Acquired melanocytic nevi manifest initially in early childhood and increase in number over the first three decades of life. These lesions evolve following two pathways initially proposed by Stegmaier: formation of soft, flesh-colored intradermal nevi that either (1) develop into pedunculated skin tags that eventually fall off or (2) gradually fade and disappear by fibrosis and fatty infiltration. Newer nevi tend to be small and flat (junctional) and either develop a raised profile or disappear with time as a consequence of fibrosis. In one study, when the nevi on adolescents were followed over a 4-year period, there was a net increase of 50% in total number of nevi, despite complete regression in 15%. Another study demonstrated that 75% of children aged 11–14 years developed at least one new nevus and 28% had disappearance of at least one nevus over a 3-year period. ^, Patients with the highest baseline nevi count were the most likely to have concurrent development of new nevi and disappearance of one or more nevi. Additionally, patients with a globular pattern on dermoscopy were also more likely to develop new nevi.

On dermoscopy, most lesions in children and adolescents demonstrate a globular pattern, especially those on the upper trunk, head, and neck. Conversely, melanocytic nevi in adults most often demonstrate a reticular pattern, especially those on the upper and middle back. Histologically, these acquired melanocytic nevi are divided into subtypes based on the location of the nests of nevus cells, and this feature corresponds with certain clinical findings ( Table 67.1 ). Junctional nevi tend to be flat with brown/black pigmentation, and the nevus cells are located at the dermal–epidermal junction. When the nevus cells extend from the junction into the dermis, the lesion is described as a compound nevus. Clinically, this corresponds to a lesion that is slightly raised and pigmented brown/black. When the nevus cells migrate completely into the dermis, the lesion is an intradermal nevus, which is raised and typically not pigmented. In general, the deeper the nests of nevus cells, the more raised and less pigmented the lesion (i.e., dark flat lesions vs. raised tan lesions).

Environmental and genetic factors predispose to the development of acquired melanocytic nevi. Sun exposure, especially recurrent intermittent sunburns, is one of the primary environmental factors associated with nevi development. In a prospective study of nevi in a cohort of 432 U.S. children, painful sunburns and cumulative sun exposure, as measured by hours spent in the sun, were associated with an increased number of nevi on multivariate analysis. In addition, some studies suggest that sun protective behaviors such as wearing sunscreen and protective clothing may not prevent the development of melanocytic nevi. ^, This may be due to increased sun exposure among children who use sunscreen, inadequate sunscreen application and reapplication, and incomplete protection from UV radiation. Additionally, the relationship between sun exposure and nevi development may be modified by skin color, with prolonged cumulative sun exposure increasing the number of nevi among children with fair skin but not darker skin. In general, children with fair skin have more nevi than children with darker skin. However, individuals with darker skin types are more likely to have nevi on the palms and soles. Other risk factors for developing melanocytic nevi include lighter hair color, blue or green eyes, and male gender ( Box 67.1 ). ^{, , ,}

Several genes associated with nevi development have been identified, including those involved in pigmentation, telomere function and senescence, and cell-cycle regulation. ^, For instance, a particular single nucleotide polymorphism (SNP) in IRF4 is strongly associated with increased counts of flat nevi and decreased counts of raised nevi and is modulated by low-to-moderate levels of sun exposure. Additionally, BRAF ^V600E mutations, which can be associated with sun exposure, are present in 88% of acquired melanocytic nevi. Approximately 60% of adult melanomas harbor BRAF ^V600 mutations. ^, However, the risk of malignant transformation does not parallel the expression of BRAF mutations.

A large number of acquired melanocytic nevi and the development of atypical nevi are significant risk factors for the development of melanoma. ^, A meta-analysis of 46 studies in adults found that individuals with >100 nevi had almost 7 times the relative risk of developing melanoma compared to individuals with ≤15 nevi. Additionally, the presence of any atypical nevus increased the relative risk of melanoma 10-fold as compared with having no atypical nevi. Atypical nevi generally begin to appear around puberty and are more likely to develop in adolescents with a higher nevi count. They are characterized as large, predominately flat or macular acquired melanocytic nevi that can have similar features as melanoma, including asymmetry, irregular border, variation in color, diameter >5 mm, and erythema. ^, Several authors use a definition of three or more of these features to identify atypical nevi. Atypical nevi are not precursors of melanoma but rather represent an overall risk factor for melanoma. Therefore, excision of an atypical nevus does not obviate the patient from close dermatologic follow-up.

Regular follow-up with total body skin examinations is recommended beginning around puberty for children with a large number of nevi, especially children with additional risk factors such as atypical nevi, family history of melanoma, history of sunburns, fair skin type, and red hair. ^{, , ,} The predisposition for a high nevi count is usually apparent by ages 11–12 years. Some will have a predominant signature nevus type, defined as a predominant group of nevi that share a similar clinical appearance. There are several types of signature nevi patterns, including solid brown, solid pink, eclipse nevi (tan centrally with brown rim), pink eclipse nevi (pink centrally with brown rim), cockade/cockarde nevi (target), nevi with perifollicular hypopigmentation, multiple halo nevi, nonpigmented (white) melanocytic nevi, compound melanocytic nevi with a fried-egg appearance, lentiginous nevi, and junction and compound (cheetah phenotype) nevi. Patients with the cheetah phenotype often have >100 nevi admixed with multiple solar lentigines. These individuals may have an increased risk of developing melanoma because of the high number of nevi present. A halo nevus ( Fig. 67.3 ), also known as a Sutton nevus, begins as a pigmented nevus surrounded by a rim of depigmentation and eventually undergoes complete regression of the nevus and repigmentation of the depigmented area. Because regressing melanoma may be gray or white, each halo nevus should be examined closely for any atypical features suggestive of melanoma.

The classic appearance of a halo nevus. These small, pigmented lesions are surrounded by a rim of hypopigmentation. They are typically seen in adolescents and are most commonly located over the trunk, especially on the back. Treatment is usually not needed unless the pigmented portion of the halo nevus appears atypical. In that case, excision of the entire lesion, including the halo, is recommended.

It is important to note that melanoma may arise de novo on any part of the body and not necessarily from a preexisting atypical lesion. Unlike in adults, growing or changing nevi are a normal part of the natural history of melanocytic nevi in children and adolescents. Whereas a biopsy is recommended in adults for a change in a nevus (in color, size, morphology, etc.), such changes among children and adolescents are poor predictors of melanoma. Several authors recommend that a change in a nevus should not be used as the sole criterion for excision. When performing a total body skin examination in a child with a large number of nevi, physicians should identify suspicious lesions based on morphologic features of individual lesions and identification of an “ugly duckling”—a lesion that is obviously different from the patient’s signature nevi pattern. ^{, ,} The ugly duckling sign has a high specificity on clinical exam and dermoscopy for detection of melanoma among providers of all levels. ^, Addition of the ugly duckling sign to traditional morphologic assessment can reduce the number of nevi considered for biopsy by a factor of 7. Dermoscopy is a critical tool in the evaluation of a skin lesion. Any lesion suspicious on clinical examination needs a clinical evaluation and dermoscopic examination by an experienced pediatric dermatologist.

In general, CMN are classified based on the projected adult size: small, <1.5 cm; medium, 1.5–20 cm; large, >20–40 cm; and giant, >40 cm ( Figs. 67.4 and 67.5 ). These lesions grow proportionally with the infant, and the projected adult size is estimated by multiplying the diameter of the lesion in infancy by the following factors: 1.7 for lesions on the head; 2.8 for lesions on the hands, feet, torso, forearms, arms, and buttocks; 3.4 for lesions on the thigh; and 3.3 for lesions on the legs. Classification of CMN was updated and standardized in 2012 by expert consensus to reflect risk factors for an unfavorable prognosis. In addition to projected adult size, this classification system includes satellite nevus count, anatomic location, color heterogeneity, surface rugosity, hypertrichosis, and dermal or subcutaneous nodules ( Table 67.2 ).

In two different patients, congenital melanocytic nevi of large (A) and giant (B) size are seen.

Photo courtesy Matthew R. Greives, MD.

Congenital melanocytic nevi seen on the neck of a child.

Photo courtesy Mary Austin, MD.

Giant CMN were historically described using garment terminology such as “bathing trunk” or “stocking-like.” Martins da Silva et al. proposed a classification scheme for the distribution patterns of giant CMN called the 6Bs: Bolero, mainly involving the upper back and neck; Back, round shape on the back with no involvement of buttocks or shoulders; Bathing trunk, lower trunk involving the genital area and buttocks with no shoulder involvement; Breast/Belly, localized to the chest or abdomen with no overlap with bolero or bathing trunk features; Body extremity, isolated to extremity with no involvement of shoulders, genital area, or buttocks; and Body, combination of bolero and bathing trunk lesions affecting almost the entire body. Using this classification scheme, most giant CMN can be categorized into one of these six distinct patterns with high interobserver and intraobserver agreement.

CMN initially appear as flat pigmented lesions ranging from tan to black and can resemble an irregular café-au-lait spot. Over 75% of lesions have overlying hypertrichosis, most commonly with dark pigmented hairs, although some contain lanugo hairs. Many of these lesions change during the first years of life. Common morphologic changes include going from a flat, soft, smooth, and evenly pigmented lesion to becoming more elevated with lighter or darker heterogeneous coloration with a pebbly, verrucous, rugose, or cerebriform surface and hair formation within the plaque. ^, Increased skin fragility in the neonatal period can cause transient ulcerations in larger CMN within the first days of life.

CMN can also develop benign proliferative nodules within the lesion. These nodules are commonly found in giant CMN and can mimic the appearance of melanoma both clinically and histologically. Although these lesions are typically benign, biopsy is sometimes indicated to exclude melanoma. ^, Other malignant and benign tumors can also develop within CMN, including rhabdomyosarcoma, dermatofibrosarcoma, and various hamartomas ( Fig. 67.6 ). Giant CMN are also associated with smaller satellite nevi that can be present at birth or arise over the following years. Other benign lesions associated with giant CMN include fascicular schwannoma and lipomas. Additionally, 42% of patients with CMN in one series had café-au-lait spots.

A large dermal hamartoma ( arrow ) is seen arising within a giant congenital melanocytic nevus.

Photo courtesy Matthew R. Greives, MD.

Melanocytic nevi that become clinically apparent during infancy and early childhood exhibit similar dermoscopic and clinical features as CMN and are referred to as tardive congenital nevi or congenital nevus-like nevi. In a series of 2-year-old children, tardive congenital nevi were similar on clinical exam to congenital nevi present at birth in symmetry, color, presence of hair, and surface area.

CMN of varying sizes are characterized by different genetic mutations. Approximately 30% of small and medium CMN have the BRAF ^V600E mutation, and 70% of have NRAS mutations. ^, In contrast, almost all large and giant CMN have NRAS mutations as the sole recurrent somatic mutation, with only a small minority harboring BRAF ^V600E mutations. ^{, ,} Additionally, large and giant CMN with NRAS mutations have been reported to exhibit cell subpopulations with specific clonogenic and tumorigenic potential, while the presence of such cell subpopulations is much rarer in medium CMN. ^, NRAS mutations, rather than BRAF mutations, may play an important role in the risk of melanoma arising in CMN. Genetic sequencing of melanoma arising in patients with CMN has demonstrated a characteristic NRAS mutation and was notable for a distinct lack of BRAF mutations. The differences in NRAS mutations among small and medium CMN compared to large and giant CMN may explain differences in malignant degeneration. Whereas the overall incidence of melanoma in CMN is 1%–2%, the lifetime risk of melanoma for patients with a giant CMN is 10%–15%. ^,

Neurocutaneous melanocytosis (NCM) is the melanocytic proliferation in the leptomeninges and brain parenchyma (cerebral neuromelanosis) in patients with CMN. Approximately 30%–45% of patients with giant CMN develop NCM. NCM is also associated with the presence of numerous CMN lesions. Approximately a third of patients with NCM have >10 lesions of small-to-medium CMN while two-thirds have large or giant CMN with multiple satellite nevi. ^, Other risk factors associated with NCM include male gender and location on the head, neck, or posterior midline. ^{, ,}

NCM can be benign or malignant, nodular or diffuse, and can affect the amygdala, cerebrum, cerebellum, pons, medulla, and spinal cord. Most symptomatic patients present by 2 years of life with symptoms of hydrocephalus, seizures, and increased intracranial pressure such as headache, lethargy, vomiting, and photophobia. ^{, ,} Patients with a discrete mass may present with more focal sensorimotor deficits. Symptomatic patients have a poor prognosis and high mortality rate secondary to mechanical obstruction and the development of CNS melanoma. Overall survival (OS) at 3 years after becoming symptomatic is <50%.

Screening MRI of the brain and spine is recommended for children at high risk for NCM within the first 6 months of life. ^{, ,} However, not all patients with evidence of NCM on imaging will become symptomatic. Asymptomatic NCM can be found in 10%–74% of patients with MRI findings of NCM. ^, These children should be followed closely with serial neurologic examinations and developmental assessments with repeat imaging if symptoms develop. ^{, ,} A screening MRI of the brain may also be of additional prognostic benefit in patients born with CMN. After adjusting for projected adult size, the odds of developing melanoma is 17 times greater in children with CMN and an abnormal screening MRI of the CNS in the first year of life as compared with children with normal screening MRI. Additionally, an abnormal screening MRI in children with two or more CMN is the best predictor of other neurodevelopmental abnormalities, seizures, and the need for subsequent neurosurgery. Therefore, an abnormal screening MRI in the first year of life may be the best predictor of all adverse outcomes in children with CMN. ^,

For many parents, CMN in conspicuous areas and large or giant CMN are of great concern. Additionally, children with CMN, especially large CMN, are more likely to suffer from psychosocial problems and suffer from anxiety and depression. Management of CMN is tailored to the individual patient with different considerations in treatment based on the size and location of the CMN.

Most authors recommend serial examinations with photographs and dermoscopy at intervals of 6 months to 2 years for small CMN and 6 months to 1 year for medium-sized CMN. ^, The frequency of monitoring and management strategy depends on the patient’s age, nevus size, location and ease of self-monitoring, attitude toward monitoring, anesthesia requirements (local vs. general), factors that affect healing/scarring, cosmetic concerns, and anxiety level. ^, The need for more frequent monitoring is especially important after puberty because the risk for melanoma increases. ^,

Routine prophylactic excision of small- and medium-sized CMN is not recommended. ^, When parents or the patient desire excision of the lesion for cosmetic reasons, complete excision of lesions <5 cm can usually be accomplished in a single procedure with simple excision and primary closure, local plasty, or a free tissue skin graft. ^, Medium-sized CMN ≥5 cm may require multiple procedures with serial excision, tissue expanders, and tissue flaps. For facial CMN not amenable to excision, laser therapy with long-pulsed and Q-switched lasers (e.g., ruby, alexandrite, and neodymium-doped yttrium aluminum garnet) may improve the cosmetic appearance. However, mild repigmentation is common, and multiple sessions may be needed. ^, It is important to counsel patients and parents about the anticipated esthetic results and potential postoperative problems such as scarring. Additionally, for smaller lesions that could potentially be removed using a local anesthetic, it is best to delay elective removal until the patient can tolerate the procedure without general anesthesia. ^,

Early and complete or partial excision of large and giant CMN does not appear to reduce the risk of developing melanoma. ^{, ,} Complete removal of every nevus cell is unlikely, and recurrence of pigmented nevi and the development of melanoma in the scar, skin grafts, and/or tissue flaps has been reported. Furthermore, excision of CMN does not reduce the risk of primary CNS melanoma, which accounts for a substantial proportion of the cases of primary melanoma in patients with CMN. ^,

Excision of large and giant CMN may be indicated to decrease symptoms, improve cosmesis, and improve perceived well-being and quality of life. Tissue expanders, rotational flaps, and skin grafting may be helpful for surgical management of large/giant CMN. ^{, ,} In one series of CMN >20 cm, serial excisions were performed in 50% of lesions and tissue expanders were frequently used to provide expanded full-thickness skin grafts. In another series of patients with large and giant CMN, the mean age for intervention was 5.1 years, and it took an average 1.3 years to complete resection and reconstruction. Although the authors expressed a preference for earlier intervention, they acknowledged that the average age of the patients referred to them was 4.7 years, and many had already undergone a prior operative procedure. Additionally, more than 50% of these patients required more than one operative intervention and/or skin grafting. Also, many patients benefitted from tissue expanders and autologous cultured skin replacements.

It is important to carefully consider the risks and benefits for each individual patient when considering operative treatment. Surgical procedures are associated with a high rate of complications, including infection, graft loss, flap necrosis, poor wound healing, and, rarely, death. ^, Additionally, patients may require immobilization and limitation of physical activity postoperatively. Scarring can result in disfigurement and joint immobility. Impaired function is also a risk. ^, In a series of 301 patients with CMN, a significant proportion of patients and families felt that the operation had worsened the appearance in children with CMN >20 cm.

Regardless of the treatment modality, close monitoring and follow-up with serial total body skin examinations and dermoscopy is necessary in patients with large and giant CMN. Any suspicious changes, including the rapid growth of papules/nodules, ulceration, discoloration, or palpation of firm nodules beneath the scar, warrant full-thickness biopsy and histologic examination to evaluate for melanoma. Management of melanoma arising in CMN will also be discussed later in the chapter.

Spitzoid lesions are melanocytic lesions composed of epithelioid and spindle cells. There is a broad spectrum of spitzoid lesions with little diagnostic agreement among dermatopathologists. Whereas some authors maintain that all lesions can be dichotomized as either a benign nevus or malignant melanoma, others suggest that spitzoid lesions, especially atypical Spitz nevi/tumors, represent a heterogeneous group of lesions with distinct clinical, pathologic, and genetic features. ^, Additionally, the distinction between benign Spitz nevus, atypical Spitz nevus/tumor, and melanoma based on clinical exam, dermoscopy, and histology is difficult because of overlapping features. ^, Church et al. attempted to better classify these lesions in the pediatric population using genetic markers; however, the results were not statistically significant. Therefore, evaluation of these lesions by a dermatopathologist with experience in spitzoid lesions is necessary to avoid misdiagnosis of spitzoid melanoma as atypical Spitz nevus/tumor.

Spitz nevi are benign neoplasms that occur primarily in children and young adults, with up to 50%–80% occurring in those less than 20 years old. These lesions are often clinically misdiagnosed as dermatofibroma, hemangioma, viral wart, xanthogranuloma, epidermal nevus, and basal cell carcinoma. The classic appearance of a Spitz nevus is a solitary pink, red, or brown papule with either a smooth or verrucous surface. ^{, ,} On dermoscopic examination, a classic starburst pattern is seen ( Fig. 67.7 ). The most common location is the lower extremities, and children are more likely than adults to have a lesion on the head or neck ( Fig. 67.8 ). ^{, ,} They can have a fast initial growth period, with one study reporting a median time to development of 12 months. Many will eventually undergo involution independent of the patient’s age, gender, location of the lesion, and pigmentation of the lesion.

This patient has a Spitz nevus on her arm (A). Dermoscopic examination shows a classic starburst pattern (B and C).

Photos courtesy Dr. Eric Ehrsam, reprinted with permission.

A pigmented Spitz nevus is seen on the left ear of a child.

Photo courtesy Matthew R. Greives, MD.

Atypical Spitz nevi/tumors have the potential to metastasize. Many believe that atypical Spitz nevi exist on a spectrum between benign Spitz nevi and malignant spitzoid melanoma. ^, While the majority of atypical Spitz nevi/tumors have a clinically benign course, approximately 40% will have nodal metastases, and the overall mortality is estimated to be 1%. ^, Proposed criteria for favoring an atypical Spitz nevus on exam and dermoscopy versus a benign Spitz nevus include a diameter >7 mm; a nodular, firm, ulcerated, hypopigmented, or amelanotic lesion, especially in children >12 years; an asymmetric growth pattern; and increased atypia on the vascular pattern.

There are no universally accepted histopathologic criteria for atypical Spitz nevi/tumors. Urso proposed nine histologic features of an atypical Spitz nevus, including nodular growth in the dermis, deep extension to the middermis or subcutaneous fat, dermal/deep mitoses, great nuclear and nucleolar pleomorphism, melanization deep within the tumor, asymmetry, necrosis, ulceration, and neoplastic cells within the lymph vessels. Additional features associated with atypical Spitz nevi include asymmetry, Clark level IV or V, high mitotic rate (2–6 mitoses/mm ²⁾ , especially in children >14 years, lack of cytologic maturation, solid growth, nuclear pleomorphism, and atypical and deep mitoses.

One of the difficulties in determining whether a lesion is an atypical Spitz nevus/tumor is that atypical Spitz nevi/tumors may have features of atypia on dermoscopy and lack features of atypia on histopathology, and vice versa. ^, Additionally, morphology on histopathology does not consistently predict biologic behavior, and many features used to distinguish conventional melanoma from nevi are not useful indicators of behavior in atypical Spitz tumors. Recently, several studies have attempted to identify molecular markers to characterize the spectrum of atypical Spitz nevi. ^{, , ,}

Whereas most Spitz nevi have a normal karyotype, atypical Spitz tumors appear to be a heterogeneous group of genetically and biologically distinct subtypes. Genetic subtypes include BRAF ^V600E / BAP1 ^neg mutant, HRAS mutant with increased copies of 11p, and homozygous 9p21 deletion with negative p16 expression. ^{, ,} Additionally, kinase fusions of ROS1, NTRK1, ALK, BRAF, and RET create chimeric proteins that stimulate oncogenic signaling pathways and are tumorigenic are found in the entire spectrum of spitzoid neoplasms, including 55% of Spitz nevi, 56% of atypical Spitz nevi/tumors, and 40% of spitzoid melanoma in a mutually exclusive pattern.

The decision to perform a biopsy of a lesion believed to be a Spitz nevus should depend on the age of the patient and characteristics of the lesion such as the pattern of growth, size, and characteristics on dermoscopy. ^{, ,} In children <12 years with plaque-like or dome-shaped pigmented and nonpigmented Spitz nevi that have classic features on dermoscopy, serial clinical examinations with dermoscopy is recommended every 3–6 months. ^{, , ,} A completely benign Spitz nevus is expected to symmetrically enlarge, often rapidly, and then slowly involute completely or into a homogenous light brown nevus. ^, In these patients, biopsy is warranted if the lesion develops any atypical features on follow-up exam, including asymmetric growth, nodular/polypoid growth, ulceration, or increased atypia in the vascular pattern on dermoscopy. In patients ≥12 years of age, the risk of melanoma increases. Because clinical exam and dermoscopy cannot reliably distinguish between a Spitz nevus and melanoma, full-thickness biopsy is recommended for suspected spitzoid lesions in patients ≥12 years of age. Additionally, the presence of any atypical features in patients of any age warrants full-thickness biopsy.

For biopsy specimens with a histopathologic diagnosis of Spitz nevi, no further intervention is needed and reexcision of lesions with positive margins is not necessary. ^, Management of spitzoid melanoma is the same as conventional melanoma. For atypical Spitz nevi/tumors, most recommend reexcision of the biopsy site with 1- to 2-mm margins, especially if the biopsy specimen showed incomplete removal of the lesion. ^{, , , ,} There is considerable debate on the prognostic and therapeutic benefit of sentinel lymph node (SLN) biopsy in pediatric patients with atypical Spitz nevi/tumors. A recent systematic review of 24 studies published between 2002 and 2013 comprising 541 patients aged 2–65 years with atypical Spitz tumors and a mean follow-up of 5 years found no prognostic benefit of performing SLN biopsy. Of 303 (56%) patients who had SNL biopsy performed, 119 (39%) had a positive SLN and 97 (82%) of those with positive SLN went on to have a completion lymph node dissection (CLND). The 5-year OS was 99% (118/119) for patients with a positive SLN as compared with 98% (233/238) for patients who did not have SLN biopsy. Thus, SLN biopsy is not recommended in patients with atypical Spitz nevi/tumors.

Mutation analysis may help guide management decisions for atypical Spitz nevi/tumors. Several genetic markers have been identified as potential prognostic indicators. Among atypical Spitz tumors, fluorescence in situ hybridization (FISH) assays demonstrating gains in 6p25 ( RREB1 ), 11q13 ( CCND1 ), and homozygous deletions of 9p21 ( CDKN2A ) are associated with a higher risk of aggressive clinical behavior. ^, TERT -p mutations, found in over 90% of conventional melanomas, may also be a marker of more aggressive behavior when present in spitzoid lesions. ^, Conversely, isolated 6q23 ( MYB ) loss and loss of 3p21 in BAP1-associated Spitz tumors are associated with a favorable clinical outcome. ^, A risk stratification system using FISH has been suggested as a prognostic tool, but has not yet been validated ( Table 67.3 ).

Melanoma in children and adolescents represents less than 1% of all melanoma cases. However, it is the second leading cause of cancer in adolescents and young adults aged 15–29 years old. Melanoma in young children is very rare, with an incidence of <1.8 per million in children <10 years old. However, the incidence sharply increases to 7.5 per million in 15–19-year-olds. Adolescents represent 73% of melanoma cases in children, followed by 10–14 year olds (17%), 5–9 year olds (6%), and 1–4 year olds (4%).

Beginning in 1973, the incidence of pediatric melanoma increased, with an average annual percent change (APC) of 2%–2.9%. This trend continued until the beginning of the 21st century, when the incidence of melanoma among children and adolescents significantly decreased. A study using the Surveillance, Epidemiology, and End Results (SEER) cancer registry found that the incidence among children <20 years decreased 12% per year from 2004 to 2010, most notably among adolescents ages 15–19 years old and among female adolescents. Similarly, a study using the 2001–09 National Program of Cancer Registries (NPCR) and SEER data demonstrated an APC of −5.1% among adolescents with melanoma. Another SEER study from 2000 to 2011 demonstrated an APC of −1.2% among adolescents and young adults 15–39 years old. ^,

There are several known genetic and environmental risk factors that predispose to the development of melanoma. Similar to adults, children with Fitzpatrick I pigmentation (pale white skin; blond or red hair; blue, gray eyes; freckles; always burns, never tans) have a higher risk of developing melanoma. ^{, ,} The strongest risk factor among adolescents 15–19 years is the presence of more than 100 nevi >2 mm in diameter. A number of disorders are also associated with the development of melanoma, including retinoblastoma, Werner syndrome, and xeroderma pigmentosum. Patients with xeroderma pigmentosum have an increased cutaneous sensitivity to light resulting in a greater than 10,000-fold increased risk of skin cancer and 2000-fold increased risk of melanoma. UV exposure early in life leads to accumulated sunlight-induced DNA damage, which can result in skin cancer within the first decade of life. For patients with xeroderma pigmentosum, the median age of onset of nonmelanoma skin cancer is 9 years and the median age of onset for melanoma is 22 years, which is more than 30 years earlier than in the general U.S. population. Although skin cancer is the leading cause of death in patients with xeroderma pigmentosum, more than 45% of patients live into their forties despite the early age of diagnosis.

Patients with inherited and acquired immunodeficiencies are at an increased risk for developing melanoma. Inherited immunodeficiencies confer a threefold to sixfold increased risk of melanoma. Pediatric organ transplant recipients are also at increased risk of skin cancer development, including nonmelanoma skin cancer, melanoma, Kaposi’s sarcoma, and anogenital carcinoma. The combination of immunosuppression, impaired DNA damage repair, and infection with oncogenic viruses contributes to the risk of malignancies following organ transplantation. Although nonmelanoma skin cancer is the second most common posttransplantation malignancy following lymphoproliferative disorders, the number of cases of melanoma following pediatric organ transplantation is small. In one series of 512 patients <18 years old who received organ transplants, 12 patients developed melanoma, with 50% occurring during childhood. Overall, melanoma represented 2% of all posttransplantation cancers but was fatal in 25% of cases. Another series of 536 pediatric transplant patients in Sweden estimated the relative risk of melanoma to be 4.5 times greater than in the age-matched population. However, only 2 of 536 patients developed melanoma in situ or melanoma, with an average time of onset 19 years after transplantation.

Melanoma can also occur in CMN, other preexisting nevi, and, in very rare cases, through placental transmission to an unborn child. The overall incidence of melanoma in CMN is 1%–2%, with a mean age of diagnosis of 12–15 years. ^{, , , ,} Approximately two-thirds of primary melanomas arise inside the nevi, and up to one-third may occur as primary central nervous system (CNS) melanoma. The risk of melanoma is closely associated with the projected adult size of the lesion. The estimated lifetime risk for a single small CMN is very low, whereas the risk for CMN >40 cm with multiple smaller satellite lesions is 8%–15%. ^, Additionally, patients with CMN and congenital abnormalities of the CNS have a higher risk of melanoma. In patients with CMN and an abnormal screening MRI of the CNS in their first year of life, the incidence of melanoma is 12% compared to 1%–2% in individuals with a normal MRI.

Familial melanoma accounts for approximately 1% of cases of melanoma. ^, Approximately 20%–40% of families with three or more individuals with melanoma have mutations in CDKN2A, whereas CDK4 mutations have been found in only 17 families. Mutations in CDKN2A and CDK4 are associated with atypical nevi, early-onset melanoma, and multiple primaries. ^,

Although several genetic and environmental risk factors are associated with the development of melanoma in children, the majority of pediatric melanoma cases are sporadic and attributed to UV radiation. ^, A review of eight single-institutional studies comprising 322 children and adolescents with melanoma found only 22% of patients had predisposing conditions such as dysplastic or numerous melanocytic nevi, large/giant CMN, family history of melanoma, xeroderma pigmentosum, inherited or acquired immunodeficiency disorder, prior malignancy, or history of irradiation. Growing molecular evidence supports that the majority of pediatric melanoma is the result of sequential acquisitions of multiple mutations from UV exposure. ^,

Whole-genome sequencing (WGS) and whole-exome sequencing (WES) of 15 conventional melanomas in pediatric patients demonstrated a high rate of somatic single-nucleotide variations (SNVs) that exceeds that of any other pediatric tumor sequenced in the Pediatric Cancer Genome Project. Furthermore, more than 80% of the SNVs were cytidine to thymidine or guanine to adenosine, reflective of UV-induced damage. Somatic mutations in the protooncogene BRAF ^V600 are found in approximately 60% of adult melanoma, and 20% carry NRAS mutations. Similarly, the series of pediatric melanoma patients by Lu et al. demonstrated that 87% of conventional pediatric melanoma contained oncogenic BRAF ^V600 mutations and commonly PTEN alterations, whereas all NRAS mutations were associated only with pediatric melanoma arising in association with CMN. BRAF mutations stimulate the MAPK signaling pathway and require additional loss of PTEN to activate the PI3K/AKT cellular signaling pathway, whereas NRAS activates both the MAPK and PI3K/AKT pathways. Additionally, all 15 conventional pediatric melanomas contained a telomerase reverse transcriptase promoter ( TERT- p) mutation. This finding is similar to that in adult melanoma, in which over 70% have TERT -p mutations. The combination of TERT -p alternations, such as methylation, and BRAF ^V600 mutations is associated with reduced disease-free survival (DFS) and OS in pediatric patients. Additional somatic mutations associated with the development of pediatric melanoma include tumor suppressor p53, CDKN2A, and proto-oncogenes RAS, KIT, and MITF. ^{, ,}

Pediatric melanoma is a rare disease, and that rarity contributes to a decreased perception of risk among parents and physicians. Delays in diagnosis or misdiagnosis have been reported to occur in 50%–60% of patients and are associated with advanced stage of disease at presentation and increased mortality risk. ^, Traditional ABCDE ( A symmetry, B order irregularity, C olor variegation, D iameter >6 mm, E volution) criteria used among adults to aid in the diagnosis of melanoma fails to identify a subset of pediatric melanoma, especially among prepubertal patients. In one study, 60% of children <10 years old and 40% of patients aged 11–19 presented with lesions that were lacking conventional ABCDE criteria. Melanoma in children can present as nodular amelanotic lesions that are red or pink and can resemble benign lesions such as pyogenic granulomas, sebaceous nevi, hemangiomas, and warts ( Fig. 67.9 ). ^, Cordoro et al. proposed a new ABCD criteria to be used in addition to the traditional ABCDE criteria ( Table 67.4 ). The pediatric modified ABCD criteria serves to heighten awareness when examining any new or persistent lesion in a child that is pigmented or amelanotic, with or without symmetry, bleeding or ulceration, irregular borders, color variegation, or of any diameter ( Fig. 67.10 ). ^{, ,} Evolution of a lesion in adults or children is a key finding that should always raise suspicion for melanoma.

A scalp sebaceous nevus (A) and a hemangioma on the forearm (B) are seen with features mimicking melanoma.

Photo courtesy Matthew R. Greives, MD.

A melanoma is seen on the lower extremity of a 6-year-old girl (A) and an amelanotic lesion with ulceration and color variegation is seen in a 12-year-old girl (B).

The demographics, site of distribution, and stage of disease at presentation differs among prepubertal children, adolescents, and young adults. Similar to adults, pediatric melanoma occurs most often in Caucasian females. ^{, ,} However, the distribution of melanoma among males and females ranges from equal to a slight male dominance in patients <10 years. ^{, ,} Additionally, the proportion of nonwhite patients is higher among younger patients. ^{, ,} In a series of 1185 patients <20 years with melanoma identified in the SEER registry, 20% of patients aged 0–14 years old were nonwhite and only 6% of patients aged 15–19 years old were nonwhite. Hispanics were the second most common race/ethnicity affected by melanoma and represented 14% of patients ages 0–14 years. It is important to keep a high index of suspicion when evaluating skin lesions among Hispanic children because Hispanic patients with melanoma are more likely than non-Hispanic white patients to have regional or distant disease at presentation and have a worse OS.

The most common primary site of melanoma is the trunk in 15–19 year-olds, and the primary site in younger children is more likely to be the head and neck or extremities. ^, Superficial spreading melanoma is the most common subtype, although in children younger than 10 years there is equal proportion of superficial spreading and nodular melanoma. ^, Overall, most melanomas are thin at presentation with a depth of ≤1 mm. ^, However, younger children (<10 years old), as compared with adolescents, are more likely to present with tumors that are thicker, ulcerated, and have at least 1 mitosis per square millimeter. ^{, , ,} Children <10 years are also more likely to present with regional lymph node involvement and distant metastases as compared with adolescents.

A full-thickness excisional biopsy with a 1- to 3-mm margin is indicated for any suspicious lesion. The incision should be oriented parallel to the lymphatics and longitudinally along extremities. A full-thickness punch biopsy of the thickest portion of the lesion is acceptable in certain anatomic areas where an elliptical incision may be difficult or interfere with cosmesis, including the palm/sole, digit, face, ear, or for very large lesions. A shave biopsy is not indicated when melanoma is suspected because it may compromise pathologic diagnosis and assessment of Breslow thickness. Because of the difficulty in differentiating benign from malignant melanocytic lesions in pediatric patients, all histologic slides should be reviewed by a dermatopathologist highly experienced in the diagnosis of pediatric melanoma.

The management for pediatric melanoma is extrapolated from adult patients with melanoma and follows the National Comprehensive Cancer Network (NCCN) guidelines. All melanomas should be treated by wide local excision with surgical margins based on the Breslow thickness ( Table 67.5 ). Wider margins in the pediatric population may confer additional morbidity and worse cosmesis; however, there is currently no data on optimal excision margins in this population. A study evaluating pediatric patients who received either wide versus narrow excision found no difference in OS. Retrospective reviews of adult patients with cutaneous melanoma of the head or neck have shown similar results. Additionally, the effect of excision margins on OS did not change based on Breslow depth. Given that pediatric patients also have a lower risk of local recurrence compared to adult patients, it may be acceptable in certain cases to obtain smaller margins in the pediatric population. SLN biopsy is recommended for patients with tumors ≥1 mm thick and in patients with tumors >0.75 mm thick who also have ulceration or a mitotic rate ≥1 per mm ² . ^,

Melanoma within an SLN is conventionally evaluated by hematoxylin and eosin histopathology and immunohistochemistry (IHC). However, molecular staging with reverse-transcriptase polymerase chain reaction (RT-PCR) for melanoma-specific mRNA has also been proposed to evaluate SLNs that are negative on histopathology and IHC. Results from the Sunbelt Melanoma Trial in adult patients with melanoma demonstrated slightly improved DFS but not OS for CLND in patients with melanoma detected in the SLN by RT-PCR but tumor-negative by conventional pathology.

SLN biopsy is very important for staging and prognosis. Results from the Multicenter Selective Lymphadenectomy Trial (MSLT-I) demonstrated that SLN biopsy with CLND for positive sentinel node(s), as compared with nodal basin observation, improved DFS among patients with intermediate and thick tumors but did not improve melanoma-specific survival. However, a positive SLN was associated with significantly worse melanoma-specific survival as compared with a negative SLN. The prognostic value of SLN biopsy in pediatric patients has also been demonstrated. A retrospective SEER study of 310 patients <20 years with melanoma and Breslow depth >0.75 mm demonstrated children with a positive SLN had worse melanoma-specific survival and that SLN biopsy did not improve melanoma-specific survival.

Current NCCN guidelines now recommend observation without surgery and mandatory radiographic nodal surveillance over CLND. While CLND was previously the standard of care for patients with a positive SLN, the MSLT-II and DeCOG-SLT trials have now demonstrated no melanoma-specific survival benefit in adult patients who underwent CLND as compared with patients observed with frequent nodal ultrasonography and dissected only after clinically detected nodal recurrence occurred. While the MSLT-II trial demonstrated that there may be some benefit of immediate CLND in reducing the rate of regional node recurrence, CLND is associated with significant complications. In the MSLT-II, 24% of patients who underwent CLND developed lymphedema, as compared with 6% in the observation group. In the DeCOG-SLT trial adverse events including lymphedema or lymph fistula occurred in 24% of patients in the CLND group. ^, The rate of complications from CLND in children is similar. In a series of 125 pediatric patients <18 years old, 20% of patients who had a CLND developed lymphedema as compared with 2% of patients who had an SNL biopsy. Additionally, the risk of lymphedema was greatest in inguinal dissections. CLND is a procedure with a significant rate of complications and long-term morbidity. Currently, there are no good data to determine the impact of CLND in pediatric patients with melanoma and a positive SLN biopsy. Whether the results from the MSLT-II or DECOG-SLT study can be extrapolated to the management of children is unclear. A recent study evaluated current nodal management in children from 2012 to 2017—after the MSLT-II—and found that fewer CLND were performed in children in 2017 compared to 2012.

Some authors question the appropriateness of SLN biopsy and CLND in children ≤10 years of age. The rate of a positive SLN among all pediatric patients is 25%–30% and is as high as 58% among patients ≤10 years old. ^{, , ,} However, single-institution retrospective studies have demonstrated no association between SLN positivity and mortality among patients ≤10 years old. ^, Additionally, an NCDB study demonstrated that OS of patients ≤10 years who underwent SLN biopsy or CLND was not significantly different from that of pediatric patients who did not have SLN biopsy. However, a known positive SLN may lead to consideration of adjuvant therapy such as interferon-α (IFN-α). A more detailed understanding of the differences in nodal disease among pediatric patients ≤10 years of age versus adolescent patients is needed to better understand the prognostic value of SLN biopsy and the therapeutic value of CLND for the younger cohort.

Before proceeding to surgery for management of melanoma identified on a biopsy, a full physical exam should be performed. Close attention should be paid to identifying additional primary melanomas, in-transit or satellite disease, and any clinically positive lymph nodes. In a patient with suspicious regional lymphadenopathy, ultrasound and fine-needle aspiration should be considered to evaluate nodal metastasis. In patients with evidence of in-transit metastasis or nodal disease, staging imaging is recommended including chest/abdominal/pelvic CT with contrast, brain MRI with contrast, and/or ¹⁸ F-fluorodeoxyglucose ( ¹⁸ FDG)–positron emission tomography (PET)/CT. ^,

When SLN biopsy is indicated, a dual-modality approach using a gamma-sensor probe that detects the technetium-99m ( ^99m Tc) and isosulfan blue dye for visual confirmation is used for SLN identification. Preoperative lymphoscintigraphy is performed to identify nodal basins and SLNs and to assist with operative planning. A radiocolloid, usually ^99m Tc, is injected into the skin surrounding the primary site. LYMPHOSEEK (Tc-99m-labeled Tilmanocept) is an alternative agent to ^99m Tc that shows more rapid transit and accumulation in lymph nodes. ^, Fig. 67.11 shows lymphoscintigraphy of a patient with a primary lesion on his left arm that demonstrates drainage from lymphatic channels to a single epitrochlear SLN and two axillary SLNs. Lymphoscintigraphy can be performed a few hours prior to the operation or several days before with a repeat injection of the radiotracer 1–2 hours preoperatively. Intraoperatively, isosulfan blue dye is injected intradermally around the biopsy site of the melanoma. The radiotracer and isosulfan blue dye enter the lymphatic channels and flow to the draining lymph nodes ( Fig. 67.12 ). The general location of the SLN is estimated with the use of a handheld gamma-sensor probe and marked. Fig. 67.13 demonstrates intradermal injection of isosulfan blue dye around the prior biopsy site of a patient with melanoma and the external marking of sites with the highest accumulation of radionuclide as detected on a handheld gamma-sensor probe. An incision is then made overlying the area with the highest count on the gamma-sensor probe. SLNs are identified visually by following blue lymphatic channels to blue-stained lymph nodes ( Fig. 67.14 ). After removal of the node, an ex-vivo count of the node is obtained using the handheld gamma-sensor probe. The basin is then rescanned with the gamma probe to evaluate for the presence of any additional SNLs. Wide local excision of the primary tumor is then performed after completion of the SNL biopsy ( Video 67.1 ).

This is a representative image from a lymphoscintigraphy study performed on a 10-year-old boy with melanoma on the skin covering his left elbow. The study shows drainage to a single epitrochlea sentinel lymph node and two axillary sentinel nodes ( arrows ).

Stay updated, free articles. Join our Telegram channel

Join