostatectomy for low volume clinically detected disease. Residual tumor was found more frequently in the anterior regions of the gland. In 9 of the 13 cases (69 per cent) with minimal residual tumor the residual carcinoma was predominantly located anteriorly. This finding may reflect a different distribution of carcinoma in stage Al disease compared to stage B tumor, which tends to be posterior and/or posterolateral in location. Alternatively, the tumor removed by transurethral resection may have come preferentially from the posterior and/ or posterolateral regions, leaving as the residual tumor that which is located anteriorly. The amount of tumor removed by transurethral resection in these cases is significant in relation to the amount of residual tumor, such that it is difficult to make conclusive statements about the distribution of stage Al tumor before transurethral resection. The classical paper by Moore on small prostatic carcinomas showed that only 14.8 per cent of 52 small incidental prostatic carcinomas found at autopsy arose in the anterior lobe, suggesting that the marked anterior distribution of tumor after transurethral resection in our study probably is partly a result of relative sparing by transurethral resection of anterior tumor. 8 In 2 of 5 cases (40 per cent) of substantial residual disease the anterior distribution of the tumors may have accounted for their having escaped clinical detection. However, 3 other cases had substantial posterior residual tumor despite benign rectal examinations by numerous urologists, which demonstrated that factors other than location have a role as to whether a sizable tumor is palpable. One option for the management of the patient with stage Al disease is repeat transurethral resection. Some urologists believe that this may provide prognostic information on whether residual tumor is present and the possibility of completely eliminating residual tumor if it is present in low volume. Studies on the use of repeat transurethral resection in the evaluation of stage Al disease have shown that in the majority of the cases
repeat transurethral resection usually demonstrates either no tumor or only a small amount of tumor. Among the larger series addressing this issue the percentages of stage Al cancer patients whose disease was upstaged were 7, 9 9.7, 10 5, 11 3.5 7 and 26. 12 While it has been proposed that patients whose disease is not upstaged need no further treatment, most of these studies have not demonstrated the significance of the findings in terms of disease progression with long-term followup data. In our study, by totally embedding and examining the prostate, and noting carefully the location of the residual tumor, certain issues regarding repeat transurethral resection can be addressed. Of our 4 patients with substantial tumor at prostatectomy who underwent only 1 transurethral resection 2 had residual tumor more centrally, which could have been sampled at repeat transurethral resection (parts A and O of figure). In the third (part Q of figure) and possibly the fourth (part N of figure) patients, despite substantial residual tumor, the disease might not have been upstaged owing to the peripheral location of the tumor. Patient 5 with substantial residual tumor had 2 resections before prostatectomy that showed peripheral tumor (part R of figure). In this case the number of foci increased from 1 to 6 and the percentage of the specimen involved by tumor increased from 0.5 to 3, which according to our definition of stage Al did not result in upstaging. If we had lowered the threshold for the definition of stage Al to 3 foci or less (with this definition two-thirds of the stage A cancer patients would be considered as having stage A2 disease), 13 2 of our 3 patients with no residual tumor and 5 of 13 with minimal residual tumor would initially have been classified as having stage A2 cancer. Therefore, with 3 foci or less to define stage Al, many older men who are at greater risk of dying of other causes will be treated aggressively for so-called stage A2 disease and they will have minimal findings at prostatectomy. Of the 13 specimens with minimal residual tumor some of the residual tumor in the majority of the cases was either at the apex of the gland or peripherally situated adjacent to the prostatic capsule, such that complete removal of tumor by repeat transurethral resection would have been unlikely. Although the fate of these individuals with only minimal residual tumor within the prostate had they not undergone resection is not known, there is some evidence to suggest that these small foci are not necessarily innocuous. Of the 8 stage Ai cancer patients in our prior study who had progression 4 had less than 1 per cent tumor in the transurethral resection specimen and 7 of the 8 had 2 tumor foci or less. This transurethral resection profile is more comparable to our 13 current patients with minimal residual tumor than the 5 with substantial tumor, of whom 4 had 2 per cent or more of the resection specimen
Correlation of transurethral resection specimen tumor volume and grade with findings at radical prostatectomy Tumor Volume and Grade in Transurethral Resection Specimens*
Findings in Prostates Removed by Radical Prostatectomy No Residual Tumor (3 cases)
Minimal Residual Tumor (13 cases)
Substantial Residual Tumor (5 cases)
1 2
8 5
4
2 0 1
7 3 3
1 2 2
2 0
7t
2t
4
1
1
1 1
1
3 5 4
1 3
1
1
No. foci: ;a,3
>3 % tumor involvement: ;;il
1-3
3-4 Tumor wt. (gm.): ;;a0.2 0.2-0.6 0.6-0.84 Combined Gleason grade: 3 4 5 6
2 0 0
1
0
* For the patients who underwent 2 transurethral resections the data were calculated by combining the number of foci and calculating the percentage accordingly.
t In 2 cases the volume of tumor could not be calculated, since the specimens were not weighed.
RESIDUAL TUMOR IN RADICAL PROSTATECTOMY SPECIMENS FOR STAGE Al CANCER
tumor and 4 had 4 o:r more tumor foci. Nevertheless, rn,t,,,nt·" with substantial residual tumor lsss than 1 per cent of low grade tumor in the initial (ransurcthral resection specimen, we cannot exclude that some With nr,no-1e,PQ
979
ADDENDUM
We have now examined 32 radical prostatectomy specimens for stage Al disease with still only 3 (9 per cent) showing no residual tumor. REFERENCES
m our and in 10 years, :respectively) also is of minimal residual tumor 1:>ctu11,u.ue; error in which substantial tumor was not .u"""""'"'"'-' at transurethral resection. Further evidence that the b stage Al adenocarcinomas that progressed in our prior study \Vere not all merely examples of ·····-·"j[••-••v 0,om,p<11rw1nt-., will have substantial residual that in some will not necessarily be upstaged repeat transurethral resection. The of the :remaining patients have minimal residual tumor is not accessible repeat transurethral resection. Data with those of Brawn 14 suggest that ,-,wo,c~,.,uu who may be expected to live for of other causes this minimal residual 1m~rE;as:ea long-term risk of progression. Given associated with the nerve-sparing radical prostatectomy in this our data support radical prosas an management of young men with
THE JOURNAL OF UROLOGY
Printed in U.S.A.
Copyright © 1988 by The Williams & Wilkins Co.
THE VOLUME AND ANATOMICAL LOCATION OF RESIDUAL TUMOR IN RADICAL PROSTATECTOMY SPECIMENS REMOVED FOR STAGE Al PROSTATE CANCER JONATHAN I. EPSTEIN,* JOSEPH E. OESTERLING
AND
PATRICK C. WALSH
From the Departments of Pathology and Urology, The Johns Hopkins University School of Medicine and the James Buchanan Brady Urological Institute, The Johns Hopkins Hospital, Baltimore, Maryland
ABSTRACT
A detailed histological analysis of 21 prostatectomy specimens for stage Al disease was performed. Three cases (14 per cent) demonstrated no tumor, although all had foci of severe dysplasia. Of the prostates 18 (86 per cent) contained residual tumor: 13 (62 per cent) had minimal residual tumor and 5 (24 per cent) had substantial tumor. The residual carcinoma in the prostates with minimal tumor was predominantly in the apex of the gland or peripherally adjacent to the capsule such that complete removal by repeat transurethral resection would have been unlikely. Of the 5 specimens with substantial tumor, although some would have been upstaged by repeat transurethral resection, others would have remained stage Al because of the peripheral location of the residual tumor. Within the definition of stage Al, using either the percentage of tumor involvement, weight of tumor or number of tumor foci in the transurethral resection specimen, one could not predict whether the prostatectomy specimen would have no, minimal or substantial residual tumor. Postoperatively, all patients are continent and 93 per cent are potent. Because some men with stage Al carcinoma of the prostate will have substantial residual tumor not necessarily detected by repeat transurethral resection and the majority will have minimal residual tumor, which also appears to have an increased long-term risk of progression, radical prostatectomy should remain an option in the management of relatively young men with stage Al disease. (J. Ural., 139: 975-979, 1988) Previously, the discovery of small amounts of adenocarcinoma of the prostate after transurethral resection or enucleation was done for benign prostatic hyperplasia was considered an insignificant, incidental finding not requiring further treatment. In a study of men with stage Al disease Cantrell and associates in 1981 demonstrated that of 49 untreated incidentally found prostatic adenocarcinomas that occupied 5 per cent of the specimen or less and were not high grade (Gleason grades 8 to 10) only 1 (2 per cent) progressed to clinically significant disease within 4 years. 1 Although these data tended to support the concept of stage Al disease as being relatively innocuous, what remained unanswered was whether stage Al cancer patients who were followed for longer intervals still would have such a low risk of progression as to require no further treatment. We recently reported our expanded series of 94 men with untreated stage Al prostate cancer. 2 While 26 men died of other causes less than 4 years after diagnosis, of the 50 men who remained at risk for 8 years or longer after diagnosis 8 (16 per cent) had progression of disease, 6 of whom died of prostatic cancer. Our rationale for excluding the remaining 18 stage Al cancer patients with 4 to 8 years of followup from the calculation of the long-term risk of progression is that these 18 men had an average of only 6 years of followup. Since the mean interval to progression in our study was 7 years, these men had not been followed for a sufficient period to be at full risk for progression. Recognizing that patients with stage Al disease are not entirely free of the risk of progression, we have been offering radical prostatectomy as an option in the management of relatively young men with stage Al cancer. We present the histological findings in 21 radical retropubic prostatectomy specimens done for stage Al disease. By serially sectioning and totally embedding the specimen, we have been able to depict Accepted for publication September 8, 1987. * Requests for reprints: Department of Pathology, The Johns Hopkins Hospital, 600 N. Wolfe St., Baltimore, Maryland 21205.
accurately the location and volume of residual tumor within the gland. MATERIALS AND METHODS
The findings of 21 radical retropubic prostatectomies done for stage Al adenocarcinoma of the prostate were reviewed. As defined previously, stage Al prostatic cancer was defined as clinically unsuspected carcinoma that occupied 5 per cent or less of the transurethral resection specimen and that was not of high grade (combined Gleason grade 8 to 10). When the amount of the specimen involved by tumor was calculated to be at 5 per cent, the cases were considered and treated as stage A2 to avoid understaging the disease. The percentage of the specimen involved by tumor was based on naked eye examination of the glass slides after the pathologist had circled all identifiable foci of carcinoma with a marking pen. Any mention of nodularity or induration of the prostate in the preoperative evaluation excluded patients from the study. Serum prostatic acid phosphatase levels and bone scans were normal in all patients. Patient age ranged from 46 to 65 years, with a mean age of 55.6 years. All diagnostic material was obtained by transurethral resection. The interval between transurethral resection and radical prostatectomy averaged 7.1 months (range 2 to 30 months). Since most of the transurethral resections in our study were performed at outside institutions, there was no uniformity in the amount of tissue processed. At our institution at least 12 gm. (10 cassettes) of prostatic chips were processed, which will identify 91 per cent of all stage Al tumors. 3 Once an incidentally discovered prostatic cancer was identified in cases from our institution and if the entire specimen had not been processed the remainder of the tissue was submitted to arrive at an accurate assessment of the percentage of tissue involved by tumor. The average weight of the prostatic tissue removed by transurethral resection was 28.5 gm. (range 3.7 to 75 gm.) and the average number of slides prepared per case was 8.3 (range
975
976
EPSTEIN, OESTERLING AND WALSH
3 to 32). Although in cases from other institutions the entire specimen was not always embedded, the percentage of the specimen involved by tumor does not vary significantly by processing more tissue, since the chips are selected randomly for processing. Of the 21 radical prostatectomies 14 were done in 1986, 5 in 1985 and 2 in 1983. Radical prostatectomy for stage Al disease accounted for 7.8 per cent (14 of 180) of the total number of radical prostatectomies done in 1986, 3.6 per cent (5 of 140) of those in 1985 and 3.8 per cent of those (2 of 52) in 1983. After surgical removal, the intact prostate was weighed, coated over its entire external surface with india ink and fixed in 10 per cent buffered formalin for 18 to 24 hours. After fixation, the distal (apical) and proximal (basal) urethral margins were removed for histological examination. The gland was step-sectioned at 2 mm. intervals perpendicular to the long (apical-basal) axis of the gland, and each section was divided in half. These sections were designated in a manner that permitted localization of each section within the prostate and they all were examined histologically. An average of 34 slides of the prostate and seminal vesicles per case was examined (range 18 to 48 slides). RESULTS
Three specimens (14 per cent) showed no residual tumor. However, all 3 cases had foci of severe dysplasia as defined by McNeal and Bostwick.4 In these foci, although the normal glandular architecture was preserved, there was severe cytological atypia with markedly increased nuclear size and numerous prominent nucleoli. Severe dysplasia has been identified only rarely in glands without carcinoma and it has been implicated as a precursor lesion to carcinoma of the prostate. A total of 18 prostates (86 per cent) contained residual tumor. In 13 prostatectomy specimens (62 per cent) the quantity of the residual tumor was less than that which usually is seen in prostates removed for early stage, clinically detected disease. The distribution and volume of tumor are represented in parts A to M of the figure. Of these cases 12 had tumor at the apex (parts A to L of figure), 7 of which also had tumor in the mid portion of the gland. The remaining case had tumor only in the mid and basal sections of the prostate (part M of figure). In 9 of the 13 cases tumor was predominantly located in the anterior regions of the gland. All but 1 (part E of figure) of the 13 cases had some tumor situated peripherally near the prostatic capsule. The remaining 5 cases (24 per cent) had tumor indistinguishable in quantity from those glands removed for clinically evident disease (parts N to R of figure). In 2 of these specimens the carcinoma was found predominantly posterolaterally towards the capsule (parts Q and R of figure) and in another case residual tumor was located predominantly anteriorly adjacent to the capsule (part N of figure). The remaining 2 cases with substantial tumor had more centrally located tumor: 1 in the anterior and posterior regions of the gland (part O of figure) and the other anteriorly (part P of figure). In 2 of these cases there was focal penetration of the capsule at the apex with focal positive margins at these sites. The volume of tumor in the transurethral resection specimens was evaluated by the number of tumor foci, percentage of tumor involvement and estimated weight of tumor. The rationale for calculating the weight of tumor is that some investigators have suggested that 1 per cent tumor involvement of a 50 gm. specimen, which is equivalent to 0.5 gm. tumor, may be worse than 1 per cent tumor involvement of a 5 gm. specimen, which contains 0.05 gm. tumor. Consequently, we calculated the estimated weight of tumor by multiplying the weight of the transurethral resection specimen by the percentage of tumor involvement. There was a tendency for the transurethral resection specimens with lower tumor volumes to have less tumor in the prostatectomy specimens (see table). In those
cases with no, minimal or substantial tumor in the radical prostatectomy specimen the transurethral resection material had a mean of 4.33, 3.8 and 5 tumor foci, respectively, a mean tumor involvement of 1.8, 1.4 and 2.3 per cent, respectively, and mean tumor weights of 0.38, 0.2 and 0.32 gm., respectively. However, regardless of the method used to quantify the tumor, there were cases with a low volume of tumor in the transurethral resection material that had substantial tumor in the prostatectomy specimen (see table). Conversely, there were transurethral resection specimens with higher volume stage Al tumor that had no or minimal residual tumor in the radical prostatectomy specimen. Therefore, we were unable to use the volume of tumor within the transurethral resection specimens to predict the amount of residual tumor within the gland. The grade of the tumor from the transurethral resection material as determined by the Gleason system also showed overlap among the 3 groups (see table). 5 The combined Gleason grade of the tumor was 3 in 4 cases, 4 in 8, 5 in 7 and 6 in 2. In only 1 case was there a difference by more than 1 in the combined Gleason grade between the grade assigned to the tumor in the transurethral resection material and that given to the cancer in the corresponding radical prostatectomy specimen. The highest combined Gleason grade seen within the radical prostatectomy specimens was 5, which was present in 6 cases. In only 3 patients was more than 1 transurethral resection performed. In 1 patient there were 4 foci of tumor in the original resection and 1 focus in the repeat transurethral resection material. The radical prostatectomy specimen obtained from this patient showed a single focus of residual tumor. One patient had 3 foci of cancer in the first transurethral resection specimen and 2 in the second, with the prostatectomy specimen showing only foci of severe dysplasia. One patient had 1 focus in the initial and 6 in the subsequent transurethral resection specimen. The prostatectomy specimen in this patient showed substantial tumor posterolaterally and anteriorly in a subcapsular distribution (part R of figure). Seventeen patients were followed for 6 months or longer. All men are continent. Fifteen patients were potent preoperatively and 14 (93 per cent) are potent postoperatively. The only impotent patient was 1 of the earliest patients to undergo surgery with the nerve-sparing technique. DISCUSSION
Recently, we reported our series of 94 men with untreated stage Al disease with extended followup. 2 While 26 men (mean age 75 years) died of other causes less than 4 years after diagnosis, of the 50 men who remained at risk for progression 8 years or longer after diagnosis 8 (16 per cent) had progression of the disease, 6 of whom died of the cancer. In a smaller series from the Mayo Clinic Blute and associates also demonstrated that men with stage Al disease (less than 1 cc and well differentiated) are at increased risk for progression. 6 Of 15 men less than 60 years old who were followed for a minimum of 10 years 4 (27 per cent) experienced progression of disease. In our series neither volume nor grade predicted progression, since of our 8 tumors that progressed 4 involved less than 1 per cent of the tissue, 7 had 2 tumor foci or less and 6 were low grade (Gleason combined grades 2 to 4). Similarly, in the Mayo Clinic series all tumors that progressed had 2 tumor foci or less and were low grade. The average interval to progression in our series was 7 years and in the Mayo Clinic series it was 10 years. Therefore, patients with stage Al prostate cancer appear to have an increased long-term risk for progression, if they do not die of intercurrent disease. Consequently, we have been offering radical prostatectomy as a treatment option for relatively young men with stage Al disease. Our study has evaluated the pathological findings of 21 radical retropubic prostatectomy specimens done for stage Al adenocarcinoma of the prostate.
R. \a.ten3.\
L. \aleY-al
C
H
R.\atera.l
R..1atera1
L.1ater'al
R.
N
Q 0 L. la/tez'al
p
Schematic diagram of prostate as seen from right and left lateral views. Prostatic urethra is represented by hatched lines. Vertical lines labeled 1 10 correspond to different regions of prostate, that is section 1 represents most apical portion of gland, section 3 shows prostate at level of vernmontanum and so forth. UR, urethra at apex. URbl, urethra at bladder neck. Veru, verumontanum. Ejd, ejaculatory duct. Residual tumor in radical prostatectomy specimen done for stage Al disease has been accurately illustrated and corresponds to schematic diagram to demonstrate location of tumor within gland. On cross-sectional view tumor is depicted by dark black areas with light gray shading used for In lateral views areas of dark shading contiguous with black areas on cross-section demonstrate tumor extension. For example, left lateral vi,;w of N shows 2 tumor foci at most apical portion of gland (section 1), which extends proximally (section 2) to form 3 discrete areas of tumor. Dark shading extending from 3 tumor foci illustrate tumor extension proximally to mid portion of gland (section 5) where tumm merges to form area of residual tumor. Diameter of prostatic urethra at all levels of gland also has been depicted accurately in each case to show ucu,n,rn,rnJJ of residual tumor to prior area of transurethral resection.
978
EPSTEIN, OESTERLING AND WALSH
Parfitt and associates, in the only other large series ofradical prostatectomy for stage Al cancer, reported that 52 per cent of the prostates contained no residual tumor. 7 However, in their study the entire gland was not embedded and examined histologically, such that cases with only a few residual foci would not have been identified. Even with their definition of stage Al (5 foci or less), only 13 per cent (2 of 15) of the radical prostatectomies in our series had no residual tumor. In our study 18 of the glands (86 per cent) contained residual tumor: 13 (62 per cent) had less tumor than what usually is seen in radical prostatectomy specimens done for early stage, clinically detected disease and 5 (24 per cent) had tumor similar to what is found in patients undergoing radical prostatectomy for low volume clinically detected disease. Residual tumor was found more frequently in the anterior regions of the gland. In 9 of the 13 cases (69 per cent) with minimal residual tumor the residual carcinoma was predominantly located anteriorly. This finding may reflect a different distribution of carcinoma in stage Al disease compared to stage B tumor, which tends to be posterior and/or posterolateral in location. Alternatively, the tumor removed by transurethral resection may have come preferentially from the posterior and/ or posterolateral regions, leaving as the residual tumor that which is located anteriorly. The amount of tumor removed by transurethral resection in these cases is significant in relation to the amount of residual tumor, such that it is difficult to make conclusive statements about the distribution of stage Al tumor before transurethral resection. The classical paper by Moore on small prostatic carcinomas showed that only 14.8 per cent of 52 small incidental prostatic carcinomas found at autopsy arose in the anterior lobe, suggesting that the marked anterior distribution of tumor after transurethral resection in our study probably is partly a result of relative sparing by transurethral resection of anterior tumor. 8 In 2 of 5 cases (40 per cent) of substantial residual disease the anterior distribution of the tumors may have accounted for their having escaped clinical detection. However, 3 other cases had substantial posterior residual tumor despite benign rectal examinations by numerous urologists, which demonstrated that factors other than location have a role as to whether a sizable tumor is palpable. One option for the management of the patient with stage Al disease is repeat transurethral resection. Some urologists believe that this may provide prognostic information on whether residual tumor is present and the possibility of completely eliminating residual tumor if it is present in low volume. Studies on the use of repeat transurethral resection in the evaluation of stage Al disease have shown that in the majority of the cases
repeat transurethral resection usually demonstrates either no tumor or only a small amount of tumor. Among the larger series addressing this issue the percentages of stage Al cancer patients whose disease was upstaged were 7, 9 9.7, 10 5, 11 3.5 7 and 26. 12 While it has been proposed that patients whose disease is not upstaged need no further treatment, most of these studies have not demonstrated the significance of the findings in terms of disease progression with long-term followup data. In our study, by totally embedding and examining the prostate, and noting carefully the location of the residual tumor, certain issues regarding repeat transurethral resection can be addressed. Of our 4 patients with substantial tumor at prostatectomy who underwent only 1 transurethral resection 2 had residual tumor more centrally, which could have been sampled at repeat transurethral resection (parts A and O of figure). In the third (part Q of figure) and possibly the fourth (part N of figure) patients, despite substantial residual tumor, the disease might not have been upstaged owing to the peripheral location of the tumor. Patient 5 with substantial residual tumor had 2 resections before prostatectomy that showed peripheral tumor (part R of figure). In this case the number of foci increased from 1 to 6 and the percentage of the specimen involved by tumor increased from 0.5 to 3, which according to our definition of stage Al did not result in upstaging. If we had lowered the threshold for the definition of stage Al to 3 foci or less (with this definition two-thirds of the stage A cancer patients would be considered as having stage A2 disease), 13 2 of our 3 patients with no residual tumor and 5 of 13 with minimal residual tumor would initially have been classified as having stage A2 cancer. Therefore, with 3 foci or less to define stage Al, many older men who are at greater risk of dying of other causes will be treated aggressively for so-called stage A2 disease and they will have minimal findings at prostatectomy. Of the 13 specimens with minimal residual tumor some of the residual tumor in the majority of the cases was either at the apex of the gland or peripherally situated adjacent to the prostatic capsule, such that complete removal of tumor by repeat transurethral resection would have been unlikely. Although the fate of these individuals with only minimal residual tumor within the prostate had they not undergone resection is not known, there is some evidence to suggest that these small foci are not necessarily innocuous. Of the 8 stage Ai cancer patients in our prior study who had progression 4 had less than 1 per cent tumor in the transurethral resection specimen and 7 of the 8 had 2 tumor foci or less. This transurethral resection profile is more comparable to our 13 current patients with minimal residual tumor than the 5 with substantial tumor, of whom 4 had 2 per cent or more of the resection specimen
Correlation of transurethral resection specimen tumor volume and grade with findings at radical prostatectomy Tumor Volume and Grade in Transurethral Resection Specimens*
Findings in Prostates Removed by Radical Prostatectomy No Residual Tumor (3 cases)
Minimal Residual Tumor (13 cases)
Substantial Residual Tumor (5 cases)
1 2
8 5
4
2 0 1
7 3 3
1 2 2
2 0
7t
2t
4
1
1
1 1
1
3 5 4
1 3
1
1
No. foci: ;a,3
>3 % tumor involvement: ;;il
1-3
3-4 Tumor wt. (gm.): ;;a0.2 0.2-0.6 0.6-0.84 Combined Gleason grade: 3 4 5 6
2 0 0
1
0
* For the patients who underwent 2 transurethral resections the data were calculated by combining the number of foci and calculating the percentage accordingly.
t In 2 cases the volume of tumor could not be calculated, since the specimens were not weighed.
RESIDUAL TUMOR IN RADICAL PROSTATECTOMY SPECIMENS FOR STAGE Al CANCER
tumor and 4 had 4 o:r more tumor foci. Nevertheless, rn,t,,,nt·" with substantial residual tumor lsss than 1 per cent of low grade tumor in the initial (ransurcthral resection specimen, we cannot exclude that some With nr,no-1e,PQ
979
ADDENDUM
We have now examined 32 radical prostatectomy specimens for stage Al disease with still only 3 (9 per cent) showing no residual tumor. REFERENCES
m our and in 10 years, :respectively) also is of minimal residual tumor 1:>ctu11,u.ue; error in which substantial tumor was not .u"""""'"'"'-' at transurethral resection. Further evidence that the b stage Al adenocarcinomas that progressed in our prior study \Vere not all merely examples of ·····-·"j[••-••v 0,om,p<