2300

Early Detection of Local Disease

Gang Zhang, MD,*,t Neil F. Wasserman, MD,S Deepak A. Kapoor, MD,**t and Pratap K. Reddy, MD*,t

Fifty-two patients with Stage A1 prostate cancer diag- nosed by transurethral resection performed between 1975 and 1989 were re-examined by transrectal ultraso- nography and ultrasonographically guided biopsies. Fol- low-up after the initial diagnosis ranged from 1 to 15 years (mean, 5.8 years). For eight patients, results of digi- tal rectal examination were abnormal. For 44 patients, results were normal or indicated a low probability of cancer. Serum prostate-specific antigen (PSA) levels (4.6 to 14.6 ng/ml) were elevated in ten patients. Ultrasonogra- phy showed from one to three hypoechoic areas in 29 patients. Locally progressive disease, defined as moder- ately to poorly differentiated cancer, was detected in five (10%) patients, three of whom underwent radical prosta- tectomy. Histopathologic examination of the specimens revealed localized disease (no capsular invasion). The re- maining two patients had radiation therapy. In three pa- tients, results of digital rectal examination and the serum PSA level were normal, but focal, well-differentiated cancer, identical to that initially diagnosed, was detected after a follow-up of 5 to 10 years. Because the clinical significance of this finding is unknown, these three pa- tients were not considered to have progressive disease and did not have additional treatment. Our data suggest that transrectal ultrasonography is valuable in early de- tection of local disease progression and should be used in the follow-up program for patients with Stage A1 pros- tate cancer. Cancer 1992; 69:2300-2305.

The management of Stage A1 adenocarcinoma of the prostate is a matter of controversy. Previous studies

From the *Department of Urologic Surgery, University of Min- nesota Hospital and Clinic, Minneapolis, and the tUrology Section and *Department of Radiology, Veterans Affairs Medical Center, Min- neapolis, Minnesota

Work performed at the Veterans Affairs Medical Center, Minne- apolis, Minnesota.

Address for reprints: Gang Zhang, MD, Urology Section, 112D, Veterans Affairs Medical Center, One Veterans Drive, Minneapolis, MN 55417.

Accepted for publication June 18, 1991.

have shown that patients with Stage A1 disease are at low risk for developing progressive disease and should be managed conservatively without additional inter- vention. 1-3 However, recent studies have indicated that 16% to 27% of patients with Stage A1 disease eventu- ally may develop locally or systemically progressive d i ~ e a s e . ~ , ~ In addition, no reports in the literature have addressed the results of definitive treatment for pros- tate cancer that has progressed locally from an initially diagnosed Stage A1 disease. We recently reported our results regarding locally or systemically progressive dis- ease in 13 of 132 patients with Stage A1 prostate cancer after long-term follow-up.6 We also reported the results of definitive treatment for locally progressive prostate cancer in patients with Stage A1 disease who had been managed expectantly.6 Our study suggests that most patients with locally progressive disease can be saved by definitive treatment if the local progression is de- tected at an early stage. However, a few of our patients had advanced disease and are unlikely to be cured by definitive therapy. Obviously, conventional follow-up methods are not able to detect all cases of progression at an early stage.

Early detection of prostate cancer has been re- ported using transrectal ultrasonography of the pros- tate and ultrasonographically guided b i ~ p s i e s . ~ - ~ How- ever, few studies address the value of transrectal ultra- sonography for early detection of local disease progression in patients with Stage A1 prostate cancer. Although the use of transrectal ultrasonography as a screening test for the general population of men older than 55 years of age is its use in the follow-up program for early detection of local disease progression in patients with Stage A1 disease seems reasonable, given the reported high risk for local or sys- temic progression from Stage A1 disease. We report our results on the use of transrectal ultrasonography and ultrasonographically guided biopsies for the early de- tection of local disease progression.

Transrectal Ultrasonography for Prostate CancerlZhang et al. 2301

Materials and Methods

Fifty-two patients in whom Stage A1 prostate cancer was diagnosed by a transurethral resection of the pros- tate performed between 1975 and 1989 were re-exam- ined by transrectal ultrasonography of the prostate and ultrasonographically guided biopsies. The length of fol- low-up after transurethral resection of the prostate ranged from 1 to 15 years (mean, 5.8 years). Ultrasonog- raphy was performed on patients for whom there was a clinical suggestion of local progression or on patients who were less than 75 years of age with a reasonable life expectancy, during which they would be at risk for dying of prostate cancer. The mean age of the patients who had transrectal ultrasonography was 68 years. All but one patient had a transurethral prostatectomy and were followed up at the Veterans Affairs Medical Center, Minneapolis, Minnesota.

Follow-up initially consisted of a measurement of serum prostatic acid phosphatase and digital rectal ex- amination repeated at intervals of 6 months to 1 year and a bone scan annually. More recently, measurement of prostatic specific antigen (PSA) became a routine test, and a bone scan is no longer routinely performed.

Stage A1 prostate cancer was defined as that hav- ing a Gleason score of 1 4 in less than 5% of the tissue removed during transurethral resection of the prostate for benign prostatic hyperplasia. Local progression was defined as detection of moderately to poorly diff eren- tiated prostate cancer with or without abnormal results for a rectal examination of the prostate. All patients were tested for the level of PSA and serum prostatic acid phosphatase and had digital rectal examination before transrectal ultrasonography. The ultrasonogra- phy machine used for this study was a Toshiba model SSA 100A (Toshiba, Schaumburg, IL) with a biplane probe equipped with 5-MHz distal radial and proximal linear transducers. Transrectal biopsies were taken in the longitudinal plane. During the ultrasonography procedure, transrectal core needle biopsies were taken with the Biopty gun (Bard, Covington, GA) from spe- cific areas demonstrated to be abnormal (usually hy- poechoic) by ultrasonography. In addition, six to eight ultrasonographically guided fractionated (or random) biopsy samples were obtained from the anterior, mid- dle, and posterior right and left sides of the gland. In cases in which the prostate appeared normal ultrasono- graphically, only fractionated biopsies were taken. In all cases, the biopsies were preceded by local infiltration of 1% lidocaine, and antibiotic therapy was given.

Results

The results of digital rectal examination of the prostate were abnormal for 8 of 52 patients; 4 had nodules and 4

had irregular induration that was considered to have a moderate to high probability of being prostate cancer. For the remaining 44 patients, results were normal or indicated a low probability of prostate cancer. Elevated serum PSA levels of from 4.6 to 14.6 pg/dl (normal range, 0 to 4 pg/dl using the Hybritech assay [Hybri- tech Inc., San Diego, CAI) were detected in ten patients.

Transrectal ultrasonography indicated abnormal echo patterns, primarily hypoechoic areas ranging from one to three, in the prostate in 29 (58%) patients. For each area that was suggestive of disease, a biopsy was done and additional fractionated biopsy samples were taken from these 29 patients. In the remaining 23 pa- tients, fractionated biopsy samples were taken. All pa- tients tolerated the procedure well, and no complica- tions occurred.

Local disease progression was verified by histopath- ologic examination of biopsy specimens in five patients. One of these patients had transurethral resection of the prostate at another hospital. In two patients with pro- gressive disease, nodules had been detected by digital rectal examination; in the remaining three patients, the results of digital rectal examination had been normal. Slightly elevated levels of PSA (4.6 to 5.6 ng/ml) were detected in three of the five patients; the levels were normal (0.5 pg/dl and 3.2 &dl, respectively) in the other two patients. All five patients with locally progres- sive disease had definitive treatment (radical prostatec- tomy in three and radiation therapy in two) for what was considered early disease. All three patients who had radical prostatectomy had pathologic Stage B dis- ease (no capsular invasion). Detailed information on these five patients appears in Table 1.

In addition to these five patients, three patients for whom the results of digital rectal examination and serum PSA levels were normal were found by transrec- tal ultrasonographically guided biopsies to have foci of well-diff erentiated cancer identical to that initially diagnosed as Stage A1 disease 5 to 10 years earlier. Because the clinical significance of this finding is un- known, these three patients were not considered to have progressive disease and did not have additional treatment. Repeat ultrasonographically guided biopsies were performed on one of these three patients and showed no cancer.

In all five patients with locally progressive disease, the hypoechoic areas demonstrated by ultrasonogra- phy were shown by biopsy to be cancerous. The frac- tionated biopsy samples taken from the ipsilateral side in the area adjacent to the hypoechoic area also demon- strated cancer in all but one patient, in whom the hy- poechoic area contained high-grade tumor (Gleason score 3 of 4). Of the other three patients who had nor- mal results for the digital rectal examination and nor-

Table 1. Course of Patients With Local Progression Detected by Transrectal Ultrasonography Patient no., age at PSA, and results diagnosis of Gleason of rectal progression grade, percent examination (yr) Year of TURP cancerous tissue before US US findings Biopsy results Treatment Pathologic type Follow-up status

1, 75 1982 11/11, 3 5.6 d d l , normal prostate

2, 69 1983 I/I, 1 0.5 p g / 4 nodule (clinical 81)

3, 75 1983 11/11, 3 4.6 pg/dl, nodule (clinical B1)

4, 68 1987 11/11, 3 3.2 pg/dl, (Another normal hospital) prostate

5, 72 1988 11/11, 1 5.0 pg/dL normal prostate

Discrete hypoechoic area in peripheral zone

Hypoechoic area in peripheral zone

Two hypoechoic areas in peripheral zone

Two hypoechoic areas in peripheral zone

area in peripheral zone

One hypoechoic

Hypoechoic area: Cap III/IV, adjacent tissue random biopsy negative

Hypoechoic area: CaP II/III, adjacent tissue random biopsy CaP 11/11

Hypoechoic area: Cap lll/IlI, adjacent tissue random biopsy CaP II/III

Hypoechoic area and adjacent random biopsy CaP II/III

Hypoechoic area and adjacent random biopsy CaP II/III

RRP Jan. 1990

RRP Oct. 1989

Radiation Oct. 1989

RRP May 1990

Radiation Jan. 1990

CaP IV/III, no capsule invasion

CaP II/III, no capsule invasion

CaP II/III, no capsule invasion

NED, PSA 0.2 pg/dl

NED, PSA 0.2

NED, PSA 0.4 pg/dl

NED, PSA 0.2

NED, PSA 0.3 pg/dl

TURP: transurethral resection of prostate; US: transrectal ultrasonography; Cap: adenocarcinoma of the prostate; RRP: radical retropubic prostatectomy; NED: no evidence of disease; PSA: prostatic-specific antigen.

Transrectal Ultrasonography for Prostate CancerlZhang et al. 2303

ma1 PSA values and were not considered to have pro- gressive disease, two had small hypoechoic areas that were determined by biopsy to contain foci of well-dif- ferentiated cancer. For the other patient, the results of ultrasonography were normal, but fractionated biopsy samples identified a focus of well-diff erentiated tumor. Repeat ultrasonography and biopsy failed to show cancer in this patient. Thus, the positive predictive value of transrectal ultrasonography for early detection of prostate cancer in this series is 27% (8 patients with cancer).

Discussion

Most patients with Stage A1 prostate cancer have a life expectancy comparable to that for men of the same age who do not have prostate cancer, but 10% to 15% of these patients may develop locally or systemically pro- gressive disease.'r6 The controversy regarding the man- agement of Stage A1 prostate cancer, especially in rela- tively young patients, centers around the following is- sues. Does Stage A1 disease, if untreated, become locally progressive before it becomes metastatic? If a close clinical follow-up program that incorporates more sensitive diagnostic tests (such as measurement of serum PSA and transrectal ultrasonography) is used, can locally progressive disease be detected early enough to save patients by definitive treatment? Is transrectal ultrasonography and ultrasonographically guided biopsy indicated during follow-up of patients with Stage A1 disease and, if so, how frequently should the procedure be performed?

In a report on the histopathologic study of 104 radi- cal prostatectomy specimens, McNeal et ~ 1 . ' ~ reported that 68% of the prostate cancers arose in the peripheral zone, 24% arose in the transition zone, and 8% arose in the central zone. Only 6% of Stage B cancers arising from the peripheral zone invaded the transition zone, and only 10% of Stage A cancers arising from the tran- sition zone invaded the peripheral zone. Stamey12 has postulated that the compressed fibromuscular bound- ary between the transition zone and the peripheral zone acts as a barrier to the invasion of Stage A cancer into the peripheral zone. He also speculates that the separa- tion of the transition and peripheral zones explains why some large Stage B and C prostate cancers detected by digital rectal examination may be associated with be- nign tissue when transurethral prostatectomy is per- formed, and large clinical Stage A2 cancers arising from the transition zone and detected by transurethral resec- tion of the prostate often are associated with normal results of digital rectal examination and negative find- ings on prostate biopsy when the needles are directed into the peripheral zone.12 If these observations are

correct, local progression from Stage A1 prostate cancer can arise in only three ways: from residual cancer in the remaining transition zone, which may involve the pe- ripheral zone after the compressed fibromuscular barrier between the transition and peripheral zones is violated by transurethral resection of the prostate; from multicentric cancers in the peripheral or central zones, which were not detected at the time of transurethral resection of the prostate; and from new cancers :hat develop in the peripheral zone, which may explain why local progression develops many years after the initial diagnosis of Stage A1 disease in some patients.

Local progression in any of these three cases proba- bly would involve the peripheral zone, from which dis- ease can be detected by digital rectal examination or transrectal ultrasonography. If a cancer is located ante- riomedially, as is sometimes the case with Stage A pros- tate cancer,14 transrectal ultrasonography would be es- pecially useful for early detection of disease progres- sion. Studies of patients with untreated prostate cancers who had been followed up for long periods showed that in most cases local disease progression preceded distant metastases.'',16

A major concern regarding close, long-term follow- up for patients whose Stage A1 disease is treated expec- tantly is that locally progressive disease may go unno- ticed until it reaches an advanced stage. However, in our experience with ten patients in whom locally pro- gressive disease developed from Stage A1 prostate cancer and who then underwent definitive therapy, eight had localized disease and a good chance for a cure.6 In the current study, all five patients in whom local progression was detected were considered to have early disease. All three patients who had radical prosta- tectomy had pathologically localized disease with no capsular invasion. The 10% incidence of progressive disease that we report here is the same as the 10% inci- dence of progressive disease that we have previously reported among 132 patients with Stage A1 prostate cancer who underwent long-term follow-up.6 Ultraso- nography demonstrated its ability to detect locally pro- gressive disease at an early stage in which the cancer was confined within the prostate gland. Thus, the use of ultrasonography potentially could improve the chances of survival for patients with Stage A1 disease who develop progressive disease. These data suggest that expectant therapy with close follow-up (or active surveillance) for patients with Stage A1 prostate cancer, even relatively young patients, would detect and save most patients with 1-1 disease progression while spar- ing most patients unnecessary definitive treatment.

The estimated probability that a male in the United States will be diagnosed with prostate cancer during his lifetime is 8.8Y0.l~ Prostate cancer is detected by

rb

2304 CANCER M a y 2 , 2992, Volume 69, No. 9

transrectal ultrasonography in patients whose prostate appears normal by digital rectal examination in 4.7% to 5.2% of patients who have the pr~cedure.’~,’~ In a large screening trial of 784 self-referred men, digital rectal examination and transrectal ultrasonography detected early cancer in 2.8% of the men.8 The value and cost- effectiveness of a screening transrectal ultrasonography for the general population of men older than 55 years of age is debatable. However, considering the higher risk for progressive disease and the higher prevalence of detectable cancer in patients with Stage A1 cancer than in the general population, routine use of transrectal ul- trasonography of the prostate as part of regular follow- up seems reasonable. Transrectal ultrasonography is not for the detection of all prostate cancers, but only for the detection of clinically significant or potentially lethal cancers, which we think includes moderately to poorly differentiated cancers.

The positive predictive value of transrectal ultraso- nography is reported to be from 17% to 36Y0.l~ Cooner et a1.” initially reported positive biopsy results in 28 of 96 (29%) patients with areas that were suggestive of disease on ultrasonography . These authors recently re- ported updated results in which biopsy results were pos- itive in 30% of 835 patients in whom transrectal ultraso- nography showed abnormal areas in the prostate.’l Lee et a L s reported positive results for 31% of biopsy sam- ples taken from hypoechoic areas.

In our series, ultrasonography demonstrated hy- poechoic areas in the prostate in all but one of the eight patients in whom prostate cancer was detected. In addi- tion, cancer was found in fractionated biopsy samples taken from the ipsilateral lobe in four of five patients with locally progressive disease. However, in 23 pa- tients with a normal echo pattern, fractionated biopsy samples showed focal well-differentiated cancer in only 1 patient in whom repeat biopsy failed to show cancer. Thus, biopsy results were positive in 7 of 29 patients (24%) in whom transrectal ultrasonography showed abnormal echo patterns. These results suggest that fractionated (or random) biopsy samples of pros- tate tissue exhibiting a normal isoechoic pattern give little positive yield and should not be taken routinely. However, biopsy samples from the area exhibiting an abnormal echoic pattern and fractionated biopsy sam- ples from adjacent areas should be taken if abnormal echoic patterns are demonstrated.

If transrectal ultrasonography is valuable in the early detection of locally progressive disease from Stage A1 cancer, how often should this procedure be per- formed and how often should biopsies be done? Small, clinically detectable prostate cancers have a slow dou- bling time. Maksimovic and Schoeder” and Stamey et ~ 1 . ‘ ~ recently have observed that the doubling time for

prostate cancer is about 2 years. Thus, transrectal ultra- sonography at intervals of 1 to 2 years and digital rectal examination and measurement of serum PSA at inter- vals of 6 months to 1 year seems a reasonable follow-up program for relatively young patients with a life expec- tancy of more than 10 years.

Summary

Of 52 patients with previously diagnosed Stage A1 pros- tate cancer who had transrectal ultrasonography and ultrasonographically guided biopsies, 5 (10%) were found to have early local disease progression and had definitive treatment. In all three patients who had radi- cal prostatectomy, histopathologic examination re- vealed localized disease with no invasion to the cap- sule. This 10% incidence of local progression detected by ultrasonography is the same as that which we re- ported for a series of 132 patients who were followed up for a long period of time. Whether 10% represents the true incidence of progressive disease in patients with Stage A1 disease remains to be confirmed. Our data support the hypothesis that the routine use of transrectal ultrasonography of the prostate in the close clinical follow-up of patients with Stage A1 prostate cancer improves detection of early locally progressive cancer and potentially may result in improved survival.

References

1. Byar DB and The Veterans Administration Cooperative Urologi- cal Research Group. Survival of patients with incidentally found microscopic cancer of the prostate: Results of a clinical trial of conservative treatment. ] Urol 1972; 108:908-913.

2. McNeal JE. Origin and development of carcinoma in the pros- tate. Cancer 1969; 23:24-34.

3. Scott R Jr, Mutchnik DL, Laskowski TZ, Schmalhorst WR. Carci- noma of the prostate in elderly men: Incidence, growth charac- teristics and clinical significance. ] Urol 1969; 101:602-607.

4. Blute ML, Zincke H, Farrow GM. Long-term follow-up of young patients with Stage A adenocarcinoma of the prostate. ] UroI

5. Epstein JI, Paul1 G, Eggleston JC, Walsh PC. Prognosis of un- treated Stage A1 prostatic carcinoma: A study of 94 cases with extended follow-up. Urol 1986; 136537-839.

6. Zhang G, Wasserman NF, Reddy PK. Is conservative manage- ment of Stage A1 adenocarcinoma of the prostate warranted? Outcome of 206 patients with Stage A1 disease (Abstr). l lrol 1990; 143:314A.

7. Cooner WH, Eggers GW, Lichtenstein P. Prostate cancer: New hope for early diagnosis. Ala Med 1987; 56:13-16.

8. Lee F, Littrup PJ, Torp-Pedersen ST et a/ . Prostate cancer: Com- parison of transrectal US and digital rectal examination for screening. Radiology 1988; 168:389-394.

9. Cohen JM, Resnick MI. The use of transrectal ultrasonography in the diagnosis of Stage A prostatic carcinoma. World Urol

1986; 136:840-843.

1983; 1112-14.

Transrectal Ultrasonography for Prostate Cancer/Zhang e t al. 2305

10. Scardino PT. Early detection of prostate cancer. Urol Clin North Am 1989; 16:635-655.

11. Chodak GW. Screening for prostate cancer: Role of ultrasonogra- phy. Urol Clin North Am 1989; 16:657-661.

12. Stamey TA. Prostate cancer: Some basic clinical and morphomet- ric observations. Monogr Urol 1989; 10:79.

13. McNeal JE, Redwine EA, Freiha FS, Stamey TA. Zonal distribu- tion of prostatic carcinoma: Correlation with histologic pattern and direction of spread. Am ] Surg Pathol 1988; 125397-906.

14. McNeal JE, Price HM, Redwine EA, Freiha FS, Stamey TA. Stage A versus Stage B adenocarcinoma of the prostate: Morphologcal comparison and biological significance. J Urol 1988; 139:61-65.

15. George NJR. Natural history of localized prostatic cancer man- aged by conservative therapy alone. Lancet 1988; 1:494-497.

16. Handley R, Cam TW, Travis D, Powell PH, Hall RR. Deferred treatment for prostate cancer. Br J Urol 1988; 62249-253.

17. Hodge KK, McNeal JE, Stamey TA. Ultrasound guided transrec- tal core biopsies of the palpably abnormal prostate. ] Urol 1989;

18. Cooner WH, Mosley BR, Rutherford CL Jr ef al. Coordination of 142:66-70.

urosonography and prostate-specific antigen in the diagnosis of nonpalpable prostate cancer. ] Endourol 1989; 3:193-199.

19. Chodak GW, Schoenberg HW. Progress and problems in screening for carcinoma of the prostate. World J Surg 1989;

20. Cooner WH, Mosley BR, Rutherford CL Jr et ai. Clinical applica- tion of transrectal ultrasonography and prostate specific antigen in the search for prostate cancer. ] Urol 1988; 139:758-761. Cooner WH, Mosley BR, Rutherford CL Jr et al . Prostate cancer detection in a clinical urological practice by ultrasonography, digital rectal examination and prostate specific antigen. Urol

22. Maksimovic P, Schoeder FH. Transrectal ultrasonometry of the prostate and of hypoechoic lesions in 6 untreated and 6 treated patients with locally confined prostatic carcinoma (Abstr). J Urol 1989; 141:512A.

23. Stamey TA, Kabalin JN. Prostate specific antigen in the diagno- sis and treatment of adenocarcinoma of the prostate. I: Un- treated patients. J Urol 1989; 141:1070-1075.

13 :60-64.

21.

1990; 143:1146-1154.