If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Melanoma Institute Australia, The University of Sydney, Sydney, NSW, AustraliaTissue Pathology and Diagnostic Oncology, Royal Prince Alfred Hospital and NSW Health Pathology, Sydney, NSW, AustraliaFaculty of Medicine and Health, The University of Sydney, Sydney, NSW, Australia
Address for correspondence: Prof Richard Scolyer, Tissue Pathology and Diagnostic Oncology, Royal Prince Alfred Hospital, Building 94 (via John Hopkins Drive), Missenden Road, Camperdown, NSW 2050, Australia.
Melanoma Institute Australia, The University of Sydney, Sydney, NSW, AustraliaTissue Pathology and Diagnostic Oncology, Royal Prince Alfred Hospital and NSW Health Pathology, Sydney, NSW, AustraliaFaculty of Medicine and Health, The University of Sydney, Sydney, NSW, AustraliaCharles Perkins Centre, The University of Sydney, Sydney, NSW, Australia
Melanoma Institute Australia, The University of Sydney, Sydney, NSW, AustraliaTissue Pathology and Diagnostic Oncology, Royal Prince Alfred Hospital and NSW Health Pathology, Sydney, NSW, AustraliaFaculty of Medicine and Health, The University of Sydney, Sydney, NSW, Australia
Acquired and congenital melanocytic naevi are common benign neoplasms. Understanding their biology and genetics will help clinicians and pathologists correctly diagnose melanocytic tumours, and generate insights into naevus aetiology and melanomagenesis.
Genomic data from published studies analysing acquired and congenital melanocytic naevi, including oncogenic driver mutations, common melanoma associated mutations, copy number aberrations, somatic mutation signature patterns, methylation profile, and single nucleotide polymorphisms, were reviewed. Correlation of genomic changes to dermoscopic features, particular anatomic sites and total body naevus counts, was also performed. This review also highlights current scientific theories and evidence concerning naevi growth arrest.
Acquired and congenital melanocytic naevi show simple genomes, typically characterised by mutually exclusive single oncogenic driver mutations in either BRAF or NRAS genes. Genomic differences exist between acquired and congenital naevi, common and dysplastic naevi, and by dermoscopic features. Acquired naevi show a higher rate of BRAF hotspot mutations and a lower rate of NRAS hotspot mutations compared to congenital naevi. Dysplastic naevi show upregulation of follicular keratinocyte-related genes compared to common naevi. Anatomical locations and DNA signatures of naevi implicates ultraviolet radiation and non-ultraviolet radiation pathways in naevogenesis.
DNA driver point mutations in acquired and congenital melanocytic naevi have been well characterised. Future research is required to better understand transcriptional and epigenetic changes in naevi, as well as those regulating naevus growth arrest and cell environment signalling.
The origin and development of acquired naevi (AN) and congenital melanocytic naevi (CMN) has been a topic of some controversy, but recent discoveries utilising modern molecular techniques have provided new important insights. As implied by the terminology, CMN are present at birth, and AN develop after birth. In this review, the term AN refers to benign common or dysplastic naevi, and excludes other specific types of naevi, including blue, Spitz, BAP-1 deficient, and deep penetrating naevi.
The evolutionary development of naevi is not yet well understood. Melanocyte progenitor cells are thought to travel to the skin from the neural crest via two pathways during embryogenesis: dorsolaterally under the epidermis, and ventrally along nerves.
Conversely, Cramer's theory, Hochsteigerung (‘upward climbing’), proposed almost one century later, involves naevi developing in the dermis and migrating upwards to the epidermis over time.
Examining the molecular findings in AN and CMN have helped to and will continue to help further elucidate naevogenesis pathways and biology.
Understanding the genetics of AN and CMN has evolved over the past decade from data generated from gene hotspot mutational analyses (such as in BRAF and NRAS genes), to a deeper understanding of broader DNA, RNA and epigenetic changes, as sophisticated molecular analysis techniques have become more readily available. In an effort to appreciate the aetiology, clinicopathological features and potentially helpful diagnostic features, in this review we focus on the genomic changes related to the biology of these naevi, incorporating recent findings additional to hotspot mutational analyses. Characterising the genomic changes in naevi can enhance understanding of melanomagenesis, and assist with the pathological interpretation of borderline melanocytic neoplasms.
Acquired naevi
Clinicopathological features
The clinical morphology of AN varies, including the colours, size, borders, symmetry, dermoscopic features, and surface profile (flat/raised).
The colour of melanocytic naevi depends on where the melanocytes are situated in the epidermis, due to the Tyndall effect of how light is reflected, whereby melanin in the upper epidermis is seen as black, at the dermo-epidermal junction seen as brown, in the papillary dermis seen as blue-grey, and in the reticular dermis seen as blue.
AN may be further classified into common or dysplastic naevi (along with grading of dysplasia), for which the distinction may be subjective and somewhat controversial.
Features of dysplastic naevi include architectural disorder and cytological atypia, elongated rete, papillary fibroplasia, a lymphocytic infiltrate and neovascularisation.
Furthermore, there are specific histological types of AN, including naevus with site-related atypia, regenerative naevus, atypical lentiginous naevus of the elderly (a specific type of dysplastic naevus), and halo naevus. AN have also been eponymously named (e.g., Unna, Miescher and Clark naevi) which reflect different histomorphological growth patterns.
Activation of the mitogen-activated protein kinase (MAPK) pathway plays an important role in the development of AN (Fig. 1). BRAF and NRAS proteins are important components of this intracellular signalling pathway. Almost all AN have a mutually exclusive activating hotspot point mutation found in either the BRAF or NRAS gene.
The BRAF and NRAS mutation prevalence in dermoscopic subtypes of acquired naevi reveals constitutive mitogen-activated protein kinase pathway activation.
The BRAF and NRAS mutation prevalence in dermoscopic subtypes of acquired naevi reveals constitutive mitogen-activated protein kinase pathway activation.
The BRAF and NRAS mutation prevalence in dermoscopic subtypes of acquired naevi reveals constitutive mitogen-activated protein kinase pathway activation.
The BRAF and NRAS mutation prevalence in dermoscopic subtypes of acquired naevi reveals constitutive mitogen-activated protein kinase pathway activation.
Fig. 1Simplified overview of the mitogen-activated protein kinase (MAPK) pathway. Binding of growth factors to tyrosine kinase receptors triggers activation of the pathway, through a series of phosphorylations. NRAS G12/13 and Q61 mutations, and BRAF V600 mutations lead to constitutional activation, without requiring RTK-GF or RAS activation respectively. GF, growth factor; RTK, receptor tyrosine kinase.
Mutational events in these MAPK pathway genes appear sufficient per se for naevogenesis. Yeh et al. showed that a single copy of a BRAF V600E mutated allele was present in the majority, if not all melanocytes in BRAF V600E AN (n=8), providing evidence for the clonal origin of AN, with a single BRAF V600E mutation occurring as an initiating event.
The BRAF and NRAS mutation prevalence in dermoscopic subtypes of acquired naevi reveals constitutive mitogen-activated protein kinase pathway activation.
Nevertheless other potential explanations for this data include factors such as keratinocyte, lymphocyte or non-neoplastic melanocyte contamination and false positives or negatives related to technical artefacts.
How and when these BRAF and NRAS mutations occur has been debated. While ultra-violet light radiation has been implicated in naevogenesis (see later section on DNA signature patterns),
Kittler and Tschandl hypothesised that a BRAF V600E mutation in melanocytes may occur in the neural crest early during fetal life, or during migration to the skin, and these mutated melanocytes then wait as ‘invisible seeds until further events initiate their proliferative potential’, as the so-called ‘seed hypothesis’.
did not find BRAF or NRAS somatic mutations in the adjacent skin of naevi, and does not explain how a BRAF V600E mutated compound naevus could arise in an ovarian teratoma.
Mutation rates in naevi at acral and mucosal anatomical locations
Insights into the aetiology of DNA mutations and aberrations in naevi may be gained by evaluating naevi in specific anatomical locations, such as at mucosal and acral sites. Indeed, the high frequency of BRAF V600E mutations in naevi from solar protected sites (Table 1) indicates that this mutation can occur independently to DNA damage caused by ultraviolet radiation, as alluded to above. Acral naevi show a similar frequency of BRAF and NRAS mutations to cutaneous AN.
In addition to BRAF and NRAS mutations in acral naevi, Moon et al. found GNAQ (n=8, 38.1%), ALK (n=5, 23.8%), KIT (n=3, 14.3%) and NF1 (n=2, 9.5%) mutations, which commonly co-occurred with a BRAF mutation.
The higher rate of NRAS mutations in conjunctival naevi (compared to cutaneous AN) reported by Francis et al. was postulated to be related to the inclusion of congenital naevi, and in support of this theory the authors found that intrinsic cysts were statistically associated with NRAS-immunoreactive naevi.
and it would seem most likely that these are passenger mutations rather than co-contributors to naevogenesis, although the spectrum of co-mutations is diverse and co-contributions/interactions are possible. The position of the mutation in these non-BRAF/NRAS genes usually varies, in contrast to the regularity in patterns seen in BRAF and NRAS genes.
In BRAF V600 mutated naevi, when some of these other less common mutations are detected, they are typically found at a similar variant allele frequency to the BRAF mutation, implying these mutations took place at the same time.
Telomerase reverse transcriptase (TERT) promoter and cyclin-dependent kinase inhibitor 2A (CDKN2A) genes have received significant attention by researchers and clinicians in recent years due to their association with melanomagenesis.
Given the emerging role of using TERT promoter mutations and CDKN2A copy number changes as diagnostic adjuncts in melanocytic pathology interpretation (with the presence of aberrations supporting a diagnosis of melanoma),
in 2013. Specific hotspot TERT promoter mutations generate new binding sites for specific (E26) transcription factors, leading to increased DNA transcription.
These specific TERT promoter mutations are rarely found in AN, but are common in melanoma, and are thought to be an important molecular event in the transformation of benign melanocytes or naevus cells to melanoma.
Colebatch et al. identified two of 11 AN that harboured subclonal TERT promoter mutations (0.2% allelic frequency), which may reflect early progression towards melanoma not seen morphologically, or adjacent keratinocyte contamination.
This also reflects the findings reported by Shain et al., where TERT promoter mutations were found as the earliest secondary alterations in intermediate melanocytic tumours progressing towards melanoma.
mRNA data indicates increased expression of this gene in AN melanocytes compared to non-naevus melanocytes, which may be important for limiting naevus cell growth potential.
Loss of heterozygosity at chromosome 9p21 (INK4-p14ARF locus): homozygous deletions and mutations in the p16 and p14ARF genes in sporadic primary melanomas.
Diffuse loss of p16 expression in melanocytes using immunohistochemistry is commonly used a surrogate marker for CDKN2A homozygous gene loss, however there are other mechanisms for loss of immunohistochemical p16 expression that do not correlate with CDKN2A homozygous gene loss.
DNA signature patterns and copy number aberrations in naevi
Further clues to understanding naevogenesis can be unearthed through evaluation more broadly across the naevus genome, both at the patterns of single-based substitution somatic mutations and copy number aberrations. Patterns of base-substitutions may arise as a consequence of different mutational processes, and these patterns have been grouped by ‘signatures’ correlating to their underlying mutational aetiology.
Signatures for defective DNA mismatch repair were relatively common (33%, 10/30 naevi), suggesting that defective DNA repair contributes to naevogenesis.
Interestingly, the ultraviolet signature (signature 7) was present in 97% of naevi, and only 10% of perilesional skin. This led Stark et al. to postulate that melanin content and breakdown products could contribute to this ultraviolet DNA pattern.
Fig. 2Naevus genome mutational signatures. (A) Absolute contribution of mutational signatures to overall single nucleotide variant count for each naevus. (B) Relative contribution of mutational signatures to total overall single nucleotide variant count for each naevus. Signature associations: 1, age related; 3, failure of DNA double-strand break repair by homologous recombination; 5,12,16, aetiology unknown; 6, defective DNA mismatch repair; 7a/b/c, ultraviolet radiation high.
and in a trinucleotide context (N[C>T]N), there are differences in transitions between intradermal naevi compared to junctional/compound naevi, but not between dysplastic and common naevi.
While similar trinucleotide C>T transitions were found in naevi compared to melanomas, there were differences in certain trinucleotide transition ratios, and this correlated with deleterious mutations in tumour suppressor genes being more common in melanomas.
Copy number aberrations in naevi, if present (see below with dermoscopic correlations), are typically large (as opposed to melanoma, which shows more focal regions of gain and loss) and implies a random mutagenic process as opposed to specific genes being targeted.
While the pathological distinction between dysplastic and common naevi has been controversial, genetic analysis has also revealed differences between them.
Mitsui et al. showed that dysplastic naevi demonstrate a unique form of dysplasia, with complex interplay between epidermal keratinocytes and melanocytes.
Dysplastic naevi showed significant upregulation of follicular keratinocyte-related genes compared to common naevi, in addition to tumour microenvironment differences, that appear to drive formation of dysplastic naevi.
Melamed et al., using whole-exome sequencing, showed that dysplastic naevi have a median non-synonymous mutation frequency of 21, compared to 10.5 for common naevi, with both frequencies much lower than melanoma.
The BRAF and NRAS mutation prevalence in dermoscopic subtypes of acquired naevi reveals constitutive mitogen-activated protein kinase pathway activation.
In dysplastic naevi, differentially methylated loci (compared to the adjacent skin), were enriched for active gene promoters and gene regulatory regions.
Furthermore, dysplastic naevi showed significant DNA methylation differences (either in the gene body or promoter regions) in candidate oncogenes and tumour suppressor genes (MAP2K1, CDKN2A, PIK3CA, MDM2).
Dermoscopic features of naevi and their genomic correlates
The dermoscopic patterns seen in naevi also correlate to underlying genetic events. The dermoscopic patterns of naevi are broadly classified as globular (round to oval structures), reticular (net-like pattern), homogenous (structureless pattern), or complex (representing overlapping and other distinct or diverse patterns not otherwise categorised).
Globular naevi correlate to dermal or large dermo-epidermal nests of melanocytes, while reticular naevi correlate to a lentiginous and small nested growth pattern along the dermo-epidermal junction, predominantly in rete ridges with relative sparing of the suprapapillary plates.
The BRAF and NRAS mutation prevalence in dermoscopic subtypes of acquired naevi reveals constitutive mitogen-activated protein kinase pathway activation.
The BRAF and NRAS mutation prevalence in dermoscopic subtypes of acquired naevi reveals constitutive mitogen-activated protein kinase pathway activation.
Globular naevi showed a higher proportion of T[C>T]A transitions, and a lower proportion of insertion/deletions compared to reticular/nonspecific naevi.
Reticular/non-specific naevi, compared to globular naevi, also demonstrated a higher proportion of copy number aberrations, which appeared to be large and regional, mainly involving loss of heterozygosity, implicating both tumour suppressor genes and oncogenes simultaneously.
Stark et al. postulated that this combined loss of both tumour suppressor genes and oncogenes in naevi with copy number alterations, gave an overall copy neutral effect, helping keep the naevus in a state of equilibrium.
DNA methylation analysis showed that globular naevi had no significantly differentially methylated CpG loci compared to the adjacent normal skin, which was not the case for reticular/non-specific naevi.
Single nucleotide polymorphisms (SNPs) in IRF4 and TERT showed an association with globular naevi, while SNPs in CDKN1B, MTAP and PARP1 were associated with reticular naevi.
Fig. 3Globular naevus on dermoscopy (A), with corresponding photomicrograph demonstrating large junctional nests of melanocytes as the histology correlate (B). BRAF V600E immunohistochemistry is positive (brown chromogen) in the naevus (C). (Dermoscopy image courtesy of Dr Genevieve Ho and Stephanie Friedlander, Melanoma Institute Australia).
Total body naevus counts and associations with genomic features
Total body naevus counts show increasing frequencies over the first three decades of life, rising rapidly during the second decade of life, and subsiding after the third decade.
Scope et al. showed volatility in naevus biology during early adolescence, with new naevi developing and naevi disappearing in the same individuals between ages 11 to 14, with greater volatility related to higher baseline naevus counts.
Interestingly, none of the naevi in the longitudinal study by Scope et al. that disappeared, grew smaller, or faded showed evidence of a halo or dermoscopic regression structures, and the authors hypothesised that non-immunological mechanisms could be causing their involution.
Understanding the correlation of total body naevus counts with genomic data also inform the nature of naevogenesis pathways. The estimated total heritability of naevus counts is 58%, with contributions from every chromosome and one-sixth from chromosome 9 alone.
A meta-analysis of 11 naevus genome-wide association studies (GWAS) (naevus type not defined), with single-nucleotide polymorphism (SNP) data from 52,806 individuals, revealed five genomic regions were significantly associated with total naevus counts, including MTAP/CDKN2A, IRF4, KITLG, DOCK8, and PLA2G6.
Statistical heterogeneity was encountered for some of these regions (IRF4, MTAP, PLA2G6, and DOCK8), which was expected for IRF4 given a known age-relationship with this gene.
IRF4 is a gene associated with phenotypic pigmentation traits, and a particular IRF4 genotype is associated with high naevus counts during childhood and adolescence, with the direction of this trend reversing with older age.
18 pleiotropic loci containing SNPs were associated with both melanoma risk and naevus count (with the strongest loci being MTAP, PLA2G6, and an intergenic region on 9q31.1).
Interestingly, no ‘pure naevus’ loci were identified when the analysis combined melanoma and naevus GWAS data, although KITLG showed more of a naevus-only pattern on bivariate analysis.
Pathway analysis (using genome-wide SNP data) from Duffy et al., found that the contribution of the telomere maintenance pathway to naevus count variation was 0.8%, and a contribution from the immune regulation/checkpoint pathway was absent.
It is a highly polymorphous gene, with certain variant alleles linked to characteristic phenotypic pigmentation traits, as well as sensitivity to ultraviolet light.
DNA methylation patterns in healthy skin are associated with total body naevus counts, with implicated genomic loci involved in melanocyte biology and cancer.
Selected SNPs associated with higher naevus counts (featured in genes PLA2G6 and NID1) were also found to be associated with higher nearby methylation levels.
CMN show a diverse range of clinical morphologies, including colour variegation, surface rugosity, hypertrichosis, subcutaneous nodules, and satellitosis.
CMN exhibit a range of sizes, and their classification according to predicted adult size reflects this (small <1.5 cm, medium 1.5–20 cm, large >20–40 cm, and giant >40 cm).
A subset of CMN are spilus-like, referring to a morphology with large lightly coloured café au lait macules with superimposed darker maculopapular speckling.
Congenital histopathological features include location in the lower dermis and subcutis, growth around and within adnexa, nerves and vessel walls, and corded growth between collagen bundles, none of which are pathognomonic for CMN (Fig. 4).
Fig. 4A predominantly intradermal naevus (A) showing congenital features including deep extension to the dermal-subcutaneous junction, and extension around adnexae, the latter highlighted by the zoomed image (B), composed of melanocytes with banal cytological appearance (C).
Mutually exclusive BRAF mutations are also found in CMN, with varying frequencies reported, usually less common than NRAS, and more typically, but not exclusively, found in smaller CMN.
Usually this is a BRAF V600E mutation, however Martins da Silva et al. did report one patient who had a spilus-type CMN with both BRAF G464E and BRAF L584F mutations (present differentially within the same CMN).
A variety of other less common gene mutations in CMN have also been described, including in GNAQ, KRAS, HRAS, PIK3CA, APC, MET, EGFR, TMEM2, LFNG, and KRT81,
Martins da Silva et al. importantly highlighted the genetic diversity within CMN, where different mutational patterns were detected in the same lesion (from three spilus-type CMN, two of which were NRAS/BRAF wildtype).
As further support to this finding, Stark et al. reported an LFNG point mutation as a driver mutation for one patient with a CMN, where it was detected in both the CMN and a satellite.
Polubothu et al. recently found PPP2R3B duplications in 12.5% of CMN (3/24), including in one wild-type BRAF/NRAS CMN, which led to the discovery of the C21orf91 gene as a MITF-independent mediator of melanocyte behaviour.
Proliferative nodules in CMN show similar rates of BRAF and NRAS mutations compared to the background CMN, supporting their evolution from a common precursor.
Chromosomal aberrations are relatively common in proliferative nodules, and typically involve whole chromosomes (as opposed to a greater mixture of partial and whole chromosomal aberrations seen in melanoma).
Understanding the mechanism of how AN and CMN stop growing is an important topic, as its failure likely relates to melanomagenesis. The clinicopathological features of naevi influence how such hypothesised models are conceptualised and tested, which include observations such as AN usually being of a consistent similar size (<6 mm),
that naevi can regenerate, can become mitotically active (such as in pregnancy, following trauma or proliferative nodules), and may change in morphology over time. Growth arrest in AN and CMN has been hypothesised to occur via a model of active oncogene-induced senescence (OIS).
Under this model, the acquisition of an oncogenic (BRAF or NRAS) somatic mutation leads to rapid initial cellular proliferation, formation of a clonal progeny, followed by a cell cycle arrest and cellular senescence.
McNeal et al. showed BRAF V600E mutated naevi induce TGF-β expression, an extracellular signalling protein, presumably via MAPK pathway activation, that in turn upregulates tumour suppressor proteins such as p15, to reduce mitotic activity.
Under this model, the rate of production, diffusion through the tissue, and half-life of TGF-β would determine the tolerated naevus mass (with feedback loops involved).
clarified and advanced this topic further, by showing that a cell-intrinsic model of growth arrest (where oncogene expression in a cell induces stress within the cell causing it to arrest) does not best explain naevus clinical behaviour, and suggest that alternative models involving networks of cell-cell communication and adjacent cell-sensing better explain naevus size distributions and development.
AN exhibit limited and well characterised somatic genomic alterations, with some variation corresponding to different dermoscopic and histological features. Most AN show single somatic point mutations in the MAPK pathway genes BRAF and NRAS as isolated oncogenic events. CMN show variable genomic features, with different BRAF and NRAS mutational frequencies compared to AN. Deeper understanding of the genetics and biological formations of these types of naevi will play an important role in understanding melanomagenesis and identifying targets for therapeutic benefit in melanoma.
Conflicts of interest and sources of funding
NM is supported by a Fellowship from Deborah McMurtrie and John McMurtrie AM, through Melanoma Institute Australia, Australia. RAS is supported by a National Health and Medical Research Council of Australia (NHMRC), Australia Practitioner Fellowship (APP1141295). RAS has received fees for professional services from MetaOptima Technology Inc., F. Hoffmann-La Roche Ltd, Evaxion, Provectus Biopharmaceuticals Australia, Qbiotics, Novartis, Merck Sharp & Dohme, NeraCare, AMGEN Inc., Bristol-Myers Squibb, Myriad Genetics, GlaxoSmithKline.
Acknowledgements
Support from colleagues at Melanoma Institute Australia and Royal Prince Alfred Hospital is gratefully acknowledged.
References
Mort R.L.
Jackson I.J.
Patton E.E.
The melanocyte lineage in development and disease.
The BRAF and NRAS mutation prevalence in dermoscopic subtypes of acquired naevi reveals constitutive mitogen-activated protein kinase pathway activation.
Loss of heterozygosity at chromosome 9p21 (INK4-p14ARF locus): homozygous deletions and mutations in the p16 and p14ARF genes in sporadic primary melanomas.
00683-3&id=gr1.jpg){kind=link}
00683-3&id=gr2.jpg){kind=link}
00683-3&id=gr3.jpg){kind=link}
00683-3&id=gr4.jpg){kind=link}