cyclin d1 gene amplification and overexpression are present in ductal carcinomain situ of the breast

J. Pathol. 187: 279–284 (1999)

CYCLIN D1 GENE AMPLIFICATION ANDOVEREXPRESSION ARE PRESENT IN DUCTAL

CARCINOMA IN SITU OF THE BREAST

. . 1, . 1, . 2, . 1 . 1,2*

1Department of Pathology, Leiden University Medical Centre, P.O. Box 9600, 2300 RC Leiden, The Netherlands2Department of Pathology, Netherlands Cancer Institute, Plesmanlaan 121, 1066 CX Amsterdam, The Netherlands

SUMMARY

Cyclin D1 (CCND1) amplification is found in 10–15 per cent of invasive breast carcinomas, but it is not well established whether thisgene alteration also occurs in the precursor of invasive breast carcinoma, ductal carcinoma in situ (DCIS). By Southern blot analysis,cyclin D1 gene amplification was detected in 10 per cent (3/32) of DCIS cases. In addition, 15 cases of DCIS were analysed using brightfield in situ hybridization (BRISH), of which 11 had already been analysed by Southern blotting. One additional case with geneamplification was found by BRISH. The use of BRISH for the detection of gene amplification is shown to be a novel and reliable in situmethod on paraffin-embedded tissue sections. By immunohistochemistry, 147 cases of DCIS were analysed for the expression of cyclinD1. Cyclin D1 overexpression was found in 9 per cent of well-differentiated, 29 per cent of intermediately differentiated, and 19 per centof poorly differentiated DCIS. No statistically significant association was found between cyclin D1 overexpression and the differentiationgrade of DCIS, although 90 per cent of the cases that show overexpression are classified as intermediately and poorly differentiated. Anassociation was found between cyclin D1 overexpression and oestrogen receptor positivity. Cyclin D1 overexpression was found in allfour cases with cyclin D1 gene amplification, but was also found in 30 per cent (8/27) of cases without detectable gene amplification. Itis concluded that cyclin D1 gene amplification is an early event in the development of breast carcinoma and occurs in poorlydifferentiated DCIS. Cyclin D1 protein overexpression is also present in tumours without cyclin D1 gene amplification and is seenpredominantly in DCIS of intermediately and poorly differentiated histological type and oestrogen receptor positivity. Copyright ?1999 John Wiley & Sons, Ltd.

KEY WORDS—DCIS; cyclin D1; breast cancer; gene amplification; ISH; protein expression

INTRODUCTION

Cyclin D1 is a critical factor for cell cycle progressionand deregulation of cyclin D1 protein expression hasbeen implicated in the development of a number ofmalignancies, including breast cancer. Cyclin D1 geneamplification is observed in 10–15 per cent of primaryinvasive breast cancers1–7 and occurs predominantlyin oestrogen receptor (ER)-positive tumours.8,9 Apossible explanation for this relationship is that cyclinD1 protein can directly bind to the ER, which thenbecomes activated and up-regulates ER-mediated tran-scription.10 In invasive breast cancer, the frequency ofcyclin D1 protein and mRNA overexpression exceedsthe frequency of DNA amplification.11–13 Recently,Weinstat-Saslow et al.14 reported increased cyclin DmRNA levels at the transition of premalignant atypicalductal hyperplasia to in situ and invasive carcinoma ofthe breast, suggesting a role for cyclin D in the initiationof a malignant phenotype. It is therefore of interest toassess the involvement of cyclin D1 in ductal carcinomain situ (DCIS).

CCC 0022–3417/99/030279–06 $17.50Copyright ? 1999 John Wiley & Sons, Ltd.

MATERIALS AND METHODS

Patients and tissues

The study was performed on 147 paraffin-embeddedcases of pure DCIS which were retrieved from thearchives of the Departments of Pathology of TheNetherlands Cancer Institute (NCI), Leiden UniversityMedical Center (LUMC), and University HospitalNijmegen. The tumours were classified according to theclassification described by Holland et al.15 as 67 poorly-,48 intermediately-, and 32 well-differentiated DCIS.

*Correspondence to: Dr M. J. van de Vijver, Department ofPathology, Leiden University Medical Centre, P.O. Box 9600, 2300RC Leiden, The Netherlands.

Contract/grant sponsor: Dutch Cancer Society; Contract/grantnumber: KWF 94-757.

Immunohistochemical analysis

Immunohistochemical assays were performed on 5 ìmtissue sections mounted on APES-coated slides. Sectionswere dewaxed in xylene and dehydrated through agraded series of alcohol, followed by quenching ofendogenous peroxidase in 0·3 per cent hydrogenperoxide/methanol for 20 min. Cyclin D1 expression wasdetected using the monoclonal antibody clone DCS6(MS-210-P1 NEO Markers, Fremont, CA, U.S.A.) at adilution of 1:500. The optimal concentration was estab-lished by a titration experiment on cyclin D1-positivetumours. Before overnight incubation with the firstantibody, antigen retrieval was performed by boilingin 10 m citrate buffer (pH 6·0) for 10 min using a

Received 13 January 1998Revised 24 April 1998

Accepted 7 September 1998

280 C. B. J. VOS ET AL.

microwave oven. The next day, incubation with aRaMbio followed by avidin–biotin complex conjugatedto horseradish peroxidase (DAKO Ltd., HighWycombe, U.K.) was carried out, followed by incu-bation with diaminobenzidene as a chromogen. Slideswere rinsed thoroughly with phosphate-buffered saline(PBS) between all steps. Finally, the sections werecounterstained with haematoxylin. Tumours withknown cyclin D1 overexpression were used as a positivecontrol.

Scoring was done using a semiquantitative systembased on the staining intensity scored as 0 (none), 1(weak), 2 (moderate), and 3 (strong), and the percentageof positive tumour cell nuclei scored as 0 (0 per cent), 1(1–25 per cent), 2 (25–50 per cent), 3 (50–75 per cent),and 4 (>75 per cent). The cyclin D1 staining score wascalculated as the sum of the intensity and the percentageof positive tumour cells. Expression of the ER wasdetermined using the mouse monoclonal antibody 1D5(DAKOpatts, Glostrup, Denmark) at a dilution of1:500. Prior to the incubation with the first antibody,antigen retrieval was performed using 10 m citratebuffer (pH 6·0)/0·05 per cent detergent (Dishclean,Bosman Chemie, Heijningen, The Netherlands). Anynuclear staining was considered ER-positive andcomplete lack of nuclear staining was consideredER-negative. The staining results were correlated totumour type and ER status using the chi-square test.

Gene amplification

Sample selection—From tissue banks of the NCI andLUMC, 80 freshly frozen samples were collected. Only32 specimens met the requirement of 30 per cent or moretumour cells and were classified as 20 poorly-, sixintermediately-, and six well-differentiated DCIS.

DNA analysis—High molecular weight DNA wasisolated by standard methods.16 DNA from eachtumour and control DNA were digested with the restric-tion endonuclease EcoRI (Pharmacia, Uppsala, Sweden)using buffers recommended by the suppliers. Thedigested DNA was size-fractionated overnight by gelelectrophoresis on a 0·8 per cent agarose gel in 1#TAE,denatured, and transferred with 1 NaCl/0·4 NaOHonto nylon membranes (Hybond N+, Amersham,U.K.). After transfer, the filters were neutralized in 0·5 Tris–HCl/1 NaCl (pH 7·2) buffer, dried, and baked at80)C for 2 h.

Hybridization conditions—Filters were pre-hybridizedfor 0·5 h at 65)C in a hybridization mix [1 Na2HPO4/1 NaH2PO (pH 7·2)/7 per cent SDS/0·5 EDTA].Cyclin D1 amplification was tested with a c-DNA cloneU21B31, a 1·45 kb EcoRI/EcoRI fragment, represent-ing the 3* non-translated region of the cyclin D1 gene.7The probe was radiolabelled with 20 ìCi of[á-32P]dCTP(>3000 Ci/mmol; Amersham, U.K.) usingthe random-primed labelling kit (Pharmacia, Uppsala,Sweden). After 17 h of hybridization, the filters werewashed at 65)C to a final stringency of 0·1#SSC/0·1 per

Copyright ? 1999 John Wiley & Sons, Ltd.

cent SDS. The filters were exposed to Kodak Xomat ARfilms with DuPont Cronex Lightning Plus screens for2–5 days at "70)C. The degree of amplification ofcyclin D1 was determined visually and with a PhosphorImager (Molecular Dynamics) by comparison with theprobe for the N-cam locus (11q23), as an internalcontrol for gene copy number. The same filters werehybridized successively. Amplification was defined asan increase of copy number of cyclin D1 comparedwith the copy number of N-cam by a factor of atleast 2.

Bright field in situ hybridization (BRISH) onparaffin-embedded tissue

Probe/cosmid and labelling—A (peri) centromericrepetitive satellite DNA probe of chromosome 11(pLC11A, alphoid sequence17) and a cosmid specific forcyclin D1 (cosmid 3123 containing the completegenomic DNA sequence of cyclin D118) were used andlabelled with digoxigenin (Boehringer Mannheim) bynick translation.

Bright field in situ hybridization (BRISH)—For 15paraffin-embedded tumour samples, of which 11 hadalso been analysed using Southern blot analysis, BRISHwas performed as described by Alers et al.19 with somemodifications. Briefly, after deparaffinization of the4 ìm tissue sections in xylene, endogenous peroxidaseactivity was blocked with 0·3 per cent H2O2 in methanol.Pretreatment was carried out for 30 min in 2#SSC at70)C. Enhancement of tissue permeability was achievedby proteolytic enzyme digestion (0·4 per cent pepsin,Sigma) in 0·2 HCl. The optimal digestion time foreach tumour block was determined by a pepsin timeseries (5–20 min). Nuclear DNA was denatured at 72)Cfor 2 min in 70 per cent deionized formamide in 2#SSC(pH 7·0), followed by dehydration in an ethanol series.The chromosome 11-specific DNA probe and cyclinD1 cosmid were denatured for 10 min at 70)C in ahybridization mixture containing 5 ng/ìl probe DNA.The cyclin D1 cosmid was pre-annealed for 30 min at37)C. The chromosome 11-specific probe and cyclin D1cosmid were applied to successive sections and incu-bated overnight at 37)C. The slides were washed in50 per cent formamide in 2#SSC (pH 7·0) at 42)C.Slides were subsequently incubated for 30 min at 37)Cwith mouse anti-digoxigenin (Sigma Bio Sciences),biotin-labelled goat anti-mouse (30 min at 37)C) andavidin–biotin complex (30 min at 37)C, DAKO ABCkit). The signal was amplified using deposition of bioti-nylated tyramine (BT) at the location of the DNAprobe/cosmid. The biotin-labelled tyramine was synthe-sized according to Adams20 and incubated at a dilutionof 1:1000 for 10 min, followed again by avidin–biotincomplex (20 min at room temperature). Visualizationwas achieved with 0·05 per cent H2O2, 0·5 g/l diamino-benzidine (DAB) in 0·05 Tris–HCl (pH 7·6). Slideswere rinsed in distilled water and counterstained withhaematoxylin. Finally, slides were dehydrated in

J. Pathol. 187: 279–284 (1999)

281CYCLIN D1 GENE AMPLIFICATION AND OVEREXPRESSION IN DCIS

ethanol, cleared in xylene, and mounted in Pertex(Klinipath).

As a reference for normal spot distribution, 100 intactand non-overlapping 4 ìm nuclei, hybridized with achromosome 11 centromere-specific probe, were countedand the number of solid DAB spots per nuclear frag-ment was scored (0, 1, 2, 3, 4, >4 spots per nuclear slice).The spot distribution was statistically evaluated bymeans of the Kolmogorov–Smirnov test21 to detectploidy status. Distribution shifts to the left (decreasingnumber of spots) or right (increasing number of spots)compared with the normal distribution were consideredaneuploid.

To detect amplification due to double minutes (DMs),we compared the number of spots present in the tumourcells with those present in the nuclei of the normalepithelial cells present on the same slide. Amplificationdue to a gain of intensity of the spots in the tumour cellscompared with the normal epithelium was considered tobe caused by homogeneously staining regions (HSRs).

RESULTS

Immunohistochemical detection of cyclin D1

Cyclin D1 protein overexpression was determined in147 cases of DCIS. Cyclin D1 staining was mainlylocated in the nucleus, showing a range of stainingintensities. Occasionally, weak staining of some of the


normal epithelial cells was observed. As the biologicalrelevance of the various levels of cyclin D1 (over)expression is not known, Table I shows the percentageof cyclin D1-positive cases for each of the different typesof DCIS, using different cut-off points. To study thecorrelation between cyclin D1 overexpression and otherparameters, the cut-off point of §4 was chosen, becausethe cases with gene amplification all had a proteinexpression level of §4.

Cyclin D1 overexpression was detected in 21 per cent(31/147) of all DCIS cases, subdivided into 9 per cent(3/32) well-differentiated, 29 per cent (14/48) inter-mediately differentiated, and 19 per cent (13/67) poorlydifferentiated DCIS. Cyclin D1 positivity was not sig-nificantly correlated with differentiation grade, although90 per cent of the cyclin D1-positive cases are inter-mediately or poorly differentiated (Table II). A signifi-cant dependent correlation was found betweenoverexpression of cyclin D1 and positive ER status(Table III). Well-differentiated DCIS shows a significantcorrelation with ER positivity (Table IV).

Gene amplification

Table I—Cyclin D1 overexpression in the different histological DCIS types. Different cut-off points for thestaining score (intensity+amount of cells expressing) were used

Staining score

§3 §4 §5 §6

Well-differentiated DCISCyclin D1+ 8 3 0 0Cyclin D1" 24 29 32 32Total 32 (25%) 32 (9%) 32 (0%) 32 (0%)

Intermediately differentiated DCISCyclin D1+ 23 14 7 3Cyclin D1" 25 34 41 45Total 48 (48%) 48 (29%) 48 (15%) 48 (6%)

Poorly differentiated DCISCyclin D1+ 30 13 6 5Cyclin D1" 37 54 61 63Total 67 (45%) 67 (19%) 67 (9%) 67 (7%)

Table II—Association between cyclin D1 overexpression and histological type of DCIS (p=0·1 usingchi-square test)

DCIS

Well differentiated Intermediately differentiated Poorly differentiated Total

Cyclin D1+ 3 14 13 30Cyclin D1" 29 34 54 117

147

Southern blotting—Estimates of DNA copy numberwere obtained by comparing the auto-radiogram signalfor cyclin D1 (7·5 kb) with that of the control markerN-cam (7·5 and 8 kb), located on a non-amplified regionof chromosome 11q23 (Fig. 1). Amplification of cyclin

J. Pathol. 187: 279–284 (1999)


D1 was observed in 10 per cent, all poorly differentiatedcases of DCIS.

BRISH—In 15 cases of DCIS, we could reveal ampli-fication of cyclin D1 in 26 per cent (4/15) of the cases. Anormal diploid distribution of chromosome 11 wasfound in all 15 cases. In all cases of amplification, a gain


of spot intensity was observed in the tumour nucleicompared with the normal epithelial nuclei present onthe same slide (Fig. 2c). When using biotin tyramine(BT), we could not make a distinction between amplifi-cation due to homogeneously staining regions (HSRs) ordouble minutes (DMs) because of the diffuse signaldistribution throughout the nucleus. However, when weperformed BRISH without amplification using BT, wecould distinguish multiple spots (indicating the presenceof DMs) in the nucleus instead of a gain of intensity ofone of the spots (indicating the presence of HSRs) (datanot shown).

The BRISH results were compared with thoseobtained by Southern blot and immunohistochemicalstaining (Table V). Concordance between gene amplifi-cation detected by Southern blotting and BRISH andIHC overexpression of cyclin D1 was shown in threecases: BT 1204, BT 1271, and BT 1282. Discordance wasseen in one case (BT 1213) in which Southern blottingfailed to detect amplification, whereas BRISH revealed again of spot intensity in the tumour cell nuclei comparedwith normal nuclei (Fig. 2c). This tumour also showedoverexpression of cyclin D1 at the protein level (Fig. 2a).Four tumours showed overexpression of cyclin D1 pro-tein (BT 1208, BT 1404, BT 1460, and BT 1462),but amplification was not detected by BRISH or bySouthern blot analysis.

Table III—Association between cyclin D1 staining and oestro-gen receptor (ER) status (p=0·02 using chi-square test)

ER status

Cyclin D1

Negative (<4) Positive (§4) Total

Negative 47 (40%) 5 (16%) 52Positive 70 (60%) 25 (84%) 95

147

Table IV—Association between oestrogen receptor (ER) statusand histological type of DCIS (p<0·0001 using chi-square test)

DCIS

ER

Positive Negative Total

Well differentiated 29 3 32Intermediately differentiated 35 13 48Poorly differentiated 27 40 67

147

Fig. 1—Detection of cyclin D1 amplification by Southern blot analysis. A Southern blot is shownof EcoRI digests of DNA from 11 samples of DCIS (lanes 1–9, 15, and 16) and from a squamouscell (UMSSC2) and two breast carcinoma cell lines (MDA134 and SKBR3), a cell line derivedfrom normal mammary tissue (HBL100) and normal placental tissue (lanes 10–14). The Southernblot was hybridized successively with a probe for cyclin D1 and n-cam. Asterisks indicate the DCISsamples that showed cyclin D1 gene amplification determined by a two-fold or more increase incopy number of cyclin D1 compared with N-cam. The relative increase in cyclin D1 copy numberis shown at the bottom of the figure

DISCUSSION

We have studied cyclin D1 aberrations in DCIS bythree different techniques: overexpression of cyclin D1

J. Pathol. 187: 279–284 (1999)

283CYCLIN D1 GENE AMPLIFICATION AND OVEREXPRESSION IN DCIS

Fig. 2—Immunohistochemical staining and BRISH of cyclin D1 in the same case of DCIS. (a) Strong staining of poorly differentiated DCIS;(b) BRISH with a chromosome 11-specific (peri)centromeric probe; (c) BRISH with the cyclin D1 specific cosmid

Table V—Comparison of BRISH results with the results from Southern blot analysis and immunohisto-chemical staining for evaluation of cyclin D1 gene amplification in DCIS

Patient No. Histological type Southern blot

IHC BRISH

int exp #11 Cyclin D1

BT 1201 Poorly diff. " 0 0 N "BT 1204 Poorly diff. + 3 4 N +BT 1208 Poorly diff. n.d. 2 2 N "BT 1211 Poorly diff. " 1 2 N "BT 1213 Poorly diff. " 3 3 N +BT 1253 Poorly diff. " 2 3 N "BT 1254 Intermediately diff. " 0 0 N "BT 1271 Poorly diff. + 3 3 n.b. +BT 1273 Intermediately diff. " n.d. n.d. N "BT 1282 Poorly diff. + 1 3 N +BT 1290 Well diff. " 0 0 N "BT 1308 Well diff. n.d. 1 2 N "BT 1404 Intermediately diff. n.d. 3 2 N "BT 1460 Well diff. n.d. 2 2 N "BT 1462 Poorly diff. " 2 2 N "

n.d.=not done; n.b.=not detectable; + =amplification found; " =no amplification found;IHC=immunohistochemistry; int=intensity of the staining; exp=amount of cells expressing cyclin D1;diff.=differentiated; BRISH=bright field in situ hybridization; #11=chromosome 11 specific probe; Cyclin D1=cyclinD1 cosmid; N=normal distribution of signals in the nucleus.

protein in 147 cases of DCIS and cyclin D1 geneamplification in 32 and 15 cases, respectively.

Recently using FISH,22 cyclin D1 gene amplificationin DCIS was reported in 8 per cent of low grade, 14 percent of intermediate-, and 32 per cent of high-gradeductal carcinoma in situ. The percentage of DCIS caseswith cyclin D1 gene amplification in our study is lower,which may be due to differences in techniques and DCISclassification.15,23 As in our study, an association ofcyclin D1 gene amplification and poorly differentiatedDCIS was found.

Immunohistochemical cyclin D1 overexpression(using monoclonal antibody DCS6 in different concen-trations) has been studied in invasive breast cancer. In astudy by Worsley et al.,13 it was found that in 73 per centof invasive breast carcinomas overexpression of cyclinD1 was present. No correlation was found betweenoverexpression and gene amplification. In a study byGillett et al.,12 it was found that 35 per cent invasive


breast carcinomas showed overexpression of cyclin D1.The scoring method used11 is comparable to oursalthough they looked only at the staining intensity of themajority of the cells and did not combine the percentageof positively stained cells; moderate, strong, and verystrong cyclin D1 protein staining was correlated withgene amplification. Simpson et al.22 also looked atprotein expression using a 1:10 dilution and found thatlow grade had a higher mean percentage of positivenuclei than intermediate- and high-grade DCIS(p=0·007). In contrast, we found protein overexpressionpredominantly in intermediately and poorly differ-entiated ER-positive DCIS. The variations in the per-centage of cyclin D1-positive cases between the differentstudies is not easily explained. There is a differencebetween the antibody dilutions used by Simpson et al.(1:10) and most other studies, including ours (1:500–1:2000). This may be the result of different antibodybatches.

J. Pathol. 187: 279–284 (1999)


There are several techniques to detect gene amplifi-cation in paraffin-embedded tissue, including FISH onisolated nuclei and paraffin sections and differentialPCR.24–27 We have used BRISH as a novel and sensitivein situ technique to analyse gene amplification. BRISHhas the advantage over the other techniques that it ispossible to analyse gene amplification using lightmicroscopy. BRISH has enabled us to study gene ampli-fication in tumours that could not be analysed usingSouthern blotting, because the percentage of tumourcells was too low (<30 per cent) or heterogeneous.Although BRISH might be a more sensitive techniqueto detect amplification, there is still a proportion oftumours overexpressing cyclin D1 protein, but in whichno DNA amplification could be detected by either of thetechniques used. These data suggest that in concordancewith invasive breast carcinoma, an alternative mech-anism, distinct from DNA amplification, may causeup-regulation of cyclin D1 expression.

We conclude that cyclin D1 deregulation by amplifi-cation and other mechanisms already occurs in DCISand is therefore an early event in breast cancer develop-ment. Furthermore, cyclin D1 overexpressing DCISrepresents an interesting subset of DCIS, since it com-prises predominantly intermediately and poorly differ-entiated ER-positive cases of DCIS. The fact thatintermediately and poorly differentiated DCIS is associ-ated with negative ER suggests that tumours withup-regulated cyclin D1 comprise a separate entity withinthe heterogeneous group of breast tumours.

ACKNOWLEDGEMENTS

We wish to acknowledge Dr E. Schuuring for provid-ing us with the c-DNA clone U21B31 and cosmid 3123;Dr J. C. Alers and Dr H. Van Dekken for sharing theBRISH technique; Dr R. Holland for providing some ofthe cases of DCIS; Dr J. Hermans for the statisticalanalysis; and Dr A.-M. Cleton-Jansen for critical read-ing of the manuscript. This work was supported by theDutch Cancer Society (KWF 94-757).

REFERENCES1. Lidereau R, Callahan R, Dickson C, Peters G, Escot C, Ali I. Amplification

of the int-2 gene in primary human breast tumors. Oncogene Res 1988; 2:285–291.

2. Adnane J, Gaudray P, Simon M-P, Simon-Lafonaine J, Jeanteur P, TheilletC. Proto-oncogene amplification and human breast tumor phenotype.Oncogene 1989; 4: 1389–1395.

3. Tsuda H, Hirohashi S, Shimosato Y, et al. Correlation between long-termsurvival in breast cancer patients and amplification of two putativeoncogene-coamplification units hst1/int-2 and c-erbB-2/ear1. Cancer Res1989; 49: 3104–3108.


4. Fantl V, Richards MA, Smith R, et al. Gene amplification on chromosomeband 11q13 and estrogen receptor status in breast cancer. Eur J Cancer1990; 26: 423–429.

5. Theillet C, Adnane J, Szepetowski P, et al. BCL-1 participates in the 11q13amplification found in breast cancer. Oncogene 1990; 5: 147–149.

6. Lammi GA, Fantl V, Smith R, et al. D11S287, a putative oncogene onchromosome 11q13, is amplified and expressed in squamous cell andmammary carcinomas and linked to BCL-1. Oncogene 1991; 6: 439–444.

7. Schuuring E, Verhoeven E, Mooi WJ, Michalides RJAM. Identification andcloning of two overexpressed genes, U21B31/PRAD1 and EMS1, within theamplified chromosome 11q13 region in human carcinomas. Oncogene 1992;7: 335–361.

8. Fantl V, Smith R, Brookes S, Dickson C, Peters G. Chromosome 11q13abnormalities in human breast cancer. Cancer Surv 1993; 18: 77–94.

9. Peters G, Fantl V, Smith R, Brookes S, Dickson C. Chromosome 11q13markers and D-type cyclins in breast cancer. Breast Cancer Res Treatment1995; 33: 125–135.

10. Zwijsen RM, Wientjens E, Klompmaker R, van der Sman J, Bernards R,Michalides RJAM. Cdk independent activation of estrogen receptor bycyclin D1. Cell 1997; 88: 405–415.

11. Buckley MF, Sweeney KJ, Hamilton JA, et al. Expression and amplificationof cyclin genes in human breast cancer. Oncogene 1993; 8: 2127–2133.

12. Gillett C, Fantl V, Fisher C, et al. Amplification and overexpression ofcyclin D1 in breast cancer detected by immunohistochemical staining.Cancer Res 1994; 54: 1812–1817.

13. Worsley SD, Jennings BA, Khalil KH, Mole M, Girling AC. Cyclin D1amplification and expression in human breast carcinoma: correlation withhistological prognostic markers and oestrogen receptor expression. J ClinPath Mol Pathol 1996; 49: M46–M50.

14. Weinstat-Saslow D, Merino MJ, Manrow RE, et al. Overexpression ofcyclin D mRNA distinguishes invasive and in situ breast carcinomas fromnon-malignant lesions. Nature Med 1995; 12: 1257–1260.

15. Holland R, Peterse JL, Millis RR, et al. Ductal carcinoma in situ: a proposalfor new classification. Semin Diagn Pathol 1994; 11: 167–180.

16. Sambrook J, Fritsch EF, Maniatis T. Molecular cloning: A LaboratoryManual. New York: Cold Spring Harbour Laboratory Press, 1989.

17. Willard HF, Smith KD, Sutherland J. Isolation and characterization of amajor tandem repeat family from the human X-chromosome. Nucleic AcidRes 1983; 11: 2017–2033.

18. Vaandrager JW, Kluin PH, Schuuring E. The t(11;14)(q13;q32) in multiplemyeloma cell line KM12 has its 11q13 breakpoint 330 kb centromeric fromthe cyclin D1 gene. Blood 1997; 89: 349–350.

19. Alers JC, Krijtenburg J, Vissers KJ, Bosman FT, van der Kwast TH, vanDekken H. Interphase cytogenetics of prostatic adenocarcinomas andprecursor lesions: analysis of 25 radical prostatectomies and 17 adjacentprostatic intra-epithelial neoplasias. Genes, Chromosomes Cancer 1995; 12:241–250.

20. Adams JC. Biotin amplification of biotin and horseradish peroxidase signalsin histochemical stains. J Histochem Cytochem 1992; 40: 1457–1463.

21. Young IT. Proof without prejudice: use of the Kolmogorov-Smirnov testfor the analysis of histograms from flow systems and other sources.J Histochem Cytochem 1977; 25: 935–944.

22. Simpson JF, Quan DE, O’Malley F, Odom-Maryon T, Clarke PE.Amplification of CCND1 and expression of its protein product, cyclin D1,in ductal carcinoma in situ of the breast. Am J Pathol 1997; 151: 161–168.

23. Silverstein MJ, Poller DN, Waisman JR, et al. Prognostic classification ofbreast ductal carcinoma-in-situ. Lancet 1995; 345: 1154–1157.

24. Misra DN, Dickman PS, Yunis EJ. Fluorescence in situ hybridization(FISH) detection of myc-N oncogene amplification in neuroblastoma usingparaffin embedded tissue. Diagn Mol Pathol 1995; 4: 128–135.

25. Pauletti G, Godolphin W, Press MF, Slamon DJ. Detection and quantita-tion of HER2/neu gene amplification in human breast cancer archivalmaterial using fluorescence in situ hybridization. Oncogene 1996; 13: 63–72.

26. Li BDL, Harlow SP, Budnick RM, Sheedy DL, Stewart CC. Detection ofHER2/neu oncogene amplification in flow cytometry-sorted breast ductalcells by competitive polymerase chain reaction. Cancer 1994; 73: 2771–2778.

27. Deng GR, Yu M, Chen LC, et al. Amplification of oncogene erbb2 andchromosome 20q in breast cancer determined by differentially competitivepolymerase chain reaction. Breast Cancer Res Treatment 1996; 40: 271–281.

J. Pathol. 187: 279–284 (1999)

cyclin d1 gene amplification and overexpression are present in ductal carcinomain situ of the breast

Documents