The Significance of Isoechoic Prostatic Carcinoma

0022-534 7/94/1526-2304$03.00/0

THE JOURNAL OF UROLOGY Copyright© 1994 by AMERICAN UROLOGICAL AssoCIATION, INC.

Vol. 152, 2304-2307, December 1994 Printed in U.S.A.

THE SIGNIFICANCE OF ISOECHOIC PROSTATIC CARCINOMA WILLIAM J. ELLIS*

AND

MICHAEL K. BRAWER

From the Department of Urology, University of Washington and Section of Urology, Seattle Veterans Affairs Medical Center, Seattle, Washington

ABSTRACT

The diagnosis of prostatic carcinoma is most commonly made today by transrectal ultrasound guided needle biopsy. Often hypoechoic and peripheral zone lesions are the only areas sampled. Recently, we showed that this approach missed a quarter of the cancers that would be detected by a systematic biopsy technique. We term these missed cancers isoechoic carcinomas. We reviewed 1,549 systematic sextant prostate needle biopsies, of which 417 cancers were detected and subdivided into hypoechoic cancers (cancers detected on biopsy of a hypoechoic sector and isoechoic cancers (cancers found only in normal [isoechoic] peripheral zone). We noted in men with only isoechoic cancers that fewer biopsy cores per prostate revealed cancer (mean 1.6 versus 3.0, p <0.0001) and that these men had lower serum prostate specific antigen levels (mean 14.4 versus 43.7, p <0.001). The Gleason scores for the isoechoic and hypoechoic cancers were indistinguishable. The pathological staging ofhypoechoic and isoechoic cancers was also similar. This study suggests that while isoechoic cancers are generally smaller than hypoechoic cancers, they do not represent low grade clinically insignificant carcinomas. A systematic approach to performing prostate biopsy is recommended. KEY WORDS: prostatic neoplasms, ultrasonography, biopsy

While the purpose of this special edition is to honor Doctor Grayhack, I (W. J.E.) in fact am the one honored by being allowed to participate in such a tribute. The contributions of Doctor Grayhack in the field of urology are many, and I am sure reflected throughout these pages. Yet the many honors bestowed upon him are not the primary source of my admiration for Doctor Grayhack. Jack Grayhack's integrity is what earned my respect long ago. He may be the most honest man I know. His constant teaching to us over the years was "when in doubt do the right thing." Yet it would not surprise me ifhe was unaware of this legacy, for his teaching was with his actions not his words. I am truly grateful to have associated with such a man.

were reviewed. All patients without a previous diagnosis of prostate cancer who underwent systematic 6-sector biopsy with documentation of ultrasonographic findings on a sector by sector basis were included in this analysis. The indications for prostate needle biopsy included an abnormal digital rectal examination, serum PSA level greater than 4.0 ng./ml., annual 20% or greater increase in serum PSA with an absolute level of greater than 1.5 ng./ml. or entry into an investigational treatment protocol for benign prostatic hyperplasia. Many of the men were participants in our ongoing PSA based screening study, including those biopsied for a 20% increase in serum PSA. 6· 7 Procedures. Serum PSA levels were determined by the The diagnosis of prostate carcinoma is being made with Hybritech Tandem-E or Tandem-R assays, or the Abbott IMx increasing frequency,1 commonly by prostate needle biopsy assay. Digital rectal examination was performed by an atperformed under transrectal ultrasound guidance. The role tending urologist or senior resident, and scored as 0-normal of transrectal ultrasound in the detection of prostate carci- examination, 1-gland asymmetric but smooth, 2-mild noma is controversial. Initially, prostatic carcinoma was to moderate induration and 3-glands believed to contain thought to produce a hypoechoic echo pattern on ultrasound.2 cancer. Informed consent was obtained from all patients. Patients Subsequent studies suggested that many carcinomas are found in areas that do not appear hypoechoic on ultrasonic received antibiotic prophylaxis consisting of 500 mg. ciproexamination. 3 • 4 We recently reported on a study demonstrat- floxacin orally for 3 doses with the first dose 1 hour before the ing that prostate specific antigen (PSA) is superior to trans- procedure and 80 mg. gentamicin intramuscularly. A Fleets rectal ultrasound in predicting which patients will harbor enema was administered before ultrasound. Transrectal ulprostatic carcinoma, since many prostate cancers are iso- trasound was performed by an attending physician or senior echoic and not detected by ultrasound. 5 In that study we resident using the Bruel & Kjar model 1846 ultrasound noted that a quarter of all cancers detected would have been equipped with a 7.0 MHz. end fire transducer. Systematic missed had only hypoechoic sectors been biopsied. In this 6-sector prostate needle biopsies were obtained using a modstudy we evaluate whether there are differences between ification of the technique of Hodge et al 8 as previously disprostate carcinomas that are hypoechoic or isoechoic on trans- cussed.6 Each sector was described as hypoechoic, if any rectal ultrasound, and compare those prostate carcinomas hypoechoic lesions were noted within the sector, or isoechoic. that were detected only by biopsy of isoechoic sectors with When a hypoechoic lesion was noted within the biopsy sector the biopsy needle was directed toward the hypoechoic lesion. those that were detected in hypoechoic sectors. If no hypoechoic lesion was noted in the sector then the biopsy was directed to sample maximally the peripheral zone MATERIALS AND METHODS within the sector. Patient population. The medical records of patients who Pathological examination of biopsies from each sector was underwent transrectal ultrasound with prostate needle bi- performed separately. Carcinoma was graded according to opsy at our institutions between April 1989 and June 1994 the criteria of Gleason. 9 Gleason scores were calculated by adding the major and minor histological pattern. For the * Requests for reprints: Section of Urology, 112UR, 1660 S. Columbian Way, Seattle, Washington 99108. purposes of this analysis prostatic intraepithelial neoplasia 2304

2305

ISOECHOIC PROSTr\ rIC Ci\RCil'·JOlVlA 1

TABLE L

Rt. base Rt. mid Rt. apex Lt. base Lt. mid Lt. apex Totals

Individual sector analysis biopsy results and ultrasound findings

No. Hypoechoic/No. Biopsy

No. Ca/No. Biopsy

No. Hypoechoic Ca/

(%)

(%)

No. Hypoechoic (%)

704/1,549 (45.4) 460/1,549 (29. 7) 492/1,549 (31.8) 654/1,549 (42.2) 483/1,549 (31.2) 507/1 549 (32. 7) 3,300/9,294 (35.5)

174/1,549 (11.2) 184/1,549 (11.8) 165/1,549 (10.6) 213/1,549 (13. 7) 202/1,549 (13.0) 17 4/1 549 (11.2) l,112/9,294 (12.0)

and atypia were categorized as benign. Radical prostatectomy specimens were step sectioned from whole mounts for histological examination. The cases of cancer identified were divided into hypoechoic and isoechoic. A hypoechoic carcinoma was defined as any carcinoma that would have been detected by biopsy of only hypoechoic sectors. Thus, a cancer that was present in hypoechoic and isoechoic sectors of the prostate was classified as hypoechoic. Cancer detected only in sonographically normal peripheral zone was classified as isoechoic. The number of positive biopsy cores, serum PSA levels and Gleason scores were compared using Mann-Whitney nonparametric analysis on software. Nonparametric analysis was chosen since much of our data did not fit a normal distribution.

121/704 103/460 81/492 135/654 111/483 91/507 642/3,300

TABLE

No. Isoechoic Ca/ No. Isoechoic (%)

(17.1) (22.3)

53/845 81/1,089 84/1,057 78/895 91/1,066 93/1,042 470/5,994

(16.4)

(20.6) (23.0) (17.9) (19.5)

(6.3) (7.4) (7.7) (8. 7) (8.5) (8.9)

(7.8)

2. Serum PSA levels in patients with hypoechoic and isoechoic cancers

Hypoechoic Ca Isoechoic Ca

No.

MeanPSA

SD

Median PSA

p Value

253 92

43.7 14.4

205.6 52.2

8.6 6.1

>0.001

30 - - - - - - - - - - - - - - - - - - - - - - -

1111 HYPOECHOIC CANCERS

D

ISOECHOIC CANCERS

RESULTS

A total of 1,549 biopsies was performed on 1,414 patients, of whom 111 men underwent 2 biopsies and 12 underwent 3. For study purposes each biopsy is analyzed separately. Indications for repeat biopsy included a change in or markedly abnormal digital rectal examination, increasing serum PSA level, unexplained, persistently high serum PSA level (greater than 10 ng./ml.), or previous biopsy showing atypia or prostatic intraepithelial neoplasia. Prostatic carcinoma was detected in 41'7 of the 1,549 biopsies (26.9%). Each of the 6 peripheral zone sectors was categorized separately based on transrectal ultrasound findings (hypoechoic or isoechoic) and biopsy results (table 1). Approximately a third of the sectors were believed to contain a hypoechoic lesion (note that hypoechoic lesions are more often described by the urologist at the base of the prostate). Cancers were

80

60

j

2

4

3

0

D

ISOECHOIC CANCERS

0

2

3

4

5

6

NUMBER POSITIVE CORES

FIG. 1. Histogram shows number of cores containing cancer for isoechoic and hypoechoic cancers.

8

10

9

SCORE

3. Comparison of Gleason score of hypoechoic and isoechoic cancers

Hypoechoic Ca Isoechoic Ca

20

7

FIG. 2. Histogram shows distribution of Gleason scores in hypoechoic and isoechoic cancers.

No.

HYPOECHOIC CANCERS

6

GLEASON

TABLE

Ill

5

275

93

Mean Gleason Score

SD

Median Score

6.9 6.6

1.8 1.9

7.0 7.0

p Value 0.2216

found in 12% of biopsy specimens, and were well distributed within the gland. Evaluating each biopsy core separately, cancers were more likely to be found in hypoechoic sectors than in isoechoic sectors by a 2.5:1 ratio. The prostate cancers detected were divided into 2 groups based on whether the prostate cancer was detected in a hypoechoic sector, with or without cancer in other sectors (306 cases), or only in isoechoic sectors (111). Figure 1 demonstrates that isoechoic carcinomas were commonly found in only 1 core of the sextant biopsy. The mean number of cores containing cancer was 1.6 and 3.0, respectively, for isoechoic and hypoechoic cancers (p <0.0001). Serum PSA levels were compared in the 345 men with pre-biopsy determinations. Hypoechoic cancers were associated with higher serum PSA levels (table 2, p <0.001). Gleason scores for the tumors when available are shown in figure 2. There was no statistical difference in the Gleason scores between isoechoic and hypoechoic cancers (table 3, p = 0.2216).

2306

ISOECHOIC PROSTATIC CARCINOMA TABLE

No. hypoechoic Ca(%) No. isoechoic Ca(%)

4. Pathological findings in the 107 patients known to have surgical staging

Organ Confined

Capsular Perforation With Neg. Margins

Pos. Margin as Evidenced by Tumor Touching Ink

Seminal Vesicle Invasion

Lymph Node Involvement

22 (29.3) 14 (43.8)

17 (22.7) 5 (15.6)

22 (29.3) 8 (25.0)

5 (6.7) 1 (3.1)

9 (12.0) 4 (12.5)

Of the 417 patients with prostate cancer 107 (25.6%) underwent surgical staging, most often by radical prostatectomy. The pathological staging distribution is shown in table 4. Combining the organ confined and Cl (capsular perforation with negative margins) groups produced a similar distribution between the 2 populations. DISCUSSION

Before the widespread introduction of transrectal ultrasound into urological practice Lee et al reported that the most common ultrasonographic appearance of prostatic carcinoma was the hypoechoic peripheral zone lesion. 2 In subsequent prostate cancer detection studies biopsies were directed only toward hypoechoic lesions of the prostate, 1°, 11 The detection bias in these studies erroneously inflated the ability of transrectal ultrasound to detect prostate carcinoma. Salo 3 and Dahnert4 et al were among the first to suggest that many prostate carcinomas were, in fact, isoechoic. In a later study of ultrasonic examinations of the prostate performed before radical prostatectomy Carter et al found that only 54% of prostatic cancers identified pathologically on the nonsuspicious side of the prostate were detected by ultrasound. 12 We recently reviewed 1,001, 6-sector prostate needle biopsies and compared the performance of digital rectal examination, transrectal ultrasound and serum PSA in the diagnosis of cancer. 5 Serum PSA proved to be superior to digital rectal examination and transrectal ultrasound. Interestingly, we noted that a strategy of biopsying only hypoechoic lesions, which is standard practice for many urologists, would miss a quarter of the cancers detected by systematic sextant biopsy. The present study examines the clinical significance of those carcinomas missed by transrectal ultrasound. We found that the isoechoic cancers have fewer positive biopsy cores and a lower pre-biopsy serum PSA than do the hypoechoic cancers. Since serum PSA is known to correlate with tumor volume, 13 both of these findings suggest that the isoechoic cancers are smaller than hypoechoic cancers, which is not unexpected. One would expect that larger cancers would be more likely to be visible on transrectal ultrasound. Furthermore, even if transrectal ultrasound had no ability to detect prostatic carcinoma, in other words the association between hypoechoic lesions and cancers was purely random, larger carcinomas that were present in multiple biopsy cores would still be more likely to appear on biopsies from hypoechoic sectors than would smaller carcinomas. More interesting is the similarity between the 2 groups of cancers in terms of Gleason scores and pathological staging. Extensive morphometric studies by the Stanford group suggested that tumor volume correlates with tumor grade and that tumors undergo a progression from small, well differentiated tumors to large poorly differentiated tumors. 14 - 16 Since the isoechoic cancers were smaller and associated with lower serum PSA levels, one would predict that the Gleason scores of these cancers would be lower. However, the Gleason scores of the hypoechoic and isoechoic cancers in this study were indistinguishable. Epstein et al recently reported less association between tumor volume and tumor grade with many high grade tumors in small cancers, 17 which may in part explain our findings. The presence of many high grade cancers in the isoechoic cancer group indicates that these cancers are clinically sig-

nificant. These are not low grade carcinomas that need not be detected and may be best managed with observation. The median PSA in the isoechoic group was 6.1 ng./ml. versus 8.6 ng./ml. in the hypoechoic group. One could even argue that detection of isoechoic cancers is preferable to detection of hypoechoic cancers, since the isoechoic cancers are more likely to have a serum PSA of less than 10.0, above which prostate cancer is less likely to be surgically resectable. 18 The patients who underwent surgical staging procedures showed a similar stage distribution for the hypoechoic and isoechoic cancers. If one combines the organ confined and Cl subsets, which probably have a similar prognosis, the distribution is almost identical. However, this may only reflect the criteria used to recommend radical prostatectomy, the staging procedure most of these patients received. Not all cases were pathologically staged and the selection criteria were not random. In summary, these data indicate that approximately 25% of cancers are found only in isoechoic sectors. While these cancers are smaller than those found in hypoechoic sectors, they appear to be significant cancers based on serum PSA, biopsy grade and pathological stage. A policy of systematic biopsy to sample all areas of the prostate is strongly recommended. REFERENCES

1. Boring, C. C., Squires, T. S., Tong, T. and Montgomery, S.: Cancer statistics, 1994. CA, 44: 7, 1994. 2. Lee, F. R., Gray, J.M. and McLeary, R. D.: Prostatic evaluation by transrectal sonography: criteria for diagnosis of early carcinoma. Radiology, 158: 91, 1986. 3. Salo, J. 0., Rannikko, S., Makinen, J. and Lehtonen, T.: Echogenie structure of prostatic cancer imaged on radical prostatectomy specimens. Prostate, 10: 1, 1987. 4. Dahnert, W. F., Hamper, U. M., Eggleston, J. C., Walsh, P. C. and Sanders, R. C.: Prostatic evaluation by transrectal sonography with histopathologic correlation: the echopenic appearance of early carcinoma. Radiology, 158: 97, 1986. 5. Ellis, W. J., Chetner, M., Preston, S. and Brawer, M. K.: Diagnosis of prostatic carcinoma: the yield of serum PSA, DRE and TRUS. J. Urol., 152: 1520, 1994. 6. Brawer, M. K., Chetner, M. P., Beatie, J., Buchner, D. M., Vessella, R. L. and Lange, P. H.: Screening for prostatic carcinoma with PSA. J. Urol., 147: 841, 1992. 7. Brawer, M. K., Beattie, J., Wener, M. H., Vessella, R. L., Preston, S. D. and Lange, P.H.: Screening for prostatic carcinoma with PSA: results of the second year. J. Urol., 150: 106, 1993. 8. Hodge, K., McNeal, S., Terris, M. and Stamey, T.: Random systematic versus directed ultrasound-guided transrectal core biopsies of the prostate. J. Urol., 142: 71, 1989. 9. Gleason, D. F.: Histologic Grading in Clinical Staging of Prostatic Carcinoma. Philadelphia: Lea & Febiger, pp. 171-197, 1977. 10. Lee, F., Torp-Pedersen, S. T., Littrup, P. J., McLeary, R. D., McHugh, T. A., Smid, A. P., Stella, P. J. and Borlaza, G. S.: Hypoechoic lesions of the prostate: clinical relevance of tumor size, DRE, and PSA. Radiology, 170: 29, 1989. 11. Cooner, W., Mosley, R., Rutherford, C. J., Jr., Beard, J. H., Pond, H. S., Terry, W. J., Igel, T. C. and Kidd, D. D.: Prostate cancer detection in a clinical urological practice by ultrasonography, digital rectal examination and prostate specific antigen. J. Urol., 143: 1146, 1990. 12. Carter, H.B., Hamper, U. M., Sheth, S., Sanders, R. C., Epstein, J. I. and Walsh, P. C.: Evaluation oftransrectal ultrasound in the early detection of prostate cancer. J. Urol., 142: 1008, 1989.

ISOECHOI-C PllOSTP.""TIC CA.RCl}·JOMA

T. A_., ~lang, J>J.> Hay Pl,_, R., lVIcNeal, J-. E., F:reiha, ;_,T. S. E.: Prostate-specific antigen as a serum marker for adenocarcinoma of the prostate. New Engl. J. Med., 317: 909, 1987. 14. McNeal, J. E., IGndrachuck, R. A. and Freiha, F.: Patterns of progression in prostate cancer. Lancet, 1: 60, 1986. 15. Stamey, T., McNeal, J., Freiha, F. and Redwine, E.: Morphometric and clinical studies on 68 consecutive radical prostatectomies. J. Urol., 139: 1235, 1988. 16. McNeal, J. E., Villers, A. A., Redwine, E. A., Freiha, F. S. and Stamey, T. A.: Histologic differentiation, cancer volume, 13.

1

2307

and pelvic lymph node metastasis in adenocarcinoma of the prostate. 66: 1225, 1990. 17. Epstein, J.E., Carmichael, M. J., Partin, A. W. and Walsh, P. C.: Small high grade adenocarcinoma of the prostate in radical prostatectomy specimens performed for nonpalpable disease: pathologenetic and clinical implications. J. Urol., Hil: 1587, 1994. 18. Catalona, W., Smith, D., Ratcliff, T., Dodds, K., Coplen, D., Yuan, J., Petros, J. and Andriole, G.: Measurement of prostate-specific antigen in serum as a screening test for prostate cancer. New Engl. J. Med., 324: 1156, 1991.