effects - from bmj and acp

96 Clin Pathol 1994;47:906-913

Effects of combination endocrine treatment onnormal prostate, prostatic intraepithelialneoplasia, and prostatic adenocarcinoma

R Montironi, C Magi-Galluzzi, G Muzzonigro, E Prete, M Polito, G Fabris

Institute ofMorbidAnatomy andHistopathologyR MontironiC Magi-GalluzziE PreteG Fabris

Urology Clinic, SchoolofMedicine,University ofAncona,Nuovo OspedaleRegionale, Torrette diAncona, ItalyG MuzzonigroM PolitoCorrespondence to:Professor RodolfoMontironi, Institute ofMorbid Anatomy andHistopathology, School ofMedicine, University ofAncona, Nuovo OspedaleRegionale, Via Conca, I-

60020 Torrette di Ancona,ItalyAccepted for publication9 March 1994

AbstractAims-To investigate the effect of combi-nation endocrine treatment (CET) orluteinising hormone releasing hormoneagonist and flutamide on non-neoplasticprostate, prostatic intraepithelial neopla-sia, and prostatic adenocarcinoma.Methods-The morphology, includingthe mitotic activity, of 12 radical prosta-tectomies from patients with prostaticadenocarcinoma pretreated for threemonths with CET was evaluated inhaematoxylin and eosin stained sectionsand compared with an untreated age andstage matched control group.Results-A differential effect on the non-neoplastic prostate was observed. In fact,the transition zone of the treated prostateshowed simplification of the glandularlobules: the ducts and acini were smallwithout undulations of the epithelial bor-der and with a prominent basal cell layer.Within the peripheral zone there wasinconspicuous branching of the ducts andacini which looked dilatated and lined byflattened atrophic epithelium. Prostaticintraepithelial neoplasia occurred inscattered ducts and acini in the periph-eral zone of 10 of the 12 patients. Theepithelial cell lining showed a prominentbasal cell layer. A certain degree of secre-tory cell type stratification was alwayspresent. However, crowding was less evi-dent than in the untreated prostatebecause of cytoplasmic clearing andenlargement as a result of coalescence ofvacuoles. The treated adenocarcinomashad neoplastic acini which looked smalland shrunken, and areas of individualinfiltrating tumour cells separated byabundant interglandular connective tis-sue. The secretory cells of the non-neoplastic, prostatic intraepithelial neo-plasia, and prostatic adenocarcinomalesions had inconspicuous nucleoli,nuclear shrinkage, chromatin condensa-tion, and cytoplasmic clearing. Apoptoticbodies were easily identifiable in all thecell layers. The lumina were rich inmacrophages, sloughed secretory cellswith degenerative features, and apoptoticbodies. Mitoses were not observed in anyof the treated non-neoplastic prostate,prostatic intraepithelial neoplasia, orprostatic adenocarcinomas, whereas themitotic frequency increased from non-neoplastic prostate through prostaticintraepithelial neoplasia up to prostatic

adenocarcinomas in the untreated speci-mens.Conclusions-CET before radical prosta-tectomy causes regressive epithelialchanges together with enhanced apopto-sis and blocked mitotic activity.

(J Clin Pathol 1994;47:906-913)

For many years, testicular androgen depriva-tion by oestrogen treatment or orchidectomyhas been the standard hormonal treatment foradvanced invasive adenocarcinoma of theprostate.' Recently, new drugs, especially ago-nists of luteinising hormone releasing hor-mone (LHRH) and anti-androgens, such asflutamide, have been introduced.2' Althoughthe efficacy of these new drugs used alone isequivalent to that of standard treatment, theiruse avoids the major side effects of oestrogentreatment and the psychological impact oforchidectomy.4 Treatment combining anLHRH agonist and a pure antiandrogen drug(combination endocrine treatment or CET)to block simultaneously adrenal and testicularandrogen has been shown to be superior toeither approach used alone.5-7Some histological changes induced by CET

in human prostate cancer have recently beendescribed.8 "' However, there is scant informa-tion on the differential effect of an LHRHagonist and flutamide on the normal prostatezones and, in particular, the influence of CETon prostatic intraepithelial neoplasia.The aim of our study was to evaluate the

effect of CET on non-neoplastic prostate,prostatic intraepithelial neoplasia, and prostaticadenocarcinomas in resected prostatectomyspecimens with emphasis on morphology andmitotic activity.

MethodsThe study included the histopathologicalmaterial from 12 consecutive patients (agerange 60 to 70 years) who had undergone aradical prostatectomy for stage B prostaticadenocarcinoma"2 after temporary CET.Before surgery the patients had received CETfor three months. Radical prostatectomy wasperformed while the patient was still receivingtreatment.The histopathological material of 60 cases

(20 of benign prostatic hyperplasia, 20 ofprostatic intraepithelial neoplasia, and 20 ofprostatic adenocarcinoma) included in a pre-vious publication"3 was used to compare the

906

on Decem

ber 25, 2021 by guest. Protected by copyright.

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.47.10.906 on 1 O

ctober 1994. Dow

nloaded from

http://jcp.bmj.com/

Prostatic intraepithelial neoplasia treated by CET

histological changes induced by CET with theprostate morphology of untreated specimens.The control group (age matched with thetreated group) underwent a radical prostatec-tomy for stage B prostatic adenocarcinomawithout receiving chemotherapy, hormone, orradiation treatment before surgery. We choseto use a separate control group of untreatedpatients instead of the pretreatment prostatebiopsy specimens because the latter type ofdiagnostic procedure does not usually yieldrepresentative material containing prostaticadenocarcinoma, prostatic intraepithelial neo-plasia, and non-neoplastic prostate for athoroughly comparative study, especially formitosis counts.The prostatectomy specimens were deliv-

ered fresh from the operating room shortlyafter excision. They were cut into slices about0 3 cm thick and then fixed for 24-48 hoursin neutral buffered formalin (4%), dehydratedin graded alcohols, cleared in xylene, andembedded in paraffin wax. The haematoxylinand eosin stained sections of the radicalprostatectomies of the treated group werereviewed by one of us (RM) to select slidescontaining prostatic adenocarcinoma, prosta-tic intraepithelial neoplasia (if present), andnon-neoplastic tissue. Slides with acute andchronic inflammation were discarded. Theslides selected for prostatic adenocarcinomaalso covered the peripheral zone of theprostate including the capsule-that is, theadvancing front of the carcinoma where theproliferative activity is thought to be higherthan in the centre of the tumour nodule.9Because the prostatic epithelium adjacent toprostatic intraepithelial neoplasia and prosta-tic adenocarcinoma usually shows a prolifera-tive pattern,'4 the sections selected torepresent the non-neoplastic tissue were at adistance from prostatic intraepithelial neoplasiaand prostatic adenocarcinoma and coveredthe peripheral and transition zones. The for-mer comprises about 70% of the mass of theglandular prostate and is adjacent to the rec-tum. The latter surrounds the urethra fromthe base of the gland tapering to the verumon-tanum.

Mitoses were assessed on conventional 3pm thick histological sections (microtome set-ting) cut from formalin fixed, paraffin waxembedded material. The research project wasconducted using haematoxylin and eosinstained sections. For the analyses of mitoticfrequency and location, the sections wereevaluated by one of our team using a LeitzOrthoplan microscope equipped with a x 63objective and an eyepiece graticule.'3 The cri-teria used to identify the mitotic figures and todistinguish them from apoptotic bodies havebeen detailed before.'5 In particular, mitosesare characterised by the absence of nuclearmembrane, absence of a clear zone at the cen-tre, by the presence of hairy instead of trian-gular or spiky projections, and basophilia ofthe surrounding cytoplasm. In contrast, apop-totic bodies are round or oval in shape andvariable in size, made up of masses ofpyknotic chromatin surrounded by narrow

cytoplasmic rims. The percentage of mitosesfrom a minimum of 1000 nuclei per case (alsocalled the mitotic index) was calculated sepa-rately for each cell layer. The time needed toanalyse each case was about 60 minutes.The data were stored in an Apple

Macintosh II computer. StatView II softwarewas used for the calculation of the mean andstandard error of mean (SEM), as well as forstatistical analyses (Kruskal-Wallis andMann-Whitney U tests). Reproducibility wastested for by duplicate evaluations of mitosesin six untreated cases (two benign prostatichyperplasia, two prostatic intraepithelial neo-plasia, and two prostatic adenocarcinoma),but no significant differences were found.

ResultsFor the treated group, the following morpho-logical patterns of prostate diseases were pre-sent and investigated in the 12 patients:non-neoplastic prostate (p = 12), prostaticintraepithelial neoplasia (n = 10), prostaticadenocarcinoma (seven cases with acinar pat-tern, four with cribriform pattern, and onewith solid or trabecular pattern). For theuntreated group, the following morphologicalpatterns of prostrate diseases were investi-gated in prostatectomy specimens: non-neoplastic prostate (n = 20), prostaticintraepithelial neoplasia of low grade(PINlow) (n = 10), and prostatic intraepithe-lial neoplasia of high grade (PINhigh) (n =10), and prostatic adenocarcinoma (five caseswith cribriform pattern or PACcri, five withsolid or trabecular or PACsolid, five withsmall acinar or PACsmac, and five with largeacinar or PAClgac). The morphology of theuntreated cases has already been extensivelyreported.'4 16 17

NON-NEOPLASTIC PROSTATEWithin the peripheral and transition zones ofthe untreated non-neoplastic prostate, ductsand acini were morphologically similar to sim-ple rounded contours that were not perfectlycircular because of prominent undulations ofthe epithelial border. Both ducts and aciniwere lined by columnar secretory cells con-taining pale cytoplasm and separated from thebasement membrane by noticeably flattenedcells having dark, slender, filiform nuclei, andusually little or no discernible cytoplasm(basal cell layer) (fig 1). Apoptotic bodieswere very seldom seen. Mitoses wereextremely rare (mean (SEM) value 0-00 1(0-001)%) and seen exclusively in the basalcell layer, where their mean (SEM) value was0-002 (0 003)%. No difference was observedbetween peripheral and transition zones. Thelumina contained occasional macrophagesand rarely apoptotic bodies.The transition zone of the treated non-neo-

plastic prostate showed important architec-tural and epithelial cell changes which werenever seen in the untreated sections.Simplification of the glandular lobules wasthe consistent major feature on all specimensat low magnification, although the lobular

907

on Decem


http://jcp.bmj.com

/J C


ctober 1994. Dow

nloaded from

http://jcp.bmj.com/

908~~~~~~~~~~~~~~~~~~~Montironi,Magi-Galluzzi, Muzzonigro, Irete, Polito, Fabris

odf ~Am~~~~~~~~

4.:',VA~ ~ 4041

....

k4V

Of A&~~~~~~~~~~

A~ ~ ~

A~~~~~~~.

wit:4!'

2iur Thrastinzoeoftetreatednon-neoplastic prostateshowiinoe) hedcs simpaciciatioeo

siplrun ed c nt ur. he a eliedbwasum a cduct hacini werepl sm

andbyafuenbtsau

r~~~~~~~~

4 ~ ~

e~~~~~~~~~~~~~~~~~~~~~~~~~~i

and~~P

-A~ ~ ~ ~

haU

la4ae1hr

and cytoplasmic clearing. The basal cell layer

was prominent (fig 2). Focal squamous and

transitional metaplasia as well as basal cellhyperplasia were seen. Within the peripheralzone there was inconspicuous branching ofthe ducts and acini which looked dilatated,star-shaped, and lined by flattened atrophicepithelium, which, in general, was single-layered and seldom double-layered (fig 3).This pattern was diffuse and involved theentire peripheral zone in all the treatedpatients. In the untreated cases this atrophicchange was predominantly subcapsular andinvolved only occasional glands, with segmen-tal distribution. Atrophic and cysticallydilatated glands were rare in the transitionzone in both untreated and treated patients.Neither squamous and transitional metaplasianor basal cell hyperplasia were seen mn theperipheral zone. Apoptotic bodies were seenmore frequently than in the untreated prostateat the level of the epithelial cell layers as wellas in the lumina. In the latter macrophageswere also increased. No mitotic figures werepresent in the treated cases.

PROSTATIG INTRAEPITHELIAL NEOPLASMAIn the untreated cases prostatic intraepithelialneoplasia always occurred multifocally andwas located generally in the peripheral zone.In PINlow, cell crowding was present andaccompanied by irregular nuclear spacing aswell as a prominent increase in nuclear size.In addition to the increased size variability,nuclei showed finely granular chromatin and afew albeit prominent nucleoli usually locatedcentrally; the cytoplasm was in general mod-erately clear and was eosinophilic in some ofthe cells (fig 4). In PINhigh, the deviationsfrom normal were more evident than inPiN.low. In particular, the degree of cytologicalabnormality was comparable with that of ade-nocarcinoma: presence in most nuclei oflarge, prominent eosinophilic nucleoli posi-tioned peripherally, chromatin margination,noticeable nuclear enlargement, and degree ofnuclear crowding, which in some areas was sosevere as to form bridges of epithelial cellsextending across the glandular lumina. Thecytoplasm was in general granular. Someapoptotic bodies, identified in all epithelialcell layers of the ducts and acini, were foundin general in the intercellular space and occa-sionally seen in the cytoplasm of epithelialcells. The lumina always contained somemacrophages, together with some apoptoticbodies. The frequency of mitoses in theepithelial cell layers was greater than in non-neoplastic prostate: PINlow, 0-053 (SEM0-024)%; PINhigh, 0-095 (SEM 0-043)%. Inboth grades the percentage of mitoses wascounted separately for the cells adjacent to thebasement membrane, for the cells borderingthe lumen, and for the cells in the positionintermediate between the basal and the luminallayers. In both prostatic intraepithelial neo-plasia grades the proportions of mitosesdecreased from the basal position through theintermediate cell layers to the cell layer bor-dering the lumen. For PINlow, the mean val-ues were 0-087 (SEM 0-04)% in the basal,0-046 (SEM 0-033)% in the intermediate,

908

on Decem


http://jcp.bmj.com

/J C


ctober 1994. Dow

nloaded from

http://jcp.bmj.com/


44

Xr'-Ug'a-z.~~*~AFigure 3 In contrast to the treated transition zone, there is inconspicuous branching of theducts and acini within the peripheral zone of the non-neoplastic prostate. These lookdilatated and lined by atrophic epithelium.

;*F#X 5 @ i4 ;wh~~~~4 IT ". -..F. .11 .. ...;............................. 3:S

Figure 4 Untreated prostatic intraepithelial neoplasia (low grade PIN). Cel crowding ispresent and accompanied by irregular nuclear spacing as weUl as an increase in nuclearsize. Nuclei show finely granular chromatin and a few albeit prominent nucleoli usuallylocated centrally.

and 0O024 (SEM 0O024)% in the luminal posi-tion. For PINhigh, the mean values were

0-194 (SEM 0O178)% in the basal position,0075 (SEM 0 06)% in the intermediate, and0049 (SEM 0 033)% in the luminal position.In PINlow the percentages were slightly lowerthan in PINhigh.

In the treated patients prostatic intraepithe-lial neoplasia was observed only in the periph-eral zone and was localised in scattered ductsin comparison with the control group inwhich the extension was greater. The epithe-lial cell lining showed a basal cell layer whichwas recognisable in most instances. A certain

degree of secretory cell type stratification wasalways present. However, crowding was lessevident than in the untreated prostate becauseof cytoplasmic clearing and enlargement as aresult of coalescence of vacuoles; the nuclei ingeneral showed small inconspicuous nucleoliwithout the preferential margination typical ofuntreated prostatic intraepithelial neoplasia.The chromatin showed different degrees ofchanges which ranged from a mild condensa-tion-which barely permitted the distinctionbetween coarse chromatin granules (corre-sponding to heterochromatin) and finely dis-persed chromatin (corresponding toeuchromatin)-to a tightly condensed statevery much like that observed in apoptosis.The luminal border of the epithelium wasirregular in outline due to cells of differentsize and some shallow hollows correspondingto the site of detached cells. Apoptotic bodieswere easily identifiable in all secretory celltype layers and were found only in the inter-cellular space. The lumina contained abun-dant cells: some were macrophages, somesloughed secretory cells with degenerative fea-tures, while others corresponded to apoptoticbodies. As the duct and acinar changes wereso severe, no clear post-treatment grading ofthe prostatic intraepithelial neoplasia lesionswas possible (fig 5). No mitotic figures werepresent in treated prostatic intraepithelial neo-plasia.

PROSTATIC ADENOCARCINOMAThe hallmark of all the untreated adenocarci-nomas was that the tumour nuclei were fre-quently multinucleolated, the nucleoli beingprominent, marginated, and with a narrowclear space around. The chromatin in generallooked finely granular with some chromatinmargination along the inner surface of thenuclear membrane. The cytoplasm of thesmall and large acinar patterns was moder-ately clear, whereas in the cribriform and solidpatterns it was generally granular. The cellboundaries were always clearly recognisable(fig 6). The closely packed glands and tumournodules with cribriform or solid or trabecularpatterns with amphophilic cytoplasm whichlooked haphazardly related to one anotherstood out on low magnification when com-pared with the pale-staining benign glands.Like the untreated prostatic intraepithelialneoplasia, some apoptotic bodies were identi-fied in all epithelial cell layers and, togetherwith macrophages, in the lumina.When considering the mitotic count in the

adenocarcinomas with a cribriform pattern,the proportions of mitoses, greater than inPINhigh, decreased from the basal position,or adjacent to the stroma, through the inter-mediate cell layers to the cell layer borderingthe lumen. The mean values were 0-154(SEM 0096)% in the basal position, 0072(SEM 0044)% in the intermediate, and0064 (SEM 004)% in the luminal position.In the solid or trabecular pattern the percent-age of mitoses was calculated separately forthe cells adjacent to the stroma and for thecells in the other cell layers. The mean value

909

on Decem


http://jcp.bmj.com

/J C


ctober 1994. Dow

nloaded from

http://jcp.bmj.com/

Montironi, Magi-Galluzzi, Muzzonigro, Prete, Polito, Fabris

* .: . . . ...................... .r~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~....... ..

o,:i.^ +.& 4

Figure S Treated prostatic intraepithelial neoplasia (peripheral zone). There isinconspicuous branching of the ducts and acini, which look dilatated. The epithelialining shows a recognisable basal cell layer. A certain degree of secretory ceU typestratification is present. Crowding is less evident than in the untreated prostate. This rich in cells, including apoptotic bodies..: . ; . .. .. ... e.

.. .. .!,^: i>: 2: i s:. ..:S,,<. w#:.j : ^::. ': .:''i.:vw } =ff] 5 : .. ... .... . '. - .S . . .:eA9.a<.:: .. .: :: . . 4 .. .:.'S,§ !w: ": , ..... .. ,: .i^:... t :::. .. :: :i . : : . :; ':

:; tek .R .: .:*aZ' :

.. §X :X: .: za : ...

e: .:. : qF. W.W.;::. .: <, ... ... 9 1//|..

'.' ;= wy. jj}S'°d04i :: zsaS.S :- n#*2F s wF.

is,,:§L'...

::: . ., i

..: se: ..:'. :S. :' .:V -aE ZWW

} eS. uo>^mRi9^ ofbdiF* ;. .°Sg ox. S#e SuSor ;,opO ^R Brx *.. ., ej? &:: 81E uf

s«jo'i#: .8' l'>: 6.u.:.'! a s r- .R.. <; .;F .... . :.!:=L':7e0wa h ::* ;B : .. ,: : .i ...... : .: ... ;R ; <R

<2^^<!: % .. ..... D w Sii :.

.: .E.:

.. ....: j: wi.:: . : :: ::: ^ :z ... .. :: w.: .. . : .. . ..... j ... ... 9.iS;S s.,

:.!: 9 :. .' !. Ob Vs.i;2 .2SX;nSRF Zei:^:>::

intraepithelial neoplasia up to prostatic ade-nocarcinoma, the differences being significantfor the following comparisons: Kruskal-Wallistest: benign prostatic hyperplasia v prostaticintraepithelial neoplasia v prostatic adeno-

-.F carcinoma p = 0 0001; Mann-Whitney Utest: benign prostatic hyperplasia v prostaticintraepithelial neoplasia p = 0-006, andbenign prostatic hyperplasia v prostatic ade-nocarcinoma p = 0 0003. Prostatic intra-epithelial neoplasia was regarded as a singlecategory without distinguishing between lowand high grade; prostatic adenocarcinoma wasalso regarded as a single category. Statisticalanalysis showed no significant differences forany of the comparisons between the differentcell layers.)

In comparison with the untreated speci-mens, the treated tumours with acinar patternshowed neoplastic acini which lookedshrunken and areas of individual infiltratingtumour cells separated by an abundant inter-glandular connective tissue. The epithelialtumour cells had inconspicuous nucleoli,

zl cell nuclear shrinkage, chromatin condensationlumina and pyknosis, cytoplasmic clearing and

enlargement by coalescence of vacuoles andrupture of cell membranes (fig 7). Like thetreated prostatic intraepithelial neoplasia,apoptotic bodies were easily identifiable in allepithelial cell layers and, together withmacrophages and sloughed epithelial cells, in

. ...:> the lumina. The treated tumours with cribri-form and solid or trabecular patterns showednuclear and cytoplasmic changes induced byGCET which looked less pronounced than inthe acinar pattern. No mitotic figures werepresent in any of the treated adenocarcino-

..

Figure 6 Untreated prostatic adenocarcinoma with an acinar patternarefrequently multinucleolated tumour nuclei, prominent and marginasurrounded by a narrow clear space. The chromatin looks finely granulchromatin marginalisation along the inner surface of the nuclear memb

in the cell layer adjacent to0-22 (SEM 0Ol l1)% whereaslayers it was 0-074 (SEM 0O0portions of mitoses were highccategories of adenocarcinomEage of mitoses in the small acisimilar to that of PINlow andthe large acinar pattern-that0-024)% and 0-068 (SEM 0tively. (The frequency ofepithelial cell layers increaseprostatic hyperplasia thrc

mas.

DiscussionThis study has shown that GET before radicalprostatectomy causes regressive epithelialchanges, together with enhanced apoptosisand reduced mitotic index on the non-neo-plastic transition and peripheral prostatezones in prostatic intraepithelial neoplasia andprostatic adenocarcinoma.

Current medical opinion shows a lot ofinterest in alternatives to prostatectomy in themanagement of non-neoplastic prostate

c. Typicalfeatures lesions called benign prostic hyperplasiaised nucleoli, which affects the transition zone. In particu-far with some7rane. lar, antiandrogens such as LHRH agonists or

flutamide have been used for reversing benignprostatic hyperplasia.18 19 Peters and Walshdemonstrated the possible effects of an

the stroma was LHRH agonist20: the prostate regressed byin the other cell about 24%, based on radiological imaging. A45)%. The pro- biopsy specimen of the prostate confirmeder than for other that the regression was primarily of glandulara. The percent- tissue, thus confirming earlier histologicalinar pattern was quantitative studies by Huggins and Stevens.21close to that in This group of authors had already observedis, 0-058 (SEM that various antiandrogen approaches*019)%, respec- decrease the epithelial component and pro-mitoses in the duce histological atrophy of the prostate. Tetu-d from benign et al reached a similar conclusion in a histo-)ugh prostatic logical study investigating the effect ofLHRH

910

...43

40"

:: :::: i.*:::

::

:w

.:

on Decem


http://jcp.bmj.com

/J C


ctober 1994. Dow

nloaded from

http://jcp.bmj.com/


4~~~~~~~~~

0*

A:~~~~~~wihaeso nivda nitaig tumou cellepaae by abndn intrganda

0611~~~~ ~ ~~~~~~~~~~:4

shikae chomti cnestion, an cyolsic clarn an enlargement

q9Nt:

difrn ci.oaut th psible

en s an h gicaNngs o the

9jf~~~~mdia cstrtoapproach.}2::t.- :.. _ .. In 'a^ stud by*2lli:i. Kean et at biopsye specimen showe no evi-

i e 8 * ~Cin et al shwe the clnia efeciens of

i~~~~~~bevd Ourophooia finins in the .

Aww the cnl drw b H a

.~~~~~~I th no-eplsi periphra zone, -.;the;*~~mi orgi si..tW-eof potai intapthla

fon df cn wi aredi4fferen

from.-B to in thetransitionis zone.Tis.is.a

Fge7radrst poorlyrcinognis fisndint the best of our

treteadnoarinoaithanaknowledge, Thows netplastiwereithec lonly phreviouswitaeao idivdulnfltatntuthoursto menartion veyabriefyathtitherglgdland

connectivetisue.7heepilattheo ce;p her ofhpriostate treat

LRagonist and flutamide0 wheresdavtate

igre7Teaedprotaicandenorineda by atomphriedn epiththeium.eae Wise, doetreteaencarinmawihnohciavepattr defisniolative explnaionic fork thrnese

different conclusions about the possible clinicalbenefits and morphological changes of themedical castration approach. In a study byKeane et al biopsy specimens showed no evi-dence of reduction in epithelial cell height.2Caine et al showed the clinical effectiveness offlutamide on benign prostatic hyperplasia.3However, no evidence of qualitative histologi-cal changes in prostatic biopsy specimens wasobserved. Our morphological findings in thetransition zone, which basically are similar tothose observed in prostatic intraepithelial neo-plasia and prostatic adenocarcinoma, agreewith the conclusions drawn by Huggins andStevens,2 Peters and Walsh,20adTt ta.t

In the non-neoplastic peripheral zone, themain origin site of prostatic intraepithelialneoplasia and prostatic adenocarcinoma, wefound diffuse changes which are differentfrom those in the transition zone. This is apoorly recognised finding. To the best of ourknowledge, Te'tu et alP° were the only previousauthors to mention very briefly that the glandsat the periphery of the prostate treated with anLHRH agonist and flutamide were dilatatedand lined by atrophied epithelium. We donot have a definitive explanation for thesemorphological differences between the non-neoplastic prostate zones. Although the differ-ential response might be linked to differencesin honmone receptors, we advocate the alter-native, traditional explanation for focal atxo-

py24~--that is, as the consequenceofdcaobstruction associated with a CET effect.

Prostatic intraepithelial neoplasia is consid-ered to be a premalignant lesion which mightprogress to prostatic adenocarcinoma.13 14

Ellison et at8 and Tetu et al'0 mentioned thatCET has some kind of effect on prostaticintraepithelial neoplasia lesions. In particular,

the former group recently affirmed that theextent of high grade prostatic intraepithelialneoplasia decreased significantly in patientsreceiving a total androgen blockade. Eventhough the extent of prostatic intraepithelialneoplasia was evaluated visually, we reachedthe same conclusion in our study. In fact, pro-static intraepithelial neoplasia lesions werelocalised in scattered ducts compared with thecontrol group in which the extension wasgreater. However, to our knowledge, the mor-phology of the regressing prostatic intra-epithelial neoplasia lesions has not beendescribed before. Our study showed regres-sive changes in the secretory cells, while themitotic activity was abolished. The nucleishowed chromatin condensation and pykno-sis. These changes are probably associatedwith qualitative and quantitative DNA alter-ations and might possibly be linked to apopto-sis. A reduced DNA content was observed byPeters and Walsh20 in non-neoplastic prostate.In their experience androgen deprivation forsix months induced a 40% decrease in nuclearDNA content. This has also been our experi-ence with flow cytometric analysis: in fact,DNA histograms with a hypodiploid modewere obtained from some of the fresh prostatespecimens (unpublished data). The cytoplas-mic changes are of the regressive type, inagreement with the study by Tetu et al,'0which showed that prostatic specific antigenimmunostaining was weaker and the numberof positive cells substantially reduced in pros-tatic adenocarcinoma treated by CET.

Apoptotic bodies were more frequentlyseen in the regressing prostatic intraepithelialneoplasia lesions than in the untreated lesionsand were observed at the level of the cell layersas well as in the lumina. This was observedafter CET in the normal and neoplasticprostate in studies in which it was suggestedthat CET induced a certain degree of epithelialregression by enhancing the cell death phe-nomenon called apoptosis.2526 The findings ofthe present investigation with the experimen-tal study of Szende et al 27 who investigatedthe effects of LHRH agonists and somato-statin on cancer in hamsters. They observedthat the tumours treated with agonists showedregressive histological changes characteristicof apoptosis. Our observation probably corre-sponds to the finding of "progressive nuclearpyknosis" observed by Helpap28 in patientsreceiving hormone treatment (the type oftreatment was not specified). Apoptosis wasevaluated by Stiens et al 29 in patients with pro-static carcinoma before and after oestrogenand radiation treatment. They observed thatthe apoptosis index can increase 10-foldwithin the first 10 days of treatment; underradiation treatment, apoptotic bodies can alsobe measured as increasing. Histologically,apoptotic bodies were seen in the intercellularspace and in the lumen where some of thebodies appeared in the macrophage cyto-plasm. This apoptotic body location supportsthe conclusions made by Kerr et al 30 on howapoptotic bodies are eliminated. This groupreported that apoptotic bodies formed in

911

on Decem


http://jcp.bmj.com

/J C


ctober 1994. Dow

nloaded from

http://jcp.bmj.com/

Montironi, Magi-Galluzzi, Muzzonigro, Prete, Polito, Fabris

tissues are dispersed from their site of originalong intercellular spaces; some of those arisingin the epithelia are extruded into the lumen,but they are also phagocytosed and degradedby the adjoining viable cells. Epithelial cells aswell as cells of the mononuclear phagocyticsystem participate in this disposal. In malig-nant neoplasms many of the bodies are takenup and digested by surrounding neoplasticcells. Savill et al clearly demonstrated howmacrophages present in the synovial cavityingested apoptotic cells, mainly neutrophils,in acutely inflamed joints."The absence of mitoses in prostatic intra-

epithelial neoplasia lesions as well as innon-neoplastic prostate and prostatic adeno-carcinoma indicates suppressed proliferationactivity as a consequence of androgen depri-vation treatment, as hypothesised by Murphyet alI and Bruchovsky et al.32 This has beendocumented in the neoplastic prostate.6283' 37For instance, Zalatnai et al described the his-tological changes in rats with prostate cancerafter treatment with somatostatin agonists andthe D-Trp-6 analog of an LHRH agonist.'8They reported that mitotic activity in theepithelium from the treated group was muchlower than from the untreated specimens, inagreement with the findings obtained byOomens et al,34 who assessed the proliferativecell fraction of human prostatic carcinomausing Ki67. These authors found that the per-centage of positive nuclei decreased signifi-cantly after androgen deprivation, dropping to7% of the initial values after three months.Recently, Sakamoto et al found that shortterm treatment with LHRH agonists sup-pressed the activity of thymidate synthetaseand thymidine kinase and substantiallyreduced the appearance of bromodeoxyuri-dine immunoreactive cells.39 Armas et alreported decreased proliferating cell nuclearantigen (PCNA) expression due to preopera-tive androgen deprivation treatment on pro-static carcinoma.6 Magi Galluzzi et al alsofound that the values of the PCNA stainednuclei in treated cases were notably lowerthan in untreated specimens, and interpretedthe low PCNA positivity not only as adecrease in proliferation but also as anattempt to repair DNA damage.9 Even thoughthe proliferation activity is greatly affected, thestem cells might survive and form part of themorphological finding of basal cell promi-nence observed in the non-neoplastic transi-tion zone and in prostatic intraepithelialneoplasia. In a study dealing with the effectsof androgen withdrawal on the stem cell com-position of Shionogi carcinoma, Bruchovskyet al experimentally documented the survivalof this type of cells.32The morphological changes do not seem to

be equally distributed over the prostatic ade-nocarcinoma lesions. Such unequal distribu-tion seems to be linked to the tumour pattern.Those adenocarcinomas with an acinar pat-tern were greatly affected by CET: in fact, thearchitectural and cell damage is obvious andeasily recognisable. CET exerted a pro-nounced effect, to the point that the tumour

acini shrunk, making it difficult to say whetherthey belonged to the large or small acinar pat-tern categories. In the cribriform and solid ortrabecular pattern the changes were not sopronounced as in the acinar pattern, andsome nuclear details were still recognisable, afeature also present in some ducts and aciniwith prostatic intraepithelial neoplasia. Thismeans that the high grade lesions, whetherinfiltrating or intraductal, might give a poorerresponse to three months of CET, and thatwhen high grade lesions are observed in a pre-treatment biopsy specimen, a longer period ofCET has to be planned. Alternatively, as pos-tulated by Akakura et al,40 androgen suppres-sion could be of the intermittent type. Theseauthors studied the effects of intermittentandrogen suppression on androgen dependenttumours and suggested that the replacementof androgens after a certain period of andro-gen suppression might result in the regenera-tion and differentiation of stem cells withfurther apoptotic potential. These cells couldthen be eliminated by successive androgenwithdrawal.

In conclusion, CET induces regressivechanges in the secretory cells and blocks themitotic activity of the non-neoplastic prostate,prostatic intraepithelial neoplasia, and prostaticadenocarcinoma. In particular, its effect onprostatic intraepithelial neoplasia may be ofparamount importance for the pharmacologi-cal treatment of prostate glands with a biopsydiagnosis of prostatic intraepithelial neoplasiawithout morphological or clinical evidence ofprostatic adenocarcinoma. Further studies arenecessary to investigate the longer term effectof CET and the importance of basal cellpreservation in the regrowth of lesions whenCET is discontinued.

This paper was read at the XIVth Congress of the EuropeanSociety of Pathology, Innsbruck, Austria, September 5-10,1993.

1 Scott WW, Menon M, Walsch PC. Hormonal therapy ofprostatic cancer. Cancer 1980;45(suppl):1929-36.

2 Labrie F, Belanger A, Cusan L, Seguin C, Pelletier G,Kelly PA, et al. Antifertility effects ofLHRH agonists inthe male. YAndrol 1980;1:209-28.

3 Sogani PC, Vagaiwala MR, Whitmore WV Jr. Experiencewith flutamide in patients with advanced prostatic can-cer without prior endocrine therapy. Cancer 1984;54:744-50.

4 Glashan RW, Franks LM. Cardiovascular complicationsin the treatment of prostic carcinoma. BrJ Urol 1981;53:624-6.

5 Crawford ED, Eisenberger MA, McLeod DG, SpauldingJT, Benson R, Dorr FA, et al. A controlled trial ofLeuprolide with and without flutamide in prostatic carci-noma. N EnglJMed 1989;321:419-24.

6 Armas OA, Melamed A, Aprikian A, Cordon-Cardo C,Fair W, Reuter WE. Effect of preoperative androgendeprivation therapy in prostatic carcinoma. Lab Invest1993;68:55A.

7 Fair WR, Aprikian A, Sogani P, Reuter V, WNhitmore WE.The role of neoadjuvant hormonal manipulation inlocalized prostatic cancer. Cancer 1993;71 (suppl):1031-8.

8 Ellison E, Ferguson J, Zincke H, Bostwick DG."Nucleolus-poor" clear cell adenocarcinoma of theprostate: a distinctive histologic variant in patientsreceiving total androgen blockade. Lab Invest 1993;68:59A.

9 Magi Galluzzi C, Montironi R, Giannulis I, Diamanti L,Scarpelli M, Muzzonigro GT, et al. Prostatic invasiveadenocarcinoma. Effect of combination endocrinetherapy (LHRH agonist and flutamide) on the expres-sion and location of proliferating cell nuclear antigen(PCNA). Path Res Pract 1994;189:1154-60.

10 Tetu B, Srigley JR, Boivin J-C, Dupont A, Monfette G,Pinault S, Labrie F. Effect of combination endocrinetherapy (LHRH agonist and flutamide) on normal

912

on Decem


http://jcp.bmj.com

/J C


ctober 1994. Dow

nloaded from

http://jcp.bmj.com/


prostate and prostatic adenocarcinoma. Am J SurgPathol 1991;15:111-20.

11 Murphy WM, Soloway MS, Barrows GH. Pathologicchanges associated with androgen deprivation therapyfor prostate cancer. Cancer 1991;68:821-8.

12 Jewett HJ. The present status of radical prostatectomy forstage A and B prostatic cancer. Urol Clin North Am1975;2: 105-23.

13 Montironi R, Magi Galluzzi C, Diamanti L, Giannulis I,Pisani E, Scarpelli M. Prostatic intra-epithelial neopla-sia. Expression and location of proliferating cell nuclearantigen (PCNA) in epithelial, endothelial and stromalnuclei. Virchows Arch (Pathol Anat) 1993;422: 185-92.

14 Bostwick DG, Brawer MK. Prostatic intra-epithelial neo-plasia and early invasion in prostate cancer. Cancer1987;59:788-94.

15 Montironi R, Collan Y, Scarpelli M, Sisti S, Barbatelli G,Camevali A, et al. Reproducibility of mitotic counts andidentification of mitotic figures in malignant glialtumors. Appl Pathol 1988;6:258-65.

16 McNeal JE. Normal histology of the prostate. Am Jf SurgPathol 1988;12:619-33.

17 Mostofi FK, Sesterhenn I, Sobin LH. Histologic typing ofprostate tumours. No. 22. Geneva: WHO 1980:17-20.

18 Geller J. Nonsurgical treatment of prostatic hyperplasia.Cancer 1992;70(suppl):339-45.

19 de Voorgt HJ, Rao BR, Geldof AA, Gooren LJG, BoumanFG. Androgen action blockade does not result in reduc-tion in size but changes histology of the normal humanprostate. Prostate 1987;11:305-11.

20 Peters CA, Walsh PC. The effect of narfarelin acetate, aluteinizing hormone/releasing hormone agonist, onbenign prostatic hyperplasia. A placebo controlled study.NEnglJfMed 1987;317:599-604.

21 Huggins C, Stevens RA. The effect of castration on benignhyperplasia of the prostate in man. J Urol 1940;43:705-14.

22 Keane PF, Timony AG, Kiely E, Williams G, Stamp G.Response of the benign hypertrophied prostate to treat-ment with an LHRH analogue. Br J Urol 1988;62:163-5.

23 Caine M, Perlberg S, Gordon R. The treatment of benignprostic hypertrophy with flutamide (SCH 13521): aplacebo-controlled study. 7 Urol 1975;114:565-8.

24 McNeal JE, Bostwick DG. Anatomy of the prostate: impli-cations for disease. In: Bostwick DG, ed. Pathology of theprostate. New York: Churchill Livingstone, 1990:1-14.

25 Montironi R, Magi Galluzzi C, Scarpelli M, Giannulis I,Diamanti L. Occurrence of cell death (apoptosis) in pro-static intra-epithelial neoplasia. Virchows Arch (PatholAnat) 1993;423:351-7.

26 Montironi R, Muzzonigro G, Magi Galluzzi C, GiannulisI, Diamanti L, Polito M. Benign prostatic hyperplasia.Effect of LHRH agonist and flutamide (combinationendocrine therapy) on the frequency and location of pro-liferating cell nuclear antigen and apoptotic bodies.J Urol Pathol 1994;2:161-71.

27 Szende B, Srkalovic G, Schally AV, Lapis K, Groot K.Inhibitory effects of analogs of luteinizing hormone-releasing hormone and somatostatin on pancreaticcancers in hamsters. Cancer 1990;65:2279-90.

28 Helpap B. Treated prostatic carcinoma. Histological,immunohistochemical and cell kinetic studies. ApplPathol 1985;3:320-41.

29 Stiens R, Helpap B, Weissbach L. Quantitative unter-suchungen zum zellverlust in prostatakarzinomen.Klinisch-morphologische aspekte. Verh Dtsch Ges Urol1981;32:73-4.

30 Kerr JFR, Bishop CJ, Searle J. Apoptosis. In: Anthony PP,MacSween NM, eds. Recent advances in histopathology.No 12. Edinburgh: Churchill Livingstone, 1984:1-15.

31 Savill JS, Wyllie AH, Henson JE, Walport MJ, HensonPM, Haslett C. Macrophage phagocytosis of aging neu-trophils in inflammation. Programmed cell death in theneutrophil leads to its recognition by macrophages. JClin Invest 1989;83:865-75.

32 Bruchovsky N, Rennie PS, Coldman AJ, Goldenberg SL,To M, Lawson D. Effects of androgen withdrawal onstem cell composition of Shionogi carcinoma. Cancer Res1990;50:2275-82.

33 Carroll PR, Waldman FM, Rosenau W, Cohen MB,Vapnek JM, Fong P, et al. Cell proliferation in prostaticadenocarcinoma: in vitro measurement by 5-bromod-eoxyuridine incorporation and proliferation cell nuclearantigen expression. 7 Urol 1993;149:403-7.

34 Oomens EHGM, Van Steenbrugge GJ, Van Der KwastTH, Schroder FH. Application of the monoclonal anti-body Ki-67 on prostate biopsies to assess the prolifera-tive cell fraction of human prostatic carcinoma. J Urol1991;145:81-5.

35 Raymond WE, Leong A S-Y, Bolt JW, Milios Y, Jose JS.Growth fractions in human prostatic carcinoma deter-mined by Ki-67 immunostaining. J Pathol 1988;156:161-7.

36 Sadi MV, Barrack ER. Determination of growth fraction inadvanced prostate cancer by Ki-67 immunostaining andits relationship to the time to tumor progression afterhormonal therapy. Cancer 199 1;67:3005-71.

37 Visakorpi T. Proliferative activity determined by DNAflow cytometry and proliferating cell nuclear antigen(PCNA) immunohistochemistry as a prognostic factor inprostatic carcinoma. J7 Pathol 1992;168:7-13.

38 Zalatnai A, Paz-Bouza JI, Redding TW, Schally AV.Histologic changes in the rat prostate cancer model aftertreatment with somatostatin analogs and D-Trp-6-LH-RH. Prostate 1988;12:85-98.

39 Sakamoto S, Tajima M, Sawaki K, Suzuki S, Kuolo M,Sassa S, et al. Effects of luteinizing hormone-relatedhormone analogue on DNA synthesis in rat prostate anduterus. In Vivo 1993;7:13-6.

40 Akakura K, Bruchovsky N, Goldenberg N, Rennie PS,Buckley AR, Sullivan LD. Effects of intermittent andro-gen suppression on androgen-dependent tumors. Cancer1993;71 :2782-90.

913

on Decem


http://jcp.bmj.com

/J C


ctober 1994. Dow

nloaded from

http://jcp.bmj.com/

effects - from bmj and acp

Documents