Central zone histology of the prostate: A mimicker of high-grade prostatic intraepithelial neoplasia

Central zone histology of the prostate: A mimicker of high-grade prostatic intraepithelial neoplasia

Central Zone Histology of the Prostate: A Mimicker of High-grade Prostatic Intraepithelial Neoplasia MONICA SRODON, MD, AND JONATHAN I. EPSTEIN, MD Th...

600KB Sizes 0 Downloads 6 Views

Central Zone Histology of the Prostate: A Mimicker of High-grade Prostatic Intraepithelial Neoplasia MONICA SRODON, MD, AND JONATHAN I. EPSTEIN, MD The central zone (CZ) is located at the base of the prostate adjacent to the seminal vesicles. Its histology as a potential mimicker of high-grade prostatic intraepithelial neoplasia (PIN) has not been formally studied. Three groups were evaluated. Group 1 comprised 30 consecutive radical prostatectomy specimens assessed for the extent of CZ and of Roman arch and/or cribriform formation in the CZ. Group 2 comprised 100 consecutive cases of nonconsult prostate needle biopsies, screened in a random blinded fashion to identify CZ histology and the specificity of its identification on biopsy. Group 3 comprised 34 consult cases (1984 to the present) with CZ histology on needle biopsy. For group 1, the average maximum diameter of CZ histology was 5 mm. Two cases (6.7%) did not contain the classic features of CZ histology. The average amount of cribriform and/or Roman arch formation in the areas with CZ histology was 16.5%. In group 2, 10% of prostate needle biopsy cases had CZ histology. Of these, 80% were located on biopsy specimens designated as the base of the prostate, 10% were located in the base and midportion of the prostate, and 10% were located in the midportion of the prostate. For group 3, CZ histology occupied on average 32% of the involved core. The 2 most common histologic features were eosin-

ophilic cytoplasm (97%) and location at the end of a core (97%). Other features were Roman arch formation (59%), a prominent basal cell layer (32%), cribriform formation (26%), and associated thick muscle bundles typical of bladder neck (24%). On average, cribriform and/or Roman arch formation occupied 22% of the CZ area seen on biopsy. Twenty-six of the consult cases were sent in with preliminary outside diagnoses. Of these, 21 (81%) were either PIN or atypical: 11 (42%) high-grade PIN, 7 (27%) PIN, and 3 (12%) atypical glands. Our findings show that CZ histology is distinctive, as seen in radical prostatectomy specimens. Less frequently it is found on needle biopsy, where the presence of Roman arch and/or cribriform formation mimics PIN. Recognition of the distinctive features of CZ histology (i.e., tall columnar cells with eosinophilic cytoplasm, prominent basal cell layer, and lack of cytologic atypia) can help avoid a misdiagnosis of PIN or “atypia” on needle biopsy. HUM PATHOL 33:518-523. Copyright 2002, Elsevier Science (USA). All rights reserved. Key words: prostate, pathology, prostatic intraepithelial neoplasia, central zone. Abbreviations: PIN, prostatic intraepithelial neoplasia; CZ, central zone.

The prostate has been anatomically divided into 4 zones: anterior fibromuscular stroma, central zone (CZ), peripheral zone, and preprostatic region, which includes the periurethral ducts and transition zone.1 The CZ is located at the base of the prostate and surrounds the ejaculatory ducts. Formal studies have not been written about the histology of the CZ, so many general surgical pathologists are not familiar with its unique features. McNeal, in his description of the CZ, described large acini with irregular contours where “epithelial-covered stromal ridges project into the gland lumens.”1 He goes on to describe cells of the CZ as being columnar, crowded, with somewhat darker granular cytoplasm than cells in the peripheral zone. However, his single illustration of CZ histology lacks cribriform or Roman arch formation and does not resemble high-grade prostatic intraepithelial neoplasia (PIN); the issue of CZ glands resembling high-grade PIN is not discussed. In 1989, Epstein, in one of the

earliest works to briefly illustrate and discuss this lesion as a mimicker of high-grade PIN, noted “prominent roman arch formation commonly seen toward the base of the prostate gland” without cytologic atypia.2 More recently published specialty texts on prostate pathology have not covered this topic in greater depth, merely stating that CZ histology, due to its architectural complexity, may mimic high-grade PIN.3,4 These works either do not illustrate CZ histology or do so briefly in no greater detail beyond what was done by Epstein in 1989. The purpose of the current study was to formally study the histology of the CZ, its frequency in prostate needle biopsy and radical prostatectomy specimens, and the features that distinguish it from high-grade PIN.

From the Departments of Urology and Pathology, The Johns Hopkins University Hospital, Baltimore, MD. Accepted for publication January 4, 2002. Supported by the National Institutes of Health (SPORE CA58236) and the Reinhard Professorship in Urologic Pathology, The Johns Hopkins University Hospital. Address correspondence and reprint requests to Jonathan I. Epstein, MD, The Johns Hopkins Hospital, Weinberg Bldg., Room 2242, 401 N. Broadway St., Baltimore, MD 21231. Copyright 2002, Elsevier Science (USA). All rights reserved. 0046-8177/02/3305-0010$35.00/0 doi:10.1053/hupa.2002.124032

MATERIALS AND METHODS In this study, 3 types of prostate specimens were evaluated. Group 1 consisted of 30 consecutive radical prostatectomy specimens obtained from our institution. These were evaluated for the presence or absence of CZ histology and the extent to which cribriform and/or Roman arch formation was seen in the CZ. The average diameter of CZ histology was also evaluated, with each slide representing a 3-mm thickness of prostatic tissue. Group 2 comprised 100 consecutive cases of prostate needle biopsies performed at our institution. Each case consisted of 6 parts designating the location of the biopsies as either the left or right half as well as the apex, middle, or base of the prostate. The slides were screened in a random, blinded fashion to obtain the frequency of CZ histology seen

518

CENTRAL ZONE HISTOLOGY (Srodon and Epstein)

FIGURE 1. Radical prostatectomy specimen with extensive CZ histology. (A) Lower magnification with prominent Roman arch and cribriform formation. (B) Higher magnification of the CZ histology showing tall columnar cells and Roman arch and cribriform formation.

FIGURE 2. Prostate needle biopsy specimen with prominent CZ histology. (A) Low magnification comparing CZ histology (top) to a core with typical histology (below). (B) Higher magnification of the prostate needle biopsy with CZ histology demonstrating tall columnar cells and Roman arch and cribriform formation.

519

HUMAN PATHOLOGY

Volume 33, No. 5 (May 2002)

FIGURE 3. (A) Lower magnification of a prostate needle biopsy specimen with CZ histology. (B) Higher magnification of the prostate needle biopsy specimen showing Roman arch formation and tall columnar cells.

FIGURE 4. (A) Prostate needle biopsy specimen with CZ histology demonstrating tall columnar cells. (B) Higher magnification showing the striking basal cell layer, tall columnar cells, and lack of cytologic atypia. Nucleoli are absent.

520

CENTRAL ZONE HISTOLOGY (Srodon and Epstein)

on sextant needle biopsy as well as the specificity of its location. Group 3 comprised 34 consult cases of prostate needle biopsies from 1984 to the present identified as having CZ histology. The biopsies were evaluated for the extent of CZ histology in the cores as well as the following histologic features: Roman arch formation, cribriform formation, eosinophilic cytoplasm, location at the end of the core, prominent basal cell layer, and association with thick muscle bundles seen in the bladder neck. When available, the preliminary diagnoses from outside institutions were reviewed to determine the diagnosis most commonly confused with CZ histology.

RESULTS In the group 1 specimens, the average diameter of CZ histology was 5 mm, indicating the presence of CZ histology on 2 consecutive glass slides. A diameter of 15 mm was seen in 2 of the 30 cases (Fig 1). In another 2 cases (6.7%), the CZ did not contain definitive cribriform or Roman arch formation. Ten of the 100 (10%) group 2 specimens were identified as having CZ histology. Of these 10, 8 (80%) were found in biopsies designated as the base of the prostate, 1 (10%) was located in a biopsy designated as the middle of the prostate, and 1 (10%) contained CZ histology in biopsies designated as both the base and middle of the prostate. The group 3 cases were evaluated for the extent and specific features of CZ histology. The average extent of CZ histology in an involved core was 32% (Fig 2). The most frequently seen features were eosinophilic cytoplasm (97%) and location at the end of the biopsy core (97%). Roman arch formation was identified in 59% of the cases; cribriform formation, in 26% (Fig 3). When Roman arch and/or cribriform formation was identified in the biopsy, this histology occupied an average of 22% of the CZ area. Less frequently seen were prominent basal cell layers (32%) and thick muscle bundles (24%) (Figs 4 and 5). Of the consult cases with CZ histology, 26 were received with preliminary diagnoses. Twenty-one (81%) were diagnosed as either PIN or atypical glands: 11 (42%) as high-grade PIN, 7 (27%) as PIN, and 3 (12%) as atypical. The remaining 5 cases (19%) were received with a benign diagnosis. DISCUSSION Several studies have shown that the incidence of high-grade PIN on prostate needle biopsies varies from 1.5% to 24%, with a median incidence of 5% to 6%.5 Such a wide range may be due to interobserver variability, use of alternative fixatives, variation in sampling techniques, variability in patient population, and diagnostic errors resulting from the difficulty in distinguishing PIN from its mimickers. The importance of accurately diagnosing high-grade PIN has been shown in several studies in which the risk of carcinoma in subsequent biopsies ranged from 27% to 79%.5 The largest

FIGURE 5. Prostate needle biopsy specimen with CZ histology (bottom) associated with thick muscle bundles (top) typically seen in the bladder neck.

studies to date have shown a risk ranging from 23% to 35% on subsequent needle biopsies.6-8 Several histologic findings on prostate needle biopsies may mimic high-grade PIN. Two malignant entities that may be confused with high-grade PIN are cribriform acinar adenocarcinoma and ductal adenocarcinoma. Although ductal adenocarcinoma was initially described as occurring exclusively within the transition zone, it is now recognized within the peripheral zone as well and may be sampled on needle biopsy.9 Benign entities that also may mimic high-grade PIN include clear cell cribriform hyperplasia, which is seen within the transition zone and only rarely identified on needle biopsy,10 and basal cell hyperplasia (and sometimes even normal basal cells). As with high-grade PIN, basal cells may have prominent nucleoli and mitoses.11,12 Florid basal cell hyperplasia is typically present within the transition zone, although minor degrees of basal cell hyperplasia mimicking high-grade PIN can be seen on needle biopsy. CZ histology is another lesion that may be misdiagnosed as high-grade PIN on needle biopsy. In our study, the frequency of CZ histology in group 2 specimens was 10%. The relatively low frequency with which CZ histology may be seen on needle biopsy may reflect the small size (mean, 5 mm) of the CZ as seen in the current study’s evaluation of radical prostatectomy specimens, and the absence of distinctive CZ histology seen in 6.7% of the prostate specimens that we studied.

521

HUMAN PATHOLOGY

Volume 33, No. 5 (May 2002)

FIGURE 6. Prostate needle biopsy specimen with high-grade PIN. (A) Low magnification showing prominent Roman arch and cribriform formation. (B) Higher magnification revealing nuclear atypia with prominent nucleoli (arrows).

Of the group 3 cases with CZ histology, 81% had a preliminary diagnosis of either PIN or atypical. CZ histology and high-grade PIN share certain features. Both may show cribriform and Roman arch formation. Both also have tall columnar cells that tend to stand out at low magnification. In contrast to highgrade PIN, however, glands of the CZ have eosinophilic cytoplasm and may reveal a prominent basal cell layer. CZ glands also tend to be located at the end of the core biopsy and may be associated with thick muscle bundles typically in the bladder neck. When biopsies are divided according to location (apex, middle, and base), CZ histology are most often located in biopsies designated as base. The uncommon case with CZ histology seen in a sample labeled from the midportion of the prostate may reflect a very prominent CZ, as was occasionally seen in our radical prostatectomy specimens, or extension of the biopsy needle into the base despite the urologist’s attempts to sample the midportion of the gland. Most important, CZ histology lacks the cytologic atypia and prominent nucleoli observed in high-grade PIN (Fig 6). Distinguishing high-grade PIN from benign mimickers is important to avoid labeling a patient as being at increased risk for prostate cancer. Such a mislabeling can result in unnecessary repeat biopsies and undue concern; we are even aware of anecdotal cases in which men have undergone radical prostatectomy for the diagnosis of high-grade PIN. In summary, CZ histology is a relatively common

benign histologic feature seen on prostate needle biopsy that may mimic PIN. Key features of CZ histology include the presence of tall columnar cells with eosinophilic cytoplasm, a prominent basal cell layer, and lack of prominent nucleoli. CZ glands may have prominent cribriform or Roman arch architecture and are typically located at the end of the core. Finally, knowledge that the cores are from the base of the prostate gland can help the pathologist arrive at a definitive diagnosis of CZ histology and avoid overdiagnosing high-grade PIN.

REFERENCES 1. McNeal JE: Normal and pathologic anatomy of prostate. Urology 17(suppl):11-16, 1981 2. Epstein JI: Prostate Biopsy Interpretation. New York, NY, Raven Press, 1989, pp 152-153 3. Bostwick DG, Dundore PA: Biopsy Pathology of the Prostate. London, UK, Chapman and Hall, 1997, pp 1, 4 4. Young RH, Srigley JR, Amin MB, et al: Tumors of the prostate gland, seminal vesicles, male urethra, and penis, in Rosai J (ed): Atlas of Tumor Pathology. Washington, DC, Armed Forces Institute of Pathology, 200, pp 9-10, 89-90, 2000 5. Epstein JI, Potter SR: The pathologic interpretation and significance of prostate biopsy findings: Implications and current controversies. J Urol 166:402-410, 2001 6. Kronz JD, Allan CH, Shaikh AA, et al: Predicting cancer following a diagnosis of high-grade prostatic intraepithelial neoplasia on needle biopsy: Data on men with more than one follow-up biopsy. Am J Surg Pathol 25:1079-1085, 2001 7. Davidson D, Bostwick D, Qian J, et al: Prostatic intraepithelial

522

CENTRAL ZONE HISTOLOGY (Srodon and Epstein) neoplasia is a risk factor for adenocarcinoma: predictive accuracy in needle biopsies. J Urol 154:1295-1299, 1995 8. O’Dowd GJ, Miller MC, Orozco R, et al: Analysis of results within 1 year after a noncancer diagnosis. Urology 55:553-559, 2000 9. Brinker DA, Potter SR, Epstein JI: Ductal adenocarcinoma of the prostate diagnosed on needle biopsy: Correlation with clinical and radical prostatectomy findings and progression. Am J Surg Pathol 23:1471-1479, 1999

10. Ayala AG, Srigley JR, Ro JY, et al: Clear cell cribriform hyperplasia of the prostate. Report of 10 cases. Am J Surg Pathol 10:665-671, 1986 11. Devaraj LT, Bostwick DG: Atypical basal cell hyperplasia of the prostate. Immunophenotypic profile and proposed classification of basal cell proliferations. Am J Surg Pathol 17:645-659, 1993 12. Epstein JI, Armas OA: Atypical basal cell hyperplasia of the prostate. Am J Surg Pathol 16:1205-1214, 1992

523