Transrectal ultrasound in the diagnosis of prostate cancer: Location, echogenicity, histopathology, and staging

The Prostate 7:117-129 (1985)

Transrectal Ultrasound in the Diagnosis of Prostate Cancer: Location, Echogenicity, Histopathology, and Staging F. Lee, J.M. Gray, R.D. McLeary, T.R. Meadows, G.H. Kumasaka, G.S. Borlaza, W.H. Straub, F. Lee, Jr., M.H. Solomon, T.A. McHugh, and R.M. Wolf

Departments of Radiology (F. L., R. D. M., G. H.K., G.S. B., W H.S.), Pathology (J.M.G., ZR.M.), and Urology (M.H.S., TA.M.) St. Joseph Mercy Hospital. Department of Urologic Oncology (R. M. W), Roswell Park Memorial Institute, Buffalo, New York, and Boston University School of Medicine, Boston (F. L. Jr.)

Adenocarcinoma of the prostate produces specific ultrasonic findings that can be used in diagnosis. We have examined 211 patients using transrectal ultrasound in both the sagittal and axial planes. Thirty-three carcinomas were detected, and 3 1 histologically confirmed; 24 by needle biopsy, six by transurethral resection, one by total prostatectomy, and two by the demonstration of distant metatases. On ultrasound, all of the carcinomas were less echogenic than normal prostate. All appeared to originate in the peripheral zone of the prostate and produced asymmetry of the gland. The majority of Carcinomas in this series showed capsular involvement and ten penetrated and extended beyond the prostatic capsule.

The results of this series indicate that transrectal ultrasound can be used to detect cancer of the prostate gland. Ultrasound demonstrated the extent of tumor involvement and enabled accurate staging of these cancers.

Key words: prostate, neoplasms, ultrasound studies, cancer staging

INTRODUCTION

Adenocarcinoma of the prostate gland is the second most common cancer in males over the age of 55 [l]. Autopsy series have shown foci of adenocarcinoma within the prostate gland in 30% of men by the age of 50 [2]. With the use of transrectal ultrasonic scanning, the internal structure of the prostate gland can be evaluated [3-171. Normal glandular tissue can be differentiated from hyperplastic and malignant tissue [3-13,15-17,24-261.

By throughly sectioning prostate glands removed at autopsy, it has been shown that the overwhelming majority of carcinomas of the prostate involve the peripheral

Received May 1, 1985; accepted May I , 1985.

Presented at “Ultrasonography of the Prostate” Roswell Park Memorial Institute, Buffalo, NY, July 30, 1985. Sponsored by the OSCC of the Organ Systems Program, NCI.

Address reprint requests to Fred Lee, MD, Department of Radiology, St Joseph Mercy Hospital, 5301 E. Huron River Drive, PO Box 995, Ann Arbor, MI 48106.

0 1985 Alan R. Liss, Inc.

118 Lee et al

T O T A L P A T I E N T S 211 M E A N A G E sfivearr

90 C A N C E R FOUND 33 M E A N AGE fiOyParr = lo01 80

5 70 w 5 60

Lc

P

g 50

: 40 z

z $ 30

20

10

0

1 31 40 41 50 51 60 61 70 7 1 80 81 90

AGE IN VEARS

Fig. I . Age distribution of patient population.

2

l C " 3

atlnq

Non palpahie

Greater fhan l c m

Non Palpable A 2

Greater than 1 Scn

Fig. 2. Staging of prostate cancer.

Transrectal Ultrasound of Prostatic Cancer 119

zone [18,19]. As the tumor grows, it extends into the capsule peripherally and grows into the central zone of the gland [ 18,191. We have used this information to develop new ultrasonic criteria for the location and appearance of carcinoma of the prostate.

MATERIALS AND METHODS

Transrectal ultrasonic examinations were performed on 211 men ranging in age from 31 to 87, with a mean age of 56 (Fig. 1). Eight prostates obtained at autopsy or following radical prostatectomy were examined in a water bath prior to pathologic study. All of the examinations were performed in both sagittal and axial planes, utilizing a Johnson & Johnson Ultrasound model 256 scanner with a 5.0 Mhz transducer for sagittal imaging, and a Bruel and Kjaer model 1846 scanner with either a 4.0- or a 5.5-Mhz transducer for axial imaging.

All examinations were recorded on film and videotape. On ultrasound all tumors were staged according to a modification of the Jewett system [20] (Fig. 2).

Tissue was fixed in neutral-buffered formalin and thin sections were stained with hematoxylin and eosin. Whole mounts were prepared of selected specimens.

RESULTS

The mean age in this series of 211 men was 56 years of age. The mean age for the patients with cancer was 60 years (Table I).

In the initial phase of this study, we biopsied only prostates with echogenic lesions using the criteria for cancer as developed by other authors [4,7,8]. Of these biopsies, four out of 24 contained neoplasm. When we compared our ultrasonic and pathologic findings, we noted that benign hyperplasia tended to be either isoechoic or hyperechoic, and usually was delineated by the surgical capsule. The peripheral zone tissue was characterized by a homogeneous isoechoic echotexture (Fig. 3). A review of the examinations of the four cancer cases showed hypoechoic lesions in the peripheral zone adjacent to the hyperechoic lesion we had biopsied. In addition, the pathology literature indicates that cancer arises in the peripheral zone [ 18,191. At this point, we redefined our criteria and biopsied only hypoechoic lesions located in the peripheral zone obtaining 18 cancers in 26 cases. A biopsy of a local recurrence following radical prostatectomy and biopsy of an extraprostatic mass gave a total of

TABLE I. Summary of Pathologic Findings in Hypoechoic Lesions of Peripheral Zone

Cancer Diagnosed by:

Transperineal biopsy Radical prostatectomy Transurethral resection Distant metastases Total

Benign Focal atypical glandular hyperplasia with prostatitis Prostatitis with leiomyomatous hyperplasia Leiomyomatous hyperplasia No apparent abnormality Total

24 1 6 2

33

Total-all cases 41

120 Lee et al

Fig. 3. Surgical specimen scanned in a water bath showing central benign hyperplasia with hyperechoic foci (B). Note the surgical capsule (arrow). Homogeneous echogenicity is characteristic of the peripheral zone (P). Arrowheads delineate the prostatic capsule. H. head; F. foot; R , right: L. left; ANT, Anterior. A. Sagittal scan. B. Axial scan. C. Whole mount. hematoxylin and eosin. X 3 .

20 cancers. Six of the patients with cancer were diagnosed by transurethral resection, one by radical prostatectomy, and two patients had known distant metastases from the prostate (Table I).

Of these 33 patients with cancer, 25 had enlargement of the gland secondary to the presence of carcinoma and benign prostatic hyperplasia. Most of the patients in this series also showed glandular enlargement secondary to benign prostatic hyperplasia. All of the 33 cases with carcinoma showed asymmetry of the gland secondary to the tumor arising in the peripheral zone. Ten patients showed capsular penetration and extraprostatic extension. Eight patients demonstrated segmental interruptions of the capsule. Fifteen showed subtle scalloped irregularities and thinning of the capsule. The carcinomas all appeared to originate in the peripheral zone of the gland, and


Fig. 4. Stage Bz by ultrasound. Axial scan shows peripheral hypoechoic tumor (T) on the right side with localized thinning of the capsule (arrowheads). The invasion of the adjacent hyperplastic tissue produces a heterogeneous lesion of intermediate echogenicity (*).

involved both the posterior and lateral portions of the peripheral zone with equal frequency. All of the cancers appeared to invade the surgical capsule and extended into the benign hyperplastic area to varying degrees, depending upon tumor size.

In retrospect, all of the carcinomas were primarily hypoechoic (less than normal peripheral zone tissue), and there was no correlation between the echo pattern and histologic grading. All of the patients with cancer had varying degrees of benign prostatic hyperplasia. In the areas of benign prostatic hyperplasia invaded by cancer, ultrasound showed a heterogeneous pattern with echogenicity greater than the tumor itself, but usually less than that of adjacent benign tissue. Four patients had biopsies of this area, and the pathologic findings were a mixture of tumor and benign prostatic hyperplasia (Fig. 4). These tumors of mixed echogenicity, due to benign prostatic hyperplasia invaded by carcinoma, were located in the central zone of the gland (Fig. 5 ) . This would explain the six of the 33 cases that were diagnosed by transurethral resection of the prostate. Two of these were initially thought to be stage A tumors following transurethral resection. However, ultrasound demonstrated a larger extent of tumor, and this was confirmed by needle biopsy. These two tumors were restaged on this basis as B2 (Fig. 6) and C (Fig. 7). The patient with the B2 lesion had a radical prostatectomy at another institution, and the pathologic findings confirmed the ultrasonic staging and location of the tumor.

Four of 13 patients with stage C neoplasms had enlargement of the seminal vesicals on one or both sides, with focal echotexture changes greater than that for normal seminal vesicals and similar to that of the primary tumor. All of these patients had loss of the normal angle between the prostatic capsule and seminal vesical.

Nine of the patients with cancer demonstrated direct tumor extension beyond the gland. These were hypoechoic in nature, similar to the carcinoma within the gland. Biopsy of one of these extracapsular extensions confirmed adenocarcinoma of prostatic origin (Fig. 8). One patient had local recurrence following a radical prostatectomy and this lesion was also hypoechoic.

122 Leeetal

Fig. 5. A. Stage B2 by ultrasound: Axial scan. There is asymmetry of the right lateral lobe with capsular interruption (arrows) by the hypoechoic tumor (T). The invaded benign tissue is central in position and of mixed echogenicity (*). Artifact (A). B. Sagittal scan: hypoechoic tumor (T), mixed echogenic area of invaded benign hyperplasia (*). and hypcrcchoic benign hyperplasia (B). Biopsy needle is in place (arrowheads). Note interruption of the surgical capsule (arrow). C. Benign hyperplastic prostate 3 0 x . D. Adenocarcinomd. histologic grade I . invades muscular stroma 3 0 X .

The prostatic capsule, the seminal vesicals and the intraprostatic extent of neoplasm are well seen with ultrasound, enabling accurate staging to be performed. Cases were staged by ultrasound criteria as either B I (three in number) (Fig. 9), B2 (17) (Fig. 4-6), or C Lesions (13) (Fig. 7, 8). Since all of our patients were referred because of a palpable mass lesion, this was to be expected.


Fig. 6. A. Stage B2 by ultrasound. Axial scan shows the hypoechoic tumor on the left (T) which infiltrates the central zone producing mixed echogenicity (*). Arrow marks site of TUR. Hyperplastic tissue (B), seen on the right side, is hyperechoic. B. Sagittal scan showing biopsy needle (arrowheads) directed toward tumor (T). Benign tissue (B), mixture of tumor invading benign tissue (*).

Fig. 7. Stage C by ultrasound. A. Axial scan shows extraprostatic spread indicating a stage C tumor. TUR defect (D). Note tumor extension (T) along right pararectal region and a mixture of invaded benign tissue and tumor (*). B. Sagittal view demonstrates the longitudinal extension of the hypoechoic tumor (T) along the right margin of the gland.

We estimated the relative volumes of carcinoma and benign tissue in each of 24 needle biopsy specimens. The seven cases biopsied with finger guidance contained an average volume of carcinoma of 40%. The four ultrasound-guided biopsies of lesions of echogenicity greater than normal prostate contained only 25% carcinoma. Ultra- sound-guided biopsies of 11 hypoechoic lesions contained 70% carcinoma. Two hypoechoic extraprostatic masses were also biopsied (Fig. 10).

Eight hypoechoic lesions located in the peripheral zone were benign on biopsy. Five showed a variable mixture of atypical glandular hyperplasia, leiomyomatous hyperplasia, or inflammation. One of the patients with atypical glandular hyperplasia showed borderline changes (Fig. 11). Three had no apparent abnormality (Table I).

124 Lee et a1

Fig. 8. Stage C by ultrasound: Axial (A) and sagittal (B) scan. There is a large right hypoechoic extraprostatic mass (T) adjacent to the rectum. The tumor inside of the gland has the same echogenicity as the tumor outside of the gland. Benign tissue (B) . C. Grade 11 adenocarcinoma virtually replaces extraprostatic connective tissue 3 0 X .

The results of our pathologic studies confirmed with tissue obtained at needle biopsy and total prostatectomy, agree with the following conclusions of previous authors:

(1) origin of carcinoma in the peripheral zone [ 191; (2) spread of the tumor to the capsule, with subsequent invasion and penetration

(3) strands of tumor growth into the adjacent benign hyperplastic tissue 118,191. Ultrasound scanning of these specimens show that the primary tumor is hypo-

echoic, poorly marginated, and sends tumor strands into the adjacent benign tissue, producing a mixed echogenic pattern in the central area (Fig. 12).

[ 18,191;

DISCUSSION

In the past, most authors have employed only axial ultrasonic imaging of the prostate [5-16,21,22]. In this study, we examined the prostate in both axial and sagittal projections as recommended by Rifkin [24]. We believe that this technique increases the sensitivity of the examination.


Y > Mars after

Finger Guided PrOrtateCtOmV

Fig. 9. Stage B , by ultrasound: axial (A) and sagittal (B) scans show a localized hypoechoic peripheral tumor (T) with an asymmetric bulge of the posterior margin and capsular thinning (arrows). C. Grade I adenocarcinoma (T) infiltrates peripheral benign glands (G) 30 X .

Fig. 10. Average volume percent of carcinoma in biopsy specimens.

126 Leeetal

Fig. 1 1 . Hypoechoic peripheral lesion. A. Low-power view demonstrates orderly tubular gland forma- tion in fibromuscular stroma 30 x . B. High-power view shows nuclear anisomorphism (arrows) and nucleoli (arrowhead) 75 X .

The cancers in our series all appeared to involve the peripheral zone. In retrospect, all of the tumors in our series were hypoechoic in echotexture.

Extraprostatic extensions and a local recurrence following radical prostatectomy had a hypoechoic appearance. This finding is expected because cancer in or out of the prostate has the same histologic appearance [18] and should have the same echogenicity. Of the benign hypoechoic lesions, three cases showed atypical glandular hyperplasia, one histologically borderline (Figs. 10,ll). These patients warrant close

Benign tissue, in the majority of cases, was either isoechoic or hyperechoic. A few of the relatively hyperechoic lesions were found to be composed of prostatic hyperplastic tissue with infiltrating carcinoma. Review of the literature suggests that many authors believe that carcinoma is relatively hyperechoic in relation to normal prostate [3,4,6-8,lO-13,16,17,21-241. We believe that the positive biopsies in these

follow-up.


Fig. 12. A. Stage B2 by ultrasound. Radical prostatectomy specimen examined in a water bath showing a peripheral hypoechoic tumor (T) with capsular disruption (arrows). Arrowheads mark needle in prostatic urethra. B,C. Sagittal whole mount of prostate at midline shows central hyperplasia (B), peripheral carcinoma in the posterior lobe (T). Arrows denote areas of capsular disruption. Arrowheads mark urethra. B, Low-power view 3.5X; C, Medium-power view 12X. D. High-power view from center of Fig. C shows tumor strands (T) invading benign tissue 75 X .

cases are due to cancer infiltrating hyperplastic prostatic tissue. Our series indicates that cancers are found less frequently when echogenic lesions are biopsied and these lesions also have a lower ratio of malignant to benign tissue.

Two patients in this series were diagnosed as incidental “stage A” cancer following transurethral resection. Their physical examinations were normal during the postoperative period. Subsequent ultrasound studies were graded as B2 and C, confirmed by transperineal biopsies. One patient had a total prostatectomy, and the pathologic findings confirmed the ultrasonic stage. The staging by transrectal ultrasound caused a change in the treatment of these incidental cancers. The current treatment of incidental “stage A” lesions [25-271 may well be inappropriate since undetected peripheral cancer may be present. Because of this, we recommend that these patients have further evaluation by transrectal ultrasound. If a hypoechoic lesion

128 Leeetal

of the peripheral zone is found, it should be biopsied so that appropriate treatment can be instituted.

CONCLUSIONS

Utilizing both axial and sagittal imaging of the prostate, we believe that it is possible to detect and properly stage carcinoma of the prostate. All of the tumors proven by pathologic examination were hypoechoic in echotexture. Areas of moderate echogenicity were found on pathologic examination to be secondary to ingrowth of tumor into the central area of benign hyperplastic tissue. Our study further questions the validity of diagnosing “stage A” cancer following transurethral resection. It is our opinion that proper staging can best be done by transrectal ultrasound.

Finally, we believe that ultrasound may have the potential to be used as a screening modality for carcinoma of the prostate, but further work is required to verify this.

ACKNOWLEDGMENTS

The authors acknowledge Charles Richison and AM Maher for their technical assistance with the ultrasound studies, Mary Ann Olson for her preparation of the illustrations, and Elizabeth Alamat and Debbra Clemons for their secretarial assistance.

REFERENCES

I . Holleb AI: Ca - A Cancer Journal for Clinicians 35: 19-21, 1985. 2. Rich AR: On the frequency of Occurrence of occult carcinoma of the prostate. J Urol 3215-233,

1935. 3. Rifkin MD, Kurtz AB: Ultrasound of the Prostate. In Sanders RC, Hill MC Ultrasound Annual

4 . Rifkin MD, Kurtz AB, Choi HY, Goldberg BB: Endoscopic ultrasonic evaluation of the prostate using a transrectal probe: Prospective evaluation and acoustic characterization. Rad 149:265-27 1, 1983.

1983. 1983 pp 95-132.

5 . Peeling WB, Griffiths GI: Imaging of the prostate by ultrasound. J Urol 132:217-224, 1984. 6. Spirnak JP, Resnick MI: Transrectal ultrasonography. Urology 23:461467, 1984. 7. Resnick MI, Willard JW, Boyce WH: Ultrasonic evaluation of prostatic nodule. J Urol 120:86-89,

1978. 8. Fritzsche PJ, Axford PD, Ching VC, Rosenquist RW, Moore RJ: Correlation of transrectal

sonographic findings in patients with suspected and unsuspected prostatic disease. J Urol 130:272- 274, 1983.

9 . Okafor PIS, Wild SR, Beynon LL, Chisholm GD: Progress in transrectal ultrasonography for prostatic disease. Br J Urol55:721-725, 1983.

10. Peeling WB, Griffiths GJ, Evans KT, Roberts EE: Diagnosis and staging of prostatic cancer by transrectal ultrasonography. A preliminary study. Br J Urol51:565-569, 1979.

11. Resnick MI, Willard JW, Boyce WH: Recent progress in ultrasonography of the bladder and prostate. J Urol 117:444-447, 1977.

12. Harada K, Tanahashi Y, Igari D, Nurnata I, Orikasa S: Clinical evaluation of inside echo patterns in gray scale prostatic echography. J Urol 124:216-220, 1980.

13. Brooman PJC, Griffiths GJ. Roberts E, Peeling WB, Evans K: Per rectal ultrasound in the investigation of prostatic disease. Clin Rad 32:669-676, 1981.

14. Pontes JE, Ohe H, Watanabe H, Murphy GP: Transrectal ultrasonography of the prostate. Cancer 53:1369-1372, 1984.

15. Watanabe H, Igari D, Tanahashi Y, Harada K, Saitoh M: Transrectal ultrasonotomography of the prostate. J Urol 114:734739, 1975.


16. Harada K, Igari D, Tanahashi Y: Gray scale transrectal ultrasonography of the prostate. J Clin

17. King WW, Wilkiemeyer RM, Boycc WH, McKinney WM: Current Status of prostatic echography.

18. Mostofi FK, Price EB: “Tumors of the Male Genital System.” Washington DC: Armed Forces

19. McNeal JE: Origin and development of carcinoma in the prostate. Cancer 23:24-34, 1969. 20. Catalona WJ, Scott WW: Carcinoma of the prostate: A revicw. J Urol 119: 1-7, 1978. 21. Watanabe H, Date S, Ohe H, Saitoh M, Tanaka S: A survey of 3,000 examinations by transrectal

22. Resnick MI: Evaluation of prostatic carcinoma: Non-invasive and preoperative techniques. The

23. Sekine H, Oka K, Takehara Y: Transrectal longitudinal ultrasonotomography of the prostate by

24. Rifkin MD: Transrectal prostatic ultrasonography : Comparison of linear array and radial scanners.

25. Sheldon CA, Williams RD, Fraley EE: Incidental carcinoma of prostate: A review of the literature

26. Heaney JA, Chang HC, Daly JJ, Prout GR: Prognosis of Clinically undiagnosed prostatic carcinoma

27. Correa RJ Jr., Anderson RG, Gibbons RP, Mason JT: Latent Carcinoma of the prostate-why the

Ultrasound 7:4549, 1979.

JAMA 226:444447, 1973.

Institute of Pathology, pp. 196-252, 1973.

ultrasonotomography . The Prostate I :27 1-278, 1980.

Prostate 1:311-320, 1980.

electronic linear scanning. J Urol 127:62-65, 1982.

J Ultrasound Med 4: 1-5, 1985.

and critical reappraisal of classification. J Urol 124:626-63 I , 1980.

and the influence of endocrine therapy. J Urol Il8:283-287, 1977.

controversy? J Urol 11 1 :644646. 1974.

Documents

Transrectal ultrasound in the diagnosis of prostate cancer: Location, echogenicity, histopathology, and staging