The Prognostic Significance of Post-Irradiation Biopsy Results in Patients with Prostatic Cancer

0022-534 7/86/1353-0510$02.00/0 Vol. 135, March

THE JOURNAL OF UROLOGY

Printed in U.S.A.

Copyright © 1986 by The Williams & Wilkins Co.

THE PROGNOSTIC SIGNIFICANCE OF POST-IRRADIATION BIOPSY RESULTS IN PATIENTS WITH PROSTATIC CANCER PETER T. SCARDINO, JEFFREY M. FRANKEL, THOMAS M. WHEELER, RANDALL B. MEACHAM, GEORGE S. HOFFMAN, CARIE SEALE, JOHN H. WILBANKS, JAMES EASLEY AND C. EUGENE CARLTON, JR. From the Roy and Lillie Cullen Department of Urologic Research, and Departments of Urology, Pathology and Radiotherapy, Baylor College of Medicine, The Methodist Hospital and St. Luke's Episcopal Hospital, Texas Medical Center, Houston, Texas

ABSTRACT

To evaluate the prognostic significance of post-irradiation biopsy results in patients with prostatic cancer, we reviewed the records of 803 patients who had been treated with pelvic lymph node dissection, radioactive gold seed implantation and external beam irradiation. Of the patients 124 had 1 or more biopsies within 6 to 36 months after completion of radiotherapy when there was no evidence of local or distant recurrence of tumor. Patients were followed for a mean of 64 months (range 14 to 175 months) and received no other therapy before relapse. Over-all, 43 of these patients (35 per cent) had a positive biopsy result. The incidence of positive biopsy results correlated directly with the initial stage of the tumor, ranging from 22 per cent of stage BlN to 50 per cent of stage Cl lesions. However, biopsy results did not correlate with the grade of the tumor. Local recurrence and distant metastases were much more common among patients with a positive biopsy result (p equals 0.0006). Local recurrence developed in 58 per cent of the patients with a positive biopsy by 5 years and in 82 per cent by 10 years. Of those in whom all biopsies were negative only 18 per cent had local recurrence by 5 years and 32 per cent by 10 years. Biopsy results retained their prognostic significance even among the more favorable subset of patients whose pelvic lymph nodes were negative initially and those with a normal prostatic examination at biopsy. These results indicate that a post-irradiation prostate biopsy 6 to 36 months after completion of treatment can be used to determine the efficacy of a particular radiotherapeutic regimen as well as the success or failure of radiotherapy in an individual patient. Although radiotherapy has become the most common form of treatment for early stage prostatic cancer, few studies have examined thoroughly the efficacy of radiotherapy in eradicating the local tumor within the treated field. The value of postirradiation biopsy of the prostate as a means to determine the success of radiotherapy continues to generate controversy. To evaluate the response of the tumor to treatment, we reviewed the results of needle biopsies of the prostate after definitive radiotherapy in patients receiving no other therapy before relapse. The results suggest that the post-irradiation biopsy within 6 to 36 months after completion of treatment can be used to determine the efficacy of a particular radiotherapeutic regimen and the success or failure of radiotherapy in the individual patient. MATERIALS AND METHODS

Treatment protocol. Between 1966 and 1979 virtually all patients seen at our institution with clinical stages A2, B or Cl prostatic cancer were treated with a combination of radioactive gold (198Au) seed implantation and external beam irradiation. 1 • 2 After appropriate staging, treatment was initiated with a bilateral pelvic lymph node dissection and implantation of 6 to 10 radioactive gold seeds directly into the prostate gland, giving a dosage of 2,500 to 3,000 rad. External beam irradiation was begun 2 to 3 weeks later with an 8 MeV. linear accelerator. If the pelvic lymph nodes were negative for metastases, a dose of 4,000 rad was delivered to the prostate through a 6 X 6 cm. portal with a rotational technique. However, if the nodes were positive for metastases, opposing anterior and posterior 12 X Accepted for publication September 11, 1985. Read at annual meeting of American Urological Association, Las Vegas, Nevada, April 17-21, 1983.

12 cm. portals were used to deliver 5,000 rad to the entire pelvis. Therefore, the total dosage to the prostate was 6,500 to 8,000 rad. Several recent reviews have described the over-all results of this treatment protocol in detail. 1 • 2 Patient population. From 1966 to 1979, 803 patients received definitive irradiation therapy for prostatic cancer. Of these patients 328 were excluded from the analysis for various reasons (table 1), including 118 who were excluded because information in the medical records was too limited for us to determine the stage of the tumor at the time of initial treatment or to confirm that the treatment regimen described actually was followed. There remained 475 patients with biopsy proved clinical stages A2 through Cl adenocarcinoma of the prostate gland who had no evidence of distant metastases at the time of treatment, and who completed the course of radiotherapy and received no systemic treatment (hormones or chemotherapy) before documented tumor recurrence. Clinical stage of disease was defined as stage Al-no tumor palpable, 3 or less high power fields and well differentiated, stage A2-no tumor palpable, more than 3 high power fields and/or less than well differentiated, stage BIN-tumor 1.5 cm. or less confined to 1 lobe and surrounded by normal gland on 3 sides, stage Bl-intracapsular tumor more than 1.5 cm. confined to 1 lobe, stage B2-tumor involving both lobes but confined to the gland, stage Cl-tumor extending beyond the capsule, with or without invasion of the lateral sulci or seminal vesicles, and 6 cm. or less in maximum diameter, and stage C2-same criteria as stage Cl disease except more than 6 cm. Among 18 urologists who treated these patients several routinely performed transrectal needle biopsies annually for 1 to 3 years in each patient who had not had local or distant recurrence as defined. There were 124 patients who had 1 or more

510

PftOG:i'"JOSTIC SIGI"liFICAl\JCE ~---TABLE

L Patient population from 1966

Total No. pts. treated Accepted for analysis Excluded: Inadequate information, 118 Lost to followup, 12 Metastases at presentation, 17 Hormonal therapy before relapse, 48 No lymph node dissection, 91 Ca not confirmed, 5 Gold seeds not implanted, 2 No external beam irradiation, 14 Stage Al or C2 disease, 21

?OST-IRRADIArfIOf~ BlOPSY

1979 803 475 328

biopsies between 6 and 36 months after the completion of radiotherapy and for whom the results of a digital rectal examination of the prostate were recorded at the time of biopsy. These patients constituted our study group. The average age of the patients was 68.3 years, with a range of 54 to 82 years. Mean followup was 64 months (range 14 to 175 months). Of the 124 patients studied 77 (62 per cent) were followed for at least 5 years and 54 (44 per cent) for at least 10 years or until recurrence. At total of 22 patients (18 per cent) died during the observation period. Data on more than 150 parameters were abstracted from the records of each patient, including stage and grade of the primary tumor, status of the pelvic lymph nodes, and results of all prostatic examinations and biopsies (whether by needle or transurethral resection) performed after therapy. The initial clinical stage of the tumor recorded in the medical record was noted, and all descriptions of prostatic examinations before treatment were reviewed individually. The clinical stage then was reassigned in accordance with the definitions noted previously. The grade of the tumor was determined by review of the original histological slides by a single pathologist (T. M. W.) via the Gleason grading system as well as the standard system used at our institution (well, moderately or poorly differentiated). Study interval. The status of each patient was reviewed 4 years after the close of accrual in 1979. A patient was considered lost to followup at the time the record was reviewed unless he had had an adequate medical evaluation within the previous 2 years, the tumor had recurred or the patient had died. For purposes of this analysis, the time to recurrence was measured from the time of gold seed implantation until the finding of local or distant recurrence. Local recurrence was defined as a clinical phenomenon causing discrete signs or symptoni.s, rather than an abnormalit; on ~iopsy or rectal examination alone. These symptoms included the development of bladder outlet obstruction, massive HCH1<2CL!He1. Of palpable local l'PCnrOmT n of tumor O.,ClUhlcH>e ureteral obstruction evident on excretory tissue diagnosis or the resoonse to hormonal It is noteworthy that neither a ·--·-.,---" biopsy nor an abnormal post-treatment prostatic examination was in and of itself considered evidence of local recurrence. Distant recurrence was defined as the development of a persistently elevated acid phosphatase, a positive bone scan or blastic lesions seen on skeletal radiographs. Prostatic biopsy. To evaluate the prognostic significance of biopsy results, we considered only those biopsies obtained by needle, since transurethral resections almost always were performed for obstructive voiding symptoms and the tissue removed was malignant more than 90 per cent of the time. Biopsy results were disregarded if obtained when local or distant recurrence already was apparent. Because prostatic cancer does not regress immediately after irradiation, we considered biopsy results only when the biopsy was performed at least 6 months after the completion of external beam therapy. Biopsies performed more than 3 years after therapy usually were done to

::;;:_::;,;:;;•;c;;, of recurrent t;..nr101\ alrr1ost invariwere and, +h,,~n,cM,A were (m;ree:ardEld in this analysis. This left 152 biopsies in 124 patients that met our criteria,_ for an average of 1.2 (range l to 3) biopsies per patient. If a patient had undergone more than 1 biopsv the result was classi~ed as negative if all biopsies were negative. However, if any b10psy was then the result was considered positive even th?ug_h an occas_ional_ patient had a positive biopsy fol~ lowed w1thm the specified mterval a negative study or vice ve~sa. T~e biopsy specimens were reviewed by a single pathologist and were considered positive if any neoplastic cells were seen, regardless of the presumed viability of the cells or the extent of radiation changes. 3 Data analysis. When the actuarial survival curves were computed treatment was considered a failure (that is an uncensored observation) at the time local or distant recurrence was established (for example a transurethral resection was necessary for obstruction or a bone scan became positive). Otherwise treatment was ~onsidered a success (that is a censored obser~ation) as of the tnne of the last complete negative medical evaluation. The treatment also was considered a success (that is the tumor had not progressed) if the patient died of other causes with no evidence of recurrent tumor. Consequently, the actuarial curves in our report indicate the progression rate of the tumor in our patient population and not the survival or survival rate free of di~e~se. The curves should be interpreted to indicate the probab1hty that the tumor has recurred at a given point in time after the gold seed implantation. Three types of curves were generated: those showing the rate of any recurrence (local or distant), those tracing the rate of local recurrence (with or w_ithout distant recurrence), and those recording the rate of distant recurrence (with or without local recurrence). Frequency distributions, cross tabulations and Kaplan-Meier life-table analyses 4 were performed with the Statistical Program for the Social Sciences. Statistical significance of cross tabulations was determined by chi-square analysis and of actuarial survival curves by the Wilcoxon test. 5 RESULTS

The distribution of clinical stages and nodal status for the 124_ patients is shown in table 2. Over-all, 24 per cent of the patients had nodal metastases at the pelvic lymph node dissection. The grade of the primary tumor could be determined by review of the original slides in 112 of the 124 patients, and 27 (24 per cent) were considered poorly differentiated. The actual dose of irradiation received by these patients was 6,860 ± 874 (mean ± standard deviation) rad, ranging from 5,000 to 10,500 rad to the prostate the combined technique. For the entire group, the actuarial survival rate was 89 ± 6 per cent (mean ± 2 standard errors) at 5 years and 52 ± 24 cent at 10 years. At 5 years 48 ± 11 per cent ± 2 :,;un1l1aru and at 10 yearn 31 ± 15 per cent remained free of local or distant recurrence of tumoL Post-irradiation Among these 124 patients who had 1 or more needle biopsies of the prostate betvveen 6 and 36 months after the completion of therapy 81 (65 per cent) had consistently negative results and 43 (35 per cent) had 1 or more results positive for cancer. The incidence of positive biopsy results correlated directly with the clinical stage of the initial tumor

TABLE 2.

Clinical stage and nodal status of disease

Clinical Stage A2 BlN Bl B2 Cl Totals

No. Pts.

Positive Nodes No.(%)

21

5 (24)

9 36

24 34

124

1 (11) 5 7 12 30

(14) (29) (35)

(24)

512

SCARDINO AND ASSOCIATES

Post-irradiation biopsy results correlated with initial clinical stage (p <0.05), initial grade of tumor (p = 0.53), pelvic lymph node dissection (p = 0.007) and simultaneous digital rectal examination (p <0.0001)

TABLE 3.

Biopsy Results Total No. Pts. (%)

Neg. No.(%)

Pos. No.(%)

Clinical stage: A2

BlN Bl 82

Cl Totals Grade:* Well differentiated Moderately differentiated Poorly differentiated Nodal status: Neg. Pos. Prostatic examination: Normal Abnormalt

21 9 36 24 34 124

15 7 27 15 17 81

(71) (78) (75) (62) (50) (65)

6 (29) 2 9 9 17 43

(22) (25) (38) (50) (35)

47 38 27

33 (70) 23 (60) 16 (59)

14 (30) 15 (39) 11 (41)

94 (76) 30 (24)

68 (72) 13 (43)

26 (28) 17 (57)

86 (69) 38 (31)

70 (81)

11 (29)

16 (19) 27 (71)

* Unknown in 12 patients. t Suspicious for cancer.

TABLE 4.

Prognostic significance of post-irradiation biopsy results in 124 patients (p = 0.001) Recurrence

Biopsy Result Neg. Pos.

Total No.(%) 81 (65) 43 (35)

Local* No.(%)

Distantt No.(%)

Any No.(%)

11 (14) 20 (47)

24 (30) 22 (51)

25 (31) 28 (65)

correlated closely with the clinical course of the disease. By the time of the analysis local recurrence and distant metastases were much more common (p = 0.001) among the patients with a positive biopsy (table 4). When computed on an actuarial basis this difference was even more pronounced. The probability of recurrence (local or distant) for patients with a positive biopsy was 70 per cent at 5 years and 87 per cent at 10 years (fig. 1). Yet, if all of the biopsies were negative the probability was only 37 per cent at 5 years and 53 per cent at 10 years (p = 0.009). When local recurrence was used as the end point 58 per cent of the patients with a positive biopsy had local recurrence by 5 years and 82 per cent by 10 years. However, if all biopsies were negative only 18 per cent of the patients had local recurrence by 5 years and 32 per cent by 10 years, a highly significant difference (p = 0.0006) (fig. 2). Patients with nodal metastases are known to have a poor prognosis 2 •6 •7 as well as a high incidence of positive postirradiation biopsies. 8 • 9 Excluding these patients, we examined biopsy results in the more favorable group of patients with proved negative nodes. Of the 124 patients 94 (76 per cent) had negative nodes (table 3). Yet 28 per cent (26 of 94) of these patients had a positive biopsy. Of the patients whose tumors were confined to the prostate gland, that is stages A and B with negative nodes, 19 to 25 per cent had a positive biopsy (table 5). The incidence of positive biopsy results increased to 45 per cent in patients with stage Cl disease. Within this favorable group of patients with negative nodes, biopsy results retained their powerful prognostic significance. The rate of local recurrence was increased markedly among these patients if the biopsy was positive (p <0.03) (fig. 3). Correlation of prostatic biopsy with post-treatment prostatic examination. The results of the post-irradiation biopsy also were compared to those of digital rectal examination of the

* With or without distant metastases. t With or without local recurrence.

PROBABILITY OF LOCAL RECURRENCE BASED ON BIOPSY RESULTS

PROBABILITY OF RECURRENCE BASED ON BIOPSY RESUl TS 0

0.2

0.2 :;,...

i

Negative

n=81

0.4

.

0.4

...0

0.6

:C ~

n=81

A.

ca

,g

n--c---..--t., Negative

~

0.6 Positive

0.8

1.0

Positive

0.8

A.

n=43 0

Time to Recurrence (Years) FIG. 1. If needle biopsy performed 6 to 36 months (shaded area) after completion of therapy was positive, probability of recurrence was significantly greater than if biopsy was negative. Brackets at 5 and 10 years represent 95 per cent confidence interval (± 2 standard errors).

n=43

1.0--~~--~--~----~~ 10 0 5 Time to Recurrence (Years) FIG. 2. Biopsy results at 6 to 36 months proved to be powerful predictor of risk of local recurrence, so that by 10 years nearly all patients with positive biopsy had clinical evidence of local recurrence. Brackets indicate 95 per cent confidence interval.

TABLE 5.

(table 3), ranging from 22 per cent for stage BlN to 50 per cent for stage Cl lesions (p <0.05, chi-square analysis). There was no significant association of biopsy results with the grade of the tumor (table 3). However, there was a significant correlation with the status of the pelvic lymph nodes (p = 0.007). Only 28 per cent of the patients with negative nodes had a positive biopsy, compared to 57 per cent of the patients with positive nodes (table 3). The results of needle biopsies in the 6 to 36-month interval

Biopsy results by surgical stage in patients with negative pelvic lymph nodes Biopsy Results

Surgical Stage A2

BlN Bl 82

Cl Totals

No. Pts.

Neg. No.(%)

Pos. No.(%)

16 8 31 17 22

12 (75) 6 (75) 25 (81) 13 (76) 12 (55) 68 (72)

4 2 6 4 10 26

94

(25) (25) (19) (24) (45) (28)

513

PROGNOSTIC SIGNIFICANCE OF POST-HmAmATION BIOPSY IN PROSTATIC CANCER

PROBABILITY OF LOCAL RECURRENCE BASED ON BIOPSY RESULTS For Patients with Negative Nodes

0.2

~

0.4

1.i ffl .Q

Cl.6

...

0 Cl.

Positi'\le n=26

0.8

<.03

1.CI 5

0

Time to Recurrence (Years) FIG. 3. Even among more favorable subset of patients whose pelvic lymph nodes were negative for metastases at time of treatment biopsy results retained powerful prognostic significance. Brackets indicate 95 per cent confidence interval.

prostate performed at the same time Of the 38 patients with an examination suggestive of cancer (abnormal) 27 (71 per cent) had a positive biopsy. Among the 86 with a normal examination 16 (19 per cent) had a positive biopsy. Over-all, the results of the examination and the biopsy were concordant in 97 patients (78 per cent) and discordant in 27 (22 per cent) (p <0.0001). To assess the prognostic significance of biopsy results in these patients, we compared the actuarial rate of recurrence in the negative biopsy group to that in the positive biopsy group, all of whom had a normal prostatic examination at biopsy (fig. 4). The risk of recurrence was substantially greater for those with a positive biopsy. At 5 years the actuarial probability of recurrence in the 16 patients with a normal examination but a positive biopsy was 66 per cent, compared to only 29 per cent in the 70 patients with a normal examination and a negative biopsy (p = 0.02). However, among the patients with an abnormal posttreatment prostatic examination those with a negative biopsy did as poorly as those with a positive biopsy, which suggested that the negative biopsy results in these patients may represent sampling errors. DISCUSSION

PROBABILITY OF RECURRENCE BASED ON BIOPSY RESULTS For Patients with a Normal Prostatic Exam 0

0.2

•• •• •• o, •• •• •

~ 0.4

:ti Cl!

J:I 0

lb

0.6

I:!..

0.8

Negative n=70 !Positive ~m,O

n=1S

p=.02

1.0

10

5

0

Time to Recurrence {Years) FIG. 4. When post-treatment prostatic examination was normal at biopsy risk of recurrence was significantly greater in those with positive biopsy. Brackets indicate 95 per cent confidence interval. TABLE 6.

Currently, the value of a post-irradiation prostatic biopsy is questioned widely on the basis that viable tumor cells cannot be distinguished histologically from lethally irradiated cells. 10- 12 The high rate of positive post-irradiation biopsies reported from several centers 3 ' 8 ' 9 ' 13' 14 seems inconsistent with the numerous clinical reports of a low rate of local recurrence of tumor after radiotherapy. 15- 18 Animal studies using shortterm tumor models have been cited to explain this discrepancy: a biopsy obtained shortly after a lethal dose of irradiation therapy to a transplanted tumor in vivo often fails to distinguish viable tumors from nonviable, lethally irradiated tumors destined to regress. Perhaps the definitive clinical study claiming no prognostic significance for post-irradiation biopsies was that of Cox and Stoffel, 10 who reported that the incidence of positive biopsies decreased steadily from 60 per cent at 6 months to 19 per cent at 30 months after irradiation therapy. There were no differences in recurrence or survival rates between patients with negative and positive biopsies 12 or more months after therapy. Several other studies appear to support their findings. 11 • 12 ' 19 However, several serious criticisms of their study have been raised. 8 Perhaps the most important concerns the unregulated use of hormonal therapy in their patients. Of the 38 patients

Incidence and prognostic significance of post-irradiation prostatic biopsy results in series in which lwrmonal therapy was not used before recurrence Result of Biopsy

Reference

Neg.

No. Pts.

Pos.

Recurrences No.(%)

No.(%)

No.(%)

Recurrences No.(%)

Followup (yrs.)

External beam 6 1 4 0 0 0 7

(39) (38) (43) (44) (39) (48) (61)

(24) (17) (14) (0) (0) (0) (18)

Freiha and Bagshaw" Sewell and associates 13 Kiesling and associates 14 Cosgrove and Kaempf21 Kurth and associates 26 Nachtsheim and associates27 Jacobi and Hohenfellner2"

64 16 68 9 23 29 64

25 6 29 4 9 14 39

Lytton and associates 19 Schellhammer and associates, 29 and Herr and Whitmore30

22

11 (50)

1 (9)

87

60 (69)

14 (23)

39 10 39 5 14 15 25

(61) (62) (57) (56) (61) (52) (39)

28 8 11 2 2 8 16

(72) (80) (28) (40) (14) (53) (64)

2:5 5 5 1-4 1.5-3 4

11 (50)

1 (9)

2

27 (31)

18 (67)

:aa5

43 (35) 228 (45)

28 (65) 122 (54)

5

8

! seeds

125

Au seeds+ external beam

198

Present study Totals

124 506

81 (65) 278 (55)

25 (31) 58 (21)

514

SCARDINO AND ASSOCIATES

18 (47 per cent) received estrogens or orchiectomy before radiotherapy. In many patients hormonal therapy was initiated or continued after irradiation. Thus, the results are not indicative of the effects of radiotherapy alone. Not only does hormonal therapy markedly delay the appearance of metastases 20 but it often results in significant shrinkage and even complete disappearance of the primary tumor by rectal examination. In fact, hormonal therapy alone will convert the prostatic biopsy to negative in 36 to 57 per cent of the patients. 21 • 22 Therefore, it becomes impossible to determine the significance of a positive post-irradiation biopsy or to assess the probability of recurrence in patients given hormonal therapy along with irradiation. Finally, although Cox and Stoffel found no differences in survival based on biopsy results, their 38 patients were followed for a median of only 4 years in their early report and 8 years in their later reports. However, differences in survival rates between different prognostic groups of patients with prostatic cancer are not likely to become apparent for 10 to 15 years if the patients are treated with hormonal therapy. 2a- 25 In contrast to these studies, we found that a prostatic biopsy performed after irradiation accurately reflects the response of the local tumor to therapy and provides important, reliable prognostic information for the patient. There was a marked difference in the risk of clinical recurrence of the tumor (whether local or distant) based on the results of the biopsy performed between 6 and 36 months after the completion of therapy. Even when the analysis was restricted to patients in the most favorable category, those whose nodes were negative initially, a positive biopsy remained a powerful predictor of eventual local recurrence. Our results are consistent with those reported in other carefully analyzed series in which hormonal therapy was not used (table 6). 9• 13 • 14• 19• 21 • 26- 30 One of the most convincing studies is that reported by Jacobi and Hohenfellner. 28 Of 98 patients who were treated with radiotherapy alone needle aspiration biopsies 2 years after the completion of radiotherapy were positive in 38 (39 per cent). By that time 7 of these 38 patients (18 per cent) had metastases, compared to only 1 of 60 (2 per cent) with a negative biopsy. Of these patients 64 were followed for 4 years. By then, metastases had developed in 16 of 25 patients (64 per cent) whose biopsy had been positive, compared to only 7 of 39 (18 per cent) whose biopsy had been negative. Kiesling and associates reported on 68 patients who were treated with external beam radiotherapy (5,000 to 7,000 rad) and followed for a mean of 5 years. 14 Disease in two-thirds of the patients was clinical stage C. There was no decrease in the rate of positive biopsies with time, and the final biopsy was positive in 57 per cent of the patients. Biopsies done 12 or more months after the completion of radiotherapy were positive in 38 of 58 patients (66 per cent) and only 1 of these later had negative biopsies. Progression of tumor was apparent in 28 per cent of the 39 patients with a positive biopsy (even though many were given hormonal therapy after the biopsy) compared to only 14 per cent of the 29 patients with a negative biopsy. In a study of biopsy results after external beam irradiation Freiha and Bagshaw reported the results of biopsies performed 18 months or more after the completion of therapy in 64 patients. 9 The biopsy was positive in 39 patients (61 per cent). In results virtually identical to our own, they found that the biopsy results correlated closely with the stage (or size) of the primary tumor, presence of pelvic nodal metastases and abnormal post-treatment prostatic examination but not with the Gleason grade of the tumor. With a mean followup of 8 years (range 5 to 11 years) 28 of 39 patients (72 per cent) with a positive biopsy have had distant metastases. Of the 11 patients without metastases 6 were treated with diethylstilbestrol for local control, so that, in fact, 87 per cent of the patients with a positive biopsy have evidence of treatment failure. However, in the group with a negative biopsy only 6 of 25 (24 per cent) have metastases and 4 of these had positive nodes initially.

Schellhammer and associates, 29 and Herr and Whitmoreao reported biopsies after 125iodine (1251) implantation in 87 patients, 27 of whom (31 per cent) had a positive biopsy. Local recurrence developed in 44 per cent of those with a positive biopsy, compared to only 8 per cent when the biopsies were negative. Over-all, local or distant recurrence of tumor was seen in 67 per cent of the patients if the biopsy was positive but in only 23 per cent if the biopsies were negative. These data and our own results strongly attest to the powerful prognostic significance of a post-irradiation biopsy, evident in virtually every series in which the results were not obscured by the concomitant use of hormonal therapy. As measured by post-irradiation biopsy, how successful is radiotherapy in destroying the tumor within the field of irradiation? Our results suggest that treatment failed locally in 35 per cent of the patients (28 per cent of those with negative nodes), a minimum estimate, since this was a highly select group that excluded those whose tumor recurred before they were due for a routine biopsy and those who had a transurethral resection. By collecting the results of post-irradiation biopsies from many large series in which a variety of radiotherapeutic techniques were used (table 6), one can estimate the efficacy of definitive radiotherapy for localized prostatic cancer. Among 506 patients the biopsy was positive in 228 (45 per cent). Of these patients 122 (54 per cent) had evidence of treatment failure by the time of the report (mean 4 years), compared to only 58 of 278 (21 per cent) if the biopsy was negative. Nevertheless, some patients with a positive biopsy do not have local recurrence. There are several possible explanations: 1) many prostatic cancers grow slowly, so that local recurrence has not become apparent during the limited period of followup, 2) some prostatic cancers tend to metastasize rather than grow locally, 3) if hormonal therapy is begun when distant metastases appear any local tumor present usually will regress 21 • 22 and 4) benign radiation-induced atypia may be misinterpreted in the biopsy specimen as persistent carcinoma.a On the other hand, why do some patients with a negative biopsy suffer a local recurrence? Again, possible explanations for such false negative biopsies might include a sampling error when the biopsy specimen is taken, late development of a tumor de novo in the organ left in situ and misinterpretation of persistent cancer as radiation-induced atypia in the pathology specimen.a Although one should not expect biopsy results to be uniformly accurate, our results do indicate that a prostatic biopsy performed at a sufficient interval after therapy can provide an accurate gauge of the success with which a particular radiotherapeutic regimen can control the local prostatic cancer. With a properly timed post-irradiation biopsy, a physician managing a patient with prostatic cancer can detect treatment failure early, when a second chance at definitive therapy still is possible. Patients who participated in this study were under the care of Drs. W. G. Guerriero, J. D. Wright, H. M. Seybold, C. D. Cawood, W. J. Wolf, G. W. Smith, H. L. Gordon, M. Marcoe, F. B. Scott, R. Scott, D. W. Franke, W. A. Grimes, S. L. Attia, J. M. Baird, F. David and S. D. Axelrad.

REFERENCES 1. Scardino, P. T., Guerriero, W. G. and Carlton, C. E., Jr.: Surgical

staging and combined therapy with radioactive gold grain implantation and external irradiation. In: Genitourinary Tumors: Fundamental Principles and Surgical Techniques. Edited by D. E. Johnson and M. A. Boileau. New York: Grune & Stratton, chapt. 6,pp. 75-80, 1982.

PROGNOSTIC SIGNIFICANCE OF POST·IRRADIATWN BIOPSY 1N PROSTA'I'IC CANCER 2. Scardino, P. T. and Carlton, C. E., Jr.: Combined interstitial and external irradiation for prostatic cancer. In: Principles and Management of Urologic Cancer. Edited by N. Javadpour. Baltimore: The Williams & Wilkins Co., chapt. 22, pp. 392-408, 1983. 3. Bostwick, D. G., Egbert, B. M. and Fajardo, L. F.: Radiation injury of the normal and neoplastic prostate. Amer. J. Surg. Path., 6: 541, 1982. 4. Kaplan, E. L. and Meier, P.: Nonparametric estimation from incomplete observations. J. Amer. Stat. Ass., 53: 457, 1958. 5. Gehan, E. A.: A generalized Wilcoxon test for comparing arbitrarily singly-censored samples. Biometrika, 52: 203, 1965. 6. Grossman, H. B., Batata, M., Hilaris, B. and Whitmore, W. F., Jr.: 1251 implantation for carcinoma of prostate. Further follow-up of first 100 cases. Urology, 20: 591, 1982. 7. Bagshaw, M.A. and Ray, G. R.: External beam radiation therapy of prostate carcinoma. In: Urological Cancer. Edited by D. G. Skinner. New York: Grune & Stratton, pp. 53- 71, 1983. 8. Scardino, P. T.: The prognostic significance of biopsies after radiotherapy for prostatic cancer. Sem. Urol., 1: 243, 1983. 9. Freiha, F. S. and Bagshaw, M. A.: Carcinoma of the prostate: results of post-irradiation biopsy. Prostate, 5: 19, 1984. 10. Cox, J. D. and Stoffel, T. J.: The significance of needle biopsy after irradiation for stage C adenocarcinoma of the prostate. Cancer, 40: 156, 1977. 11. Kagan, A. R., Gordon, J., Cooper, J. F., Gilbert, H., Nussbaum, H. and Chan, P.: A clinical appraisal of post-irradiation biopsy in prostatic cancer. Cancer, 39: 637, 1977. 12. Leach, G. E., Cooper, J. F., Kagan, A. R., Snyder, R. and Forsythe, A.: Radiotherapy for prostatic carcinoma: post-irradiation prostatic biopsy and recurrence patterns with long-term followup. J. Urol., 128: 505, 1982. 13. Sewell, R. A., Braren, V., Wilson, S. K. and Rhamy, R. K.: Extended biopsy followup after full course radiation for resectable prostatic carcinoma. J. Urol., 113: 371, 1975. 14. Kiesling, V. J., McAninch, J. W., Goebel, J. L. and Agee, R. E.: External beam radiotherapy for adenocarcinoma of the prostate: a clinical followup. J. Urol., 124: 851, 1980. 15. Neglia, W. J., Hussey, D. H. and Johnson, D. E.: Megavoltage radiation therapy for carcinoma of the prostate. Int. J. Rad. Oncol. Biol. Phys., 9: 873, 1977. 16. Bagshaw, M.A., Ray, G. R., Pistenma, D. A., Castellino, R. A. and Meares, E. M., Jr.: External beam radiation therapy of primary carcinoma of the prostate. Cancer, 36: 723, 1975. 17. Perez, C. A., Walz, B. H., Zivnuska, F. R., Pilepich, M., Prasad, K. and Bauer, W.: Irradiation of carcinoma of the prostate localized to the pelvis: analysis of tumor response and prognosis. Int. J. Rad. Oncol. Biol. Phys., 6: 555, 1980. 18. Leibel, S. A., Pino y Torres, J. L. and Order, S. E.: Improved quality of life following radical radiation therapy for early stage carcinoma of the prostate. Urol. Clin. N. Amer., 7: 593, 1980. 19. Lytton, B., Collins, J. T., Weiss, R. M., Schiff, M., Jr., McGuire, E. J. and Livolsi, V.: Results of biopsy after early stage prostatic cancer treatment by implantation of 125 I seeds. J. Urol., 121: 306, 1979. 20. Byar, D. P.: The Veterans Administration Cooperative Urological Research Group's studies of cancer of the prostate. Cancer, 32: 1126, 1973. 21. Cosgrove, M. D. and Kaempf, M. J.: Prostatic cancer revisited. J. Urol., 115: 79, 1976. 22. Dhom, G. and Degro, S.: Therapy of prostatic cancer and histopathologic followup. Prostate, 3: 531, 1982. 23. Byar, D. P.: VACURG studies on prostatic cancer and its treatment. In: Urologic Pathology: The Prostate. Edited by M. Tannenbaum. Philadelphia: Lea & Febiger, part III, chapt. 13, pp. 241-267, 1977. 24. Barnes, R. W. and Ninan, C. A.: Carcinoma of the prostate: biopsy and conservative therapy. J. Urol., 108: 897, 1972.

515

F., Jr.: Radical perinea! 25. Williams, J., Marshall, V. F. and prostatectomy with bilateral v«.rn•cLcvu.,y for carcinoma of the prostate. J. Urol., 113: 380, 1975. 26. Kurth, K. H., Altwein, J. E., Skoluda, D. and Hohenfe!lner, R.: Followup of irradiated prostatic carcinoma by aspiration biopsy. J. Urol., 117: 615, 1977. 27. Nachtsheim, D. A., Jr., McAninch, J. W., Stutzman, R. E. and Goebel, J. L.: Latent residual tumor following external radiotherapy for prostate adenocarcinoma. J. Urol., 120: 312, 1978. 28. Jacobi, G. H. and Hohenfellner, R.: Staging, management and posttreatment reevaluation of prostate cancer: dogma questioned. In: Prostate Cancer. Edited by G. H. Jacobi and R. Hohenfellner. Baltimore: The Williams & Wilkins Co., pp. 31-56, 1982. 29. Schellhammer, P. F., Lagada, L. E. and El-Mahdi, A.: Histological characteristics of prostatic biopsies after 125iodine implantation. J. Urol., 123: 700, 1980. 30. Herr, H. W. and Whitmore, W. F., Jr.: Significance of prostatic biopsies after radiation therapy for carcinoma of the prostate. Prostate, 3: 339, 1982. EDITORIAL COMMENT There is little doubt that patients with a favorable response to x-ray therapy do better in terms of distant metastases than those in whom a favorable response cannot be demonstrated. No convincing data exist that this is a cause-and-effect relationship. Of 30 patients with positive lymph nodes at staging lymphadenectomy 17 (57 per cent) had a positive prostatic biopsy at a later date, compared to 26 of 94 (28 per cent) with negative lymph nodes only. The fact that lymph node status was assessed before radiation and prostatic biopsy subsequently was done suggests that the sensitivity to x-ray therapy, as defined by the ability of x-ray therapy to sterilize the prostate, is greater in patients whose tumor has a lesser metastatic potential. In other words, in patients with an aggressive (metastatic) tumor x-ray therapy is only likely to be 50 per cent or less as effective in treating a local lesion as it is in those tumors that have not demonstrated metastatic potential. Thus, the radioresponsiveness of the primary tumor may be an indicator of the metastatic ability of a particular tumor but the local effect and metastatic potential may be independent factors determining the eventual success or failure of treatment. It also appears that in patients with positive lymph nodes the results of subsequent biopsy are unrelated to the initial clinical stage. Thus, 10 of 18 patients with low stage (A and B) disease and positive lymph nodes had a positive biopsy following x-ray therapy, compared to 7 of 12 with stage C cancer. Patients with stage C disease and negative lymph nodes also had a high percentage of positive biopsy (10 of 22). Thus, in patients with stage C tumor the result of subsequent prostatic biopsy is not related to the status of the lymph nodes before therapy. However, in patients with negative lymph nodes the volume of initial tumor appears to be the most important predictor of eventual success of treatment. Since 50 per cent of the patients with stage C disease and negative lymph nodes, and fewer than 25 per cent of 72) with stages A2 and B2 disease and negative lymph nodes a positive followup biopsy, the conclusion to be drawn is that in those tumors that have not metastasized before the initiation of treatment the most important determinant of eventual local recurrence and/ 01' metastatic disease is the original volume of the tumor treated. To judge truly the effect of x-ray therapy on localized disease it would be most appropriate to examine the relationship between biopsy results, and the probability of local recurrence and metastases in patients with stages A and B disease in whom lymph nodes are negative. Since only 15 per cent of the total patients treated had a subsequent biopsy one wonders whether there is any indication that patients who did not undergo biopsy had a much lesser incidence of local recurrence or distant disease. William R. Fair Department of Urology Memorial Hospital New York, New York