- Email: [email protected]
Lymph node melanocytic nevi: Pathogenesis and differential diagnoses, with special reference to p16 reactivity
G Model PRP-51359; No. of Pages 8
ARTICLE IN PRESS Pathology – Research and Practice xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Pathology – Research and Practice journal homepage: www.elsevier.com/locate/prp
Original article
Lymph node melanocytic nevi: Pathogenesis and differential diagnoses, with special reference to p16 reactivity Simonetta Piana a,∗ , Elena Tagliavini a , Moira Ragazzi a , Magda Zanelli a , Iris Zalaudek b , Alessia Ciarrocchi c , Riccardo Valli a a
Pathology Unit, IRCCS-Arcispedale Santa Maria Nuova, Reggio Emilia, Italy Department of Dermatology, Medical University of Graz, Graz, Austria c Laboratory of Translational Research, IRCCS-Arcispedale Santa Maria Nuova, Reggio Emilia, Italy b
a r t i c l e
i n f o
Article history: Received 4 August 2014 Received in revised form 23 September 2014 Accepted 9 January 2015 Keywords: Melanocytic nevi Lymph node Nodal nevi Nevogenesis Metastasis P16
a b s t r a c t Lymph node nevi (NN) have been occasionally described, yet little is currently known on their origin. According to a theoretical model of nevogenesis, the dissemination of nevus progenitor cells through lymphatic routes is responsible for the development of both nodal and skin nevi. The true incidence of NN is largely unknown but it has been reported to vary from 0.017% to as high as 22%. The frequency of NN nevi has increased since the introduction of sentinel lymph node mapping as a routine prognostic procedure in breast cancer and melanoma. The aim of this study was to analyze the frequency and morphological findings of NN, to discuss possible pathogenetic pathways in their evolution, and to verify the consistency of p16 immunostaining in the critical differential approach between NN and melanoma metastases. We therefore morphologically and immunohistochemically evaluated a series of 60 NN from 58 patients. In 21 patients, the lymph nodes had been removed during the staging for a skin melanoma; in all these patients NN immunostaining with p16 was strongly positive and p16 proved to be a reliable marker for the crucial differential diagnosis between NN and melanoma metastasis, strongly reacting in NN and lacking in melanoma deposits. A deeper knowledge on NN could help to clarify some important topics such as lymph node metastatic melanoma with unknown primary and the current debate on the lymph node involvement from atypical spitzoid tumors. © 2015 Elsevier GmbH. All rights reserved.
Introduction Since the original description by Stewart and Copeland in 1931 [1], melanocytic inclusions in lymph nodes, also called node nevi (NN), have been reported episodically [2–21] and their histological variants have been further elucidated [22–28]. The true incidence of NN is largely unknown but it has been reported to vary from 0.017% to as high as 22% [13,15,16,19]. The reported frequency of NN nevi has increased since the introduction of sentinel lymph node (SLN) mapping as a routine prognostic procedure in breast cancer and melanoma. This is because this technique allows for the detection of melanocytic rests and epithelial inclusions [29] within metastatic and non-metastatic lymph
∗ Corresponding author. Tel.: +0039 0522 295919; fax: +0039 0522 296945. E-mail address: [email protected] (S. Piana).
nodes; the presence of melanocytic inclusions increases if sectioning search is extensive. NN mostly present as small aggregates of normal-looking melanocytes located within the fibrous components of the lymph node, mainly in the capsule and the trabeculae. Though rarely, they can also be found in the parenchyma or in the lymphatic channels [18]. Due to their location, besides being an incidental histological oddity, they can represent a diagnostic challenge in lymph nodes excised during the workup for a malignancy, especially for melanoma. In order to discriminate NN from metastatic melanoma, p16 immunostaining has been proposed as a reliable tool, since it has been shown that NN display an extensive nuclear and cytoplasmic staining for p16, at variance with melanomas [30]. The aim of this study was to analyze the frequency and morphological findings of NN in neoplastic and non-neoplastic clinical settings, to discuss possible pathogenetic pathways in their evolution, and to verify the consistency of p16 immunostaining in the critical differential approach between NN and melanoma metastases.
http://dx.doi.org/10.1016/j.prp.2015.01.003 0344-0338/© 2015 Elsevier GmbH. All rights reserved.
Please cite this article in press as: S. Piana, et al., Lymph node melanocytic nevi: Pathogenesis and differential diagnoses, with special reference to p16 reactivity, Pathol. – Res. Pract (2015), http://dx.doi.org/10.1016/j.prp.2015.01.003
G Model
ARTICLE IN PRESS
PRP-51359; No. of Pages 8
S. Piana et al. / Pathology – Research and Practice xxx (2015) xxx–xxx
2
Materials and methods The computerized files of the Pathology Unit of IRCCS-S.Maria Nuova Hospital in Reggio Emilia were retrospectively searched for all the lymph node specimens between January 1, 2000 and January 31, 2014. Among these, all cases of lymph nodes harboring one or more melanocytic rests were retrieved and histopathologically revised. For all retrieved cases of NN, demographic data, the anatomic basin, the major associated diseases, the number of the revised slides, and the reason for excision (i.e. SLN or non-SLN procedure) were recorded. The SLNs were processed as previously described according to the breast cancer screening and melanoma staging protocols [31,32]. The non-SLNs corresponded to the elective lymphadenectomies, after a positive SLN. As a rule, non-SLNs have not been examined with step sections and immunohistochemical stains. This issue, together with the different protocols employed for breast SLN and melanoma SLN (a serial section evaluation of tissue in the former and four serial sections in the latter) may represent an underestimation bias of NN frequency in the SLNs from melanoma patients and in non-SLNs. On histological review, the number of melanocytic aggregates and their location within the lymph node were evaluated. Size estimation of the NN was performed by integrating the major axis of the lesions with the number of step sections in cases of SLNs. For SLNs from melanoma patients, one or more slides stained with S100 (polyclonal, Ventana, Tucson, AZ) were routinely available for analysis. In cases of melanoma SLNs harboring NN, the primary melanomas were revised to compare the morphologic features. Using an automated immunostainer (Benchmark, Ventana, Tucson, AZ), all cases of NN in patients with melanoma were stained with p16 (E6H4, Ventana, Tucson, AZ) and ki67 (clone MIB1, Dako, Glostrup, Denmark) to evaluate proliferative activity. Selected cases were also stained with Melan A/MART-1 (A103, Ventana, Tucson, AZ, to confirm the presence of melanocytic nests. Results During the study period, the Pathology Unit of IRCCS-S.Maria Nuova Hospital accepted a total of 26,240 lymph node specimens, including all types of lymph node resection, i.e. tru-cut biopsies, lymphadenectomies for lymphomas or reactive lymphadenitis, SLNs and extended lymphadenectomies for oncologic staging. Among all these specimens there were 416 (1.5%) SLNs from 399 patients with melanoma and 4276 (17%) SLNs from 4205 patients with breast cancer. Among all cases, the presence of an NN was recorded in 60 (0.2%) cases from 58 patients, including 17 (29.3%) men and 41 (70.6%) women, with a mean age of 55 years (range 29–84 yr). Among all the SLNs analyzed by our Institution in the study period, NN were present in 18 out of 416 melanoma SLNs (4.3%)
and in 12 out of 4,276 breast SLNs (0.3%). In our series, therefore, NN in SLNs represented 50% of all the cases. The clinical features in relation to the primary diseases are listed in Table 1. In 55 patients, the lymph nodes had been removed during staging for a neoplasia. In 32 of these patients (55%), the reported primary disease was breast cancer (26 invasive ductal carcinoma [IDC], 5 in situ ductal carcinoma [ISDC], 1 invasive lobular carcinoma [ILC]), in 21 patients (27%) cutaneous melanoma, in 1 patient (1.6%) invasive squamous cell carcinoma of the larynx, and in 1 further patient (1.6%) invasive adenocarcinoma of the prostate. In 3 patients (5%), a single enlarged lymph node was removed with the clinical suspicion of a lymphoma and with the final pathological report of reactive lymphadenopathy. In all the 58 patients, only 1 lymph node harbored an NN, except in two female patients, one of whom with breast cancer and the other with melanoma. They featured NN in 2 lymph nodes, one SLN and one non-SLN, in axillary and inguinal region, respectively. The lymph node site was axillary in 45 cases (75%; 23 right and 22 left), inguinal in 10 cases (16.6%; 6 right and 4 left), cervical in 4 cases (6.6%; 1 right and 3 left), and right obturator in 1 case (1.6%). The histologic features in relation to the primary diseases are listed in Table 2. The number of melanocytic nests within the same lymph node varied from 1 to 6, with a mean number greater in nodes with reactive lymphadenopathies. The mean size of NN was 1.4 mm (range 0.1 to 7 mm). NN morphologic features were generally repetitive. As a rule, nests were composed of monomorphous bland-appearing melanocytes, mostly located in the lymph node supporting stroma, the capsule, or along the trabeculae (Fig. 1A). A peculiar perivascular arrangement was often seen (Fig. 1B and C), while lymph node parenchyma involvement was present in only 3 cases (Fig. 1E). Mitotic figures, nucleoli, pycnotic nuclei, and necrotic foci were always absent. Immunohistochemical stainings with S100 and Melan A were diffusely positive in all cases, whereas immunohistochemical staining with ki67 confirmed the absence of proliferative activity (Fig. 1D). Four cases showed distinct characteristics: in two cases (a reactive lymphadenopathy and one SLN during breast cancer workup), the melanocytic rests had blue nevus-like features, with dendritic, hyperpigmented melanocytes disrupting the collagenous fibers of the lymph node capsule (Fig. 2A and B). In the other two cases, the melanocytic aggregates were particularly numerous and large. One of these last two cases was a non-SLN metastatic node in the axillary region of a 61-year-old woman with a mucinous invasive carcinoma of the breast. The melanocytic nests were multiple and grew along the capsule and the trabeculae, with few collections present even in the parenchyma. The second case was an inguinal SLN in a 57-year-old female patient with a 2.5 mm thick nodular melanoma arising on a giant congenital nevus occupying the entire abdominal wall. The entire circumference of the lymph node was
Table 1 Clinical features of patients with NN. n. Lymph nodes* (%)
Males
Females*
Mean age (yr)
Axillary (right)
SLN breast cancer Non-SLN breast cancer SLN melanoma Non-SLN melanoma Non-SLN other neoplasias Lymphadenopathies
12 (20%) 21 (35%) 18 (30%) 4 (6.6%) 2 (3.3%) 3 (5%)
0 0 12 2 2 1 17
12 20 6 1 0 2 41
59 59 52 44 58 31
12 (6) 21 (12) 10 (4) 1 (0)
Total
60
58
1 (1) 45
Inguinal (right)
6 (4) 2 (1) 2 (1) 10
Cervical (right)
Obturator (right)
2 (1) 1 (0) 1 (0)
1 (1)
4
1
60
SLN: sentinel lymph node; yrs: years. * Two female patients had NN in two adjacent lymph nodes.
Please cite this article in press as: S. Piana, et al., Lymph node melanocytic nevi: Pathogenesis and differential diagnoses, with special reference to p16 reactivity, Pathol. – Res. Pract (2015), http://dx.doi.org/10.1016/j.prp.2015.01.003
G Model
ARTICLE IN PRESS
PRP-51359; No. of Pages 8
S. Piana et al. / Pathology – Research and Practice xxx (2015) xxx–xxx
3
Table 2 Histologic features of NN. n. Lymph nodes
n. Nests * (mean)
Dimension* (mean, mm)
SLN breast cancer Non SLN breast cancer SLN melanoma Non SLN melanoma Non SLN other neoplasias lymphadenopathies
12 21 18 4 2 3
1.1 1.3 1.4 1 1.5 6
1.2 1.5 1.1 1.2 1.2 2.6
TOT
60
2.0
1.4
*
Parenchymal involvement
Special features 1 Blue nevus-like
1 1
1 3
1 Blue nevus-like
Metastasis 1 4 3 2 0 0 10
Except two cases with exceptionally large NN.
Table 3 Clinicopathologic features of primary skin melanomas. Melanoma SLN n. Patients Mean age (yr) (range) M/F Mean Breslow thickness (mm) vascular invasion Associated nevi Metastasis
18 52 (32–78) 12/6 2.3 1 6 3
Melanoma non-SLN 3 44 (29–73) 2/1 2.6 0 1 2
occupied by multiple confluent aggregates of epithelioid or weakly pigmented spindle melanocytes extending to the inner trabeculae and, focally, to the parenchyma (Fig. 2C and D). The same aggregates, highlighted by the immunohistochemical stain with S100, were also present in the perinodal fat as short fascicles or nodular collections and around the vessels, whose walls were focally infiltrated (Fig. 2E and F). Metastasis from the primary malignancy coexisted in 10 cases of NN (16.6%): 5 IDCs of the breast and 5 melanomas. The revision of the cutaneous primary melanomas confirmed its morphologic similarity with the metastatic lymph node deposits and the unquestionable histologic differences with the bland looking melanocytic nevi. Notably, in 7 (30%) out of 21 patients, melanoma arose within or adjacent to a congenital-type dermal nevus. Vascular invasion was recorded in only 1 case of melanoma. Immunostaining with p16 was strongly positive in all the 22 NN in melanoma patients with a strong nuclear and cytoplasmic staining; in 5 lymph nodes harboring both one or more NN and melanomatous metastatic deposits, melanoma aggregates were consistently negative (Figs. 3 and 4). Ki67 immunostaining confirmed the presence of variable proliferative activity in melanoma metastases. Table 3 reports the main clinical and histologic features of the 21 primary skin melanomas with NN in SLN and non-SLN. One female patient with melanoma had NN in 2 lymph nodes in inguinal region, one SLN and one non-SLN (see above). Discussion The presence of normal looking heterotopic epithelial tissue is well known in specific lymph node basins, i.e. in submandibular, axillary, or in pelvic nodes, where it represents an important diagnostic pitfall and a “false positive” finding during the staging of a malignancy. While it is accepted that Müllerian-derived organs and breast are the major source of epithelial inclusions in axillary and pelvic lymph nodes, respectively [33], little is currently known on the origin of NN. The two most reliable theories, still en vogue after more than 40 years, argue that NN are caused by embolic transfer of skin melanocytes to the draining lymph node [2,16] or, alternatively, by an aberrant embryologic migration from the neural crest to their permanent anatomic sites [2,3].
The former theory seems to be confirmed by the frequent finding of melanocyte aggregates protruding into both the lymphatic and hematic vascular lumina, in the context of congenital nevi. It is conceivable that the same aggregates can detach from the wall and be transported to the corresponding lymph nodes. This interpretation is in line with the up to 30% rate of melanoma-associated congenital nevi we report in our series. Indeed, the presence of NN supports a theoretical model of nevogenesis, according to which the dissemination of nevus progenitor cells through lymphatic routes is responsible for the development of both nodal and cutaneous nevi [34]. In other words, nevogenesis is considered a benign metastatic process that can stop in the lymph node capsule or in the skin. While the theory on embolic transfer to draining nodes does not explain the presence of NN in deep locations, i.e. in an obturator lymph node, the “benign metastasis” approach does: melanocytes enter systemic lymphatic circulation and implant at a distance. The second hypothesis relies on the perturbation of the embryogenic melanoblast migration to the developing skin [35], with consequent deposition of melanocytic precursors into the developing lymph nodes. The latter incorporate melanoblasts in their framework and allow their morphologic maturation. As it has been postulated that the formation of multiple, common or atypical (dysplastic) nevi results from an alterated embryogenic melanoblast migration [36], this interpretation explains well why we find melanocytic rests more frequently in lymph node basins draining the teguments of patients with melanoma, who presumably have more moles (high melanocytic burden), typical or atypical. Moreover, this model plausibly explains both the involvement of deeply located nodes and the intracapsular position of the melanocytic rests, which is almost by definition remarkably distinct from the intrasinusoidal location of the metastatic emboli. A third point of view could combine the previous two: the extravascular cellular migration may be an important pathway in the development of congenital nevi. In other words, melanoblasts reach their destination by migrating around the vessels, not inside them [37]. This theory could explain the “pseudoinfiltrative” and extranodal pattern of distribution of the melanocytic rests in the cases associated with the giant congenital nevus. Indeed, we noted an unusual perivascolar location of the melanocytic rests within the perinodal adipose tissue. This view has also been called upon to justify the presence of SLN metastasis from atypical spitzoid melanocytic neoplasms [38], a very controversial issue that probably shares some points with the etiopathogenesis of melanocytic rests. However, although extremely intriguing, any discussion of this is beyond the aim of the present work. These theories notwithstanding, many points remain unexplained. First, why is the relative frequency of nodal rests higher in melanoma patients than in patients with other malignancies, i.e. breast cancer, despite a SLN procedure? In our series, the incidence of nodal rests was 4.3% in melanoma patients compared to only 0.3% in breast cancer patients. If one accepts that there is an association between nodal nevus cells and congenital nevi [17], one could claim that patients with
Please cite this article in press as: S. Piana, et al., Lymph node melanocytic nevi: Pathogenesis and differential diagnoses, with special reference to p16 reactivity, Pathol. – Res. Pract (2015), http://dx.doi.org/10.1016/j.prp.2015.01.003
G Model PRP-51359; No. of Pages 8 4
ARTICLE IN PRESS S. Piana et al. / Pathology – Research and Practice xxx (2015) xxx–xxx
Fig. 1. (A) Melanocytic nests are composed of monomorphous bland-appearing melanocytes, mostly located in the lymph node supporting stroma, the capsule, or along the trabeculae, with a typical perivascular arrangement (insert). (B)–(C) At higher power view, melanocytic infiltration of the vessel walls is evident. (D) The immunohistochemical staining with ki67 confirms the absence of proliferative activity; as a control, two stained lymphocytic nuclei are evident. (E) The lymph node parenchyma involvement is rare (arrows) but can be seen besides the capsular melanocytic aggregates. (F)–(G) Immunohistochemical stainings with Melan A (F) and p 16 (G) confirm the presence of bland melanocytes within the lymphoid tissue (arrows).
melanoma have a higher incidence of NN because they generally have a greater melanocytic burden, i.e. a higher number of nevi, including nevi of the congenital type, than do non-melanoma patients. In addition, it is unclear why NN are more frequent than epithelial inclusions in axillary nodes; in fact, the latter, although
they have not been found in one large series [39], have frequently been described in the literature [29]. Do these heterotopic melanocytic rests possess any malignant potential? A recent paper has demonstrated that NN can harbor the BRAF V600E mutation [40]. This finding is not surprising as we
Please cite this article in press as: S. Piana, et al., Lymph node melanocytic nevi: Pathogenesis and differential diagnoses, with special reference to p16 reactivity, Pathol. – Res. Pract (2015), http://dx.doi.org/10.1016/j.prp.2015.01.003
G Model PRP-51359; No. of Pages 8
ARTICLE IN PRESS S. Piana et al. / Pathology – Research and Practice xxx (2015) xxx–xxx
5
Fig. 2. (A)–(B) NN can show blue nevus-like features, with dendritic, hyperpigmented melanocytes disrupting the collagenous fibers of the lymph node capsule. (C)–(D) In rare cases, the melanocytic aggregates are particularly large and can be noticed in the perinodal fat. They can be easily highlighted around the vessels by the immunohistochemical stain with S100 (E)–(F)
expect that up to 80% of melanocytic nevi carry the same mutation [41]. We do agree with those authors who claim that neither morphology nor the presence of BRAF mutation predicts the chance of malignant transformation of a nevus into a melanoma [42] and we arbitrarily transport this observation to NN as well. However, it is conceivable that melanocytic rests can undergo neoplastic progression. Malignant transformation of heterotopic epithelium in lymph nodes has been described in the major anatomic draining sites [43–47]; few cases of bona fide primary lymph node Merkel cell carcinoma have been described, too [48,49]. Although malignant transformation has not been reported for NN, as it has been for nodal epithelial inclusions, the concept of lymph node metastatic melanoma with unknown primary site has commonly been accepted since 1952 [50] and has been repeatedly
analyzed [51]. In 1977, Ridolfi et al. [5] speculated that nodal nevi can account for some of the lymph node melanomas with unknown primary. In 1986, Shenoy et al. described a putative case of melanoma primary in the lymph node [52], with a morphologic transition of melanocytes which they considered the missing link between NN and melanoma. The development of a melanoma from ectopic melanocytes located in the lymph node capsule could explain those cases of metastatic melanoma without a primary but this hypothesis has never been re-examined, since it is dogmatically assumed that a skin or mucosal primitive localization must exist; if it does not, then the primary lesion must have regressed. Therefore, to date, Shenoy et al.’s theory remains speculative. Despite many pathogenetic questions still to be answered, pathologists know that the main problem related to NN is the
Please cite this article in press as: S. Piana, et al., Lymph node melanocytic nevi: Pathogenesis and differential diagnoses, with special reference to p16 reactivity, Pathol. – Res. Pract (2015), http://dx.doi.org/10.1016/j.prp.2015.01.003
G Model PRP-51359; No. of Pages 8 6
ARTICLE IN PRESS S. Piana et al. / Pathology – Research and Practice xxx (2015) xxx–xxx
Fig. 3. (A)–(B) A capsular NN in a patient with melanoma. The melanocytes are small sized and strongly immunoreactive to p16 with an intense nuclear pattern (B). (C) The melanoma micrometastasis is made up by neoplastic atypical, epithelioid melanocytes (arrows), completely negative with p16 immunostain (D).
differential diagnosis with a metastatic deposit. This is why some papers have searched for further ancillary methods to assist in the diagnosis of NN [30,53–55]. In our experience, p16 has proven to be a reliable marker to distinguish NN from melanoma metastases, strongly reacting in NN and lacking in melanoma deposits. We can thus conclude that p16, together with ki67 and eventually
other routine melanocytic markers, as HMB45 and MelanA [56], can reasonably be used in everyday life for the crucial differential diagnosis between NN and melanoma metastasis. This scenario is coherent with the physiological role of p16, which is a tumor suppressor protein involved in the regulation of cell cycle and senescence. P16 overexpression seems to represent a
Fig. 4. (A) A capsular NN in a patient with melanoma. (B) At higher magnification, NN cells are monomorphous and cytologically bland. (C)–(D) Immunohistochemical stainings with p16 (C) and Melan A are diffusely positive in nodal nevi; p16 shows a diffuse nuclear and cytoplasmic stain, with few negative cells. (E) The same lymph node also harbors a small melanoma metastatic deposit. (F)–(G)–(H) The micrometastasis is immunoreactive with S100 (F) and Melan A (H) but negative with p16 (G).
Please cite this article in press as: S. Piana, et al., Lymph node melanocytic nevi: Pathogenesis and differential diagnoses, with special reference to p16 reactivity, Pathol. – Res. Pract (2015), http://dx.doi.org/10.1016/j.prp.2015.01.003
G Model PRP-51359; No. of Pages 8
ARTICLE IN PRESS S. Piana et al. / Pathology – Research and Practice xxx (2015) xxx–xxx
compensatory cell mechanism to BRAFV600E-driven proliferation in human melanocytes [57]; the presence of both BRAFV600E [39] and p16 [30] is therefore likely in lesions like NN which are in a apparently growth-arrested state, as confirmed by immunohistochemical staining with ki67. In conclusion, after more than 80 years since the original description, the definitive diagnosis of NN relies on an accurate morphologic analysis in hematoxylin-eosin stained sections and on the cautious employment of the immunohistochemical markers. Despite the growing interest on this topic, the ontogenesis of NN is still unclear. This knowledge that may appear insignificant in a surgical pathology practice, could have a key role to solve some important topics such as lymph node metastatic melanoma with unknown primary site and the current debate on the presence of melanocytic deposits in SLN from atypical spitzoid tumors. In this context, the pathogenesis of NN may be a promising area of study in cancer behavior research. Source of funding None. References [1] F.W. Stewart, M.M. Copeland, Neurogenic sarcoma, Am. J. Cancer 15 (1931) 1235–1320. [2] W.T. Johnson, E.B. Helwig, Benign nevus cell in the capsule of lymph nodes, Cancer 23 (1969) 747–753. [3] W.R. Hart, Primary nevus of a lymph node, Am. J. Clin. Pathol. 55 (1971) 88–92. [4] S.W. McCarthy, A.A. Palmer, P.M. Bale, E. Hirst, Naevus cells in lymph nodes, Pathology 6 (1974) 351–358. [5] R.L. Ridolfi, P.P. Rosen, H. Thaler, Nevus cell aggregates with lymph nodes: estimated frequency and clinical significance, Cancer 39 (1977) 164–171. [6] J.L. Jensen, L.W. Correll, Nevus cell aggregates in submandibular lymph nodes, Oral Surg. Oral Med. Oral Pathol. 50 (1980) 552–556. [7] G. Ioannides, Lymph node aggregates of nevus cells, in: A.B. Ackerman (Ed.), Pathology of Malignant Melanoma, Masson Publishing, New York, NY, 1981, pp. 297–300. [8] S. Andreola, C. Clemente, Nevus cells in axillary lymph nodes from radical mastectomy specimens, Pathol. Res. Pract. 179 (1985) 616–618. [9] H.M. Yadzi, Nevus cell aggregates associated with lymph nodes. Immunohistochemical observations, Arch. Pathol. Lab. Med. 109 (1985) 1044–1046. [10] C. Subramony, J.R. Lewin, Nevus cells within lymph nodes: possible metastases from a benign intradermal nevus, Am. J. Clin. Pathol. 84 (1985) 220–223. [11] R.H. Hruban, F. Eckert, B. Baricevic, Melanocytes of a melanocytic nevus in a lymph node from a patient with a primary cutaneous melanoma associated with a small congenital nevus, Am. J. Dermatopathol. 12 (1990) 402–407. [12] K. Hara, Melanocytic lesions in lymph nodes associated with congenital naevus, Histopathology 23 (1993) 445–451. [13] N.C. Bautista, S. Cohen, K.H. Anders, Benign melanocytic nevus cells in axillary lymph nodes: a prospective incidence and immunohistochemical study with literature review, Am. J. Clin. Pathol. 102 (1994) 102–108. [14] C.J. Fisher, S. Hill, R.R. Millis, Benign lymph node inclusions mimicking metastatic carcinoma, J. Clin. Pathol. 47 (1994) 245–247. [15] K.F. Carson, D.R. Wen, P.X. Li, et al., Nodal nevi and cutaneous melanomas, Am. J. Surg. Pathol. 20 (1996) 834–840. [16] D. Fontaine, W. Parkhill, W. Greer, N. Walsh, Nevus cells in lymph nodes: an association with congenital cutaneous nevi, Am. J. Dermatopathol. 24 (2002) 1–5. [17] D.A. Biddle, H.L. Evans, B.L. Kemp, A.K. El-Naggar, J.D. Harvell, W.L. White, S.S. Iskandar, V.G. Prieto, Intraparenchymal nevus cell aggregates in lymph nodes: a possible diagnostic pitfall with malignant melanoma and carcinoma, Am. J. Surg. Pathol. 27 (2003) 673–681. [18] J.W. Patterson, Nevus cell aggregates in lymph node, Am. J. Clin. Pathol. 121 (2004) 13–15. [19] J.B. Holt, O.P. Sangueza, E.A. Levine, P. Shen, S. Bergman, K.R. Geisinger, A.J. Creager, Nodal melanocytic nevi in sentinel lymph nodes. Correlation with melanoma-associated cutaneous nevi, Am. J. Clin. Pathol. 121 (2004) 58–63. [20] T. Gambichler, L. Scholl, M. Stücker, F.G. Bechara, K. Hoffmann, P. Altmeyer, N. Othlinghaus, Clinical characteristics and survival data of melanoma patients with nevus cell aggregates within sentinel lymph nodes, Am. J. Clin. Pathol. 139 (2013) 566–573. [21] J.G. Azzopardi, C.M. Ross, G. Frizzera, Blue naevi of lymph node capsule, Histopathology 1 (1977) 451–461. [22] J. Roth, Ectopic blue nevi in lymph nodes, in: A.B. Ackerman (Ed.), Pathology of malignant melanoma, Masson Publishing, New York, NY, 1981, pp. 293–296. [23] R.L. Goldman, Blue nevus of lymph node capsule: report of a unique case, Histopathology 5 (1981) 445.
7
[24] J. Epstein, R.A. Erlandson, P.P. Rosen, Nodal blue nevi; a study of three cases, Am. J. Surg. Pathol. 8 (1984) 907–915. [25] C. Urso, Nodal melanocytic nevus with baloon-cell change (nodal baloon-cell nevus), J. Cutan. Pathol. 35 (2008) 672–676. [26] N. Misago, K. Nagase, S. Toda, Y. Shinoda, S. Koba, Y. Narisawa, Cellular blue nevus with nevus cells in a sentinel lymph node, Eur. J. Dermatol. 18 (2008) 586–589. [27] S.M.K. Nahar Begum, M. Lomme, M. Ruhul Quddus, Nodal combined blue nevus and benign nevus cells in multiple axillary sentinel nodes in a patient with breast carcinoma: report of a case, Int. J. Surg. Pathol. (2013), http://dx.doi.org/10.1177/1066896913509008. [28] L. Dohese, T. Ferringer, Nodal blue nevus: a pitfall in lymph node biopsies, J. Cutan. Pathol. 37 (2010) 102–104. [29] G. Fellegara, M.L. Carcangiu, J. Rosai, Benign epithelial inclusions in axillary lymph nodes: report of 18 cases and review of the literature, Am. J. Surg. Pathol. 35 (2011) 1123–1133. [30] D. Mihic-Probst, P. Saremaslani, P. Komminoth, P.U. Heitz, Immunostaining for the tumour suppressor gene p16 product is a useful marker to differentiate melanoma metastasis from lymph-node nevus, Virchow Arch. 443 (2003) 745–751. [31] N. Perry, M. Broeders, de Törnberg, S. Wolf C., R. Holland, L. von Karsa, European guidelines for quality assurance in breast cancer screening and diagnosis. Fourth edition-summary document, Ann. Oncol. 19 (2008) 614–622. [32] V.G. Prieto, Sentinel Lymph nodes in cutaneous melanoma, Clin. Lab. Med. 31 (2011) 301–310. [33] R.A. Warnke, L.M. Weiss, J.K.C. Chan, M.L. Cleary, R.F. Dorfman, Non hematolymphoid tumors and tumor-like lesions, in: J. Rosai, L. Sobin (Eds.), Atlas of Tumor Pathology, Third Series. Tumors of the Lymph Nodes and Spleen, Armed Forces Institute of Pathology, Washington DC, 1994, pp. 474–475. [34] A.L. Ross, M.I. Sanchez, J.M. Grichnik, Nevogenesis: a benign metastatic process? ISRN Dermatol. (2011), http://dx.doi.org/10.5402/2011/813513. [35] I. Zalaudek, C. Catricalà, E. Moscarella, G. Argenziano, What dermoscopy tells us about nevogenesis, J. Dermatol. 38 (2011) 16–24. [36] T. Saida, Histogenesis of congenital and acquired melanocytic nevi: a unifying concept, Am. J. Dermatopathol. 4 (2006) 377–379. [37] V. Kokta, T. Hung, R. Al Dhaybi, C. Lugassy, R.L. Barnhill, High prevalence of angiotropism in congenital malenocytic nevi: an analysis of 53 cases, Am. J. Dermatopathol. 35 (2013) 180–183. [38] R.L. Barnhill, H. Kutzner, B. Schmidt, L. Ali, M. Bagot, A. Janin, C. Lugassy, Atypical spitzoid melanocytic neoplasms with angiotropism: a potential mechanism of locoregional involvement, Am. J. Dermatopathol. 33 (2011) 236–243. [39] S. Iken, M. Schmidt, C. Braun, A. Valentino, H.A. Lehr, S.C. Schaefer, Absence of ectopic epithelial inclusions in 3904 axillary lymph nodes examined in sentinel technique, Breast Cancer Res. Treat. 132 (2012) 621–624. [40] J.M. Traube, S. Begum, C. Shi, J.R. Eshleman, W.H. Westra, Benign nodal nevi frequently harbour the activating V600R BRAF mutation, Am. J. Surg. Pathol. 33 (2009) 568–571. [41] P.M. Pollock, U.L. Harpern, K.S. Hansen, L.M. Yudt, M. Stark, C.M. Robbins, T.Y. Moses, G. Hostetter, U. Wagner, J. Kakareka, G. Salem, Y. Pohida, P. Heenan, P. Duray, High frequency of BRAF mutations in nevi, Nat. Genet. 33 (2003) 19–20. [42] P. Tschandl, A.S. Berghoff, M. Preusser, S. Burgstaller-Muehlbacher, H. Pehamberger, I. Okamoto, H. Kittler, NRAS and BRAF Mutations in melanomaassociated nevi and uninvolved nevi, PLoS ONE 8 (2013) e69639. [43] M. Prade, A. Spatz, R. Bentley, P. Duvillard, C. Bognel, S.J. Robboy, Borderline and malignant serous tumor arising in pelvic lymph nodes: evidence of origin in benign glandular inclusions, Int. J. Gynecol. Pathol. 14 (1995) 87–91. [44] R.E. Fechner, Mammary carcinoma arising in benign axillary lymph node inclusions, Histopathology 14 (1989) 434–435. [45] S. Jaffer, R. Lin, I.J. Bleiweiss, C. Nagi, Intraductal carcinoma arising in intraductal papilloma in an axillary lymph node: review of the literature and proposed theories of evolution, Arch. Pathol. Lab. Med. 132 (2008) 1940–1942. [46] A. Kr, P. Sebastian, T. Somanathan, N.A. George, K. Jayasree, Significance of incidentally detected thyroid tissue in lymph nodes of neck dissections in patients with head and neck carcinoma, Int. J. Surg. Pathol. 20 (2012) 564–569. [47] F.I. Boulos, N.M. Granja, J.F. Simpson, F.P. O’Malley, M.M. Saadeldine, D.L. Page, M.E. Sanders, Intranodal papillary epithelial proliferations. A local process with a spectrum of morphologies and frequent association with papillomas in the breast, Am. J. Surg. Pathol. 38 (2014) 383–388. [48] V. Eusebi, C. Capella, A. Cossu, J. Rosai, Neuroendocrine carcinoma within lymph nodes in the absence of a primary tumor, with special reference to Merkel cell carcinoma, Am. J. Surg. Pathol. 16 (1992) 658–666. [49] I. Cozzolino, R. Zeppa, P. Zeppa, Lymph nodal Merkel cell carcinoma: primary tumor or metastasis from unknown primary site? J. Cutan. Pathol. 38 (2011) 836–837. [50] G.T. Pack, D.M. Gerber, I.M. Scharnagel, End results in the treatment of malignant melanoma, a report of 1190 cases, Ann. Surg. 136 (1952) 905–911. [51] J.N. Cormier, Y. Xing, L. Feng, X. Huang, L. Davidson, J.E. Gershenwald, J.E. Lee, P.F. Mansfield, M.I. Ross, Metastatic melanoma to lymph nodes in patients with unknown primary sites, Cancer 106 (2006) 2012–2020. [52] B.V. Shenoy, L. Fort III, S.P. Benjamin, Malignant melanoma primary in lymph node. The case of a missing link, Am. J. Surg. Pathol. 11 (1987) 140–146. [53] C.M. Lohman, K. Iversen, A.A. Jungbluth, M. Berwick, K.J. Busam, Expression of melanocyte differentiation antigens and Ki-67 in nodal nevi and comparison of Ki-67 expression with metastatic melanoma, Am. J. Surg. Pathol. 26 (2002) 1351–1357.
Please cite this article in press as: S. Piana, et al., Lymph node melanocytic nevi: Pathogenesis and differential diagnoses, with special reference to p16 reactivity, Pathol. – Res. Pract (2015), http://dx.doi.org/10.1016/j.prp.2015.01.003
G Model PRP-51359; No. of Pages 8 8
ARTICLE IN PRESS S. Piana et al. / Pathology – Research and Practice xxx (2015) xxx–xxx
[54] M.J. Mentrikoski, L. Ma, J.G. Pryor, L.A. McMahon, Q. Yang, B.O. Spaulding, G.A. Scott, H.L. Wang, H. Xu, Diagnostic utility of IMP3 in segregating metastatic melanoma from benign nevi in lymph nodes, Mod. Pathol. 22 (2009) 1582–1587. [55] W.A. Kanner, C.I. Barry, C.N. Smart, D.P. Frishberg, S.W. Binder, M.R. Wick, Reticulin and NM23 staining in the interpretation of lymph nodal nevus rests, Am. J. Dermatopathol. 35 (2013) 452. [56] T.J. de Vriesl, M. Smeets, R. de Graaf, K. Hou-Jensen, E.B. Bröcker, N. Renard, A.M.M. Eggermont, G.N.P. van Muijenl, D.J. Riuter, Expression of gp100,
MART-1, Tyrosinase, and S100 in paraffin embedded primary melanomas and locoregional, lymph node and visceral metastases: implications for dioagnosis and immunotherapy. A study conducted by the EORTC Melanoma Cooperative Group, J. Pathol. 193 (2001) 13–20. [57] C. Michaloglu, L.C. Vredeveld, M.S. Soengas, C. Denoyelle, T. Kuilman, C.M. van der Horst, D.M. Majoor, J.W. Shay, W.J. Mooi, D.S. Peeper, BRAFVE600associated senescence-like cell cycle arrest of human naevi, Nature 436 (2005) 720–724.
Please cite this article in press as: S. Piana, et al., Lymph node melanocytic nevi: Pathogenesis and differential diagnoses, with special reference to p16 reactivity, Pathol. – Res. Pract (2015), http://dx.doi.org/10.1016/j.prp.2015.01.003
ARTICLE IN PRESS Pathology – Research and Practice xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Pathology – Research and Practice journal homepage: www.elsevier.com/locate/prp
Original article
Lymph node melanocytic nevi: Pathogenesis and differential diagnoses, with special reference to p16 reactivity Simonetta Piana a,∗ , Elena Tagliavini a , Moira Ragazzi a , Magda Zanelli a , Iris Zalaudek b , Alessia Ciarrocchi c , Riccardo Valli a a
Pathology Unit, IRCCS-Arcispedale Santa Maria Nuova, Reggio Emilia, Italy Department of Dermatology, Medical University of Graz, Graz, Austria c Laboratory of Translational Research, IRCCS-Arcispedale Santa Maria Nuova, Reggio Emilia, Italy b
a r t i c l e
i n f o
Article history: Received 4 August 2014 Received in revised form 23 September 2014 Accepted 9 January 2015 Keywords: Melanocytic nevi Lymph node Nodal nevi Nevogenesis Metastasis P16
a b s t r a c t Lymph node nevi (NN) have been occasionally described, yet little is currently known on their origin. According to a theoretical model of nevogenesis, the dissemination of nevus progenitor cells through lymphatic routes is responsible for the development of both nodal and skin nevi. The true incidence of NN is largely unknown but it has been reported to vary from 0.017% to as high as 22%. The frequency of NN nevi has increased since the introduction of sentinel lymph node mapping as a routine prognostic procedure in breast cancer and melanoma. The aim of this study was to analyze the frequency and morphological findings of NN, to discuss possible pathogenetic pathways in their evolution, and to verify the consistency of p16 immunostaining in the critical differential approach between NN and melanoma metastases. We therefore morphologically and immunohistochemically evaluated a series of 60 NN from 58 patients. In 21 patients, the lymph nodes had been removed during the staging for a skin melanoma; in all these patients NN immunostaining with p16 was strongly positive and p16 proved to be a reliable marker for the crucial differential diagnosis between NN and melanoma metastasis, strongly reacting in NN and lacking in melanoma deposits. A deeper knowledge on NN could help to clarify some important topics such as lymph node metastatic melanoma with unknown primary and the current debate on the lymph node involvement from atypical spitzoid tumors. © 2015 Elsevier GmbH. All rights reserved.
Introduction Since the original description by Stewart and Copeland in 1931 [1], melanocytic inclusions in lymph nodes, also called node nevi (NN), have been reported episodically [2–21] and their histological variants have been further elucidated [22–28]. The true incidence of NN is largely unknown but it has been reported to vary from 0.017% to as high as 22% [13,15,16,19]. The reported frequency of NN nevi has increased since the introduction of sentinel lymph node (SLN) mapping as a routine prognostic procedure in breast cancer and melanoma. This is because this technique allows for the detection of melanocytic rests and epithelial inclusions [29] within metastatic and non-metastatic lymph
∗ Corresponding author. Tel.: +0039 0522 295919; fax: +0039 0522 296945. E-mail address: [email protected] (S. Piana).
nodes; the presence of melanocytic inclusions increases if sectioning search is extensive. NN mostly present as small aggregates of normal-looking melanocytes located within the fibrous components of the lymph node, mainly in the capsule and the trabeculae. Though rarely, they can also be found in the parenchyma or in the lymphatic channels [18]. Due to their location, besides being an incidental histological oddity, they can represent a diagnostic challenge in lymph nodes excised during the workup for a malignancy, especially for melanoma. In order to discriminate NN from metastatic melanoma, p16 immunostaining has been proposed as a reliable tool, since it has been shown that NN display an extensive nuclear and cytoplasmic staining for p16, at variance with melanomas [30]. The aim of this study was to analyze the frequency and morphological findings of NN in neoplastic and non-neoplastic clinical settings, to discuss possible pathogenetic pathways in their evolution, and to verify the consistency of p16 immunostaining in the critical differential approach between NN and melanoma metastases.
http://dx.doi.org/10.1016/j.prp.2015.01.003 0344-0338/© 2015 Elsevier GmbH. All rights reserved.
Please cite this article in press as: S. Piana, et al., Lymph node melanocytic nevi: Pathogenesis and differential diagnoses, with special reference to p16 reactivity, Pathol. – Res. Pract (2015), http://dx.doi.org/10.1016/j.prp.2015.01.003
G Model
ARTICLE IN PRESS
PRP-51359; No. of Pages 8
S. Piana et al. / Pathology – Research and Practice xxx (2015) xxx–xxx
2
Materials and methods The computerized files of the Pathology Unit of IRCCS-S.Maria Nuova Hospital in Reggio Emilia were retrospectively searched for all the lymph node specimens between January 1, 2000 and January 31, 2014. Among these, all cases of lymph nodes harboring one or more melanocytic rests were retrieved and histopathologically revised. For all retrieved cases of NN, demographic data, the anatomic basin, the major associated diseases, the number of the revised slides, and the reason for excision (i.e. SLN or non-SLN procedure) were recorded. The SLNs were processed as previously described according to the breast cancer screening and melanoma staging protocols [31,32]. The non-SLNs corresponded to the elective lymphadenectomies, after a positive SLN. As a rule, non-SLNs have not been examined with step sections and immunohistochemical stains. This issue, together with the different protocols employed for breast SLN and melanoma SLN (a serial section evaluation of tissue in the former and four serial sections in the latter) may represent an underestimation bias of NN frequency in the SLNs from melanoma patients and in non-SLNs. On histological review, the number of melanocytic aggregates and their location within the lymph node were evaluated. Size estimation of the NN was performed by integrating the major axis of the lesions with the number of step sections in cases of SLNs. For SLNs from melanoma patients, one or more slides stained with S100 (polyclonal, Ventana, Tucson, AZ) were routinely available for analysis. In cases of melanoma SLNs harboring NN, the primary melanomas were revised to compare the morphologic features. Using an automated immunostainer (Benchmark, Ventana, Tucson, AZ), all cases of NN in patients with melanoma were stained with p16 (E6H4, Ventana, Tucson, AZ) and ki67 (clone MIB1, Dako, Glostrup, Denmark) to evaluate proliferative activity. Selected cases were also stained with Melan A/MART-1 (A103, Ventana, Tucson, AZ, to confirm the presence of melanocytic nests. Results During the study period, the Pathology Unit of IRCCS-S.Maria Nuova Hospital accepted a total of 26,240 lymph node specimens, including all types of lymph node resection, i.e. tru-cut biopsies, lymphadenectomies for lymphomas or reactive lymphadenitis, SLNs and extended lymphadenectomies for oncologic staging. Among all these specimens there were 416 (1.5%) SLNs from 399 patients with melanoma and 4276 (17%) SLNs from 4205 patients with breast cancer. Among all cases, the presence of an NN was recorded in 60 (0.2%) cases from 58 patients, including 17 (29.3%) men and 41 (70.6%) women, with a mean age of 55 years (range 29–84 yr). Among all the SLNs analyzed by our Institution in the study period, NN were present in 18 out of 416 melanoma SLNs (4.3%)
and in 12 out of 4,276 breast SLNs (0.3%). In our series, therefore, NN in SLNs represented 50% of all the cases. The clinical features in relation to the primary diseases are listed in Table 1. In 55 patients, the lymph nodes had been removed during staging for a neoplasia. In 32 of these patients (55%), the reported primary disease was breast cancer (26 invasive ductal carcinoma [IDC], 5 in situ ductal carcinoma [ISDC], 1 invasive lobular carcinoma [ILC]), in 21 patients (27%) cutaneous melanoma, in 1 patient (1.6%) invasive squamous cell carcinoma of the larynx, and in 1 further patient (1.6%) invasive adenocarcinoma of the prostate. In 3 patients (5%), a single enlarged lymph node was removed with the clinical suspicion of a lymphoma and with the final pathological report of reactive lymphadenopathy. In all the 58 patients, only 1 lymph node harbored an NN, except in two female patients, one of whom with breast cancer and the other with melanoma. They featured NN in 2 lymph nodes, one SLN and one non-SLN, in axillary and inguinal region, respectively. The lymph node site was axillary in 45 cases (75%; 23 right and 22 left), inguinal in 10 cases (16.6%; 6 right and 4 left), cervical in 4 cases (6.6%; 1 right and 3 left), and right obturator in 1 case (1.6%). The histologic features in relation to the primary diseases are listed in Table 2. The number of melanocytic nests within the same lymph node varied from 1 to 6, with a mean number greater in nodes with reactive lymphadenopathies. The mean size of NN was 1.4 mm (range 0.1 to 7 mm). NN morphologic features were generally repetitive. As a rule, nests were composed of monomorphous bland-appearing melanocytes, mostly located in the lymph node supporting stroma, the capsule, or along the trabeculae (Fig. 1A). A peculiar perivascular arrangement was often seen (Fig. 1B and C), while lymph node parenchyma involvement was present in only 3 cases (Fig. 1E). Mitotic figures, nucleoli, pycnotic nuclei, and necrotic foci were always absent. Immunohistochemical stainings with S100 and Melan A were diffusely positive in all cases, whereas immunohistochemical staining with ki67 confirmed the absence of proliferative activity (Fig. 1D). Four cases showed distinct characteristics: in two cases (a reactive lymphadenopathy and one SLN during breast cancer workup), the melanocytic rests had blue nevus-like features, with dendritic, hyperpigmented melanocytes disrupting the collagenous fibers of the lymph node capsule (Fig. 2A and B). In the other two cases, the melanocytic aggregates were particularly numerous and large. One of these last two cases was a non-SLN metastatic node in the axillary region of a 61-year-old woman with a mucinous invasive carcinoma of the breast. The melanocytic nests were multiple and grew along the capsule and the trabeculae, with few collections present even in the parenchyma. The second case was an inguinal SLN in a 57-year-old female patient with a 2.5 mm thick nodular melanoma arising on a giant congenital nevus occupying the entire abdominal wall. The entire circumference of the lymph node was
Table 1 Clinical features of patients with NN. n. Lymph nodes* (%)
Males
Females*
Mean age (yr)
Axillary (right)
SLN breast cancer Non-SLN breast cancer SLN melanoma Non-SLN melanoma Non-SLN other neoplasias Lymphadenopathies
12 (20%) 21 (35%) 18 (30%) 4 (6.6%) 2 (3.3%) 3 (5%)
0 0 12 2 2 1 17
12 20 6 1 0 2 41
59 59 52 44 58 31
12 (6) 21 (12) 10 (4) 1 (0)
Total
60
58
1 (1) 45
Inguinal (right)
6 (4) 2 (1) 2 (1) 10
Cervical (right)
Obturator (right)
2 (1) 1 (0) 1 (0)
1 (1)
4
1
60
SLN: sentinel lymph node; yrs: years. * Two female patients had NN in two adjacent lymph nodes.
Please cite this article in press as: S. Piana, et al., Lymph node melanocytic nevi: Pathogenesis and differential diagnoses, with special reference to p16 reactivity, Pathol. – Res. Pract (2015), http://dx.doi.org/10.1016/j.prp.2015.01.003
G Model
ARTICLE IN PRESS
PRP-51359; No. of Pages 8
S. Piana et al. / Pathology – Research and Practice xxx (2015) xxx–xxx
3
Table 2 Histologic features of NN. n. Lymph nodes
n. Nests * (mean)
Dimension* (mean, mm)
SLN breast cancer Non SLN breast cancer SLN melanoma Non SLN melanoma Non SLN other neoplasias lymphadenopathies
12 21 18 4 2 3
1.1 1.3 1.4 1 1.5 6
1.2 1.5 1.1 1.2 1.2 2.6
TOT
60
2.0
1.4
*
Parenchymal involvement
Special features 1 Blue nevus-like
1 1
1 3
1 Blue nevus-like
Metastasis 1 4 3 2 0 0 10
Except two cases with exceptionally large NN.
Table 3 Clinicopathologic features of primary skin melanomas. Melanoma SLN n. Patients Mean age (yr) (range) M/F Mean Breslow thickness (mm) vascular invasion Associated nevi Metastasis
18 52 (32–78) 12/6 2.3 1 6 3
Melanoma non-SLN 3 44 (29–73) 2/1 2.6 0 1 2
occupied by multiple confluent aggregates of epithelioid or weakly pigmented spindle melanocytes extending to the inner trabeculae and, focally, to the parenchyma (Fig. 2C and D). The same aggregates, highlighted by the immunohistochemical stain with S100, were also present in the perinodal fat as short fascicles or nodular collections and around the vessels, whose walls were focally infiltrated (Fig. 2E and F). Metastasis from the primary malignancy coexisted in 10 cases of NN (16.6%): 5 IDCs of the breast and 5 melanomas. The revision of the cutaneous primary melanomas confirmed its morphologic similarity with the metastatic lymph node deposits and the unquestionable histologic differences with the bland looking melanocytic nevi. Notably, in 7 (30%) out of 21 patients, melanoma arose within or adjacent to a congenital-type dermal nevus. Vascular invasion was recorded in only 1 case of melanoma. Immunostaining with p16 was strongly positive in all the 22 NN in melanoma patients with a strong nuclear and cytoplasmic staining; in 5 lymph nodes harboring both one or more NN and melanomatous metastatic deposits, melanoma aggregates were consistently negative (Figs. 3 and 4). Ki67 immunostaining confirmed the presence of variable proliferative activity in melanoma metastases. Table 3 reports the main clinical and histologic features of the 21 primary skin melanomas with NN in SLN and non-SLN. One female patient with melanoma had NN in 2 lymph nodes in inguinal region, one SLN and one non-SLN (see above). Discussion The presence of normal looking heterotopic epithelial tissue is well known in specific lymph node basins, i.e. in submandibular, axillary, or in pelvic nodes, where it represents an important diagnostic pitfall and a “false positive” finding during the staging of a malignancy. While it is accepted that Müllerian-derived organs and breast are the major source of epithelial inclusions in axillary and pelvic lymph nodes, respectively [33], little is currently known on the origin of NN. The two most reliable theories, still en vogue after more than 40 years, argue that NN are caused by embolic transfer of skin melanocytes to the draining lymph node [2,16] or, alternatively, by an aberrant embryologic migration from the neural crest to their permanent anatomic sites [2,3].
The former theory seems to be confirmed by the frequent finding of melanocyte aggregates protruding into both the lymphatic and hematic vascular lumina, in the context of congenital nevi. It is conceivable that the same aggregates can detach from the wall and be transported to the corresponding lymph nodes. This interpretation is in line with the up to 30% rate of melanoma-associated congenital nevi we report in our series. Indeed, the presence of NN supports a theoretical model of nevogenesis, according to which the dissemination of nevus progenitor cells through lymphatic routes is responsible for the development of both nodal and cutaneous nevi [34]. In other words, nevogenesis is considered a benign metastatic process that can stop in the lymph node capsule or in the skin. While the theory on embolic transfer to draining nodes does not explain the presence of NN in deep locations, i.e. in an obturator lymph node, the “benign metastasis” approach does: melanocytes enter systemic lymphatic circulation and implant at a distance. The second hypothesis relies on the perturbation of the embryogenic melanoblast migration to the developing skin [35], with consequent deposition of melanocytic precursors into the developing lymph nodes. The latter incorporate melanoblasts in their framework and allow their morphologic maturation. As it has been postulated that the formation of multiple, common or atypical (dysplastic) nevi results from an alterated embryogenic melanoblast migration [36], this interpretation explains well why we find melanocytic rests more frequently in lymph node basins draining the teguments of patients with melanoma, who presumably have more moles (high melanocytic burden), typical or atypical. Moreover, this model plausibly explains both the involvement of deeply located nodes and the intracapsular position of the melanocytic rests, which is almost by definition remarkably distinct from the intrasinusoidal location of the metastatic emboli. A third point of view could combine the previous two: the extravascular cellular migration may be an important pathway in the development of congenital nevi. In other words, melanoblasts reach their destination by migrating around the vessels, not inside them [37]. This theory could explain the “pseudoinfiltrative” and extranodal pattern of distribution of the melanocytic rests in the cases associated with the giant congenital nevus. Indeed, we noted an unusual perivascolar location of the melanocytic rests within the perinodal adipose tissue. This view has also been called upon to justify the presence of SLN metastasis from atypical spitzoid melanocytic neoplasms [38], a very controversial issue that probably shares some points with the etiopathogenesis of melanocytic rests. However, although extremely intriguing, any discussion of this is beyond the aim of the present work. These theories notwithstanding, many points remain unexplained. First, why is the relative frequency of nodal rests higher in melanoma patients than in patients with other malignancies, i.e. breast cancer, despite a SLN procedure? In our series, the incidence of nodal rests was 4.3% in melanoma patients compared to only 0.3% in breast cancer patients. If one accepts that there is an association between nodal nevus cells and congenital nevi [17], one could claim that patients with
Please cite this article in press as: S. Piana, et al., Lymph node melanocytic nevi: Pathogenesis and differential diagnoses, with special reference to p16 reactivity, Pathol. – Res. Pract (2015), http://dx.doi.org/10.1016/j.prp.2015.01.003
G Model PRP-51359; No. of Pages 8 4
ARTICLE IN PRESS S. Piana et al. / Pathology – Research and Practice xxx (2015) xxx–xxx
Fig. 1. (A) Melanocytic nests are composed of monomorphous bland-appearing melanocytes, mostly located in the lymph node supporting stroma, the capsule, or along the trabeculae, with a typical perivascular arrangement (insert). (B)–(C) At higher power view, melanocytic infiltration of the vessel walls is evident. (D) The immunohistochemical staining with ki67 confirms the absence of proliferative activity; as a control, two stained lymphocytic nuclei are evident. (E) The lymph node parenchyma involvement is rare (arrows) but can be seen besides the capsular melanocytic aggregates. (F)–(G) Immunohistochemical stainings with Melan A (F) and p 16 (G) confirm the presence of bland melanocytes within the lymphoid tissue (arrows).
melanoma have a higher incidence of NN because they generally have a greater melanocytic burden, i.e. a higher number of nevi, including nevi of the congenital type, than do non-melanoma patients. In addition, it is unclear why NN are more frequent than epithelial inclusions in axillary nodes; in fact, the latter, although
they have not been found in one large series [39], have frequently been described in the literature [29]. Do these heterotopic melanocytic rests possess any malignant potential? A recent paper has demonstrated that NN can harbor the BRAF V600E mutation [40]. This finding is not surprising as we
Please cite this article in press as: S. Piana, et al., Lymph node melanocytic nevi: Pathogenesis and differential diagnoses, with special reference to p16 reactivity, Pathol. – Res. Pract (2015), http://dx.doi.org/10.1016/j.prp.2015.01.003
G Model PRP-51359; No. of Pages 8
ARTICLE IN PRESS S. Piana et al. / Pathology – Research and Practice xxx (2015) xxx–xxx
5
Fig. 2. (A)–(B) NN can show blue nevus-like features, with dendritic, hyperpigmented melanocytes disrupting the collagenous fibers of the lymph node capsule. (C)–(D) In rare cases, the melanocytic aggregates are particularly large and can be noticed in the perinodal fat. They can be easily highlighted around the vessels by the immunohistochemical stain with S100 (E)–(F)
expect that up to 80% of melanocytic nevi carry the same mutation [41]. We do agree with those authors who claim that neither morphology nor the presence of BRAF mutation predicts the chance of malignant transformation of a nevus into a melanoma [42] and we arbitrarily transport this observation to NN as well. However, it is conceivable that melanocytic rests can undergo neoplastic progression. Malignant transformation of heterotopic epithelium in lymph nodes has been described in the major anatomic draining sites [43–47]; few cases of bona fide primary lymph node Merkel cell carcinoma have been described, too [48,49]. Although malignant transformation has not been reported for NN, as it has been for nodal epithelial inclusions, the concept of lymph node metastatic melanoma with unknown primary site has commonly been accepted since 1952 [50] and has been repeatedly
analyzed [51]. In 1977, Ridolfi et al. [5] speculated that nodal nevi can account for some of the lymph node melanomas with unknown primary. In 1986, Shenoy et al. described a putative case of melanoma primary in the lymph node [52], with a morphologic transition of melanocytes which they considered the missing link between NN and melanoma. The development of a melanoma from ectopic melanocytes located in the lymph node capsule could explain those cases of metastatic melanoma without a primary but this hypothesis has never been re-examined, since it is dogmatically assumed that a skin or mucosal primitive localization must exist; if it does not, then the primary lesion must have regressed. Therefore, to date, Shenoy et al.’s theory remains speculative. Despite many pathogenetic questions still to be answered, pathologists know that the main problem related to NN is the
Please cite this article in press as: S. Piana, et al., Lymph node melanocytic nevi: Pathogenesis and differential diagnoses, with special reference to p16 reactivity, Pathol. – Res. Pract (2015), http://dx.doi.org/10.1016/j.prp.2015.01.003
G Model PRP-51359; No. of Pages 8 6
ARTICLE IN PRESS S. Piana et al. / Pathology – Research and Practice xxx (2015) xxx–xxx
Fig. 3. (A)–(B) A capsular NN in a patient with melanoma. The melanocytes are small sized and strongly immunoreactive to p16 with an intense nuclear pattern (B). (C) The melanoma micrometastasis is made up by neoplastic atypical, epithelioid melanocytes (arrows), completely negative with p16 immunostain (D).
differential diagnosis with a metastatic deposit. This is why some papers have searched for further ancillary methods to assist in the diagnosis of NN [30,53–55]. In our experience, p16 has proven to be a reliable marker to distinguish NN from melanoma metastases, strongly reacting in NN and lacking in melanoma deposits. We can thus conclude that p16, together with ki67 and eventually
other routine melanocytic markers, as HMB45 and MelanA [56], can reasonably be used in everyday life for the crucial differential diagnosis between NN and melanoma metastasis. This scenario is coherent with the physiological role of p16, which is a tumor suppressor protein involved in the regulation of cell cycle and senescence. P16 overexpression seems to represent a
Fig. 4. (A) A capsular NN in a patient with melanoma. (B) At higher magnification, NN cells are monomorphous and cytologically bland. (C)–(D) Immunohistochemical stainings with p16 (C) and Melan A are diffusely positive in nodal nevi; p16 shows a diffuse nuclear and cytoplasmic stain, with few negative cells. (E) The same lymph node also harbors a small melanoma metastatic deposit. (F)–(G)–(H) The micrometastasis is immunoreactive with S100 (F) and Melan A (H) but negative with p16 (G).
Please cite this article in press as: S. Piana, et al., Lymph node melanocytic nevi: Pathogenesis and differential diagnoses, with special reference to p16 reactivity, Pathol. – Res. Pract (2015), http://dx.doi.org/10.1016/j.prp.2015.01.003
G Model PRP-51359; No. of Pages 8
ARTICLE IN PRESS S. Piana et al. / Pathology – Research and Practice xxx (2015) xxx–xxx
compensatory cell mechanism to BRAFV600E-driven proliferation in human melanocytes [57]; the presence of both BRAFV600E [39] and p16 [30] is therefore likely in lesions like NN which are in a apparently growth-arrested state, as confirmed by immunohistochemical staining with ki67. In conclusion, after more than 80 years since the original description, the definitive diagnosis of NN relies on an accurate morphologic analysis in hematoxylin-eosin stained sections and on the cautious employment of the immunohistochemical markers. Despite the growing interest on this topic, the ontogenesis of NN is still unclear. This knowledge that may appear insignificant in a surgical pathology practice, could have a key role to solve some important topics such as lymph node metastatic melanoma with unknown primary site and the current debate on the presence of melanocytic deposits in SLN from atypical spitzoid tumors. In this context, the pathogenesis of NN may be a promising area of study in cancer behavior research. Source of funding None. References [1] F.W. Stewart, M.M. Copeland, Neurogenic sarcoma, Am. J. Cancer 15 (1931) 1235–1320. [2] W.T. Johnson, E.B. Helwig, Benign nevus cell in the capsule of lymph nodes, Cancer 23 (1969) 747–753. [3] W.R. Hart, Primary nevus of a lymph node, Am. J. Clin. Pathol. 55 (1971) 88–92. [4] S.W. McCarthy, A.A. Palmer, P.M. Bale, E. Hirst, Naevus cells in lymph nodes, Pathology 6 (1974) 351–358. [5] R.L. Ridolfi, P.P. Rosen, H. Thaler, Nevus cell aggregates with lymph nodes: estimated frequency and clinical significance, Cancer 39 (1977) 164–171. [6] J.L. Jensen, L.W. Correll, Nevus cell aggregates in submandibular lymph nodes, Oral Surg. Oral Med. Oral Pathol. 50 (1980) 552–556. [7] G. Ioannides, Lymph node aggregates of nevus cells, in: A.B. Ackerman (Ed.), Pathology of Malignant Melanoma, Masson Publishing, New York, NY, 1981, pp. 297–300. [8] S. Andreola, C. Clemente, Nevus cells in axillary lymph nodes from radical mastectomy specimens, Pathol. Res. Pract. 179 (1985) 616–618. [9] H.M. Yadzi, Nevus cell aggregates associated with lymph nodes. Immunohistochemical observations, Arch. Pathol. Lab. Med. 109 (1985) 1044–1046. [10] C. Subramony, J.R. Lewin, Nevus cells within lymph nodes: possible metastases from a benign intradermal nevus, Am. J. Clin. Pathol. 84 (1985) 220–223. [11] R.H. Hruban, F. Eckert, B. Baricevic, Melanocytes of a melanocytic nevus in a lymph node from a patient with a primary cutaneous melanoma associated with a small congenital nevus, Am. J. Dermatopathol. 12 (1990) 402–407. [12] K. Hara, Melanocytic lesions in lymph nodes associated with congenital naevus, Histopathology 23 (1993) 445–451. [13] N.C. Bautista, S. Cohen, K.H. Anders, Benign melanocytic nevus cells in axillary lymph nodes: a prospective incidence and immunohistochemical study with literature review, Am. J. Clin. Pathol. 102 (1994) 102–108. [14] C.J. Fisher, S. Hill, R.R. Millis, Benign lymph node inclusions mimicking metastatic carcinoma, J. Clin. Pathol. 47 (1994) 245–247. [15] K.F. Carson, D.R. Wen, P.X. Li, et al., Nodal nevi and cutaneous melanomas, Am. J. Surg. Pathol. 20 (1996) 834–840. [16] D. Fontaine, W. Parkhill, W. Greer, N. Walsh, Nevus cells in lymph nodes: an association with congenital cutaneous nevi, Am. J. Dermatopathol. 24 (2002) 1–5. [17] D.A. Biddle, H.L. Evans, B.L. Kemp, A.K. El-Naggar, J.D. Harvell, W.L. White, S.S. Iskandar, V.G. Prieto, Intraparenchymal nevus cell aggregates in lymph nodes: a possible diagnostic pitfall with malignant melanoma and carcinoma, Am. J. Surg. Pathol. 27 (2003) 673–681. [18] J.W. Patterson, Nevus cell aggregates in lymph node, Am. J. Clin. Pathol. 121 (2004) 13–15. [19] J.B. Holt, O.P. Sangueza, E.A. Levine, P. Shen, S. Bergman, K.R. Geisinger, A.J. Creager, Nodal melanocytic nevi in sentinel lymph nodes. Correlation with melanoma-associated cutaneous nevi, Am. J. Clin. Pathol. 121 (2004) 58–63. [20] T. Gambichler, L. Scholl, M. Stücker, F.G. Bechara, K. Hoffmann, P. Altmeyer, N. Othlinghaus, Clinical characteristics and survival data of melanoma patients with nevus cell aggregates within sentinel lymph nodes, Am. J. Clin. Pathol. 139 (2013) 566–573. [21] J.G. Azzopardi, C.M. Ross, G. Frizzera, Blue naevi of lymph node capsule, Histopathology 1 (1977) 451–461. [22] J. Roth, Ectopic blue nevi in lymph nodes, in: A.B. Ackerman (Ed.), Pathology of malignant melanoma, Masson Publishing, New York, NY, 1981, pp. 293–296. [23] R.L. Goldman, Blue nevus of lymph node capsule: report of a unique case, Histopathology 5 (1981) 445.
7
[24] J. Epstein, R.A. Erlandson, P.P. Rosen, Nodal blue nevi; a study of three cases, Am. J. Surg. Pathol. 8 (1984) 907–915. [25] C. Urso, Nodal melanocytic nevus with baloon-cell change (nodal baloon-cell nevus), J. Cutan. Pathol. 35 (2008) 672–676. [26] N. Misago, K. Nagase, S. Toda, Y. Shinoda, S. Koba, Y. Narisawa, Cellular blue nevus with nevus cells in a sentinel lymph node, Eur. J. Dermatol. 18 (2008) 586–589. [27] S.M.K. Nahar Begum, M. Lomme, M. Ruhul Quddus, Nodal combined blue nevus and benign nevus cells in multiple axillary sentinel nodes in a patient with breast carcinoma: report of a case, Int. J. Surg. Pathol. (2013), http://dx.doi.org/10.1177/1066896913509008. [28] L. Dohese, T. Ferringer, Nodal blue nevus: a pitfall in lymph node biopsies, J. Cutan. Pathol. 37 (2010) 102–104. [29] G. Fellegara, M.L. Carcangiu, J. Rosai, Benign epithelial inclusions in axillary lymph nodes: report of 18 cases and review of the literature, Am. J. Surg. Pathol. 35 (2011) 1123–1133. [30] D. Mihic-Probst, P. Saremaslani, P. Komminoth, P.U. Heitz, Immunostaining for the tumour suppressor gene p16 product is a useful marker to differentiate melanoma metastasis from lymph-node nevus, Virchow Arch. 443 (2003) 745–751. [31] N. Perry, M. Broeders, de Törnberg, S. Wolf C., R. Holland, L. von Karsa, European guidelines for quality assurance in breast cancer screening and diagnosis. Fourth edition-summary document, Ann. Oncol. 19 (2008) 614–622. [32] V.G. Prieto, Sentinel Lymph nodes in cutaneous melanoma, Clin. Lab. Med. 31 (2011) 301–310. [33] R.A. Warnke, L.M. Weiss, J.K.C. Chan, M.L. Cleary, R.F. Dorfman, Non hematolymphoid tumors and tumor-like lesions, in: J. Rosai, L. Sobin (Eds.), Atlas of Tumor Pathology, Third Series. Tumors of the Lymph Nodes and Spleen, Armed Forces Institute of Pathology, Washington DC, 1994, pp. 474–475. [34] A.L. Ross, M.I. Sanchez, J.M. Grichnik, Nevogenesis: a benign metastatic process? ISRN Dermatol. (2011), http://dx.doi.org/10.5402/2011/813513. [35] I. Zalaudek, C. Catricalà, E. Moscarella, G. Argenziano, What dermoscopy tells us about nevogenesis, J. Dermatol. 38 (2011) 16–24. [36] T. Saida, Histogenesis of congenital and acquired melanocytic nevi: a unifying concept, Am. J. Dermatopathol. 4 (2006) 377–379. [37] V. Kokta, T. Hung, R. Al Dhaybi, C. Lugassy, R.L. Barnhill, High prevalence of angiotropism in congenital malenocytic nevi: an analysis of 53 cases, Am. J. Dermatopathol. 35 (2013) 180–183. [38] R.L. Barnhill, H. Kutzner, B. Schmidt, L. Ali, M. Bagot, A. Janin, C. Lugassy, Atypical spitzoid melanocytic neoplasms with angiotropism: a potential mechanism of locoregional involvement, Am. J. Dermatopathol. 33 (2011) 236–243. [39] S. Iken, M. Schmidt, C. Braun, A. Valentino, H.A. Lehr, S.C. Schaefer, Absence of ectopic epithelial inclusions in 3904 axillary lymph nodes examined in sentinel technique, Breast Cancer Res. Treat. 132 (2012) 621–624. [40] J.M. Traube, S. Begum, C. Shi, J.R. Eshleman, W.H. Westra, Benign nodal nevi frequently harbour the activating V600R BRAF mutation, Am. J. Surg. Pathol. 33 (2009) 568–571. [41] P.M. Pollock, U.L. Harpern, K.S. Hansen, L.M. Yudt, M. Stark, C.M. Robbins, T.Y. Moses, G. Hostetter, U. Wagner, J. Kakareka, G. Salem, Y. Pohida, P. Heenan, P. Duray, High frequency of BRAF mutations in nevi, Nat. Genet. 33 (2003) 19–20. [42] P. Tschandl, A.S. Berghoff, M. Preusser, S. Burgstaller-Muehlbacher, H. Pehamberger, I. Okamoto, H. Kittler, NRAS and BRAF Mutations in melanomaassociated nevi and uninvolved nevi, PLoS ONE 8 (2013) e69639. [43] M. Prade, A. Spatz, R. Bentley, P. Duvillard, C. Bognel, S.J. Robboy, Borderline and malignant serous tumor arising in pelvic lymph nodes: evidence of origin in benign glandular inclusions, Int. J. Gynecol. Pathol. 14 (1995) 87–91. [44] R.E. Fechner, Mammary carcinoma arising in benign axillary lymph node inclusions, Histopathology 14 (1989) 434–435. [45] S. Jaffer, R. Lin, I.J. Bleiweiss, C. Nagi, Intraductal carcinoma arising in intraductal papilloma in an axillary lymph node: review of the literature and proposed theories of evolution, Arch. Pathol. Lab. Med. 132 (2008) 1940–1942. [46] A. Kr, P. Sebastian, T. Somanathan, N.A. George, K. Jayasree, Significance of incidentally detected thyroid tissue in lymph nodes of neck dissections in patients with head and neck carcinoma, Int. J. Surg. Pathol. 20 (2012) 564–569. [47] F.I. Boulos, N.M. Granja, J.F. Simpson, F.P. O’Malley, M.M. Saadeldine, D.L. Page, M.E. Sanders, Intranodal papillary epithelial proliferations. A local process with a spectrum of morphologies and frequent association with papillomas in the breast, Am. J. Surg. Pathol. 38 (2014) 383–388. [48] V. Eusebi, C. Capella, A. Cossu, J. Rosai, Neuroendocrine carcinoma within lymph nodes in the absence of a primary tumor, with special reference to Merkel cell carcinoma, Am. J. Surg. Pathol. 16 (1992) 658–666. [49] I. Cozzolino, R. Zeppa, P. Zeppa, Lymph nodal Merkel cell carcinoma: primary tumor or metastasis from unknown primary site? J. Cutan. Pathol. 38 (2011) 836–837. [50] G.T. Pack, D.M. Gerber, I.M. Scharnagel, End results in the treatment of malignant melanoma, a report of 1190 cases, Ann. Surg. 136 (1952) 905–911. [51] J.N. Cormier, Y. Xing, L. Feng, X. Huang, L. Davidson, J.E. Gershenwald, J.E. Lee, P.F. Mansfield, M.I. Ross, Metastatic melanoma to lymph nodes in patients with unknown primary sites, Cancer 106 (2006) 2012–2020. [52] B.V. Shenoy, L. Fort III, S.P. Benjamin, Malignant melanoma primary in lymph node. The case of a missing link, Am. J. Surg. Pathol. 11 (1987) 140–146. [53] C.M. Lohman, K. Iversen, A.A. Jungbluth, M. Berwick, K.J. Busam, Expression of melanocyte differentiation antigens and Ki-67 in nodal nevi and comparison of Ki-67 expression with metastatic melanoma, Am. J. Surg. Pathol. 26 (2002) 1351–1357.
Please cite this article in press as: S. Piana, et al., Lymph node melanocytic nevi: Pathogenesis and differential diagnoses, with special reference to p16 reactivity, Pathol. – Res. Pract (2015), http://dx.doi.org/10.1016/j.prp.2015.01.003
G Model PRP-51359; No. of Pages 8 8
ARTICLE IN PRESS S. Piana et al. / Pathology – Research and Practice xxx (2015) xxx–xxx
[54] M.J. Mentrikoski, L. Ma, J.G. Pryor, L.A. McMahon, Q. Yang, B.O. Spaulding, G.A. Scott, H.L. Wang, H. Xu, Diagnostic utility of IMP3 in segregating metastatic melanoma from benign nevi in lymph nodes, Mod. Pathol. 22 (2009) 1582–1587. [55] W.A. Kanner, C.I. Barry, C.N. Smart, D.P. Frishberg, S.W. Binder, M.R. Wick, Reticulin and NM23 staining in the interpretation of lymph nodal nevus rests, Am. J. Dermatopathol. 35 (2013) 452. [56] T.J. de Vriesl, M. Smeets, R. de Graaf, K. Hou-Jensen, E.B. Bröcker, N. Renard, A.M.M. Eggermont, G.N.P. van Muijenl, D.J. Riuter, Expression of gp100,
MART-1, Tyrosinase, and S100 in paraffin embedded primary melanomas and locoregional, lymph node and visceral metastases: implications for dioagnosis and immunotherapy. A study conducted by the EORTC Melanoma Cooperative Group, J. Pathol. 193 (2001) 13–20. [57] C. Michaloglu, L.C. Vredeveld, M.S. Soengas, C. Denoyelle, T. Kuilman, C.M. van der Horst, D.M. Majoor, J.W. Shay, W.J. Mooi, D.S. Peeper, BRAFVE600associated senescence-like cell cycle arrest of human naevi, Nature 436 (2005) 720–724.
Please cite this article in press as: S. Piana, et al., Lymph node melanocytic nevi: Pathogenesis and differential diagnoses, with special reference to p16 reactivity, Pathol. – Res. Pract (2015), http://dx.doi.org/10.1016/j.prp.2015.01.003