Pathology (June 2020) 52(4), pp.

431–438

A N ATO M I C A L PAT H O L O G Y

Large nested melanoma: a clinicopathological,

morphometric and cytogenetic study of 12 cases
TAMAZIN N. LEECY1, PETER MCQUILLAN2, NATHAN T. HARVEY1,3,
NIMA MESBAH ARDAKANI1,3,4, CHRIS VAN VLIET1, JOANNE PEVERALL5,
DAGMARA KENNEDY5, SORUBA SIVAMOORTHY5, ALYSSA FRUVALL5,
HASHIKA RIJHUMAL5, JOAN UZARAGA5, SHALINDER SINGH1, BENJAMIN A. WOOD1,3
1
Department of Anatomical Pathology, PathWest Laboratory Medicine, Perth, WA, Australia;
2
Melbourne Pathology, Melbourne, Vic, Australia; 3Division of Pathology and Laboratory
Medicine, Medical School, University of Western Australia, Perth, WA, Australia; 4Murdoch
5
University, Perth, WA, Australia; Department of Diagnostic Genomics, PathWest Labora-
tory Medicine, Perth, WA, Australia

Summary typically employed architectural criteria for the diagnosis of

A group of melanomas characterised by predominant melanoma such as predominant lentiginous growth, pagetoid
growth as large nests within the epidermis has been extension of melanocytes and poor circumscription are diffi-
described. These cases present a diagnostic challenge, as cult to appreciate or absent in such lesions. Despite the paucity
many traditional architectural criteria for the recognition of of conventional architectural criteria, multiple copy number
melanoma are absent. We report the clinical, histological, variations typical of melanoma were demonstrated in these
immunohistochemical, morphometric and cytogenetic lesions by comparative genomic hybridisation (CGH) and/or
features of a series of 12 cases of large nested melanoma. fluorescence in situ hybridisation (FISH). Supported by this
In this series, large nested melanoma accounted for 0.2% molecular data, Kutzner et al. proposed that histological
of cases of melanoma. The majority occurred on the trunk criteria such as asymmetry, large nests with variation in shape
of middle aged patients with absent or minimal solar and size, confluence of nests and moderate to marked cyto-
elastosis and 42% were associated with a component of logical atypia are of value in the recognition of this variant of
benign intradermal melanocytic naevus, speaking to clas- melanoma. In addition, they noted that many cases display a
sification of these melanomas as falling within the spec- distinct cuff of keratinocytes beneath the nests of melanocytes,
trum of lesions developing in skin with low cumulative sun implying that the nests are not truly junctional.
damage. In 67% of cases invasive melanoma was present. In our routine and consultation practices, we have identi-
Criteria such as asymmetry, variation in nest size and fied a number of cases with a striking population of large
intraepidermal nests with an underlying rim of junctional nests, corresponding broadly to the group described as MLN
keratinocytes appear to be highly specific, and are strongly by Kutzner et al. We describe the clinical, morphological,
predictive of typical cytogenetic abnormalities of mela- immunophenotypic and cytogenetic features of these cases.
noma, which were identified in 92% of cases. Conversely,
in addition to features which are definitionally absent or MATERIALS AND METHODS
limited, features such as solar elastosis and cytological The archives of the Department of Anatomical Pathology at PathWest Lab-
atypia do not appear to be particularly helpful in recogni- oratory Medicine for the period 2013 to 2018 were searched for cases of
tion of this variant. melanoma coded as ‘large nested’ using AP Software (Version 8.4.1.4).
Similar cases identified in community practice (Melbourne Pathology, Vic-
Key words: Melanoma; melanocytic naevus; FISH; array CGH; large nested toria) by one of the authors (PM) were also retrieved. A total of 14 cases were
melanoma. identified.
Clinical diagnosis was obtained from review of the request forms. The
Received 17 October 2019, revised 6 February, accepted 17 February 2020 diagnostic material, including haematoxylin and eosin stained sections,
Available online 21 April 2020 immunohistochemically stained sections and cytogenetic results [including
FISH and array CGH (aCGH) results] were reviewed by four dermatopa-
thologists (TNL, NMA, NTH and BAW). After review, one case was
excluded due to incomplete sections and another case was excluded as it was
INTRODUCTION
felt to show prominent features of conventional melanoma (significant
In 2012, Kutzner et al. described the histological and cyto- lentiginous growth and pagetoid spread) in addition to large nests, leaving a
genetic features of melanomas characterised by a striking study group of 12 lesions. Of these cases, six were internal cases from
component of large nests within the epidermis with a PathWest, two were external cases received in consultation at PathWest and
‘cannonball’ appearance. These tumours were described as four were identified from community practice. Over the same time period,
‘melanoma composed exclusively or predominantly of large 2954 cases of melanoma were received at PathWest, giving an incidence of
nests’ (MLN).1 Such lesions present a diagnostic challenge, as approximately 0.2% of cases of melanoma.

Print ISSN 0031-3025/Online ISSN 1465-3931 © 2020 Royal College of Pathologists of Australasia. Published by Elsevier B.V. All rights reserved.
DOI: https://doi.org/10.1016/j.pathol.2020.02.006
432 LEECY et al. Pathology (2020), 52(4), June

Histological and morphometric analysis two DNA samples were mixed and co-hybridised to a SurePrint G3 Human
Haematoxylin and eosin stained sections were produced by standard tech- CGH 8x60K array (Agilent, USA), and digitally scanned to capture and
niques. Solar elastosis was graded from 1 to 3 using the criteria and images quantify the relative fluorescence intensities. The ratio of the fluorescence
provided in the WHO Classification, 4th edition.2 Pagetoid spread and intensities was analysed by Cytogenomics software (Agilent) for further
lentiginous growth were classified as present or absent based on assessment of analysis and detection of potential copy number variations (CNVs).
a representative slide. Cytological atypia was graded as mild, moderate or
severe. The lesions were additionally assessed for the presence of circum- RESULTS
scription of the periphery (defined as the presence of a terminal nest of me-
lanocytes), presence of a rim of basal keratinocytes separating large nests of
Clinical characteristics
melanocytes from the dermis and epidermal consumption. The demographic data are presented in Table 1. The 12 pa-
The stained sections were scanned using Aperio ScanScope AT (Leica tients included 10 males and two females. The age at diag-
Biosystems, Germany). One representative section from each case was nosis ranged from 29 to 83 years, with a median age of 59
selected for morphometric analysis. All the junctional nests within this section
years. The majority of lesions occurred on the back and
were identified by one of the authors (TNL) and the greatest diameter
measured using Aperio Imagescope v12.1.0.5029 (Leica Biosystems, Ger-
shoulder region, with three lesions involving a lower limb.
many). The data were recorded in Microsoft Excel 2010 Spreadsheet No cases involved the head and neck region. The lesions
(Microsoft, USA). ranged from 4 to 14 mm in diameter, with a median size of
9.5 mm. Only one lesion measured under 6 mm. The prof-
Immunohistochemistry fered clinical diagnoses included a range of considerations for
Immunohistochemical staining was available for seven cases, including an- pigmented lesions with varying degrees of clinical concern.
tibodies against p16 (E6H4, cat no. 705-4713; Ventana, USA) in seven cases, In one case, the clinical diagnosis of recurrent naevus was
CKAE1/3 to assess for a rim of basal keratinocytes (Clones AE1 and AE3, cat proffered. This case had previously been diagnosed on
no. M3515; Dako/Agilent, USA) in six cases and V600E mutant BRAF excision biopsy as naevus. On review of those previous
protein (V600E, cat no. E19294; Abcam, UK) in seven cases.
slides, the lesion was reclassified as large nested melanoma
Cytogenetic analysis and subsequent FISH was abnormal.
FISH testing was performed on eight cases. For FISH studies, 5 mm thick,
unstained, formalin fixed, paraffin embedded (FFPE) tissue sections were Histological features
subject to in situ hybridisation. All eight cases were evaluated with a mela-
The histological features are presented in Table 2. The le-
nocytic four-colour FISH kit including probes for RREB1 (6p25), CEP 6
(6p11.1-q11.1), MYB (6q23), and CCND1 (11q13). Seven cases were eval-
sions were generally broad, with striking large nests (Fig. 1).
uated with probes against CDKN2A/Cep9, and probes for MYC and chro- They predominantly involved skin without evidence of
mosome 7q (all Vysis, USA) were additionally employed in one case each. A significant chronic sun damage; seven of the cases showed
standard hybridisation protocol was followed, as previously reported.3 A absent solar elastosis, with no cases showing grade 3 elas-
minimum of 60 tumour nuclei were counted for FISH analysis. Four colour tosis. Nine of the 12 lesions were well circumscribed. A rim
FISH results were considered abnormal if at least one of the following criteria of keratinocytes beneath the nests was present, at least
were met: gain of RREB1 in >29% of nuclei; gain of RREB1 signals relative focally, in all cases (Fig. 2). Pagetoid spread was identified
to centromere 6 signals in >55% of nuclei; loss of MYB signals relative to focally on close inspection in four of 12 cases. Lentiginous
centromere 6 signals in >40% of nuclei; gain of CCND1 signals in >38% of
growth was identified focally in two of 12 cases (Fig. 3).
nuclei. Homozygous loss of CDKN2A was defined as absence of both 9p21
The lesions showed mild to moderate cytological atypia,
signals in the presence of at least one CEP9 signal in at least 20 of 60 nuclei.
aCGH was performed in two cases as previously described.4 In brief, five
with no cases showing high grade atypia. Invasive mela-
25 mm thick unstained FFPE tissue sections were used to extract genomic noma was present in eight cases, in all but one case less than
DNA in each sample. The test DNA and gender mismatched human reference 1 mm in thickness (Fig. 4). A co-existing melanocytic
DNA were differentially labelled with fluorophores. Equal quantities of the naevus was identified in five cases (Fig. 5).

Table 1 Clinical characteristics of the study group

Case Age Sex Site Macroscopic diameter Clinical diagnosis

1 58 M Deltoid 8 mm ? Melanoma
2 69 M Shoulder 11 mm Irregular patch of pigmentation
3 29 M Deltoid 6 mm None provided
4 58 F Thigh/hip 11 mm Recurrent naevus/mole with change
5 31 M Leg 12 mm ? Dysplastic naevus
6 39 M Back 12 mm ? Melanoma
? Pigmented seborrhoeic keratosis
7 47 M Chest wall 7 mm ? Melanoma
8 69 M Back 4 mm Pigmented mole
9 83 M Upper back 14 mm Suspicious lesion
10 71 F Back 12 mm None provided
11 60 M Thigh 6 mm ? Irritated seborrhoeic keratosis
12 68 M Back 8 mm Atypical pigmented lesion,
? basal cell carcinoma,
? melanoma

Case 4 was previously excised and diagnosed as naevus, recurrent lesion 4 mm. On review, the original excision represented melanoma in situ, 11 mm macroscopic
diameter.
 LARGE NESTED MELANOMA 433

Immunohistochemistry

Negative
Negative

Negative

Negative
Positive
Positive
VE1
Immunohistochemistry for BRAF V600E was positive in two
Immunostaining

N/A

N/A

N/A
N/A
N/A
N/A
of six cases tested. Interestingly, there was complete loss of
p16 stain in the same two cases.

Negative
Negative
Positive
Positive
Positive

Positive
p16

Cytogenetic analysis

N/A
N/A
N/A
N/A
N/A
N/A
The cytogenetic features are presented in Table 3. FISH
testing was performed on eight cases, of which seven showed
naevus
naevus

naevus

naevus

naevus
cytogenetic abnormalities typical of melanoma. One case
showed monosomy 9, without abnormalities on melanocytic
Co-existing
lesion

4-colour FISH analysis. Two cases showed homozygous loss

Intradermal
Intradermal

Intradermal

Intradermal

Intradermal
of CDKN2A; both were p16 negative by immunohisto-
chemistry. aCGH was performed on two cases. One case
Nil
Nil
Nil

Nil

Nil
Nil
Nil
showed no significant copy number variation, while the other
showed multiple abnormalities, including gain of 7q which
was also detected on confirmatory FISH testing (Table 3).
thickness

>0.6 mm
Breslow

N/A

N/A
N/A
N/A
0.7
0.3
0.5
0.3

1.1
0.4

0.5

Morphometry
A median of 37.5 nests per case were measured (range
19–90). The nests ranged from 20.6 mm to 856.1 mm (median
elastosis
Solar

116 mm). The distributions of nest size are displayed in Fig. 6.

2
2
0
0
0
0
0
2
2
1
0
0

DISCUSSION
keratinocytes
Rim of basal

A pattern of melanoma characterised by a predominance of

Present
Present
Present
Present
Present
Present
Present
Present
Present
Present
Present
Present

large ‘cannon-ball’ nests within the epidermis was

described in 2012 by Kutzner et al.1 The morphological
diagnosis of melanoma in these cases was supported by the
identification of copy number variation by aCGH. This
description did not include detailed information as to the
Moderate

Moderate
Moderate
Moderate
Moderate
Moderate
Moderate
Moderate
Moderate
Atypia

degree of cumulative sun damage, though it was noted that

Mild
Mild

Mild

‘some’ cases showed solar elastosis. It was proposed that

these cases represented a morphological variant of super-
ficial spreading melanoma,1 corresponding to the group
consumption

classified as ‘melanoma arising in skin with a low degree of

Epidermal

Absent
Absent
Absent
Absent
Absent
Absent
Absent
Absent
Absent
Absent
Absent
Absent

cumulative sun damage’ (low-CSD melanoma) in the

current WHO classification.2
A number of authors have described cases of melanoma
arising in skin with high levels of cumulative sun damage
showing large nests as ‘nested melanoma of the elderly’.5,6
Lentiginous

Present

Present
growth

Absent
Absent
Absent

Absent
Absent
Absent
Absent

Absent
Absent
Absent

It has been proposed that the term ‘nested melanoma in

sun-damaged skin’ might better describe these cases.7
While it has been suggested that these represent the same
lesions as those described by Kutzner et al.,8 some of these
cases may conform more closely to ‘melanoma arising in
invasion
Dermal

Present
Present
Present
Present

Present
Present

Present
Present
Absent

Absent
Absent
Absent

skin with a high degree of cumulative sun damage’ (high-

Histological and immunohistochemical features

CSD melanoma) in the current WHO classification.2

Further studies including molecular analysis are needed
to clarify this.
Pagetoid

Present

Present
Present
Present

Present
Absent

Absent
Absent
Absent
Absent
Absent

Absent
spread

The majority of cases of large nested melanoma identified

in our practice occurred on the upper trunk, with 7/12 cases
(58%) showing no evidence of solar elastosis and a compo-
nent of melanocytic naevus present in 42%. We did not
Circumscription

encounter cases involving severely sun damaged skin of the

head and neck region. The current WHO classification sep-
Yes
Yes
Yes

Yes
Yes
Yes

Yes
Yes
Yes
No

No
No

arates melanocytic neoplasms utilising a multidimensional

pathway model, proposed by Bastian.2,9 Within this system,
N/A, not available.

the clinical and histological features of the cases of large

nested melanoma identified in this series appear to fall
Case no.

broadly within that of Pathway 1, melanomas arising in skin

Table 2

with intermittent sun-exposure, frequently arising from

10
11
12
1
2
3
4
5
6
7
8
9

melanocytic naevus and corresponding broadly to lesions

434 LEECY et al. Pathology (2020), 52(4), June

Fig. 1 The lesions were generally broad and composed of a prominent population of large nests within the epidermis. (A) Case 2, (B) Case 5.

traditionally described as superficial spreading melanoma. is mild to moderate, falling within the spectrum where
Future study to determine the pattern of driver mutations in interobserver variation in interpretation would be expected.
these lesions may be of interest. Importantly, solar elastosis is present in only the minority of
We propose that a predominance of large nests is an un- such cases.
common morphological variant, most often seen in mela- However, when the large nested pattern is recognised,
noma of skin with low levels of cumulative sun damage/ close attention to a number of features can lead to accurate
superficial spreading melanoma and accounting for approxi- recognition. Although often not available at the time of initial
mately 0.2% of cases of melanoma. A predominance of pathological assessment, careful integration with the exami-
nested growth can also be seen in melanoma of skin with high nation/dermoscopic findings appears to be an important
levels of cumulative sun damage. In most such cases, the very confirmatory feature, with most cases in other studies
large ‘cannon-ball’ nests which characterise the large nested presenting with identifiable clinical features of melanoma.1
variant of melanoma are not present and areas of lentiginous Histologically, the lesions are typically asymmetrical, with
growth are frequently described.6 The combination of these significant variation in nest size; typically including nests
features typically renders the diagnosis of melanoma more much larger than the average nest size for a particular lesion.
straightforward in such cases, and likely explains why such The nests are frequently intraepidermal rather than truly
lesions have not been identified as ‘large nested’ variant in junctional, with a surrounding cuff of basal keratinocytes
our practice. identified on close inspection and/or staining with cytokeratin
The significance of the large nested pattern lies in the po- in all cases in our series. When these architectural features are
tential difficulty of applying many of the morphological appreciated, subtle additional clues, such as the presence of
criteria for the diagnosis of melanoma to lesions composed small foci of lentingous growth or pagetoid extension may be
exclusively or predominantly of large nests. Specifically, by evident in some cases.
definition such cases will lack prominent lentiginous growth Seven of eight cases tested in our series showed cytogenetic
and pagetoid extension and are likely to be well circum- abnormalities typical of melanoma, while one showed loss of
scribed, leading to an initial impression of a melanocytic one copy of CDKN2A in 43% of cells studied without other
naevus.10 In our experience, the degree of cytological atypia changes, a finding associated with, but not specific for,
 LARGE NESTED MELANOMA 435

Fig. 2 A peripheral rim of keratinocytes beneath the melanocytic nests was identified at least focally in all cases. (A) Case 5, (B) Case 10.

Fig. 3 This lesion was originally diagnosed as melanocytic naevus. After a recurrence showing features of large nested melanoma developed 3 years later, this case was
reviewed. In retrospect there is striking asymmetry, with poor circumscription and lentiginous growth demonstrated on the left side and large nested growth with basal
keratinocyte rimming demonstrated on the right side (Case 4).

melanoma.11 Given the admixed component of melanocytic the criteria identified above is highly specific for the diagnosis
naevus it is likely that this change was beneath the threshold of melanoma, it is not possible to accurately determine
of detection for aCGH. While this suggests that application of whether cases with normal cytogenetic findings classified as
436 LEECY et al. Pathology (2020), 52(4), June

Fig. 4 Invasive melanoma was seen in eight cases. (A) Case 7, (B) Case 6.

Fig. 5 A component of benign intradermal melanocytic naevus was present in five cases (Case 1).
 LARGE NESTED MELANOMA 437

Table 3 Cytogenetic features of the study group

Case no. FISH aCGH

1 Normal RREB1, MYB, CCND1; hemizygous loss of CDKN2A (43% of cells) No CNVs
2 Absolute and relative gain of RREB1 (90% and 63% of cells respectively), loss of N/A
MYB (50% of cells), normal CCND1, CDKN2A and MYC
3 Absolute and relative gain RREB1(63% and 57% of cells respectively), normal N/A
MYB, CCND1 and CDKN2A
4 Loss of MYB (60% of cells), homozygous loss of CDKN2A (75% of cells), normal N/A
RREB1 and CCND1
5 Absolute gain of RREB1 (40% of cells), homozygous loss of CDKN2A (38% of N/A
cells), normal MYB and CCND1
6 Gain of RREB1 (83% of cells), MYB (83% of cells), normal CCND1 and CDKN2A, 7q gain, 12q gain, 17p loss, 19q loss
gain of chromosome 7q (85% of cells)
7 Absolute gain of RREB1 (43% of cells), normal MYB, CCND1 N/A
8 Absolute and relative gain of RREB1(58% and 77% of cells respectively), normal N/A
MYB, CCND1 and CDKN2A
9 N/A N/A
10 N/A N/A
11 N/A N/A
12 N/A N/A

aCGH, array comparative genomic hybridisation; FISH, fluorescence in situ hybridisation; N/A, not available.

Fig. 6 Graph of nest diameters.

atypical naevi have been excluded. Anecdotally, these cases is extremely unusual in this context. Similarly, the architec-
are likely to be uncommon. In the series reported by Kutzner tural features do not conform well to those of dysplastic
et al.,1 while all cases showed cytogenetic abnormalities by naevus; the bridging growth and lamellar fibroplasia which
aCGH, only four of 11 cases were FISH positive. Given the characterise dysplastic naevus are not features of large nested
potential for misinterpretation as a melanocytic naevus, it variant melanoma.
remains difficult to determine the true incidence of large In conclusion, the pattern of large nested melanoma ap-
nested melanoma and the sensitivity of FISH testing. One of pears to be a recognisable variant of low-CSD melanoma,
the cases in our series had previously been reported as likely representing the most striking end of a spectrum of
melanocytic naevus by a general surgical pathologist, and was nested growth. Recognition of this architectural pattern at
only recognised as large nested melanoma in retrospect after low power is critical to allow application of a distinct subset
recurrence of in situ disease and dermatopathological review. of morphological criteria used to diagnose melanoma. With
The differential diagnosis of the large nested variant of application of these criteria, this group of cases can be
melanoma includes common acquired naevus and dysplastic identified with a high degree of predictive specificity, as
naevus. While the former might be conceptually tempting, defined by cytogenetic support for the diagnosis of
critical reflection would suggest that the large nested pattern melanoma.
438 LEECY et al. Pathology (2020), 52(4), June

Conflicts of interest and sources of funding: Dr Nathan 4. Mesbah Ardakani N, Thomas C, Robinson C, et al. Detection of copy
number variations in melanocytic lesions utilising array based
Harvey is supported by a Raine Medical Research Founda- comparative genomic hybridisation. Pathology 2017; 49: 285–91.
tion Clinician Research Fellowship. The authors have no 5. Longo C, Zalaudek I, Piana S, et al. Dermoscopy and confocal micro-
other funding sources or conflicts of interest to declare. scopy of nested melanoma of the elderly: recognizing a newly defined
entity. JAMA Dermatol 2013; 149: 941–5.
6. Pennacchia I, Garcovich S, Gasbarra R, et al. Morphological and mo-
Address for correspondence: Dr Benjamin A. Wood, Dermatopathology lecular characteristics of nested melanoma of the elderly (evolved
Group, Department of Anatomical Pathology, PathWest Laboratory Medi- lentiginous melanoma). Virchows Arch 2012; 461: 433–9.
cine, QEII Medical Centre, Hospital Avenue, Nedlands, WA 6009, 7. Garbe C, Metzler G. Nested melanoma, a newly defined entity. JAMA
Australia. E-mail: [email protected] Dermatol 2013; 149: 905–6.
8. Petkovic M, Jurakic Toncic R. Nested melanoma, a new morphological
variant of superficial spreading melanoma with characteristic dermo-
scopic features. Acta Dermatovenerol Croat 2017; 25: 80–1.
References 9. Bastian BC. The molecular pathology of melanoma: an integrated tax-
1. Kutzner H, Metzler G, Argenyi Z, et al. Histological and genetic evi- onomy of melanocytic neoplasia. Ann Rev Pathol 2014; 9: 239–71.
dence for a variant of superficial spreading melanoma composed pre- 10. Harvey NT, Wood BA. A practical approach to the diagnosis of mela-
dominantly of large nests. Mod Pathol 2012; 25: 838–45. nocytic lesions. Arch Pathol Lab Med 2019; 143: 789–810.
2. Elder DE, Massi D, Scolyer RA, et al. WHO Classification of Skin 11. Casorzo L, Luzzi C, Nardacchione A, et al. Fluorescence in situ hy-
Tumours. Lyon: IARC Press, 2018. bridization (FISH) evaluation of chromosomes 6, 7, 9 and 10 throughout
3. Gerami P, Jewell SS, Morrison LE, et al. Fluorescence in situ hybrid- human melanocytic tumorigenesis. Melanoma Res 2005; 15: 155–60.
ization (FISH) as an ancillary diagnostic tool in the diagnosis of mel-
anoma. Am J Surg Pathol 2009; 33: 1146–56.