Pathology – Research and Practice 211 (2015) 78–82
Contents lists available at ScienceDirect
Pathology – Research and Practice journal homepage: www.elsevier.com/locate/prp
Original article
Sebaceous neoplasms and the immunoprofile of mismatch-repair proteins as a screening target for syndromic cases夽 Marie Boennelycke a,∗ , Birthe M. Thomsen a , Susanne Holck b a b
Department of Pathology, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark Department of Pathology, Copenhagen University Hospital Hvidovre, Denmark
a r t i c l e
i n f o
Article history: Received 19 June 2014 Received in revised form 7 September 2014 Accepted 15 October 2014 Keywords: Muir–Torre syndrome Sebaceous adenoma Sebaceous carcinoma Mismatch-repair proteins p16
a b s t r a c t Introduction: Muir–Torre syndrome (MTS), a subset of Lynch syndrome, is characterized by concurrent or sequential development of sebaceous neoplasms, and internal malignancies, specifically colorectal carcinoma (CRC), and can be related to mismatch-repair (MMR)-protein deficiency. In CRC context, p16negativity in MLH1-deficient cases may denote methylation rather than mutation. The prime aim of this study was to evaluate the mismatch-repair (MMR)-protein deficiency and the p16 status among sebaceous neoplasms. Material and method: From January 1990 through October 2012, 26 sebaceous adenomas (SAs) and 6 sebaceous carcinomas (SCs) were accrued. The expression of MLH1, MSH2, MSH6, and PMS2 was recorded. MLH1-deficient cases were tested for p16 status. Results: Eighteen (56%) of the 32 specimens with SA or SC displayed MMR-protein deficiency, comprising 17 (65.4%) SAs (MSH2/MSH6 loss in 12, MLH1/PMS2 loss in 3, MSH6 loss only in 2 cases) and 1 (16.7%) SC (MLH1/PMS2 loss). All 4 MLH1 deficient cases were p16-positive. Conclusion: A substantial proportion of sebaceous neoplasms were MMR-protein deficient and thus likely MTS candidates. Given the low prevalence of sebaceous neoplasms in Denmark, immunohistochemistry for the four MMR-proteins is recommended in the initial diagnostic approach. The addition of p16 was none-informative, but evaluation of its utility in larger series is warranted. © 2014 Elsevier GmbH. All rights reserved.
Introduction Muir–Torre syndrome (MTS) refers to an autosomal-dominant inherited condition that predisposes to concurrent or sequential development of some skin lesions, most commonly sebaceous neoplasms and a variety of internal malignancies, specifically colorectal carcinoma (CRC). This rare syndrome was independently described by Muir and Torre [1,2]. Considering the spectrum of neoplasms and the germline mutations of MTS, Lynch et al. [3] called attention to the notion of MTS as a phenotypic subset, albeit minor (1–3%), of Lynch syndrome. Though mismatch-repair (MMR)-protein proficiency does not exclude MTS [4,5], the immunohistochemistry (IHC) analysis of MMR-proteins remains an approach to be considered in the workup of sebocytic lesions. The prevalence of MMR-protein deficiency
夽 Part of this study has been presented in abstract form (APMIS 2013, vol. 121, Suppl. 135, 03-1). ∗ Corresponding author. Tel.: +4527632333. E-mail address: [email protected] (M. Boennelycke). http://dx.doi.org/10.1016/j.prp.2014.10.002 0344-0338/© 2014 Elsevier GmbH. All rights reserved.
among neoplasms with sebocytic differentiation as a group has been addressed in several unselected series [4–10], the reported values ranging from roughly 25% to 80%. Additionally, in a series of 25 patients with sebaceous neoplasms, examined for microsatellite instability (MSI), 60% were found to be MSI-high (MSI-H) [11]. Some MMR-protein IHC studies of sebaceous neoplasms were based on MLH1- and MSH2 status only [6,12–15], others added MSH6 to the staining panel [4,7–10,16], and some of these communications additionally report the PMS2-protein status [4,5,7–10,17]. The interpretation of the MMR-protein immunostainings of the sebaceous neoplasms has by some investigators been reported as straight forward [9], whereas others experienced equivocal and/or weak expression, especially with MSH6 [4,5,8,15]. For this reason, the implementation of a 2-panel approach, confined to MSH6 and PMS2, previously addressed [8] as a cost-effective alternative in screening for MMR-protein deficiency among sebaceous neoplasms, might prove problematic. In the CRC context, a significant fraction of all MMR protein deficient cases (specifically MLH1 loss) [18–20] are caused by somatic hypermethylation of the MLH1 promoter region rather than a
M. Boennelycke et al. / Pathology – Research and Practice 211 (2015) 78–82
deleterious germline mutation. This event needs consideration prior to initiating genetic counselling and time-consuming mutation analysis. Molecular examination of BRAF-600E mutation is a mean of identifying some of these cases. An alternative, less exigent approach concerns IHC for BRAF or p16 protein. The utility of the latter has been demonstrated for CRCs [21,22]. Thus, loss of p16 in MLH1 deficient CRCs has consistently been linked with hypermethylation, whereas p16 expression has been none-informative. The yield of this approach in sebaceous neoplasms is unknown, but it is noteworthy that 6 (40%) of 15 sebaceous neoplasms with MSIH have been found to be incidental, sporadic, not MTS-associated lesions [11]. The primary objective of this study was to investigate the prevalence of MTS candidates among a retrospective cohort of SAs and SCs, respectively, using a 4-panel approach, to record their MMRprotein status. Further, the utility of determining the p16 status in case of MLH1 protein deficiency was explored, in an attempt to differentiate MTS-linked cases from sporadic counterparts. Additionally, extent of labelling and interpretative challenges of the MMR-protein-stained slides including aberrant staining patterns that are well-described in some CRCs, were addressed. Materials and methods Cases signed-out as SA or SC in the pathology database of the Department of Pathology, Bispebjerg Hospital, extracted retrospectively between the time interval of January 1990 and October 2012, produced a total of 52 hits (45 SAs from 34 patients and 7 SCs from 7 patients). The slides were retrieved and reviewed. The diagnosis of SA (defined as multiple incompletely differentiated lobules of mature sebaceous cells centrally, surrounded by basaloid type cells [9]) could not be confirmed in 2 cases. On the basis of availability of non-exhausted paraffin blocks among the remaining cases, 32 sebaceous neoplasms (26 SAs and 6 SCs) from 32 patients were selected and processed for study (in cases of more than a single neoplasm in a given patient, the most recently excised specimen was used, here referred to as the index specimen). Sebocytic neoplasms and Lynch syndrome related internal malignancies, diagnosed prior to the index specimen were recorded, as was available relevant information on family history. Upon collection of the study material all patient data were deidentified. On this background the study was approved by the ethical scientific committee of the Region Hovedstaden, Denmark. Immunohistochemical stainings Stainings were performed on one tumour block from each tumour. Four-micron thick sections of the formalin-fixed, paraffinembedded tissue blocks were cut and immunostained. Antigen
79
Table 1 Data on antibodies. Antibody
Clone
Dilution
Manufacturer
MLH1 MSH2 MSH6 PMS2 P16
ES05/E805a G219-1129/FE11a 44/MSH6/EP49a A16-4/EP51b Cintec p16
RTU/RTUa 1:600/RTUa 1:2000/RTUa 1:150/RTUb RTU
Dako/Dakoa Becton Dickinson/Dakoa Epitomic/Dakoa BD Pharmigen/Dakob Roche
a b
Repeat procedure (one case). Repeat procedure (two cases).
retrieval was achieved by PT-Link at high pH (Dako). Subsequently, the sections were processed in a Dako autostainer, applying the antibodies that target the MMR-proteins MLH1, MSH2, MSH6, and PMS2. In cases of MLH1 deficiency, additional immunostaining for p16 was performed in a Ventana autostainer (Roche). Details on the antibodies are given in Table 1. The antibodies were detected with a two-layer Envision (Dako) peroxidase detection system. Tissue sections were counterstained with Meyer’s hematoxylin, dehydrated, mounted on coated slides, and dried one hour at 60◦ . For MMR-proteins, expression of non-neoplastic cells included in the sections served as positive control. For p16 controls, appropriate sections of uterine cervical carcinoma were run in parallel. Interpretation of the specimens The MMR-protein expression was considered normal when neoplastic nuclei (of any extent) were labelled. Slides with intact MMR-proteins were further characterized according to the proportion of stained neoplastic nuclei, conducted in a semi quantitative manner. Histological details that correlated with the extent of labelling were noticed. Tumours that showed total absence of nuclear staining in the presence of immunostained stromal/inflammatory/non-neoplastic epithelial cells, indicated loss of expression. In case of uncertain interpretation, the procedure was repeated. Specimens were considered p16 positive when nuclear with or without cytoplasmic staining of the tumour cells was present. Examples are shown in Figs. 1 and 2. Results The interpretation of the protein stainings was considered unequivocal in all but two specimens. The latter applied to all four MMR-proteins in one SA specimen, which was small and focally crushed and to the PMS2 staining of another SA, in which internal control as well as lesional tissue were none reactive. Upon repeated IHC, both cases became assessable. Fourteen of the 32 studied cases displayed normal expression of all 4 markers. The proportion of the labelled neoplastic nuclei
Fig. 1. Example of sebaceous adenoma with normal expression of MLH1 (A) (and of PMS2, not shown), loss of MSH2 (B) (and of MSH6, not shown).
80
M. Boennelycke et al. / Pathology – Research and Practice 211 (2015) 78–82
Fig. 2. Sebaceous adenoma with normal expression of MSH6 (A) (and MSH2, not shown), loss of MLH1 (B) (and of PMS2, not shown), and patchy expression of p16 (C).
ranged from 10–100% for MLH1, PMS2, and MSH6, 20–100% for MSH2. Median values for the four markers were 90% (MLH1), 80% (PMS2), 90% (MSH2), and 70% (MSH6). The values of SAs and SCs were comparable. Unstained nuclei were largely sited near cysts. Accordingly, prominent cysts characterized the four slides displaying expression of only 10% of the nuclei. In the remaining 18 study patients with SA or SC, MMR-protein deficiency was demonstrated, comprising 17 (65.4%) of the 26 SAs and one (16.7%) of the 6 SCs. Loss of MSH2 and MSH6 was recorded in 12 SAs; loss of MLH1 and PMS2 in 3 SAs and in one SC. In two SAs loss was confined to MSH6. Loss of PMS2 only was not present. Personal history in relation to MMR-protein status is summarized in Table 2. Thus, one or more sebocytic neoplasms had been diagnosed prior to the index specimen in 5 of the MMRprotein deficient cases as well as in 3 proficient cases. Three of these patients were additionally previously diagnosed with relevant internal malignancies. In none of the cases was a relevant family history available. Aberrant staining patterns of the MMR-proteins (i.e. cytoplasmic, perinuclear, or dot-like nucleolar) as well as extremely patchy pattern were not present. All four MLH1 deficient cases were p16-positive.
determination of the MMR-protein status remains an important approach in the characterization of patients with sebaceous neoplasms. A similar view has previously been advanced by others [10,14,17] and was most recently formulated by Everett et al. [5], who proposed a practical algorithm for patients with sebaceous neoplasms, including IHC as a first-line screening. Additionally, Everett et al. commented on the limited utility of family history alone. Accordingly, in none of our study cases were a relevant family history available. Among our 8 cases with a positive personal history (all having prior sebaceous neoplasms, a strong MTS marker), three had intact MMR-proteins, corroborating the notion that MTS-associated genetic events other than MMR gene mutations occasionally leads to sebaceous neoplasms [17].
Discussion
MSH6 deficiency
Prevalence of MMR-protein deficiency among sebaceous neoplasms
Value of MMR-protein status of sebaceous neoplasms in identifying MTS candidates
Several available IHC analyses of sebaceous neoplasms have included only MLH1 and MSH2 in the antibody panel [6,12–15]. The addition of a MSH6 marker is, however, advisable. In our SA cohort, MSH6 deficiency was the most prevalent aberration, occurring in combination with MSH2 deficiency in the vast majority of cases. This was also the observation in prior reports that include MSH6 examination [7–10]. Notably, in two of our 17 MMR-protein deficient SAs, loss of expression was confined to the MSH6, probably indicating MSH6 mutation, as was the case in three of the specimens (sebaceous adenomas and epitheliomas) in Chhibbar’s series [7], as well as in 2 of 25 deficient sebaceous neoplasms of Cornejo et al. [17], and in one of 26 SAs reported by Mojtahed et al. [8]. None of 45 SAs and 17 SCs presented by Lee et al. [4] included that constellation. The sole case of exclusive MSH6 loss in the Orta’s series [9] was a SC.
Even though a personal history provided a clue to a possible MTS diagnosis in 8 (5 deficient, 3 proficient) of the studied patients,
PMS2 deficiency
The yield of deficiency in at least one of the key MMR-proteins by IHC staining was noticed in a substantial proportion (65.4%) of our patients with SA. In concert, SA is considered the most specific marker of MTS [23]. Benign neoplasms of other organs, specifically the colorectum (i.e. adenomatous lesions) are less commonly deficient [24]. In the current series of sebaceous neoplasms, SC was less frequently MMR-protein deficient, corresponding to data appearing in most [3,4,6,8,25] albeit not in all [9] prior reports.
Table 2 Correlation of MMR-protein status of SNs with personal history. MMR-protein status
Personal history (prior MTS-associated neoplasms)
Deficiency (n = 18) Proficiency (n = 14)
CRC and SN: 2 cases (SN only: 3 cases) UroC and SN: 1 case (SN only: 2 cases)
MMR: mismatch-repair; SN: sebaceous neoplasm; MTS: Muir–Torre syndrome; LS: Lynch syndrome; CRC: colorectal carcinoma; UroC: urothelial carcinoma.
Distribution pattern of MMR-protein deficiencies We found, as have others [6–10,13,14,17,26] that MSH2/MSH6 deficiency, suggestive of MSH2 germline mutation, overrides other constellations. However, in the study by Entius et al. [12], based on eight MMR-protein deficient SCs, the proportion of cases with loss of MSH2 and MLH1 was similar, which, as the authors acknowledged, might reflect bias by familiar clustering.
As anticipated, loss of PMS2 accompanied all cases of MLH1 deficiency. In none of the presently studied SAs and SCs was PMS2 loss the sole abnormality. Indeed, such profile, suggestive of PMS2 mutation, has been recorded in merely one prior report [9]. Thus, Orta and co-workers [9] recorded loss of PMS2 unaccompanied by MLH1 loss in 2 of 12 deficiency cases (one SA and one SC). In a recent review paper by Fernandez-Flores [27] a figure demonstrating this uncommon immunoprofile was, however, included.
M. Boennelycke et al. / Pathology – Research and Practice 211 (2015) 78–82
81
The utility of P16 analysis
References
Though mutational status was not obtained, the present IHC results are suggestive of MTS in 14 SAs (12 cases of MSH2/MSH6 deficiency and two cases of MSH6 only deficiency). In the 4 cases of combined deficiency of MLH1 and PMS2 (3 SAs and one SC), transcriptional silencing caused by abnormal methylation needs consideration. For this reason we recorded the p16 status. In the current study all four MLH1-deficient cases were p16-positive. This immunoprofile is none-informative. The constellation of MLH1 deficiency and p16-negativity, which in CRCs has been found highly suggestive of methylation [21,22], justifying sparing of genetic counselling, was not a component of the present, admittedly small series. This issue was recently addressed by Cornejo et al. [17], who recorded the BRAF status of 7 MLH1 deficient sebocytic neoplasms, all of which yielded wild-type BRAF genotype. Thus, that study was none-informative like our p16 analysis. The putative role of hypermethylation in MLH1 deficient sebaceous neoplasms needs further exploration by adding methylation analyses.
[1] E.G. Muir, A.J. Bell, K.A. Barlow, Multiple primary carcinomata of the colon, duodenum, and larynx associated with kerato-acanthomata of the face, Br. J. Surg. 54 (1967) 191–195. [2] D. Torre, Multiple sebaceous tumors, Arch. Dermatol. 98 (1968) 549–551. [3] H.T. Lynch, P.M. Lynch, J. Pester, R.M. Fusaro, The cancer family syndrome. Rare cutaneous phenotypic linkage of Torre’s syndrome, Arch. Int. Med. 141 (1981) 607–611. [4] B.A. Lee, L. Yu, L. Ma, A.C. Lind, D. Lu, Sebaceous neoplasms with mismatch repair protein expressions and the frequency of co-existing visceral tumors, J. Am. Acad. Dermatol. 67 (2012) 1228–1234. [5] J.N. Everett, V.M. Raymond, M. Dandapani, M. Marvin, W. Kohlmann, A. Chittenden, et al., Screening for germline mismatch repair mutations following diagnosis of sebaceous neoplasm, JAMA Dermatol. (2014), http://dx.doi. org/10.1001/jamadermatol.2014.1217. [6] A.M. Cesinaro, A. Ubiali, P. Sighinolfi, G.P. Trentini, F. Gentili, F. Facchetti, Mismatch repair proteins expression and microsatellite instability in skin lesions with sebaceous differentiation: a study in different clinical subgroups with and without extracutaneous cancer, Am. J. Dermatopathol. 29 (2007) 351–358. [7] V. Chhibber, K. Dresser, M. Mahalingam, MSH-6: extending the reliability of immunohistochemistry as a screening tool in Muir–Torre syndrome, Mod. Pathol. 21 (2008) 159–164. [8] A. Mojtahed, I. Schrijver, J.M. Ford, T.A. Longacre, R.K. Pai, A two-antibody mismatch repair protein immunohistochemistry screening approach for colorectal carcinomas, skin sebaceous tumors, and gynecologic tract carcinomas, Mod. Pathol. 24 (2011) 1004–1014. [9] L. Orta, D.S. Klimstra, J. Qin, P. Mecca, L.H. Tang, K.J. Busam, et al., Towards identification of hereditary DNA mismatch repair deficiency: sebaceous neoplasm warrants routine immunohistochemical screening regardless of patient’s age or other clinical characteristics, Am. J. Surg. Pathol. 33 (2009) 934–944. [10] E.F. Plocharczyk, W.L. Frankel, H. Hampel, S.B. Peters, Mismatch repair protein deficiency is common in sebaceous neoplasms and suggests the importance of screening for Lynch syndrome, Am. J. Dermatopathol. 35 (2013) 191–195. [11] R. Kruse, A. Rutten, N. Schweiger, E. Jakob, M. Mathiak, P. Propping, et al., Frequency of microsatellite instability in unselected sebaceous gland neoplasias and hyperplasias, J. Invest. Dermatol. 120 (2003) 858–864. [12] M.M. Entius, J.J. Keller, P. Drillenburg, K.C. Kuypers, F.M. Giardiello, G.J. Offerhaus, Microsatellite instability and expression of hMLH-1 and hMSH-2 in sebaceous gland carcinomas as markers for Muir–Torre syndrome, Clin. Cancer Res. 6 (2000) 1784–1789. [13] P. Machin, L. Catasus, C. Pons, J. Munoz, J.M. Conde-Zurita, J. Balmana, et al., Microsatellite instability and immunostaining for MSH-2 and MLH-1 in cutaneous and internal tumors from patients with the Muir–Torre syndrome, J. Cutan. Pathol. 29 (2002) 415–420. [14] M. Mathiak, A. Rutten, E. Mangold, H.P. Fischer, T. Ruzicka, W. Friedl, et al., Loss of DNA mismatch repair proteins in skin tumors from patients with Muir–Torre syndrome and MSH2 or MLH1 germline mutations: establishment of immunohistochemical analysis as a screening test, Am. J. Surg. Pathol. 26 (2002) 338–343. [15] A. Morales-Burgos, J.L. Sanchez, L.D. Figueroa, W.E. De Jesus-Monge, M.R. CruzCorrea, C. Gonzalez-Keelan, et al., MSH-2 and MLH-1 protein expression in Muir–Torre syndrome-related and sporadic sebaceous neoplasms, P. R. Health Sci. J. 27 (2008) 322–327. [16] G. Ponti, L. Losi, G.C. Di, L. Roncucci, M. Pedroni, A. Scarselli, et al., Identification of Muir–Torre syndrome among patients with sebaceous tumors and keratoacanthomas: role of clinical features, microsatellite instability, and immunohistochemistry, Cancer 103 (2005) 1018–1025. [17] K.M. Cornejo, L. Hutchinson, A. Deng, K. Tomaszewicz, M. Welch, S. Lyle, et al., BRAF/KRAS gene sequencing of sebaceous neoplasms after mismatch repair protein analysis, Hum. Pathol. 45 (2014) 1213–1220. [18] E. Glogowski, S. Boyar, K. Sarrel, C. LeBlang, E. Utay, J. Shia, et al., Unexplained immunohistochemical DNA mismatch repair deficiency, Fam. Cancer 10 (2011) 726. [19] H. Hampel, W.L. Frankel, E. Martin, M. Arnold, K. Khanduja, P. Kuebler, et al., Feasibility of screening for Lynch syndrome among patients with colorectal cancer, J. Clin. Oncol. 26 (2008) 5783–5788. [20] L.H. Jensen, J. Lindebjerg, L. Byriel, S. Kolvraa, D.G. Cruger, Strategy in clinical practice for classification of unselected colorectal tumours based on mismatch repair deficiency, Color Dis. 10 (2008) 490–497. [21] A. Paya, C. Alenda, L. Perez-Carbonell, E. Rojas, J.L. Soto, C. Guillen, et al., Utility of p16 immunohistochemistry for the identification of Lynch syndrome, Clin. Cancer Res. 15 (2009) 3156–3162. [22] T. Agander, L. Lindberg, L. Brixen, H. Okkels, M. Haska, I. Bernstein, et al., The utility of P16 in identifying Lynch syndrome (LS), APMIS 121 (2013) 22. [23] O. Abbas, M. Mahalingam, Cutaneous sebaceous neoplasms as markers of Muir–Torre syndrome: a diagnostic algorithm, J. Cutan. Pathol. 36 (2009) 613–619. [24] B. Halvarsson, A. Lindblom, L. Johansson, K. Lagerstedt, M. Nilbert, Loss of mismatch repair protein immunostaining in colorectal adenomas from patients with hereditary nonpolyposis colorectal cancer, Mod. Pathol. 18 (2005) 1095–1101. [25] R.S. Singh, W. Grayson, M. Redston, A.H. Diwan, C.L. Warneke, P.H. McKee, et al., Site and tumor type predicts DNA mismatch repair status in cutaneous sebaceous neoplasia, Am. J. Surg. Pathol. 32 (2008) 936–942.
Interpretation of MMR-protein staining The challenge in the interpretation of MMR-protein immunostains of CRC has repeatedly been reported, contributing to observer variations [28], specifically in MSI tumours [29]. Such difficulties seem less prominent in case of SA/SC, a quality, which has been ascribed to the high proliferative activity of epidermally derived lesions [30] and less common fixation problems, given the limited dimensions of these specimens. Accordingly, Machin et al. [13] recorded equivocal MLH1 staining of CRC despite unequivocal loss of MLH1 of a subsequently examined SA, corroborating the low sensitivity in detecting MLH1 germline mutation by IHC of CRCs [31]. Crush artefact of most lesional cells in some small specimens may, on the other hand, prove a limiting factor as noticed in one of our cases, in which the determination of the MMR protein status required additional deeper sections. Aberrant staining patterns, such as cytoplasmic staining, perinuclear staining and “dot-like” nucleolar staining occasionally appear in the CRC context [30] with the risk of confounding the interpretation. These qualities were not a feature in the current series and have not, to our knowledge, been addressed in prior series of sebocytic neoplasms.
Recommended histopathological handling of sebaceous neoplasms Given the high yield of MMR-protein deficiency among sebaceous neoplasms, coupled with their relatively low occurrence in the general population (in Denmark some 45 newly diagnosed cases per year), we consider IHC screening as a pivotal initial step and recommend the use of a 4-panel approach as routinely used in CRCs, designed to capture MTS candidate lesions. The fact that these neoplasms may predate visceral malignancies [6] provides an added incentive to follow this approach. Furthermore, immunostaining is an inexpensive, widely available method, compared to the more costly microsatellite instability (MSI) analyses. Our preliminary p16 investigation of MLH1deficient cases proved none-informative. Additional research efforts on larger cohorts directed at this issue are, however, desirable. In this manner the awareness on MTS in the pathology community will be increased, whereby genetic counselling as well as lifelong cancer surveillance of both probands and at-risk family members more likely will be undertaken, reducing cancer mortality.
82
M. Boennelycke et al. / Pathology – Research and Practice 211 (2015) 78–82
[26] E. Mangold, N. Rahner, N. Friedrichs, R. Buettner, C. Pagenstecher, S. Aretz, et al., MSH6 mutation in Muir–Torre syndrome: could this be a rare finding? Br. J. Dermatol. 156 (2007) 158–162. [27] A. Fernandez-Flores, Considerations on the performance of immunohistochemistry for mismatch repair gene proteins in cases of sebaceous neoplasms and keratoacanthomas with reference to Muir–Torre syndrome, Am. J. Dermatopathol. 34 (2012) 416–422. [28] L. Klarskov, S. Ladelund, S. Holck, K. Roenlund, J. Lindebjerg, J. Elebro, et al., Interobserver variability in the evaluation of mismatch repair protein immunostaining, Hum. Pathol. 41 (2010) 1387–1396. [29] L.I. Overbeek, M.J. Ligtenberg, R.W. Willems, R.P. Hermens, W.A. Blokx, S.V. Dubois, et al., Interpretation of immunohistochemistry for mismatch repair
proteins is only reliable in a specialized setting, Am. J. Surg. Pathol. 32 (2008) 1246–1251. [30] J. Shia, S. Holck, G. Depetris, J.K. Greenson, D.S. Klimstra, Lynch syndrome-associated neoplasms: a discussion on histopathology and immunohistochemistry, Fam. Cancer 12 (2013) 241–260, http://dx.doi.org/ 10.1007/s10689-013-9612-4. [31] J. Shia, D.S. Klimstra, K. Nafa, K. Offit, J.G. Guillem, A.J. Markowitz, et al., Value of immunohistochemical detection of DNA mismatch repair proteins in predicting germline mutation in hereditary colorectal neoplasms, Am. J. Surg. Pathol. 29 (2005) 96–104.
Contents lists available at ScienceDirect
Pathology – Research and Practice journal homepage: www.elsevier.com/locate/prp
Original article
Sebaceous neoplasms and the immunoprofile of mismatch-repair proteins as a screening target for syndromic cases夽 Marie Boennelycke a,∗ , Birthe M. Thomsen a , Susanne Holck b a b
Department of Pathology, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark Department of Pathology, Copenhagen University Hospital Hvidovre, Denmark
a r t i c l e
i n f o
Article history: Received 19 June 2014 Received in revised form 7 September 2014 Accepted 15 October 2014 Keywords: Muir–Torre syndrome Sebaceous adenoma Sebaceous carcinoma Mismatch-repair proteins p16
a b s t r a c t Introduction: Muir–Torre syndrome (MTS), a subset of Lynch syndrome, is characterized by concurrent or sequential development of sebaceous neoplasms, and internal malignancies, specifically colorectal carcinoma (CRC), and can be related to mismatch-repair (MMR)-protein deficiency. In CRC context, p16negativity in MLH1-deficient cases may denote methylation rather than mutation. The prime aim of this study was to evaluate the mismatch-repair (MMR)-protein deficiency and the p16 status among sebaceous neoplasms. Material and method: From January 1990 through October 2012, 26 sebaceous adenomas (SAs) and 6 sebaceous carcinomas (SCs) were accrued. The expression of MLH1, MSH2, MSH6, and PMS2 was recorded. MLH1-deficient cases were tested for p16 status. Results: Eighteen (56%) of the 32 specimens with SA or SC displayed MMR-protein deficiency, comprising 17 (65.4%) SAs (MSH2/MSH6 loss in 12, MLH1/PMS2 loss in 3, MSH6 loss only in 2 cases) and 1 (16.7%) SC (MLH1/PMS2 loss). All 4 MLH1 deficient cases were p16-positive. Conclusion: A substantial proportion of sebaceous neoplasms were MMR-protein deficient and thus likely MTS candidates. Given the low prevalence of sebaceous neoplasms in Denmark, immunohistochemistry for the four MMR-proteins is recommended in the initial diagnostic approach. The addition of p16 was none-informative, but evaluation of its utility in larger series is warranted. © 2014 Elsevier GmbH. All rights reserved.
Introduction Muir–Torre syndrome (MTS) refers to an autosomal-dominant inherited condition that predisposes to concurrent or sequential development of some skin lesions, most commonly sebaceous neoplasms and a variety of internal malignancies, specifically colorectal carcinoma (CRC). This rare syndrome was independently described by Muir and Torre [1,2]. Considering the spectrum of neoplasms and the germline mutations of MTS, Lynch et al. [3] called attention to the notion of MTS as a phenotypic subset, albeit minor (1–3%), of Lynch syndrome. Though mismatch-repair (MMR)-protein proficiency does not exclude MTS [4,5], the immunohistochemistry (IHC) analysis of MMR-proteins remains an approach to be considered in the workup of sebocytic lesions. The prevalence of MMR-protein deficiency
夽 Part of this study has been presented in abstract form (APMIS 2013, vol. 121, Suppl. 135, 03-1). ∗ Corresponding author. Tel.: +4527632333. E-mail address: [email protected] (M. Boennelycke). http://dx.doi.org/10.1016/j.prp.2014.10.002 0344-0338/© 2014 Elsevier GmbH. All rights reserved.
among neoplasms with sebocytic differentiation as a group has been addressed in several unselected series [4–10], the reported values ranging from roughly 25% to 80%. Additionally, in a series of 25 patients with sebaceous neoplasms, examined for microsatellite instability (MSI), 60% were found to be MSI-high (MSI-H) [11]. Some MMR-protein IHC studies of sebaceous neoplasms were based on MLH1- and MSH2 status only [6,12–15], others added MSH6 to the staining panel [4,7–10,16], and some of these communications additionally report the PMS2-protein status [4,5,7–10,17]. The interpretation of the MMR-protein immunostainings of the sebaceous neoplasms has by some investigators been reported as straight forward [9], whereas others experienced equivocal and/or weak expression, especially with MSH6 [4,5,8,15]. For this reason, the implementation of a 2-panel approach, confined to MSH6 and PMS2, previously addressed [8] as a cost-effective alternative in screening for MMR-protein deficiency among sebaceous neoplasms, might prove problematic. In the CRC context, a significant fraction of all MMR protein deficient cases (specifically MLH1 loss) [18–20] are caused by somatic hypermethylation of the MLH1 promoter region rather than a
M. Boennelycke et al. / Pathology – Research and Practice 211 (2015) 78–82
deleterious germline mutation. This event needs consideration prior to initiating genetic counselling and time-consuming mutation analysis. Molecular examination of BRAF-600E mutation is a mean of identifying some of these cases. An alternative, less exigent approach concerns IHC for BRAF or p16 protein. The utility of the latter has been demonstrated for CRCs [21,22]. Thus, loss of p16 in MLH1 deficient CRCs has consistently been linked with hypermethylation, whereas p16 expression has been none-informative. The yield of this approach in sebaceous neoplasms is unknown, but it is noteworthy that 6 (40%) of 15 sebaceous neoplasms with MSIH have been found to be incidental, sporadic, not MTS-associated lesions [11]. The primary objective of this study was to investigate the prevalence of MTS candidates among a retrospective cohort of SAs and SCs, respectively, using a 4-panel approach, to record their MMRprotein status. Further, the utility of determining the p16 status in case of MLH1 protein deficiency was explored, in an attempt to differentiate MTS-linked cases from sporadic counterparts. Additionally, extent of labelling and interpretative challenges of the MMR-protein-stained slides including aberrant staining patterns that are well-described in some CRCs, were addressed. Materials and methods Cases signed-out as SA or SC in the pathology database of the Department of Pathology, Bispebjerg Hospital, extracted retrospectively between the time interval of January 1990 and October 2012, produced a total of 52 hits (45 SAs from 34 patients and 7 SCs from 7 patients). The slides were retrieved and reviewed. The diagnosis of SA (defined as multiple incompletely differentiated lobules of mature sebaceous cells centrally, surrounded by basaloid type cells [9]) could not be confirmed in 2 cases. On the basis of availability of non-exhausted paraffin blocks among the remaining cases, 32 sebaceous neoplasms (26 SAs and 6 SCs) from 32 patients were selected and processed for study (in cases of more than a single neoplasm in a given patient, the most recently excised specimen was used, here referred to as the index specimen). Sebocytic neoplasms and Lynch syndrome related internal malignancies, diagnosed prior to the index specimen were recorded, as was available relevant information on family history. Upon collection of the study material all patient data were deidentified. On this background the study was approved by the ethical scientific committee of the Region Hovedstaden, Denmark. Immunohistochemical stainings Stainings were performed on one tumour block from each tumour. Four-micron thick sections of the formalin-fixed, paraffinembedded tissue blocks were cut and immunostained. Antigen
79
Table 1 Data on antibodies. Antibody
Clone
Dilution
Manufacturer
MLH1 MSH2 MSH6 PMS2 P16
ES05/E805a G219-1129/FE11a 44/MSH6/EP49a A16-4/EP51b Cintec p16
RTU/RTUa 1:600/RTUa 1:2000/RTUa 1:150/RTUb RTU
Dako/Dakoa Becton Dickinson/Dakoa Epitomic/Dakoa BD Pharmigen/Dakob Roche
a b
Repeat procedure (one case). Repeat procedure (two cases).
retrieval was achieved by PT-Link at high pH (Dako). Subsequently, the sections were processed in a Dako autostainer, applying the antibodies that target the MMR-proteins MLH1, MSH2, MSH6, and PMS2. In cases of MLH1 deficiency, additional immunostaining for p16 was performed in a Ventana autostainer (Roche). Details on the antibodies are given in Table 1. The antibodies were detected with a two-layer Envision (Dako) peroxidase detection system. Tissue sections were counterstained with Meyer’s hematoxylin, dehydrated, mounted on coated slides, and dried one hour at 60◦ . For MMR-proteins, expression of non-neoplastic cells included in the sections served as positive control. For p16 controls, appropriate sections of uterine cervical carcinoma were run in parallel. Interpretation of the specimens The MMR-protein expression was considered normal when neoplastic nuclei (of any extent) were labelled. Slides with intact MMR-proteins were further characterized according to the proportion of stained neoplastic nuclei, conducted in a semi quantitative manner. Histological details that correlated with the extent of labelling were noticed. Tumours that showed total absence of nuclear staining in the presence of immunostained stromal/inflammatory/non-neoplastic epithelial cells, indicated loss of expression. In case of uncertain interpretation, the procedure was repeated. Specimens were considered p16 positive when nuclear with or without cytoplasmic staining of the tumour cells was present. Examples are shown in Figs. 1 and 2. Results The interpretation of the protein stainings was considered unequivocal in all but two specimens. The latter applied to all four MMR-proteins in one SA specimen, which was small and focally crushed and to the PMS2 staining of another SA, in which internal control as well as lesional tissue were none reactive. Upon repeated IHC, both cases became assessable. Fourteen of the 32 studied cases displayed normal expression of all 4 markers. The proportion of the labelled neoplastic nuclei
Fig. 1. Example of sebaceous adenoma with normal expression of MLH1 (A) (and of PMS2, not shown), loss of MSH2 (B) (and of MSH6, not shown).
80
M. Boennelycke et al. / Pathology – Research and Practice 211 (2015) 78–82
Fig. 2. Sebaceous adenoma with normal expression of MSH6 (A) (and MSH2, not shown), loss of MLH1 (B) (and of PMS2, not shown), and patchy expression of p16 (C).
ranged from 10–100% for MLH1, PMS2, and MSH6, 20–100% for MSH2. Median values for the four markers were 90% (MLH1), 80% (PMS2), 90% (MSH2), and 70% (MSH6). The values of SAs and SCs were comparable. Unstained nuclei were largely sited near cysts. Accordingly, prominent cysts characterized the four slides displaying expression of only 10% of the nuclei. In the remaining 18 study patients with SA or SC, MMR-protein deficiency was demonstrated, comprising 17 (65.4%) of the 26 SAs and one (16.7%) of the 6 SCs. Loss of MSH2 and MSH6 was recorded in 12 SAs; loss of MLH1 and PMS2 in 3 SAs and in one SC. In two SAs loss was confined to MSH6. Loss of PMS2 only was not present. Personal history in relation to MMR-protein status is summarized in Table 2. Thus, one or more sebocytic neoplasms had been diagnosed prior to the index specimen in 5 of the MMRprotein deficient cases as well as in 3 proficient cases. Three of these patients were additionally previously diagnosed with relevant internal malignancies. In none of the cases was a relevant family history available. Aberrant staining patterns of the MMR-proteins (i.e. cytoplasmic, perinuclear, or dot-like nucleolar) as well as extremely patchy pattern were not present. All four MLH1 deficient cases were p16-positive.
determination of the MMR-protein status remains an important approach in the characterization of patients with sebaceous neoplasms. A similar view has previously been advanced by others [10,14,17] and was most recently formulated by Everett et al. [5], who proposed a practical algorithm for patients with sebaceous neoplasms, including IHC as a first-line screening. Additionally, Everett et al. commented on the limited utility of family history alone. Accordingly, in none of our study cases were a relevant family history available. Among our 8 cases with a positive personal history (all having prior sebaceous neoplasms, a strong MTS marker), three had intact MMR-proteins, corroborating the notion that MTS-associated genetic events other than MMR gene mutations occasionally leads to sebaceous neoplasms [17].
Discussion
MSH6 deficiency
Prevalence of MMR-protein deficiency among sebaceous neoplasms
Value of MMR-protein status of sebaceous neoplasms in identifying MTS candidates
Several available IHC analyses of sebaceous neoplasms have included only MLH1 and MSH2 in the antibody panel [6,12–15]. The addition of a MSH6 marker is, however, advisable. In our SA cohort, MSH6 deficiency was the most prevalent aberration, occurring in combination with MSH2 deficiency in the vast majority of cases. This was also the observation in prior reports that include MSH6 examination [7–10]. Notably, in two of our 17 MMR-protein deficient SAs, loss of expression was confined to the MSH6, probably indicating MSH6 mutation, as was the case in three of the specimens (sebaceous adenomas and epitheliomas) in Chhibbar’s series [7], as well as in 2 of 25 deficient sebaceous neoplasms of Cornejo et al. [17], and in one of 26 SAs reported by Mojtahed et al. [8]. None of 45 SAs and 17 SCs presented by Lee et al. [4] included that constellation. The sole case of exclusive MSH6 loss in the Orta’s series [9] was a SC.
Even though a personal history provided a clue to a possible MTS diagnosis in 8 (5 deficient, 3 proficient) of the studied patients,
PMS2 deficiency
The yield of deficiency in at least one of the key MMR-proteins by IHC staining was noticed in a substantial proportion (65.4%) of our patients with SA. In concert, SA is considered the most specific marker of MTS [23]. Benign neoplasms of other organs, specifically the colorectum (i.e. adenomatous lesions) are less commonly deficient [24]. In the current series of sebaceous neoplasms, SC was less frequently MMR-protein deficient, corresponding to data appearing in most [3,4,6,8,25] albeit not in all [9] prior reports.
Table 2 Correlation of MMR-protein status of SNs with personal history. MMR-protein status
Personal history (prior MTS-associated neoplasms)
Deficiency (n = 18) Proficiency (n = 14)
CRC and SN: 2 cases (SN only: 3 cases) UroC and SN: 1 case (SN only: 2 cases)
MMR: mismatch-repair; SN: sebaceous neoplasm; MTS: Muir–Torre syndrome; LS: Lynch syndrome; CRC: colorectal carcinoma; UroC: urothelial carcinoma.
Distribution pattern of MMR-protein deficiencies We found, as have others [6–10,13,14,17,26] that MSH2/MSH6 deficiency, suggestive of MSH2 germline mutation, overrides other constellations. However, in the study by Entius et al. [12], based on eight MMR-protein deficient SCs, the proportion of cases with loss of MSH2 and MLH1 was similar, which, as the authors acknowledged, might reflect bias by familiar clustering.
As anticipated, loss of PMS2 accompanied all cases of MLH1 deficiency. In none of the presently studied SAs and SCs was PMS2 loss the sole abnormality. Indeed, such profile, suggestive of PMS2 mutation, has been recorded in merely one prior report [9]. Thus, Orta and co-workers [9] recorded loss of PMS2 unaccompanied by MLH1 loss in 2 of 12 deficiency cases (one SA and one SC). In a recent review paper by Fernandez-Flores [27] a figure demonstrating this uncommon immunoprofile was, however, included.
M. Boennelycke et al. / Pathology – Research and Practice 211 (2015) 78–82
81
The utility of P16 analysis
References
Though mutational status was not obtained, the present IHC results are suggestive of MTS in 14 SAs (12 cases of MSH2/MSH6 deficiency and two cases of MSH6 only deficiency). In the 4 cases of combined deficiency of MLH1 and PMS2 (3 SAs and one SC), transcriptional silencing caused by abnormal methylation needs consideration. For this reason we recorded the p16 status. In the current study all four MLH1-deficient cases were p16-positive. This immunoprofile is none-informative. The constellation of MLH1 deficiency and p16-negativity, which in CRCs has been found highly suggestive of methylation [21,22], justifying sparing of genetic counselling, was not a component of the present, admittedly small series. This issue was recently addressed by Cornejo et al. [17], who recorded the BRAF status of 7 MLH1 deficient sebocytic neoplasms, all of which yielded wild-type BRAF genotype. Thus, that study was none-informative like our p16 analysis. The putative role of hypermethylation in MLH1 deficient sebaceous neoplasms needs further exploration by adding methylation analyses.
[1] E.G. Muir, A.J. Bell, K.A. Barlow, Multiple primary carcinomata of the colon, duodenum, and larynx associated with kerato-acanthomata of the face, Br. J. Surg. 54 (1967) 191–195. [2] D. Torre, Multiple sebaceous tumors, Arch. Dermatol. 98 (1968) 549–551. [3] H.T. Lynch, P.M. Lynch, J. Pester, R.M. Fusaro, The cancer family syndrome. Rare cutaneous phenotypic linkage of Torre’s syndrome, Arch. Int. Med. 141 (1981) 607–611. [4] B.A. Lee, L. Yu, L. Ma, A.C. Lind, D. Lu, Sebaceous neoplasms with mismatch repair protein expressions and the frequency of co-existing visceral tumors, J. Am. Acad. Dermatol. 67 (2012) 1228–1234. [5] J.N. Everett, V.M. Raymond, M. Dandapani, M. Marvin, W. Kohlmann, A. Chittenden, et al., Screening for germline mismatch repair mutations following diagnosis of sebaceous neoplasm, JAMA Dermatol. (2014), http://dx.doi. org/10.1001/jamadermatol.2014.1217. [6] A.M. Cesinaro, A. Ubiali, P. Sighinolfi, G.P. Trentini, F. Gentili, F. Facchetti, Mismatch repair proteins expression and microsatellite instability in skin lesions with sebaceous differentiation: a study in different clinical subgroups with and without extracutaneous cancer, Am. J. Dermatopathol. 29 (2007) 351–358. [7] V. Chhibber, K. Dresser, M. Mahalingam, MSH-6: extending the reliability of immunohistochemistry as a screening tool in Muir–Torre syndrome, Mod. Pathol. 21 (2008) 159–164. [8] A. Mojtahed, I. Schrijver, J.M. Ford, T.A. Longacre, R.K. Pai, A two-antibody mismatch repair protein immunohistochemistry screening approach for colorectal carcinomas, skin sebaceous tumors, and gynecologic tract carcinomas, Mod. Pathol. 24 (2011) 1004–1014. [9] L. Orta, D.S. Klimstra, J. Qin, P. Mecca, L.H. Tang, K.J. Busam, et al., Towards identification of hereditary DNA mismatch repair deficiency: sebaceous neoplasm warrants routine immunohistochemical screening regardless of patient’s age or other clinical characteristics, Am. J. Surg. Pathol. 33 (2009) 934–944. [10] E.F. Plocharczyk, W.L. Frankel, H. Hampel, S.B. Peters, Mismatch repair protein deficiency is common in sebaceous neoplasms and suggests the importance of screening for Lynch syndrome, Am. J. Dermatopathol. 35 (2013) 191–195. [11] R. Kruse, A. Rutten, N. Schweiger, E. Jakob, M. Mathiak, P. Propping, et al., Frequency of microsatellite instability in unselected sebaceous gland neoplasias and hyperplasias, J. Invest. Dermatol. 120 (2003) 858–864. [12] M.M. Entius, J.J. Keller, P. Drillenburg, K.C. Kuypers, F.M. Giardiello, G.J. Offerhaus, Microsatellite instability and expression of hMLH-1 and hMSH-2 in sebaceous gland carcinomas as markers for Muir–Torre syndrome, Clin. Cancer Res. 6 (2000) 1784–1789. [13] P. Machin, L. Catasus, C. Pons, J. Munoz, J.M. Conde-Zurita, J. Balmana, et al., Microsatellite instability and immunostaining for MSH-2 and MLH-1 in cutaneous and internal tumors from patients with the Muir–Torre syndrome, J. Cutan. Pathol. 29 (2002) 415–420. [14] M. Mathiak, A. Rutten, E. Mangold, H.P. Fischer, T. Ruzicka, W. Friedl, et al., Loss of DNA mismatch repair proteins in skin tumors from patients with Muir–Torre syndrome and MSH2 or MLH1 germline mutations: establishment of immunohistochemical analysis as a screening test, Am. J. Surg. Pathol. 26 (2002) 338–343. [15] A. Morales-Burgos, J.L. Sanchez, L.D. Figueroa, W.E. De Jesus-Monge, M.R. CruzCorrea, C. Gonzalez-Keelan, et al., MSH-2 and MLH-1 protein expression in Muir–Torre syndrome-related and sporadic sebaceous neoplasms, P. R. Health Sci. J. 27 (2008) 322–327. [16] G. Ponti, L. Losi, G.C. Di, L. Roncucci, M. Pedroni, A. Scarselli, et al., Identification of Muir–Torre syndrome among patients with sebaceous tumors and keratoacanthomas: role of clinical features, microsatellite instability, and immunohistochemistry, Cancer 103 (2005) 1018–1025. [17] K.M. Cornejo, L. Hutchinson, A. Deng, K. Tomaszewicz, M. Welch, S. Lyle, et al., BRAF/KRAS gene sequencing of sebaceous neoplasms after mismatch repair protein analysis, Hum. Pathol. 45 (2014) 1213–1220. [18] E. Glogowski, S. Boyar, K. Sarrel, C. LeBlang, E. Utay, J. Shia, et al., Unexplained immunohistochemical DNA mismatch repair deficiency, Fam. Cancer 10 (2011) 726. [19] H. Hampel, W.L. Frankel, E. Martin, M. Arnold, K. Khanduja, P. Kuebler, et al., Feasibility of screening for Lynch syndrome among patients with colorectal cancer, J. Clin. Oncol. 26 (2008) 5783–5788. [20] L.H. Jensen, J. Lindebjerg, L. Byriel, S. Kolvraa, D.G. Cruger, Strategy in clinical practice for classification of unselected colorectal tumours based on mismatch repair deficiency, Color Dis. 10 (2008) 490–497. [21] A. Paya, C. Alenda, L. Perez-Carbonell, E. Rojas, J.L. Soto, C. Guillen, et al., Utility of p16 immunohistochemistry for the identification of Lynch syndrome, Clin. Cancer Res. 15 (2009) 3156–3162. [22] T. Agander, L. Lindberg, L. Brixen, H. Okkels, M. Haska, I. Bernstein, et al., The utility of P16 in identifying Lynch syndrome (LS), APMIS 121 (2013) 22. [23] O. Abbas, M. Mahalingam, Cutaneous sebaceous neoplasms as markers of Muir–Torre syndrome: a diagnostic algorithm, J. Cutan. Pathol. 36 (2009) 613–619. [24] B. Halvarsson, A. Lindblom, L. Johansson, K. Lagerstedt, M. Nilbert, Loss of mismatch repair protein immunostaining in colorectal adenomas from patients with hereditary nonpolyposis colorectal cancer, Mod. Pathol. 18 (2005) 1095–1101. [25] R.S. Singh, W. Grayson, M. Redston, A.H. Diwan, C.L. Warneke, P.H. McKee, et al., Site and tumor type predicts DNA mismatch repair status in cutaneous sebaceous neoplasia, Am. J. Surg. Pathol. 32 (2008) 936–942.
Interpretation of MMR-protein staining The challenge in the interpretation of MMR-protein immunostains of CRC has repeatedly been reported, contributing to observer variations [28], specifically in MSI tumours [29]. Such difficulties seem less prominent in case of SA/SC, a quality, which has been ascribed to the high proliferative activity of epidermally derived lesions [30] and less common fixation problems, given the limited dimensions of these specimens. Accordingly, Machin et al. [13] recorded equivocal MLH1 staining of CRC despite unequivocal loss of MLH1 of a subsequently examined SA, corroborating the low sensitivity in detecting MLH1 germline mutation by IHC of CRCs [31]. Crush artefact of most lesional cells in some small specimens may, on the other hand, prove a limiting factor as noticed in one of our cases, in which the determination of the MMR protein status required additional deeper sections. Aberrant staining patterns, such as cytoplasmic staining, perinuclear staining and “dot-like” nucleolar staining occasionally appear in the CRC context [30] with the risk of confounding the interpretation. These qualities were not a feature in the current series and have not, to our knowledge, been addressed in prior series of sebocytic neoplasms.
Recommended histopathological handling of sebaceous neoplasms Given the high yield of MMR-protein deficiency among sebaceous neoplasms, coupled with their relatively low occurrence in the general population (in Denmark some 45 newly diagnosed cases per year), we consider IHC screening as a pivotal initial step and recommend the use of a 4-panel approach as routinely used in CRCs, designed to capture MTS candidate lesions. The fact that these neoplasms may predate visceral malignancies [6] provides an added incentive to follow this approach. Furthermore, immunostaining is an inexpensive, widely available method, compared to the more costly microsatellite instability (MSI) analyses. Our preliminary p16 investigation of MLH1deficient cases proved none-informative. Additional research efforts on larger cohorts directed at this issue are, however, desirable. In this manner the awareness on MTS in the pathology community will be increased, whereby genetic counselling as well as lifelong cancer surveillance of both probands and at-risk family members more likely will be undertaken, reducing cancer mortality.
82
M. Boennelycke et al. / Pathology – Research and Practice 211 (2015) 78–82
[26] E. Mangold, N. Rahner, N. Friedrichs, R. Buettner, C. Pagenstecher, S. Aretz, et al., MSH6 mutation in Muir–Torre syndrome: could this be a rare finding? Br. J. Dermatol. 156 (2007) 158–162. [27] A. Fernandez-Flores, Considerations on the performance of immunohistochemistry for mismatch repair gene proteins in cases of sebaceous neoplasms and keratoacanthomas with reference to Muir–Torre syndrome, Am. J. Dermatopathol. 34 (2012) 416–422. [28] L. Klarskov, S. Ladelund, S. Holck, K. Roenlund, J. Lindebjerg, J. Elebro, et al., Interobserver variability in the evaluation of mismatch repair protein immunostaining, Hum. Pathol. 41 (2010) 1387–1396. [29] L.I. Overbeek, M.J. Ligtenberg, R.W. Willems, R.P. Hermens, W.A. Blokx, S.V. Dubois, et al., Interpretation of immunohistochemistry for mismatch repair
proteins is only reliable in a specialized setting, Am. J. Surg. Pathol. 32 (2008) 1246–1251. [30] J. Shia, S. Holck, G. Depetris, J.K. Greenson, D.S. Klimstra, Lynch syndrome-associated neoplasms: a discussion on histopathology and immunohistochemistry, Fam. Cancer 12 (2013) 241–260, http://dx.doi.org/ 10.1007/s10689-013-9612-4. [31] J. Shia, D.S. Klimstra, K. Nafa, K. Offit, J.G. Guillem, A.J. Markowitz, et al., Value of immunohistochemical detection of DNA mismatch repair proteins in predicting germline mutation in hereditary colorectal neoplasms, Am. J. Surg. Pathol. 29 (2005) 96–104.
