Melanoma in children

INTRODUCTION — Pediatric melanoma, usually defined as melanoma occurring in patients younger than 20 years, is rare, accounting for approximately 1 percent of all melanomas [1,2]. Because of its rarity, the biology and clinical behavior, as well as the histopathologic features of pediatric melanoma, are not well characterized. The diagnosis is often extremely difficult to establish, especially in prepubertal children in whom melanoma may present as a nonspecific, amelanotic lesion with a morphology resembling a benign growth, resulting in frequent misdiagnosis, delays in diagnosis, thicker tumors, and delayed treatment [3].

This topic will discuss the clinical presentation, diagnosis, and management of melanoma in children. Melanoma in adults is discussed separately. Spitz nevi and atypical Spitz tumors are also discussed separately.

●(See "Melanoma: Clinical features and diagnosis".)

●(See "Pathologic characteristics of melanoma".)

●(See "Surgical management of primary cutaneous melanoma or melanoma at other unusual sites".)

●(See "Spitz nevus and atypical Spitz tumors".)

TERMINOLOGY — The term "juvenile melanoma" is only of historical relevance. Introduced by Sophie Spitz in 1948, it describes childhood melanocytic lesions with histologic features of melanomas but less aggressive clinical behavior compared with adult melanomas [4]. These lesions remain distinct from banal nevi and conventional melanoma and are now collectively regarded as Spitz tumors, a disease spectrum including benign Spitz nevi, spitzoid melanomas, and variants with uncertain malignant potential. These lesions are discussed in detail separately. (See "Spitz nevus and atypical Spitz tumors".)

EPIDEMIOLOGY

Incidence — Melanoma is rare in individuals younger than 20 years, with an estimated annual incidence rate of nine per million in those aged 15 to 19 years, according to the Surveillance, Epidemiology, and End Results (SEER) Cancer Statistics Review 1975-2014 [5]. Melanoma is even rarer in younger children, with estimated annual incidence rates of one, two, and three per million in the age groups 1 to 4, 5 to 9, and 10 to 14, respectively [5]. Lower incidence rates of pediatric melanoma have been reported among Hispanic and American Indian children <19 years (2.1 and 3.3 per million individuals, respectively) [6]. Due to the low incidence and lack of comprehensive registries, the precise incidence of clinical subtypes of pediatric melanoma, such as congenital and infantile melanoma, melanoma associated with congenital nevi, or spitzoid melanoma, is unknown.

Progressive increases in melanoma incidence in children have been reported in the United States over the last few decades, with approximately 90 percent of cases occurring among children older than 10 years [1,7]. However, one study utilizing SEER data from 2000 to 2010 has reported the first decrease in age-adjusted incidence rates of melanoma in children and adolescents in the United States [8]. In this report, the decrease in incidence was most notable between 2003 and 2010 among adolescents aged 15 to 19 years. This finding is consistent with trends in Australia and Sweden, where reductions in pediatric melanoma incidence have also been reported [9,10]. This decrease may be related to improved awareness regarding melanoma, changes in sun protection practices, and restricting indoor tanning access for minors. However, it is also possible that the decreased incidence of pediatric melanoma may be due, in part, to the improved histopathologic and molecular classification of some benign lesions (eg, some Spitz nevi and proliferative nodules in giant congenital nevi) that were misdiagnosed as melanomas in the past. Ongoing data collection and analysis are necessary to determine the cause for this apparent decline in incidence and whether this trend will persist.

Risk factors — Risk factors for adult melanoma are also significant for pediatric patients. They include genetic factors (eg, family history of melanoma, light skin phenotype and propensity to sunburn, high nevus counts, congenital nevi, inherited deoxyribonucleic acid [DNA] repair defects), environmental factors (eg, excessive exposure to sunlight, history of sunburns, indoor tanning), and iatrogenic or acquired immunosuppression. In a cohort of 70 pediatric patients with melanoma, 40 percent had high nevus counts, 27 percent had a positive family history of melanoma, and 25 percent reported a history of sunburn [11-13].

Variants of the melanocortin-1 receptor (MC1R) gene, a key regulator of skin pigmentation, are likely genetic risk factors for melanoma in children and adolescents.

●In a multicenter, case-control study including 233 children/adolescents (≤20 years) and 932 adults with melanoma and 932 controls, children and adolescents with melanoma had higher frequencies of any MC1R variants, r variants, V60L variants, and D294H variants than adults with melanoma and unaffected controls [14].

●In an Italian cohort of 123 patients younger than 21 years with histologically confirmed melanoma (116 conventional melanomas, 4 spitzoid, and 3 of other type), MC1R pathogenic variants were found in 67 percent of patients, whereas pathogenic variants affecting CDKN2A were extremely rare in sporadic cases (2 percent) [15].

Other risk factors include hereditary retinoblastoma and Werner syndrome [12,16]. (See "Retinoblastoma: Clinical presentation, evaluation, and diagnosis" and "Werner syndrome".)

Melanoma arising in congenital nevi — A systematic review of melanoma in patients with congenital melanocytic nevi of any size found a mean age at diagnosis of 15.5 years (median 7 years) and estimated a 465-fold higher risk of developing melanoma in childhood or adolescence for patients with congenital nevi compared with the general population [17]. It should be underscored that while most of these melanomas developed within large congenital nevi (67 percent), approximately 8 percent of primary melanomas developed in extracutaneous sites, including the central nervous system in patients with neurocutaneous melanocytosis. (See "Congenital melanocytic nevi", section on 'Neurocutaneous melanosis'.)

The risk of melanoma is not evenly distributed across all congenital nevi since the magnitude of risk correlates with congenital nevus size and number. The lifetime risk for the development of melanoma within a small congenital nevus (<1.5 cm adult diameter) to medium-sized congenital nevus (1.5 to 20 cm adult diameter) is less than 1 percent [18]. Children with large congenital nevi (>20 cm adult diameter) have a slightly increased risk of developing melanoma, estimated to be between 2 and 10 percent [19]. Patients with giant congenital nevi (>40 cm adult diameter) and multiple satellites (multiple small congenital nevi) represent the highest risk group, with a lifetime risk as high as 12 percent for developing cutaneous or extracutaneous melanoma [19-22]. (See "Congenital melanocytic nevi".)

In a 2013 meta-analysis of 14 studies including 2578 patients with large congenital melanocytic nevi, melanoma developed in 51 patients (2 percent) at a mean age of 13 years (range 0 to 58 years), with a mortality rate of 55 percent [19]. Most melanomas developed in giant congenital nevi or in large congenital nevi associated with multiple satellite nevi [19].

Melanomas developing in patients with congenital melanocytic nevi may arise in the skin, the central nervous system, or in other extracutaneous locations. In an analysis of a United Kingdom cohort of 448 children with congenital melanocytic nevi observed between 1988 and 2016, 12 children (2.7 percent) developed a melanoma [23]. Of these 12 primary melanomas, seven were primary melanomas of the central nervous system, two were primary cutaneous, one was primarily lymph nodal, and two were melanomas with unknown primary. The authors found that an abnormal screening magnetic resonance imaging (MRI) of the central nervous system in the first year of life in these patients was associated with a higher melanoma risk compared with a normal screening MRI (12 versus 2 percent, respectively). (See "Congenital melanocytic nevi", section on 'Neurocutaneous melanosis'.)

Spitzoid melanoma — Spitzoid melanomas are predominantly seen in preadolescent children [24-26]. In clinical practice, the choice of treatment is made based upon the lesion size and type (eg, homogeneous versus nonhomogeneous), presence and grade of histopathologic dysplasia, patient's preference, and clinical experience [27]. These lesions have histologic features of Spitz tumors; those with characteristic genetic alterations of Spitz tumors, such as activating HRAS mutations or kinase fusions, are termed "Spitz melanomas" [28,29].

Spitzoid melanoma in children has clinicopathologic features, biologic behavior, and a prognosis distinct from adult-type spitzoid melanoma.

●A multisite investigation spanning over two decades in 12 large academic institutions did not identify any fatal cases of spitzoid melanoma in prepubertal children [30].

●Similarly, there were no fatal outcomes among the Spitz melanomas and atypical Spitz tumors reported in the 2022 prospective registry Molecular Analysis of Childhood Melanocytic Tumors, which integrates molecular analysis of pediatric melanomas [31].

The clinical features, diagnosis, and management of atypical Spitz tumors are discussed separately. (See "Spitz nevus and atypical Spitz tumors", section on 'Spitz melanoma'.)

Congenital and infantile melanoma — Congenital and infantile melanoma, defined as melanoma present at birth or developing in the first year of life, is exceedingly rare, with only a few cases reported in the literature [32]. In a retrospective review of 87 women with placental or fetal metastases, 27 cases were attributed to melanoma [33]. The fetus was affected in six cases, and five of six infants succumbed to disease. In one report of transplacental metastasis of terminal maternal melanoma, the infant experienced spontaneous regression of all lesions and was disease free at the time of the report publication [34].

There are no long-term follow-up data on unaffected children born to mothers with metastatic melanoma to exclude the possibility of a delayed presentation of the disease. However, neonates who do not present melanoma metastases at birth should be considered at high risk and undergo close clinical monitoring. Some experts recommend repeated clinical evaluation during the first year of life and even beyond. Pathologic examination of the placenta for melanoma is also warranted in these cases.

PATHOGENESIS — Pediatric melanoma is considered to be biologically distinct from adult melanoma due to several distinctive features: greater thickness at presentation, higher frequency of amelanotic lesions, greater rate of sentinel node positivity, and overall less aggressive clinical course [3]. Whether the greater thickness at presentation is due to delayed diagnosis or more rapid growth in children has not been determined.

Congenital melanoma can arise de novo, in association with a congenital nevus, or from transplacental metastases from metastatic maternal melanoma. The pathophysiology of transplacental spread of melanoma is unclear. Factors involved may include the high vascularity of the placenta, placental production of angiogenic and growth factors, and impaired fetal immune response.

Molecular testing has provided insights in the molecular pathogenesis of pediatric melanoma. The results of a study of whole-genome or whole-exome sequencing and targeted sequencing of 23 pediatric melanoma samples, including 15 conventional (superficial spreading and nodular) melanomas, 3 melanomas associated with congenital nevi, and 5 spitzoid melanomas, are summarized below [35]:

●Conventional melanomas (ie, superficial spreading and nodular) from 15 adolescent patients aged 10 to 20 years, 10 of whom died of disease, were histologically similar to adult melanoma and showed a high burden of somatic single nucleotide mutations, of which over 80 percent were consistent with ultraviolet radiation (UVR) damage. The activating BRAF V600 mutation was found in 13 of 15 patients (87 percent) and TERT promoter mutations in 12 of 13 patients. Alterations (deletion or inactivating mutation) were common in tumor suppressor genes CDKN2A (mutation in 15 percent and biallelic deletion in 21 percent) and PTEN (mutation in 23 percent and biallelic deletion in 14 percent).

●All three melanomas associated with congenital nevi had NRAS (but not BRAF or TERT promoter) mutations, which is not surprising, based on the known NRAS mutant status of most large and giant congenital nevi [36]. Although the size of the three congenital nevi was not reported, their mutation profile and the fact that all children were <5 years of age and had a fatal outcome suggest that the congenital nevi were likely large lesions.

●Two of the five spitzoid melanomas (from three preadolescent children, one adolescent, and one patient of borderline age, of whom one succumbed to disease) demonstrated kinase fusions but not BRAF/NRAS mutations. In other reports, kinase fusions that include MAP3K8, ALK, NTRK1, NTRK3, MET, RET, ROS1, and BRAF have been found with variable frequency. Of note, kinase fusions have also been identified in a significant number of banal Spitz nevi and atypical Spitz tumors [37-40].

Similar results were seen in another whole genome sequencing study of tumor samples from 50 adolescent and young adult patients with melanoma (superficial spreading in 43 of 50 patients) [41]. Patients ranged from age 10 to 30, with a median age of 20 years, and had stage III/IV melanoma. Melanomas showed a high mutation burden, with BRAF mutations being present in 96 percent of cases, 82 percent of which were p.V600E. TERT was the second most frequently mutated gene, with promoter hotspot mutations present in 80 percent of cases, followed by PTEN (36 percent), CDKN2A (24 percent), TP53 (24 percent), and RBI (12 percent). Ninety percent of the mutations found were UVR signature mutations.

These findings suggest that conventional melanoma in adolescents is similar to BRAF-mutant adult melanoma and confirm that UVR is likely a major mutagenic factor in melanomas occurring in adolescence. Melanomas associated with congenital nevi and spitzoid melanomas are instead distinct entities, with a different genetic profile, clinical course, and prognosis. (See "Spitz nevus and atypical Spitz tumors", section on 'Spitz melanoma' and "Spitz nevus and atypical Spitz tumors", section on 'Mutational analysis'.)

CLINICAL FEATURES — The clinical presentation of melanoma in children often does not conform to the conventional ABCDE criteria (asymmetry, border irregularities, color variegation, diameter ≥6 mm, evolving lesion) (picture 4B). Especially in prepubertal children, melanoma may present as a symmetric, pigmented lesion with regular borders simulating a banal nevus; as a pink or reddish, symmetric, dome-shaped papule simulating a Spitz nevus (picture 1); or as an amelanotic, nodular lesion simulating a pyogenic granuloma (picture 2) or common wart. Lesions may be uniform in color and smaller than 6 mm. Most importantly, the "evolving lesion" criterion of the ABCDE rule is not as helpful in children and adolescents, in whom the onset of new nevi as well as lesion evolution (ie, increase in diameter and elevation) are the rule.

In a series of 70 children with melanoma, 60 percent of lesions in children <11 years and 40 percent of those in adolescents lacked the conventional ABCDE criteria [11]. Lesions were more likely to be small and amelanotic in young children and frequently uniform in color in adolescents, with lesion evolution being a nearly universal finding.

In a study of 52 pediatric melanoma cases [42], approximately 52 percent of lesions manifested the conventional ABCD criteria. However, the clinical morphology differed significantly between cases of spitzoid melanoma and nonspitzoid melanomas. While less than 25 percent of all cases of pediatric melanoma (spitzoid and nonspitzoid) fulfilled the modified ABCD criteria (amelanotic, bleeding/bump, color uniformity, de novo/any diameter), 40 percent of the spitzoid and only 13.5 percent of nonspitzoid melanoma cases revealed the modified ABCD features (see 'Physical examination' below). In contrast, almost 60 percent of nonspitzoid melanoma cases and only 33 percent of spitzoid melanoma cases had the conventional ABCD features.

In another series of pediatric melanomas, the prebiopsy clinical diagnosis was a benign lesion in 40 of 68 cases (59 percent) and atypical nevus or melanoma in 28 [43]. Forty-five percent of the tumors were amelanotic, and the most common presentation sign was a change in size and color. Pediatric melanomas that mimicked benign-appearing lesions were more likely to have a Breslow thickness >1 mm (odds ratio 5.48, 95% CI 1.66-20.12).

The atypical appearance of pediatric melanomas can cause a diagnostic challenge, leading to a delay in diagnosis and thicker lesions (picture 3). In one study of 70 pediatric melanoma cases, the time from detection to diagnosis was ≥6 months in 82 percent and >12 months in 62 percent of patients, and, in keeping with the diagnostic delay, more than 90 percent of childhood cases and 50 percent of adolescent cases were diagnosed with stage IIa disease or higher [11]. (See "Tumor, node, metastasis (TNM) staging system and other prognostic factors in cutaneous melanoma".)

DERMOSCOPIC FEATURES — Dermoscopy has emerged as a useful tool to differentiate the morphology of various pediatric skin lesions that resemble melanoma. Evaluation using a dermatoscope is painless and well tolerated by children in the clinic [44].

Ten melanoma-specific dermoscopic structures have been identified: atypical network; negative network; streaks; shiny, white structures; atypical dots and globules; irregular blotch; blue-white veil; regression structures; peripheral, brown, structureless areas; and atypical vessels. The presence of at least 1 of these 10 structures should raise suspicion of melanoma. Since melanomas in children are often amelanotic, the examination of vascular structures and presence of shiny, white structures is of primary importance (table 1). Any nodular, pink lesion with dotted vessels; milky, red globules; serpentine vessels; shiny, white structures seen with polarized light dermoscopy; or negative (inverse) network should be considered suspicious for melanoma. (See "Dermoscopic evaluation of skin lesions", section on 'Criteria for melanoma'.)

However, it is important to acknowledge that many of the dermoscopic features observed in melanoma can also be seen in Spitz nevi, making the clinical differential diagnosis between Spitz nevi and melanoma challenging.

An analysis of 49 dermoscopic images from 52 pediatric melanoma cases (mean age, range 2 to 20) found that the dermoscopic patterns most frequently associated with nonspitzoid melanomas were the multicomponent pattern and nevus-like pattern (58 and 25 percent, respectively) [42]. The multicomponent pattern is the "classic" dermoscopic melanoma pattern encountered in superficial spreading-type melanoma, with most cases presenting with a disorganized distribution of dots/globules, atypical network, blue-white veil, vessels, and negative network. Spitzoid melanomas showed in most cases a "vascular, pink, Spitz-like pattern" with atypical vessels and shiny, white structures or a "pigmented, Reed-like" pattern with black, blue-gray, and dark brown colors, peripheral streaks, and dark blotches. In this study, all cases of melanoma revealed at least one of the melanoma-specific structures.

However, even in the absence of melanoma-specific structures and atypical vascular morphology, all lesions that do not manifest an unequivocal benign pattern and lesions that appear atypical should be considered for histopathologic examination. A useful approach may also include measurement, photography, and close clinical follow-up for evaluation of lesion evolution.

DIAGNOSIS — The diagnosis of melanoma in children is challenging due to its rarity and often banal clinical presentation. It requires a high index of suspicion based upon history, clinical, and dermoscopic findings, as well as a low threshold for the decision to perform a biopsy. The definitive diagnosis is reliant upon histopathologic evaluation.

History — A prominent feature in pediatric melanoma is lesion evolution over time. However, melanoma evolution in children may be misinterpreted as the expected and natural evolution of benign, pigmented lesions during childhood and adolescence. Patients or parents/caregivers may report a history of bleeding, particularly for ulcerated lesions. Additional complaints may include pain, itch, or discomfort, all of which can prompt evaluation in the clinic. Moreover, patients and parents/caregivers should be queried for recent treatments (eg, wart removal preparations) or trauma, as melanoma may have been misdiagnosed and treated as a benign entity.

Physical examination — In older adolescents and young adults, most melanomas are of the nonspitzoid variety, and most of these will manifest the conventional "ABCDE criteria" (asymmetry, border irregularity, color variegation, large diameter, evolving lesion). In contrast, melanomas in prepubertal and pubertal children are often of the spitzoid melanoma type, and many of these will lack the conventional ABCD features. In addition, the "evolving lesion" criterion may not be helpful at an age in which the new onset/evolution of common nevi is normal. Additional sets of pediatric criteria (the pediatric ABCD and the CUP criteria) have therefore been proposed to be used in combination with the conventional ABCDE rule [11,45]:

●A = Amelanotic

●B = Bleeding, bump

●C = Color uniformity

●D = De novo, any diameter

●C = Color is pink/red, changing

●U = Ulceration, upward thickening

●P = Pyogenic granuloma-like lesions, pop-up of new lesions

While these additional criteria may increase sensitivity for melanoma detection, they are not specific for melanoma. In a study of 52 pediatric melanoma cases, of which 15 were spitzoid and 37 nonspitzoid melanomas, only 40 percent of spitzoid melanomas and 13 percent of nonspitzoid melanomas fulfilled the proposed pediatric ABCD criteria [42].

When assessing a new pink/red papule in a child, it is important to also consider associated features, such as bleeding, history of trauma, itch or pain, growth or change over time, or similar lesions elsewhere, in the decision of whether to clinically monitor or biopsy a particular lesion.

If a lesion suspicious for melanoma is found on skin examination, palpation of the regional lymph nodes should be performed to evaluate for enlarged nodes. Additionally, evaluation of congenital nevi should include palpation for subcutaneous changes, especially in large congenital nevi since melanomas arising in large congenital nevi often develop below the dermo-epidermal junction.

Biopsy — Biopsy is warranted when there is clinical suspicion of melanoma. An excisional biopsy including the subcutaneous fat with a small (2 mm) rim of normal-appearing skin is the preferred biopsy method. While incisional biopsy, including deep shave biopsy, is acceptable (especially if the lesion is large and/or in a cosmetically sensitive area), superficial shave biopsies should be avoided since the full depth of the lesion is required to accurately determine the degree of maturation of the melanocytes, evaluate the base of the lesion, and obtain an accurate measure of tumor thickness.

Pathology — Not only are pediatric melanomas difficult to diagnose clinically, but they may also be challenging histopathologically. It can be difficult to distinguish a spitzoid melanoma from an atypical Spitz tumor because they have similar architectural and cytologic features, although the number and severity of atypical features are increased for melanomas (picture 1). For example, a large lesion diameter (especially >1 cm), increased mitotic rate (especially >6 per mm²), high cellular density, and ulceration raise concern for malignancy in Spitz tumors. Moreover, there are pigmented tumors of uncertain malignant potential, and there are challenging pigmented lesions for which expert dermatopathologists cannot reach consensus on a diagnosis [46]. (See "Spitz nevus and atypical Spitz tumors", section on 'Histopathology'.)

Data on the frequency of each of the histologic subtypes of pediatric melanoma have not collected systematically, but it is estimated that approximately 40 to 50 percent of pediatric melanomas are of the conventional subtype (adult melanoma-like), which shares similar morphologic features with adult cutaneous melanoma (ie, superficial spreading and nodular) [47]. The same criteria for the histologic diagnosis, classification, and stage grouping for adult melanoma are applied to the pediatric conventional subtype. The affected individuals typically are adolescents or postpubertal children. Lentigo maligna melanoma (the histologic subtype that typically arises on chronically sun-damaged skin) almost never occurs in children/adolescents, and acral lentiginous melanoma (melanoma on glabrous skin) is exceedingly rare in this population [48]. (See "Pathologic characteristics of melanoma".)

Among pediatric melanomas, childhood tumors may show more aggressive histopathologic features than adolescent tumors (picture 4A-B), despite a paradoxically better survival [11]. A retrospective review of published fatal pediatric melanomas reported that childhood melanomas (up to age 10 years) were thicker than adolescent melanomas, with a mean Breslow thickness of 8.5 mm compared with adolescent melanoma mean Breslow thickness of 3.7 mm [49]. In another cohort of 12 children and 20 adolescents with melanoma, the median Breslow thickness was 3.5 mm for lesions in children versus 1.5 mm in adolescents [24]. In another series of 25 fatal cases of pediatric melanoma, the median Breslow thickness was 2.2 mm, with a range of 0.9 to 36 mm [30].

Molecular tests — Genomic criteria (eg, multiple chromosomal copy number variations observed by comparative genomic hybridization [CGH] or fluorescence in situ hybridization [FISH]) and alterations in the TERT promoter [25,37] may provide support for the diagnosis of melanoma in atypical melanocytic proliferations. Laboratory diagnostic approaches, including FISH, CGH, and genomic sequencing, promise to further our understanding and diagnosis of these lesions, but studies have provided inconsistent results for pediatric tumors. Although histopathology continues to play an important role in the diagnosis of pediatric melanocytic lesions, integration of genomic findings improves the accurate pathologic classification of these lesions, which may ultimately affect the clinical management of these patients [31,50,51].

●Next-generation sequencing – Next-generation sequencing using melanoma-specific gene panels is increasingly used for the genomic characterization of melanoma, including presence or absence of BRAF mutations; TERT promoter mutations; and the presence of kinase fusions involving MAP3K8, RET, ALK, BRAF, NTRK, and ROS1 [52].

●Fluorescence in situ hybridization – The commercial fluorescence in situ hybridization (FISH) tests and those used in research settings utilize a variety of probes to identify copy number gains and losses in melanoma and Spitz tumors. Some of the probes that identify copy number alterations include chromosomes 6p25, 6q23, 11q13, 9p21, and 8q24 [53]. It is often helpful to first examine for 9p21 deletions. Since this region contains the CDKN2A gene, which encodes P16, immunohistochemistry staining for p16 can be performed to confirm expression and therefore lack of homozygous deletion. Complete absence of p16 expression can be due to homozygous deletion or inactivating mutations, or a combination of both.

●Comparative genomic hybridization – Comparative genomic hybridization (CGH) allows detection of amplifications and deletions of smaller regions of DNA along the lengths of all chromosomes. CGH analysis typically shows copy number alterations of whole chromosomes in proliferative nodules in congenital nevi, whereas copy number variations in regions or segments of chromosomes are observed in melanomas arising in association with giant congenital nevi [54].

Imaging — Imaging is typically reserved for evaluation of regional and distant disease spread for cases with deep tumor involvement. In adults, ultrasound has been proposed as an alternative to sampling the sentinel lymph node (SLN), although studies have provided inconsistent results. One study including 716 patients with melanoma participating in the Multicenter Selective Lymphadenectomy Trials-I and -II found that the sensitivity of routine preoperative high-resolution targeted ultrasonography to detect a positive node was low (24.3 percent, 95% CI 19.5-28.7 percent), suggesting that the routine use of this technique cannot be used reliably as an alternative to sentinel lymph node biopsy (SLNB) [55]. In contrast, some European centers have demonstrated value of routine high-resolution ultrasound examination of the regional node field in the detection of nodal recurrence among patients who did not have SLNB or had positive biopsy without completion lymph node dissection [56].

There are no standards for imaging pediatric melanoma, and in many cases, adult recommendations are followed. In one study, extensive pretreatment imaging with computerized tomography (CT), MRI, and positron emission tomography (PET) followed by clinical and intensive radiographic follow-up with CT and MRI suggests that children with spitzoid melanocytic lesions who lack TERT promoter mutations do not require extensive imaging at diagnosis or follow-up, given the very low risk of recurrence [57].

Brain metastases are best evaluated with MRI. Lung and liver metastases are best evaluated by CT. PET imaging is not routinely utilized, due to low rates of detection of metastatic disease and high false-positive rates associated with inflammation and presence of "brown fat" that is typical of younger patients.

DIFFERENTIAL DIAGNOSIS — The differential diagnosis of melanoma in children includes a variety of cutaneous lesions, including:

●Common or dysplastic nevi – These nevi present as pigmented papules and macules that may be characteristically round and symmetric (common nevi (picture 5)) or include clinically atypical features of asymmetry, border irregularity, color variegation, or large diameter (atypical nevi (picture 6)). Nevi are present in nearly all pediatric patients, and their number increases over the first three decades of life, with a peak occurring during childhood and adolescence. Histologically, these lesions lack criteria for melanoma. These lesions are expected to bleed only with trauma, and unexplained bleeding should be investigated further. (See "Acquired melanocytic nevi (moles)" and "Atypical (dysplastic) nevi".)

●Spitz nevi and atypical Spitz tumors – These pigmented (picture 7) or pink papules (picture 8) are more common among young pediatric patients. Classic Spitz nevi are typically monitored clinically without biopsy or intervention, while those with concerning features on clinical or dermoscopic examination should be biopsied for histopathologic evaluation [58]. (See "Spitz nevus and atypical Spitz tumors".)

●Blue nevi – Blue nevi are dark blue-gray papules (picture 9) and macules that may raise concern due to their characteristic dark color, corresponding to the presence of melanocytes deeper in the dermis. (See "Acquired melanocytic nevi (moles)", section on 'Blue nevi'.)

●Proliferative nodules – Proliferative nodules may appear as papules within larger congenital nevi and have atypical histology (picture 10) but reassuring genomic studies [59]. In a comparison of proliferative nodules and lethal melanomas arising in congenital nevi of children, one study reported that proliferative nodules more frequently arise from dermis, occur in multiple sites instead of one single focus, and have infrequent ulceration (3 of 22 proliferative nodules versus 2 of 2 lethal melanomas) [60]. These authors also reported that histologically proliferative nodules had lower mitotic rates (<5/mm²) than that generally seen in melanomas. Proliferative nodules more often have whole chromosomal aberrations, whereas lethal melanomas more often have copy number aberrations of 6p25 without gains of the long arm of chromosome 6 [60]. (See "Congenital melanocytic nevi".)

●Pyogenic granulomas – Pyogenic granuloma, also called lobular capillary hemangioma, is a benign vascular tumor that presents as a growing pink or red papule that can bleed, simulating an amelanotic melanoma. Histology shows a distinct vascular proliferation. (See "Pyogenic granuloma (lobular capillary hemangioma)".)

●Warts – Warts are viral papules that are often characterized by hyperkeratosis and/or visible pinpoint capillary vessels. They can arise on cutaneous or mucosal surfaces and within congenital melanocytic nevi. The presence of multiple lesions and characteristic clinical features suggests the diagnosis. Histopathology can confirm the diagnosis. (See "Cutaneous warts (common, plantar, and flat warts)".)

MANAGEMENT — The staging system and management of pediatric melanoma is the same as for adult melanoma. (See "Tumor, node, metastasis (TNM) staging system and other prognostic factors in cutaneous melanoma".)

Surgery — The first-line therapy for pediatric melanoma is wide local excision to the deep fascia. The margin size depends on the thickness of melanoma and on whether the anatomic location permits removal with the designated/recommended margins, which are the same as for melanoma in adults. Margins of excision specific to pediatric patients with melanoma have not been defined. While in most cases adult guidelines are followed, in younger children consideration can be given to narrower margins of excision than those used in adults. Thin (<1 mm) pediatric melanomas typically do not require further treatment beyond wide excision, though the children will enter a monitoring phase with regular skin and lymph node examinations. (See "Surgical management of primary cutaneous melanoma or melanoma at other unusual sites".)

Sentinel lymph node biopsy — Lymphatic mapping and sentinel lymph node biopsy (SLNB) is controversial in pediatric patients. This is a procedure that allows identification of lymph node micrometastases and pathologic staging of melanoma involving the regional lymph node basin. In adults, randomized trials have demonstrated the utility of this method in prognostication and risk stratification [61]. (See "Evaluation and management of regional nodes in primary cutaneous melanoma", section on 'SLNB timing and technique'.)

The controversy in the use of SLNB in pediatric patients stems from data indicating that younger patients, who in most cases have spitzoid melanomas (or indeterminate or borderline tumors), have higher rates of positive sentinel lymph node (SLN) compared with adult patients and that positivity does not predict an aggressive clinical course [62].

●One study reporting on a series of 109 patients <18 years found that SLN positivity did not predict prognosis or outcome in pediatric melanoma [63]. In this study, 57 patients had a SLNB, with a positive result in 52 percent of patients under age 10 years and 26 percent of patients aged 10 to 17 years. In addition to younger age, tumor thickness >2 mm was also associated with SLN positivity. In this cohort, none of the patients under age 10 years died from melanoma.

●Tumor thickness was associated with SLN positivity in a separate series of 126 patients <21 years with melanoma [64]. Of 62 patients who underwent SLNB, 29 percent were positive, and positivity was correlated with tumor thickness. The five-year melanoma-specific survival was 78 percent for patients with positive SLN compared with 97 percent for those with negative SLN. These rates are similar to those reported for the corresponding stages in the American Joint Committee on Cancer (AJCC) seventh edition staging system, which predominantly included adults [65].

●In a systematic review of 541 patients with atypical Spitz tumors who were followed for a mean of 59.4 months, 56 percent underwent a sentinel node biopsy and 39 percent had a positive node. Eighty-two percent of these patients had a completion node dissection, and 19 percent had additional positive nodes. Despite this, 99 percent of the patients with atypical Spitz tumors and a positive sentinel node who were treated with surgery alone were alive at a median follow-up of 59 months [66].

Thus, the decision to pursue SLNB should be made for the individual patient, taking into consideration the patient's age, histopathologic and molecular diagnosis, anatomic location, lymph drainage, and risk of false negativity and false positivity. False positivity may be due to the presence of nevus cells in the lymph node [67]. For pediatric patients with a diagnosis of melanoma that has qualifying features per adult data, SLNB is recommended. Tumors that are indeterminate or spitzoid melanomas in children diagnosed before age 11 require additional consideration.

Although completion dissection of all involved nodal basins previously was considered the standard treatment for patients with a positive SLNB, this approach is no longer recommended. The Multicenter Selective Lymphadenectomy Trial-II (MSLT-II) comparing immediate completion lymph node dissection with observation followed by lymph node dissection in the event of a regional lymph node recurrence did not demonstrate any improvement in melanoma-specific survival or distant metastasis-free survival for patients who underwent completion lymph node dissection [68].

If the SLNB is positive, surveillance of the nodal basins with clinical observation and serial ultrasound imaging is appropriate. (See "Evaluation and management of regional nodes in primary cutaneous melanoma", section on 'Subsequent management'.)

If nonsentinel nodes become enlarged and are histologically proven to be involved, therapeutic lymph node dissection may be warranted to prevent the development of bulky adenopathy, although the procedure has no impact on survival. Lymph node dissection confers significant morbidity and complication risk. In a series of 125 pediatric patients, complications occurred in 52 percent of inguinal dissections and 17 percent of axillary dissections [69]. Lymphedema occurred in 2 percent of patients who had undergone SLNB alone compared with 20 percent when completion dissection was performed. (See "Evaluation and management of regional nodes in primary cutaneous melanoma".)

Adjuvant therapy — As in adults, adjuvant therapy has generally been used for patients with localized disease at increased risk for disease dissemination due to regional lymph node disease or a high-risk primary tumor. Data on the efficacy of adjuvant therapy in children and adolescents are scant [70]. Adjuvant targeted therapies against MAPK pathway have proven to be highly effective in high-risk resected adult melanoma [71] and may be considered for patients whose tumors demonstrate relevant genomic alterations, such as a BRAF V600E mutation [72,73]. (See "Systemic treatment of metastatic melanoma with BRAF and other molecular alterations".)

Immunotherapy with checkpoint inhibitors, such as pembrolizumab and nivolumab, have produced improvements in relapse survival in adult patients with high-risk resected disease, and these therapies have been studied in pediatric patients, providing a safe and rationale alternative to adjuvant therapies in selected patients with pediatric melanoma [74-78]. (See "Systemic treatment of metastatic melanoma lacking a BRAF mutation".)

Metastatic melanoma — Our approach for pediatric and adolescent patients is to enroll patients in formal clinical protocols whenever possible or to pattern therapy after that in adults. (See "Overview of the management of advanced cutaneous melanoma", section on 'Pediatric and adolescent patients' and "Systemic treatment of metastatic melanoma lacking a BRAF mutation" and "Systemic treatment of metastatic melanoma with BRAF and other molecular alterations".)

PREVENTION — Although interventions to reduce individual exposure to ultraviolet (UV) light are universally recommended for the primary prevention of melanoma, their impact on melanoma risk among patients with giant congenital nevi or prepubertal cases is uncertain. Nonetheless, since childhood and adolescence are critical times to establish and practice sun protection behaviors that can modify future melanoma risk, patients, parents, and caregivers should be educated to adopt "safe sun" behaviors.

Based on evidence of benefit of behavioral counseling interventions on children's parents, adolescents, and young adults, the United States Preventive Services Task Force (USPSTF) recommends counseling young adults, adolescents, children, and parents of young children about minimizing exposure to UV radiation for persons aged 6 months to 24 years with fair skin types to reduce their risk of skin cancer [79,80].

Sun protection measures include minimizing outdoor exposure between 10 AM and 2 PM; avoiding indoor tanning; wearing protective clothing and large hats; and using broad-spectrum sunscreens, including physical blockers such as zinc oxide and titanium dioxide, which are not absorbed by the skin, and reapplying sunscreen every few hours, as needed, to prevent sunburn. (See "Primary prevention of melanoma".)

PROGNOSIS — As for adult melanoma, the most important predictors of prognosis for non-Spitz pediatric melanoma are tumor thickness, presence of ulceration, and stage at diagnosis [3,81,82].

●The analysis of an international registry of 365 patients aged 1 to 21 years with invasive melanoma reported a 10-year overall survival rate of 81 percent; the survival rate was 100 percent for patients up to age 10 years, 70 percent for patients aged 10 to 15 years, and 80 percent for patients aged 15 to 20 years [3]. Tumor thickness at diagnosis was strongly associated with the 10-year survival, with 97 percent survival among patients with lesions <1 mm and 80 percent among patients with lesions >4 mm.

●In an international cohort of 62 children and 452 adolescents treated for invasive melanoma between 2000 and 2014, the 10-year recurrence-free survival was 91.5 percent (95% CI 82.4-100) in children and 86.4 percent (95% CI 82.7-90.3) in adolescents [83]. Among adolescents, head and neck location, Breslow thickness >4 mm, and ulceration were associated with poorer survival.

●An Italian prospective registry study of 54 patients younger than 18 years with melanoma found an overall five-year survival of 85 percent. Survival was correlated with tumor stage and ulceration [81].

●In another study of 100 patients aged 11 to 22 years with conventional (superficial spreading and nodular) melanomas followed up for a median of 7.6 years, 16 patients developed a recurrence, and 8 eventually died of melanoma [84]. The 20-year overall survival, recurrence rate, and melanoma-specific mortality were 77, 34, and 20 percent, respectively. Of 21 patients who underwent sentinel lymph node biopsy, all were negative, and none developed a recurrence.

●In a systematic review and individual patient meta-analysis that included 1002 children with melanoma, 108 patients (median age 8 years [range 4 to 12]) had spitzoid/spindle cell melanomas with a median Breslow thickness of 3.3 mm [85]. Seventy-three (67 percent) developed metastasis (73 percent nodal), and 12 percent died of melanoma. In contrast, the mortality rates for superficial spreading (n = 210) and nodular melanoma (n = 139) were 15 and 45 percent, respectively.

Fatal outcome from pediatric melanoma can occur with a variety of presenting types and is more common among patients diagnosed with melanoma during adolescence than in childhood [30].

FOLLOW-UP — Pediatric melanoma follow-up protocols align with the adult recommendations for regular total-body skin examination and lymph node surveillance to evaluate for recurrence or metastasis. Surveillance with ionizing radiation (CT or positron emission tomography [PET] scans) is generally pursued less aggressively than in adults due to concerns about the cumulative impact of radiation exposure [86]. Typically, physical examination with special attention to regional recurrences is performed every three months for one year, then every six months for five years, followed by annual exams thereafter. (See "Staging work-up and surveillance of cutaneous melanoma".)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Melanoma screening, prevention, diagnosis, and management".)

SUMMARY AND RECOMMENDATIONS

●Epidemiology and risk factors – Pediatric melanoma is rare, representing only 1 to 4 percent of all melanomas, with an annual incidence estimated at six per million individuals. Children and adults share similar risk factors for developing melanoma, including genetic factors (eg, family history of melanoma, light phenotype, high nevus counts, congenital nevi, high-risk germline mutations) and environmental factors (eg, excessive ultraviolet exposure from direct sunlight or indoor tanning, history of sunburns, immunosuppression). (See 'Epidemiology' above.)

●Clinical presentation – The clinical presentation of melanoma in children often does not conform to the conventional ABCDE criteria (asymmetry, border irregularity, color variegation, large diameter, and evolution). Especially in prepubertal children, melanoma may simulate a banal nevus, pyogenic granuloma, or common wart. (See 'Clinical features' above.)

●Diagnosis – The diagnosis of melanoma in children is challenging due to its rarity and often atypical clinical presentation. It requires a high index of suspicion based upon history, clinical, and dermoscopic findings. When melanoma is suspected, the preferred biopsy is an excisional biopsy deep to the subcutaneous fat with 2 mm rim of normal-appearing skin. This provides an adequate specimen for histopathologic examination. (See 'Diagnosis' above.)

●Staging – The staging system of pediatric melanoma is the same as for adult melanoma. The decision to pursue sentinel lymph node biopsy should be made for the individual patient, taking into consideration the patient's age, clinical context, subtype of melanoma and certainty in its diagnosis, anatomic location, lymph drainage, prognostic information desired, and risk of false-positive and false-negative results. (See 'Sentinel lymph node biopsy' above and "Staging work-up and surveillance of cutaneous melanoma".)

●Management – The first-line therapy for pediatric melanoma is wide local excision to the deep fascia. The margin size depends on the thickness of melanoma and on whether the anatomic location permits removal with the designated margins. Thin (<1 mm) pediatric melanomas typically do not require further treatment beyond wide excision. Once excised, children will enter a monitoring phase with regular skin and lymph node examinations. (See 'Management' above and "Surgical management of primary cutaneous melanoma or melanoma at other unusual sites".)