Genetics of synchronous uterine and ovarian endometrioid carcinoma: Combined analyses of loss of heterozygosity, PTEN mutation, and microsatellite instability

Genetics of Synchronous Uterine and Ovarian Endometrioid Carcinoma: Combined Analyses of Loss of Heterozygosity, PTEN Mutation, and Microsatellite Instability HIROAKI FUJII, MD, TOSHIHARU MATSUMOTO, MD, MANABU YOSHIDA, MD, YOSHIAKI FURUGEN, MD, TETSUYA TAKAGAKI, MT, KEIICHI IWABUCHI, MD, YASUO NAKATA, MD, YOSHITAKE TAKAGI, MD, TAKUYA MORIYA, MD, NAOMI OHTSUJI, MT, MAREKI OHTSUJI, MT, SACHIKO HIROSE, MD, AND TOSHIKAZU SHIRAI, MD Synchronous development of carcinomas in the endometrium and ovaries is a fairly common phenomenon, but distinction of a single clonal tumor with metastasis from 2 independent primary tumors may present diagnostic problems. To determine clonality and the occurrence of progression, we microdissected multiple foci from 17 cases of synchronous endometrioid carcinomas and studied loss of heterozygosity (LOH), microsatellite instability (MI), and PTEN mutations. In 14 of the 17 cases, genetic alterations were either homogeneous or found in only some of the foci. LOH was detected for 10q (4 cases), 17p (2 cases), and 2p, 5q, 6q, 9p, 11q, 13q, and 16q (1 case each). Four cases had the MI phenotype with discordant MI patterns between both tumor sites, thus indicating a biclonal or triple clonal process. In 3 of 6 cases with PTEN mutations, identical mutations in both tumor sites indicated a single clonal neoplasm. Altogether, 14 synchronous tumors were genetically diagnosed as follows: single clonal tumor, characterized by concordant genetic alterations in both tumor sites, including identical LOH, identical PTEN mutations, and/or identical sporadic allelic instability patterns (4 cases); single

clonal tumor with genetic progression, homogeneous LOH or identical PTEN mutations in both tumor sites and progressive LOH in ovarian metastatic foci (2 cases); and double (7 cases) or triple clonal tumors (1 case), determined by discordant PTEN mutations, heterogeneous LOH, and/or discordant MI patterns. Thus, 35% of synchronous tumors were monoclonal, 47% were polyclonal, and 18% were undetermined. The favorable prognosis of synchronous endometrioid carcinomas may be due to the occurrence of PTEN mutations in both independent and metastatic tumors, the MI-positive independent primary tumors, and the low frequency of LOH. HUM PATHOL 33:421-428. Copyright 2002, Elsevier Science (USA). All rights reserved. Key words: loss of heterozygosity, microsatellite instability, PTEN, endometrioid carcinoma, synchronous. Abbreviations: LOH, loss of heterozygosity; MI, microsatellite instability; PCR, Polymerase chain reaction; FIGO, International Federation of Gynecology and Obstetrics; SSCP, single-strand conformational polymorphism.

Carcinoma of the endometrium and the ovary can coexist; and the incidence of such cases is 34% to 40% in autopsies and 5% to 15% in surgically resected specimens.1-3 Sometimes these cases present a diagnostic difficulty when attempting to determine whether the 2 tumors involving different organs are independent primary tumors or represent a single clonal neoplasm with metastasis from 1 site to the other.3-5 The differential

diagnosis is critical, since the patient prognosis, International Federation of Gynecology and Obstetrics (FIGO) stage, and the choice of therapy differs in the 2 circumstances.3-6 When the simultaneous tumors are endometrioid adenocarcinomas, the most common combination for synchronous neoplasms, the patients often have a good prognosis2,5,7; in such cases, the neoplasms are considered independent primary tumors.1,2,6,8 In contrast, when the tumors show a non endometrioid histology, such as serous papillary carcinoma, clear cell carcinoma, or mucinous carcinoma, patients usually have a poor prognosis, and their tumors are assumed to be clonal neoplasms with metastasis from one site to the other.8,9 In some cases, however, differential diagnosis can be difficult; recently, distinct genetic pathways have been proposed for endometrioid versus nonendometrioid cancers.9-11 Bilaterality, size and extent of tumor, depth of myometrial invasion, vascular invasion, tubal involvement, presence of atypical hyperplasia in the uterus, endometriosis in the ovary, and nodular growth patterns facilitate the diagnosis in most cases.3,8,9,12 With advances in molecular biologic and other

From the Departments of Pathology and Obstetrics & Gynecology, Juntendo University School of Medicine, Tokyo, Japan; Department of Pathology, Kitasato University, Sagamihara, Japan; Department of Pathology, Hyogo College of Medicine, Nishinomiya, Japan; Division of Pathology, Koshigaya Hospital, Dokkyo University, Koshigaya, Japan; the Pathology Division, Tohoku University Hospital, Sendai, Japan. This work was supported in part by a research grant from the Ministry of Education, Science, Technology, Sports, and Culture of Japan. Address correspondence and reprint requests to Hiroaki Fujii, MD, PHd, Department of Pathology II, Juntendo University School of Medicine, 2-1-1, Hongo, Bunkyo-ku, Tokyo 113-8421, Japan. Copyright 2002, Elsevier Science (USA). All rights reserved. 0046-8177/02/3304-0007$35.00/0 doi:10.1053/hupa.2002.124118

421

HUMAN PATHOLOGY

Volume 33, No. 4 (April 2002)

techniques, including DNA flow cytometry, immunohistochemistry, X chromosomal inactivation, loss of heterozygosity (LOH), microsatellite instability (MI), and PTEN mutational studies, several investigators have attempted, using these various methods, to distinguish between primary and metastatic endometrial/ovarian cancers.13-17 Concordant genetic changes have been interpreted as indications of a single clonal process; discordant genetic alterations, as evidence of an independent neoplastic process. However, concern has been expressed that pathologic diagnosis based on these genetic findings alone may be misleading, because the tumors in both situations may undergo genetic progression and/or divergence and genetic heterogeneity may occur within each tumor, primary or metastatic.18 Also, relying only on a single genetic marker may not be conclusive for clonal determination. Actually, few of these previous studies have considered the occurrence of genetic heterogeneity within individual neoplasms. In our previous genetic studies on breast cancer,19 pancreatic cancer,20 and tumors with complex histologic patterns, including combined hepatocellularcholangiocarcinomas21 and gynecologic carcinosarcomas,22 we microdissected multiple neoplastic foci and carried out LOH analyses. We noted various patterns of genetic progression and heterogeneity that parallel histopathological progression and diversity. Thus this approach may provide a better understanding of temporospatial evolutional patterns of synchronous tumors as well. In the present study, we investigated the clonality and evolutionary course of synchronous endometrioid carcinoma of the uterine corpus and ovary by microdissecting multiple foci of these tumors and subsequently evaluating LOH, MI, and PTEN mutations. These genetic patterns, which correlated with the clinicopathologic findings, led to a better understanding of each case. MATERIALS AND METHODS Tissue Samples The 17 cases of synchronous endometrial and ovarian cancers used for this genetic analysis were obtained from files of the Pathology Department of Juntendo University, Hyogo College of Medicine, Koshigaya Hospital of Dokkyo University, and Kitasato University. All selected cases had been tested for the quality of DNA preservation and for adequate and reproducible polymerase chain reaction (PCR) amplification. The Japanese patients were age 31 to 61 years with a mean age of 48 years. None of the patients had a family history of cancer-related syndromes or hereditary nonpolyposis colorectal cancer. All of the collected samples were formalin-fixed, paraffinembedded tissue blocks from surgically resected tumors. Histologic tumor grading was scored according to FIGO as G1 (well differentiated), G2 (moderately differentiated), or G3 (poorly differentiated). Myometrial invasion was graded as a (tumor limited to the endometrium), b (invasion up to or less than one-half of the myometrial thickness), or c (invasion of more than one-half of the myometrial thickness). In 4 cases,

the ovarian tumors were associated with endometriosis, and in 3 cases there was bilateral ovarian involvement by the tumor. Because the clonality of synchronous tumors had not been determined, a FIGO stage was not assigned. Clinicopathologic profiles are summarized in Table 1.

Tissue Microdissection and DNA Extraction Multiple tumor foci in the uterus and the ovary were selected for microdissection. Some 8-␮m serial sections were cut, deparaffinized, stained with hematoxylin and eosin, visualized using an inverted microscope, and microdissected using a 27-gauge needle. Microscopically, at least 80% to 90% of the microdissected cells were estimated to be tumor cells. These highly purified samples facilitated subsequent detection of allelic losses and mutations, as shown in Figures 1 through 4. Six to 10 foci were microdissected from 17 cases. Stromal and inflammatory cells were collected separately and served as normal controls. Microdissected tissue was digested overnight at 50°C in buffer containing 0.5% Nonidet P-40, 50 mM Tris-HCl pH 8.0, 1 mM ethylenediamine tetraacetic acid, and 200 ␮g/mL proteinase K. The lysate was heated at 95°C for 10 minutes and stored at ⫺20°C until subjected to PCR reaction.

Detection of LOH PCR reactions contained 1 ␮L of DNA lysate, 0.4 ␮M of [␥-32p] ATP-radiolabeled microsatellite primers, 0.2mM of deoxyribonucleoside triphosphate, 10 mM of Tris-HCl pH 8.3, 1.5mM of MgCl2, 50 mM of KCl, and 0.4 U of Taq polymerase in a total reaction volume of 10 ␮L. Taq was added to the reactions prewarmed to 94°C (hot-start PCR), and the samples were amplified with 35 cycles of PCR amplification. Each PCR reaction contained an average of 100 to 200 cells. The PCR products were separated on a 5% denaturing polyacrylamide-urea-formamide gel, and LOH was determined based on a more than 75% reduction of relative intensity in 1 of the 2 alleles compared to those in the normal controls. Elimination of PCR artifacts was done as described.20 When only a portion of the microdissected foci showed LOH, PCR reactions were repeated at least 3 times to confirm the LOH and exclude spurious PCR reactions. Other informative microsatellite markers located on the same chromosomal arm also confirmed LOH. Microdissection was repeated as required.

Microsatellite instability Markers BAT 25, BAT 26, D2S123, D5S346, and D17S250 were used to identify tumors of the MI phenotype as recommended by the National Cancer Institute Workshop on Microsatellite Instability.23 Tissues with novel bands of microsatellite repeats in the gels with at least 2 of these 5 markers were considered to be MI positive. All of the markers used for LOH were also assessed for MI patterns. Novel alleles of microsatellite repeats were confirmed by repeating PCR reactions to rule out artefacts.

Microsatellite Markers All PCR primers for microsatellite markers were purchased from Research Genetics Co., Ltd. (Huntsville, AL). The microsatellite markers were selected to cover chromosomal regions commonly deleted in endometrial, ovarian, and various types of human cancers. The following primers were used: 1p and 1q (D1S228, D1S158, D1S318, and

422

GENETICS OF SYNCHRONOUS GYNECOLOGIC CANCER (Fujii et al)

TABLE 1. Clinicopathologic Summary Uterine tumor

Case

Age

Grade

Depth

Cervical involvement

1 2 3

51 49 54

G1 G1 G1

a a b

None Inv⫹, deep Inv⫹

4 5 6

33 49 54

G1 G3 G1

b c b

7

31

G1

8

49

9

Ovarian tumor Associated findings

Rt/Lt

Grade

Rt Rt Rt

G1 G1 G1

Inv⫹ Inv⫹ Surface⫹

Rt Rt Rt

G2 G3 G1

b

None

Bil Lt ⬎ Rt

G1

G2

a

None

Lt

G2

35

G1

b

None

Lt

G1

10 11 13 18

57 50 58 42

G2 G1 G1 G1

b b b b

Inv⫹ None None Inv⫹

Bil Lt ⫽ Rt Rt Lt Lt

G2 G1 G1 G2

20 21 22 23

61 48 48 49

G2 G1 G1 G1

b c b c

Inv⫹ None Inv⫹ None

Rt Rt Rt Bil Lt ⫽ Rt

G2 G1 G1 G1

Focal S

Focal S Adenomyosis Adenomyosis

Serosal deposits

Associated findings

Focal muci diff Capsular invasion Capsular invasion Bilateral ovarian endometriosis Surface deposits on the Lt Lt endometriosis Focal S Capsular invasion Capsular invasion Bilateral endometriosis Capsular invasion Capsular invasion Focal S Lt endometriosis

Capsular invasion

Metastasis to other sites None None None None LN, omentum None None None None LN, parametrium None None None None LN Parametrium LN, parametrium

Abbreviations: Inv⫹, cervical stromal invasion; Surface⫹, cervical glandular involvement; S, squamous differentiation; Rt, right ovary; Lt, left ovary; muci diff, mucinous differentiation; LN, lymph node; Bil, bilateral; ⫽, sizes were almost equal; ⬎, one tumor was much larger than the other.

D1S197), 3p (D3S1234, D3S1286, and D3S1293), 4q (D4S424, D4S415, and D4S413), 5q (D5S1956, D5S346, D5S2072, D5S644, D5S647, and D5S421), 6q (D6S292 and D6S264), 8p (D8S264, D8S255, D8S258, and D8S261), 9p (D9S1748 and D9S1749), 10q (D10S221, D10S219, and D10S574), 11p (D11S1324), 11q (D11S29 and Int2), 13q (D13S166, D13S263, and D13S171), 16q (D16S265, D16S541,

FIGURE 1. Representative gels showing LOH and deduced clonal evolution for case 11. N, normal control DNA; T#, microdissected tumor DNA; UT, endometrial tumor foci; OV, ovarian tumor foci; dot, normal alleles; arrow in the gel, LOH detected throughout tumor foci; arrowheads, LOH detected only in ovarian tumor foci. The tumor is considered to be a single clone originating in the endometrium with LOH of 10q and 11q. The tumor metastasized to the ovary with progressive LOH of 17p.

and D16S261), 17p (TP53, CHRNB1, and D17S786), 18q (D18S46, D18S55, D18S474, and D18S487), and 22q (D22S284). Additional markers, BAT 25, BAT 26, D2S123, D5S346, and D17S250, were also used to detect microsatellite instability.

Mutations of the PTEN Gene The following primer pairs were used for single-strand conformational polymorphism (SSCP) and sequencing analysis of exons 5 and 8 of the PTEN gene: 5⬘-acctgttaagtttgtatgcaac-3⬘/5⬘-ctttacagtgaattgctgcaac-3⬘ for the proximal portion of exon 5, 5⬘-accacagctagaacttatcaaa-3⬘/5⬘-tccaggaagaggaaaggaaa-3⬘ for the distal portion of exon 5, 5⬘-tgtcatttcatttctttttcttttc-3⬘/5⬘-tctgcacgctctatactgcaaatgctatc-3 for the proximal portion of exon 8, and 5⬘-gaagtctatgtgatcaagaaatc-3⬘/5⬘-cacacatcacatacatacaagtcacc-3⬘ for the distal portion of exon 8. For SSCP, the PCR products were diluted in formamide dye and run on 5% acrylamide and 5% glycerol gels at 4°C and at 23°C. Cases with aberrant bands on SSCP were further submitted for direct sequencing analysis using the same PCR primers. Cycle sequencing using the ABI PRISM Dye Terminator Cycle Sequencing Ready Reaction Kit (Applied Biosystems, Foster City, CA) was done using a gel-purified PCR product as the template. Each sequencing product was run on an ABI 373A sequencer (Applied Biosystems).

RESULTS Of the 17 cases studied, 3 (cases 9, 18, and 20) showed no allelic deletion, MI, or PTEN mutation, and clonal determination was impossible. The remaining 14

423

HUMAN PATHOLOGY

Volume 33, No. 4 (April 2002)

FIGURE 3. Representative microsatellite amplifications, histology, and clonal evolution of case 6. N, normal control DNA; T#, microdissected tumor DNA; UT, endometrial tumor foci; OV, ovarian tumor foci; A, MI pattern A; B, MI pattern B; C, MI pattern C; dot, normal alleles; arrows, novel alleles due to instability. (Hematoxylin and eosin, original magnification ⫻ 200.)

FIGURE 2. Representative gels showing MI and LOH for case 21 and the deduced double-clonal process. (A) Gels showing 2 discordant MI patterns and LOH of 10q for only the ovarian tumor foci. N, normal control DNA; T#, microdissected tumor DNA; UT, endometrial tumor foci; OV, ovarian tumor foci; dot, normal alleles; arrows, novel alleles due to MI; arrowheads, LOH. (B) Representative histology of case 21 and schema showing double-clonal tumors. (Hematoxylin and eosin original magnification ⫻200.)

13q, and 16q (1 case each; 6%). LOH was detected either homogeneously throughout the microdissected uterine and ovarian tumor foci or in only some of the foci. Homogeneous LOH with other genetic changes strongly indicate a single clonal process (cases 2, 3, and 5). Together with other genetic alterations and clinicopathologic status, heterogeneous LOH was considered either a genetic progression of the single-clonal or double-clonal process. In case 11, all of the microdissected endometrial and ovarian foci showed homoge-

cases showed genetic alterations (see Table 2). Cases with concordant genetic alterations (often multiple parameters) were interpreted to be a single clonal tumor. Dissimilar genetic alterations were interpreted to be a multiclonal tumor or a single-clonal tumor with genetic progression, as based on other genetic markers and the clinicopathologic findings. Loss of Heterozygosity LOH was detected in 7 cases. The highest frequency of LOH was detected for 10q (4 cases; 24%), followed by 17p (2 cases; 12%), and 2p, 5q, 6q, 9p, 11q,

FIGURE 4. Representative histology, antisense sequencing reaction of PTEN exon 5 at codon 112, LOH of 6q in ovarian foci and deduced genetic pathways for case 8. (Hematoxylin and eosin, original magnification ⫻ 200.)

424

GENETICS OF SYNCHRONOUS GYNECOLOGIC CANCER (Fujii et al)

TABLE 2. Summary of Genetic Analysis Case

LOH/MI

1 2

10q LOH (both UT and OV, homogeneous)

3

2p LOH/sporadic MI of DIS158 (both UT and OV, homogeneous) 10q LOH (focal in UT, homogeneous in OV)

4 5 6 7 8

5q, 9p, 13q, 16q, 17p LOH (both UT and OV, homogeneous) M1, 3 lineages (2 clones in UT, 1 in OV) Sporadic MI, (DIS158, TP53), 2 lineages (1⫻ in UT, 1⫻ in bil OV) 6q LOH (OV only)

9 10 11

22 23

Exon 8, OV only (codon 335, arg to stop) Exon 8, both UT and OV (codon 323 ins t) Exon 5, OV only (codon 129 silent mut)

Double clones Single clone

Double clones Single clone

ND

Triple clones (2⫻ in UT, OV, third primary) Double clones (1⫻ in UT, 1⫻ in bil OV)

Exon 5, both UT and OV (codon 112, leu to val)

Single clone, UT primary, progression in OV Not determined Single clone

10q, 11q LOH (both UT and OV, homogeneous), 17p LOH (only OV) Exon 8, UT only/Ex5, OV only (SSCP only) MI, 2 lineages (1⫻ in UT, 1⫻ in OV), 10q LOH (only OV) MI, 2 lineages (1⫻ in UT, 1⫻ in Ov) MI, 2 lineages (1⫻ in UT, 1⫻ in bil OV)

Clonality and Progression

Single clone

Exon 5, both UT and OV (codon 130, arg to gly)

13 18 20 21

PTEN Mutation

Single clone, UT primary, progression in OV Double clones

ND

Not determined Not determined Double clones

ND ND

Double clones Double clones (1⫻ in UT, 1⫻ in bil OV)

Abbreviations: OV, ovary; UT, uterine endometrium; MI, microsatellite instability; ND, not done; bil, bilateral; SSCP only, only SSCP was done.

neous LOH of 10q and 11q, strongly indicating a single clonal process. LOH of 17p was detected in all of the ovarian foci but in none of the endometrial foci. Thus this case was considered to be a single tumor originating in the uterus with ovarian metastasis and genetic progression, as shown in Figure 1. Patterns of High-Level Microsatellite Instability With a panel of markers frequently used for detection of MI phenotype tumors (BAT 25, BAT 26, D2S123, D5S346, and D17S250),23 4 cases (cases 6, 21, 22, and 23) showed 2 to 5 markers with MI. All of these 4 had multiple microsatellite markers with extensive allelic instability. No other case showed a single marker with MI, determined using the 5 standard markers. Because patterns of allelic instability can trace clonal evolution of the neoplasm, and sharing of the number of loci with instability indicates a common clonal origin, allelic instability patterns from multiple microdissected foci were compared. In cases 21, 22, and 23, instability patterns indicated double-clonal tumors. In these cases, MI patterns were markedly different in endometrial and ovarian foci. However, multiple foci in each tumor site shared many novel alleles of microsatellite repeats. Several new alleles were also detected in individual microdissected foci. In case 23, the bilateral ovarian tumors shared many of the novel repeats, indicating metastasis from one side to the other. In case 21,

along with two distinct MI patterns, LOH of 10q was detected in only the ovarian tumor, as shown in Figure 2. In case 6, three distinct MI patterns were detected, as shown in Figure 3. Two different patterns were seen in the endometrium, and an additional pattern was detected in the ovary. Within each pattern, multiple microdissected foci had many shared novel repeats. Thus, case 6 was considered to be a triple cancer. Low-Level (Sporadic) Microsatellite Instability In two cases (cases 3 and 7), 1 or 2 microsatellite markers showed novel alleles of microsatellite repeats in microdissected foci. In case 3, exactly the same novel allele for D1S158 was uniformly detected in all of the microdissected endometrial and ovarian tumor foci, but not in the normal control DNA. LOH of D2S123 was also detected in both uterine and ovarian foci. Thus case 3 was considered to be a single clonal tumor. In case 7, different new repeats for TP53 were detected in endometrial and ovarian tumor foci. In the ovarian foci, an additional novel repeat was detected on D1S158. The bilateral ovarian tumors shared these novel alleles of microsatellite repeats. Thus the case was considered to be a double endometrial/ovarian cancer. The bilateral ovarian tumors were interpreted as metastases from the left to the right ovary, which contained the small tumor.

425

HUMAN PATHOLOGY

Volume 33, No. 4 (April 2002)

TABLE 3. Clinicopathological profiles and clonality of the tumors Clnicopathologic Profiles Grade of endometrial tumor Grade of ovarian tumor Depth of myometrial invasion Ovarian tumor Ovarian endometriosis Metastases to other sites

Single

Double

Triple

UD

3 2 1 3 2 1 2 3 1 5 1 6

7

1

2 1

6 1

1

1 2

1

3

1

3

1 1

1 2 3

G1 G2 G3 G1 G2 G3 a b c Unilateral Bilateral None Present None Present

1 4 2 5 2 6 1 4 3

4 2

Note. Data represent the number of cases in each category. Abbreviation: UD, undetermined.

PTEN Mutation Initial SSCP screening of exons 5 and 8 and subsequent cycle sequencing revealed 3 cases of mutation in exon 8 (cases 1, 2, 4 and 13), and 4 cases of mutation in exon 5 (cases 4, 8, 10, and 13). Mutations were homogeneously detected in samples from both endometrial and ovarian tumors (2, 8, and 10) or in samples from only 1 of the 2 neoplasms (cases 1, 4, and 13), as summarized in Table 2. In case 8, both endometrial and ovarian foci showed the same mutation in exon 5. At codon 112, amino acid substitution from leucine (cta) to valine (gta) was detected in both foci but not in the normal control DNA, thus excluding a germline mutation. LOH of 6q was detected only in ovarian foci. Clinically, the endometrial tumor was confined to the endometrium but was a moderately differentiated tumor (G2) in both uterine and ovarian sites. Based on the identical PTEN mutation, we considered that this tumor had originated in the uterus and metastasized to the ovary, where genetic progression had probably occurred (6q LOH only in the ovarian foci), as shown in Figure 4. In case 4, PTEN mutation in exon 5 was detected in all of the microdissected ovarian foci but in none of the endometrial foci. 10q LOH was detected in all of the ovarian foci but only focally in the endometrial foci. Thus case 4 was interpreted to be a double clonal cancer with coincidental LOH of 10q in both organs. Summary of Genetic Alterations and Clonality Altogether, we classified 14 synchronous endometrial and ovarian tumors into the following 3 patterns: 1. Single clonal tumor with concordant genetic alterations, defined by identical LOH, identical PTEN mutation, and/or sporadic instability patterns (4 cases: 2, 3, 5, and 10). 2. Single clonal tumor with genetic progression, defined by concordant LOH or identical PTEN

mutation and progressive LOH in ovarian metastatic foci (2 cases: 8 and 11). 3. Double or triple clonal tumors, defined by discordant PTEN mutations, heterogeneous LOH, and/or discordant MI patterns (7 cases of double cancer: 1, 4, 7, 13, 21, 22, and 23 and 1 case of triple clonal tumors: 6). Relationship Between Clonality and Clinicopathologic Profiles Table 3 summarizes the clonality versus clinicopathologic profiles, including the grade of each tumor, depth of myometrial invasion, unilateral or bilateral presence of ovarian endometriosis, and metastasis to other sites. DISCUSSION In this study, a combination of multiple genetic parameters corresponding to the multiple microdissected foci of synchronous tumors allowed us to more clearly infer the probable clonality as well as the clonal progression pathways of these tumors. These findings are summarized in Table 2. Although we investigated microsatellite markers on many chromosomal arms, LOH of alleles other than 10q were found infrequently. These findings are similar to the reported allelic loss patterns of sporadic endometrioid carcinomas and contrast with those of tumors exhibiting other histologic types, which usually show LOH more often. 10,11 The PTEN gene was found to be mutated in 30% to 50% of the endometrial endometrioid cancers.24-26 In contrast to other cancers in which PTEN mutations occur in advanced tumors, including glioblastomas, melanomas, and breast cancers, PTEN mutations in endometrial cancers have been found as early as the precancerous stages,27 and most are closely associated with favorable endometrioid histology.26 In ovarian car-

426

GENETICS OF SYNCHRONOUS GYNECOLOGIC CANCER (Fujii et al)

cinomas, PTEN mutations have been encountered in 21% of endometrioid carcinomas, but rarely in carcinomas of other histologic types.28 We found 3 cases with identical PTEN mutations in both the endometrial and ovarian tumors, indicating the metastatic nature of these tumors. In case 8, although the uterine tumor was confined to the endometrium, both the endometrial and the ovarian foci showed the same PTEN mutation. Thus we postulated that the cells of a single clonal endometrial tumor containing an early inactivating PTEN mutation were shed from the endometrium and implanted in the ovary, where clonal progression with 6q LOH most likely occurred. Similar patterns of clonal expansions have been genetically identified in neoplasms of other organs as well.29,30 Instability of microsatellite sequences, one of the most common genetic alterations of cancers, has been found in 17% to 25% of uterine endometrioid carcinomas,31 and in 11% to 17% of ovarian endometrioid and clear cell carcinomas.32 In our series, 4 of 17 cases (24%) exhibited such instability, a frequency similar to that of MI in sporadic endometrial carcinomas. Microsatellite alterations are not a clonally stable phenomenon, and thus novel alleles of microsatellite repeats appear and evolve in clonal neoplastic processes.33 These microsatellite alterations can be detected even in early lesions, such as endometrial hyperplasia.27,33 Thus we compared MI patterns of multiple foci with multiple genetic markers. Although we detected the presence of new sequences with some of the microsatellite markers even within the same tumor, thus indicating progressive expansion of neoplastic processes, we found that MI patterns generally were maintained within individual tumors. In contrast, multiple genetic markers showed remarkably discordant allelic changes between different neoplastic clones, a finding that makes MI analysis highly effective in clonality analysis. In addition to MI phenotypes, we also detected sporadic instability patterns of a few microsatellite markers in 2 cases (3 and 7), a result that could also be used as a marker. Because all of our 4 cases with MI had instability in both the endometrial and ovarian tumors, and all tumors proved to be independent primary tumors, there appears to be a common underlying genetic abnormality for carcinogenesis in these patients. In this regard, promoter hypermethylation and loss of expression of the mismatch-repair gene hMLH1 may be causally related to sporadic endometrial cancer with MI.34 Predisposing abnormalities of other, as-yet to be identified mismatch repair genes may also be involved in these cases of synchronous double cancer with the MI phenotype. As summarized in Table 3, there were noteworthy findings related to the clinicopathologic profiles and the clonality of tumors. The presence of deep myometrial invasion of the endometrial carcinoma has been considered one histologic criterion favoring the interpretation that the synchronous ovarian tumor is metastatic.8,12 In the present study, however, 5 cases with genetically proven metastatic carcinomas to the ovary

showed no myometrial invasion or myometrial invasion, but of less than half of the myometrial thickness (cases 2, 3, 8, 10, and 11). Case 2 showed deep cervical invasion and likely metastatic potential. Case 3 showed invasion of almost one-half of the myometrial wall, and case 10 showed lymphatic invasion. Thus these 3 cases may have metastasized to the ovary despite their lack of deep myometrial invasion. However, case 11 showed superficial myometrial invasion with no angiolymphatic invasion, and case 8 was confined to the endometrium. It appears that the lack of deep myometrial invasion does not exclude the metastatic origin of the simultaneous ovarian tumors. Histologically, the presence of endometriosis in the ovary generally means that the ovarian tumor is a primary lesion rather than a metastatic lesion.3,8,12 In our series, 4 cases showed concomitant ovarian endometriosis, and in fact 2 were genetically primary ovarian tumors; the other 2 cases did not show signs of genetic alteration. Genetic investigations have recently shown that endometriosis is a genetic precursor of ovarian cancer.35,36 Thus our findings provide additional evidence that the coexistence with ovarian endometriosis is indicative of the primary nature of the adjacent ovarian carcinoma. The presence of bilateral ovarian tumors has been considered one of the features favoring ovarian metastases.3,8,12 Of the 3 such cases that we identified, only 1 case proved to be metastatic from the endometrial carcinoma; the other 2 cases turned out to be primary ovarian carcinoma with metastasis to the contralateral ovary. In conclusion, our investigation of the clonality and genetic pathways of synchronous ovarian and endometrioid cancers indicated that 35% of the synchronous tumors were monoclonal, 47% were polyclonal, and 18% were undetermined. The favorable prognosis of these cases may be related to involvement of the PTEN tumor-suppressor gene in metastatic and independent carcinomas, MI-positive independent primary tumors, and low frequency of LOH. Acknowledgment. The authors thank M. Ohara for comments and assistance with the language.

REFERENCES 1. Eisner RF, Nieberg RK, Berek JS: Synchronous primary neoplasms of the female reproductive tract. Gynecol Oncol 33:335-339, 1989 2. Kline RC, Wharton JT, Atkinson EN, et al: Endometrioid carcinoma of the ovary: Retrospective review of 145 cases. Gynecol Oncol 39:337-346, 1990 3. Young RH, Scully RE: Metastatic tumors of the ovary, in R. J. Kurman (ed): Blaustein’s Pathology of the Female Genital Tract (ed 4). New York, NY: Springer Verlag, 1994, pp 939-974 4. Russell P, Bannatyne PM, Solomon HJ, et al: Multifocal tumorigenesis in the upper female genital tract—Implications for staging and management. Int J Gynecol Pathol 4:192-210, 1985 5. Pearl ML, Johnston CM, Frank TS, et al: Synchronous dual primary ovarian and endometrial carcinomas. Int J Gynaecol Obstet 43:305-312, 1993 6. Jambhekar NA, Sampat MB: Simultaneous endometrioid car-

427

HUMAN PATHOLOGY

Volume 33, No. 4 (April 2002)

cinoma of the uterine corpus and ovary: A clinicopathologic study of 15 cases. J Surg Oncol 37:20-23, 1988 7. Falkenberry SS, Steinhoff MM, Gordinier M, et al: Synchronous endometrioid tumors of the ovary and endometrium. A clinicopathologic study of 22 cases. J Reprod Med 41:713-718, 1996 8. Eifel P, Hendrickson M, Ross J, et al: Simultaneous presentation of carcinoma involving the ovary and the uterine corpus. Cancer 50:163-170, 1982 9. Ulbright TM, Roth LM: Metastatic and independent cancers of the endometrium and ovary: A clinicopathologic study of 34 cases. HUM PATHOL 16:28-34, 1985 10. Pere H, Tapper J, Wahlstrom T, et al: Distinct chromosomal imbalances in uterine serous and endometrioid carcinomas. Cancer Res 58:892-895, 1998 11. Tritz D, Pieretti M, Turner S, et al: Loss of heterozygosity in usual and special variant carcinomas of the endometrium. HUM PATHOL 28:607-612, 1997 12. Choo YC, Naylor B: Multiple primary neoplasms of the ovary and uterus. Int J Gynaecol Obstet 20:327-334, 1982 13. Prat J, Matias-Guiu X, Barreto J: Simultaneous carcinoma involving the endometrium and the ovary. A clinicopathologic, immunohistochemical, and DNA flow cytometric study of 18 cases. Cancer 68:2455-2459, 1991 14. Shenson DL, Gallion HH, Powell DE, et al: Loss of heterozygosity and genomic instability in synchronous endometrioid tumors of the ovary and endometrium. Cancer 76:650-657, 1995 15. Fujita M, Enomoto T, Wada H, et al: Application of clonal analysis. Differential diagnosis for synchronous primary ovarian and endometrial cancers and metastatic cancer. Am J Clin Pathol 105: 350-359, 1996 16. Emmert-Buck MR, Chuaqui R, Zhuang Z, et al: Molecular analysis of synchronous uterine and ovarian endometrioid tumors. Int J Gynecol Pathol 16:143-148, 1997 17. Lin WM, Forgacs E, Warshal DP, et al: Loss of heterozygosity and mutational analysis of the PTEN/MMAC1 gene in synchronous endometrial and ovarian carcinomas. Clin Cancer Res 4:2577-2583, 1998 18. Matias-Guiu X, Bussaglia E, Catasus L, et al: Comment on: Clin Cancer Res 4:2577-2583, 1998. Clin Cancer Res 6:1598-1600, 2000 19. Fujii H, Marsh C, Cairns P, et al: Genetic divergence in the clonal evolution of breast cancer. Cancer Res 56:1493-1497, 1996 20. Yamano M, Fujii H, Takagaki T, et al: Genetic progression and divergence in pancreatic carcinoma. Am J Pathol 156:2123-2133, 2000 21. Fujii H, Zhu XG, Matsumoto T, et al: Genetic classification of combined hepatocellular-cholangiocarcinoma. HUM PATHOL 31: 1011-1017, 2000 22. Fujii H, Yoshida M, Gong ZX, et al: Frequent genetic heterogeneity in the clonal evolution of gynecological carcinosarcoma and its influence on phenotypic diversity. Cancer Res 60:114-120, 2000

23. Boland CR, Thibodeau SN, Hamilton SR, et al: A National Cancer Institute Workshop on Microsatellite Instability for cancer detection and familial predisposition: Development of international criteria for the determination of microsatellite instability in colorectal cancer. Cancer Res 58:5248-5257, 1998 24. Bussaglia E, del Rio E, Matias-Guiu X, et al: PTEN mutations in endometrial carcinomas: A molecular and clinicopathologic analysis of 38 cases. HUM PATHOL 31:312-317, 2000 25. Ali IU, Schriml LM, Dean M: Mutational spectra of PTEN/ MMAC1 gene: A tumor suppressor with lipid phosphatase activity. J Natl Cancer Inst 91:1922-1932, 1999 26. Risinger JI, Hayes K, Maxwell GL, et al: PTEN mutation in endometrial cancers is associated with favorable clinical and pathologic characteristics. Clin Cancer Res 4:3005-3010, 1998 27. Levine RL, Cargile CB, Blazes MS, et al: PTEN mutations and microsatellite instability in complex atypical hyperplasia, a precursor lesion to uterine endometrioid carcinoma. Cancer Res 58:3254-3258, 1998 28. Obata K, Morland SJ, Watson RH, et al: Frequent PTEN/ MMAC mutations in endometrioid but not serous or mucinous epithelial ovarian tumors. Cancer Res 58:2095-2097, 1998 29. Franklin WA, Gazdar AF, Haney J, et al: Widely dispersed p53 mutation in respiratory epithelium. A novel mechanism for field carcinogenesis. J Clin Invest 100:2133-2137, 1997 30. Barrett MT, Sanchez CA, Prevo LJ, et al: Evolution of neoplastic cell lineages in Barrett oesophagus. Nat Genet 22:106-109, 1999 31. Risinger JI, Berchuck A, Kohler MF, et al: Genetic instability of microsatellites in endometrial carcinoma. Cancer Res 53:51005103, 1993 32. Sood AK, Holmes R, Hendrix MJ, et al: Application of the National Cancer Institute International Criteria for Determination of Microsatellite Instability in Ovarian Cancer. Cancer Res 61:43714374, 2001 33. Mutter GL, Boynton KA, Faquin WC, et al: Allelotype mapping of unstable microsatellites establishes direct lineage continuity between endometrial precancers and cancer. Cancer Res 56:44834486, 1996 34. Esteller M, Catasus L, Matias-Guiu X, et al: hMLH1 promoter hypermethylation is an early event in human endometrial tumorigenesis. Am J Pathol 155:1767-1772, 1999 35. Jiang X, Morland SJ, Hitchcock A, et al: Allelotyping of endometriosis with adjacent ovarian carcinoma reveals evidence of a common lineage. Cancer Res 58:1707-1712, 1998 36. Sato N, Tsunoda H, Nishida M, et al: Loss of heterozygosity on 10q23.3 and mutation of the tumor suppressor gene PTEN in benign endometrial cyst of the ovary: Possible sequence progression from benign endometrial cyst to endometrioid carcinoma and clear cell carcinoma of the ovary. Cancer Res 60:7052-7056, 2000

428