ELSEVIER
Unusual DNA mismatch repair-deficient tumors in Lynch syndrome: a report of new cases and review of the literature
Yevgeniy Karamurzin MDª, Zhaoshi Zeng MDb, Zsofia K. Stadler MDC, Liying Zhang MDª, Ihsane Ouansafi MDa, Hikmat A. Al-Ahmadie MDa, Christine Sempoux MDd, Leonard B. Saltz MDC, Robert A. Soslow MDa, Eileen M. O’Reilly MDC, Philip B. Paty MDb, Daniel G. Coit MDb, Jinru Shia MDa,*, David S. Klimstra MDª
aDepartment of Pathology, Memorial Sloan-Kettering Cancer Center, New York, NY 10065
Department of Surgery, Memorial Sloan-Kettering Cancer Center, New York, NY 10065
“Department of Medicine, Memorial Sloan-Kettering Cancer Center, New York, NY 10065
Department of Pathology, School of Medicine, Université Catholique de Louvain, Cliniques Universitaires Saint-Luc, B-1200 Brussels, Belgium
Received 30 September 2011; revised 9 December 2011; accepted 14 December 2011
Keywords:
Microsatellite instability; Immunohistochemistry; Hereditary nonpolyposis colorectal cancer
Summary Immunohistochemical detection of DNA mismatch repair proteins and polymerase chain reaction detection of microsatellite instability have enhanced the recognition of mismatch repair-deficient neoplasms in patients with Lynch syndrome and, consequently, led to the identification of tumors that have not been included in the currently known Lynch syndrome tumor spectrum. Here, we report 4 such unusual tumors. Three of the 4, a peritoneal mesothelioma, a pancreatic acinar cell carcinoma, and a pancreatic well-differentiated neuroendocrine tumor, represented tumor types that, to the best of our knowledge, have not been previously reported in Lynch syndrome. The fourth tumor was an adrenocortical carcinoma, which has rarely been reported previously in Lynch syndrome. Three of our 4 patients carried a pathogenic germ-line mutation in a mismatch repair gene. The unusual tumor in each of the 3 patients showed loss of the mismatch repair protein corresponding to the mutation. The fourth patient did not have mutation information but had a history of colonic and endometrial carcinomas; both lacked MSH2 and MSH6 proteins. Interestingly, none of the 4 unusual tumors revealed microsatellite instability on polymerase chain reaction testing, whereas an appendiceal carcinoma from 1 of the study patients who was tested simultaneously did. The recognition of such tumors expands the repertoire of usable test samples for the workup of high-risk families. As yet, however, there are no data to support the inclusion of these tumors into general screening guidelines for detecting Lynch syndrome, nor are there data to warrant surveillance for these tumors in patients with Lynch syndrome. @ 2012 Elsevier Inc. All rights reserved.
# Disclosures: The authors have no conflicts of interest or funding to disclose.
** Christine Sempoux was supported by the Foundation Saint Luc, Brussels, Belgium.
* Corresponding author.
E-mail address: Shiaj@MSKCC.org (J. Shia).
1. Introduction
Over the last 2 decades, advances in our understanding of colorectal carcinogenesis have refined the definitions of familial colorectal cancer syndromes. The Amsterdam criteria, first developed in 1991, defined the clinical condition known as hereditary nonpolyposis colorectal cancer syndrome (HNPCC) [1,2]. More recently, it has been recognized that a germ-line mutation in 1 of a set of DNA mismatch repair (MMR) genes-MLH1, MSH2, MSH6, and PMS2 [3]-accounts for a significant proportion of patients with clinical HNPCC as well as some patients with colorectal cancer without a well-defined family history [4]. The term Lynch syndrome (LS), originally put forth by Boland and Troncale [5] in the 1980s, has now been chosen for the subset of patients with cancer and families that have an underlying pathogenic germ-line mutation in a DNA MMR gene, irrespective of family history [6,7].
Tumors that develop in patients with LS typically involve certain characteristic organs and have specific morphologic and molecular phenotypes [3,8,9]. Morpho- logically, as exemplified by colorectal and endometrial carcinomas, LS-associated tumors tend to have a solid growth pattern and/or increased lymphocytic infiltration. Molecularly, they show high-frequency microsatellite instability (MSI-H). Furthermore, the tumors show loss of the affected MMR proteins that can be detected by immunohistochemistry (IHC). These tumor characteristics have significantly enhanced our ability to detect LS, and MSI testing and MMR IHC techniques have been incorporated into routine practice in many institutions.
With the increasingly wider application of screening tests, tumors that have not yet been included in the conventional LS tumor spectrum are being increasingly discovered in patients with this syndrome. Conventionally, the extracolo- rectal LS tumor spectrum encompassed carcinomas of the endometrium, ovary, small bowel, stomach, hepatobiliary system, central nervous system, renal pelvis, and skin [3,10]. In recent years, tumors outside this spectrum have been discovered in patients with LS; examples include sarcomas (such as malignant fibrous histiocytoma [11,12], leiomyo- sarcoma [13-15], and liposarcoma [15]), melanoma [16], and epithelial malignancies of organs such as the breast [17], prostate [18], lung [19], thyroid [20], and adrenal cortex [20- 22]. All such tumors have been shown to be MMR deficient.
The occurrence of such unusual tumors in patients with LS bears practical implications because it suggests that these unusual tumors may also be used as test samples in the workup of families at high risk for LS. At the same time, these tumors pose many new and important questions: for example, how often do these tumors occur in patients with LS; is the risk of developing these tumors increased in LS individuals such that surveillance is warranted; do they occur in the sporadic setting; and what would the MMR deficiency imply in terms of both tumor pathogenesis and clinical behavior ?. In this report, we describe 4 unusual MMR-
deficient tumor entities occurring in patients with LS, 1 previously reported and 3 as yet unreported, and we conduct a review of the relevant literature. Our aim is 2-fold: (1) to document the association of the 3 previously unreported tumor entities with LS and (2) to provide a discussion about the current knowledge regarding the relationship of these tumors with LS.
2. Materials and methods
2.1. Patient data
The study cases were identified from the databases of the Department of Pathology and the clinical genetics service of Memorial Sloan-Kettering Cancer Center (New York, NY) and the Department of Pathology, Cliniques Universitaires Saint-Luc, UCL (Brussels, Belgium). Clinical and pathologic information was retrieved from the hospital information systems. The study was approved by the institutional review board of Memorial Sloan-Kettering Cancer Center and the ethics committee of the Medical Faculty of the University of Louvain, and the study was performed in agreement with the Belgian law.
2.2. Histologic and immunohistochemical analysis
The histologic slides were reviewed, and the diagnoses were confirmed. The specific morphologic findings were recorded, including the presence of a diffuse, solid growth pattern, and the extent of lymphocytic infiltration. Immuno- histochemical staining for MLH1 (Ventana, Tucson, AZ; clone G168-728, prediluted, pretreatment with citrate buffer at pH 9.0), PMS2 (clone A16-4, 1:25 dilution, pretreatment with citrate buffer at pH 9.0; Ventana), MSH2 (Ventana; clone AB-2, prediluted, pretreatment with citrate buffer at pH 9.0), and MSH6 (Ventana; clone 44, prediluted, pretreatment with citrate buffer at pH 9.0) was performed on the unusual tumors of all the study patients. Additional immunohisto- chemical studies performed at the time of diagnosis as part of the routine diagnostic workup were reviewed.
2.3. Detection of tumor microsatellite instability
Formalin-fixed, paraffin-embedded (FFPE) tissue blocks containing tumor and nonneoplastic epithelial areas were sectioned at 10 um and mounted on regular histology slides. Both tumor and normal tissue were harvested by manual dissection and placed in 50 uL DNA digestion buffer. Each sample was further admixed with 5 uL 10% sodium dodecyl sulfate, 1 µL proteinase K (10 mg/mL), and 0.1 uL ß- mercaptoethanol (Sigma, St. Louis, MO) under the chemical fume hood. The sample tubes were incubated at 56℃ and with constant rotating agitation for 72 hours, followed by spike 2 uL proteinase K (10 mg/mL) digestion under the
same incubation parameters for another 24 hours. QIAmp DNA FFPE Tissue Kit (QIAGEN, Valencia, CA) was used for manual DNA extraction and purification. The Promega MSI Multiplex System Version 1.1 (Promega, Madison, WI) was used. This system consisted of 5 nearly monomorphic mononucleotide markers (BAT-25, BAT-26, NR-21, NR-24, and MONO-27) for MSI determination and 2 polymorphic pentanucleotide markers (Penta C and Penta D) for sample identification. The method has been shown to be superior compared with the National Cancer Institute (NCI)-recom- mended Bethesda MSI panel as being more robust in separating tumors with MSI-H and microsatellite stable (MSS) profiles [23]. MSI analysis was performed according to the manufacturer’s instructions (Promega). Fluorescently labeled polymerase chain reaction products were separated using the ABI 3730 Genetic Analyzer (CD Genomics, Shirley, NY). The data were analyzed using GeneMapper 3.7 Software (Applied Biosystems, Carlsbad, CA).
3. Results
3.1. Patient data
Four patients were identified.
Patient 1 was a 56-year-old man of European ancestry with a family history fulfilling the Amsterdam criteria II: 2 first-degree relatives diagnosed as having colon cancer at ages 46 and 56 years, respectively, and 1 second-degree relative diagnosed as having colon cancer at age 24 years. His personal history included a right-sided colon adenocar- cinoma at age 35 years and numerous skin and nasal sebaceous carcinomas. The patient had undergone genetic counseling, and mutation analysis revealed a 1456insT in MLH1, which, by linkage analysis, was shown to be deleterious. The patient now presented with a hypermeta- bolic perisplenic lesion on positron emission tomography scan, and a resection of this lesion was performed.
Patient 2 was a 65-year-old woman of Ashkenazi Jewish ancestry with a family history fulfilling the Amsterdam criteria II: 2 first-degree relatives diagnosed as having colon cancer at ages 35 and 46 years, respectively, and 1 second- degree relative diagnosed as having colon cancer at age 60 years. Additional family cancer history included breast cancer in several relatives: 3 maternal first cousins, 1 maternal second cousin, 1 maternal aunt, and maternal grandmother. The patient’s personal history included breast cancer at age 52 years and sebaceous carcinoma at age 60 years. The patient had undergone genetic counseling. Her sebaceous tumor lost MSH6 staining by IHC, and mutation analysis on her blood cells revealed 1312insA in MSH6. She now presented with chronic abdominal pain, which led to the discovery of a 4-cm pancreatic mass with multiple liver lesions. A biopsy of the liver lesions was obtained.
Patient 3 was a 48-year-old Portuguese woman with a family history fulfilling the revised Bethesda Guidelines: 1
first-degree relative (her mother) with both colonic and endometrial carcinomas, 3 second-degree relatives with colonic carcinoma, 2 diagnosed at age younger than 50 years, and 1 diagnosed under age 30 years. One of the 3 second-degree relatives also had an endometrial carcinoma. The patient’s personal history was significant for an endometrial carcinoma treated by radical hysterectomy at age 46 years and a colonic villous adenoma with high-grade dysplasia removed at age 47 years. She now presented with abdominal discomfort, which led to the discovery of a well- circumscribed 2.2-cm mass in the body of the pancreas, which was resected.
Patient 4 was a 29-year-old Ashkenazi Jewish man with a family history significant for his grandfather with colon cancer diagnosed at age 59 years and his father found to have a few “colon polyps” at age younger than 60 years. The father had undergone genetic workup and was found to have a germ-line mutation, MSH2*1906G > C (A636P). The patient had no significant personal history until age 29 years when he presented with persistent right flank pain and was found to have a large right retroperitoneal tumor with central necrosis displacing the right kidney. The mass was then removed via an en bloc resection including right adrenalectomy, nephrectomy, and retroperitoneal lymph node dissection. Four months later, the patient presented with symptoms of acute appendicitis, and a subsequent appendectomy with follow-up right colectomy revealed a pT1N0 moderately differentiated appendiceal adenocarcinoma. The patient underwent genetic counseling, and mutation analysis revealed the MSH2*1906G>C (A636P) as seen in the patient’s father. The material available for this study consisted of the en bloc resection of the patient’s retroperitoneal mass and the appendiceal tumor sample.
3.2. Pathologic findings of the unusual tumors
The unusual tumors detected in the 4 study patients are outlined in Table 1 and illustrated in Figs. 1 to 4.
In case 1, the tumor showed morphologic and immuno- phenotypical features characteristic of epithelioid malignant mesothelioma, with tumor cells showing immunoreactivity to keratin AE1:AE3, calretinin, WT-1, and D2-40 (Fig. 1). Interestingly, the tumor showed a conspicuous stromal lymphocytic infiltration with formation of lymphoid aggre- gates. Scattered intratumoral lymphocytic infiltration was also noted.
In case 2, the liver biopsy material demonstrated a tumor with cytologic and immunophenotypical features diagnostic of pancreatic acinar cell carcinoma (Fig. 2). This tumor had a solid pattern of growth typical of acinar cell carcinoma. The cells were monotonous with a suggestion of basal nuclear polarization and cytoplasmic granularity. There were prominent nucleoli in the tumor cells. There was no apparent lymphocytic infiltration in the tumor. Immunohistochemically, the tumor cells were positive for
| Reference | Sex | Age (y) | Unusual tumor | Germ-line mutation | ||
|---|---|---|---|---|---|---|
| Diagnosis | Lost expression by IHC | MSI status | ||||
| Multiple [17] | NA | NA | Breast carcinoma | Variable | MSI-H | Variable |
| Berends et al [21] | F | NA | Adrenal cortical carcinoma | None | MSS | MSH2 |
| Sijmons et al [12] | M | 45 | MFH, leg | MSH2 | MSI-H | MSH2 |
| Soravia et al [18] | M | 61 | Prostate carcinoma | MSH2 | MSI-H | MLH1 |
| Broaddus et al [20] | M | 34 | Adrenal cortical carcinoma | MSH2 | MSS | MSH2 |
| F | 39 | Anaplastic carcinoma of thyroid | MSH2 | MSI-L | MSH2 | |
| Ponti et al [16] | M | 43 | Malignant melanoma | MSH2 | MSI-H | MSH2 |
| Clyne et al [13], Yu et al [14] | M | 36 | Leiomyosarcoma, thigh | MLH1 | MSI-H | MLH1 |
| Nilbert et al [15 ] | M | 38 | Liposarcoma, thigh | MSH2/MSH6 | NA | MSH2 |
| M | 32 | Gliosarcoma, brain | MSH2/MSH6 | NA | MSH2 | |
| F | 44 | Leiomyosarcoma, uterus | MSH2/MSH6 | MSI-H | NA | |
| Nolan et al [19] | M | 64 | Lung adenocarcinoma | MSH2 | MSI-H | MSH2 |
| Petersen et al [24] | M | 41 | Pancreatic serous cystadenoma | MLH1 | MSI-H | MLH1 |
| Brieger et al [11] | M | 43 | MFH, left psoas muscle | MSH2 | MSI-H | MSH2 c.2038C ≥ T |
| M | 50 | MFH, Right upper leg | MSH2 | MSI-H | MSH2 c.942 ± 3A ≥T | |
| Medina-Arana et al [22] | F | 60 | Adrenal cortical carcinoma | MSH2 | MSS | MSH2 |
| Our cases | ||||||
| 1 | M | 56 | Peritoneal mesothelioma | MLH1/PMS2 | MSS | MLH1*1456insT |
| 2 | F | 65 | Acinar cell carcinoma of pancreas | MSH6 | MSS | MSH6*1312insA |
| 3 | F | 50 | Pancreatic endocrine neoplasm | MSH2/MSH6 | MSS | NA |
| 4 | M | 29 | Adrenal cortical carcinoma | MSH2/MSH6 | MSS | MSH2*1906GC |
Abbreviations: MFH, malignant fibrous histiocytoma; NA, not available; F, female; M, male.
trypsin and chymotrypsin and negative for synaptophysin and chromogranin.
In case 3, the pancreatic resection showed a low-grade epithelial tumor with features characteristic of pancreatic well-differentiated neuroendocrine tumor (Fig. 3). There was no mitotic activity, and lymphocytic infiltration was not evident. By IHC, the cells were diffusely positive for chromogranin and synaptophysin, and negative for trypsin and chymotrypsin.
In case 4, the sections from the retroperitoneal mass demonstrated an epithelioid neoplasm originating from the adrenal gland. The tumor showed a solid pattern of growth and oncocytic cytologic features (Fig. 4). The mitotic count reached 20 per 50 high-power fields. Scattered intratumoral lymphocytes were noted. Immunohistochemically, the tumor cells were positive for melan-A, inhibin, and synaptophysin, and negative for chromogranin, HepPar-1, and TTF1. There was focal positivity for AE1:AE3. These findings led to a diagnosis of adrenal cortical carcinoma.
3.3. Evaluation of DNA MMR proteins and microsatellite instability status
The unusual tumors from all 4 patients demonstrated immunohistochemical evidence of MMR protein abnormal-
ities as shown in Table 1 and illustrated in Figs. 1 to 4. Cases 1 and 4 had deleterious germ-line mutations in MLH1 and MSH2, respectively; their tumors lost not only the protein encoded by the mutated gene but also the partner protein, that is, loss of both MLH1 and PMS2 in the case of MLH1 mutation and loss of both MSH2 and MSH6 in the case of MSH2 mutation. In addition, the appendiceal carcinoma in case 4 also showed complete loss of MSH2 and MSH6 in the tumor. In case 2, the pathogenic mutation occurred in MSH6, and on IHC, there was only isolated loss of MSH6 protein. In case 3, the germ-line mutation status was unknown; however, the patient’s prior endometrial carcinoma tested negative for MSH2 and MSH6 staining by IHC, and the current pancreatic endocrine tumor also lost these 2 proteins, whereas the staining for MLH1 and PMS2 was retained.
Polymerase chain reaction testing for microsatellite instability using 5 mononucleotide markers revealed stable microsatellites in the unusual tumors from all 4 patients. The appendiceal carcinoma from patient 4 was tested simulta- neously, and it showed high-frequency MSI. The results are outlined in Table 1, and examples of electropherograms (the mesothelioma showing lack of instability and the appendi- ceal tumor from case 4 showing high-frequency instability) are illustrated in Fig. 5.
A
B
C
D
E
F
4. Discussion
A literature search revealed a variety of unusual tumors that show IHC or molecular evidence of MMR deficiency and occur in individuals that carry a germ-line pathogenic MMR gene mutation (Table 1) [11-13,15-22,24]. These entities include malignant fibrous histiocytoma, leiomyosar-
coma of the extremity or uterus, liposarcoma, malignant melanoma, breast carcinoma, prostatic carcinoma, lung adenocarcinoma, anaplastic carcinoma of the thyroid, adrenal cortical carcinoma, and pancreatic serous cystade- noma. Our report added 3 more tumors to this list: peritoneal mesothelioma, pancreatic acinar cell carcinoma, and pancre- atic well-differentiated neuroendocrine tumor.
A
B
C
D
E
F
In our series, all the unusual tumors were of epithelial origin. The histologic diagnosis in each case was established by the characteristic tumor morphology as well as lineage or organ-specific immunohistochemical profiles. It is to be noted that our case of pancreatic acinar cell carcinoma was also included in a clinical analysis of a series of 40 pancreatic acinar cell carcinomas that we performed simultaneously.
The emphasis of our clinical analysis was on our experience of treating acinar cell tumors over the last 10 years [25]. Among the 40 cases we studied, this case represented the only one that was associated with the clinical condition of HNPCC syndrome.
Our cases, along with those reported in the literature and illustrated in Table 1, serve to expand the tumor sample
A
B
C
D
E
F
repertoire for the workup of high-risk family members susceptible for LS. In our case 4, for example, the patient was deemed to be at high risk for LS because his father carried a pathogenic germ-line MSH2 mutation. Our patient’s adrenal cortical carcinoma, the first malignancy he developed, served as a test sample for the genetic workup, and the abnormal MSH2/MSH6 IHC results led to further
germ-line mutation analysis that indeed identified the same MSH2 mutation his father carried. It is to be pointed out that, similar to colon adenomas, normal MMR protein expression in such unusual tumors does not exclude the possibility of LS. Another scenario in which knowledge of such tumors can be helpful is a family deemed to be at high risk for LS by clinical history of multiple cancers, but the
A
B
C
D
E
F
relatives with more conventional tumors are not available for testing. In such a scenario, MMR IHC testing of these unusual tumors may reveal whether MMR is abnormal and which gene or genes are likely affected.
The observation of these unusual MMR-deficient tumors in LS also raises a variety of new issues. One consideration regards the incidence of such unusual tumors in patients
with LS and whether the abnormal MMR protein and/or function in these tumors is a secondary event or causally related to the patient’s germ-line MMR defect. Answers to such questions will then provide guidance as to whether these unusual tumor types should be surveyed for in LS or whether they should be tested for MMR deficiency for the purpose of identifying LS.
00
-
.
…
-
3
NR-24
Panta C
BAT-25
NR-21
BAT-26
MONO-27
(T)
7
Panta D
:-
MSS
0
1
9
2
A
(N)
-
-
1833
.
A
A
ML
M
A
A
A
a
AN
N
190
…
1%
170
Panta C
Panta D
-
BAT-25
(T)
NR-21
BAT-26
NR-24
7
-
MONO-27
1833
cees
MSI-H
0
A
A
MSI-H
MSI-H
MSI-H MSI-H
MSS
-
-
-
-
-
-
-
-
(N)
.
-
-
.
A
1
M
N
1
N
A
A
At the current stage, studies that address the frequency of unusual tumors in LS are largely limited to breast cancer. After scattered case reports documenting the occurrence of MMR-deficient breast cancer in patients with LS, the frequency of breast cancer in familial cancer registries has been studied. Walsh et al [26] examined the Colon Cancer Family Registry for breast cancers from families with a history of early-onset (<50 years) colorectal cancer and breast cancer occurring at any age. Among 35 breast cancers identified in individuals who carried a pathogenic germ-line MMR gene mutation, the authors detected loss of MMR expression consistent with their respective germ-line mutations in 18 tumors, thus documenting a 51% frequency of MMR protein deficiency in breast cancers occurring in genetically defined LS individuals. A study by Jensen et al [27] using the Danish HNPCC registry similarly found MMR protein loss by IHC in 7 (44%) of 16 breast cancers
that occurred in individuals carrying a pathogenic germ-line MMR gene mutation. Thus, it seems that among high-risk individuals, particularly members of genetically defined LS families, the chance of a breast cancer being MMR deficient is significant, a finding further suggesting the potential use of breast tumor as test samples in the workup of high-risk families as described previously. As yet, however, the available data seem insufficient to determine an accurate frequency of MMR-deficient breast carcinomas in LS individuals in general. Data regarding the frequency of other unusual MMR-deficient tumors in LS are essentially lacking.
Similarly, whether such unusual MMR-deficient tumors in LS are merely coincidental occurrences or components of a genetic syndrome is to be clarified. Studies on breast cancers suggest that the lifetime risk of breast cancer is not increased in individuals with LS, despite the rather consistent
observation of relatively young age of breast cancer onset [10,28]. Such observations would argue that breast carcinoma (and potentially other unusual tumors as well) is not a component of the syndrome. It is possible that the MMR deficiency arises after transformation to carcinoma has already been initiated through other mechanisms. The lack of detectable MSI-H phenotype in some such tumors, as shown in our series and in the literature [20-22], may be a reflection of MMR deficiency being a secondary event. It has been suggested that the accumulation of detectable unstable microsatellites requires a sufficient amount of cell cycling [29], and in cases where MMR deficiency occurs after the cancer has already been initiated, it is conceivable that a tumor could have lost the MMR protein but not yet accumulated detectable microsatellite instability.
The specific molecular mechanisms underlying the MMR deficiency in these tumors notwithstanding, from a practical point of view, the lack of evidence of an increased lifetime risk fails to support the need for surveillance for these unusual tumors in LS. Conversely, patients diagnosed as having any of these tumors should not necessarily be routinely tested for MMR deficiency for the purpose of detecting LS. No data, thus far, provided evidence to support the inclusion of these unusual tumors into screening guidelines such as the Bethesda Guidelines for detecting LS. However, there may be specific circumstances when the decision to offer surveillance may differ. For example, age-specific cancer risk, cancer susceptibility in the family, and the availability of sensitive and specific screening tools may all alter the decision for surveillance [28].
Another question posed by the occurrence of such unusual MMR-deficient tumors in LS relates to whether and to what extent MMR deficiency affects the develop- ment and progression of the sporadic form of these tumor types. As illustrated in the case of colorectal cancer, hereditary tumors may serve as a shortcut to identify genes or pathways that are subsequently found to be important in sporadic cancers as well. It follows that some sporadic forms of these unusual tumor entities may have MMR deficiency as well. These subsets may be small, but one might hypothesize that because of their MMR defects, these subsets may share a specific biologic behavioral pattern that can be translated to specific clinical management. Indeed, there are already some data supporting such a hypothesis. In a study of 48 pancreatic neuroendocrine tumors, House et al [30] detected 5 MSI-H tumors, and all 5 were found to have MLH1 promoter hypermethylation, supporting that these MSI-H tumors are sporadic in nature. Furthermore, this study suggested that MSI-H tumors were associated with improved survival compared with MSI-negative tumors at 5 years. Similar data were presented by another study that focused on insulinomas [31]. A study of pancreatic acinar cell carcinoma found that the MSI-H phenotype (unstable in 4 of the 5 NCI microsatellite markers) was detected in 1 of 13 cases and MSI-low
(unstable in 1 of the 5 markers tested) in 2 cases [32]. The MSI-H case from this study was a 75-year-old male patient with no known family or personal cancer history. We performed MMR protein IHC on this patient’s tumor after the publication and found the tumor lacking both MLH1 and PMS2 proteins (unpublished data). These are obviously only preliminary observations, but they may serve to guide further research exploring subsets of MMR-deficient tumors across various organ systems.
In summary, this report describes 4 unusual MMR- deficient tumors in patients with LS, including 3 that have not been reported previously. These unusual entities can serve to expand the test sample repertoire in the workup of high-risk family members. As yet, however, there is no evidence to recommend general surveillance for any of the unusual tumors in LS or routine testing of these tumors for MMR deficiency for the purpose of detecting LS. A better understanding of the biologic and clinical implication of MMR deficiency in these tumors awaits further investigation.
Acknowledgment
The authors thank Ruben Bacares of the Diagnostic Molecular Genetics Laboratory at Memorial Sloan-Kettering Cancer Center for his technical assistance.
References
[1] Vasen HF, Mecklin JP, Khan PM, Lynch HT. The International Collaborative Group on Hereditary Non-Polyposis Colorectal Cancer (ICG-HNPCC). Dis Colon Rectum 1991;34:424-5.
[2] Park JG, Vasen HF, Park KJ, et al. Suspected hereditary nonpolyposis colorectal cancer: International Collaborative Group on Hereditary Non-Polyposis Colorectal Cancer (ICG-HNPCC) criteria and results of genetic diagnosis. Dis Colon Rectum 1999;42:710-5 [discussion 715-6].
[3] Umar A, Boland CR, Terdiman JP, et al. Revised Bethesda Guidelines for hereditary nonpolyposis colorectal cancer (Lynch syndrome) and microsatellite instability. J Natl Cancer Inst 2004;96:261-8.
[4] Lindor NM, Rabe K, Petersen GM, et al. Lower cancer incidence in Amsterdam-I criteria families without mismatch repair deficiency: familial colorectal cancer type X. JAMA 2005;293:1979-85.
[5] Boland CR, Troncale FJ. Familial colonic cancer without antecedent polyposis. Ann Intern Med 1984;100:700-1.
[6] Recommendations from the EGAPP Working Group: can UGT1A1 genotyping reduce morbidity and mortality in patients with metastatic colorectal cancer treated with irinotecan? Genet Med 2009;11:15-20.
[7] Lynch HT, Lynch PM, Lanspa SJ, Snyder CL, Lynch JF, Boland CR. Review of the Lynch syndrome: history, molecular genetics, screening, differential diagnosis, and medicolegal ramifications. Clin Genet 2009;76:1-18.
[8] Boland CR, Shike M. Report from the Jerusalem workshop on Lynch syndrome-hereditary nonpolyposis colorectal cancer. Gastroenterology 2010;138:2197.
[9] Shia J. Immunohistochemistry versus microsatellite instability testing for screening colorectal cancer patients at risk for hereditary nonpolyposis colorectal cancer syndrome. Part I. The utility of immunohistochemistry. J Mol Diagn 2008;10:293-300.
[10] Watson P, Vasen HF, Mecklin JP, et al. The risk of extra-colonic, extra- endometrial cancer in the Lynch syndrome. Int J Cancer 2008;123:444-9.
[11] Brieger A, Engels K, Schaefer D, et al. Malignant fibrous histiocytoma is a rare Lynch syndrome-associated tumor in 2 German families. Fam Cancer 2011;10:591-5.
[12] Sijmons R, Hofstra R, Hollema H, et al. Inclusion of malignant fibrous histiocytoma in the tumour spectrum associated with hereditary non- polyposis colorectal cancer. Genes Chromosomes Cancer 2000;29: 353-5.
[13] Clyne M, Offman J, Shanley S, et al. The G67E mutation in hMLH1 is associated with an unusual presentation of Lynch syndrome. Br J Cancer 2009;100:376-80.
[14] Yu VP, Novelli M, Payne SJ, et al. Unusual presentation of Lynch syndrome. Hered Cancer Clin Pract 2009;7:12.
[15] Nilbert M, Therkildsen C, Nissen A, Akerman M, Bernstein I. Sarcomas associated with hereditary nonpolyposis colorectal cancer: broad anatomical and morphological spectrum. Fam Cancer 2009;8:209-13.
[16] Ponti G, Losi L, Pellacani G, et al. Malignant melanoma in patients with hereditary nonpolyposis colorectal cancer. Br J Dermatol 2008;159: 162-8.
[17] D’Arcy C, Wen YH, Stadler ZK, Brogi E, Shia J. Synchronous breast cancers with different morphologic and molecular phenotypes occurring in lynch syndrome: what does the heterogeneity imply? Am J Surg Pathol 2011;35:1743-8.
[18] Soravia C, van der Klift H, Brundler MA, et al. Prostate cancer is part of the hereditary non-polyposis colorectal cancer (HNPCC) tumor spectrum. Am J Med Genet A 2003;121A:159-62.
[19] Nolan L, Eccles D, Cross E, et al. First case report of Muir-Torre syndrome associated with non-small cell lung cancer. Fam Cancer 2009;8:359-62.
[20] Broaddus RR, Lynch PM, Lu KH, Luthra R, Michelson SJ. Unusual tumors associated with the hereditary nonpolyposis colorectal cancer syndrome. Mod Pathol 2004;17:981-9.
[21] Berends MJ, Cats A, Hollema H, et al. Adrenocortical adenocarcinoma in an MSH2 carrier: coincidence or causal relation? HUM PATHOL 2000;31:1522-7.
[22] Medina-Arana V, Delgado L, Gonzalez L, et al. Adrenocortical carcinoma, an unusual extracolonic tumor associated with Lynch II syndrome. Fam Cancer 2011;10:265-71.
[23] Murphy KM, Zhang S, Geiger T, et al. Comparison of the microsatellite instability analysis system and the Bethesda panel for the determination of microsatellite instability in colorectal cancers. J Mol Diagn 2006;8:305-11.
[24] Petersen M, Evert M, Schneider-Stock R, et al. Serous oligocystic adenoma (SOIA) of the pancreas-first reported case of a genetically fixed association in a patient with hereditary non-polyposis colorectal cancer (HNPCC). Pathol Res Pract 2009;205:801-6.
[25] Lowery MA, Klimstra DS, Shia J, et al. Acinar cell carcinoma of the pancreas: new genetic and treatment insights into a rare malignancy. Oncologist 2011;16:1714-20.
[26] Walsh CS, Blum A, Walts A, et al. Lynch syndrome among gynecologic oncology patients meeting Bethesda guidelines for screening. Gynecol Oncol 2010;116:516-21.
[27] Jensen UB, Sunde L, Timshel S, et al. Mismatch repair defective breast cancer in the hereditary nonpolyposis colorectal cancer syndrome. Breast Cancer Res Treat 2010;120:777-82.
[28] Vasen HF, Morreau H, Nortier JW. Is breast cancer part of the tumor spectrum of hereditary nonpolyposis colorectal cancer? Am J Hum Genet 2001;68:1533-5.
[29] Blake C, Tsao JL, Wu A, Shibata D. Stepwise deletions of polyA sequences in mismatch repair-deficient colorectal cancers. Am J Pathol 2001;158:1867-70.
[30] House MG, Herman JG, Guo MZ, et al. Prognostic value of hMLH1 methylation and microsatellite instability in pancreatic endocrine neoplasms. Surgery 2003;134:902-8 [discussion 909].
[31] Mei M, Deng D, Liu TH, et al. Clinical implications of microsatellite instability and MLH1 gene inactivation in sporadic insulinomas. J Clin Endocrinol Metab 2009;94:3448-57.
[32] Abraham SC, Wu TT, Hruban RH, et al. Genetic and immunohisto- chemical analysis of pancreatic acinar cell carcinoma: frequent allelic loss on chromosome 11p and alterations in the APC/beta-catenin pathway. Am J Pathol 2002;160:953-62.