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
With the elucidation of the genetics of melanocytic tumours, new concepts have emerged. An important one is the identification of ‘intermediate’ melanocytic tumours, those with genetic progression events beyond those of melanocytic naevi but that are not fully malignant. Thus, melanocytic tumours exist on a genetic spectrum that likely corresponds to biological behaviour. There are multiple pathways to melanoma development with different initiating events and characteristic benign melanocytic neoplasms and the precise placement of tumours on these pathways remains to be established and the corresponding risks of progression quantified.
In this review, I discuss the classification of melanocytic naevi based on clinical, histopathological and genetic features, as well as the concept of melanocytomas with discussion of specific recognisable subtypes.
A melanocytic naevus is a benign tumour of melanocytes. While melanomas can arise within naevi, the risk of progression is so low that removal of the melanocytic naevus is generally not considered beneficial for the patient, although an increased number of melanocytic naevi is associated with increased risk of melanoma.
Genetic mutations continue to accumulate within each of the melanocytes of a naevus, and those mutations that provide a survival or growth advantage are selected for and can lead to melanoma. Thus, the risk of melanoma development from a melanocytic naevus is related to the number of melanocytes present, explaining the increased risk of melanoma in giant congenital naevi and in patients with more acquired melanocytic naevi.
While the ability for cells to leave a tumour, enter the vascular circulation, exit the vasculature and persist at a distant site was for a long time considered to be a characteristic of cancer,
In melanocytic naevi, clusters of melanocytes can be observed within vascular structures, and this is thought the be the source of melanocytic naevi that occur within the subcapsular space of lymph nodes.
In this section, we will discuss the clinical, histopathological and genetic features of melanocytic naevi. The mitogen-activated protein (MAP) kinase pathway is of central importance in melanocytic neoplasia (Fig. 1) and mutational activation of this pathway is found in most melanocytic naevi.
This is the most common melanocytic naevus seen both in the clinic and as biopsy specimens. They arise starting in childhood and typically accumulate throughout the first three decades of life and then undergo involution in later life.
Increased sun exposure (particularly intermittent exposure and a history of sunburn) and genetic factors (including polymorphisms in pigment synthesis genes) are associated with increased numbers of them.
Histopathologically, they are composed of melanocytes with small ovoid nuclei and moderate to scant cytoplasm that are arrayed in rounded collections (nests). When melanocytes are present in the dermis, they display diminishing pigmentation, size of nests, and cell and nuclear size with distance from the epidermis, features that are part of what is known as maturation. Most common acquired melanocytic naevi harbour a BRAF V600E mutation in every constituent melanocyte and a minority harbour an activating NRAS mutation or other alteration.
The presence of a BRAF V600E mutation is associated with predominantly nested melanocytes within the epidermis, larger nests of melanocytes within the epidermis and larger melanocytes.
Common acquired melanocytic naevi arise from melanocytes that are exposed to ultraviolet (UV) light prior to their initiation, as evidenced by the high burden of UV signature mutations present within them.
The superficial aspects of them continue to be exposed to ultraviolet light, with each melanocyte acquiring different mutations. The mutations that occur after initiation are unique to each melanocyte and are not detectable in formalin fixed samples by current methods unless there is subsequent clonal expansion.
A subset of common acquired melanocytic naevi have a superficial congenital growth pattern which is characterised by extension of dermal melanocytes into the reticular dermis and extension around adnexal structures within the adventitial dermis. It is unclear if melanocytic naevi with a superficial congenital growth pattern differ from those without by their genetic profile, cell of origin, or age.
Most nodal naevi (small deposits of melanocytes in the capsule or subcapsular space of lymph nodes) consist of melanocytes that resemble those of common acquired melanocytic naevi. That these nodal naevi remain relatively small indicate that they are limited in their ability to proliferate.
Blue naevus
Blue naevi are characterised by dendritic pigmented melanocytes and melanophages, often positioned between sclerotic collagen bundles. There are both hypopigmented and hyperpigmented variants. The cellular variant contains central fascicles of spindled melanocytes with ovoid nuclei.
Blue naevi are typically intradermal without changes in their cellular morphology with depth in the dermis (they lack maturation). They are more common on the dorsal surfaces of the hands and feet as well as the scalp and present as blue/black uniform papules.
Blue naevi are initiated by activating mutations of the Gαq pathway, typically point mutations in GNAQ or GNA11 and less commonly hotspot mutations in CYSLTR2 of fusions of protein kinase C (PKC) isoforms.
CYSLTR2-mutant cutaneous melanocytic neoplasms frequently simulate “pigmented epithelioid melanocytoma,” expanding the morphologic spectrum of blue tumors: a clinicopathologic study of 7 cases.
They share initiating mutations with uveal naevi and uveal melanoma. Uveal melanocytic neoplasms arise from melanocytes of the uveal tract which are pigmented, dendritic and not associated with epithelium. Mutations of the Gαq pathway are also found in melanocytic proliferations of the leptomeninges.
Melanocytes in the skin exist in multiple microenvironmental niches and can develop from disparate developmental pathways. In the dorsolateral developmental pathway, melanoblasts from the neural crest migrate dorsolaterally through the mesenchyme and enter the epithelium.
In the ventromedial pathway, bipotent stem cells exist along developing nerves with the potential to develop into Schwann cells or melanocyte stem cells that migrate from developing nerves into the epithelium.
In mice, melanocytes from the ventromedial pathway are increased on the face and limbs. In a mouse model, melanocyte stem cells of the ventromedial pathway can be initiated by Gαq mutation, suggesting that these melanocytes may be the cell of origin for blue naevi.
Spitz naevi are less common than common acquired melanocytic naevi and have a lower peak age of incidence, thought to occur during early childhood. Many are reported to have a more noticeable initial growth phase and greater final size.
Spitz naevi contain polygonal and fusiform melanocytes, with larger nuclei and more abundant cytoplasm than common acquired naevi, particularly in their superficial aspect. They often display single melanocytes within the upper levels of the epidermis (pagetoid scatter) and clefting around melanocytes, features that are concerning in melanocytic tumours of the low-CSD pathway.
Cytological atypia may be notable, particularly in children, and maturation may be minimal. The differential diagnosis includes melanoma, but as melanoma is exceedingly rare in children and Spitz naevi more common, careful consideration of these two entities is typical in paediatric cases. However, Spitz naevi are not limited to children and we suspect they are more likely classified as melanoma in the adult setting. While different diagnostic criteria for malignancy are used for tumours of the low-CSD versus Spitz sublineage, in a study of diagnostic reproducibility in assessment of melanocytic tumours, little agreement was seen between dermatopathologists in classifying tumours along these lines. As expected, this agreement is improved with ancillary molecular assessment.
The majority of initiating mutations in Spitz naevi are a consequence of structural rearrangements (rather than point mutations) and include fusions of receptor tyrosine kinases, including ALK, ROS1, NTRK1/2/3, RET, MET and MERTK and fusions of kinases of the MAPK pathway (BRAF, MAP3K8).
Not only is there a diverse spectrum of kinases that are rearranged in Spitz naevi, but the fusion partners can be highly diverse. The regulatory domain of the kinases is absent in the fusion genes. All of the kinase genes that are rearranged, except for MAP3K8, have a 5′ regulatory domain that is replaced by the fusion partner whose promoter also regulates transcription of the fusion gene. In many cases, the fusion partner contributes domains that promote dimerisation of the fusion kinase and increase signaling output.
Specific histopathological features have been associated with ALK, BRAF, NTRK1, and NTRK3 fusions (Fig. 2). While these features do not allow precise segregation of Spitz naevi by initiating oncogene based on histopathology, they can guide the application of ancillary methods of fusion detection, such as immunohistochemistry. ALK and ROS1 are not expressed in normal melanocytes, but the fusion kinase is under control of the promoter of the fusion partner. Thus, immunohistochemistry for the kinase domains of ALK or ROS1 is highly sensitive and specific for the presence of a fusion of the respective kinase. MET, NTRK1 and NTRK3 are expressed in melanocytes, but strong uniform expression of the kinase domain is associated with the presence of a fusion kinase but with less sensitivity and specificity. For a subset of kinase fusions, the expression level of athe kinase domain does not appear to differ significantly in naevi with or without the fusion and thus the corresponding immunohistochemistry is not informative (i.e., BRAF or MAP3K8).
Fig. 2Histopathological features of Spitz naevi. (A) Fascicular Spitz naevus with ALK fusion. (B) ALK immunohistochemistry highlights the fascicular nests in A. (C) Spitz naevus with NTRK1 fusion and rosette-like structures. (D) NTRK1 immunohistochemistry of C demonstrates strong uniform expression throughout the melanocytes. (E) Spitz naevus with BRAF fusion and storiform fibrosis.
The association of different initiating mutations with different histopathological features and the diversity of initiating oncogenic mutations in Spitz naevi may contribute to the lack of interobserver agreement in classification of a tumour of Spitz sublineage by histopathology, as well as a lack of interobserver agreement on the classification of tumours on the spectrum of benign to malignant.
There are no known risk factors for the development of Spitz naevi; they are not known to be related to a history of sun-exposure. We did report a patient with ring chromosome 7 syndrome with multiple Spitz naevi initiated by rearrangements of BRAF on the ring chromosome 7.
Thus, a propensity for structural rearrangements that can give rise to Spitz naevi (such as ring chromosome 7) could be associated with increased risk of Spitz naevus. The reason that Spitz naevi have a lower peak age of incidence as compared to common acquired naevi is not known. It may be related to the type of mutation that is most common (structural rearrangement versus point mutation) or the cell of origin (not identified, but perhaps a distinct cell type).
Recently we showed that in a patient with eruptive Spitz naevus arising in the fifth decade of life, the too numerous to count small individual eruptive naevi were clonal with the same ROS1 fusion, suggesting that wide-spread haematogenous dissemination of Spitz naevus can occur without portending a poor outcome.
The patient remains alive and well years after onset.
Congenital melanocytic naevus
Congenital melanocytic naevi are melanocytic naevi in which the initiating mutation was acquired by the ancestral cell prior to birth. One can consider two scenarios at opposite ends of the spectrum with respect to timeframe. The earliest developing congenital melanocytic naevus would occur by mutation in the first cell that could give rise to a melanocyte naevus without significantly affecting viability. In fact, some congenital naevi arise from a pluripotent stem cell that is not committed to the melanocytic lineage, giving rise to birthmarks affecting multiple developmental lineages. For example, phacomatosis pigmentokeratotica is characterised by co-occurring and overlapping epidermal naevus and naevus spilus and the same HRAS activating mutations have been shown to be present in both portions, suggesting transformation of a ectodermal pluripotent stem cell that can give rise to epithelium and neural crest.
Similarly, phakomatosis pigmentovascularis, characterised by overlapping vascular birthmark and dermal melanocytosis is caused by point mutations in GNAQ or GNA11.
At the opposite end of the spectrum, one could imagine a melanocyte acquiring a naevus initiating mutation in the moment before birth. This initiated melanocyte would grow into a melanocytic naevus after birth. Because pigmentation of melanocytes increases in the newborn period, some congenital melanocytic naevi may not be clinically apparent at birth.
Congenital melanocytic naevi can be classified by their clinical features including size, localisation, number of satellite naevi and morphological characteristics (colour, rugosity, nodularity, hypertrichosis).
This classification should help stratify patients by their risk for adverse outcomes, including neurocutaneous melanosis and progression to melanoma. Interestingly, for congenital melanocytic naevi, the spectrum of initiating mutations appears correlated to the size of the naevus. The majority of giant congenital melanocytic naevi harbour NRAS mutations with a minority harbouring kinase fusions or BRAF mutation,
Giant congenital melanocytic naevi highlight the relationship between the overall risk of malignant transformation within a given tumour corresponding both to how many additional mutations are required for malignant transformation and how many cells are at risk. While each melanocyte within a giant congenital naevus has a low risk of malignant transformation, the risk in sum over many melanocytes is significant, with a risk of melanoma estimated at ∼5–10% in giant congenital melanocytic naevi.
At the current time, congenital melanocytic naevi are considered a single entity and the benign neoplasm of the melanoma arising in congenital naevus pathway. However, rare giant congenital naevi harbour initiating mutations characteristic of Spitz naevi, including ALK fusion, BRAF fusion and HRAS mutation.
Some of these are indistinguishable from giant congenital melanocytic naevus with NRAS mutation based on histopathological features. Some HRAS mutated congenital naevi present as naevus spilus, a tan patch (which does not show any increase of melanocytes) within which melanocytic naevi develop. In one such case, the mutant HRAS allele was demonstrated within the tan patch with gain of the allele within the melanocytes of the naevi within it.
What accounts for the phenotypic differences between the subtypes of melanocytic naevi? One possibility is the function of the initiating oncogene without any relationship to the cell of origin. If this were the case, the differences in peak age of incidence and anatomical distribution would reflect the propensity for developing different types of mutations at different ages or different locations.
However, it may be that cell type or microenvironmental niche constrain the oncogenes that result in melanocytic naevi. We hypothesise that larger congenital melanocytic naevi result from the acquisition of an initiating oncogenic mutation occurring within a melanocyte precursor early during embryonic or fetal development in a melanocyte precursor destined to multiply significantly during development. Perhaps these melanocyte precursors are more susceptible to transformation by NRAS mutation. Small congenital naevi and common acquired naevi are indistinguishable on a histopathological and genetic basis and may reflect the same tumourigenic process that can occur in cells that are present both before and after birth.
Perhaps blue naevi arise from melanocytes that develop from the ventromedial pathway, and common acquired naevi develop from melanocytes that develop from the dorsolateral pathway. In adult skin, there are at least three known subtypes of melanocyte stem cells. One exists in the follicular bulge, one in the dermis, and another exists within the eccrine sweat gland.
The latter is proposed to be the cell of origin for acral lentiginous melanoma. Additionally, there appear to be distinct epidermal melanocytes with variable distribution by anatomical site.
Melanocytomas are melanocytic tumours with pathogenic mutations in addition to those found in melanocytic naevi but they are not fully transformed or malignant. Some of the genetic alterations found in melanocytomas lead to distinct cytomorphological features in addition to providing a growth or survival advantage that allows for clonal outgrowth. While some melanocytomas are recognisable by their histopathological features, there are many different pathogenic mutations that may contribute to melanoma pathogenesis and it is seems likely that some of these do not result in distinctive phenotypical changes in neoplastic melanocytes (i.e., TERT promoter mutations). Thus, there may be many more distinct types of melanocytomas than are currently recognised. The World Health Organization classification of skin tumours outlines low-grade and high-grade melanocytomas, accounting for multiple progression steps between naevus and melanoma, with high-grade melanocytomas requiring fewer additional hits for malignant transformation and thus a higher risk for progression to melanoma.
Many melanocytomas occur next to their precursor melanocytic naevus, and together they were previously considered ‘combined naevi’. However, in our current paradigm, the term combined naevus should be reserved for melanocytic naevi that have two distinct morphologies within them or two independently initiated melanocytic naevi that occur in close proximity.
Melanocytomas of the low-CSD pathway
Melanocytomas of the low-CSD pathway are the most commonly encountered, likely reflecting the prevalence of their common acquired naevus precursors.
Wnt-activated (deep penetrating) melanocytoma
Wnt-activated (deep penetrating) melanocytomas harbour mutations that result in constitutive activation of the Wnt/β-catenin pathway, most often point mutations in CTNNB1 that prevent degradation, but alternatively biallelic inactivation of APC, a major component of the CTNNB1 degradation complex, has been observed. This is in addition to naevus initiating mutations such as activating mutations of BRAF or MAP2K1. Wnt activated melanocytomas have been known by different names including deep penetrating naevus,
and when limited to the superficial portion of a common acquired naevus precursor, focal atypical epithelioid cell components within a melanocytic naevus.
The melanocytes of Wnt-activated melanocytoma have abundant granular pigmented cytoplasm and ovoid nuclei that are larger than those of common acquired naevi. The melanocytes are usually arrayed as elongated nests with many interspersed melanophages. A notable feature is that the melanocytes do not demonstrate maturation with depth within the dermis. Occasional dermal melanocytes in mitosis are not uncommonly observed. The differential diagnosis of Wnt-activated melanocytoma includes melanoma as well as pigmented Spitz naevus.
The melanocytes of common acquired naevi demonstrate a gradient pattern of β-catenin expression with increased expression in melanocytes within or near epithelium. In contrast, melanocytes in Wnt-activated melanocytoma have increased β-catenin expression and absence of a gradient of expression levels within the tumour (Fig. 3). Similar expression patterns are seen for CCND1 and LEF1 which are transcriptional targets of CTNNB1.
Fig. 3Wnt-activated melanocytoma with associated precursor melanocytic naevus. (A) At low power, the precursor melanocyte's naevus is in the upper left and nests of Wnt activated melanocytoma are present in the lower right. (B) β-catenin immunohistochemistry demonstrates a gradient pattern in the melanocytic naevus and strong uniform staining in the melanocytoma. (C) Higher power demonstrates the cytomorphology of the Wnt-activated melanocytoma with larger ovoid nuclei and expanded pigmented cytoplasm. (D) β-catenin expression is cytoplasmic and occasionally nuclear in the Wnt-activated melanocytoma. Some melanocytes of the melanocytic naevus show nuclear β-catenin expression.
The absence of maturation in melanocytes with constitutive Wnt activity suggests that maturation reflects a response of naevomelanocytes to a Wnt gradient in the skin.
Mutational activation of the Wnt pathway is associated with increased β-catenin expression and the melanocytes typically demonstrate cytoplasmic and occasional nuclear positivity. The melanocytes of common acquired naevi can show similar levels of β-catenin expression as Wnt-activated melanocytomas when they are near the epidermis and evaluation of CTNNB1 immunohistochemistry in superficial tumours can be challenging and often non-informative.
There are reports of cases diagnosed as deep penetrating or plexiform naevus (Wnt-activated melanocytoma) that have disseminated with lethal outcomes. A subset of the metastases have been genetically evaluated and determined to have additional pathogenic mutations in addition to naevus initiating and Wnt-activating mutations. Thus, these cases may reflect misdiagnosed melanomas or may be the result of additional progression in the residual tumour as some recurrences reportedly occurred many years after initial diagnosis.
BAP1-inactivated melanocytoma
BAP1-inactivated melanocytomas have distinct histopathological features and are typically predominantly intradermal, with large epithelioid melanocytes with well-defined cellular membranes, pleomorphism of cell and nuclear size, and eccentric placement of nuclei in some cells (Fig. 4).
There is often an associated lymphocytic infiltrate. Before our detailed understanding of the genetics of these tumours, they were often classified as halo Spitz naevi or atypical Spitz tumours. They harbour BRAF V600E or NRAS activating mutations at similar proportions as common acquired melanocytic naevi.
Fig. 4BAP1-inactivated melanocytic tumours. (A) BAP1-inactivated melanocytoma and precursor naevus. (B) BAP1 immunohistochemistry highlights the precursor melanocytic naevus that retains BAP1 nuclear expression. (C) BAP1-inactivated melanoma with crowded melanocytes arranged in sheets. (D) BAP1 expression is absent.
Inactivation of BAP1 typically occurs due to a truncating mutation in one allele and loss of the wild-type allele. Thus, the tumours often show loss of part or all of chromosome 3 by copy number analysis. BAP1 immunohistochemistry can be used to confirm inactivation of BAP1, though it is not 100% sensitive, as loss of function missense mutations in BAP1 result in the expression of non-functional BAP1 protein that is detectable.
BAP1-inactivated melanomas of the low-CSD pathway may have higher cellularity with sheets of cells and increased mitotic activity. They may show increased PRAME expression, loss of p16 by immunohistochemistry, additional pathogenic mutations and copy number aberrations.
BAP1-inactivating mutations can occur in the germline and predispose to melanoma (particularly uveal melanoma) and other cancers including mesothelioma and renal cell carcinoma.
The presence of a BAP1-inactivated melanocytoma in a young person or multiple tumours from one person increases the likelihood of a BAP1 germline alteration. Some authors suggest testing for germline BAP1 mutations in patients with two or more confirmed BAP1-inactivated tumours, or one tumour and a first or second degree family member with a BAP1-inactivated tumour.
Biallelic PRKAR1A-inactivation is present in a subset of tumours classified by histopathology as pigmented epithelioid melanocytoma. This entity was defined before the use of the term melanocytoma to describe melanocytic tumours with intermediate genetics. It appears that some tumours classified by their histopathological features as pigmented epithelioid melanocytoma may be considered by their genetic features as blue naevi as they only have a single genetic alteration of the Gαq pathway.
PRKAR1A-inactivated melanocytomas are composed of epithelioid melanocytes with abundant granular and pigmented cytoplasm (Fig. 5). In contrast to Wnt-activated melanocytomas, the melanocytes in PRKAR1A-inactivated melanocytomas are typically arranged in small nests or as single cells and there may be a predominance of melanophages. The melanocytes have larger nuclei with a more irregular chromatin pattern and prominent nucleoli. Absence of cytoplasmic PRKAR1A in the melanocytes by immunohistochemistry can provide confirmation of PRKAR1A-inactivation.
Fig. 5PRKAR1A-inactivated melanocytic tumours. (A) PRKAR1A-inactivated melanocytoma at low power. (B) Higher power view shows melanocytes with larger nuclei with prominent nucleoli with an increase in melanophages in the surrounding dermis. (C) PRKAR1A immunohistochemistry shows that epithelioid melanocytes lack expression while melanophages demonstrate cytoplasmic expression. (D) PRKAR1A-inactivated melanoma shows a nodular and asymmetric growth pattern. (E) High power view shows cytological atypia and mitotic activity. (F) The melanocytes lack PRKAR1A expression by immunohistochemistry.
Germline inactivating mutations of PRKAR1A cause Carney complex, and this syndrome should be considered in patients with PRKAR1A-inactivated melanocytoma, particularly children. Carney syndrome is associated with so called ‘epithelioid blue naevus’ but these are typically PRKAR1A-inactivated melanocytomas of the low-CSD pathway and not in the blue naevus pathway.
Secondary alterations are not necessarily limited to the low-CSD pathway
The additional pathogenic alterations in the low-CSD melanocytomas discussed above are not strictly limited to the low-CSD pathway, they can be observed in other contexts. BAP1 is an important tumour suppressor that is inactivated in a subset of uveal melanoma and melanoma arising in blue naevus. BAP1-inactivated melanocytomas of these pathways have not been described, exemplifying how the diagnostic significance of a specific genetic finding depends on its context.
A minority of the Wnt-activated melanocytomas in our initial report harboured Spitz naevus initiating mutations (HRAS mutation and BRAF fusion) and we have come across rare Wnt-activated Spitz melanocytomas since that time. These tumours display hybrid features of Spitz naevus and Wnt activated morphology.
Biallelic PRKAR1A-inactivation has also been reported in tumours of the blue or Spitz pathway.
Attempting to solve the pigmented epithelioid melanocytoma (PEM) conundrum: PRKAR1A inactivation can occur in different genetic backgrounds (common, blue, and Spitz subgroups) with variation in their clinicopathologic characteristics.
Genetic studies of Spitz tumours indicate that there are true Spitz melanocytomas, tumours with an initiating mutation of Spitz naevus with additional pathogenic alterations that provide a selective advantage resulting in clonal proliferation but that are not fully transformed or malignant. Previous studies have shown that Spitz tumours with homozygous CDKN2A deletion typically demonstrate benign behaviour but are more likely than those without homozygous CDKN2A deletion to lead to distant metastasis.
Comparative analysis of atypical spitz tumors with heterozygous versus homozygous 9p21 deletions for clinical outcomes, histomorphology, BRAF mutation, and p16 expression.
However, the aggressive Spitz tumours were not assessed for additional known pathogenic alterations, and it is possible that the tumours with aggressive biological behaviour harboured additional pathogenic alterations such as TERT promoter mutation, which was demonstrated to predict aggressive outcome in a small series of spitzoid tumours.
Additionally, identification of the initiating mutations was not possible at the time of the study, so some aggressive tumours may have been from other sublineages. Spitz naevi typically demonstrate expression of p16, a tumour suppressor encoded by CDKN2A, and complete absence of p16 expression is associated with homozygous deletion of CDKN2A and other methods of inactivation (i.e., nonsense point mutation and loss of the wild-type allele). In our experience, CDKN2A-inactivated melanocytomas are not uncommon in a consultation practice and we typically perform gene panel testing to exclude the presence of additional genetic alterations before making this diagnosis. There are also Spitz melanomas, tumours with frankly malignant histopathological features, a Spitz initiating oncogene and multiple additional pathogenic events including subchromosomal copy number losses and gains.
who importantly pointed out that Spitz tumours with atypical features typically had benign behaviour, even those with local lymph node metastasis, and are distinct from melanoma. At the time, initiating oncogenes in common acquired or Spitz naevi had not yet been identified and the concept of melanocytic tumours with genetic progression beyond naevus but without full malignant transformation had not been firmly established. Because of the difficulty in accurately distinguishing tumours of the common acquired and Spitz pathways and discriminating between naevus, melanocytoma and melanoma within the Spitz pathway without ancillary immunophenotypic or molecular techniques, series of atypical Spitz tumours in the literature include tumours of the common acquired pathway (with BRAF V600E mutation, many of which may be BAP1-inactivated tumours) as well as Spitz naevi and Spitz melanoma.
Thus, for the majority of its existence as an entity, the diagnosis atypical Spitz tumour was defined by purely histopathological features and carried some uncertainty, both as to the sublineage and its place on the spectrum from benign to malignant.
Thus, one cannot simply assign an equivalence between CDKN2A-inactivated Spitz melanocytoma and atypical Spitz tumour. With more understanding of the genetics of Spitz tumours, our diagnostic paradigm is shifting. For example, we now understand that Spitz naevi with ALK fusion are relatively uncommon, accounting for less than 20% of Spitz naevi and often contain larger melanocytes, larger nests, and attain a larger clinical size than other subtypes of Spitz naevus.
When the presence of an ALK fusion is confirmed by immunohistochemistry or other methods, tumours that would have been classified as atypical Spitz tumour in the pre-molecular era may be accepted as Spitz naevus. For a Spitz tumours with CDKN2A-inactivation, it is unclear if histopathological features alone can reliably distinguish them from Spitz melanoma, and ancillary genetic testing is often performed to screen for additional melanoma associated genetic findings. As comprehensive molecular assessment is not feasible or practical for all Spitz tumours, there is still a role for the diagnostic term ‘atypical Spitz tumour’ that includes uncertainty as to the classification as naevus, melanocytoma or melanoma.
Many tumours classified as atypical Spitz tumours travel to local lymph nodes but in the vast majority of cases, even when a clinically palpable lymph node deposit that disrupts lymph node architecture is present, distant metastasis does not occur.
Clinical management of melanocytomas and future directions
Melanocytomas for the most part have benign biological behaviour, with a few reported cases with distant metastases or lethal outcomes. It is unclear if these cases reflect (1) melanomas misdiagnosed as melanocytomas, tumours with genetic features of melanoma that are miscategorised by histopathological evaluation; (2) melanocytomas with a small focus of melanoma within them that was missed or perhaps not even sampled on sectioning; or (3) the result of additional progression in any residuum (either at the primary site or at a distant site as a result of benign metastasis). Theoretically, any residuum of a melanocytic tumour has some chance of progressing to melanoma and this chance increases with an increasing number of melanocytes and the state of progression of the melanocytes. The fewer mutations needed by each melanocyte for malignant transformation, the higher the risk of such transformation. Outcome studies to demonstrate how much risk reduction results from excision of melanocytomas is lacking and there is not an evidence-based guideline for when to recommend re-excision of melanocytomas.
While still not fully enumerated and defined, specific types of melanocytoma should be considered distinct diagnostic entities and melanocytoma should not be used to capture ambiguity when a case is diagnostically challenging. Thus, it would be incorrect to assign a diagnosis of melanocytoma to a case in which one is considering unusual naevus versus naevoid melanoma, for example. As we do not understand the full complexity of melanocytic tumours or their biological potential, there are still cases that present diagnostic uncertainty, even with sophisticated molecular analysis. Uncertainty in the diagnosis should be explicitly discussed in the pathology report.
Additional studies of melanocytic tumours, with identification of specific subtypes of melanocytomas of different pathways and outcomes studies to understand their risks of progression and key features that distinguish them from melanoma, will improve our ability to stratify melanocytic tumours.
Conflicts of interest and sources of funding
The author states that there are no conflicts of interest to disclose. No special funding was received by the author of this review.
References
Bataille V.
Bishop J.A.
Sasieni P.
et al.
Risk of cutaneous melanoma in relation to the numbers, types and sites of naevi: a case-control study.
CYSLTR2-mutant cutaneous melanocytic neoplasms frequently simulate “pigmented epithelioid melanocytoma,” expanding the morphologic spectrum of blue tumors: a clinicopathologic study of 7 cases.
Attempting to solve the pigmented epithelioid melanocytoma (PEM) conundrum: PRKAR1A inactivation can occur in different genetic backgrounds (common, blue, and Spitz subgroups) with variation in their clinicopathologic characteristics.
Comparative analysis of atypical spitz tumors with heterozygous versus homozygous 9p21 deletions for clinical outcomes, histomorphology, BRAF mutation, and p16 expression.
00679-1&id=gr1.jpg){kind=link}
00679-1&id=gr2.jpg){kind=link}
00679-1&id=gr3.jpg){kind=link}
00679-1&id=gr4.jpg){kind=link}
00679-1&id=gr5.jpg){kind=link}