Molecular genetics and cytogenetics of breast carcinomas: Comparison of the two methods

ELSEVIER

Molecular Genetics and Cytogenetics of Breast Carcinomas: Comparison of the Two Methods

Rut Valgardsdottir, Margret Steinarsdottir, Kesara Anamthawat-Jonsson, Ingibj6rg Petursdottir, Helga M. ()gmundsdottir, and J6runn E. Eyfj6rd

ABSTRACT: Molecular genetics and cytogenetics are two different approaches to studying genetic changes in breast carcinoma. We have used karyotype analysis, fluorescence in situ hybridization, and molecular analy.~is of allelic imbalance on chromosomes 7q and 16q and on both arms of chromosome 17, to study 85 breast carcinomas. Twenty-five of these samples gave results that could be used to compare the two methods. S&ty-nh3e chromosome arms were compared, of which 48 (70%) gave concordant molecular ,~nd cytogenetical results. Samples were processed for karyotyping both by harvesting directly from the fresh tissue and after selective culture for a few days. Karyotypes among the direct harvest samples mc!tched significantly better with the molecular genetics results than karyotypes among the cultured cell preparations. © Elsevier Science Inc., 1996

INTRODUCTION

Over the past decade, increasing insight has been gained into the molecular genetics of breast cancer. It has been shown that specific mutations, deletions of tumor suppressor genes, or amplification of oncogenes occur in can- cers, although the order in which these changes take place does not seem to be fixed. Conventional cytogenetic methods and, recently, fluorescent in situ hybridization (FISH) have shown that the breast cancer cell karyotype can be very complex [1, 2]. It has not yet been possible to associ- ate breast cancer with specific chromosomal aberrations, but some candidates are beginning to emerge [3]. The cytogenetic findings can sometimes lead to the discovery of specific genes, as was the case with the Philadelphia chromosome in leukemia [4], and it is widely believed that cytogenetics can be used to identify molecular events and sites of possible tumor suppressor genes and oncogenes. The question thus arises whether these two methods, molecular genetics and cytogenetics, reflect each other.

The nature of the results obtained using these two methods, cytogenetics arid molecular genetics, are quite different. The strength of cytogenetics is that changes can

From the Molecular and Cell Biology Research Laboratory, Icelandic Cancer Society (R. V., L P., H. M. 0., ]. E. E.), and Cyto- genetics Laboratory, Department of Pathology, University Hospi- tal of Iceland (M. S., K. A.-J.), Reykjavik, Iceland.

Address reprint requests to: Jdrunn E. Eyfj6rd, Molecular and Cell Biology Research Labon~tory, Icelandic Cancer Society, P.O. Box 5420, 125 Reykjavik, Iceland.

Received December 20, 1995; accepted May 13, 1996.

Cancer Genet Cytogenet 92:37-42 (1996) © Elsevier Science Inc., 1996 655 Avenue of the Americas, New York, NY 10010

be observed on all chromosomes simultaneously and, since individual cells are screened, different clones can be identified and, thus, the tumor heterogeneity demonstrated. The weakness, however, is that only structural changes of certain magnitude can be seen, such as losses or gains of whole chromosome bands. Furthermore, since only those cells that are in metaphase of mitosis can be success- fully karyotyped, the analysis is based on a very small pro- portion of each tumor. Molecular genetics, on the other hand, give the opportunity to diagnose very small changes, down to the change of one base, but usually only for one or a few chromosomes for each sample, since allelotyping all the chromosomes is very labor- and DNA-consuming. As the DNA is obtained from a mixture of tumor cells, all changes observed represent an idealized average of the tumor [5]. The tumor 's heterogeneity and complex but balanced changes are not seen. Thus, the strength of molecular genetics is the weakness of cytogenetics and vice versa. It has recently been suggested that the technique of molecular cytogenetics, i.e., FISH, is sensitive and can be used for the detection of chromosome changes in breast cancer [6, 7]. It combines the advantages of the other two methods, solving many of the problems encountered.

We have analyzed breast carcinomas with conventional karyotyping as well as whole chromosome painting for selected chromosomes [1, 8], and found that these methods support and complement each other [8]. The same tumors were also analyzed for allelic imbalance on chromosomes 7, 16, and 17. Results with these two approaches, molecular genetics and cytogenetics, were compared in order to see if they gave similar results. This is a question that has been raised several times [5, 9] but only a few attempts

0165-4608/96/$15.00 PII S0165-4608(96)00150-1

38 R. Valgardsdott i r et al.

have been made to answer it [10, 11]. So far, the compari - son has been made mainly by searching for a possible rela- t ionship between homogeneous ly staining regions (HSRs) and amplif icat ions observed with molecular genetic methods. Our goal, on the other hand, was to see whether the two methods can be compared and if so, whether they match in general, focusing on four chromosome arms.

MATERIALS AND METHODS

Patients and Samples Consecutive samples from 85 breast carcinomas were obtained, immedia te ly after surgery, from the depar tment of Pathology, Universi ty Hospi tal of Iceland. Half of each sample was put aside for DNA analysis and the remainder was processed either by direct harvest or t issue culture, for use in chromosome analysis [1].

Only cases wi th clonal chromosomal abnormali t ies and molecular genetic results were included in this study. Cases with single cell abnormali t ies, only, were excluded. Fur- thermore, cases with normal karyotypes were excluded, as they might represent karyotypes obtained only from the normal cells in the sample. This left 25 samples for comparison of cytogenetic and molecular genetic results.

Cytogenetic Analysis Tumors were harvested direct ly or cul tured for 6-7 days on average before harvesting, as descr ibed previously [1]. Chromosomes were G-banded with Wright stain fol lowing the method of Pandis et al. [12]. Analysis of karyotypes was performed according to ISCN [13].

Fluorescence In Situ Hybridization PCR ampl i f ied whole chromosome paint ing probes for chromosomes 16 and 17 (Cambio Ltd, UK) were used for FISH of selected samples [8]. The probes were direct ly labeled with FITC over anti-FITC. Chromosomes were counters ta ined with DAPI.

Molecular Analysis Allel ic imbalance was s tudied with six different Southern blot probes and 11 different PCR po lymorph i sms (Table 1). Information on markers was obtained from the Genome Data Base (GDB).

netics results. The results were said to match when there was a loss or gain of whole chromosome or chromsome band and imbalance at most or all of the markers, or when there was no cytogenetic change observed, and imbalance at none of the markers. Since small changes are not detect- able wi th the cytogenetic methods used here, imbalance at one of 4 -6 markers was cons idered as site-specific change, and, when this co inc ided with no cytogenetical change, the results were said to match. Samples wi th imbalance at one out of two informative markers were excluded, as they give insufficient information for compar ison with the cytogenetic results, whether or not cytogenetical changes were found. The criteria used for combining results for each sample (Table 2) were that a sample was said to match if more than half of the chromosome arms matched (e.g. two out of three or three out of four) (Table 3).

Statistics Fisher ' s exact test was app l i ed for testing for association.

RESULTS

Comparison Cytogenetical results were obtained for 100 chromosome arms 7q, 16q and 17p/q (Table 2). For seven of these, the cytogenetical results were too complex for compar ison (i.e. it was not possible to determine if there had been gains or losses) and 12 chromosome arms had no informative molecular markers and, therefore, could not be used for the comparison. Twelve chromosome arms were further exc luded as they showed imbalance at only one out of two informative markers. This left 69 chromosome arms that could be used for the comparison.

Matching molecular and cytogenetical results were obtained for 48 chromosome arms (70%). Aneuploid samples

Table I Markers used for molecular analysis

Locusname Locussite Type °

D7S522 D7S495 D7S523 D16S261

Comparison of Molecular Genetic and D16S265 Cytogenetic Results D16S422

The number of whole chromosomes or chromosome parts D17S34 was counted for chromosomes number 7, 16, and 17. If D17S5

TP53 numer ica l or structural changes were observed in these p53MS b chromosomes, we evaluated whether they were ba lanced THRA1 or not, using Dutri l laux et al. 's hypothesis on chromosomal D17S520 alterations occurring during tumor progression [14]; this D17S579 assumes that chromosome losses occur first, then endoredu- D17S74 plication, fol lowed by further chromosome losses, a sec- D17S4 ond endoredupl ica t ion , and so on. D17S24

As we have previously shown that karyotypes and D17S802 FISH give s imilar results for the samples used here [8], cytogenetics results, as dec ided from combined karyotyping and FISH analysis, were compared with the molecular ge-

7q31 M 7q31-35 M 7q31 M

16q12.21 M 16q21 M 16q23-24 M 17p13.3 S 17p13.3 S 17p13.1 M 17p13.1 M 17q11.2 S 17q12-21 M 17q12-21 M 17q12-21 S 17q23-25 S 17q23-25 S 17q23-35 M

aTypes of markers are M=Microsatellite markers and S=Southern Blotting markers

bSantibanez-Koref, personal information

M o l e c u l a r Gene t i c s and Cytogene t i c s of Breast C a r c i n o m a

Tab le 2 C o m p a r i s o n of the cy togene t i c a n d m o l e c u l a r gene t ic resu l t s

39

Range of chromo- Cytogenetic results a Molecular results b

No. somes/no, cells 7 16 17 7q 16q 17p 17q

DHPs 1 45/5 1 1 2 1/26 2/2 c 0/2 0/4 2 48-51/10 2 1 1 0/1 1/1 1/1 1/1 3 33-47/3 f 2 2 2 1/2 0/3 1/1 1/5 4 45-47/2 2 2 2 0/1 0/3 0/1 1/6 5 ?80/2 f 4 2id 3, +q12-qter 0/1 3/3 1/2 1/4 6 63-69/5 f 3 3 4, (2 ×HSR) e - - 0/2 3/3 4/5

7 59-65/14 f 0 NI 2id, +2xq 0/1 2/2 - - 1/4 8 50-63/12 f NI NI NI 0/1 - - 3/3 3/3

9 92/5 4 4 4 1/2 0/1 1/2 1/2

CCPs 10 44-45/9 2 2 1, -p13 0/2 - - - - 0/1 11 43-46/9, idemx2/5 f 2, (HSR q32-36) e 2 1 1/1 - - 2/3 2/3 12 44-47/24 2 1 1, - p l l 2/2 0/1 1/1 3/3 13 46/11 2 2 2 - - - - 1/2 0/1 14 48/7 2 2 2 2/2 0/3 0/1 1/5 15 47/2 2 2 2 1/1 1/2 2/2 2/5 16 46/5 2 2 2 0/2 0/1 1/1 0/4 17 45/3 2 2 2 1/1 2/2 0/1 2/4 18 39-40/4 f 2 2 2 0/1 1/2 2/2 4/5 19 45-53/4 2 2 2 0/1 1/2 - - 0/3 20 75-86/50 cp 2, +p 2id, +q 0/1 0/1 - - 1/3 21 64-71/5 f G 2id, +q 3-4 1/1 2/2 - - 0/2 22 75-84/8 f 2, -q11, -q32 3, +p 2id 1/2 0/1 1/1 1/2

23 76-87/18 2id, +p 1, -q21, -q21, -q24 2id, +2xq 3/3 1/3 1/1 2/3 24 106-112/33 f L L G 1/1 2/2 - - 3/3 25 g L cp G 1/3 0/2 1/2 0/4

Abbreviationsl NI = results that could not be interpreted, cp = complex result, L (loss), G (Gain) = loss/gain of some material but difficult to give simple de- scription, or based on FISH results, id = 2 identical chromosomes (of same parental origin) determined using Dutrillaux hypothesis (see Materials and Methods), DHP = direct harvest preparations, CCP = cell culture preparations, - - = no molecular results.

~Cytogenetic results are interpreted as a number of whole chromosomes -~ chromosome parts. Where breakpoints are given, they indicate loss (-)/gain (+) from the breakpoint to terminal end.

~Molecular results are interpreted as a number of markers with imbalance per informative markers for each arm.

~Bold indicates where the two methods match.

Qtalicized numbers indicate where the two methods could not be compared.

eHSR refers to one or more (× 2) of the whole chromosomes.

fClone could not be fully analyzed.

~Three clones were found in 1his sample. Chromosome numbers were 42-43, a3-74(idemx2), and 128-131 (idemx4).

were more difficult to compare than the d ip lo id and near- d ip lo id samples , bu t gave an equal p ropor t ion of ma tch ing

molecu la r gene t ic a n d cy togene t i c resu l t s (72 a n d 68% re- spect ive ly) . Cy togene t i c resu l t s a m o n g the d i rec t ha rves t p r e p a r a t i o n s (DHPs) gave a s ign i f i can t ly be t te r m a t c h to the m o l e c u l a r resu l t s t h a n ka ryo types f rom the c u l t u r e d cel l p r e p a r a t i o n s (CCPs) [p = 0.006] {Table 3a). If resu l t s for each s a m p l e w e r e c o m b i n e d to i nd ica t e w h e t h e r the m a t c h f o u n d for i n d i v i d u a l c h r o m o s o m e a rms corre- s p o n d e d to the ka ryo :ypes , the d i f fe rence o b s e r v e d be- t w e e n m o l e c u l a r gene t i c a n d cy togene t i c resu l t s a m o n g DHPs and CCPs was stil l s ign i f i can t [p = 0.022] (Table 3a).

A more conservat ive evaluat ion exc ludes all c h r o m o s o m e arms w i t h only one infc,rmafive marker. This leaves 40 chro- m o s o m e a rms for the c o m p a r i s o n {Table 3b). W h e n co n s i d - er ing i n d i v i d u a l c h r o m o s o m e arms, the DHP resu l t s m a t c h signif icantly better wi th the molecu la r results than the CCPs

resu l t s [p = 0.013], a n d the d i f f e rence is c lose to be ing sig- n i f i can t [p -- 0.051] w h e n each s a m p l e is c o n s i d e r e d .

C h r o m o s o m e 17

Figure 1 s h o w s a de ta i l ed c o m p a r i s o n of m o l e c u l a r a n d cy togene t i ca l resu l t s for c h r o m o s o m e 17, w h e r e in forma- t ion was o b t a i n e d on 10 loci. The figure reveals m a t c h i n g m o l e c u l a r a n d cy togene t i c resu l t s for 20 out of 25 samples . The e x c e p t i o n s are s a m p l e s 12, 15, 18, 23, a n d 25. These have e i ther i m b a l a n c e at s i tes w h e r e no i m b a l a n c e is e x p e c t e d or b a l an ce w h e r e i m b a l a n c e is expec t ed . Sam- p les 6 a n d 11 have i m b a l a n c e s at m o s t of the i n fo rma t ive markers , but , b e t w e e n the i m b a l a n c e d markers , the re are s i tes w h e r e no i m b a l a n c e is de tec ted . The reverse si tua- t ion is e v i d e n t in o the r s a m p l e s s u c h as 3, 14, 16, a n d 17.

A n e x a m p l e of good fit is s a m p l e 6, w i t h m o l e c u l a r im- b a l an ce on s e v e n out of e ight i n fo rma t ive marke r s on chro-

40 R. Valgardsdottir et al.

T a b l e 3 Comparison of molecular genetic and cytogenetic results between direct harvest and cultured preparations

Proportions with matching results

DHP CCP p

a. All chromosomes arms counted

Chromosome arms 21/23 (91%) 27/46 (59%) 0.006 Samples ° 8/8 (100%) 8/16 (50%) 0.022

b. Only chromosome arms with more than one informative marker counted

Chromosome arms 12/13 (92%) 13/27 (48%) 0.013 Samples a 6/6 (100%) 7/14 (50%) 0.051

Abbreviations: DHP = direct harvest preparations, CCP = cell culture preparations.

OResults for each sample are combined.

mosome 17. One of the Southern blot probes clearly detects amplif icat ion rather than loss of one allele. This is in concordance with the cytogenetic results as HSR was found on chromosome 17 (Table 2). A possible case of non-fit is sample 9 with 4 chromosomes 17, but only imbalance on two out of four informative loci. The "lost"

markers are located on two different sites of the long arm and, thus, might represent site-specific imbalance.

DISCUSSION

This study has demonstrated that, in the majority of com- parable cases, cytogenetic and molecular genetic results are concordant. Where differences were observed, these two methods often complement each other. In some cases the DNA results are complemented, especially where complex balanced changes or rearrangements have occurred. This was illustrated by sample 20, where no molecular changes were observed on chromosome 16 but a translocation was found, or sample 25, a mult iclonal and very complex tumor with almost no molecular changes on the four chromosome arms tested. In other cases, the cytogenetic results are complemented, e.g., where site-specific imbal- ances were found. Since only a small proport ion of the cancer cells are karyotyped, the clones found may only partially reflect the tumor. This is a possible explanation for the discrepancy found in some samples, and in some instances it can be explained by the fact that some clones could not be completely analyzed.

Dutril laux et al. have put forward a hypothesis on chromosomal alterations occurring during tumor progression [14] (see Materials and Methods). We used this hypothesis

Figure I Molecular genetic and cytogenetic results for chromosome 17. The molecular results are represented with dots and bars. Filled dot = allelic imbalance, empty dot = heterozygous but no imbalance, bar = homozy- gous, blank = not determined. The boxes represent affected parts of the chromosome (filled = lost and striped = gained), according to cytogenetic results (only presumed unbalanced changes are shown), aResults for the two p53 markers are combined.

13

12

11.2

11.1 11.1

11.2

12

21.1 21.2

22

23

24

25

17

~ THRA1 0

mfdl5 0

mfd188 l

CMM86 l

THH59 "

RMU30

$802 0

sample number 1 2 3 4 5 6 7

144D6

t'NZ22 0 n

PS3" 0 • 0

_

o °_ • 0 0

© • 0

O 0

© -

0 © -

10 11 9 12 13 14 15 16 17 18 19 20

"!!i:" © 0 - -

.o . o 0 • O 0

• 0 0 - - 0 - - 0

- - 0 - - 0

0 0 0 0 - - - -

- - 0 0 0 0 0

• 0 0 0 0 0 - -

21 22 23

O O

24 25

Molecular Genetics and Cytogenetics of Breast Carcinoma 41

to evaluate whether the changes observed were ba lanced or not. Where this was done (e.g., samples 5, 21, and 23) the molecular genetic and cytogenetic results match and thus seem to suppor t Dutr i l laux 's hypothesis .

Assuming that results wi th the molecular genetic and cytogenetic methods are as s imilar as possible, keeping in mind their inherent diffe::ences, we can suggest a "missing l ink" in samples wi th complex changes. Sample 11, which came from a c l in ical ly advanced case, is an example of this. It is a d ip lo id tumor which has, according to karyotyping, lost one whole :h romosome 17, and imbalance was observed with all markers except those that map close to the p53 (17p13.1) and BRCA-1 (17q21) tumor suppressor genes (Fig. 1), sites where a high frequency of allelic imbalance has been repor ted in breast tumors [2, 15]. Nei- ther p53 muta t ion nor abnormal p53-protein expression were found in this sample (unpubl ished results), which might indicate that this represents an advanced tumor that has lost both copies of the gene [16] and thus no dele t ion at the p53 site can be detected wi th molecular genetic methods. The same might apply for the BRCA-1 locus. This hypothet ica l sequence of events is depic ted in Figure 2. If this was the case, a]lelic imbalance, found wi th molecular genetic methods in this tumor and other advanced tumors, wou ld not necessar i ly indicate sites of impor tance for tumor progression.

Technical difficulties in est imating allel ic imbalance might poss ibly expla in why we d id not always observe a perfect fit for our results. This is, however, probably not the main cause for the d iscrepancy between molecular genetic and cytogenetic results. It is more l ikely that we missed the most advanced samples, where both copies of an impor tant locus have been lost, as descr ibed above. The first copy is lost in a locus-specific, non- random loss as suggested in Knudson ' s tumor suppressor theory [17]. When the tumor progresses further, a second non-random, site-specific loss occurs or, perhaps more frequently, random loss of a whole chromosome removes the remaining locus. Since losing both copies of an impor tant tumor suppressor gene causes still :Further progression of the tumor, this random loss is re ta ined non-randomly.

This impl ies that nei[her molecular genetics nor cytogenetics alone should be trusted to provide a complete picture. Information on allelic imbalance should be inter- pre ted with some caut ion and, perhaps more important ly , one should not draw strong conclusions from data on allelic imbalance regarding ind iv idua l tumors. Molecular cytogenetics, such as locus-specific FISH, can solve some of the problems encountered with molecular and cytogenetical methods.

Two or more unre la ted clones were found in five of the 25 tumors analyzed (excluding cases showing polyp- lo idizat ion of clones). For each tumor, the unrela ted clones gave the same cytogenetic results for chromosomes 7, 16, and 17. Thus comparabi l i ty of the tumor heterogeneity could not be tested. Neither could the accuracy of breakpoint locat ion be tested as there was not enough breakpoint information on the three chromosomes under investigation.

progression of the tumor

i ii

) v

mm mm

Ill

m m mm m e~e~

Figure 2 Hypothetical progression of tumor sample ]1. a) cytogenetical events, b) autoradiograph showing allelic imbalance results which could apply either to the p53 or the BRCA-1 loci, c) autoradiograph of other chromosome 17 loci. In the first stage (I), nothing has happened cytogenetically and the autoradiographs show no change either. In the next stage (II), the p53 and BRCA-1 loci are lost, but the loss is too small to be seen in G-banded cells. Allelic imbalance is seen on the p53 and BRCA-1 loci but not on any other loci. A weak striped band is still seen on blI, because tumor cells are only about 60% of the tumor sample. In the last stage, III, a whole chromosome is lost, as can be seen in the G- banded cells. The loss of the p53 and BRCA-1 sites on the other chromosome is now balanced and, thus, cannot be seen, but the other remaining loci appear lost. Abbreviations: T = DNA from the tumor, N = DNA from normal tissue.

It is an interest ing quest ion whether clones found among DHPs or CCPs show better agreement wi th the molecular results. There have been different views on which method should be p re fe r red- -d i rec t harvesting or short term cul tur ing- -genera ted by the fact that the types of clonal chromosomal abnormali t ies detected differ according to the method used [1, 18, 19]. The cells used for molecular analysis represent the tumor as it was at the t ime of surgery, i.e. the cells are processed immedia te ly and DNA extracted. The same applies to the DHPs, but when the tumor is cultured, the tumor cells are s t imulated to enter mitosis and more cells can be analyzed cytogenet ical ly than in the direct method.

The present s tudy revealed significantly greater simi- lari ty between karyotypes found among DHPs and molecular genetic results than between molecular genetic results and the CCPs. This does not, however, mean that the clones found among the DHPs are the only impor tant clones, since noncycl ing neoplast ic cells can not be analyzed from direct harvest. Analyz ing in terphase cells, us-

42 R. Valgardsdottir et al.

ing chromosome painting, may circumvent this problem, as almost all cells can be screened using this technique [8].

We conclude that the molecular genetic results are in agreement with the cytogenetic findings where enough information was obtained. The two methods complement each other, and important steps in development of the tumors can be deduced by combining the data.

We thank the staff of the Department of Pathology and Oncology, University Hospital of Iceland, for supplying the samples. We also wish to thank the staff of the Cytogenetics Laboratory for technical assistance, the staff of the Molecular and Cell Biology Research Laboratory for sample preparations and technical assistance, and the Agricultural Research Institute for molecular cytogenetics facilities. This work was supported by grants from The National Hospital Science Fund, The Icelandic Science Fund, NORFA, the memorial fund of Bergthora Magm~sd6ttir and Jakob B. Bjarnason, and the Icelandic Research Fund for Higher Education.

REFERENCES

1. Steinarsdottir M, Petursdottir I, Snorradottir S, Ey~6rd JE, Ogmundsdottir HM (1995): Cytogenetic studies of breast carcinomas: different karyotypic profiles detected by direct harvesting and short-term culture. Genes Chromosom Cancer 13:239-248.

2. Devilee P, Cornelisse CJ (1994): Somatic genetic changes in human breast cancer. Biochim Biophys Acta 1198:113-130.

3. Pandis N, Jin Y, Gorunova L, Petersen C, Bardi G, Idvall I, Johansson B, Ingvar C, Mandahl N, Mitelman F, Heim S (1995): Chromosome analysis of 97 primary breast carcinomas: Identification of eight karyotypic subgroups. Genes Chromosom Cancer 12:173-185.

4. Heisterkamp N, Groffen J (1991): Molecular insights into the Philadelphia translocation. Hematol Pathol 5:1-10.

5. Heim S (1992): Is cancer cytogenetics reducible to the molecular genetics of cancer cells? Genes Chromosom Cancer 5:188-196.

6. Gray JW, Pinkel D (1992): Molecular cytogenetics in human cancer diagnosis. Cancer 69:1539-1542.

7. Buckle V], Kearney L (1994): New methods in cytogenetics. Current Opn Genet Dev 4:374-382.

8. Anamthawat-lonsson K, Eyi]6rd JE, (3gmundsdottir HM, Petursdottir I, Steinarsdottir M (1996): Instability of chromosomes 1, 3, 16 and 17 in primary breast carcinomas inferred by fluorescent in situ hybridization. Cancer Genet Cytogenet 88:1-7.

9. Sandberg AA, Berger CS (1994): Review of chromosome studies in urological tumors. II. Cytogenetics and molecular genetics of bladder cancer. J Urol 151:545-560.

10. Saint-Ruf C, Gerbault-Seureau M, Viegas-P6quignot E, Zaf- rani B, Cassingena R, Dutrillaux B (1990): Proto-oncogene amplification and homogeneously staining regions in human breast carcinomas. Genes Chromosom Cancer 2:18-26.

11. Saint-Ruf C, Gerbault-Seureau M, Viegas-P6quignot E, Zaf- rani B, Malfoy B, Dutrillaux B (1991): Recurrent homogeneously staining regions in 8pl in breast cancer and lack of amplification of POLB, LHRH, and PLAT genes. Cancer Genet Cytogenet 52:27-35.

12. Pandis N, Helm S, Bardi G, Limon J, Mandahl N, Mitelman F (1992): Improved technique for short-term culture and cytogenetic analysis of human breast cancer. Genes Chromosom Cancer 5:14-20.

13. ISCN (1991): Guidelines for Cancer Cytogenetics, Supple- ment to an International System for Human Cytogenetic Nomenclature. Mitelman F (ed). Basel: S. Karger.

14. Dutrillaux B, Gerbault-Seureau M, Remvikos Y, Zafrani B, Prieur M (1991): Breast cancer genetic evolution: I. Data from cytogenetics and DNA content. Breast Cancer Res Treat 19:245-255.

15. Thorlacius S, lonasdottir O, Eyfj6rd JE (1991): Loss of het- erozygosity at selective sites on chromosomes 13 and 17 in human breast carcinoma. Anticancer Res 11:1501-1508.

16. Fisher CJ, Gillett CE, Vojtesek B, Barnes DM, Millis RR (1994): Problem with p53 immunohistochemica] staining: the effect of fixation and variation in the methods of evalua- tion. Br J Cancer 69:26-31.

17. Knudson AG] (1971): Mutation and cancer: statistical study of retinoblastoma. Proc Natl Acad Sci USA 68:820-823.

18. Pandis N, Bardi G, Heim S (1994): Interrelationship between methodological choices and conceptual models in solid tumor cytogenetics. Cancer Genet Cytogenet 76:77-84.

19. Thompson F, Emerson J, Dalton W, Yang J-M, McGee D, Vil- lar H, Knox S, Massey K, Weinstein R, Bhattacharyya A, Trent J (1993): Clonal chromosome abnormalities in human breast carcinomas I. Twenty-eight cases with primary dis- ease. Genes Chromosom Cancer 7:185-193.

Documents

Molecular genetics and cytogenetics of breast carcinomas: Comparison of the two methods