ORIGINAL ARTICLE
Locus/Chromosome Aberrations in Intraductal Papillary Mucinous Neoplasms Analyzed by Fluorescence In Situ Hybridization Katsuyuki Miyabe, MD, PhD,*w Yasuki Hori, MD,*w Takahiro Nakazawa, MD, PhD,* Kazuki Hayashi, MD, PhD,* Itaru Naitoh, MD, PhD,* Shuya Shimizu, MD,* Hiromu Kondo, MD,* Yuji Nishi, MD,* Michihiro Yoshida, MD, PhD,* Shuichiro Umemura, MD,* Akihisa Kato, MD,* Hirotaka Ohara, MD, PhD,z Takashi Joh, MD, PhD,* and Hiroshi Inagaki, MD, PhDw
Abstract: Locus and chromosome abnormalities have not been well clarified in intraductal papillary mucinous neoplasms (IPMNs). The aim of this study was to retrospectively examine these abnormalities using fluorescence in situ hybridization. IPMNs (n = 28) were histopathologically classified into noninvasive IPMN (n = 17) and IPMN with an associated invasive carcinoma (invasive IPMN, n = 11) groups. Noninvasive IPMNs possessed non-neoplastic and noninvasive spots in their tissues, and invasive IPMN cases possessed non-neoplastic, noninvasive, and invasive spots. Non-neoplastic (n = 28), noninvasive (n = 28), and invasive (n = 11) spots were then analyzed for aneuploidy of chromosomes 3, 6, 7, 8, 17, and 18 and deletions of p16 and p53 loci. Polysomy 6 and p16 deletion were significantly more frequent in noninvasive than in non-neoplastic spots. Polysomy 7, polysomy 18, p16 deletion, and p53 deletion were significantly more frequent in invasive than in noninvasive spots. Detection of polysomy 7 and p53 deletion gave a high diagnostic accuracy for invasive IPMN (sensitivity, 90.9%; specificity, 94.1%; and accuracy, 92.5%). Our study suggests that: (1) polysomy 6 and p16 deletion may contribute to
From the Departments of *Gastroenterology and Metabolism; wAnatomic Pathology and Molecular Diagnostics; and zCommunitybased Medical Education, Nagoya City University Graduate School of Medical Sciences, Nagoya, Japan. Conflicts of Interest and Source of Funding: Supported in part by a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science, and Technology (MEXT) of Japan (26461046 to Itaru Naitoh) and a research grant from Pancreas Research Foundation of Japan and Aichi Cancer Research Foundation. The authors have disclosed that they have no significant relationships with, or financial interest in, any commercial companies pertaining to this article. Correspondence: Hiroshi Inagaki, MD, PhD, Department of Anatomic Pathology and Molecular Diagnostics, Nagoya City University Graduate School of Medical Sciences, Japan, 1 Kawasumi, Mizuhocho, Mizuho-ku, Nagoya 467-8601, Japan (e-mail: hinagaki@med. nagoya-cu.ac.jp). Supplemental Digital Content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Website, www.ajsp.com. Copyright r 2014 Wolters Kluwer Health, Inc. All rights reserved.
512 | www.ajsp.com
adenomatous change of IPMN; (2) polysomy 7, polysomy 18, p16 deletion, and p53 deletion play roles in malignant transformation of noninvasive IPMN; and (3) polysomy 7 and p53 deletion may be excellent diagnostic markers for invasive IPMN. Key Words: pancreatic tumor, intraductal papillary mucinous neoplasm, fluorescence in situ hybridization (Am J Surg Pathol 2015;39:512–520)
I
ntraductal papillary mucinous neoplasm (IPMN) is an established type of pancreatic neoplasm that accounts for 0.5% to 9.8% of all pancreatic exocrine neoplasms.1,2 It is characterized by intraductal proliferation of mucinproducing epithelial cells and cystic dilatation of the pancreatic ducts.1 On the basis of the histology and immunohistochemistry (IHC) for mucin glycoproteins (MUCs) and CDX2, IPMNs can be subclassified into 4 types: gastric, intestinal, pancreatobiliary, and oncocytic.1,2 Most IPMN patients can be cured by complete resection when the tumors are noninvasive. However, some IPMNs progress to malignant or invasive IPMNs whose prognosis is poor.3 Preoperative diagnosis of invasive IPMN is still a challenge. The International Consensus Guidelines for the Management of pancreatic IPMN were revised in 2012.4 These guidelines use a combination of criteria of clinical history, sex, imaging characteristics, and cytology for preoperative diagnosis of invasive IPMNs. Branch duct (BD)-IPMN patients with any of the worrisome features (cyst Z3 cm, thickened/enhancing cyst walls, main duct size 5 to 9 mm, a nonenhancing mural nodule, and abrupt change in caliber of pancreatic duct with distal pancreatic atrophy) can be observed without immediate resection if the tumors do not show endoscopic ultrasound findings of definite mural nodules, main duct features suspicious for involvement, or cytologic findings suspicious or positive for malignancy. However, resection of the tumor is recommended for patients with main duct Am J Surg Pathol
Volume 39, Number 4, April 2015
Am J Surg Pathol
Volume 39, Number 4, April 2015
(MD)-IPMN and for those with BD-IPMN with any of the high-risk stigmata for malignancy (obstructive jaundice in a patient with cystic lesion of the head of the pancreas, enhancing solid component within cyst, and main pancreatic duct [MPD] Z10 mm in size). Although these criteria are useful for identifying patients recommended for surgery, the diagnostic accuracy of these criteria for invasive IPMNs before surgery does not seem to be very high.5–7 Detection of locus and chromosome abnormalities responsible for the development and progression of IPMN may help clarify IPMN oncogenesis and identify IPMNs at risk for malignant progression as well as develop novel therapeutic strategies for this tumor. However, to the best of our knowledge, there has been only 1 such study that focused on IPMN.8 The aims of the present study were to retrospectively examine locus and chromosome abnormalities in paraffin-embedded specimens of IPMN using the fluorescence in situ hybridization (FISH) technique and to identify any molecular markers that may aid in increasing the accuracy of preoperative diagnosis of invasive IPMN.
MATERIALS AND METHODS Case Selection IPMN cases (n = 28) were retrieved from the pathology files of the Department of Anatomic Pathology and Molecular Diagnostics, Nagoya City University Graduate School of Medical Sciences. All tumor samples were resected specimens and were fixed in formalin and embedded in paraffin. Informed consent was obtained, and the study was approved by the institutional review board of Nagoya City University (approval No. 140) and was conducted in accordance with the Declaration of Helsinki. Clinicopathologic data were obtained from medical records. All hematoxylin and eosin (H&E)stained slides were reviewed by 2 authors (Y.H. and H.I.) blinded to the clinical information.
Clinicopathologic Data IPMN cases were classified into 2 groups: noninvasive IPMN (including low-grade, intermediate-grade, and high-grade dysplasia) and invasive IPMN (IPMN with an associated invasive carcinoma) according to the World Health Organization classification.2 The following clinicopathologic factors were analyzed: age, sex, primary tumor site (head, body/tail, and multifocal), tumor type (MD-IPMN, BD-IPMN, and mixed type), tumor size, MPD dilatation, IPMN subtype, dysplasia grade, overall survival, and the existence of mural nodules, of lymph node metastasis, of vascular invasion, and of perineural invasion.
Immunohistochemistry Tissue sections were deparaffinized and rehydrated. After antigen retrieval by heat treatment, IHC was performed using an automated immunostainer (Bond-Max; Leica MicroSystems, Wetzlar, Germany) with monoCopyright
r
2014 Wolters Kluwer Health, Inc. All rights reserved.
FISH Analysis in Intraductal Papillary Mucinous Neoplasm
clonal antibodies against MUC1 (clone Ma695; dilution, 1:100), MUC2 (clone Ccp58; 1:200), MUC5AC (clone CLH2; 1:100), MUC6 (clone CLH5; 1:100), CDX2 (clone AMT28; 1:100), P53 (clone DO7; 1:200) (all from Leica MicroSystems), and P16 (clone JC8; Santa Cruz Biotechnology, Santa Cruz, CA; 1:200), and P53 (clone DO7; Leica MicroSystems; 1:200). Expression of MUC1, MUC2, MUC5AC, MUC6, and CDX2 was considered as positive when >10% of the tumor cells were stained. On the basis of the results of MUC1, MUC2, MUC5AC, MUC6, and CDX2 expression, IPMN cases were subclassified as gastric, intestinal, pancreatobiliary, or oncocytic subtypes, respectively (Table 1S, Supplementary Digital Content 1, http://links.lww.com/ PAS/A247).2 For analysis of P53 expression, a homogenous staining pattern in >10% of the cells was considered as positive.9 P16 was scored as absent when no nucleus in the IPMN epithelium was stained.10
FISH Procedure and Data Analysis FISH analysis of tissue was performed as previously described.11 To detect monosomy and polysomy in chromosomes 3, 6, 7, 8, 17, and 18, commercially available FISH probes (Kreatech Diagnostics, Amsterdam, The Netherlands), which hybridize to the centromeric regions of their respective chromosomes, were used. To detect p16 and p53 deletions, FISH was performed using probes (Kreatech) for p16 (9p21)/9q12 and p53 (17p13)/SE 17, respectively. Using serial H&E sections as a guide, non-neoplastic, noninvasive, and invasive spots (each 10 mm in diameter) were identified, and the FISH signals were counted in >200 tumor cells/ spot under a fluorescent microscope (BX53; Olympus Medical Systems, Tokyo, Japan). In the preliminary study, cutoff values were determined by counting FISH signals in >200 nuclei of nonneoplastic pancreatic glands obtained from surgeries for pancreatic ductal adenocarcinomas (n = 18). The upper cutoff values (%) were determined as the mean+3 SD. The upper cutoff values (%) thus obtained for polysomy in chromosomes 3, 6, 7, 8, 17, and 18 were 12%, 12%, 10%, 12%, 10%, and 12%, respectively. When the percentage of cells containing >2 FISH signals exceeded the respective cutoff values, the case was interpreted as positive for polysomy. The upper cutoff values (%) for monosomy in chromosomes 3, 6, 7, 8, 17, and 18 were 52%, 59%, 50%, 53%, 50%, and 55%, respectively. Similar to the judgment of polysomy, when the percentage of cells containing 6 mm).13 The 1 case that did not fulfill these criteria was surgically resected, because MPD dilatation and elevation of serum CA-19-9 level (2007.0 U/mL) were found during follow-up after gastric cancer surgery. On the basis of histology and IHC analysis of MUC1, MUC2, MUC5AC, MUC6, and CDX2, the IPMN cases were subclassified into gastric (n = 18), intestinal (n = 8), pancreatobiliary (n = 1), or oncocytic (n = 1) types. The frequency of invasive IPMN in MD-IPMN was 5/16 (31.3%) cases, whereas that in BD-IPMN was 5/11 (45.5%) cases. There was no significant difference in clinicopathologic characteristics between noninvasive and invasive IPMNs (Table 1). Patients with noninvasive IPMNs received no further treatment after tumor resection. Of 11 invasive IPMN patients, 3 patients received chemotherapy, 1 received radiotherapy, and 1 received chemoradiotherapy after the tumors were resected. The follow-up period ranged from 4 to 88 months. The overall 1-year survival rate of invasive IPMN patients was 72.7%, with a median survival time of 20 months. There was no significant difference in survival time (P = 0.2532) between patients with invasive MD-IPMN (n = 5) and invasive BD-IPMN (n = 5).
FISH Analysis and Comparison of Nonneoplastic, Noninvasive, and Invasive Spots FISH analysis was successfully carried out on all 67 tumor tissue spots (28 non-neoplastic spots, 28 noninvasive spots, and 11 invasive spots). Representative images of H&E and FISH analysis are shown in Figures 1 and 2, respectively. The FISH and IHC abnormalities observed in
514 | www.ajsp.com
Volume 39, Number 4, April 2015
TABLE 1. Cliniopathologic Characteristics of the IPMN Patients Noninvasive IPMN (n = 17) Age (mean [range]) (y) Sex (male/female) Tumor location Head Body or tail Multifocal IPMN type MD-IPMN BD-IPMN Mixed IPMN Tumor size (mm) < 30 Z30 MPD size (mm) 30 mm (P = 0.0025) and enhanced solid component (P = 0.0329). Sex, age, primary tumor site (head, body/tail, and multifocal), and tumor type (MD-IPMN, BD-IPMN, and mixed type) did not show any significant association with any factor. Locus and chromosome abnormalities and IHC expression of P16 and P53 had no impact on overall survival of invasive IPMN patients.
Preoperative Diagnosis of Invasive IPMN
FIGURE 1. Histological findings of noninvasive IPMN (A, case #15) and invasive IPMN (B, case #5). H&E staining.
2 homozygous spots, P = 0.0253), and p53 deletion (8 hemizygous spots and no homozygous spot, P = 0.0143). The mean polysomy number was significantly greater in invasive than in noninvasive spots (1.73 vs. 0.54, P = 0.0351), whereas there was no such difference between non-neoplastic and noninvasive spots. No increased monosomy was detected for any chromosome examined, and the mean monosomy number did not differ significantly between noninvasive and invasive spots.
IHC Analysis of P16 and P53, and Comparison With FISH Results IHC analysis of P16 and P53 was successfully performed for all spots. Loss of P16 expression was not significantly different between normal, noninvasive, and invasive IPMN spots. P53 overexpression was more frequent in noninvasive than in non-neoplastic spots (P = 0.0339) and was also more frequent in invasive than in noninvasive spots (P = 0.0455). p16 and p53 locus deletions were not related to P16 loss and P53 overexpression, respectively. Copyright
r
2014 Wolters Kluwer Health, Inc. All rights reserved.
Four FISH and 1 IHC factors were significantly different between noninvasive and invasive IPMNs (Table 3). To clarify whether these factors were useful in distinction between the 2 types of IPMN, we calculated the diagnostic accuracies of various patterns of locus/ chromosome and IHC abnormalities, including up to 2 factors, with a view to practical clinical application. When 1 factor was selected, p53 deletion gave the highest diagnostic accuracy of 86.4% (sensitivity, 72.7% and specificity, 100%), and the diagnostic accuracy increased to 92.5% (sensitivity, 90.9% and specificity, 94.1%) when p53 deletion and polysomy 7 were considered (Table 5).
DISCUSSION In the present study, we performed FISH and IHC analyses of 67 spots (28 non-neoplastic, 28 noninvasive, and 11 invasive) in 28 IPMN cases (17 noninvasive and 11 invasive). Chromosomes and loci analyzed by FISH in this study included chromosomes 3, 6, 7, 8, 17, and 18, as well as p16 and p53 gene loci, which have been previously studied in IPMN or pancreatic ductal adenocarcinoma.8,14–19 In addition, aberrant expression of P16 and P53 was examined using IHC.20–22 FISH and IHC results were compared among non-neoplastic, noninvasive, and invasive spots, and the clinicopathologic significance of the detected abnormalities was examined. We then evaluated the usefulness of FISH and IHC results for the differential diagnosis between noninvasive and invasive IPMNs, which we consider to be the most crucial point in deciding the patients’ treatment. www.ajsp.com |
515
Miyabe et al
Am J Surg Pathol
Volume 39, Number 4, April 2015
FIGURE 2. A, A FISH image using a probe against centromeric regions of chromosome 8 (case #4) shows a decreased number of red signals, indicating monosomy. B, A FISH image using a probe against chromosome 6 (case #5) shows an increased number of red signals, indicating polysomy. C, A FISH image using a probe for 17p13 (red) and a centromeric region of chromosome 17 (green) shows 1 red and 2 green signals (case #11), indicating hemizygous p53 deletion (arrows). D, A FISH image using a probe for 9p21 (red) and 9q21 (green) shows no red and 2 green signals (case #15), indicating homozygous p16 deletion (arrows).
The most important finding of this study is that we defined the locus and chromosome abnormalities that may be responsible for the advancement of IPMN. A schematic diagram of the identified abnormalities is depicted in Figure 3. Although none of the non-neoplastic spots showed these abnormalities, polysomy 6 and p16 deletion were found more frequently in noninvasive IPMNs. We also showed that polysomies 7 and 18 and p16 and p53 deletions were more frequent in invasive than
516 | www.ajsp.com
in noninvasive IPMNs. There would be a direct pathway from normal pancreatic gland to invasive IPMN without passing through a noninvasive IPMN condition. Unfortunately, we could not investigate such a pathway, as all invasive IPMNs studied had noninvasive IPMN lesions concurrently. Genetic abnormalities responsible for the transition from non-neoplastic to noninvasive lesions seem to be heterogenous, and there may be no common or major abnormalities in this process. In contrast, in the Copyright
r
2014 Wolters Kluwer Health, Inc. All rights reserved.
Am J Surg Pathol
Volume 39, Number 4, April 2015
FISH Analysis in Intraductal Papillary Mucinous Neoplasm
TABLE 2. Patterns of Locus/Chromosome Aberrations Non-neoplastic Spot Case No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
Noninvasive Spot
Invasive Spot
FISH
IHC
Dysplasia
FISH
IHC
FISH
IHC
— — — — — — — — — — — — — — — — — — — — — — — — — — — —
— P16 P16 , P53+ P53+ — P16 — P53+ P16 P53+ P53+ P16 P16 , P53+ — P16 — P16 — — P16 — P53+ P53+ P53+ P53+ — — —
High High High Int. Int. Int. Int. Low Low Low Low High High High Int. Int. Int. Int. Int. Int. Int. Int. Int. Low Low Low Low Low
p53 hemi 6+ — p16 hemi, p53 hemi 6+ 6+ — — — — — 3+, 6+, 8+ 6+, 17+ 6+ 6+, p16 homo 3+, 18+ 7+ 8+ 3+ p16 hemi p16 hemi — — — — — — —
— P16 P53+ P53+ P16 , P53+ P16 , P53+ P16 P16 , P53+ P16 P53+ P16 P53+ P16 , P53+ P16 , P53+ P16 , P53+ — P16 , P53+ P53+ — — — P53+ P53+ P53+ P53+ P16 — —
8 , p16 hemi, p53 hemi 6+, 7+, 18+ 7+ 8, 17+, p16 hemi, p53 hemi 6+, 17+, 18+, p53 hemi 3+, 6+, 18+, p16 hemi, p53 hemi — 3+, 6+, 7+, 17+, 18+, p53 hemi 18+, p16 hemi, p53 hemi p16 homo, p53 hemi 6+, 7+, p16 homo, p53 hemi
— P16 , P53+ P53+ P53+ P16 , P53+ P16 , P53+ P53+ P16 , P53+ P16 , P53+ P16 , P53+ P16 , P53+
The results on the gray background show data of cases with invasive IPMN, and those on the white background show data of cases with noninvasive IPMN. Minus and plus signs after a number in FISH columns indicate monosomy and polysomy, respectively. Hemi indicates hemizygous deletion; high, high-grade dysplasia; homo, homozygous deletion; int., intermediate-grade dysplasia; low, low-grade dysplasia; P16 , loss of P16 IHC expression; P53, P53 IHC overexpression.
TABLE 3. Comparison of FISH and IHC Results for IPMN Development and Progression Locus/Chromosome Aberration FISH outcomes Polysomy 3 Polysomy 6 Polysomy 7 Polysomy 8 Polysomy 17 Polysomy 18 Monosomy 3 Monosomy 6 Monosomy 7 Monosomy 8 Monosomy 17 Monosomy 18 p16 deletion Hemi./homo. p53 deletion Hemi./homo. IHC outcomes Loss of P16 P53 overexpression
Non-neoplastic Spot (n = 28)
Noninvasive Spot (n = 28)
0 0 0 0 0 0 0 0 0 0 0 0 0 0/0 0 0/0
3 7 1 2 1 1 0 0 0 0 0 0 4 3/1 2 2/0
NS 0.0082 NS NS NS NS — — — — — — 0.0455
19 10
17 16
NS 0.0339
P
NS
Noninvasive Spot (n = 11)
Invasive Spot (n = 11)
0 3 0 0 0 0 0 0 0 0 0 0 1 1/0 2 2/0
2 5 4 0 3 5 0 0 0 2 0 0 6 4/2 8 8/0
NS NS 0.0455 — NS 0.0253 — — — NS — — 0.0253
4 6
4 10
NS 0.0455
P
0.0143
Values represent the number of spots. The McNemar test was used for comparison of 2 spots in the same case. Hemi. indicates hemizygous deletion; homo., homozygous deletion; NS, not significant.
Copyright
r
2014 Wolters Kluwer Health, Inc. All rights reserved.
www.ajsp.com |
517
Miyabe et al
Am J Surg Pathol
Volume 39, Number 4, April 2015
TABLE 4. The Relationship Between FISH/IHC Results and Clinicopathologic Parameters Clinicopathologic Features
Polysomy 6
Polysomy 7
P
P
n
Clinical factor Tumor size (mm) < 30 15 Z30 13 MPD size (mm) < 10 18 Z10 10 Mural nodule None/nonenhanced 14 Enhanced 14 Lymph node metastasis Yes 24 No 4 Pathologic factor Histologic types Gastric type 18 Intestinal type 8 Dysplasia* Low/intermediate 22 High 6 Vascular invasion Yes 5 No 23 Perineural invasion Yes 7 No 21
Polysomy 18
p16 Deletion
p53 Deletion
P
P
P
P53 IHC P
3 (20) 6 (46.2)
NS
3 (20) 2 (15.4)
NS
2 (13.3) 4 (30.8)
NS
5 (33.3) 4 (19.0)
NS
3 (20) 5 (38.5)
NS
7 (46.7) 13 (100)
0.0025
3 (16.7) 6 (60.0)
0.0346
1 (5.6) 4 (15.4)
NS
4 (22.2) 2 (20.0)
NS
7 (38.9) 2 (20.0)
NS
7 (38.9) 1 (10.0)
NS
11 (61.1) 9 (90.0)
NS
3 (21.4) 6 (42.9)
NS
0 (0) 5 (35.7)
0.0407
2 (14.3) 4 (28.8)
NS
4 (28.8) 5 (35.7)
NS
2 (14.3) 6 (42.9)
NS
7 (50.0) 13 (92.9)
0.0329
6 (25.0) 3 (75.0)
NS
3 (12.5) 2 (50)
NS
4 (16.7) 2 (50.0)
NS
6 (25.0) 3 (75.0)
NS
4 (100) 4 (16.7)
0.0034
16 (66.7) 4 (100)
NS
3 (16.7) 4 (50)
NS
1 (5.6) 3 (37.5)
NS
3 (16.7) 1 (12.5)
NS
7 (39.9) 1 (12.5)
NS
6 (33.3) 0 (0)
NS
11 (61.1) 7 (87.5)
NS
3 (13.6) 4 (66.7)
0.0207
1 (4.5) 0 (0)
NS
1 (4.5) 0 (0)
NS
4 (18.2) 0 (0)
NS
1 (4.5) 1 (16.7)
NS
12 (54.5) 4 (66.7)
NS
2 (40.0) 7 (30.4)
NS
2 (40) 3 (13)
NS
2 (40.0) 4 (17.4)
NS
4 (80.0) 5 (21.7)
0.0256
5 (100) 3 (13.0)
0.0006
4 (80) 16 (69.6)
NS
4 (57.1) 5 (23.8)
NS
2 (28.6) 3 (14.3)
NS
4 (57.1) 2 (9.5)
0.0207
5 (57.1) 4 (9.5)
0.0196
7 (100) 1 (4.8)
< 0.0001
6 (85.7) 14 (66.7)
NS
Data represented as the number of cases (percentages among each parameter). *Noninvasive spots in low-grade plus intermediate-grade dysplasia were compared with those in high-grade dysplasia. NS indicates not significant; p16 hemi, p16 hemizygous deletions; p53 hemi, p53 hemizygous deletions.
transition from noninvasive to invasive lesions, p53 gene deletion was found in 73% of the invasive lesions. As a major genetic abnormality, p53 gene deletion is considered to play an important role in adenomatous to malignant transition. In this study, we showed that polysomies 6, 7, and 18 were involved in oncogenesis or malignant transformation in IPMN. To the best of our knowledge, involvement of these abnormalities has not been pointed out previously. The precise oncogenetic mechanism of
these polysomies remains to be elucidated. However, their significance in transformation may be validated by their significant association with clinicopathologic factors: polysomy 6 was associated with MPD dilatation, the presence of mural nodules, and high-grade dysplasia in noninvasive spots; polysomy 7 with enhanced solid component; and polysomy 18 with perineural invasion of the tumor. p16/CDKN2A is a well-known tumorsuppressor gene. High prevalence of p16 inactivation (loss of heterozygosity in the 9p allele) has been described
TABLE 5. Outcomes for the Diagnosis of Invasive IPMN in the Present Study Diagnostic Criteria The criteria based on FISH or IHC data Polysomy 7 Polysomy 18 p16 deletion p53 deletion P53 IHC positive Polysomy 7 or polysomy 18 Polysomy 7 or p16 deletion Polysomy 7 or p53 deletion Polysomy 7 or P53 IHC positive Polysomy 18 or p16 deletion Polysomy 18 or p53 deletion Polysomy 18 or P53 IHC positive p16 deletion or p53 deletion p16 deletion or P53 IHC positive p53 deletion or P53 IHC positive
518 | www.ajsp.com
Sensitivity (%)
Specificity (%)
Accuracy (%)
36.4 45.5 54.5 72.7 90.9 63.6 81.8 90.9 90.9 81.8 81.8 45.5 72.7 100 100
94.1 94.1 82.4 100 57.1 88.2 76.5 94.1 41.2 76.5 94.1 100 82.4 29.4 41.2
65.2 69.8 68.5 86.4 74.0 75.9 79.2 92.5 66.1 79.2 88.0 72.8 77.6 64.7 70.6
Copyright
r
2014 Wolters Kluwer Health, Inc. All rights reserved.
Am J Surg Pathol
Volume 39, Number 4, April 2015
FIGURE 3. The schematic diagram of locus and chromosome abnormalities in this study. Polysomy 6 and p16 deletion are associated with the transition from non-neoplastic epithelium to noninvasive IPMN epithelium, and polysomies 7 and 18 and p16 and p53 deletions are associated with progression from the noninvasive to invasive IPMN epithelium.
in invasive IPMN.19 We demonstrated that p16 inactivation was associated with both the transition from non-neoplastic to noninvasive lesions and the transition from noninvasive to invasive lesions. p53 is another important tumor-suppressor gene. This molecule was found to play a major role in progression from noninvasive to invasive IPMNs and to be a potent diagnostic marker of invasive IPMNs. These findings are supported by previous microsatellite polymerase chain reaction studies that detected loss of heterogeneity of p53 in invasive IPMNs but not in noninvasive IPMNs.14 Regarding numeric abnormalities in chromosomes in IPMN, a pioneer study was reported by Soldini et al.8 Using 12 IPMN cases, they described that monosomy 6 may represent an early event necessary for neoplastic transformation of the pancreatic duct epithelium and that monosomy 18 may be associated with the progression from borderline to malignant IPMN. In the present study, some polysomies did emerge as genetic factors associated with the development or progression of IPMN; however, no monosomy was implicated. The discrepancies between the present study and that of Soldini and colleagues are difficult to explain. However, one plausible explanation is the difference in the cutoff values defining polysomies and monosomies in the respective studies. In our study, each upper cutoff value was calculated as the mean+3 SD after counting >200 nuclei of normal duct epithelium (18 cases). The cutoff values thus obtained ranged between 10% and 12% and 50% and 59% for polysomies and monosomies, respectively. In contrast, in the study by Soldini, et al,8 cutoff values were set as >20% and >40% for polysomies and monosomies, respectively, which were calculated from the mean+2 SD. Copyright
r
2014 Wolters Kluwer Health, Inc. All rights reserved.
FISH Analysis in Intraductal Papillary Mucinous Neoplasm
Another important finding of the present study is that FISH detection of p53 gene deletion and polysomy 7 may provide a high diagnostic accuracy for identification of invasive IPMNs. The International Consensus Guidelines for the Management of pancreatic IPMN were revised in 2012,4 and these guidelines have been used worldwide. However, there may be some issues that remain to be resolved: the levels of evidence for all items addressed in the guidelines are low, and the positive predictive value for high-grade dysplasia/invasive carcinoma is not very high, reportedly 62.5% in 1 study7 necessitating more accurate markers. Cytology might be the most useful test for the detection of malignancy. However, in addition to the typical impediments to cytologic interpretation of scant cellularity and poor cellular preservation, there is a lack of standardized, reproducible, diagnostic criteria for IPMNs.23 As suggested in this study, locus and chromosome abnormalities detected using the FISH technique may be a promising ancillary test for the diagnosis of invasive IPMN lesions. As shown in Table 5, detection of both p53 deletion and polysomy 7 provided a high diagnostic accuracy of 92.5%. The usefulness of FISH assays for the diagnosis of invasive IPMN should be further validated in prospective large-scale studies. Regarding p53 and p16, IHC results did not correlate with their deletion statuses by FISH analysis. It is difficult to explain these discrepancies. Plausible explanations may include the involvement of other molecular mechanisms such as gene mutation, aberrant methylation, and transcriptional and posttranscriptional factors.24,25 Cytologic examinations are sometimes performed for IPMN cases suspected of malignancy; however, its diagnostic accuracy has not been so high, ranging from 74% to 86%.26–28 In addition, stromal invasion of the tumor is difficult to diagnose. The diagnostic accuracy of IPMN cytology is expected to be improved using additional FISH analysis for p53 locus and polysomy 7, as has been shown in this study. Some researchers have mentioned that the diagnostic significance of trisomy 7 for malignancy remains uncertain. Fritcher et al15 performed a FISH study for malignant pancreatic and/or biliary strictures and described that trisomy 7 was occasionally detected in non-neoplastic and neoplastic tissues. In the present study, however, none of non-neoplastic spots had trisomy 7 or polysomy 7. This result is supported by a study of Kubiliun et al18 describing that none of the patients with benign disease had trisomy 7. In conclusion, our study demonstrated that some specific locus and chromosome abnormalities may be important in the advancement from normal gland to noninvasive IPMN and from noninvasive to invasive IPMN. In addition, FISH may be a useful modality for differentiating invasive from noninvasive IPMNs and should be evaluated by using cytologic specimen in clinical practice. These findings provide insights into the oncogenesis and molecular diagnosis of IPMN.
REFERENCES 1. Hruban RH, Pitman MB, Klimstra DS. Intraductal neoplasms. In: Silverberg SG, Sobin LH, eds. AFIP Atlas of Tumor Pathology
www.ajsp.com |
519
Miyabe et al
2.
3. 4. 5. 6.
7.
8.
9. 10. 11. 12. 13.
14. 15.
Series 4. Washington, DC: American Registry of Pathology; 2007: 75–110. Adsay NV, Fukushima N, Furukawa T, et al. Intraductal neoplasms of the pancreas. In: Bosman FT, Carneiro F, Hruban RH, et al, eds. WHO Classification of Tumours of the Digestive System. Lyon: IARC; 2010:304–313. Hibi Y, Fukushima N, Tsuchida A, et al. Pancreatic juice cytology and subclassification of intraductal papillary mucinous neoplasms of the pancreas. Pancreas. 2007;34:197–204. Tanaka M, Fernandez-del Castillo C, Adsay V, et al. International consensus guidelines 2012 for the management of IPMN and MCN of the pancreas. Pancreatology. 2012;12:183–197. Jang JY, Park T, Lee S, et al. Validation of international consensus guidelines for the resection of branch duct-type intraductal papillary mucinous neoplasms. Br J Surg. 2014;101:686–692. Amato E, Molin MD, Mafficini A, et al. Targeted next-generation sequencing of cancer genes dissects the molecular profiles of intraductal papillary neoplasms of the pancreas. J Pathol. 2014;233:217–227. Goh BK, Thng CH, Tan DM, et al. Evaluation of the Sendai and 2012 International Consensus Guidelines based on cross-sectional imaging findings performed for the initial triage of mucinous cystic lesions of the pancreas: a single institution experience with 114 surgically treated patients. Am J Surg. 2014;208:202–209. Soldini D, Gugger M, Burckhardt E, et al. Progressive genomic alterations in intraductal papillary mucinous tumours of the pancreas and morphologically similar lesions of the pancreatic ducts. J Pathol. 2003;199:453–461. Bova RJ, Quinn DI, Nankervis JS, et al. Cyclin D1 and p16INK4A expression predict reduced survival in carcinoma of the anterior tongue. Clin Cancer Res. 1999;5:2810–2819. Wilentz RE, Geradts J, Maynard R, et al. Inactivation of the p16 (INK4A) tumor-suppressor gene in pancreatic duct lesions: loss of intranuclear expression. Cancer Res. 1998;58:4740–4744. Noda H, Okumura Y, Nakayama T, et al. Clinicopathological significance of MAML2 gene split in mucoepidermoid carcinoma. Cancer Sci. 2013;104:85–92. Nomoto J, Hiramoto N, Kato M, et al. Deletion of the TNFAIP3/ A20 gene detected by FICTION analysis in classical Hodgkin lymphoma. BMC Cancer. 2012;12:457. Tanaka M, Chari S, Adsay V, et al. International consensus guidelines for management of intraductal papillary mucinous neoplasms and mucinous cystic neoplasms of the pancreas. Pancreatology. 2006;6:17–32. Wada K. p16 and p53 gene alterations and accumulations in the malignant evolution of intraductal papillary-mucinous tumors of the pancreas. J Hepatobiliary Pancreat Surg. 2002;9:76–85. Fritcher EG, Kipp BR, Halling KC, et al. A multivariable model using advanced cytologic methods for the evaluation of indeter-
520 | www.ajsp.com
Am J Surg Pathol
16. 17.
18.
19. 20.
21.
22.
23. 24. 25.
26.
27. 28.
Volume 39, Number 4, April 2015
minate pancreatobiliary strictures. Gastroenterology. 2009;136: 2180–2186. Schleicher C, Poremba C, Wolters H, et al. Gain of chromosome 8q: a potential prognostic marker in resectable adenocarcinoma of the pancreas? Ann Surg Oncol. 2007;14:1327–1335. Levy MJ, Oberg TN, Campion MB, et al. Comparison of methods to detect neoplasia in patients undergoing endoscopic ultrasoundguided fine-needle aspiration. Gastroenterology. 2012;142:1112– 1121. e1112. Kubiliun N, Ribeiro A, Fan YS, et al. EUS-FNA with rescue fluorescence in situ hybridization for the diagnosis of pancreatic carcinoma in patients with inconclusive on-site cytopathology results. Gastrointest Endosc. 2011;74:541–547. Fujii H, Inagaki M, Kasai S, et al. Genetic progression and heterogeneity in intraductal papillary-mucinous neoplasms of the pancreas. Am J Pathol. 1997;151:1447–1454. Biankin AV, Biankin SA, Kench JG, et al. Aberrant p16(INK4A) and DPC4/Smad4 expression in intraductal papillary mucinous tumours of the pancreas is associated with invasive ductal adenocarcinoma. Gut. 2002;50:861–868. Biankin AV, Kench JG, Biankin SA, et al. Pancreatic intraepithelial neoplasia in association with intraductal papillary mucinous neoplasms of the pancreas: implications for disease progression and recurrence. Am J Surg Pathol. 2004;28:1184–1192. Abe K, Suda K, Arakawa A, et al. Different patterns of p16INK4A and p53 protein expressions in intraductal papillary-mucinous neoplasms and pancreatic intraepithelial neoplasia. Pancreas. 2007;34:85–91. Pitman MB. Revised international consensus guidelines for the management of patients with mucinous cysts. Cancer Cytopathol. 2012;120:361–365. Ribeiro A, Peng J, Casas C, et al. EUS FNA with FISH analysis in 104 patients with pancreatic mass. J Gastroenterol Hepatol. 2014;29:1654–1658. Moore PS, Orlandini S, Zamboni G, et al. Pancreatic tumours: molecular pathways implicated in ductal cancer are involved in ampullary but not in exocrine nonductal or endocrine tumorigenesis. Br J Cancer. 2001;84:253–262. Ono J, Yaeger KA, Genevay M, et al. Cytological analysis of small branch-duct intraductal papillary mucinous neoplasms provides a more accurate risk assessment of malignancy than symptoms. CytoJournal. 2011;8:21. Cizginer S, Turner BG, Bilge AR, et al. Cyst fluid carcinoembryonic antigen is an accurate diagnostic marker of pancreatic mucinous cysts. Pancreas. 2011;40:1024–1028. Pais SA, Attasaranya S, Leblanc JK, et al. Role of endoscopic ultrasound in the diagnosis of intraductal papillary mucinous neoplasms: correlation with surgical histopathology. Clin Gastroenterol Hepatol. 2007;5:489–495.
Copyright
r
2014 Wolters Kluwer Health, Inc. All rights reserved.
Locus/Chromosome Aberrations in Intraductal Papillary Mucinous Neoplasms Analyzed by Fluorescence In Situ Hybridization Katsuyuki Miyabe, MD, PhD,*w Yasuki Hori, MD,*w Takahiro Nakazawa, MD, PhD,* Kazuki Hayashi, MD, PhD,* Itaru Naitoh, MD, PhD,* Shuya Shimizu, MD,* Hiromu Kondo, MD,* Yuji Nishi, MD,* Michihiro Yoshida, MD, PhD,* Shuichiro Umemura, MD,* Akihisa Kato, MD,* Hirotaka Ohara, MD, PhD,z Takashi Joh, MD, PhD,* and Hiroshi Inagaki, MD, PhDw
Abstract: Locus and chromosome abnormalities have not been well clarified in intraductal papillary mucinous neoplasms (IPMNs). The aim of this study was to retrospectively examine these abnormalities using fluorescence in situ hybridization. IPMNs (n = 28) were histopathologically classified into noninvasive IPMN (n = 17) and IPMN with an associated invasive carcinoma (invasive IPMN, n = 11) groups. Noninvasive IPMNs possessed non-neoplastic and noninvasive spots in their tissues, and invasive IPMN cases possessed non-neoplastic, noninvasive, and invasive spots. Non-neoplastic (n = 28), noninvasive (n = 28), and invasive (n = 11) spots were then analyzed for aneuploidy of chromosomes 3, 6, 7, 8, 17, and 18 and deletions of p16 and p53 loci. Polysomy 6 and p16 deletion were significantly more frequent in noninvasive than in non-neoplastic spots. Polysomy 7, polysomy 18, p16 deletion, and p53 deletion were significantly more frequent in invasive than in noninvasive spots. Detection of polysomy 7 and p53 deletion gave a high diagnostic accuracy for invasive IPMN (sensitivity, 90.9%; specificity, 94.1%; and accuracy, 92.5%). Our study suggests that: (1) polysomy 6 and p16 deletion may contribute to
From the Departments of *Gastroenterology and Metabolism; wAnatomic Pathology and Molecular Diagnostics; and zCommunitybased Medical Education, Nagoya City University Graduate School of Medical Sciences, Nagoya, Japan. Conflicts of Interest and Source of Funding: Supported in part by a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science, and Technology (MEXT) of Japan (26461046 to Itaru Naitoh) and a research grant from Pancreas Research Foundation of Japan and Aichi Cancer Research Foundation. The authors have disclosed that they have no significant relationships with, or financial interest in, any commercial companies pertaining to this article. Correspondence: Hiroshi Inagaki, MD, PhD, Department of Anatomic Pathology and Molecular Diagnostics, Nagoya City University Graduate School of Medical Sciences, Japan, 1 Kawasumi, Mizuhocho, Mizuho-ku, Nagoya 467-8601, Japan (e-mail: hinagaki@med. nagoya-cu.ac.jp). Supplemental Digital Content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Website, www.ajsp.com. Copyright r 2014 Wolters Kluwer Health, Inc. All rights reserved.
512 | www.ajsp.com
adenomatous change of IPMN; (2) polysomy 7, polysomy 18, p16 deletion, and p53 deletion play roles in malignant transformation of noninvasive IPMN; and (3) polysomy 7 and p53 deletion may be excellent diagnostic markers for invasive IPMN. Key Words: pancreatic tumor, intraductal papillary mucinous neoplasm, fluorescence in situ hybridization (Am J Surg Pathol 2015;39:512–520)
I
ntraductal papillary mucinous neoplasm (IPMN) is an established type of pancreatic neoplasm that accounts for 0.5% to 9.8% of all pancreatic exocrine neoplasms.1,2 It is characterized by intraductal proliferation of mucinproducing epithelial cells and cystic dilatation of the pancreatic ducts.1 On the basis of the histology and immunohistochemistry (IHC) for mucin glycoproteins (MUCs) and CDX2, IPMNs can be subclassified into 4 types: gastric, intestinal, pancreatobiliary, and oncocytic.1,2 Most IPMN patients can be cured by complete resection when the tumors are noninvasive. However, some IPMNs progress to malignant or invasive IPMNs whose prognosis is poor.3 Preoperative diagnosis of invasive IPMN is still a challenge. The International Consensus Guidelines for the Management of pancreatic IPMN were revised in 2012.4 These guidelines use a combination of criteria of clinical history, sex, imaging characteristics, and cytology for preoperative diagnosis of invasive IPMNs. Branch duct (BD)-IPMN patients with any of the worrisome features (cyst Z3 cm, thickened/enhancing cyst walls, main duct size 5 to 9 mm, a nonenhancing mural nodule, and abrupt change in caliber of pancreatic duct with distal pancreatic atrophy) can be observed without immediate resection if the tumors do not show endoscopic ultrasound findings of definite mural nodules, main duct features suspicious for involvement, or cytologic findings suspicious or positive for malignancy. However, resection of the tumor is recommended for patients with main duct Am J Surg Pathol
Volume 39, Number 4, April 2015
Am J Surg Pathol
Volume 39, Number 4, April 2015
(MD)-IPMN and for those with BD-IPMN with any of the high-risk stigmata for malignancy (obstructive jaundice in a patient with cystic lesion of the head of the pancreas, enhancing solid component within cyst, and main pancreatic duct [MPD] Z10 mm in size). Although these criteria are useful for identifying patients recommended for surgery, the diagnostic accuracy of these criteria for invasive IPMNs before surgery does not seem to be very high.5–7 Detection of locus and chromosome abnormalities responsible for the development and progression of IPMN may help clarify IPMN oncogenesis and identify IPMNs at risk for malignant progression as well as develop novel therapeutic strategies for this tumor. However, to the best of our knowledge, there has been only 1 such study that focused on IPMN.8 The aims of the present study were to retrospectively examine locus and chromosome abnormalities in paraffin-embedded specimens of IPMN using the fluorescence in situ hybridization (FISH) technique and to identify any molecular markers that may aid in increasing the accuracy of preoperative diagnosis of invasive IPMN.
MATERIALS AND METHODS Case Selection IPMN cases (n = 28) were retrieved from the pathology files of the Department of Anatomic Pathology and Molecular Diagnostics, Nagoya City University Graduate School of Medical Sciences. All tumor samples were resected specimens and were fixed in formalin and embedded in paraffin. Informed consent was obtained, and the study was approved by the institutional review board of Nagoya City University (approval No. 140) and was conducted in accordance with the Declaration of Helsinki. Clinicopathologic data were obtained from medical records. All hematoxylin and eosin (H&E)stained slides were reviewed by 2 authors (Y.H. and H.I.) blinded to the clinical information.
Clinicopathologic Data IPMN cases were classified into 2 groups: noninvasive IPMN (including low-grade, intermediate-grade, and high-grade dysplasia) and invasive IPMN (IPMN with an associated invasive carcinoma) according to the World Health Organization classification.2 The following clinicopathologic factors were analyzed: age, sex, primary tumor site (head, body/tail, and multifocal), tumor type (MD-IPMN, BD-IPMN, and mixed type), tumor size, MPD dilatation, IPMN subtype, dysplasia grade, overall survival, and the existence of mural nodules, of lymph node metastasis, of vascular invasion, and of perineural invasion.
Immunohistochemistry Tissue sections were deparaffinized and rehydrated. After antigen retrieval by heat treatment, IHC was performed using an automated immunostainer (Bond-Max; Leica MicroSystems, Wetzlar, Germany) with monoCopyright
r
2014 Wolters Kluwer Health, Inc. All rights reserved.
FISH Analysis in Intraductal Papillary Mucinous Neoplasm
clonal antibodies against MUC1 (clone Ma695; dilution, 1:100), MUC2 (clone Ccp58; 1:200), MUC5AC (clone CLH2; 1:100), MUC6 (clone CLH5; 1:100), CDX2 (clone AMT28; 1:100), P53 (clone DO7; 1:200) (all from Leica MicroSystems), and P16 (clone JC8; Santa Cruz Biotechnology, Santa Cruz, CA; 1:200), and P53 (clone DO7; Leica MicroSystems; 1:200). Expression of MUC1, MUC2, MUC5AC, MUC6, and CDX2 was considered as positive when >10% of the tumor cells were stained. On the basis of the results of MUC1, MUC2, MUC5AC, MUC6, and CDX2 expression, IPMN cases were subclassified as gastric, intestinal, pancreatobiliary, or oncocytic subtypes, respectively (Table 1S, Supplementary Digital Content 1, http://links.lww.com/ PAS/A247).2 For analysis of P53 expression, a homogenous staining pattern in >10% of the cells was considered as positive.9 P16 was scored as absent when no nucleus in the IPMN epithelium was stained.10
FISH Procedure and Data Analysis FISH analysis of tissue was performed as previously described.11 To detect monosomy and polysomy in chromosomes 3, 6, 7, 8, 17, and 18, commercially available FISH probes (Kreatech Diagnostics, Amsterdam, The Netherlands), which hybridize to the centromeric regions of their respective chromosomes, were used. To detect p16 and p53 deletions, FISH was performed using probes (Kreatech) for p16 (9p21)/9q12 and p53 (17p13)/SE 17, respectively. Using serial H&E sections as a guide, non-neoplastic, noninvasive, and invasive spots (each 10 mm in diameter) were identified, and the FISH signals were counted in >200 tumor cells/ spot under a fluorescent microscope (BX53; Olympus Medical Systems, Tokyo, Japan). In the preliminary study, cutoff values were determined by counting FISH signals in >200 nuclei of nonneoplastic pancreatic glands obtained from surgeries for pancreatic ductal adenocarcinomas (n = 18). The upper cutoff values (%) were determined as the mean+3 SD. The upper cutoff values (%) thus obtained for polysomy in chromosomes 3, 6, 7, 8, 17, and 18 were 12%, 12%, 10%, 12%, 10%, and 12%, respectively. When the percentage of cells containing >2 FISH signals exceeded the respective cutoff values, the case was interpreted as positive for polysomy. The upper cutoff values (%) for monosomy in chromosomes 3, 6, 7, 8, 17, and 18 were 52%, 59%, 50%, 53%, 50%, and 55%, respectively. Similar to the judgment of polysomy, when the percentage of cells containing 6 mm).13 The 1 case that did not fulfill these criteria was surgically resected, because MPD dilatation and elevation of serum CA-19-9 level (2007.0 U/mL) were found during follow-up after gastric cancer surgery. On the basis of histology and IHC analysis of MUC1, MUC2, MUC5AC, MUC6, and CDX2, the IPMN cases were subclassified into gastric (n = 18), intestinal (n = 8), pancreatobiliary (n = 1), or oncocytic (n = 1) types. The frequency of invasive IPMN in MD-IPMN was 5/16 (31.3%) cases, whereas that in BD-IPMN was 5/11 (45.5%) cases. There was no significant difference in clinicopathologic characteristics between noninvasive and invasive IPMNs (Table 1). Patients with noninvasive IPMNs received no further treatment after tumor resection. Of 11 invasive IPMN patients, 3 patients received chemotherapy, 1 received radiotherapy, and 1 received chemoradiotherapy after the tumors were resected. The follow-up period ranged from 4 to 88 months. The overall 1-year survival rate of invasive IPMN patients was 72.7%, with a median survival time of 20 months. There was no significant difference in survival time (P = 0.2532) between patients with invasive MD-IPMN (n = 5) and invasive BD-IPMN (n = 5).
FISH Analysis and Comparison of Nonneoplastic, Noninvasive, and Invasive Spots FISH analysis was successfully carried out on all 67 tumor tissue spots (28 non-neoplastic spots, 28 noninvasive spots, and 11 invasive spots). Representative images of H&E and FISH analysis are shown in Figures 1 and 2, respectively. The FISH and IHC abnormalities observed in
514 | www.ajsp.com
Volume 39, Number 4, April 2015
TABLE 1. Cliniopathologic Characteristics of the IPMN Patients Noninvasive IPMN (n = 17) Age (mean [range]) (y) Sex (male/female) Tumor location Head Body or tail Multifocal IPMN type MD-IPMN BD-IPMN Mixed IPMN Tumor size (mm) < 30 Z30 MPD size (mm) 30 mm (P = 0.0025) and enhanced solid component (P = 0.0329). Sex, age, primary tumor site (head, body/tail, and multifocal), and tumor type (MD-IPMN, BD-IPMN, and mixed type) did not show any significant association with any factor. Locus and chromosome abnormalities and IHC expression of P16 and P53 had no impact on overall survival of invasive IPMN patients.
Preoperative Diagnosis of Invasive IPMN
FIGURE 1. Histological findings of noninvasive IPMN (A, case #15) and invasive IPMN (B, case #5). H&E staining.
2 homozygous spots, P = 0.0253), and p53 deletion (8 hemizygous spots and no homozygous spot, P = 0.0143). The mean polysomy number was significantly greater in invasive than in noninvasive spots (1.73 vs. 0.54, P = 0.0351), whereas there was no such difference between non-neoplastic and noninvasive spots. No increased monosomy was detected for any chromosome examined, and the mean monosomy number did not differ significantly between noninvasive and invasive spots.
IHC Analysis of P16 and P53, and Comparison With FISH Results IHC analysis of P16 and P53 was successfully performed for all spots. Loss of P16 expression was not significantly different between normal, noninvasive, and invasive IPMN spots. P53 overexpression was more frequent in noninvasive than in non-neoplastic spots (P = 0.0339) and was also more frequent in invasive than in noninvasive spots (P = 0.0455). p16 and p53 locus deletions were not related to P16 loss and P53 overexpression, respectively. Copyright
r
2014 Wolters Kluwer Health, Inc. All rights reserved.
Four FISH and 1 IHC factors were significantly different between noninvasive and invasive IPMNs (Table 3). To clarify whether these factors were useful in distinction between the 2 types of IPMN, we calculated the diagnostic accuracies of various patterns of locus/ chromosome and IHC abnormalities, including up to 2 factors, with a view to practical clinical application. When 1 factor was selected, p53 deletion gave the highest diagnostic accuracy of 86.4% (sensitivity, 72.7% and specificity, 100%), and the diagnostic accuracy increased to 92.5% (sensitivity, 90.9% and specificity, 94.1%) when p53 deletion and polysomy 7 were considered (Table 5).
DISCUSSION In the present study, we performed FISH and IHC analyses of 67 spots (28 non-neoplastic, 28 noninvasive, and 11 invasive) in 28 IPMN cases (17 noninvasive and 11 invasive). Chromosomes and loci analyzed by FISH in this study included chromosomes 3, 6, 7, 8, 17, and 18, as well as p16 and p53 gene loci, which have been previously studied in IPMN or pancreatic ductal adenocarcinoma.8,14–19 In addition, aberrant expression of P16 and P53 was examined using IHC.20–22 FISH and IHC results were compared among non-neoplastic, noninvasive, and invasive spots, and the clinicopathologic significance of the detected abnormalities was examined. We then evaluated the usefulness of FISH and IHC results for the differential diagnosis between noninvasive and invasive IPMNs, which we consider to be the most crucial point in deciding the patients’ treatment. www.ajsp.com |
515
Miyabe et al
Am J Surg Pathol
Volume 39, Number 4, April 2015
FIGURE 2. A, A FISH image using a probe against centromeric regions of chromosome 8 (case #4) shows a decreased number of red signals, indicating monosomy. B, A FISH image using a probe against chromosome 6 (case #5) shows an increased number of red signals, indicating polysomy. C, A FISH image using a probe for 17p13 (red) and a centromeric region of chromosome 17 (green) shows 1 red and 2 green signals (case #11), indicating hemizygous p53 deletion (arrows). D, A FISH image using a probe for 9p21 (red) and 9q21 (green) shows no red and 2 green signals (case #15), indicating homozygous p16 deletion (arrows).
The most important finding of this study is that we defined the locus and chromosome abnormalities that may be responsible for the advancement of IPMN. A schematic diagram of the identified abnormalities is depicted in Figure 3. Although none of the non-neoplastic spots showed these abnormalities, polysomy 6 and p16 deletion were found more frequently in noninvasive IPMNs. We also showed that polysomies 7 and 18 and p16 and p53 deletions were more frequent in invasive than
516 | www.ajsp.com
in noninvasive IPMNs. There would be a direct pathway from normal pancreatic gland to invasive IPMN without passing through a noninvasive IPMN condition. Unfortunately, we could not investigate such a pathway, as all invasive IPMNs studied had noninvasive IPMN lesions concurrently. Genetic abnormalities responsible for the transition from non-neoplastic to noninvasive lesions seem to be heterogenous, and there may be no common or major abnormalities in this process. In contrast, in the Copyright
r
2014 Wolters Kluwer Health, Inc. All rights reserved.
Am J Surg Pathol
Volume 39, Number 4, April 2015
FISH Analysis in Intraductal Papillary Mucinous Neoplasm
TABLE 2. Patterns of Locus/Chromosome Aberrations Non-neoplastic Spot Case No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
Noninvasive Spot
Invasive Spot
FISH
IHC
Dysplasia
FISH
IHC
FISH
IHC
— — — — — — — — — — — — — — — — — — — — — — — — — — — —
— P16 P16 , P53+ P53+ — P16 — P53+ P16 P53+ P53+ P16 P16 , P53+ — P16 — P16 — — P16 — P53+ P53+ P53+ P53+ — — —
High High High Int. Int. Int. Int. Low Low Low Low High High High Int. Int. Int. Int. Int. Int. Int. Int. Int. Low Low Low Low Low
p53 hemi 6+ — p16 hemi, p53 hemi 6+ 6+ — — — — — 3+, 6+, 8+ 6+, 17+ 6+ 6+, p16 homo 3+, 18+ 7+ 8+ 3+ p16 hemi p16 hemi — — — — — — —
— P16 P53+ P53+ P16 , P53+ P16 , P53+ P16 P16 , P53+ P16 P53+ P16 P53+ P16 , P53+ P16 , P53+ P16 , P53+ — P16 , P53+ P53+ — — — P53+ P53+ P53+ P53+ P16 — —
8 , p16 hemi, p53 hemi 6+, 7+, 18+ 7+ 8, 17+, p16 hemi, p53 hemi 6+, 17+, 18+, p53 hemi 3+, 6+, 18+, p16 hemi, p53 hemi — 3+, 6+, 7+, 17+, 18+, p53 hemi 18+, p16 hemi, p53 hemi p16 homo, p53 hemi 6+, 7+, p16 homo, p53 hemi
— P16 , P53+ P53+ P53+ P16 , P53+ P16 , P53+ P53+ P16 , P53+ P16 , P53+ P16 , P53+ P16 , P53+
The results on the gray background show data of cases with invasive IPMN, and those on the white background show data of cases with noninvasive IPMN. Minus and plus signs after a number in FISH columns indicate monosomy and polysomy, respectively. Hemi indicates hemizygous deletion; high, high-grade dysplasia; homo, homozygous deletion; int., intermediate-grade dysplasia; low, low-grade dysplasia; P16 , loss of P16 IHC expression; P53, P53 IHC overexpression.
TABLE 3. Comparison of FISH and IHC Results for IPMN Development and Progression Locus/Chromosome Aberration FISH outcomes Polysomy 3 Polysomy 6 Polysomy 7 Polysomy 8 Polysomy 17 Polysomy 18 Monosomy 3 Monosomy 6 Monosomy 7 Monosomy 8 Monosomy 17 Monosomy 18 p16 deletion Hemi./homo. p53 deletion Hemi./homo. IHC outcomes Loss of P16 P53 overexpression
Non-neoplastic Spot (n = 28)
Noninvasive Spot (n = 28)
0 0 0 0 0 0 0 0 0 0 0 0 0 0/0 0 0/0
3 7 1 2 1 1 0 0 0 0 0 0 4 3/1 2 2/0
NS 0.0082 NS NS NS NS — — — — — — 0.0455
19 10
17 16
NS 0.0339
P
NS
Noninvasive Spot (n = 11)
Invasive Spot (n = 11)
0 3 0 0 0 0 0 0 0 0 0 0 1 1/0 2 2/0
2 5 4 0 3 5 0 0 0 2 0 0 6 4/2 8 8/0
NS NS 0.0455 — NS 0.0253 — — — NS — — 0.0253
4 6
4 10
NS 0.0455
P
0.0143
Values represent the number of spots. The McNemar test was used for comparison of 2 spots in the same case. Hemi. indicates hemizygous deletion; homo., homozygous deletion; NS, not significant.
Copyright
r
2014 Wolters Kluwer Health, Inc. All rights reserved.
www.ajsp.com |
517
Miyabe et al
Am J Surg Pathol
Volume 39, Number 4, April 2015
TABLE 4. The Relationship Between FISH/IHC Results and Clinicopathologic Parameters Clinicopathologic Features
Polysomy 6
Polysomy 7
P
P
n
Clinical factor Tumor size (mm) < 30 15 Z30 13 MPD size (mm) < 10 18 Z10 10 Mural nodule None/nonenhanced 14 Enhanced 14 Lymph node metastasis Yes 24 No 4 Pathologic factor Histologic types Gastric type 18 Intestinal type 8 Dysplasia* Low/intermediate 22 High 6 Vascular invasion Yes 5 No 23 Perineural invasion Yes 7 No 21
Polysomy 18
p16 Deletion
p53 Deletion
P
P
P
P53 IHC P
3 (20) 6 (46.2)
NS
3 (20) 2 (15.4)
NS
2 (13.3) 4 (30.8)
NS
5 (33.3) 4 (19.0)
NS
3 (20) 5 (38.5)
NS
7 (46.7) 13 (100)
0.0025
3 (16.7) 6 (60.0)
0.0346
1 (5.6) 4 (15.4)
NS
4 (22.2) 2 (20.0)
NS
7 (38.9) 2 (20.0)
NS
7 (38.9) 1 (10.0)
NS
11 (61.1) 9 (90.0)
NS
3 (21.4) 6 (42.9)
NS
0 (0) 5 (35.7)
0.0407
2 (14.3) 4 (28.8)
NS
4 (28.8) 5 (35.7)
NS
2 (14.3) 6 (42.9)
NS
7 (50.0) 13 (92.9)
0.0329
6 (25.0) 3 (75.0)
NS
3 (12.5) 2 (50)
NS
4 (16.7) 2 (50.0)
NS
6 (25.0) 3 (75.0)
NS
4 (100) 4 (16.7)
0.0034
16 (66.7) 4 (100)
NS
3 (16.7) 4 (50)
NS
1 (5.6) 3 (37.5)
NS
3 (16.7) 1 (12.5)
NS
7 (39.9) 1 (12.5)
NS
6 (33.3) 0 (0)
NS
11 (61.1) 7 (87.5)
NS
3 (13.6) 4 (66.7)
0.0207
1 (4.5) 0 (0)
NS
1 (4.5) 0 (0)
NS
4 (18.2) 0 (0)
NS
1 (4.5) 1 (16.7)
NS
12 (54.5) 4 (66.7)
NS
2 (40.0) 7 (30.4)
NS
2 (40) 3 (13)
NS
2 (40.0) 4 (17.4)
NS
4 (80.0) 5 (21.7)
0.0256
5 (100) 3 (13.0)
0.0006
4 (80) 16 (69.6)
NS
4 (57.1) 5 (23.8)
NS
2 (28.6) 3 (14.3)
NS
4 (57.1) 2 (9.5)
0.0207
5 (57.1) 4 (9.5)
0.0196
7 (100) 1 (4.8)
< 0.0001
6 (85.7) 14 (66.7)
NS
Data represented as the number of cases (percentages among each parameter). *Noninvasive spots in low-grade plus intermediate-grade dysplasia were compared with those in high-grade dysplasia. NS indicates not significant; p16 hemi, p16 hemizygous deletions; p53 hemi, p53 hemizygous deletions.
transition from noninvasive to invasive lesions, p53 gene deletion was found in 73% of the invasive lesions. As a major genetic abnormality, p53 gene deletion is considered to play an important role in adenomatous to malignant transition. In this study, we showed that polysomies 6, 7, and 18 were involved in oncogenesis or malignant transformation in IPMN. To the best of our knowledge, involvement of these abnormalities has not been pointed out previously. The precise oncogenetic mechanism of
these polysomies remains to be elucidated. However, their significance in transformation may be validated by their significant association with clinicopathologic factors: polysomy 6 was associated with MPD dilatation, the presence of mural nodules, and high-grade dysplasia in noninvasive spots; polysomy 7 with enhanced solid component; and polysomy 18 with perineural invasion of the tumor. p16/CDKN2A is a well-known tumorsuppressor gene. High prevalence of p16 inactivation (loss of heterozygosity in the 9p allele) has been described
TABLE 5. Outcomes for the Diagnosis of Invasive IPMN in the Present Study Diagnostic Criteria The criteria based on FISH or IHC data Polysomy 7 Polysomy 18 p16 deletion p53 deletion P53 IHC positive Polysomy 7 or polysomy 18 Polysomy 7 or p16 deletion Polysomy 7 or p53 deletion Polysomy 7 or P53 IHC positive Polysomy 18 or p16 deletion Polysomy 18 or p53 deletion Polysomy 18 or P53 IHC positive p16 deletion or p53 deletion p16 deletion or P53 IHC positive p53 deletion or P53 IHC positive
518 | www.ajsp.com
Sensitivity (%)
Specificity (%)
Accuracy (%)
36.4 45.5 54.5 72.7 90.9 63.6 81.8 90.9 90.9 81.8 81.8 45.5 72.7 100 100
94.1 94.1 82.4 100 57.1 88.2 76.5 94.1 41.2 76.5 94.1 100 82.4 29.4 41.2
65.2 69.8 68.5 86.4 74.0 75.9 79.2 92.5 66.1 79.2 88.0 72.8 77.6 64.7 70.6
Copyright
r
2014 Wolters Kluwer Health, Inc. All rights reserved.
Am J Surg Pathol
Volume 39, Number 4, April 2015
FIGURE 3. The schematic diagram of locus and chromosome abnormalities in this study. Polysomy 6 and p16 deletion are associated with the transition from non-neoplastic epithelium to noninvasive IPMN epithelium, and polysomies 7 and 18 and p16 and p53 deletions are associated with progression from the noninvasive to invasive IPMN epithelium.
in invasive IPMN.19 We demonstrated that p16 inactivation was associated with both the transition from non-neoplastic to noninvasive lesions and the transition from noninvasive to invasive lesions. p53 is another important tumor-suppressor gene. This molecule was found to play a major role in progression from noninvasive to invasive IPMNs and to be a potent diagnostic marker of invasive IPMNs. These findings are supported by previous microsatellite polymerase chain reaction studies that detected loss of heterogeneity of p53 in invasive IPMNs but not in noninvasive IPMNs.14 Regarding numeric abnormalities in chromosomes in IPMN, a pioneer study was reported by Soldini et al.8 Using 12 IPMN cases, they described that monosomy 6 may represent an early event necessary for neoplastic transformation of the pancreatic duct epithelium and that monosomy 18 may be associated with the progression from borderline to malignant IPMN. In the present study, some polysomies did emerge as genetic factors associated with the development or progression of IPMN; however, no monosomy was implicated. The discrepancies between the present study and that of Soldini and colleagues are difficult to explain. However, one plausible explanation is the difference in the cutoff values defining polysomies and monosomies in the respective studies. In our study, each upper cutoff value was calculated as the mean+3 SD after counting >200 nuclei of normal duct epithelium (18 cases). The cutoff values thus obtained ranged between 10% and 12% and 50% and 59% for polysomies and monosomies, respectively. In contrast, in the study by Soldini, et al,8 cutoff values were set as >20% and >40% for polysomies and monosomies, respectively, which were calculated from the mean+2 SD. Copyright
r
2014 Wolters Kluwer Health, Inc. All rights reserved.
FISH Analysis in Intraductal Papillary Mucinous Neoplasm
Another important finding of the present study is that FISH detection of p53 gene deletion and polysomy 7 may provide a high diagnostic accuracy for identification of invasive IPMNs. The International Consensus Guidelines for the Management of pancreatic IPMN were revised in 2012,4 and these guidelines have been used worldwide. However, there may be some issues that remain to be resolved: the levels of evidence for all items addressed in the guidelines are low, and the positive predictive value for high-grade dysplasia/invasive carcinoma is not very high, reportedly 62.5% in 1 study7 necessitating more accurate markers. Cytology might be the most useful test for the detection of malignancy. However, in addition to the typical impediments to cytologic interpretation of scant cellularity and poor cellular preservation, there is a lack of standardized, reproducible, diagnostic criteria for IPMNs.23 As suggested in this study, locus and chromosome abnormalities detected using the FISH technique may be a promising ancillary test for the diagnosis of invasive IPMN lesions. As shown in Table 5, detection of both p53 deletion and polysomy 7 provided a high diagnostic accuracy of 92.5%. The usefulness of FISH assays for the diagnosis of invasive IPMN should be further validated in prospective large-scale studies. Regarding p53 and p16, IHC results did not correlate with their deletion statuses by FISH analysis. It is difficult to explain these discrepancies. Plausible explanations may include the involvement of other molecular mechanisms such as gene mutation, aberrant methylation, and transcriptional and posttranscriptional factors.24,25 Cytologic examinations are sometimes performed for IPMN cases suspected of malignancy; however, its diagnostic accuracy has not been so high, ranging from 74% to 86%.26–28 In addition, stromal invasion of the tumor is difficult to diagnose. The diagnostic accuracy of IPMN cytology is expected to be improved using additional FISH analysis for p53 locus and polysomy 7, as has been shown in this study. Some researchers have mentioned that the diagnostic significance of trisomy 7 for malignancy remains uncertain. Fritcher et al15 performed a FISH study for malignant pancreatic and/or biliary strictures and described that trisomy 7 was occasionally detected in non-neoplastic and neoplastic tissues. In the present study, however, none of non-neoplastic spots had trisomy 7 or polysomy 7. This result is supported by a study of Kubiliun et al18 describing that none of the patients with benign disease had trisomy 7. In conclusion, our study demonstrated that some specific locus and chromosome abnormalities may be important in the advancement from normal gland to noninvasive IPMN and from noninvasive to invasive IPMN. In addition, FISH may be a useful modality for differentiating invasive from noninvasive IPMNs and should be evaluated by using cytologic specimen in clinical practice. These findings provide insights into the oncogenesis and molecular diagnosis of IPMN.
REFERENCES 1. Hruban RH, Pitman MB, Klimstra DS. Intraductal neoplasms. In: Silverberg SG, Sobin LH, eds. AFIP Atlas of Tumor Pathology
www.ajsp.com |
519
Miyabe et al
2.
3. 4. 5. 6.
7.
8.
9. 10. 11. 12. 13.
14. 15.
Series 4. Washington, DC: American Registry of Pathology; 2007: 75–110. Adsay NV, Fukushima N, Furukawa T, et al. Intraductal neoplasms of the pancreas. In: Bosman FT, Carneiro F, Hruban RH, et al, eds. WHO Classification of Tumours of the Digestive System. Lyon: IARC; 2010:304–313. Hibi Y, Fukushima N, Tsuchida A, et al. Pancreatic juice cytology and subclassification of intraductal papillary mucinous neoplasms of the pancreas. Pancreas. 2007;34:197–204. Tanaka M, Fernandez-del Castillo C, Adsay V, et al. International consensus guidelines 2012 for the management of IPMN and MCN of the pancreas. Pancreatology. 2012;12:183–197. Jang JY, Park T, Lee S, et al. Validation of international consensus guidelines for the resection of branch duct-type intraductal papillary mucinous neoplasms. Br J Surg. 2014;101:686–692. Amato E, Molin MD, Mafficini A, et al. Targeted next-generation sequencing of cancer genes dissects the molecular profiles of intraductal papillary neoplasms of the pancreas. J Pathol. 2014;233:217–227. Goh BK, Thng CH, Tan DM, et al. Evaluation of the Sendai and 2012 International Consensus Guidelines based on cross-sectional imaging findings performed for the initial triage of mucinous cystic lesions of the pancreas: a single institution experience with 114 surgically treated patients. Am J Surg. 2014;208:202–209. Soldini D, Gugger M, Burckhardt E, et al. Progressive genomic alterations in intraductal papillary mucinous tumours of the pancreas and morphologically similar lesions of the pancreatic ducts. J Pathol. 2003;199:453–461. Bova RJ, Quinn DI, Nankervis JS, et al. Cyclin D1 and p16INK4A expression predict reduced survival in carcinoma of the anterior tongue. Clin Cancer Res. 1999;5:2810–2819. Wilentz RE, Geradts J, Maynard R, et al. Inactivation of the p16 (INK4A) tumor-suppressor gene in pancreatic duct lesions: loss of intranuclear expression. Cancer Res. 1998;58:4740–4744. Noda H, Okumura Y, Nakayama T, et al. Clinicopathological significance of MAML2 gene split in mucoepidermoid carcinoma. Cancer Sci. 2013;104:85–92. Nomoto J, Hiramoto N, Kato M, et al. Deletion of the TNFAIP3/ A20 gene detected by FICTION analysis in classical Hodgkin lymphoma. BMC Cancer. 2012;12:457. Tanaka M, Chari S, Adsay V, et al. International consensus guidelines for management of intraductal papillary mucinous neoplasms and mucinous cystic neoplasms of the pancreas. Pancreatology. 2006;6:17–32. Wada K. p16 and p53 gene alterations and accumulations in the malignant evolution of intraductal papillary-mucinous tumors of the pancreas. J Hepatobiliary Pancreat Surg. 2002;9:76–85. Fritcher EG, Kipp BR, Halling KC, et al. A multivariable model using advanced cytologic methods for the evaluation of indeter-
520 | www.ajsp.com
Am J Surg Pathol
16. 17.
18.
19. 20.
21.
22.
23. 24. 25.
26.
27. 28.
Volume 39, Number 4, April 2015
minate pancreatobiliary strictures. Gastroenterology. 2009;136: 2180–2186. Schleicher C, Poremba C, Wolters H, et al. Gain of chromosome 8q: a potential prognostic marker in resectable adenocarcinoma of the pancreas? Ann Surg Oncol. 2007;14:1327–1335. Levy MJ, Oberg TN, Campion MB, et al. Comparison of methods to detect neoplasia in patients undergoing endoscopic ultrasoundguided fine-needle aspiration. Gastroenterology. 2012;142:1112– 1121. e1112. Kubiliun N, Ribeiro A, Fan YS, et al. EUS-FNA with rescue fluorescence in situ hybridization for the diagnosis of pancreatic carcinoma in patients with inconclusive on-site cytopathology results. Gastrointest Endosc. 2011;74:541–547. Fujii H, Inagaki M, Kasai S, et al. Genetic progression and heterogeneity in intraductal papillary-mucinous neoplasms of the pancreas. Am J Pathol. 1997;151:1447–1454. Biankin AV, Biankin SA, Kench JG, et al. Aberrant p16(INK4A) and DPC4/Smad4 expression in intraductal papillary mucinous tumours of the pancreas is associated with invasive ductal adenocarcinoma. Gut. 2002;50:861–868. Biankin AV, Kench JG, Biankin SA, et al. Pancreatic intraepithelial neoplasia in association with intraductal papillary mucinous neoplasms of the pancreas: implications for disease progression and recurrence. Am J Surg Pathol. 2004;28:1184–1192. Abe K, Suda K, Arakawa A, et al. Different patterns of p16INK4A and p53 protein expressions in intraductal papillary-mucinous neoplasms and pancreatic intraepithelial neoplasia. Pancreas. 2007;34:85–91. Pitman MB. Revised international consensus guidelines for the management of patients with mucinous cysts. Cancer Cytopathol. 2012;120:361–365. Ribeiro A, Peng J, Casas C, et al. EUS FNA with FISH analysis in 104 patients with pancreatic mass. J Gastroenterol Hepatol. 2014;29:1654–1658. Moore PS, Orlandini S, Zamboni G, et al. Pancreatic tumours: molecular pathways implicated in ductal cancer are involved in ampullary but not in exocrine nonductal or endocrine tumorigenesis. Br J Cancer. 2001;84:253–262. Ono J, Yaeger KA, Genevay M, et al. Cytological analysis of small branch-duct intraductal papillary mucinous neoplasms provides a more accurate risk assessment of malignancy than symptoms. CytoJournal. 2011;8:21. Cizginer S, Turner BG, Bilge AR, et al. Cyst fluid carcinoembryonic antigen is an accurate diagnostic marker of pancreatic mucinous cysts. Pancreas. 2011;40:1024–1028. Pais SA, Attasaranya S, Leblanc JK, et al. Role of endoscopic ultrasound in the diagnosis of intraductal papillary mucinous neoplasms: correlation with surgical histopathology. Clin Gastroenterol Hepatol. 2007;5:489–495.
Copyright
r
2014 Wolters Kluwer Health, Inc. All rights reserved.
