Prostatic intraepithelial neoplasia: A premalignant lesion

Prostatic intraepithelial neoplasia: A premalignant lesion

Prostatic lntraepithelial Neoplasia: A Premalignant lesion MICHAEL K. BRAWER, MD Putative premalignant changes in the prostate have been recognized ...

3MB Sizes 15 Downloads 40 Views

Prostatic lntraepithelial Neoplasia: A Premalignant lesion MICHAEL K. BRAWER,


Putative premalignant changes in the prostate have been recognized for a number of years. A variety of synonyms have been given to the most commonly described

acini and ductules. Currently, “prostatic intraepithelial is the term most commonly

this review, the requirements

Prostatic intraepithelial neoplasia consists of dysplasia and proliferation of the normal lumenal cell layer lining prostatic ducts and acini. Features of note in PIN include cellular crowding and stratification, inequality in cell size, variation in nuclear size, hyperchromatism, and nucleomegaly. The atypical features of the cells making up high-grade PIN are indistinguishable from those of invasive cancer. The major distinction between carcinoma and PIN is that the latter is confined to a basal cell layer and, thus, there is preservation of at least the semblance of the normal acinar system with two distinct epithelial cell types, basal and lumenal. The proliferation of the “lumenal” cells in PIN may be quite severe. In the most extreme case there may be a cribriform pattern within the PIN acini or ductule. This may make this variant of PIN difficult to distinguish from cribriform Gleason grade 3 adenocarcinoma.28,2Y Prostatic intraepithelial neoplasia is distinctly different from atypical adenomatous hyperplasia. This condition is felt by some investigators to represent another premalignant change within the human prostate. Atypical adenomatous hyperplasia consists of circumscribed proliferation of small round glands similar in appearance to Gleason grade 1 carcinoma. However, atypical adenomatous hyperplasia fails to demonstrate cytologic atypia. It is my belief that data supporting the premalignant nature of atypical adenomatous hyperplasia are lacking; thus, this condition will not be discussed further in this review. McNea130 emphasized the premalignant nature of PIN (referred to as intraductal dysplasia) and subsequently, along with Bostwick in 1986, described reproducible diagnostic criteria.‘” These investigators described a three-step grading system for PIN, and I have modified it slightly as shown in Table 1.” Although the three-step grading system has enjoyed considerable use, a recent consensus meeting elected to combine grades 2 and 3 PIN into a common grade 2 lesion.‘7 Recent clinical observations as described below underscore the utility of such a two-step system as the biologic significance of grades 2 and 3 PIN far exceeds that of grade 1 and may be indistinguishable one from the other (Figs 1 through 3). As noted above, one of the major characteristics of PIN is the presence of a basal cell layer. This serves as a readily distinguishing feature between invasive carcinoma and PIN as the former is characterized by absence of basal cells. The evanescent nature of basal cells as depicted on hematoxylin-eosin preparations makes their identification difficult at times. Myself and colleagues previously have described immunohistochemical


used to describe this entity. In

for a premalignant

lesion, the reason

why PIN fulfills the majority of these requirements, description of the morphologic

and a detailed

and phenotypic similarities between

PIN and carcinoma will be outlined. Finally, the clinical significance of PIN


be reviewed.


lesion, one characterized by prolif-

eration and dysplasia of the normal two cell layers lining prostatic (PIN)





0 1992 by W.B. Saunders Company

A number of organs, including the uterine cervix and endometrium, the gastrointestinal tract, the urothelium, and the respiratory epithelium, have well-recognized premalignant lesions.‘,’ These lesions meet several criteria for premaiignant change; specifically, they have morphologic features similar to invasive carcinoma, there is spatial association between these entities and cancer, foci of microinvasive cancer can be seen to arise from the putative premalignant change, and the frequency, severity, and extent of the premalignant conditions occur to a greater degree in organs harboring invasive cancer. Finally and most significantly, progression into invasive carcinoma can be described in a number of these conditions. The concept of premalignant lesions supports the multi-step theory of neoplasia.“-5 Oerteil provided the first description of a prostatic premalignant change.6 Subsequently, a number of reports have appeared in the literature describing these conditions.7- 6 One source of great confusion in the literature as well as in surgical pathology reporting has been the wide range of synonyms used to describe prostatic premalignant conditions. Terms such as “atypical epithelial hyperplasia, ” “atypical glandular hyperplasia,” “cytologic atypia, ” “duct-acinar dysplasia,” “glandular atypia, ” “intraductal dysplasia,” “intraglandular dysplasia,” and “large acinar atypical hyperplasia” have been used. Partly because of this confusion, a consensus conference was held in 1989 and the term “prostatic intraepithelial neoplasia” (PIN) was considered to be the most appropriate nomenclature for the most common premalignant prostatic change.27 From the Department of Urology, University of Washington. Seattle, WA; and the Section of Urology, Seattle VA Medical Center, Seattle, WA. Supported in part by a grant from the Veterans Administration. M.K.B. is a recipient of a Career Development Award from the American Cancer Society. k?y wr~rrf~:prostate, carcinoma, premalignant change. Address correspondence and reprint requests to Michael K. Brawer, MD, Department of Urology, RI.- IO, University of Washington, Seattle, WA 98195. Copyright 0 1992 by W.B. Saunders Company 0046~8177/92/2303-0004$.5.00/O




I’1N (;r.ldr



(:I-owding. stratification, intact basal cell layer Similar to PIN 1. increased c-rowding. may have basal tell layer disruption Similar to PIN 2. may be crihriform. lrequent hasal cell layer disruption

from Rostwitk

(Michael K. Bower)

Grading Criteria for Prostatic lntraepithelial

.9rchlteC turr



Variable Variable

SIX size





Normal Increased

Kare May br large, prominent





and Rrawet-.‘o


labeling of the basal cells of normal and hyperplastic prostatic .tissue using anticytokeratin antibodies.20,3’“3 5 and The antibj )dy used is directed against cytokeratins 13. Others have reported similar observations.“” The premalignant requirement of morphologic similarity between the putative premalignant change and invasive carcinoma can be extended to similar phenotypic expression as evidenced by immunohistochemistry. Kecently. for example, we described immunohistochemical similarities among an antibody directed against cvtokeratins l-1, 15, 16. and 19 (JL44) (developed by Dr K. B. Nagle, Department of Pathology, University of Arizona, Tucson, AZ), the lectin Ulex euroapaeus (Ulex), and antibodies to the intermediate filament vimentin.‘” McNeal et al’“’ noted other phenotypic similarities between PIN and invasive carcinoma. For example, they noted decreased immunoreactivity with prostate-specific antigen (F’SA), prostatic acid phosphatase, and Leu-7 (a differentiation marker). Enhanced immunoreactivity was observed with antibodies directed against pepsinogen 2 and tissue plasminogen activator. Finally, Perlnran and with Ulex as Epstein” have confirmed our observations well as noted absence of the blood group antigens ABH and 1,ewis in c,arcinoma and PIN.



Another requirement for a lesion tct be considered premalignant is that its incidence be higher in organs with carcinoma than in those without. Table 2 demonstrates a compilation of several series from the literature OI in which radical prostatectomy, cystoprostatectomy, prostates obtained at post mortem examination were serially sectioned and examined for the presence of PIN. As is readily apparent, PIN occurs much more commonly in prostates with invasive carcinoma than in those without. For example, McNeal and Bostwick compared the incidence of PIN in 100 prostates with cancer and in 100 prostates without invasive carcinoma in an auintraepithelial neoplasia was topsy beries.“’ Prostatic found in 82% of the former but in only 43% of the latter. Oyasu et al’” conducted a similar study comparing 5 1 specimens obtained from radical prostatectomies performed for invasive carcinoma with 5 1 prostates obtained at postmortem examination. Ten of the postmortem examination specimens had incidental carcinoma. These investigators described PIN in 94% of the radical prostatectomy specimens but in only 59% of the

FIGURE 1. Grade 1 PIN. Note extensive field exhibitIng this change. (Hematoxylin-eosin stain: magnification r81 )



Volume 23, No. 3 (March 1992)

FIGURE 2. Grade 2 PIN. Note extensive proliferation of dysplastic cells with scattered prominent nucleoli. (Hemotoxylin-eosin stain; magnification X 160.)

case of invasive carcinoma in which PIN could not be demonstrated when sufficient material for examination was available. The relative rarity of PIN in the transition zone, as discussed below, may suggest that pure transition zone carcinoma may not be associated with PIN. However, in most of the cases of stage A carcinoma that my colleagues and I have examined, we have found foci of PIN (unpublished observation). Further study of this phenomena is important because this may serve as an important distinguishing feature between carcinomas

autopsy cases without invasive cancer. All autopsy prostates with carcinoma had PIN. Troncoso et al’” conducted an interesting study in which step sections of prostates obtained at cystoprostatectomy for invasive urothelial carcinoma were studied for the presence of PIN. Prostatic intraepithelial neoplasia was identified in all 61 specimens without invasive carcinoma and in 72% of those without. These findings support my observation of PIN in all specimens obtained at radical prostatectomy. 2o My colleagues and I have not observed a single

FIGURE 3. Grade 3 PIN. Note increased onisonucleosis, prominent nucleoli, and clumping of chromatin. The basal cell layer is prominently displayed in this section. (Hematoxylin-eosin stain; magnification x 162.)






Prostatic lntraepithelial Neoplasia Incidence (Literature Compilation)

prostatrs. All berirs are based on examination of whok prostates ta~nrd ‘It radital prf~statcctomy, ~ystoprostatectom~, or autopsy.

The volume of PIN also appears to c.orrelatr with the presence of carcinoma. McNeal and Bostwick’” noted that whereas only 4% of prostates without invasive carcinoma had an area of PIN greater than 1 low-power microscopic field, a third of the prostates with invasive carcinoma had PIN in such volume. These investigators also noted that multicentric carcinoma was more commonly associated with prostates with multiple fields of PIN as compared with those prostates without PIN. Troncoso et al”’ reported that 7 1YOof the prostates with carcinoma had more than 10 foci of PIN and 28Y0 had more than PO foci. Only 15%’ of the prostates without invasive carcinoma in their cystoprostatectomy series had multifocal PIN. These investigators also noted that large foci of PIN (defined as > 1 low-power microscopic field) were much more commonly associated with prostates demonstrating invasive cancer.


arising in the tr.ansition zone and the far more common peripheral Lone carcinoma. The incidence of PIN in organs without carcinoma is considerably lower than the incidence in prostates with cancer, but still is quite high. The disparity in the reported incidence of PIN in benign glands may be related in part to differences in the thoroughness of specimen handling and also to the age of the patients from whom the prostates were obtained. If PIN represents a premalignant change, one would anticipate that this lesion would be found more commonly in younger men without carcinoma, and the incidence in glands without cancer should decrease as men age (proportionately to the increasing incidence of carcinoma in glands with PIN). In keeping with this hypothesis, McNeal and Bostwick’” noted that 45% of men aged 50 to 59 years and 52% of those aged 60 to 69 years had PIN in prostates without carcinoma. This percentage dropped to 37% in men aged 70 to 79 years and 38% in those a ed X(J years or more. Kovi et al noted a similar trend. w It seems from these investigations that the peak incidence of PIN appears to antedate that of invasive carcinoma. PROSTATIC GRADE


(Michael K. Brawer)



Prost,itic intraepithelial neoplasia grade seems to correlate with the presence of carcinoma. For example, if. the worst grade PIN present in the specimen is quantitated, grade 1 PIN is found in between 7% and 46% of prostates with carcinoma (mean, ~~YoO).~~~‘~~‘~~~~ Grade 2 PIN occurs more commonly in glands with carcinoma, for a mean incidence of 35% (range, 2 1% to 48%). 16,18,‘4 Finally, grade 3 PIN, which my colleagues and I believe is synonymous with carcinoma in situ, is found in between 6% and 90Y0 of organs with invasive carcinoma, for an ave-rage among all reported series of 54%.‘“.‘6,18.‘4 In contrast, grade 1 PIN is much more commonly found in prostates without invasive carcinoma. Investigators have reported these findings in 14% to 8 1% of prostates (mean, 55’%). Grade 2 PIN is found in glands without carcinoma in between 14% and 68% of prostates (mean, 35%). Grade 3 PIN is rarely found without invasive carcinoma (mean, 17%; range, 5% to 32%).1s.‘ti.‘8.n’ In general. it appears that grade 3 PIN occurs three times more frequently in prostates with invasive carcinoma ttlan in those without. 245

Similarly to carcinoma, PIN is much more commonly found in the peripheral zone of the prostate.“” For example. Troncoso et al”’ noled that 81% of the foci of PIN occurred in the peripheral zone. In 1,068 PIN lesions noted in the prostates without invasive carcinoma. 86% were found in the peripheral zone. Similar observations have been made by Kovi et al.lx Furthermore, Troncoso et al noted that PIN was found in the transition zone in only 5% of the organs with cancel and in 1% of those without cancer. This calls into question the importance of PIN in so-called transition zone carcinomas (ie, those that arise wholly within the transition zone). In contrast, Kovi et allX noted that 39% of the PINS were located in the “inner prostate.” The exact zonal site of this region was not specified. Not only does there appear to be a Tonal relationship to carcinoma with PIN, there also appears to be a spatial relationship between invasive carcinoma and PIN on a subzonal level. For example, we noted that of 1,093 ducts and acini with PIN, 41% (453) were found within 1 high-power microscopic field of invasive carcinoma in a radical prostatectomy series.“’ The proximity to carcinoma appears to correlate with PIN grade. Eightyseven percent of the 453 acini were grade 3. Similar findings have been reported by Troncoso et al.“” This relationship between PIN and carcinoma is further apparent on the microscopic level. One of the requirements of a premalignant lesion is the presence of microinvasive carcinoma arising from it. This was first described in PIN by McNeal in 1965.“” -My colleagues and I have confirmed McNeal’s findings of microinvasive carcinoma in PIN using the basal cell-specific immunohistochemistry described above.“” It also appears that there is a strong spatial association between PIN and invasive carcinoma on zonal, subzonal (regional), and microscopic (microinvasion) levels. The relative rarity of PIN in the transition zone as noted above gives some credence to the suggestion that transition zone carcinomas may significantly diffet from the more common peripheral zone carcinoma.


Volume 23, No. 3 (March 1992)

Prostatic intraepithelial neoplasia can be identified in simple prostatectomy specimens in which, theoretically, the majority of the tissue arises from the transition zone. For example, we noted PIN in 39 of 81 simple prostatectomy samples in which all removed tissue was embedded and sectioned. Prostatic intraepithelial neoplasia was present in each of the 14 carcinomas; in addition, 25 cases without carcinoma had PIN. It is possible that these findings reflect the actual PIN in the transition zone; however, another possibility is that a portion of the resection included peripheral zone tissue. Further work in this area is clearly indicated. An interesting observation from our investigation of the basal cell layer in PIN is that there appeared to be a strong correlation between grade of PIN and disruption of the circumferential basal cell layer.” While PIN always exhibited some basal cells, as evidenced by our anticytokeratin immunohistochemical preparations, we noted that 56% of the acini and ductules with grade 3 PIN had disruption of the basal cell layer. More than 30% of the basal cell layer was disrupted in 52% of the high-grade PIN. In contrast, less than 1% of grade 1 PIN was associated with disruption of the basal cell layer. This disruption of the basal cell layer may represent the earliest stage of invasion or progression of PIN into invasive carcinoma.



The definitive requirement for a lesion to be considered premalignant is observation of its progression into invasive carcinoma on serial biopsy. This is readily achieved in two-dimensional epithelial structures with or without the use of endoscopic instrumentation. Unfortunately, due to the three-dimensional characteristic of the prostate, it is not possible to serially biopsy the exact duct acinar system, even with the most modern imaging modalities. Thus, it would be impossible to demonstrate histologically the progression of PIN into invasive carcinoma. Despite this limitation, it appears that the evidence is overwhelming that PIN is at least one of the premalignant changes in the human prostate. The concern of some investigators, that this may be a tumor-associated condition but not premalignant, does not reflect the frequent observation of PIN in organs without invasive carcinoma.

CLINICAL IMPORTANCE OF PROSTATIC INTRAEPITHELIAL NEOPLASIA Prostatic intraepithelial neoplasia is obviously interesting on a number of grounds. The clinical importance of PIN is in three primary areas: the effect of PIN on serum PSA level, the results of repeat prostatic biopsy in men who have PIN identified on previous biopsy or simple prostatectomy specimens, and the ultrasonographic appearance of PIN. 246

PROSTATIC INTRAEPITHELIAL NEOPLASIA AND PROSTATE-SPECIFIC ANTIGEN Prostate-specific antigen presents a major advance in our tumor marker armamentarium. This 34-kd lycoprotein is prostate specific but not cancer specific. &.,, Prostate-specific antigen has important applications, primarily in monitoring patients with established prostatic carcinoma after therapy. The utility of PSA for staging or early detection of prostate cancer remains the subject of intense investigation. One difficulty in the application of early detection of prostate cancer with this analyte has been the frequent observation that PSA was elevated in approximately 20% of patients with benign prostatic hyperplasia.40-4’ Two requirements are necessary before PSA, elaborated in the lumenal cells of the prostatic epithelium, can be detected in the serum. Cells elaborating this protein must be present and the protein must gain access to the systemic circulation. For the latter to occur, this not insignificantly sized protein must transgress a number of tissue layers, namely, the basal cell layer, the basement membrane, the intervening prostatic stroma, the capillary basement membrane, and the endothelial cell layer. Whereas benign hyperplasia is not associated with disruption of these layers, other conditions certainly could be. In an effort to investigate this theory we performed serial step sections of all tissue removed from simple prostatectomy specimens in patients with presumed benign prostatic hyperplasia giving rise to their bladder outlet obstructive symptomatology.4J In this series the mean PSA in the 26 patients with benign hyperplasia alone was 2.1 ng/mL (range, 0.3 to 4.7 ng/ mL). Fourteen patients had incidental carcinoma. Their mean PSA was 35.1 ng/mL (range, 0.9 to 234.0 ng/ mL); however, of the 25 patients who had PIN without invasive carcinoma, the mean PSA was 5.6 ng/mL (range, 0.3 to 22.0 ng/mL). Approximately half of the patients with PIN had PSAs above the 4.0 ng/mL gram upper normal limit recommended by the manufacturer of this assay (Hybritech Inc, San Diego, CA). As noted above, PIN is frequently associated with disruption of the basal cell layer. Similarly, I have noted in a series of men undergoing ultrasound-guided prostate needle biopsy for palpable prostatic abnormality that the mean PSA of patients with PIN in their biopsy specimen was intermediate between those with pure benign histology and those with carcinoma.“” Of 36 patients with benign histology, the mean PSA was 5.8 ng/mL (range, 0.4 to 24.9 ng/mL). The PIN group comprised 15 patients and had a mean PSA of 7.8 ng/mL (range, 0.2 to 19.2 ng/mL) and the patients with carcinoma had a mean PSA of 18.3 ng/mL (range, 0.2 to 161 ng/mL). Lee et al”” noted similar observations in patients undergoing ultrasound-guided biopsy for PIN: the mean PSA of these patients was 9.5 ng/mL, intermediate between those with benign histology (4.7 ng/mL) and those with carcinoma (84 ng/mL). Whether these series, based on the simple prostatectomy or ultrasound-guided biopsy, reflect a true correlation of PIN and elevation of PSA or whether they reflect the fact that PIN is frequently





PIN on the initial biopsy had invasive carcinoma on the subsequent ultrasound-guided biopsy. Prostate-specific antigen levels were obtained in 16 patients prior to the second biopsy. There was a stratification of higher PSAs in those patients with the subsequent diagnoses of carcinoma. The relatively small number of patients in this series precluded PSA alone from realizing statistical significance in discriminating between those with and without carcinoma.

with carcinoma and undetected carcinoma caused the increased PSA level remains a matter of conjecture. Nevertheless, the correlation of PSA and the finding of PIN will gain increasing importance as PSA is used more frequently in an early detection or screening modalitv. PROSTATIC INTRAEPITHELIAL NEOPLASIA AND PROSTATE ULTRASOUND Transrectal ultrasound of the prostate is felt by many to currently represent the best method of imaging the prosta.te. The most common presentation of carcinoma is a hypoechoic area in the peripheral zone. Unfortunately, the sensitivity and specificity of such a finding on ultrasound are, in general, rather poor. Prostatic intraepithelial neoplasia, as noted above, is a relatively common finding on ultrasound-guided prostate needle biopsy of such hypoechoic peripheral zone lesions. My colleagues and I have previously reported eight men undergoing ultrasound-guided biopsy of hypoechoic peripheral zone lesions who had PIN as their only abnormality “’ Prostatic intraepithelial neoplasia was confirmed using cytokeratin immunohistochemistry. Of note, three of the eight patients on repeat ultrasoundguided biopsy or transurethral section of the prostate had invasive carcinoma. Whether PIN is actually giving rise to the hypoechoic lesion or whether other entities, such as undetected carcinoma, are the culprits remains to be proven. Nevertheless, the extensive use of transrectal ultrasound of the prostate mandates an undersranding of the association of PIN and these findings. PROSTATIC INTRAEPITHELIAL AND REPEAT BIOPSY

(Michael K. Brawer)



The importance of further evaluation of men undergoing prostate biopsy who have PIN was first suggested in the report of a small series in which ultrasoundguided pr’ostate needle biopsy was performed after negative digitally guided biopsy for palpable prostatic abnormality. Among 22 patients, my colleagues and 1 noted invasive carcinoma in 11 who had previous negative biopsy. Of note, of the six patients with PIN identified on their initial biopsy specimen performed under digital guidance. five had invasive carcinonla.‘7 Recently. my colleagues and I have extended our examinaticm of patients with biopsy proven PIN without invasive carcinoma in a series of 21 patients who had PIN identified on a biopsy performed because of an abnormal digital rectal examination.“’ All patients underwent repeat ultrasound-guided biopsy and either a systematic random biopsy in which six prostate sections were sampled with a spring-loaded biopsy gun or, if the site of previous biopsy demonstrating PIN was known, repeat biopsy of that region alone. Twelve of the patients (57Yo) had carcinoma identified on the second procedure. Of particular note was the correlation of invasive carcinoma on the repeat biopsy and PIN grade. Of the 11 patients with grade 1 PIN, only two had invasive carcinoma. In contrast. all patients with grade 2 or Z3 247


It appears that PIN has a significant Importance in the clinical management of patients with prostatic disease. Prostatic intraepithelial neoplasia appears to be associated with an elevation of serum PSA in a high percentage of patients. The yield of repeat biopsy in patients who have PIN noted on needle biopsy of the prostate is significant. Finally, sonographically, PIN appears to be indistinguishable from the most common presentation of carcinoma. How, then, should the surgical pathologist advise the clinician in the management of patients who have PIN identified either in simple prostatectomy specimens or on needle biopsy? First, the surgic:al pathologist should be sure that all tissue is examined. We have seen several cases in which submission of additional simple prostatectomy tissue or levels obtained on needle biopsy in patients with PIN have demonstrated carcinoma. Prostate-specific antigen could be a useful adjunct to the further management of patients with PIN. An elevated PSA may be an indication that underlying malignancy is likely. However, as noted above, we did not demonstrate statistical significance for this assa) in this regard. Clearly, clinicians and their patients need to know that PIN is identified. While the management of such patients is controversial, close monitoring is clearly warranted. Our current procedure is to perform immediate repeat ultrasound-guided biopsies in patients who have grade 2 or 3 PIN identified on simple prostatectomy or needle biopsy specimens. If this result is negative, we perform digital rectal examination and PSA determination in 6 months. If there is any significant change we perform repeat ultrasound. (;rade 1 PIN identified on prostate needle biopsy is probably best managed with semiannual or annual rectal examination and repeat PSA. A strong association of high-grade PIN with carcinoma is not an indication for treatment. Clearly, before such radical therapy as radiation or radical prostatectomy is performed, inva:sive carcinoma must be conclusively demonstrated. Prostatic intraepithelial neoplasia appears to he an important prostatic lesion. It fulfills the mqjority of requirements for premalignant prostatic c-hange. Moreover, PIN has important clinical ramifications in the management of patients with prostatic disorders. Further investigation on the role of this lesion in prostatic. carcinogenesis, etiologic issues, and the optimum management of patients with PIN is clearlv indicated.


Volume 23, No. 3 (March 1992)

REFERENCES I. Ikl Kept~~j.A.Ackrt-mart L.\‘:kttttoloq ofcmccr. in Harstlherger SE. Kasper R (ed): (:ant cr: Diagnoses. Treatment and Prognosis (ed 6). St Louis, MO. Moshy. I!)%?,. pp L’7-L’X L’. Sctwlc KOK. Swi:nq J: PI c-c an< crot1\ c hangrs in hladdet epithclium. I,anrct 2:9-l:{-!ll(i. I!)68 3. Rvwlin AM: ‘I’ermiltolo~~ of premalignant lesions in light of the mulris;ep thrrrrv ~rfc-~trcinogeltesis. H~IM P,trHOt. ! 5:8O&Xf)7,I%% 4. Medtine A, Farber E: The multistep thcol-v of ncoplasia, itI Antl~otty PP. MxSween KNM (eds): Recent Adv;mces in Histopatholog\. New I’ork. NY. (:hurc hill I.i\inghtone. 1981, pp I O-20 5. Hamlvn P, Sikora K: Om ~+y~es. I.;u~rt 2:X26. I!#3 6. Orrtril H: Imc)luticmatv change% in prostate and female hl-east in rclatitrn to cancer developmrnr. Can Med Assoc J l6:237-241, 1926 7. Moor.e RA: Benign hypcrtrophy of the prostate. A morphoIogic.ll study. J Ilrol 50:680-il(J. 19-13 8. Aticlrews GS: Lttetit carcinoma of the pr-ostate. J Clin Path01 2: l!l7-“08. I Q-19 0. Tant~cnhau~nl M: Histopatholo&~ of the 1”“state gland, in I‘annenhamn bl (cd): Llrologic Patholou: The Prostate. Philadelphia, PA. Lea & Fehiger. 1977. pp 309-397 IO. Helpap B: The hiotogical significance of atypicalhyprrplasia of the prostate. Virchow~ Arch [A] 387:307-5 17. I980 I 1. I,iavag J: Atrophy and regeneration on pathogenesis of prostatic carcinoma. Acta Pathol Mic robiol Scatld 73:338-350. 1968 l?. Tannenbaum MI: .ttypical epithelial hyperplasia or carcinoma of prostate gland: The surgical pathologist at arr impass? Urology 4: 758-760. 1974 13. Oyasu R. Bahnson RR, Nowels K. et at: Cytological atypia in the prostate gland: Frequency, distribution and possible relevance to carcinoma. J Urot 135:959-962, 1986 14. Harbitz TB. Haugen OA: Histology of the prostate in elderly men. A study in an autopsy series. Acta Pathol Stand 80:756-768, 1972 15. Baron E, Angriat A: Incidence of occuh adenocarcinoma of the prostate after fifty years of age. Arch Pathot 32:787-793, 1941 16. McNeal JE, Bostwick DG: Intraductal dysplasia: A premalignant lesion of the prostate. HUM PATHOI. 17:64-71, 1986 17. Miller A, SeIjelid R: Cellular atypia in the prostate. Stand J Ural Nephrol 5:17-21, 1971 18. Kovi J, Mostofi FK, Heshmat MY, et al: Large acinar atypical hyperplasia and carcinoma of the prostate. Cancer 61:555-561, 1988 IQ. McNeal JE: Significance of duct-acinar dysplasia in prostatic carcinogenesis. Prostate 13:91-102, 1988 20. Bostwick DG, Brawer MK: Prostatic intra-epithelial neoplasia and early invasion in prostate cancer. Cancer 59:788-794. 1987 21. Garnett JE, Oyasu R: The urologic evaluation of atypical prostatic hyperplasia. Urology 34:66-69, 1989 22. Brawn PN: Adenosis of the prostate. A dysplastic lesion that can he confused with prostate adenocarcinoma. Cancer 49:826-833, 1982 ?3. Kastendiek H: Correlations between atypical primary hyperplasia and carcinoma of the prostate: A histological study of 180 total prostatectomies. Pathol Res Pratt 169:366-387, 1980 24. Troncoso P, Babaian RJ, Ro JY, et al: Prostatic intraepithetial neoplasia and invasive prostatic adenocarcinoma in cystoprostatectomy specimens. Urology 34:52, 1989 25. Srigley J, King S. Van Nostrand AWP. et al: The “preneoplastic” prostate: A giant section whole organ study of 72 radical prostatectomies. Lab Invest 54:61A, 1986 (abstr) 26. Helpap B: Pathologic der ableitenden Harnwege und der Prostate. Berlin, Germany, Springer-Vertag, 1989 27. Drago JR, Mostofi FK, Lee F: Introductory remarks and workshop summary. Urol 6:2-3, 1989 (suppl)


28. Gleason DF: Histologic grading in clinical staging of prostatic carcinoma, in Tannenhaum M (ed): Urologic Pathology: The Prostate. Philadelphia, PA. I.ea & Fehiger, 1977 pp 17 I- I97 29, McNeal JE. Reece JH, Redwine EA. et al: Cribriform ad+ nocarcinoma of the prostate. Cancer 58: 17 14-17 I!). 1986 30. McNeal JE: Morphogenesis of prostatic carcinoma. reactivity in the benign and neoplastic human prostate. <:ancer Res 45:3663-3667. 1985 32. Brawer MK, Bostwick DG. Peehl DM. et al: Keratin pt-stein in human prostatir tissue and cell culture. Ann N Y Acad Sci 455: 729-733, 1985 33. Brawer MK, Nagle RB, Pitts W. et al: Keratin immunol-cattivity as an aid to the diagnosis of persistent adenocarcinoma in irradiated human prostates. Cancer 63:454-460. 1989 34. Nagle RB, Ahmann FR, McDaniel KM, et al: Cytokeratin characterization of human prostatic carcinoma and its derived cell lines. Cancer Res 47:281-28S, 1987 35. Nagle RB. Brawel- MK. Kittrlscm J. CI ‘11: Phcnc~t~pic rclationships of prostatic intraepithelial neoplasia to invasivr pt’ostati(. carcinoma. Am ,J Pat hol 138: I IQ- 128, 199 I 36. McNeal JE. Alroy J, Leav I, et al: Immunohistocher~lic-al cvidence for impaired cell differentiation in the premalignant phase of prostate carcinogenesis. Am J Clin Pathol 90:23-32. 1988 37. Perlman EJ. Epstein JT: Blood group antigen expression in dysplasia and adenocarcinoma of the prostate. Lab Invest 60:72A, 1989 (abstr) 38. Nadji M, Tabei SZ, Castro A, et al: Prostatic-specific antigen: An immunohistotogic marker for prostatic neaplasms. Cancer 48: 12?91232, 1981 39. Stein BS, Peterson RO, Vangore S. et al: Inirnurioperoxidase localization of prostate-specific antigen. Am J Surg Pathol 6:553-557. 1982 40. Ercole CJ, 1,ange PH, Math&n M, et al: Prostatic specific. antigen and prostatic acid phosphatase in the monitoring and staging of patients with prostatic cancer. ,J UroI 138: 118 l-l 184, I987 41. Stamey TA. Yang N, Hay AR, et al: Prostate-specific antigen as a serum marker for adenocarcinoma of the prostate. N Engl J Meti 317:909-916. 1987 42. Oesterling JE. Chan DW, Epsrein Jl, er al: Prostate specific antigen in the preoperative and postoperative evaluation of localized prostatic cancer treated with radical prostatectomy. J Ural 139:766772, 1988 43. Brawer MK, Rennels MA, Schifman RA. et al: Serum PSA in men undergoing simple prostatectomy. Am J Clin Pathol 92:760. 1989 44. Brdwer MK: Prostatic intraepithelial neoplasia: Signilicance and management, in Williams RD (ed): Problems in Urolow: Controversies in Prostate Cancer Management. Philadelphia, PA, I.ippincott, 1989 pp 392-407 45. Lee F, Torp-Pedersen ST, Carroll JT, et al: The use of tr-ansrectal ultrasound and prostate-specific antigen in the diagnosis of prostatic intraepithelial neoplasia. Urology 34:4-8, 1989 46. Brawer MK, Rennels MA, NagIe RB, et al: Prostatic intraepltheliat neoptasla: A lesion which may be confused with cancer on prostate ultrasound. J Ural 142:151 O-1512, 1989 47. Brdwer MK, Nagle RB: Transrectal ultrasound guided prostate needle biopsy following negative digitally guided biopsy. J IJrot 141:182, 1989 (abstr) 48. Brawer MK, Bigler SA, Sohlberg OE, et al: The significance of prostatic intraepitheli; neoplasia on prostate needle biopsy. Urology 38:103-107. 1991