Chromosomal instability in fallopian tube precursor lesions of serous carcinoma and frequent monoclonality of synchronous ovarian and fallopian tube mucosal serous carcinoma

Available online at www.sciencedirect.com

Gynecologic Oncology 110 (2008) 408 – 417 www.elsevier.com/locate/ygyno

Chromosomal instability in fallopian tube precursor lesions of serous carcinoma and frequent monoclonality of synchronous ovarian and fallopian tube mucosal serous carcinoma ☆ Shannon Salvador a,⁎,1 , Allan Rempel b,1 , Robert A. Soslow c , Blake Gilks b , David Huntsman b,d , Dianne Miller e a

d

Department of Obstetrics and Gynecology, University of British Columbia, Vancouver, BC, Canada b Department of Pathology, University of British Columbia, Vancouver, BC, Canada c Department of Pathology, Memorial Sloan-Kettering Cancer Center, New York, NY 10021, USA Genetic Pathology Evaluation Centre of the Prostate Centre and Department of Pathology of Vancouver Coastal Health Research Institute, British Columbia Cancer Agency, and University of British Columbia, Vancouver, BC, Canada e Department of Gynecology Oncology, Vancouver Coastal Health Authority, British Columbia Cancer Agency, BC, Canada Received 3 April 2008 Available online 1 July 2008

Abstract Objectives. Pelvic serous carcinomas are classified according to the location of greatest mass of tumor as ovarian, peritoneal or fallopian tube. Recent studies suggest these cancers may arise in the fallopian tube. This study explores the relationship between ovarian cancers and fallopian tube mucosal involvement. Methods. Sixteen consecutive cases of epithelial ovarian malignancy were prospectively identified and the fallopian tubes submitted in toto for histopathological examination for tubal mucosal involvement. Immunohistochemical staining for p53 and Ki-67, and fluorescent in situ hybridization (FISH) analysis for chromosomal copy number changes were performed on 10 cases. Three cases of mucosal epithelial abnormalities identified in riskreducing salpingectomy specimens were similarly characterized. Results. Of sixteen cases, twelve were high-grade serous carcinoma, stage III, and four cases were stage I, two borderline mucinous, one borderline serous, and one low-grade mucinous carcinoma. Ten cases of high-grade serous carcinoma showed either unilateral fallopian tube mucosal involvement (n = 7) or tubal obliteration ipsilateral to the dominant ovarian mass (n = 3), compared to none of the other carcinomas. FISH analysis showed similar copy number changes in the ovarian and fallopian tube mucosal carcinoma in 3 cases, suggesting a unifocal origin; one case had differences suggesting multifocal origin of cancer. One case had equivocal FISH results. From risk-reducing salpingectomy cases, the multiple foci of tubal intraepithelial carcinoma and focus of invasive carcinoma showed similar gene copy number changes within each case, suggesting monclonality. Both cases of epithelial atypia/dysplasia showed gene copy number changes. Conclusions. Fallopian tube mucosal and ovarian tumors have similar genetic abnormalities in most cases, indicating a monoclonal origin that may originate either from the ovary, peritoneum or fallopian tube. In situ epithelial lesions of the fallopian tube from risk-reducing salpingectomies show gene copy abnormalities consistent with these being early lesions of serous carcinoma and suggest that chromosomal instability is a very early event in serous carcinogenesis. © 2008 Elsevier Inc. All rights reserved. Keywords: Pelvic serous carcinoma; Fallopian tube carcinoma; Origin of disease; Ovarian carcinoma

☆

This abstract was presented at the 15th International Meeting of the European Society of Gynaecological Oncology, Oct 29, 2007. ⁎ Corresponding author. Room 2H30, BC Women's Hospital, 4490 Oak Street, Vancouver, BC, V6H 3V5. Fax: +1 604 875 2725. E-mail address: [email protected] (S. Salvador). 1 These authors contributed equally to this work. 0090-8258/$ - see front matter. © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.ygyno.2008.05.010

Introduction Primary fallopian tube carcinoma is believed to be a rare malignancy that occurs with an annual incidence of 3.6 per million women [1]. Most of these carcinomas have serous histology [2]. Due to the difficulty in distinguishing between primary

S. Salvador et al. / Gynecologic Oncology 110 (2008) 408–417

409

Table 1 Tubal involvement in ovarian cancer cases studied Unilateral tubal mucosal involvement

Unilateral obliterated tube

Bilateral tubal serosal involvement only

No tubal involvement

Number

7

3

2

4

Histology High-grade serous Borderline mucinous Borderline serous Low-grade mucinous

7 0 0 0

3 0 0 0

2 0 0 0

0 1 2 1

tubal carcinoma and advanced ovarian carcinoma, the incidence is probably higher. Currently the diagnosis of primary fallopian tube carcinoma is based on criteria set forth by Hu et al. [3], and modified by Sedlis [4] and Yoonessi [5]. This requires that the main tumor mass be present in the fallopian tube and arise from the endosalpinx. The histologic features of the tumor must reflect a tubal pattern and the transition from malignant to benign epithelium should be detectable. Finally, the tube must contain more tumor than the ovary or endometrium. These criteria are still used today with recent guidelines published by FIGO and IGCS [6]. With the difficulty of pathologically separating fallopian tube carcinoma from serous ovarian adenocarcinoma, some primary fallopian tube carcinomas are likely being misdiagnosed as ovarian or peritoneal in origin. The possibility of the fallopian tube as the primary source for these cancers was noted by Alban Doran in 1884, in which he states, “I must observe that it is possible that some of the papillomatous masses that infest the surface of the ovary and broad ligament may originate from the fallopian tube” [7]. Recent studies have shown that a higher incidence of fallopian tube carcinoma may occur than previously thought, when more extensive examination of the fallopian tube is undertaken [8,9]. These studies have shown a 10 fold increase in the number of cases of fallopian tube carcinoma diagnosed, compared to the expected number, even adhering to the strict definition previously discussed. Women with germline BRCA mutations who undergo prophylactic salpingo-oophorectomy have occult carcinoma in 4–17% of cases, and tubal involvement in 42–100% of these cancers [10–12]. A recent study by Piek et al. showed that 50% of prophylactically removed fallopian tubes had dysplastic areas with increased proliferation and atypia [13]. This study reports on results of extensive pathological examination of the fallopian tubes from 16 consecutive cases of ovarian cancer to look for evidence that the fallopian tube may have been the primary source of the cancer. Three cases in which risk-reducing salpingectomies were performed in women at high risk for breast and ovarian cancer, as well as five normal control cases, were also characterized, with immunostaining for p53 and Ki-67, and fluorescence in situ hybridization (FISH) to identify chromosomal copy number changes. Materials and methods This study was approved by the Clinical Research Ethics Board of the University of British Columbia. Sixteen consecutive cases of ovarian surface epithelial carcinoma or borderline tumor (tumor of low malignant potential) were collected prospectively in the Department of Pathology, Vancouver

General Hospital (VGH). The fallopian tubes were submitted in toto in each case, and slides were reviewed for evidence of tubal mucosal involvement. Three risk-reducing salpingo-oophorectomy specimens showing epithelial atypia/dysplasia (EAD) (2 cases) or tubal intraepithelial carcinoma (TIC) (1 case) were identified in the pathology archives of Memorial Sloan-Kettering Cancer Center. Areas of TIC were identified by criteria used by previous authors that included the loss of cell polarity, elevated nucleus to cell ratio, prominent nucleoli and absence of ciliated cells whereas areas of EAD had p53 positivity with nuclear enlargement and prominent nucleoli but maintained cell polarity [14]. The 5 control cases – fallopian tubes removed for non-cancer-related medical reasons from women at low risk for developing ovarian cancer – were selected from the archives of the Department of Pathology, VGH.

Section preparation The fallopian tube was submitted in toto and serially sectioned every 3–4mm. Sections from each formalin-fixed, paraffin-embedded tissue block selected for study were mounted on Fisherbrand Superfrost Plus slides (Thermo Fisher Scientific). Four micrometer thick sections for H&E and immunostaining were baked overnight at 37°C. Six micrometer thick sections for FISH were baked overnight at 37°C and then overnight again at 60°C.

Immunomarker staining Following deparaffinization and hydration, slides were immunostained for p53 using anti-p53 (clone D07, dilution 1:400 with heat antigen retrieval) antibody or Ki-67 using anti-Ki-67 (clone SP6, dilution 1:200 with heat antigen retrieval) Table 2 Ovarian carcinoma, immunochemistry Case code

Ovarian tumor

Fallopian tube

p53◇

Ki67^

V1

High-grade serous

V2

High-grade serous

V3

High-grade serous

V4

High-grade serous

V5

High-grade serous

V6

High-grade serous

V7

High-grade serous

V8 V9

Serous borderline Mucinous borderline

Serosal carcinoma Mucosal carcinoma Normal Mucosal carcinoma Normal Mucosal carcinoma Normal Mucosal carcinoma Normal Mucosal carcinoma Normal Serosal carcinoma Normal Serosal carcinoma Normal Normal Normal

Low Low Low High Low Low Low High Low High Low High Low High Low Low Low

Intermediate Intermediate Low High Low High Low Intermediate Low High Low High Low High Low Low Intermediate

Per 500 cells ◇p53 ^Ki-67

low ≤50% reactive. high N50% reactive. low ≤ 5% reactive. intermediate N5 and ≤ 50% reactive. high N50% reactive.

410

S. Salvador et al. / Gynecologic Oncology 110 (2008) 408–417

antibody (both from Ventana Medical Systems Inc., Tuscon, AZ, USA), and a Ventana Discovery XT immunostainer as previously described [15]. Antigen retrieval and indirect antibody detection were performed following manufacturer's directions using Ventana CC-1 pretreatment reagent and DABMap™ detection kit, respectively.

Immunostaining Interpretation Each 3–4mm section had a single p53 and Ki-67 slide prepared and examined. Scoring of the immunostains was done at X400 in 10 randomly

chosen areas in each of the lesions. The number of positive cells for 50 cells scored in each area was recorded and then summed. The p53 and Ki-67 staining was assessed as follows: p53 was low if 0–50% of cells per 500 cells counted were immunoreactive and high if N 50% were immunoreactive. P53 staining is bimodal, as previously decribed [16], with a cutoff of 50% separating the p53 “high” and p53 “low” populations. In ovarian carcinoma, p53 expression correlates with missense mutations in p53 [17]. Ki-67 had low reactivity if 0–5% of cells per 500 cells counted were immunoreactive, intermediate if 5–50% were immunoreactive and high if N 50% were immunoreactive (Table 2). Immunoreactivity of benign-appearing (normal) tubal

Fig. 1. a. Ovarian cancer with fallopian tube mucosal involvement: percent of cells showing polysomy (≥5 FISH signals/nucleus). Cases V1, V2, and V4 show monoclonality between the ovarian carcinoma and the fallopian tube mucosal carcinoma. b. Ovarian cancer with no fallopian tube mucosal involvement: percent of cells showing polysomy (≥5 FISH signals/nucleus). Case V6 shows polysomy in the ovarian carcinoma that was not present in the fallopian tube serosal implants. V7 show monoclonality of the tumors.

S. Salvador et al. / Gynecologic Oncology 110 (2008) 408–417 epithelium on the same sections, but away from the tubal lesions, was also assessed.

FISH The Vysis Breast Aneusomy Probe Set (Vysis Inc., Downer Grove, IL, USA) was used to assay for chromosome aneusomy. The four probes of the set are directed towards complementary chromosome-specific sequences: 1p12 (LSI 1) on chromosome 1 and pericentric repetitive sequences of chromosomes 8, 11, and 17. In studies previous to this one [18],these chromosomes had been shown to have aberrant numbers in breast carcinomas and was used in this study given the link between breast and ovarian cancer in BRCA mutation positive high-risk women. The probes were directly labeled with SpectrumGold, SpectrumRed, SpectrumGreen, and SpectrumAqua, respectively. Deparaffinized and hydrated sections were pretreated in 40mL solution volumes at room temperature or at indicated temperature, with no more than four slides per coplin jar. The order of pretreatment was: 0.2N HCl for 20 min; deionized water for 10 min; 2× SSC for 3 min; 1M NaSCN for 60 min at 80°C; and 0.01M HCl/pepsin (36,000U) for 10 min at 37°C. Pretreated sections were washed at room temperature for 30s in 2X SSC and then 30s in deionized water, followed by dehydration in 2 changes of absolute alcohol, 1 min each, and air-drying. Ten microlitres of probe mixture was applied to each section, then covered with a 22 × 40 coverslip (VWR International, West Chester, PA, USA), sealed with rubber cement and placed in a HYBright Denaturation/Hybridization System (Vysis Inc.). Denaturation was run at 73°C for 5 min followed by hybridization at 37°C for 16 h. Posthybridization washing was done in 40mL volumes of 2× SSC/3% NP-40: first for 5 min at room temperature followed by fresh post-hybridization solution at 72°C for 2 min. Sections were briefly washed at room temperature in 2X SSC, then dehydrated in 70% and 80% ethanol (2 min each) and 2 changes of absolute alcohol (1 min each). Slides were placed in the dark to air-dry. Ten microliters of DAPI I (4,6-diamidino-2-phenylindole; Vysis, Inc.) were applied to each air-dried section and then coverslipped. Stained FISH slides were stored at 4°C in the dark until scoring and warmed to room temperature prior to assessment.

411

FISH interpretation Tubal epithelia cases were scored to determine the percentage of epithelial cells aneusomic for chromosomes 1, 8, 11, and 17. DAPI-stained nuclei of dysplastic tubal epithelial cells and tubal intraepithelial carcinomas cells were easily distinguishable from other normal epithelial and non-epithelial cell types. For this study, a polysomic cell is defined as having ≥ 5 of any of the 4 chromosomes being enumerated, while a monosomic cell is defined as having ≤ 1 of any of the chromosomes being enumerated. Monosomy for any of the 4 loci implies loss of heterozygosity (LOH), however, as this is only one of several mechanisms through which LOH can be acquired this is an insensitive assay for LOH and thus that terminology was not used. Forty cells were assessed for each score and the percentage of cells aneusomic for each chromosome determined (Figs. 1a, b, 2, 4a and b). Although the finding of aneusomy (polysomy or monosomy at any locus) indicates chromosomal instability, the sensitivity of this assay is likely considerably less than 100% as only 4 loci were assessed.

Results Of the 16 prospectively collected cases of ovarian surface epithelial carcinoma or borderline tumor, twelve cases were highgrade serous carcinoma, stage III, with bilateral ovarian involvement. The remaining four cases were two borderline mucinous, one borderline serous, and one low-grade mucinous carcinoma; all were stage I. Of the high-grade serous adenocarcinoma cases, seven were found to have mucosal involvement of one fallopian tube by TIC ipsilateral to the dominant ovarian mass. Three of the cases had one obliterated fallopian tube, again ipsilateral to the dominant ovarian mass in each case. The fallopian tubes on the contralateral side showed no evidence of mucosal involvement in any of these cases but they did have bilateral tubal serosal involvement. Two cases of high-grade

Fig. 2. Ovarian carcinomas with and without fallopian tube mucosal involvement: percent of cells showing monosomy (≤ 1 FISH signal/nucleus). Case V6 show 100% monosomy in chromosome 17 in the serosal implants that was not present in the ovarian carcinoma. The remaining cases displayed no monosomy relative to normal FT controls.

412

S. Salvador et al. / Gynecologic Oncology 110 (2008) 408–417

serous carcinoma had only serosal involvement of both fallopian tubes. The four cases that were not high-grade serous carcinoma had no tubal involvement (Table 1). Nine representative cases were selected from the above 16 cases in which to perform further studies and these results are shown in Table 2. Five were high-grade serous carcinomas with mucosal involvement (TIC). The remaining four cases included two high-grade serous cases with fallopian tube serosal involvement but no mucosal involvement, one case of serous borderline tumor and one case of mucinous borderline tumor, neither of which had tubal involvement. Five of seven cases of high-grade serous carcinoma showed diffuse, strong p53 immunoreactivity, and all cases showed intermediate (n = 2) or high (n = 5) Ki-67 immunoreactivity. The benign-appearing mucosa, whether adjacent to the intratubal lesions, or away from such lesions, was uniformly negative for p53, while Ki-67 immunoreactivity was low in all but one case, the mucinous borderline tumor, where the fallopian tube epithelium showed intermediate immunoreactivity for Ki-67, with 5% of nuclei staining positively. Figs. 1 and 2 show the FISH results in these cases. Fig. 1 illustrates the frequency of cells showing polysomy; the normal fallopian tube controls show no cells with polysomy, as expected, as even in a dividing cell the maximum number of signals per cell is four. The cytologically normal fallopian tube epithelium from the cancer cases also shows negligible levels of polysomy. Rare cells with 5 or more chromosomes per nucleus may reflect overlapping nuclei in tissue sections, which are not

easily resolved by fluorescence microscopy. In contrast, the ovarian and fallopian tube tumors showed abnormalities in most cases. For the five cases with high-grade serous carcinoma of both the ovary and fallopian tube mucosa, cases V1, V2 and V4 show similar abnormalities in copy number at each locus examined. In case V3, the ovarian carcinoma does not show polysomy while the fallopian tumor does at all four loci tested. Finally, in case V5, the ovarian carcinoma is polysomic for chromosome 1 but the fallopian tube carcinoma is not while both sites show similar abnormalities in chromosomes 8 and 11. In summary: the gene copy number differences are similar in synchronous ovarian and fallopian tube tumors in three cases, while they are different in one case and indeterminate in another case. Fig. 1b shows the tumors without fallopian tube mucosal involvement. In the high-grade serous carcinoma cases, one shows similar gene copy number abnormalities in ovary and tubal serosal implants (case V7) and one shows differences between the ovarian carcinoma and fallopian tube serosal tumor (case V6). The fallopian tube from the two cases of borderline tumor (cases V8 and V9) shows no evidence of polysomy. Fig. 2 shows the percentage of cells with a single FISH signal per cell (monosomy) for the four probes, from four representative cases of these nine studied cases. As expected, occasional cells show a single signal in both the control cases and normal fallopian tube from the study cases, reflecting sectioning through the nuclei in the tissue sections, with one allele not appearing on the slide. The only case with significant monosomy is case V6, a high-grade

Fig. 3. Representative slides of case V1, tubal mucosal involvement seen in a. with tubal serosal (TS) and tubal mucosal (arrow) involvement, b. tubal mucosal involvement (arrow), c. tubal intraepithelial carcinoma, d. p53 immunostaining, e. Ki-67 immunostaining, f. abnormal aneusomic FISH results in a single cell with aneusomic chromosome 1 (gold), 8 (red), and 11 (green).

S. Salvador et al. / Gynecologic Oncology 110 (2008) 408–417

serous carcinoma without fallopian tube mucosal involvement, which shows monosomy for the chromosome 17 probe in the serosal implant but not in the ovarian carcinoma. Fig. 3 shows representative images of p53 and Ki-67 immunostaining and FISH staining from one of the above cases with tubal mucosal involvement. Of the three risk-reducing salpingectomy cases, two had EAD and one had serous TIC of the fallopian tubes. This latter case had bilateral tubal mucosal involvement with a b 1.0cm focus of invasive carcinoma of fimbriated end of the left fallopian tube and cytologically malignant groups of cells within the lumen of the right fallopian tube (Fig. 4). Results of p53 and Ki-67

413

staining on these three cases are shown in Table 3, with representative staining shown in Fig. 4. The TIC in case 10 stained highly positive for p53, with intermediate to high Ki-67 staining. The two cases of EAD (cases V11 and V12) showed high p53 positivity, with low or intermediate Ki-67 immunoreactivity. Fig. 5 shows the results of FISH analysis in these 3 cases. In case 10, there is no polysomy, but the focus of invasive carcinoma, TIC, and intraluminal clusters of tumor cells all show chromosome 17 monosomy. The EAD cases show polysomy for 3 out of the 4 chromosomes tested for case V11 and 2 out of the 4 for case V12. As expected, given the relatively low Ki-67 indices in these cases of EAD, the number of cells with 5 or more signals

Fig. 4. Representative slides of case V10 a. left tubal intraepithelial carcinoma with invasion (inv) and two areas of TIC: TIC1 (✸) and TIC2 (♦), b. p53 staining, ( ) indicates abrupt boundary between p53 positive lesion and benign-appearing p53 negative tubal epithelium, c. Ki-67 staining, d. invasive carcinoma from case V10, e. high power of area indicated by ( ) in b showing abrupt change from normal appearing tubal epithelium to TIC, f. high power of TIC 2 in case V10, g. p53 staining of area of tumor cells within contralateral tubal lumen, h. high power showing case V10 with cytologically malignant groups of cells within the lumen of the contralateral fallopian tube.

414

S. Salvador et al. / Gynecologic Oncology 110 (2008) 408–417

Table 3 Lesions from BRCA mutation carriers, immunochemistry Case code

Diagnosis

V10L

Left tubal intraepithelial carcinoma (TIC)

V10R

V11

V12

Fallopian tube epithelium

Invasive carcinoma TIC 1 TIC 2 Adjacent a Normal b Right tubal intraepithelial Intralumenal tumor carcinoma (TIC) TIC Adjacent Normal Epithelial EAD atypia/dysplasia (EAD) Adjacent Normal Epithelial EAD atypia/dysplasia (EAD) Adjacent Normal

p53◇ Ki67^ High High High Low Low High High Low Low High Low Low High Low Low

High Intermediate Intermediate Low Low Intermediate Intermediate Low Low Low Low Low Intermediate Low Low

Per 500 cells ◇p53

low ≤50% reactive. high N50% reactive. ^Ki-67 low ≤ 5% reactive. intermediate N5 and ≤ 50% reactive. high N50% reactive. a adjacent = morphologically normal epithelium immediately adjacent to TIC/EAD. b normal = morphologically normal epithelium distant from TIC/EAD.

per nucleus (which presumably represent cells in undergoing cell division, in G2 M phase) is relatively low (less than 20% for most probes). Discussion Currently, pelvic serous carcinoma is classified according to the location of the bulk of the tumor as being ovarian, peritoneal, or fallopian tube in origin. There are no morphological or molecular differences between serous carcinoma considered to be primary at any of these three sites. In contrast, uterine serous carcinoma differs in being WT-1 negative in most cases [19] while serous carcinoma of the cervical origin is positive for HPV (Nucci, unpublished data). The most prominent theory about the histogenesis of pelvic serous carcinoma is that incessant ovulation results in recurrent damage and repair cycles that lead to increased risk of carcinoma [20]. Related to this is the hypothesis that gonadotropin excess stimulates the ovarian surface epithelium leading to malignant transformation in inclusion cysts [21]. Another theory of the origin of pelvic serous carcinoma involves remnants of the Müllerian ducts. The secondary Müllerian system is composed of areas lined by Müllerian epithelium outside of the cervix, uterus, and fallopian tubes and it has been suggested that such remnants of the Müllerian ducts may undergo malignant transformation [22]. More recently, the fallopian tube has been implicated as the primary site for pelvic serous carcinoma [8,23] and it has been demonstrated that gene expression between tubal epithelium and serous carcinoma resemble one another [24,25]. As ovulation can cause follicular fluid containing ovarian cells to flow from the ovary to the fallopian tube and potential lead to cell seeding on the tubes[26], during menstruation endometrial cells are carried in a retrograde fashion through the fallopian tubes

thereby retrograde flow could also result in the direct seeding of tubal carcinoma cells onto the ovary [27]. In this case series there was unilateral tubal mucosal involvement by serous carcinoma in of seven out of twelve cases of high-grade serous carcinoma (58%, p = 0.069). If the cases with unilateral obliterated tubes are included, then this increases to ten out of twelve (83%). All of these cases had bilateral serosal involvement of the tubes while the mucosal involvement of the fallopian tubes was unilateral and all occurred at the fimbriated end, a pattern of involement noted in other studies [11]. The tubal mucosal involvement was ipsilateral to the dominant ovarian mass in the presence of bilateral ovarian involvement in every case. One could speculate that if this tubal mucosal tumor was implantation from the ovary, both fallopian tubes would likely show mucosal involvement instead of only serosal implantation. These findings support a recent study by Kindelberger et al., 2007 who described 55 cases of pelvic serous carcinoma in which 41 (75%) had mucosal involvement of the fallopian tube [28]. This study shows that pelvic serous carcinoma may spread either from the dominant ovarian mass to the ipsilateral fallopian tube or conversely from the fallopian tube to the dominant ovarian mass and subsequently the remainder of the abdominal cavity. Only by examining early lesions, which are unfortunately rare, will we be able to definitively document the site of origin. It is possible that pelvic serous carcinoma could have multiple sites of origin. In all cases in this study, the ovaries were the largest tumor mass, leading to their classification as primary ovarian carcinoma. Other types of cancer that metastasize to the ovary creating large tumors show how the location of the largest tumor does not equate to the primary site. For example, a low-grade mucinous appendiceal neoplasm can result in bulky mucinous tumors of the ovaries [29]. This is also seen in cases of simultaneous adenocarcinomas of the cervix and the ovary in which the ovarian tumors were initially thought to represent independent primary ovarian surface epithelial tumors [30]. Through HPV typing of both tumors, it was determined that the vastly larger tumors in the ovaries were metastases from small endocervical adenocarcinomas. FISH analysis was used to assess gene copy number. The normal copy number per cell is between two to four, with the latter present in cells in G2 and M phase of mitotic division. Thus, there should never be normal cells with five or more copies of a nonrepetitive sequence per cell. The observation that only a minority of cells have five or more copies, for most cases where increased copy number was detected, indicates that there maybe only a low level of amplification, with increased copy number only detected in cells in G2 and M phase. In contrast, monosomy was seen in the large majority of cells; an expected result as monosomic cells in G0 and G1 phase (the majority of cells) will contain a single allelic copy per cell. There was likely some inter-nuclear variability between tissue sections using this assay, as only part of each nucleus will appear on a six micrometer thick tissue section. Three of the high-grade serous adenocarcinomas showed a probable monoclonal origin of carcinoma involving the ovary and the fallopian tube, and one case had a possible multiclonal origin. Monoclonality implies that there is one primary site with secondary spread to the other pelvic and abdominal organs. This has

S. Salvador et al. / Gynecologic Oncology 110 (2008) 408–417

been shown in previous studies of ovarian carcinoma [28,31]. Lee et al. [32] studied TICs with their associated ovarian cancers and found they shared the same p53 mutation suggesting monoclonality. A large majority of cases of high-grade synchronous and

415

morphologically similar tumors at different sites are examples of a single primary with metastasis, such as serous carcinoma of the endometrium with extrauterine disease [33]. Multiclonal origin indicates that synchronous primaries have arisen at separate sites,

Fig. 5. a. TIC and epithelial dysplasia identified in risk-reducing salpingectomy specimens: percent of cells showing polysomy (≥5 FISH signals/nucleus). Case V10 has no polysomy while V11 has three out of four chromosomes with polysomy and V12 has two out of four. Normal FT controls not shown. b. TIC and epithelial dysplasia identified in risk-reducing salpingectomy specimens: percent of cells showing monosomy (≤1 FISH signal/nucleus). Case V10 shows monosomey in both the left and right fallopian tube while V11 and 12 have no monosomy. Normal FT controls not shown.

416

S. Salvador et al. / Gynecologic Oncology 110 (2008) 408–417

in this case the fallopian tube and the ovarian surface. This has been seen in the study of cases that are believed to be papillary serous carcinoma of the peritoneum [34,35]. All of the tubal lesions showed some degree of aneusomy. This data, particularly the finding of chromosomal abnormalities in the EAD and TIC lesions identified in prophylactic specimens, suggest that the acquisition of chromosomal instability is a very early event in the development of high-grade serous cancers of the fallopian tube and ovary. The acquisition of chromosomal instability as an early or even an initiating event in the oncogenesis of these lesions is similar to that described in ductal carcinoma in situ of the breast [36]. In women with germline BRCA mutations this chromosomal instability could be attributed to loss of BRCA function with acquired p53 abnormalities shielding chromosomally unstable cells from apoptosis. Although somatic abnormalities of the BRCA genes, including somatic mutations, BRCA1 promoter hypermethylation and loss of heterozygosity, have been found in the majority of sporadic high-grade serous cancers of the ovary [37,38], the role of such events in sporadic fallopian TIC lesions is not yet known. Careful examination of prohylactically removed fallopian tubes and ovaries from BRCA mutation carriers reveals a high incidence of fallopian tube mucosal atypia and occult cancers [10–13,39]. In contrast to familial ovarian cancer, precursor or early lesions of sporadic ovarian cancers are poorly understood. Early stage high-grade serous carcinoma, confined to the ovary and without surface involvement, is very rare in patients who are carefully staged [40]. The population of BRCA mutation carriers, who are at increased risk of developing high-grade serous carcinomas, may be representative of the general population with respect to the primary location of pelvic serous cancers. It is possible that the primary tumor in the fallopian tube has been consistently missed as pelvic serous cancers tend to present with tumor spread throughout the peritoneal cavity. The potential of the fallopian tube as a primary source of these cancers has been advanced by Crum and colleagues in several recent publications [11,14,23,28,32,41]. They have suggested that cancers currently classified as ovarian, tubal, or primary peritoneal serous carcinomas should be pelvic serous carcinomas [41]. We agree with this suggestion; while the early events during oncogenesis of pelvic serous cancers will become better understood in the coming years, there is significant evidence that the tubal mucosa is the primary site in many cases. This has implication for early detection and prevention as reviewed by Salvador et al. [42]. Conflict of interest statement The authors declare that there are no conflicts of interest.

References [1] Rosenblatt KA, Weiss NS, Schwartz SM. Incidence of malignant fallopian tube tumors. Gynecol Oncol 1989;35(2):236–9. [2] Alvarado-Cabrero I, Young RH, Vamvakas EC, Scully RE. Carcinoma of the fallopian tube: a clinicopathological study of 105 cases with observations on staging and prognostic factors. Gynecol Oncol 1999;72(3):367–79. [3] Hu CY, Taylor ML, Hertig A. Primary carcinoma of the fallopian tube. Am J Obstet Gynecol 1950;59(1):58–67.

[4] Sedlis A. Primary carcinoma of the fallopian tube. Obstet Gynecol Surv 1961;16:209–26. [5] Yoonessi M. Carcinoma of the fallopian tube. Obstet Gynecol Surv 1979;34(4):257–70. [6] Benedet JL, Bender H, Jones III H, Ngan HY, Pecorelli S. FIGO staging classifications and clinical practice guidelines in the management of gynecologic cancers. FIGO Committee on Gynecologic Oncology. Int J Gynaecol Obstet 2000;70(2):209–62. [7] Doran A. Tumours of the ovary, fallopian tube, and broad ligament. London: Smith, Elder, & Co.; 1884. [8] Woolas R, Jacobs I, Davies AP, Leake J, Brown C, Grudzinskas JG, et al. What is the true incidence of primary fallopian tube carcinoma? Int J Gynecol Cancer 1994;4(6):384–8. [9] Clayton NL, Jaaback KS, Hirschowitz L. Primary fallopian tube carcinoma — the experience of a UK cancer centre and a review of the literature. J Obstet Gynaecol 2005;25(7):694–702. [10] Powell CB, Kenley E, Chen LM, Crawford B, McLennan J, Zaloudek C, et al. Risk-reducing salpingo-oophorectomy in BRCA mutation carriers: role of serial sectioning in the detection of occult malignancy. J Clin Oncol 2005;23(1): 127–32. [11] Medeiros F, Muto MG, Lee Y, Elvin JA, Callahan MJ, Feltmate C, et al. The tubal fimbria is a preferred site for early adenocarcinoma in women with familial ovarian cancer syndrome. Am J Surg Pathol 2006;30(2):230–6. [12] Finch A, Shaw P, Rosen B, Murphy J, Narod SA, Colgan TJ. Clinical and pathologic findings of prophylactic salpingo-oophorectomies in 159 BRCA1 and BRCA2 carriers. Gynecol Oncol 2006;100(1):58–64. [13] Piek JM, van Diest PJ, Zweemer RP, Jansen JW, Poort-Keesom RJ, Menko FH, et al. Dysplastic changes in prophylactically removed Fallopian tubes of women predisposed to developing ovarian cancer. J Pathol 2001;195(4): 451–6. [14] Jarboe EMD, Folkins A, Nucci MR, Kindelberger D, Drapkin R, Miron A, et al. Serous carcinogenesis in the fallopian tube: a descriptive classification. Int J Gynecol Pathol 2008;27(1):1–9. [15] Alkushi A, Clarke BA, Akbari M, Makkretsov N, Lim P, Miller D, et al. Identification of prognostically relevant and reproducible subsets of endometrial adenocarcinoma based on clustering analysis of immunostaining data. Mod Pathol 2007;20(11):1156–65. [16] Alkushi A, Lim P, Coldman A, Huntsman D, Miller D, Gilks B. Interpretation of p53 immunoreactivity in endometrial carcinoma: establishing a clinically relevant cut-off level. Int J Gynecol Pathol 2004;23(2):129–37. [17] Havrilesky L, Darcy M, Hamdan H, Priore RL, Leon J, Bell J, et al. Prognostic significance of p53 mutation and p53 overexpression in advanced epithelial ovarian cancer: a Gynecologic Oncology Group Study. J Clin Oncol 2003;21(20):3814–25. [18] Sneige N, Liu B, Yin G, Gong Y, Arun BK. Correlation of cytologic findings and chromosomal instability detected by fluorescence in situ hybridization in breast fine-needle aspiration specimens from women at high risk for breast cancer. Mod Pathol 2006;19(5):622–9. [19] Al-Hussaini M, Stockman A, Foster H, McCluggage WG. WT-1 assists in distinguishing ovarian from uterine serous carcinoma and in distinguishing between serous and endometrioid ovarian carcinoma. Histopathology 2004;44(2):109–15. [20] Fleming JS, Beaugie CR, Haviv I, Chenevix-Trench G, Tan OL. Incessant ovulation, inflammation and epithelial ovarian carcinogenesis: revisiting old hypotheses. Mol Cell Endocrinol 2006;247(1–2):4–21. [21] Cramer DW, Welch WR. Determinants of ovarian cancer risk. II. Inferences regarding pathogenesis. J Natl Cancer Inst 1983;71(4):717–21. [22] Dubeau L. The cell of origin of ovarian epithelial tumors and the ovarian surface epithelium dogma: does the emperor have no clothes? Gynecol Oncol 1999;72(3):437–42. [23] Crum CP, Drapkin R, Miron A, Ince TA, Muto M, Kindelberger DW, et al. The distal fallopian tube: a new model for pelvic serous carcinogenesis. Curr Opin Obstet Gynecol 2007;19(1):3–9. [24] Bowen NJ, Logani S, Dickerson EB, Kapa LB, Akhtar M, Benigno BB, et al. Emerging roles for PAX8 in ovarian cancer and endosalpingeal development. Gynecol Oncol 2007;104(2):331–7. [25] Marquez RT, Baggerly KA, Patterson AP, Liu J, Broaddus R, Frumovitz M, et al. Patterns of gene expression in different histotypes of epithelial

S. Salvador et al. / Gynecologic Oncology 110 (2008) 408–417

[26] [27]

[28]

[29]

[30]

[31]

[32]

[33]

ovarian cancer correlate with those in normal fallopian tube, endometrium, and colon. Clin Cancer Res 2005;11(17):6116–26. Jansen RP. Endocrine response in the fallopian tube. Endocr Rev 1984;5(4): 525–51. Piek JM, Kenemans P, Verheijen RH. Intraperitoneal serous adenocarcinoma: a critical appraisal of three hypotheses on its cause. Am J Obstet Gynecol 2004;191(3):718–32. Kindelberger DW, Lee Y, Miron A, Hirsch MS, Feltmate C, Medeiros F, et al. Intraepithelial carcinoma of the fimbria and pelvic serous carcinoma: evidence for a causal relationship. Am J Surg Pathol 2007;31(2):161–9. Young RH, Gilks CB, Scully RE. Mucinous tumors of the appendix associated with mucinous tumors of the ovary and pseudomyxoma peritonei. A clinicopathological analysis of 22 cases supporting an origin in the appendix. Am J Surg Pathol 1991;15(5):415–29. Elishaev E, Gilks CB, Miller D, Srodon M, Kurman RJ, Ronnett BM. Synchronous and metachronous endocervical and ovarian neoplasms: evidence supporting interpretation of the ovarian neoplasms as metastatic endocervical adenocarcinomas simulating primary ovarian surface epithelial neoplasms. Am J Surg Pathol 2005;29(3):281–94. Mok CH, Tsao SW, Knapp RC, Fishbaugh PM, Lau CC. Unifocal origin of advanced human epithelial ovarian cancers. Cancer Res 1992;52(18): 5119–22. Lee Y, Miron A, Drapkin R, Nucci MR, Medeiros F, Saleemuddin A, et al. A candidate precursor to serous carcinoma that originates in the distal fallopian tube. J Pathol 2007;211(1):26–35. Baergen RN, Warren CD, Isacson C, Ellenson LH. Early uterine serous carcinoma: clonal origin of extrauterine disease. Int J Gynecol Pathol 2001;20(3):214–9.

417

[34] Muto MG, Welch WR, Mok SC, Bandera CA, Fishbaugh PM, Tsao SW, et al. Evidence for a multifocal origin of papillary serous carcinoma of the peritoneum. Cancer Res 1995;55(3):490–2. [35] Nishimura M, Wakabayashi M, Hashimoto T, Shibata J, Ueyama H, Ebina M, et al. Papillary serous carcinoma of the peritoneum: analysis of clonality of peritoneal tumors. J Gastroenterol 2000;35(7):540–7. [36] Chin K, de Solorzano CO, Knowles D, Jones A, Chou W, Rodriguez EG, et al. In situ analyses of genome instability in breast cancer. Nat Genet 2004;36(9):984–8. [37] Press JZ, DeLuca A, Boyd N, Young S, Troussard A, Ridge Y, et al. Ovarian carcinomas with genetic and epigenetic BRCA1 loss have distinct molecular abnormalities. Int J Gynaecol Cancer in press. [38] Hilton JL, Geisler JP, Rathe JA, Hattermann-Zogg MA, DeYoung B, Buller RE. Inactivation of BRCA1 and BRCA2 in ovarian cancer. J Natl Cancer Inst 2002;94(18):1396–406. [39] Piek JM, Torrenga B, Hermsen B, Verheijen RH, Zweemer RP, Gille JJ, et al. Histopathological characteristics of BRCA1- and BRCA2-associated intraperitoneal cancer: a clinic-based study. Fam Cancer 2003;2(2): 73–8. [40] Leitao Jr MM, Boyd J, Hummer A, Olvera N, Arroyo CD, Venkatraman E, et al. Clinicopathologic analysis of early-stage sporadic ovarian carcinoma. Am J Surg Pathol 2004;28(2):147–59. [41] Crum CP, Drapkin R, Kindelberger D, Medeiros F, Miron A, Lee Y. Lessons from BRCA: the tubal fimbria emerges as an origin for pelvic serous cancer. Clin Med Res 2007;5(1):35–44. [42] Salvador S, Gilks B, Köbel M, Huntsman D, Rosen B, Miller D. The Fallopian Tube: Primary Site of Most Pelvic High-Grade Serous Carcinomas. Int J Gynaecol Cancer in press.