- Email: [email protected]
Microsatellite Instability in Endometrial Cancer: Relation to Histological Subtypes
Gynecologic Oncology 73, 247–252 (1999) Article ID gyno.1999.5351, available online at http://www.idealibrary.com on
Microsatellite Instability in Endometrial Cancer: Relation to Histological Subtypes M. G. Tibiletti,* D. Furlan,* M. Taborelli,* C. Facco,* C. Riva,* M. Franchi,† A. Cossu,‡ M. Trubia,§ R. Taramelli, ¶ and C. Capella* *Department of Clinical and Biological Sciences and Service of Pathological Anatomy, University of Insubria and Ospedale Di Circolo, 21100 Varese, Italy; †Obstetric and Gynecological Clinic and ¶Department of Structural and Functional Biology, University of Insubria, Varese, Italy; ‡Department of Pathology, University of Sassari, Italy; and §IRCCS H. S. Raffaele, Milan, Italy Received August 14, 1998
to be neutral or nonselective; some of them, however, may affect critical oncogenic or tumor suppressor loci [2]. Endometrial carcinomas exhibit many of the genetic features also present in RER-positive colorectal carcinomas [7, 13], such as MI and germline mutations in mismatch repair genes. However, the clinical and biological significance of MI in endometrial carcinomas has not yet been clarified. In particular, it is interesting to ascertain whether the association of the RER1 phenotype with a special histotype, as observed for colorectal carcinomas [14], might also exist for cancers of the endometrium. We have studied the prevalence of MI in endometrial cancers and its relationship to clinicopathological characteristics such as tumor grade, stage, histological type, age of onset, follow-up data, and familial pattern. In addition MI was correlated with the immunohistochemical expression of p53, progesterone receptor (PgR), and estrogen receptor (ER) by the tumors.
Fifty-one endometrial cancers were analyzed with regard to whether or how microsatellite instability (MI) was associated with the development of different types of endometrial malignant neoplasms. We investigated 6 loci previously reported as informative for colorectal cancer and a group of 8 loci located on 6q. Replication error (RER1) phenotype was detected in 10 of 51 (19.6%) endometrial cancers (ECs), all but one of which showed endometrioid differentiation. On the contrary, the RER1 phenotype was not detected in serous carcinomas and malignant mixed Mu¨llerian tumors. MI was present in both early and advanced stage ECs. No correlation was found between age, grade, stage, familial pattern, mitotic index, and the RER1 phenotype of ECs. Only 1 of 8 endometrial carcinomas showing MI was associated with mutant p53 expression, while the majority of RER1 tumors were positive for estrogen and progesterone receptors. Our findings suggest that MI plays an early role in endometrial tumorigenesis and is significantly correlated with adenocarcinomas showing endometrioid features (EAs). The frequent involvement of the telomeric region of chromosome 6 in the MI of EA is an indication that this region may be crucial in the process of EA tumorigenesis. © 1999 Academic Press Key Words: endometrial cancer; microsatellite instability; histotype; chromosome 6; endometrioid adenocarcinoma.
INTRODUCTION The instability of microsatellite DNA sequences (MI) was first described in a subset of sporadic colorectal cancers and in tumors associated with hereditary nonpolyposis colorectal cancer (HNPCC) [1–3]. Endometrial carcinoma is the most common noncolorectal carcinoma occurring in women affected by HNPCC [4, 5]. MI has been observed both in the inherited form (HNPCC) and in about 20% of presumably sporadic endometrial carcinomas [6 –11]. Microsatellite alterations represent a marker for a “mutator phenotype” (termed RER1, from replication error), which is characterized by widespread genetic instability and by an increased mutation rate [12]. Most of the microsatellite alterations identified in tumors are likely 247
MATERIAL AND METHODS Histopathological Study Formalin-fixed, paraffin-embedded tissue samples from endometrial cancers were obtained from 50 patients who had undergone primary surgery at the Varese or Sassari Hospital between 1983 and 1997. A total of 51 cases were studied; in 1 case two samples from the same tumor were isolated. Hematoxylin and eosin stained sections from each case were evaluated to assess the depth of myometrial invasion, grade, mitotic index, and presence of necrosis. Histological classification was based on WHO criteria [15] and stage was assessed following recomendations of the International Federation of Gynecology and Obstetrics [16]. The tumors were grouped as adenocarcinomas with endometrioid features (30 cases) and nonendometrioid cancers (21 cases), the latter ones including serous (11 cases), mucinous (1 case), clear cell (3 cases), undifferentiated carcinomas (2 cases), adenosarcoma (1 case), and malignant 0090-8258/99 $30.00 Copyright © 1999 by Academic Press All rights of reproduction in any form reserved.
248
TIBILETTI ET AL.
Mu¨llerian mixed tumors (3 cases). Clinical information and follow-up data for each case were obtained from clinicians, hospital charts, clinical reports or death certificates. Immunohistochemical Study Immunohistochemical staining was performed on 3-mm newly cut routinely processed sections using the avidin– biotin peroxidase complex procedure according to Hsu et al. [17]. Antigen unmasking using a citrate buffer, pH 6.0, and a microwave oven for 10 min at 650 W was performed prior to incubation with the primary antibodies. Antibodies employed were directed against p53 (monoclonal, clone D07, 1:1000, DAKO, Copenaghen, Denmark), estrogen receptor (monoclonal, clone 1D5, 1:1000, DAKO), and progesterone receptor (monoclonal, clone, 1A6, 1:20, Novocastra, Newcastle, UK). Immunohistochemical results were semiquantitatively evaluated as a percentage of positive nuclei. For p53 expression a positivity cutoff of 10% immunoreactive cells was established. Specimen Collection and DNA Extraction Fifty-one samples from 50 endometrial tumors and nonadjacent, histologically normal specimens (ovary or nonmetastatic lymph nodes) or peripheral blood lymphocytes were collected and frozen at 280°C for MI analysis. Genomic DNA was isolated from frozen tissues or from 10-mm sections of formalin-fixed paraffin-embedded tissue blocks using QIAamp tissue kits (Qiagen GmbH, Germany) or from whole blood using QIAamp blood kits (Qiagen GmbH). In order to exclude contaminations, both frozen and paraffin-embedded samples were examined histologically prior to their use for DNA extraction and extensive areas of tumor necrosis and normal epithelial or stromal components were eliminated using a surgical blade. The scraped tissues were transferred to a microfuge tube, and the identity of the tissue removed was verified by microscopic analysis of the remaining tissue on the glass slide. MI Analysis DNA from paired normal and tumor tissues were PCRamplified at 14 markers: D6S87 (6q21–23), D6S255 (6q24 – 25.1), D6S193 (6q27), D6S297 (6q27), D6S149 (6q27), D6S281 (6q27), D6S446 (6q27), TBP (6q27), D2S123 (2q16), D31611 (3p24.2p22), D17S787, D2S119 (2p16), D7S481 (7pter–p15), and D13S175 (13q11). The PCR mixture (50 ml) included 100 –200 ng of genomic DNA, 10 mM Tris, 50 mM KCl, 1–2 mM MgCl 2, 200 mM dNTPs, and 50 pmol of each primer, 2 U Taq polymerase (Perkin–Elmer, Italy). Samples were denatured at 94°C for 5 min followed by 30 cycles of denaturation (94°C, 1 min) annealing (55°C, 1 min) extension (72°C, 1 min) and a final extension of 5 min at 72°C in a Perkin–Elmer Gene Amp Thermal Cycler 480. PCR products were electrophoresed on 6% denaturing polyacrylamide se-
FIG. 1. Genetic instability in endometrial cancer patients. PCR products from tumor (T) and normal (N) DNAs were loaded in parallel; arrows indicate additional bands in tumor samples compared to normal samples.
quencing gels for 2 h at 1600 V, 40 mA, 50°C. Polyacrylamide gels fixed on glass plates with methacryloxypropyltrimethosysilane were silver stained using a DNA silver staining system (Promega, Italy). Silver-stained gels were reproduced on EDF films (Kodak). The RER-positive phenotype was assigned when MI was present in at least 3 of the 14 loci tested (Fig. 1). The clinicopathological parameters and immunohistochemical results were correlated considering MI of at least 3 loci (RER1 cases). Pedigree Analysis Psychologically assisted genetic counseling was carried out at the first clinical follow-up of each patient. At least three generations were analyzed, and a careful validation of cancer cases recorded in clinical reports or on death certificates was conducted. Statistical Analysis The relationship between variables was tested by the x 2 and Fisher’s exact tests. A P value of less than 0.05 was considered statistically significant.
MICROSATELLITE INSTABILITY IN ENDOMETRIAL CANCER
TABLE 1 Distribution of Microsatellite Allelic Shifts among RER1 Cases Cases Loci
1
2
3
4
5
6
7
8
9
10
D6S87 D6S255 D6S193 D6S297 D6S149 D6S281 D6S446 TBP D2S123 D3S1611 D17S787 D2S119 D7S481 D13S175
1 1 2 2 2 1 2 1 1 2 1 1 1 1
2 2 2 1 2 2 2 2 1 2 2 2 1 1
2 1 2 1 2 1 2 2 2 2 2 2 2 2
1 1 2 2 2 1 2 2 2 2 2 2 2 2
2 2 1 1 2 2 2 2 1 2 1 2 1 2
1 1 1 1 2 1 1 1 1 2 1 1 1 1
2 1 2 1 2 1 2 1 1 2 2 1 1 2
1 2 2 2 2 2 2 2 1 2 1 2 1 2
2 2 1 1 2 1 2 2 1 1 2 2 2 1
2 2 2 1 2 1 2 2 2 2 2 2 1 2
Note. Minus (2) or plus (1) signs indicate, respectively, no alteration or instability.
RESULTS Alterations of microsatellite sizes in at least 3 of the 14 loci tested were detected in 10 of 51 (19.6%) endometrial cancers (ECs) and normal pairs (Table 1). MI at 1 or 2 loci was observed in 1 and in 4 cases, respectively. Therefore we found a group of 10 ECs showing high instability (at 3 or more loci, which were considered RER-positive cases) and a small group of 5 ECs showing low instability (at 1–2 loci). The alterations in sequence length of microsatellites involved from 1 to 10 dinucleotides, the majority consisting of additions rather than deletions.
249
The loci more frequently involved in MI were D2S123, D7S481, D6S255, D6S193, D6S281, and D6S297 (Table 1 and Fig. 2). Markers D2S123, D7S481, D6S281, and D6S87 presented the highest expansion of dinucleotides. Histologically 6/10 RER1 ECs were pure endometrioid adenocarcinomas, 2 were mixed endometrioid and clear cell carcinomas, 1 a mixed endometrioid and mucinous adenocarcinoma, and 1 an undifferentiated carcinoma. In 2 of the mixed RER1 adenocarcinomas the different components appeared strictly intermingled and could not be analyzed separately for MI, while in one case of mixed endometrioid– clear cell adenocarcinoma, both components showed a high MI after microdissection and separated molecular study. No serous papillary carcinoma, Mu¨llerian mixed tumor, or adenosarcoma was RER1. The presence of MI was statistically correlated with the presence of an endometrioid pattern (P 5 0.025) and presence of necrosis (P 5 0.004) (see Table 2). In contrast, RER phenotype did not show any statistical correlation with stage, size, grade, and mitotic index of the tumors. p53 protein was expressed more in RER-negative (14/37) ECs than in RER1 (1/8) tumors, and PgR and ER immunoreactivity was detected more frequently in RER-positive (6/8 for ER, 7/8 for PgR) ECs than in RER-negative (20/37 for ER, 19/36 for PgR) tumors, although the differences were not statistically significant. The mean age at clinical presentation of RER1 ECs was 70 years (range: 54 to 85 years), whereas the mean age of patients with RER2 tumors was 69 years (range: 39 to 88 years). Of two patients showing RER1 ECs, one had a history of colorectal cancer and the other was affected by an ovarian syncronous mucinous carcinoma. Follow-up data were obtained only for the endometrioid ECs. No information were available for the other histotypes.
FIG. 2. Distribution of unstable loci in endometrial cancers.
250
TIBILETTI ET AL.
TABLE 2 Correlations between MI Clinicopathologic and Immunohistochemical Characteristics in 51 Endometrial Cancers MI at $3 loci (RER1 ECs)
MI at ,3 loci (RER2 ECs)
P
Clinicopathologic characteristics Histological type Endometrioid (pure or mixed) Nonendometrioid Grade G1–G2 G3 Stage I–II III–IV Tumor necrosis Minimal/absent Present
9 1
21 20
0.025*
8 2
24 16
0.21
9 1
30 10
0.28
6 4
40 1
0.004*
7 1
23 14
0.17
2 6
17 20
0.25
1 7
17 19
0.07
Immunohistochemical features p53 Negative Positive ER Negative Positive PgR Negative Positive
Note. P, P value referred to MI at $3 loci. * Significant P value.
The mean follow-up period of cases of endometrioid ECs was 34 months ranging from 15 to 128 months. Eight of nine patients (88.8%) with RER1 endometrioid cancers were alive a mean follow-up of 31.4 months (range: 19 – 49); 1 (11.1%) patient died of disease 19 months after diagnosis. Among the 21 RER2 endometrioid ECs, 17 (80.0%) patients were alive after a mean follow-up of 34.7 months (range: 15–128), while 4 (19%) patients died of disease (mean follow-up: 31 months). Concerning the pedigree analysis, data were collected on three generations of all cases, but no typical HNPCC family was found. In four families, ECs were detected in 2 sisters (3 families) and in 3 sisters (1 family), respectively; these tumors showed either RER1 or RER2 phenotypes. DISCUSSION The RER1 phenotype has been reported in 20 –30% [6 –11, 18] of sporadic endometrial cancers, making it one of the most common alterations observed in this type of tumor. We classified as RER1 10/51 (19.6%) endometrial cancers in which at least 3 unstable loci were detected. Regarding the criteria for
defining RER1 phenotype, several authors [8, 9, 18] considered the cases showing 2 or more loci with MI to be RER1. Recently Zhou et al. [19] proposed a more stringent definition for RER positivity considering as significant an instability involving $30% of loci examined and recommended the use of BAT 26 mononucleotide repeat microsatellite for tumors of various origin. Eight cases in this study fulfilled this more stringent definition and showed BAT 26 size variations; the other 2 cases showed MI at 3 loci and no BAT 26 size variations (D. Furlan, unpublished results, 1998). The incidence of RER1 phenotype was different in the various tumors types examined, with significantly higher values (P 5 0.025) for adenocarcinomas with endometrioid features (9/30: 30%) in comparison to other (1/21: 4.8%) histotypes of ECs. Other authors have analyzed MI in different types of ECs, but they have not reported a positive correlation between RER1 phenotype and endometrioid adenocarcinomas [4, 7, 10, 18]. This is the first report in which the RER1 phenotype has been clearly demonstrated as characteristic of ECs with endometrioid differentiation. Interestingly, a significant correlation of MI with special histotypes, such as mucinous colorectal adenocarcinoma and intestinal-type gastric carcinoma, has been previously reported [14, 20 –22]. It is noteworthy that these colorectal and gastric RER1 adenocarcinomas seem to behave less aggressively than RER2 adenocarcinomas of the same sites. These findings favor the interpretation that MI characterizes a subgroup of carcinomas of different sites with a different tumor biology which is apparently due to a novel mechanism of carcinogenesis. In our series of ECs there was no correlation between MI and grade, stage, or mitotic index of the tumors. Concerning survival outcomes, no significant difference was observed between patients with RER1 endometrioid ECs and patients with RER2 endometrioid ECs. These results appear partially in contrast with previous observations [18] demonstrating a positive association of RER1 phenotype with high grade and poor prognosis of ECs. These discrepancies may be explained by the fact that the few studies dealing with this topic examine a small number of cases with a short follow-up [4, 7, 18]; therefore any conclusion about the relationship between MI, tumor grade, and survival outcome of ECs has to be drawn from the study of larger series of cases with adequate follow-up. In our study the detection of MI, both in early and in advanced ECs, confirms the suggestion that this genetic alteration, which has also been found in atypical precancerous endometrial hyperplasia [13], is probably an early event in endometrial tumorigenesis. The majority of RER1 EC with endometrioid aspects appeared to be p53 negative and expressed PgR and ER, although no statistical correlation was found. This trend appears to be in agreement with recent results reported by Lax and Kurman [23]. None of the RER1 ECs was associated with HNPCC syndrome or with any other form of cancer aggregation. This clearly demonstrates that ECs with MI do not reflect an under-
MICROSATELLITE INSTABILITY IN ENDOMETRIAL CANCER
lying genetic predisposition to ECs and suggests that MI may be an alternative mutation pathway in sporadic endometrial carcinogenesis. We evaluated MI at different loci, but only a few of them (D2S123, D7S481, D6S297, DS6255, D6S193, D6S281) were involved more frequently. Of note, the loci D2S123 and D7S481 are reported to be frequently involved in colorectal cancer MI [24, 25], while the other loci located on 6q (D6S255, D6S193, D6S281) have been first examined in this study and have been shown to be informative for ECs. In this context it is of interest to recall that the telomeric region of chromosome 6 between markers D6S193 and D6S281 is frequently deleted in ovarian [26 –29], breast [30], and endometrial cancers [31] and may harbor tumor suppressor genes [27, 29, 31]. The high involvement of chromosome 6 loci in endometrial cancer MI correlates well with the presence of candidate oncogenes in this region. In conclusion, our findings suggest that a mutator phenotype (RER positive) might play a crucial role in the early steps of the tumorigenesis in a subset of the endometrioid adenocarcinomas of the uterine corpus characterized by low histological grade, frequent expression of ER and PgR, and rare expression of mutant p53 protein. ACKNOWLEDGMENTS R.T. was supported by a grant from the Associazione Italiana Ricerca Cancro. C.C. was supported by grants from the University of Insubria and the Italian MURST. The authors thank Dr. J. H. Fentem (Euratom Center, Ispra) for language support in this work.
REFERENCES 1. Aaltonen L, Peltomaki P, Leach F, Sistonen P, Pylkkanen L, Mecklin J, Jarvinen H, Powell S, Jen J, Hamilton S, Peterson G, Kinzler K, Vogelstein B, De La Chapelle A: Clues to the pathogenesis of familial colorectal cancer. Science 260:812– 816, 1993 2. Eshleman JR, Markowitz SD: Mismatch repair defects in human carcinogenesis. Hum Mol Genet 5:1489 –1494, 1996 3. Moslein G, Tester DJ, Lindor NM, Honchel R, Cunningham JM, French AJ, Halling KC, Schwab M, Goretzki P: Microsatellite instability and mutation analysis of hMSH2 and hMLH1 in patients with sporadic, familial and hereditary colorectal cancer. Hum Mol Genet 5:1245–1252, 1996 4. Risinger JI, Berchuck A, Kohler MF, Watson P, Lynch HT, Boyd J: Genetic instability of microsatellites in endometrial carcinoma. Cancer Res 53:5100 –5103, 1993 5. Sumoi R, Hakala Ala Pietila T, Leminen A, Mecklin JP, Lehtovirta P: Hereditary aspects of endometrial adenocarcinoma. Int J Cancer 62:132– 137, 1995 6. Watson P, Lynch HT: Extracolonic cancer in hereditary nonpolyposis colorectal cancer. Cancer 71:677– 685, 1993 7. Duggan BD, Felix JC, Muderspach LI, Tourgeman D, Zheng J, Shibata D: Microsatellite instability in sporadic endometrial carcinoma. J Natl Cancer Inst 86:1216 –1220, 1994 8. Kobayashi K, Sagae S, Kudo R, Saito H, Koi S, Nakamura Y: Microsatellite instability in endometrial carcinomas: frequent replication errors in
251
tumors of early onset and/or of poorly differentiated type. Genes Chromosomes Cancer 14:128 –132, 1995 9. Katabuchi H, Van Rees B, Lambers AR, Ronnett BM, Blazes MS, Leach FS, Kathleen RC, Hedrick L: Mutations in DNA mismatch repair genes are not responsible for microsatellite instability in most sporadic endometrial carcinomas. Cancer Res 55:5556 –5560, 1995 10. Tashiro H, Lax SF, Gaudin PB, Isacson C, Cho KR, Hedrick L: Microsatellite instability is uncommon in uterine serous carcinoma. Am J Pathol 150:75–78, 1997 11. Helland A, Borresen-Dale AL, Peltomaki P, Hektoen M, Kristensen GB, Nesland JM, De La Chapelle A, Lothe RA: Microsatellite instability in cervical and endometrial carcinomas. Int J Cancer 70:499 – 501, 1997 12. Loeb LA: Microsatellite instability: marker of a mutator phenotype in cancer. Cancer Res 54:5059 –5063, 1994 13. Mutter GL, Boynton KA, Faquin WC, Ruiz RE, Jovanovic AS: Allelotype mapping of unstable microsatellites establishes direct lineage continuity between endometrial precancers and cancers. Cancer Res 56:4483– 4486, 1996 14. Kim H, Jen J, Vogelstein B, Hamilton SR: Clinical and pathological characteristics of sporadic colorectal carcinomas with DNA replication errors in microsatellite sequences. Am J Pathol 145:148 –156, 1994 15. Scully RE, Poulson H, Sobin LH: International Histological Classification and Histologic Typing of Female Genital Tract Tumors. Berlin, SpringerVerlag (1994) 16. Creasman WT: Announcement, FIGO stages: 1988 revisions. Gynecol Oncol 35:125–127, 1988 17. Hsu SM, Raine L, Fanger H: Use of avidin– biotin peroxidase complex (ABC) in immunoperoxidase techniques. J Histochem Cytochem 25:577– 580, 1981 18. Caduff RF, Johnston CM, Snoboda-Newman SM, Poy EL, Merajver SD, Frank TS: Clinical and pathological significance of microsatellite instability in sporadic endometrial carcinoma. Am J Pathol 148:1671–1678, 1996 19. Zhou X, Hoang J, Li Y, Seruca R, Carneiro F, Sobrinho-Simoes M, Lothe R, Gleeson C, Russell SEH, Muzeau F, Fle´jou J, Hoang-Xuan K, Lidereau R, Thomas G, Hamelin R: Determination of the replication errors phenotype in human tumors without the requirement for matching normal DNA by analysis of monucleotide repeat microsatellites. Genes Chromosomes Cancer 21:101–107, 1998 20. Bocker T, Schlegel J, Kullmann F, Stumm G, Zirngibl H, Epplen JT, Ruschoff J: Genomic instability in colorectal carcinomas: comparison of different evaluation methods and their biological significance. J Pathol 179:15–19, 1996 21. Dos Santos NR, Seruca R, Costancia M, Seixas M, Sobrinho-Simoes M: Microsatellite instability at multiple loci in gastric carcinoma: clinicopathological implications and prognosis. Gastroenterology 110:38 – 44, 1996 22. Renault B, Calistri D, Buonsanti G, Nanni O, Amadori D, Ranzani GN: Microsatellite instability and mutations of p53 and TGF-b RII genes in gastric cancer. Hum Genet 98:601– 607, 1996 23. Lax SF, Kurman RJ: A dualistic model for endometrial carcinogenesis based on immunohistochemical and molecular genetic analyses. Verh Dtsch Ges Pathol 81:228 –232, 1997 24. Leach FS, Nicolaides NC, Papadopoulos N, Liu B, Jen J, Parsons R, Peltomaki P, Sistonen P, Aaltonen LA, Nystrom-Lahti M, Guan XY, Zhang J, Meltzer PS, Yu JW, Kao FT, Chen DJ, Cerosaletti KM, Fournier REK, Todd S, Lewis T, Leach RJ, Naylor SL, Weissenbach J, Mecklin JP, Jarvinen H, Petersen GM, Hamilton SR, Green J, Jass J, Watson P, Lynch H, Trent JM, De La Chapelle A, Kinzler KW, Vogelstein B: Mutation of a mutS homolog in hereditary nonpolyposis colorectal cancer. Cell 75: 1215–1225, 1993
252
TIBILETTI ET AL.
25. Nicolaides N, Papadopoulos N, Liu B, Wel Y, Carter K, Ruben S, Rosen C, Haseltine W, Fleischmann R, Fraser C, Adams M, Venter J, Dunlopo M, Hamilton S, Petersen G, De La Chapelle A, Vogelstein B, Kinzler K: Mutation of two PMS homologues in hereditary nonpolyposis colon cancer. Nature 371:75– 80, 1994 26. Cooke IE, Shelling AN, Le Meuth VG, Charnock ML, Ganesan TS: Allele loss on chromosome arm 6q and fine mapping of the region at 6q27 in epithelial ovarian cancer. Genes Chromosomes Cancer 15:223–233, 1996 27. Saito S, Saito H, Koi S, Sagae S, Kudo R, Saito J, Noda K, Nakamura Y: Fine scale deletion mapping of the distal long arm of chromosome 6 in 70 human ovarian cancers. Cancer Res 52:5815–5817, 1992 28. Foulkes WD, Ragoussis J, Stamp GWH, Allan GJ, Trowsdale J: Frequent
loss of heterozygosity on chromosome 6 in human ovarian carcinoma. Br J Cancer 67:551–559, 1993 29. Tibiletti MG, Bernasconi B, Furlan D, Riva C, Trubia M, Buraggi G, Franchi M, Bolis PF, Mariani A, Frigerio L, Capella C, Taramelli R: Early envolvement of 6q in surface epithelial ovarian tumors. Cancer Res 56:4493– 4498, 1996 30. Noviello C, Courjal F, Theillet C: Loss of heterozygosity on the long arm of chromosome 6 in breast cancer: possibly four regions of deletion. Clin Cancer Res 2:1601–1606, 1996 31. Tibiletti MG, Bernasconi B, Taborelli M, Furlan D, Fabbri A, Franchi M, Taramelli R, Capella C: Involvement of chromosome 6 in endometrial cancer. Br J Cancer 75:1831–1835, 1997
Microsatellite Instability in Endometrial Cancer: Relation to Histological Subtypes M. G. Tibiletti,* D. Furlan,* M. Taborelli,* C. Facco,* C. Riva,* M. Franchi,† A. Cossu,‡ M. Trubia,§ R. Taramelli, ¶ and C. Capella* *Department of Clinical and Biological Sciences and Service of Pathological Anatomy, University of Insubria and Ospedale Di Circolo, 21100 Varese, Italy; †Obstetric and Gynecological Clinic and ¶Department of Structural and Functional Biology, University of Insubria, Varese, Italy; ‡Department of Pathology, University of Sassari, Italy; and §IRCCS H. S. Raffaele, Milan, Italy Received August 14, 1998
to be neutral or nonselective; some of them, however, may affect critical oncogenic or tumor suppressor loci [2]. Endometrial carcinomas exhibit many of the genetic features also present in RER-positive colorectal carcinomas [7, 13], such as MI and germline mutations in mismatch repair genes. However, the clinical and biological significance of MI in endometrial carcinomas has not yet been clarified. In particular, it is interesting to ascertain whether the association of the RER1 phenotype with a special histotype, as observed for colorectal carcinomas [14], might also exist for cancers of the endometrium. We have studied the prevalence of MI in endometrial cancers and its relationship to clinicopathological characteristics such as tumor grade, stage, histological type, age of onset, follow-up data, and familial pattern. In addition MI was correlated with the immunohistochemical expression of p53, progesterone receptor (PgR), and estrogen receptor (ER) by the tumors.
Fifty-one endometrial cancers were analyzed with regard to whether or how microsatellite instability (MI) was associated with the development of different types of endometrial malignant neoplasms. We investigated 6 loci previously reported as informative for colorectal cancer and a group of 8 loci located on 6q. Replication error (RER1) phenotype was detected in 10 of 51 (19.6%) endometrial cancers (ECs), all but one of which showed endometrioid differentiation. On the contrary, the RER1 phenotype was not detected in serous carcinomas and malignant mixed Mu¨llerian tumors. MI was present in both early and advanced stage ECs. No correlation was found between age, grade, stage, familial pattern, mitotic index, and the RER1 phenotype of ECs. Only 1 of 8 endometrial carcinomas showing MI was associated with mutant p53 expression, while the majority of RER1 tumors were positive for estrogen and progesterone receptors. Our findings suggest that MI plays an early role in endometrial tumorigenesis and is significantly correlated with adenocarcinomas showing endometrioid features (EAs). The frequent involvement of the telomeric region of chromosome 6 in the MI of EA is an indication that this region may be crucial in the process of EA tumorigenesis. © 1999 Academic Press Key Words: endometrial cancer; microsatellite instability; histotype; chromosome 6; endometrioid adenocarcinoma.
INTRODUCTION The instability of microsatellite DNA sequences (MI) was first described in a subset of sporadic colorectal cancers and in tumors associated with hereditary nonpolyposis colorectal cancer (HNPCC) [1–3]. Endometrial carcinoma is the most common noncolorectal carcinoma occurring in women affected by HNPCC [4, 5]. MI has been observed both in the inherited form (HNPCC) and in about 20% of presumably sporadic endometrial carcinomas [6 –11]. Microsatellite alterations represent a marker for a “mutator phenotype” (termed RER1, from replication error), which is characterized by widespread genetic instability and by an increased mutation rate [12]. Most of the microsatellite alterations identified in tumors are likely 247
MATERIAL AND METHODS Histopathological Study Formalin-fixed, paraffin-embedded tissue samples from endometrial cancers were obtained from 50 patients who had undergone primary surgery at the Varese or Sassari Hospital between 1983 and 1997. A total of 51 cases were studied; in 1 case two samples from the same tumor were isolated. Hematoxylin and eosin stained sections from each case were evaluated to assess the depth of myometrial invasion, grade, mitotic index, and presence of necrosis. Histological classification was based on WHO criteria [15] and stage was assessed following recomendations of the International Federation of Gynecology and Obstetrics [16]. The tumors were grouped as adenocarcinomas with endometrioid features (30 cases) and nonendometrioid cancers (21 cases), the latter ones including serous (11 cases), mucinous (1 case), clear cell (3 cases), undifferentiated carcinomas (2 cases), adenosarcoma (1 case), and malignant 0090-8258/99 $30.00 Copyright © 1999 by Academic Press All rights of reproduction in any form reserved.
248
TIBILETTI ET AL.
Mu¨llerian mixed tumors (3 cases). Clinical information and follow-up data for each case were obtained from clinicians, hospital charts, clinical reports or death certificates. Immunohistochemical Study Immunohistochemical staining was performed on 3-mm newly cut routinely processed sections using the avidin– biotin peroxidase complex procedure according to Hsu et al. [17]. Antigen unmasking using a citrate buffer, pH 6.0, and a microwave oven for 10 min at 650 W was performed prior to incubation with the primary antibodies. Antibodies employed were directed against p53 (monoclonal, clone D07, 1:1000, DAKO, Copenaghen, Denmark), estrogen receptor (monoclonal, clone 1D5, 1:1000, DAKO), and progesterone receptor (monoclonal, clone, 1A6, 1:20, Novocastra, Newcastle, UK). Immunohistochemical results were semiquantitatively evaluated as a percentage of positive nuclei. For p53 expression a positivity cutoff of 10% immunoreactive cells was established. Specimen Collection and DNA Extraction Fifty-one samples from 50 endometrial tumors and nonadjacent, histologically normal specimens (ovary or nonmetastatic lymph nodes) or peripheral blood lymphocytes were collected and frozen at 280°C for MI analysis. Genomic DNA was isolated from frozen tissues or from 10-mm sections of formalin-fixed paraffin-embedded tissue blocks using QIAamp tissue kits (Qiagen GmbH, Germany) or from whole blood using QIAamp blood kits (Qiagen GmbH). In order to exclude contaminations, both frozen and paraffin-embedded samples were examined histologically prior to their use for DNA extraction and extensive areas of tumor necrosis and normal epithelial or stromal components were eliminated using a surgical blade. The scraped tissues were transferred to a microfuge tube, and the identity of the tissue removed was verified by microscopic analysis of the remaining tissue on the glass slide. MI Analysis DNA from paired normal and tumor tissues were PCRamplified at 14 markers: D6S87 (6q21–23), D6S255 (6q24 – 25.1), D6S193 (6q27), D6S297 (6q27), D6S149 (6q27), D6S281 (6q27), D6S446 (6q27), TBP (6q27), D2S123 (2q16), D31611 (3p24.2p22), D17S787, D2S119 (2p16), D7S481 (7pter–p15), and D13S175 (13q11). The PCR mixture (50 ml) included 100 –200 ng of genomic DNA, 10 mM Tris, 50 mM KCl, 1–2 mM MgCl 2, 200 mM dNTPs, and 50 pmol of each primer, 2 U Taq polymerase (Perkin–Elmer, Italy). Samples were denatured at 94°C for 5 min followed by 30 cycles of denaturation (94°C, 1 min) annealing (55°C, 1 min) extension (72°C, 1 min) and a final extension of 5 min at 72°C in a Perkin–Elmer Gene Amp Thermal Cycler 480. PCR products were electrophoresed on 6% denaturing polyacrylamide se-
FIG. 1. Genetic instability in endometrial cancer patients. PCR products from tumor (T) and normal (N) DNAs were loaded in parallel; arrows indicate additional bands in tumor samples compared to normal samples.
quencing gels for 2 h at 1600 V, 40 mA, 50°C. Polyacrylamide gels fixed on glass plates with methacryloxypropyltrimethosysilane were silver stained using a DNA silver staining system (Promega, Italy). Silver-stained gels were reproduced on EDF films (Kodak). The RER-positive phenotype was assigned when MI was present in at least 3 of the 14 loci tested (Fig. 1). The clinicopathological parameters and immunohistochemical results were correlated considering MI of at least 3 loci (RER1 cases). Pedigree Analysis Psychologically assisted genetic counseling was carried out at the first clinical follow-up of each patient. At least three generations were analyzed, and a careful validation of cancer cases recorded in clinical reports or on death certificates was conducted. Statistical Analysis The relationship between variables was tested by the x 2 and Fisher’s exact tests. A P value of less than 0.05 was considered statistically significant.
MICROSATELLITE INSTABILITY IN ENDOMETRIAL CANCER
TABLE 1 Distribution of Microsatellite Allelic Shifts among RER1 Cases Cases Loci
1
2
3
4
5
6
7
8
9
10
D6S87 D6S255 D6S193 D6S297 D6S149 D6S281 D6S446 TBP D2S123 D3S1611 D17S787 D2S119 D7S481 D13S175
1 1 2 2 2 1 2 1 1 2 1 1 1 1
2 2 2 1 2 2 2 2 1 2 2 2 1 1
2 1 2 1 2 1 2 2 2 2 2 2 2 2
1 1 2 2 2 1 2 2 2 2 2 2 2 2
2 2 1 1 2 2 2 2 1 2 1 2 1 2
1 1 1 1 2 1 1 1 1 2 1 1 1 1
2 1 2 1 2 1 2 1 1 2 2 1 1 2
1 2 2 2 2 2 2 2 1 2 1 2 1 2
2 2 1 1 2 1 2 2 1 1 2 2 2 1
2 2 2 1 2 1 2 2 2 2 2 2 1 2
Note. Minus (2) or plus (1) signs indicate, respectively, no alteration or instability.
RESULTS Alterations of microsatellite sizes in at least 3 of the 14 loci tested were detected in 10 of 51 (19.6%) endometrial cancers (ECs) and normal pairs (Table 1). MI at 1 or 2 loci was observed in 1 and in 4 cases, respectively. Therefore we found a group of 10 ECs showing high instability (at 3 or more loci, which were considered RER-positive cases) and a small group of 5 ECs showing low instability (at 1–2 loci). The alterations in sequence length of microsatellites involved from 1 to 10 dinucleotides, the majority consisting of additions rather than deletions.
249
The loci more frequently involved in MI were D2S123, D7S481, D6S255, D6S193, D6S281, and D6S297 (Table 1 and Fig. 2). Markers D2S123, D7S481, D6S281, and D6S87 presented the highest expansion of dinucleotides. Histologically 6/10 RER1 ECs were pure endometrioid adenocarcinomas, 2 were mixed endometrioid and clear cell carcinomas, 1 a mixed endometrioid and mucinous adenocarcinoma, and 1 an undifferentiated carcinoma. In 2 of the mixed RER1 adenocarcinomas the different components appeared strictly intermingled and could not be analyzed separately for MI, while in one case of mixed endometrioid– clear cell adenocarcinoma, both components showed a high MI after microdissection and separated molecular study. No serous papillary carcinoma, Mu¨llerian mixed tumor, or adenosarcoma was RER1. The presence of MI was statistically correlated with the presence of an endometrioid pattern (P 5 0.025) and presence of necrosis (P 5 0.004) (see Table 2). In contrast, RER phenotype did not show any statistical correlation with stage, size, grade, and mitotic index of the tumors. p53 protein was expressed more in RER-negative (14/37) ECs than in RER1 (1/8) tumors, and PgR and ER immunoreactivity was detected more frequently in RER-positive (6/8 for ER, 7/8 for PgR) ECs than in RER-negative (20/37 for ER, 19/36 for PgR) tumors, although the differences were not statistically significant. The mean age at clinical presentation of RER1 ECs was 70 years (range: 54 to 85 years), whereas the mean age of patients with RER2 tumors was 69 years (range: 39 to 88 years). Of two patients showing RER1 ECs, one had a history of colorectal cancer and the other was affected by an ovarian syncronous mucinous carcinoma. Follow-up data were obtained only for the endometrioid ECs. No information were available for the other histotypes.
FIG. 2. Distribution of unstable loci in endometrial cancers.
250
TIBILETTI ET AL.
TABLE 2 Correlations between MI Clinicopathologic and Immunohistochemical Characteristics in 51 Endometrial Cancers MI at $3 loci (RER1 ECs)
MI at ,3 loci (RER2 ECs)
P
Clinicopathologic characteristics Histological type Endometrioid (pure or mixed) Nonendometrioid Grade G1–G2 G3 Stage I–II III–IV Tumor necrosis Minimal/absent Present
9 1
21 20
0.025*
8 2
24 16
0.21
9 1
30 10
0.28
6 4
40 1
0.004*
7 1
23 14
0.17
2 6
17 20
0.25
1 7
17 19
0.07
Immunohistochemical features p53 Negative Positive ER Negative Positive PgR Negative Positive
Note. P, P value referred to MI at $3 loci. * Significant P value.
The mean follow-up period of cases of endometrioid ECs was 34 months ranging from 15 to 128 months. Eight of nine patients (88.8%) with RER1 endometrioid cancers were alive a mean follow-up of 31.4 months (range: 19 – 49); 1 (11.1%) patient died of disease 19 months after diagnosis. Among the 21 RER2 endometrioid ECs, 17 (80.0%) patients were alive after a mean follow-up of 34.7 months (range: 15–128), while 4 (19%) patients died of disease (mean follow-up: 31 months). Concerning the pedigree analysis, data were collected on three generations of all cases, but no typical HNPCC family was found. In four families, ECs were detected in 2 sisters (3 families) and in 3 sisters (1 family), respectively; these tumors showed either RER1 or RER2 phenotypes. DISCUSSION The RER1 phenotype has been reported in 20 –30% [6 –11, 18] of sporadic endometrial cancers, making it one of the most common alterations observed in this type of tumor. We classified as RER1 10/51 (19.6%) endometrial cancers in which at least 3 unstable loci were detected. Regarding the criteria for
defining RER1 phenotype, several authors [8, 9, 18] considered the cases showing 2 or more loci with MI to be RER1. Recently Zhou et al. [19] proposed a more stringent definition for RER positivity considering as significant an instability involving $30% of loci examined and recommended the use of BAT 26 mononucleotide repeat microsatellite for tumors of various origin. Eight cases in this study fulfilled this more stringent definition and showed BAT 26 size variations; the other 2 cases showed MI at 3 loci and no BAT 26 size variations (D. Furlan, unpublished results, 1998). The incidence of RER1 phenotype was different in the various tumors types examined, with significantly higher values (P 5 0.025) for adenocarcinomas with endometrioid features (9/30: 30%) in comparison to other (1/21: 4.8%) histotypes of ECs. Other authors have analyzed MI in different types of ECs, but they have not reported a positive correlation between RER1 phenotype and endometrioid adenocarcinomas [4, 7, 10, 18]. This is the first report in which the RER1 phenotype has been clearly demonstrated as characteristic of ECs with endometrioid differentiation. Interestingly, a significant correlation of MI with special histotypes, such as mucinous colorectal adenocarcinoma and intestinal-type gastric carcinoma, has been previously reported [14, 20 –22]. It is noteworthy that these colorectal and gastric RER1 adenocarcinomas seem to behave less aggressively than RER2 adenocarcinomas of the same sites. These findings favor the interpretation that MI characterizes a subgroup of carcinomas of different sites with a different tumor biology which is apparently due to a novel mechanism of carcinogenesis. In our series of ECs there was no correlation between MI and grade, stage, or mitotic index of the tumors. Concerning survival outcomes, no significant difference was observed between patients with RER1 endometrioid ECs and patients with RER2 endometrioid ECs. These results appear partially in contrast with previous observations [18] demonstrating a positive association of RER1 phenotype with high grade and poor prognosis of ECs. These discrepancies may be explained by the fact that the few studies dealing with this topic examine a small number of cases with a short follow-up [4, 7, 18]; therefore any conclusion about the relationship between MI, tumor grade, and survival outcome of ECs has to be drawn from the study of larger series of cases with adequate follow-up. In our study the detection of MI, both in early and in advanced ECs, confirms the suggestion that this genetic alteration, which has also been found in atypical precancerous endometrial hyperplasia [13], is probably an early event in endometrial tumorigenesis. The majority of RER1 EC with endometrioid aspects appeared to be p53 negative and expressed PgR and ER, although no statistical correlation was found. This trend appears to be in agreement with recent results reported by Lax and Kurman [23]. None of the RER1 ECs was associated with HNPCC syndrome or with any other form of cancer aggregation. This clearly demonstrates that ECs with MI do not reflect an under-
MICROSATELLITE INSTABILITY IN ENDOMETRIAL CANCER
lying genetic predisposition to ECs and suggests that MI may be an alternative mutation pathway in sporadic endometrial carcinogenesis. We evaluated MI at different loci, but only a few of them (D2S123, D7S481, D6S297, DS6255, D6S193, D6S281) were involved more frequently. Of note, the loci D2S123 and D7S481 are reported to be frequently involved in colorectal cancer MI [24, 25], while the other loci located on 6q (D6S255, D6S193, D6S281) have been first examined in this study and have been shown to be informative for ECs. In this context it is of interest to recall that the telomeric region of chromosome 6 between markers D6S193 and D6S281 is frequently deleted in ovarian [26 –29], breast [30], and endometrial cancers [31] and may harbor tumor suppressor genes [27, 29, 31]. The high involvement of chromosome 6 loci in endometrial cancer MI correlates well with the presence of candidate oncogenes in this region. In conclusion, our findings suggest that a mutator phenotype (RER positive) might play a crucial role in the early steps of the tumorigenesis in a subset of the endometrioid adenocarcinomas of the uterine corpus characterized by low histological grade, frequent expression of ER and PgR, and rare expression of mutant p53 protein. ACKNOWLEDGMENTS R.T. was supported by a grant from the Associazione Italiana Ricerca Cancro. C.C. was supported by grants from the University of Insubria and the Italian MURST. The authors thank Dr. J. H. Fentem (Euratom Center, Ispra) for language support in this work.
REFERENCES 1. Aaltonen L, Peltomaki P, Leach F, Sistonen P, Pylkkanen L, Mecklin J, Jarvinen H, Powell S, Jen J, Hamilton S, Peterson G, Kinzler K, Vogelstein B, De La Chapelle A: Clues to the pathogenesis of familial colorectal cancer. Science 260:812– 816, 1993 2. Eshleman JR, Markowitz SD: Mismatch repair defects in human carcinogenesis. Hum Mol Genet 5:1489 –1494, 1996 3. Moslein G, Tester DJ, Lindor NM, Honchel R, Cunningham JM, French AJ, Halling KC, Schwab M, Goretzki P: Microsatellite instability and mutation analysis of hMSH2 and hMLH1 in patients with sporadic, familial and hereditary colorectal cancer. Hum Mol Genet 5:1245–1252, 1996 4. Risinger JI, Berchuck A, Kohler MF, Watson P, Lynch HT, Boyd J: Genetic instability of microsatellites in endometrial carcinoma. Cancer Res 53:5100 –5103, 1993 5. Sumoi R, Hakala Ala Pietila T, Leminen A, Mecklin JP, Lehtovirta P: Hereditary aspects of endometrial adenocarcinoma. Int J Cancer 62:132– 137, 1995 6. Watson P, Lynch HT: Extracolonic cancer in hereditary nonpolyposis colorectal cancer. Cancer 71:677– 685, 1993 7. Duggan BD, Felix JC, Muderspach LI, Tourgeman D, Zheng J, Shibata D: Microsatellite instability in sporadic endometrial carcinoma. J Natl Cancer Inst 86:1216 –1220, 1994 8. Kobayashi K, Sagae S, Kudo R, Saito H, Koi S, Nakamura Y: Microsatellite instability in endometrial carcinomas: frequent replication errors in
251
tumors of early onset and/or of poorly differentiated type. Genes Chromosomes Cancer 14:128 –132, 1995 9. Katabuchi H, Van Rees B, Lambers AR, Ronnett BM, Blazes MS, Leach FS, Kathleen RC, Hedrick L: Mutations in DNA mismatch repair genes are not responsible for microsatellite instability in most sporadic endometrial carcinomas. Cancer Res 55:5556 –5560, 1995 10. Tashiro H, Lax SF, Gaudin PB, Isacson C, Cho KR, Hedrick L: Microsatellite instability is uncommon in uterine serous carcinoma. Am J Pathol 150:75–78, 1997 11. Helland A, Borresen-Dale AL, Peltomaki P, Hektoen M, Kristensen GB, Nesland JM, De La Chapelle A, Lothe RA: Microsatellite instability in cervical and endometrial carcinomas. Int J Cancer 70:499 – 501, 1997 12. Loeb LA: Microsatellite instability: marker of a mutator phenotype in cancer. Cancer Res 54:5059 –5063, 1994 13. Mutter GL, Boynton KA, Faquin WC, Ruiz RE, Jovanovic AS: Allelotype mapping of unstable microsatellites establishes direct lineage continuity between endometrial precancers and cancers. Cancer Res 56:4483– 4486, 1996 14. Kim H, Jen J, Vogelstein B, Hamilton SR: Clinical and pathological characteristics of sporadic colorectal carcinomas with DNA replication errors in microsatellite sequences. Am J Pathol 145:148 –156, 1994 15. Scully RE, Poulson H, Sobin LH: International Histological Classification and Histologic Typing of Female Genital Tract Tumors. Berlin, SpringerVerlag (1994) 16. Creasman WT: Announcement, FIGO stages: 1988 revisions. Gynecol Oncol 35:125–127, 1988 17. Hsu SM, Raine L, Fanger H: Use of avidin– biotin peroxidase complex (ABC) in immunoperoxidase techniques. J Histochem Cytochem 25:577– 580, 1981 18. Caduff RF, Johnston CM, Snoboda-Newman SM, Poy EL, Merajver SD, Frank TS: Clinical and pathological significance of microsatellite instability in sporadic endometrial carcinoma. Am J Pathol 148:1671–1678, 1996 19. Zhou X, Hoang J, Li Y, Seruca R, Carneiro F, Sobrinho-Simoes M, Lothe R, Gleeson C, Russell SEH, Muzeau F, Fle´jou J, Hoang-Xuan K, Lidereau R, Thomas G, Hamelin R: Determination of the replication errors phenotype in human tumors without the requirement for matching normal DNA by analysis of monucleotide repeat microsatellites. Genes Chromosomes Cancer 21:101–107, 1998 20. Bocker T, Schlegel J, Kullmann F, Stumm G, Zirngibl H, Epplen JT, Ruschoff J: Genomic instability in colorectal carcinomas: comparison of different evaluation methods and their biological significance. J Pathol 179:15–19, 1996 21. Dos Santos NR, Seruca R, Costancia M, Seixas M, Sobrinho-Simoes M: Microsatellite instability at multiple loci in gastric carcinoma: clinicopathological implications and prognosis. Gastroenterology 110:38 – 44, 1996 22. Renault B, Calistri D, Buonsanti G, Nanni O, Amadori D, Ranzani GN: Microsatellite instability and mutations of p53 and TGF-b RII genes in gastric cancer. Hum Genet 98:601– 607, 1996 23. Lax SF, Kurman RJ: A dualistic model for endometrial carcinogenesis based on immunohistochemical and molecular genetic analyses. Verh Dtsch Ges Pathol 81:228 –232, 1997 24. Leach FS, Nicolaides NC, Papadopoulos N, Liu B, Jen J, Parsons R, Peltomaki P, Sistonen P, Aaltonen LA, Nystrom-Lahti M, Guan XY, Zhang J, Meltzer PS, Yu JW, Kao FT, Chen DJ, Cerosaletti KM, Fournier REK, Todd S, Lewis T, Leach RJ, Naylor SL, Weissenbach J, Mecklin JP, Jarvinen H, Petersen GM, Hamilton SR, Green J, Jass J, Watson P, Lynch H, Trent JM, De La Chapelle A, Kinzler KW, Vogelstein B: Mutation of a mutS homolog in hereditary nonpolyposis colorectal cancer. Cell 75: 1215–1225, 1993
252
TIBILETTI ET AL.
25. Nicolaides N, Papadopoulos N, Liu B, Wel Y, Carter K, Ruben S, Rosen C, Haseltine W, Fleischmann R, Fraser C, Adams M, Venter J, Dunlopo M, Hamilton S, Petersen G, De La Chapelle A, Vogelstein B, Kinzler K: Mutation of two PMS homologues in hereditary nonpolyposis colon cancer. Nature 371:75– 80, 1994 26. Cooke IE, Shelling AN, Le Meuth VG, Charnock ML, Ganesan TS: Allele loss on chromosome arm 6q and fine mapping of the region at 6q27 in epithelial ovarian cancer. Genes Chromosomes Cancer 15:223–233, 1996 27. Saito S, Saito H, Koi S, Sagae S, Kudo R, Saito J, Noda K, Nakamura Y: Fine scale deletion mapping of the distal long arm of chromosome 6 in 70 human ovarian cancers. Cancer Res 52:5815–5817, 1992 28. Foulkes WD, Ragoussis J, Stamp GWH, Allan GJ, Trowsdale J: Frequent
loss of heterozygosity on chromosome 6 in human ovarian carcinoma. Br J Cancer 67:551–559, 1993 29. Tibiletti MG, Bernasconi B, Furlan D, Riva C, Trubia M, Buraggi G, Franchi M, Bolis PF, Mariani A, Frigerio L, Capella C, Taramelli R: Early envolvement of 6q in surface epithelial ovarian tumors. Cancer Res 56:4493– 4498, 1996 30. Noviello C, Courjal F, Theillet C: Loss of heterozygosity on the long arm of chromosome 6 in breast cancer: possibly four regions of deletion. Clin Cancer Res 2:1601–1606, 1996 31. Tibiletti MG, Bernasconi B, Taborelli M, Furlan D, Fabbri A, Franchi M, Taramelli R, Capella C: Involvement of chromosome 6 in endometrial cancer. Br J Cancer 75:1831–1835, 1997