Int. J. Cancer: 72, 599–603 (1997) r 1997 Wiley-Liss, Inc. Publication of the International Union Against Cancer Publication de l’Union Internationale Contre le Cancer LOSS OF HETEROZYGOSITY AT THE TP53 GENE: INDEPENDENT OCCURRENCE FROM GENETIC INSTABILITY EVENTS IN NODE-NEGATIVE BREAST CANCER Sarab LIZARD-NACOL1*, Jean-Marc RIEDINGER2, Gérard LIZARD3, Anne-Lise GLASSER4, Nolwenn COUDRAY5, Gilles CHAPLAIN6 and Jacques GUERRIN7 1Laboratory of Molecular Genetics, Centre G.F. Leclerc, Dijon, France 2Laboratory of Medical Biology, Centre G.F. Leclerc, Dijon, France 3INSERM CJF 93-10, Hospital of Bocage, Dijon, France 4INSERM U-71, Clermont-Ferrand, France 5Institute of Cellular and Molecular Biology, Strasbourg, France 6Côte d’Or Cancer Registry, BP 728, Dijon, France 7Oncological Service, Centre G.F. Leclerc, Dijon, France TP53 abnormalities have been reported as an early event in the process of cellular transformation of human breast cancers, and involved in mammary-tumor evolution, from in situ to invasive disease. In this study, node-negative (N2) tumors were examined for TP53 allelic loss in relation to different genetic instability events, including allelic loss at chromosome 17p13.3 and c-H-ras-1 loci, as well as alteration of the c-myc and c-erbB-2/neu oncogenes. TP53 allelic loss was analyzed to determine whether such an abnormality was the more important, among other genetic events, in the N2 tumors, whether it appeared independently of these genetic events, and whether accumulation of genetic events arises in this group of breast tumors. Clinicopathological parameters were also examined. Loss of heterozygosity (LOH) at the TP53 gene appears the most frequent alteration detected (26% vs. 13%, 8%, 9% and 3% for LOH at D17S30 and c-H-ras-1 loci, and amplification of c-myc and c-erbB-2/neu respectively). There was no association between LOH at the TP53 locus and other genetic events. Among clinicopathological parameters, significant associations were observed only with estrogen-receptor-negative tumors (p 5 0.05). Our results demonstrate that LOH at TP53 arises more frequently in the N2 breast cancer, thus supporting earlier findings suggesting that TP53 abnormality has a role early in the pathogenesis of breast lesions. Moreover, the data indicate that accumulation of many genetic events occurs at a low level in N2 breast tumors, and that TP53 abnormality occurs independently of these genetic events. Int. J. Cancer 72:599–603, 1997. r 1997 Wiley-Liss, Inc. In breast cancer, a model taking into account accumulation of relevant mutations and their order of occurrence is still not clearly defined. However, certain mutations appear to be more important than others in the process of breast malignant transformation. Loss of heterozygosity (LOH) at the TP53 gene (Radford et al., 1993), as well as abnormal TP53 protein expression (Poller et al., 1993), were found in a pre-invasive form of breast cancer, the ductal carcinoma in situ, involving TP53 in mammary-tumor evolution from in situ to invasive disease. The invasive-breast-cancer group includes 2 sub-groups of patients: those with lymph-node-negative and those with lymph-node-positive cancers (N2 and N1 respectively). There are fundamental differences in the biological potential between these populations of tumor cells. Pre-invasive breast cancer is rarely fatal, N2 invasive breast cancer results in about 20% mortality, whereas mortality from cancer in the N1 sub-group is much higher (Mackey et al., 1990). It is widely held, but not proven, that the non-invasive phase precedes tumor invasion, that the N2 invasive phase appears before the N1 form, and that one leads to the other (Mackey et al., 1990). Inactivation of the TP53 tumor-suppressor gene is an important step in the development of the majority of human cancers, including those of the breast. Functional inactivation occurs most commonly as a result of mis-sense mutation, but can also result from nonsense or deletion mutations (Hollstein et al., 1991). A number of solid tumors contain a wild-type allelic deletion with mutations in the remaining TP53 allele. In breast cancer, about 30 to 50% of the cases carry a mutant TP53 gene and/or an LOH, suggesting that loss of the wild-type allele may be a necessary step in the oncogenic activation of TP53 in this cancer (Deng et al., 1994). Associations between TP53 abnormalities and genetic instability, measured as allelic loss on chromosome 17 and amplification of the c-erbB-2/neu oncogene, have been recently reported in breast cancer (Eyfjörd et al., 1995). Interestingly, there was no association between TP53 abnormalities and amplification of the c-erbB-2/neu oncogene in the in situ lesions, whereas such an association was found in invasive cancers (Poller et al., 1993). Although TP53 abnormalities play a role early in human mammary tumor formation, one may expect that their frequencies and their associations with other genetic changes are relatively similar, in the N2 sub-group, to those found in the in situ lesions. An association between TP53 abnormalities and genetic instability was often observed in the N1 sub-group of patients. Such an association has not been reported in the N2 sub-group, so that a study including the simultaneous analysis of these changes on the same tumor specimens is of interest. In the present study, TP53 LOH alterations were analyzed in a group of 161 patients with N2 breast cancer. Clinicopathological parameters (hormone-receptor status, pathological grade, tumor differentiation and cellular density), and biological parameters (cathepsin-D, cath-D and pS2 protein) were known for a majority of the patients. TP53 LOH frequencies in this group of tumors were compared with several genetic instability factors that were more frequently involved in breast cancer. These abnormalities included LOH at chromosome 17p13.3 and c-H-ras-1 loci, as well as amplification of the c-myc and c-erbB-2/neu oncogenes. PATIENTS AND METHODS Patients Tumor specimens from 161 patients with surgically confirmed node-negative breast cancer were included in this study. The patients were all women, ranging in age from 30 to 86 years (median, 61 years). Tumor samples were drawn from a pool of frozen specimens (liquid nitrogen) originally submitted for steroidreceptor analysis. Clinical information related to each patient was available. Only 16 patients received adjuvant treatment by chemotherapy and/or tamoxifen therapy. The median clinical follow-up was 20 months (range, 10 to 36 months). During this period, 9 Contract grant sponsors: Ligue Bourguignone and the Comité Départemental de Saone et Loire de la Ligue Contre le Cancer. *Correspondence to: Laboratory of Molecular Genetics, Centre GeorgesFrançois Leclerc, 1, rue du Professeur Marion, 21O34 Dijon Cedex, France. Fax: (33) 3 80 73 77 26. Received 21 January 1997; revised 18 April 1997 LIZARD-NACOL ET AL. 600 patients developed tumor recurrence, 8 developed metastatic lesions, and 7 died of breast cancer. protein) was based on the results reported by Foekens et al. (1990), as limited concentration of hormone-sensitivity determination. DNA analysis Cellular DNA was extracted from tissues and from blood leukocytes by a standard proteinase-K and phenol/chloroform method, and Southern blotting of restriction-enzyme-digested DNA was performed (Southern, 1975). Briefly, 10 µg of DNA were digested with appropriate restriction endonucleases (Table I), size fractionated on a 0.8% agarose gel, transferred to a nylon membrane (Hybond N1, Amersham, Aylesbury, UK), and hybridized overnight at 65°C with randomly primed 32P-labeled probes. The probes used for the analyses presented in Table I were obtained from the ATCC (Rockville, MD). Autoradiography was performed on membrane filters for 2 days at 270°C using Kodak XAR65 films. Statistical analysis The association between the different parameters studied was examined by a standard univariate x2 test. A p value of 0.05 or less was considered statistically significant. Determination of allele loss Paired normal and tumor DNA from each patient was analyzed using probe-enzyme combinations which identify restrictionfragment-length polymorphism in a large proportion of individuals. Normal DNA samples that were polymorphic at a given locus were considered to be ‘‘informative,’’ whereas the homozygotes were declared ‘‘uninformative’’. The signal intensity of fragments was determined by visual examination and confirmed by densitometry. Because the tumor sample used contained a certain number of non-neoplastic cells, Southern blotting techniques may have limited resolution for LOH in comparison with PCR-based techniques. Allelic loss was therefore considered if the intensity in the tumor was less than 50% of that in its corresponding normal tissue. Determination of gene amplification Restriction-enzyme-digested tumor DNA was compared with matching lymphocyte DNA in the same agarose gels. Blots of these gels were first hybridized with the oncogene probes. Rehybridation of the same blots with the human a-actin probe provided a control for the amount of DNA transferred to the nylon membranes. The oncogene and control-gene autoradiographs were measured by visualization examination and/or densitometry of Southern blots. RESULTS LOH analysis Because the amounts of normal and tumoral tissues available were not sufficient, not all of the samples were tested for LOH at all the different chromosomal loci. Thus, after digestion with appropriate restriction endonucleases, informative cases identified in a total number of patients were: 30/117 for TP53 at 17p13.1, 72/86 for D17S30 at 17p13.3, and 72/126 for c-H-ras-1 at 11p15 (Fig. 1). Among these informative cases, LOH was observed in 8 cases (26%) for TP53, 10 cases (13%) for D17S30, and 6 cases (8%) for c-H-ras-1 (Fig. 2). Oncogene alterations Amplification of the 12-kb c-myc fragment was detected in 4 tumors, and a rearrangement of this gene was found in another tumor by the presence of an additional 4.5-kb-EcoRI restriction fragment. This band was absent in DNA from the patient’s lymphocytes, indicating that this alteration was specific for the tumor tissue (Fig. 3). Alterations of the c-myc gene were observed in a total of 3% of cases (5/161), and amplification of the c-erbB-2/neu gene in 9% of cases (15/161) (Fig. 2). None of the tumors examined showed c-erbB-2/neu gene rearrangement. Analysis of transcription levels of the c-erbB-2/neu gene was performed in 3 of the 15 amplified tumors. High levels of a 4.8-kb transcript RNA analysis Total RNA was isolated from the frozen tumor tissue as described by Chomczynski and Sacchi (1987), run on 1.2% agarose gels with formaldehyde, blotted and hybridized as described for DNA analysis. Hormone-receptor assay The estrogen(ER)- and progesterone(PgR)-receptor contents were determined in cytosolic tumors using ER-EIA and PgR-EIA methods respectively (Abbott, North Chicago, IL). The cutoff level used for ER and PgR was 20 fmol/mg of cytosolic proteins. Assay of cath-D and pS2 protein Cath-D and pS2 levels were determined in breast-tumor cytosols using immunoradiometric assay (Elsa-Cath-D and Elsa pS2 respectively), according to the instructions of the manufacturer (CIS Bio, Gif sur Yvette, France). The Cath-D assay measured the total amount of Cath-D (52-, 48- and 34-kDa species). The cutoff level used was 40 pmol/mg of cytosolic protein (Kute et al., 1992). The choice of a cutoff value for the pS2 protein (1 ng/mg of total TABLE I – DNA PROBES USED Locus Probe Chromosomal location Restriction enzyme TP53 D17S30 c-H-ras-1 c-myc c-erbB-2/neu php53B pYNZ22 J77 pTyc 7.4 MAC117 17p13.1 17p13.3 11pter-p15.5 8q24 17q11.2-q12 ScaI BamHI BamHI EcoRI EcoRI FIGURE 1 – Southern-blot analysis of paired tumor (T) and normal lymphocyte (L) DNA, demonstrating loss of heterozygosity (LOH) in N2 breast tumor. Loci detected were: c-H-ras-1, D17S3O (multiple alleles detected polymorphism with bands between 0.5 kb and 1.3 kb), and TP53. For more details of probes and restriction enzymes used, see Table II. LOH AT TP53 LOCUS IN NODE-NEGATIVE BREAST CANCER were detected in the tumor cells in comparison with those obtained from normal tissue (Fig. 4). No co-amplification of the c-myc and c-erbB-2 genes was found in the same tumor. Thus, among all these genetic alterations tested, LOH at the TP53 locus was the most frequently alteration found in the N2 breast tumors analyzed (Fig. 2). LOH at the TP53 gene and genetic alterations There was no association between LOH at the TP53 and other loci for which LOH was detected, even between LOH at the TP53 601 and D17S30 loci, both localized on the short arm of chromosome 17 (17p13.1 and 17p13.3 respectively). On the contrary, an inverse relationship was found between these 2 loci ( p 5 0.0195). An inverse relationship was also found between LOH at TP53 and amplification of c-erbB-2/neu ( p 5 0.0132). No association was observed between LOH at the TP53 locus and oncogene amplifications (Table II). LOH at the TP53 gene and clinicopathological characteristics of the tumors tested LOH at the TP53 gene was associated only with estrogenreceptor-negative tumors ( p 5 0.05). Recent biological parameters, such as cath-D and pS2 protein, known to be involved in the N2 breast cancer, have also been analyzed in this study. Cath-D values ranged between 13 and 114 pmol/mg of protein. The levels of pS2 varied from 0 to 80 ng/mg of protein. There were no statistically significant associations between LOH at the TP53 locus and the levels of cath-D or of pS2 (Table III). Relationship between other genetic events and clinicopathological characteristics of the patients tested LOH at the D17S30 locus was significantly associated with poorly differentiated tumors ( p 5 0.05), and LOH at the c-H-ras -1 gene was strongly associated with elevated pathological grade ( p 5 0.01) (Table III). c-myc-oncogene alterations were significantly associated with poorly differentiated tumors ( p 5 0.025), as well as with elevated tumor-cell density ( p 5 0.025), and with progesterone-receptor-negative tumors ( p 5 0.05). A significant correlation was also found for c-erbB-2/neu-oncogene amplification with poorly differentiated tumors ( p 5 0.01) and with elevated pathological grade ( p 5 0.05). No significant correlations of any of FIGURE 2 – Frequencies of genetic alterations detected in the 161 N2 breast tumors analyzed: TP53 (26%), D17S30 (13%), c-H-ras-1 (8%), c-myc (3%) and c-erbB-2/neu (9%). FIGURE 4 – DNA: representative Southern-blot analysis of paired tumor (T) and normal lymphocytes (L) corresponding to sample 2 showing amplification of the c-erbB-2/neu gene in N2 breast tumor (lane 2). RNA: representative Northern-blot analysis of c-erbB-2/neu and of b-actin corresponding to the same sample, showing a high level of c-erbB-2/neu transcript in comparison with normal tissue (N). TABLE II – RELATIONSHIP BETWEEN ABSENCE OF LOH AT TP53 AND GENETIC ALTERATIONS IN THE N2 BREAST TUMORS ANALYZED FIGURE 3 – Southern-blot analysis of c-myc gene in N2 breast tumor. DNA from tumors (T) or peripheral-blood cells (L) were digested with EcoRI. Additional restriction fragments are found in tumor tissue from patient 2. Hybridization with b-actin probe provides a control for the amount of DNA transferred to the nylon membranes. The size of each band is indicated on the left. TP53 N D17S30 c-H-ras-1 c-myc c-erbB-2/neu With LOH Without LOH p 8 22 0 (0) 10 (45) 0.0195 0 (0) 6 (27) 0.0986 0 (0) 5 (23) 0.1396 1 (12) 14 (64) 0.0132 N, total number of cases; number in parentheses, percent of cases with genetic alteration. LIZARD-NACOL ET AL. 602 TABLE III – RELATIONSHIP BETWEEN GENETIC ALTERATIONS AND CLINICOPATHOLOGICAL PARAMETERS IN THE N2 BREAST TUMORS ANALYZED Parameters Tumor differentiation Well 1 moderately Poorly 1 undifferentiated p Tumor grade I 1 II III p Tumor-cell density 1 2 3 p Hormonal receptors ER2 ER1 p PR2 PR1 p Cath-D #40 pmol/mg protein .40 pmol/mg protein p pS2 #1 ng/mg protein .1 ng/mg protein p TP53 n 5 30 D17S30 n 5 72 c-H-ras-1 n 5 72 c-myc 84 66 5/20 (25) 3/10 (30) 0.7703 3/40 (7) 7/32 (21) 0.05 4/46 (8) 2/26 (7) 0.8824 0 (0) 5 (7) 0.01 4 (4) 11 (16) 0.01 123 22 6/23 (26) 2/7 (28) 0.4254 7/48 (14) 3/24 (12) 0.8096 0/50 (0) 6/22 (27) 0.01 3 (2) 2 (9) 0.1153 10 (8) 5 (22) 0.03 24 103 34 1/4 (25) 5/19 (25) 2/7 (28) 0.9901 1/7 (14) 6/43 (13) 3/22 (13) 0.9989 1/11 (9) 3/40 (7) 2/21 (9) 0.9591 0 (0) 1 (1) 4 (11) 0.025 2 (8) 9 (8) 4 (11) 0.8566 41 120 4/7 (57) 4/23 (17) 0.05 3/13 (23) 5/17 (29) 0.6477 3/22 (13) 7/50 (14) 0.9672 5/33 (15) 5/39 (12) 0.7757 2/27 (7) 4/45 (8) 0.8257 3/32 (9) 3/40 (7) 0.7748 2 (4) 3 (2) 0.4486 4 (5) 1 (1) 0.05 4 (12) 10 (8) 0.7801 6 (8) 7 (7) 0.9511 91 57 2 (2) 3 (5) 0.3152 9 (9) 1 (1) 0.0550 4 (4) 1 (1) 0.5868 3 (3) 1 (1) 0.5734 5 (5) 3 (5) 0.9517 48 94 0 (0) 5 (5) 0.1038 5 (10) 3 (3) 0.0773 2 (4) 3 (3) 0.7655 1 (2) 3 (3) 0.7058 4 (8) 2 (2) 0.0821 N 73 88 c-erbB-2/neu N, total number of cases; n, number of informative cases for the locus analyzed; number in parentheses, percent of cases with genetic alteration. these genetic alterations were observed with the levels of cath-D or pS2. DISCUSSION In breast cancer, the number of genetic events needed for tumor initiation, progression and metastasis, depends on several parameters and remains unclear. It is unknown which, if any, of the genetic alterations underly this change, or whether their combinations are essential to the malignant process. TP53 abnormalities have been reported as an early event in the process of cellular transformation in certain malignancies (Hasegawa et al., 1995) including human breast cancers (Radford et al., 1993). In a pre-invasive form of breast cancer, the ductal carcinoma in situ, these abnormalities appear to occur independently of other genetic instability events (Poller et al., 1993). In this study, we evaluated TP53 status in the N2 breast tumors, examining frequencies of TP53 LOH alterations and their relationships to other genetic events. These events include LOH at chromosome 17p13.3 and c-H-ras-1 loci as well as alteration of the c-myc and c-erbB-2/neu oncogenes. The c-H-ras-1 gene, located at 11p15, is frequently (30%) involved in deletion or LOH in breast cancer (Champène et al., 1995); however, this alteration has never been determined in N2 patients. The c-myc gene, which encodes a nuclear-DNA-binding protein, has been reported to be amplified in 15 to 31% of N1 breast carcinomas (Escot et al., 1986; Berns et al., 1992), and rearrangement of c-myc was observed in 5% of cases (Escot et al., 1986). However, few data are available for c-myc alterations in N2 patients. Because the c-ras and c-myc oncogenes were both involved in cellular transformation and/or apoptosis pathway in cooperation with TP53 (Finlay et al., 1989), we analyzed these 2 genes in the N2 breast tumors. The c-erbB-2/neu gene, which encodes a transmembrane protein with extensive homology with the epidermal-growth-factor receptor (Dobashi et al., 1991), has been found to be amplified and over-expressed in 10- to 50% of N1 breast cancers. In N2 breast cancer, the prevalence is somewhat lower (10 to 20%), but the prognostic value of c-erbB-2/neu is now well established in this group of patients (Press et al., 1993). In the N2 tumors analyzed in this study, LOH at the c-H-ras-1 gene was observed in only 8% of informative cases. This small proportion of LOH at the c-H-ras-1 gene suggests that LOH at this locus may not be a causal factor in breast carcinogenesis but rather a marker for tumor aggressiveness (Champène et al., 1995). We found an alteration of the c-myc gene in 3% of cases, and an amplification of the c-erbB-2/neu gene in 9% of cases. The prevalence of these 2 oncogene alterations in our study is in agreement with earlier data concerning the N2 patient group (Berns et al., 1992). In our group of patients, it is still too soon after surgery to investigate the prognostic significance of the genetic alterations tested. Nevertheless, these alterations are known to be associated with other cellular parameters reflecting poor prognosis. We have therefore made comparisons with a variety of factors: clinical (steroid-receptor status), morphological (histoprognostic grade and tumor differentiation) and biological (Cath-D and pS2). Cath-D is a protease which facilates cancer-cell migration and invasion, and pS2 is an estrogen-regulated protein found indicative for hormonal sensitivity. Both of these 2 proteins have been reported in N2 breast tumors (Kute et al., 1992; Foekens et al., 1990). Our results (LOH at the D17S30 and c-H-ras-1 loci and amplification of c-myc and c-erbB-2/neu) confirm earlier observations (Berns et al., 1992; Press et al., 1993; Coles et al., 1992), and show significant associations between the genetic events tested and some clinicopathological factors (poorly differentiated tumors, elevated pathological grade, elevated tumor-cell density and progesterone-receptornegative tumors). LOH AT TP53 LOCUS IN NODE-NEGATIVE BREAST CANCER Among the different genetic alterations analyzed, LOH at the TP53 gene is the most frequent alteration detected (26%). LOH at chromosome 17p13.3, near to the TP53 gene localization (17p13.1), was the second most prevalent alteration observed in these tumors (13%). No association was found between LOH at these 2 loci, suggesting that the use of the pYNZ22 probe provides additional information about regions of loss on the short arm of chromosome 17, and agrees with Coles et al. (1990), who reported the existence of (a) tumor-suppressor gene(s) located distally to the TP53 gene. There was no association between LOH at the TP53 locus and other genetic events, or with cath-D and pS2 levels. These findings demonstrate that this alteration arises not only more frequently in the N2 breast cancer, but also independently of other genetic events. We have shown that no LOH at the TP53 gene was found in benign breast samples among the heterozygous patients tested (Lizard-Nacol et al., 1995). However, mutations in the TP53 gene, determined by the PCR-SSCP method, have been identified in 8% of the benign breast tissues (data not shown). These results are in agreement with the findings of Millikan et al. (1995) and Younes et al. (1995), and suggest that point mutations in the TP53 gene occur 603 before loss of the wild-type allele in breast carcinomas (Coles et al., 1992). In view of the multi-step nature of carcinogenesis, an initial genetic change in breast-tumor cells may be an alteration at TP53, beginning with point mutation in the pre-cancerous lesions, followed by LOH in the more developed lesions. Thus, additional genetic changes, such as oncogene amplifications and/or LOH at another chromosomal locus, might contribute to breast-tumor progression. Our study supports earlier findings suggesting that TP53 abnormality has a role early in the pathogenesis of breast lesions. Moreover, the data indicate that accumulation of many other genetic events occurs at a low level in N2 breast tumors, and that TP53 abnormalities occur independently of these genetic events. ACKNOWLEDGEMENTS This work was supported by the Ligue Bourguignonne and the Comité Départemental de Saône et Loire de la Ligue contre le Cancer. We thank Mrs. M. Arnal and Mrs. L. Hahnel for technical assistance. REFERENCES BERNS, E.M., KLIJN, J.G., VAN PATTEN, W.L., STAVEREN, P.L., PORTENGEN, H. and FOEKENS, J.A., c-myc amplification is a better prognostic factor than HER2/neu amplification in primary breast cancer. Cancer Res., 52, 1107–1113 (1992). CHAMPÈNE, H., BIÈCHE, I., LIZARD, S. and LIDEREAU, R., 11q13 amplification in local recurrence of human primary cancer. Genes Chromosomes Cancer, 12, 128–133 (1995). CHOMCZYNSKI, P. and SACCHI, N., Single-step method of RNA isolation by acid-guanidinium-thiocyanate-phenol-chloroform extraction. Ann. Biochem., 162, 156–159 (1987). COLES, C., CONDIE, A., CHETTY, U., STEEL, C.M., EVANS, H.J. and PROSSER, J., p53 mutations in breast cancer. Cancer Res., 52, 5291–5298 (1992). COLES, C., THOMPSON, A.M., ELDER, P.A., COHEN, B.B., MACKENZIE, I.M., CRANSTON, G., CHETTY, U., MACKAY, J., MACDONALD, M., NAKAMURA, Y., HOYHEIM, B. and STEEL, C.M., Evidence implicating at least two genes on chromosome 17p in breast carcinogenesis. Lancet, 336, 761–763 (1990). DENG, G., CHEN, L.C., SCHOTT, D.R., THOR, A., BHARGAVA, V., LJUNG, B.M., CHEW, K. and SMITH, H.S., Loss of heterozygosity and p53 gene mutations in breast cancer. Cancer Res., 54, 499–505 (1994). DOBASHI, M.D., DAVIS, J.G., MIKAMI, Y., FREEMAN, J.K., HAMURO, J. and GREENE, M.P., Characterisation of a neu/c-erbB-2-protein-specific activating factor. Proc. nat. Acad. Sci. (Wash.), 88, 8582–8586 (1991). ESCOT, C., THEILLET, C., LIDEREAU, R., SPYRATOS, F., CHAMPÈNE, M.H., GEST, J. and CALLAHAN, R., Genetic alteration of the c-myc proto-oncogene (MYC) in human primary breast carcinomas. Proc. nat. Acad. Sci. (Wash.), 83, 4834–4838 (1986). EYFJÖRD, J.E., THORLACIUS, S., STEINARSDOTTIR, M., VALGARDSDOTTIR, R., ÖGMUNDSDOTTIR, H.M. and JONSSON, K., p53 abnormalities and genomic instability in primary human breast carcinomas. Cancer Res., 55, 646–651 (1995). FINLAY, C.A., HINDS, P.W. and LEVINE, A.J., The p53 proto-oncogene can act as a suppressor of transformation. Cell, 57, 1083–1093 (1989). FOEKENS, J.A., RIO, M.C., SEGUIN, P., VAN PUTTEN, W.L.J., FAUQUE, J., NAP, M., KLIJN, J.G.M. and CHAMBON, P., Prediction of relapse and survival in breast-cancer patients by pS2 protein status. Cancer Res., 50, 3832–3837 (1990). HASEGAWA, H., UEDA, M., FURUKAWA, K., WATANABE, M., TERAMOTO, T., MUKAI, M. and KITAJIMA, M., p53 gene mutations in early colorectal carcinoma. De novo vs. adenoma-carcinoma sequence. Int. J. Cancer, 64, 47–51 (1995). HOLLSTEIN, M., SIDRANSKY, D., VOGELSTEIN, B. and HARRIS, C.C., p53 mutations in human cancers. Science, 253, 49–53 (1991). KUTE, T.E., SHAO, Z.M., SUGG, N.K., LONG, R.T., RUSSELL, G.B. and CASE, L.D., Cathepsin D as a prognostic indicator for node-negative breast-cancer patients using both immuno-assays and enzymatic assays. Cancer Res., 52, 5198–5203 (1992). LIZARD-NACOL, S., LIDEREAU, R., COLLIN, F., ARNAL, M., HAHNEL, L., ROIGNOT, P., CUISENIER, J. and GUERRIN, J., Benign breast disease: absence of genetic alterations at several loci implicated in breast-cancer malignancy. Cancer Res., 55, 4416–4419 (1995). MACKEY, J., THOMPSON, A.M. and COLES, C., Molecular lesions in breast cancer. Int. J. Cancer, Suppl. 5, 47–50 (1990). MILLIKAN, R. and 12 OTHERS, p53 mutations in benign breast tissue. J. clin. Oncol., 13, 2293–2300 (1995). POLLER, D.N., ROBERTS, E.C., BELL, J.A., ELSTON, C.W., BLAMEY, R.W. and ELLIS, I.O., p53 protein expression in mammary ductal carcinoma in situ: relationship to immunohistochemical expression of estrogen receptor and c-erbB-2 protein. Hum. Pathol., 24, 463–468 (1993). PRESS, M.F., PIKE, M.C., CHAZIN, V.R., HUNG, G., UDOVE, J.A., MAEKOWICZ, M., DANYLUK, J., GODOLPHIN, W., SLIWKOWSKI, M., AKITA, R., PATERSON, M.C. and SLAMON, D.J., her-2/neu expression in node-negative breast cancer: direct tissue quantitation by computerized image analysis and association of over-expression with increased risk of recurrent disease. Cancer Res., 53, 4960–4970 (1993). RADFORD, D.M., FAIR, K., THOMPSON, A.M., RITTER, J.H., HOLT, M., STEINBRUECK, T., WALLACE, M., WELLS, S.A. JR. and DONIS-KELLER, H.R., Allelic loss on chromosome 17 in ductal carcinoma in situ of the breast. Cancer Res., 53, 2947–2950 (1993). SOUTHERN, E.M., Detection of a specific sequence among DNA fragments separated by gel electrophoresis. J. mol. Biol., 98, 503–517 (1975). YOUNES, M., LEBOVITZ, R.M., BOMMER, K.E., CAGLE, P.T., MORTON, D., KHAN, S. and LAUCIRICA, R., p53 accumulation in benign breast-biopsy specimens. Hum. Pathol., 26, 155–158 (1995).