Case Presentation LM, a healthy 2-year-old boy, presented with a nevus noted at birth. The nevus has a darker central area and measures 4 cm by 4 cm (see below). It’s located over the occipital area to the left of midline. LM’s parents have heard from friends that there’s a chance this will turn into skin cancer. The family is referred to a dermatologist who is concerned that the posterior location may indicate potential associated problems. C ongenital melanocytic nevi (CMN) are benign proliferations of nevus cells present from birth. They vary considerably in morphology with respect to color, shape, size and border regularity (see photos throughout). On histopathology, the nevus cells tend to be deep in the dermis and may be present within dermal appendages such as hair follicles and nerves. This is in contrast to acquired nevi in which the nevus cells are typically more superficial. So-called “tardive” congenital nevi are comparable to congenital melanocytic nevi in their clinical and pathologic features but are not present at birth. They may appear up to about 2 years of age. Overall, CMN occur in about 1% of children. CMN can generate concern amongst parents and physicians regarding the risk of malignancy and cosmesis. A third concern, of which dermatologists should be aware, is the issue of neurocutaneous melanosis (NCM), which can accompany certain congenital melanocytic nevi. These three concerns will be reviewed in this article. The risks of melanoma or NCM and the ease of cosmetic repair are often approached by considering the size of the nevus. The system most commonly used to categorize the size of a CMN is based on the predicted final adult size of the lesion as follows: • <1.5 cm is a small CMN • 1.5 cm to 19.9 cm is a medium or intermediate nevus • >20cm is considered large (LCMN).1 A rough estimate suggests a 9-cm CMN on the head or 6-cm CMN on the body in childhood will achieve an adult size of >20 cm.2 Ruiz-Moldanado3 has recently suggested a further modification to the “small, medium, large” system that would allow for more groupings. If adopted, this classification may enable better prognostication by narrowing the range of sizes incorporated in any one group. Some patients with large CMN also have multiple satellite lesions. These are smaller melanocytic nevi that often accompany a LCMN, and they may or may not be numerous. Their presence may also have some predictive value. Additional nevi can also appear after birth. Neurocutaneous Melanosis Nevus cells in the skin and elements of the central nervous system share a common origin from the neural crest. Dysregulated growth of melanoblasts can lead to melanocytic proliferations in the central nervous system as well as the skin. This phenomenon is called neurocutaneous melanosis. It has been more formally defined by Kadonaga and Frieden as melanocytic proliferations in the central nervous system in association with one LCMN or at least three smaller CMN.2 These proliferations are usually benign but can become malignant. NCM can be asymptomatic or manifest with neurologic symptoms. Manifest or symptomatic NCM often presents in the first few years.4 Symptoms can include hydrocephalus, seizures, increased head circumference or developmental delay. Manifest NCM typically has a very poor prognosis with a very high mortality rate.4 The importance of establishing the diagnosis of symptomatic NCM is in its ability to help direct management. A decision to undergo complicated surgical excision of a LCNM may be withheld if the overall prognosis for the patient is extremely poor. Asymptomatic NCM is a diagnosis made by imaging or autopsy as the patient will not clinically manifest the neurological signs. It is unclear how prevalent this problem is because not all patients with CMN undergo neuroimaging. Frieden et al found that 30% of patients with a LCMN had asymptomatic NCM.5 Routine MRIs will likely pick up an increasing number of cases of asymptomatic NCM in the context of a LCMN or multiple smaller CMN. However, this approach will also raise new issues: • What follow-up should these patients receive? • Should they have repeat imaging, and if so, when? • Does the information gained justify the risk of anesthesia for the MRI? Some might argue that a normal MRI might be very reassuring to parents. It might also be argued that knowing in advance that the CNS is involved would allow doctors and parents to closely monitor for symptomatic neurologic involvement. However, most cases are found incidentally, and outcomes appear to be good. Are certain patients more at risk for developing NCM? Certain features do increase the likelihood. Nevus size is a risk factor as most reported cases occur in patients with large or multiple smaller congenital nevi. There is no evidence that single small CMN (the majority of congenital nevi) have any association with NCM. In respect to large CMN, one study showed a 5-year cumulative risk of NCM of 2.5%.4 The risk may be higher in those with lesions >50 cm. Location also appears to be a risk factor. In most reported cases, the LCMN has been in a posterior axial location.6 This holds true for the cases of asymptomatic or symptomatic NCM. Number-One Predictor of NCM The single factor that is the most important predictive factor for NCM is the number of satellite lesions.7 Patients with NCM had significantly more satellite lesions (68.5) versus patients without NCM (18 satellites). Marghoob showed that having >20 satellites was associated with a 5-fold increased risk of NCM.7 Another association to consider in patients with a CMN over the lumbosacral area in the midline is spinal cord/column abnormalities. Skin lesions in this location may be a marker for deeper abnormalities.8 An MRI can rule out this association. Risk of Melanoma Overall, melanoma is very rare in childhood. One of the known risk factors is a large congenital nevus. The incidence of melanoma has been reported to be between 3.3% and 6.0% 9,10,11 in several prospective and retrospective studies of melanomas arising within large congenital nevi. A systemic review of published data found that overall 2.8% of patients with LCMN develop melanoma.12 Melanoma has been reported to occur in the actual CMN as well as other sites. DeDavid found 12% of patients with LCMN had primary melanoma within a CMN, and all patients had their CMN in an axial location.13 Patients with CMN on the extremities did not develop melanoma. Melanomas have also occurred at cutaneous sites remote from the CMN. It is difficult to know if these melanomas are de novo or whether the patients were at higher risk by virtue of their CMN. No study has found melanomas occurring in satellite lesions. Noncutaneous melanoma has been reported as well. It has occurred in the context of NCM (malignant neurocutaneous melanosis). One study found 21 of 289 patients with NCM had malignant growths. A review of the NYU-LCMN registry in 2000 found eight melanomas of which 3/8 occurred in LCMN; 1/8 occurred in normal skin, 1/8 were in the retroperitoneum and 2/8 were in the CNS. The origin of one melanoma was unknown. At least half of all melanomas that arise within CMN occur at an early age, often before age 5.13 This knowledge has led to relatively aggressive attempts by some centers to remove the nevus in infancy in an attempt to prevent melanoma development. CNS melanoma also tends to occur at an early age. Management of CMN is very difficult, and one often needs to weigh aggressive surgery against watchful waiting in the face of a small, but real, risk of melanoma. Predictive factors to isolate which patients most likely to develop melanoma would be very useful. As with the risk of NCM, the axial location and larger number of satellites are indicators of increased risk of melanoma. Bittencourt et al found that in eight patients with melanoma, all had more than 20 satellites.4 There are several reasons that melanoma can be difficult to detect in CMN. These lesions often have variegated texture and color, and subtle change is difficult to detect. The depth of the lesion can also obscure a subtle change. Just as the benign nevus cells can be quite deep in a CMN so can the malignant cells. This may lead to a deep nodule that is not easy to detect until it is large and palpable. CMN that occur in the scalp may be difficult to observe and follow. Finally, benign nodular growths that simulate melanoma can arise within CMN. These often regress. Pathologic distinction can often be made between proliferative nodules and melanoma. One study showed that a PET scan was useful in identifying malignant nodules.14 This may be a useful tool in the future. Evaluating the Size of CMN in Relation to Melanoma Risk Many dermatologists question the risk of melanoma in small- and medium-sized nevi as they are more commonly encountered in clinical practice. Swerdlow11 examined 232 medium CMN and found no melanomas. Sahin et al, in a short-term follow-up, found no melanomas arising in medium CMN.15 There were however three patients (all over age 35) who developed melanoma at other cutaneous sites. There have also been reports of melanoma in adulthood arising from a CMN.16,17 It’s worth considering the morphology of the smaller CNM in decision-making. It may be much harder to observe a dark, irregular, nodular CMN, and earlier excision may be favored. There is clearly not enough prospective data to definitively assess the risk of melanoma in small and small-medium congenital nevi. The risk appears very low and has not been reported to occur prior to puberty.16,18 This would support a “wait-and-watch” approach with removal when the child is old enough to have a local anaesthetic. Cosmesis Parents and older patients are often concerned about the cosmetic appearance of the nevus. This is especially so for the large and large-medium nevi as well as ones located in such critical areas as the face, regardless of size. Larger nevi are often difficult to remove and may require staged excision, grafting or skin expanders. Despite these techniques, some of the very large nevi cannot be entirely removed. Many of the nevi that are successfully removed leave heavy scarring. If some excised nevus cells are left behind, the scar may repigment which can be cosmetically unappealing. Managing CMN Large CMN often require a team approach that includes dermatologists, plastic surgeons and pediatricians. Neurologists, oncologists and psychiatrists may need to get involved in certain cases. Parents need to be made aware of the issues and participate in decisions. Whether or not an MRI should be done early on can be a difficult decision. Clearly if a patient is symptomatic then imaging should be performed. A patient whose LCMN is located over the posterior axis or who has many satellite lesions should be considered strongly for MRI. It should be made clear, however, if the MRI shows melanosis and the patient is asymptomatic, then the results may not help direct management and may add significant distress for the family. However, if the child remains asymptomatic, then usually the family can be cautiously optimistic about outcome. It is unclear if and when a follow-up MRI should be done if an initial MRI suggests NCM in an asymptomatic child. Timing of an MRI may also be an issue. Some authors suggest that over time it may be harder to detect neurocutaneous melanosis. It is unclear whether this happens because melanin decreases or the myelin obscures the melanosis.19,20 Some clinicians suggest MRIs as early as possible to pick up possible changes. A child whose CMN is on an extremity or who has few or no satellite nevi may not require imaging. Also, no evidence suggests that patients with small or intermediate nevi face increased risk of NCM. They do not require routine MRIs. The next major issue is how to manage the risk of melanoma within the LCMN. Many authors suggest early excision within the first year of life.21,22 The goal is to remove the at-risk lesion as early as possible because melanomas within LCMN tend to occur at a young age. For those lesions that cannot be completely excised, many surgeons will still debulk the lesion believing that it will still help in reducing the risk. There is no clear data to support this contention. Other centers use clinical observation until children are pre-school age, and then provide surgical intervention. Clinical observation can be a challenge, and repeat photographs and reassessments are helpful. Any suspicious change or nodules should be biopsied. Children who have undergone complete excision should also have follow-up for re-pigmentation of the scar, new satellites and melanomas that may occur elsewhere. For small and small-intermediate nevi, it is reasonable to wait until the child can undergo local anaesthesia for surgical removal. Melanoma has not been reported in childhood in these lesions so waiting until late childhood or adolescence shouldn’t be problematic. For a small nevus in a cosmetically-sensitive area, the decision may be made for removal prior to school entry. Other treatment options include cosmetic cover-up, curettage and laser therapy. Laser therapy may be of some benefit in selected patients, but the lesions can repigment. It is presumed that lightening of the nevus does not necessarily alter the risk of melanoma. There is also the unproven concern that laser therapy may alter the biology of the nevus cells and in some way induce cellular proliferation. A Round-Up of the Facts LCMN can be associated with risks of melanoma and neurocutaneous melanosis. The risks for melanoma include: large size of the congenital nevus (>20 cm); multiple satellite nevi and a posterior axis location. Smaller nevi likely have a much smaller risk. Neurocutaneous melanosis may accompany large CMN and again, multiple satellite nevi and posterior axis are risk factors for its development. When NCM manifests symptomatically, the outcome is very poor. Asymptomatic NCM appears to have a better outcome. Management of LCMN includes excision, debulking procedures and careful lifelong follow-up. Timing of removal should take into account the surgical complexity, presence of NCM and likelihood of complete excision. Intermediate and smaller nevi can be followed and removed when the child reaches puberty. Back to the case . . . The family was reassured that the nevus would unlikely achieve an adult size >20 cm. The risk of melanoma was discussed as being extremely low. An MRI showed no underlying neurocutaneous melanosis as would be expected with a single medium CMN. The family was counselled that the nevus could be removed by serial excision when the child was older and closer to puberty. Digital photos were taken, and the patient will be followed yearly.
Congenital Melanocytic Nevi
Case Presentation LM, a healthy 2-year-old boy, presented with a nevus noted at birth. The nevus has a darker central area and measures 4 cm by 4 cm (see below). It’s located over the occipital area to the left of midline. LM’s parents have heard from friends that there’s a chance this will turn into skin cancer. The family is referred to a dermatologist who is concerned that the posterior location may indicate potential associated problems. C ongenital melanocytic nevi (CMN) are benign proliferations of nevus cells present from birth. They vary considerably in morphology with respect to color, shape, size and border regularity (see photos throughout). On histopathology, the nevus cells tend to be deep in the dermis and may be present within dermal appendages such as hair follicles and nerves. This is in contrast to acquired nevi in which the nevus cells are typically more superficial. So-called “tardive” congenital nevi are comparable to congenital melanocytic nevi in their clinical and pathologic features but are not present at birth. They may appear up to about 2 years of age. Overall, CMN occur in about 1% of children. CMN can generate concern amongst parents and physicians regarding the risk of malignancy and cosmesis. A third concern, of which dermatologists should be aware, is the issue of neurocutaneous melanosis (NCM), which can accompany certain congenital melanocytic nevi. These three concerns will be reviewed in this article. The risks of melanoma or NCM and the ease of cosmetic repair are often approached by considering the size of the nevus. The system most commonly used to categorize the size of a CMN is based on the predicted final adult size of the lesion as follows: • <1.5 cm is a small CMN • 1.5 cm to 19.9 cm is a medium or intermediate nevus • >20cm is considered large (LCMN).1 A rough estimate suggests a 9-cm CMN on the head or 6-cm CMN on the body in childhood will achieve an adult size of >20 cm.2 Ruiz-Moldanado3 has recently suggested a further modification to the “small, medium, large” system that would allow for more groupings. If adopted, this classification may enable better prognostication by narrowing the range of sizes incorporated in any one group. Some patients with large CMN also have multiple satellite lesions. These are smaller melanocytic nevi that often accompany a LCMN, and they may or may not be numerous. Their presence may also have some predictive value. Additional nevi can also appear after birth. Neurocutaneous Melanosis Nevus cells in the skin and elements of the central nervous system share a common origin from the neural crest. Dysregulated growth of melanoblasts can lead to melanocytic proliferations in the central nervous system as well as the skin. This phenomenon is called neurocutaneous melanosis. It has been more formally defined by Kadonaga and Frieden as melanocytic proliferations in the central nervous system in association with one LCMN or at least three smaller CMN.2 These proliferations are usually benign but can become malignant. NCM can be asymptomatic or manifest with neurologic symptoms. Manifest or symptomatic NCM often presents in the first few years.4 Symptoms can include hydrocephalus, seizures, increased head circumference or developmental delay. Manifest NCM typically has a very poor prognosis with a very high mortality rate.4 The importance of establishing the diagnosis of symptomatic NCM is in its ability to help direct management. A decision to undergo complicated surgical excision of a LCNM may be withheld if the overall prognosis for the patient is extremely poor. Asymptomatic NCM is a diagnosis made by imaging or autopsy as the patient will not clinically manifest the neurological signs. It is unclear how prevalent this problem is because not all patients with CMN undergo neuroimaging. Frieden et al found that 30% of patients with a LCMN had asymptomatic NCM.5 Routine MRIs will likely pick up an increasing number of cases of asymptomatic NCM in the context of a LCMN or multiple smaller CMN. However, this approach will also raise new issues: • What follow-up should these patients receive? • Should they have repeat imaging, and if so, when? • Does the information gained justify the risk of anesthesia for the MRI? Some might argue that a normal MRI might be very reassuring to parents. It might also be argued that knowing in advance that the CNS is involved would allow doctors and parents to closely monitor for symptomatic neurologic involvement. However, most cases are found incidentally, and outcomes appear to be good. Are certain patients more at risk for developing NCM? Certain features do increase the likelihood. Nevus size is a risk factor as most reported cases occur in patients with large or multiple smaller congenital nevi. There is no evidence that single small CMN (the majority of congenital nevi) have any association with NCM. In respect to large CMN, one study showed a 5-year cumulative risk of NCM of 2.5%.4 The risk may be higher in those with lesions >50 cm. Location also appears to be a risk factor. In most reported cases, the LCMN has been in a posterior axial location.6 This holds true for the cases of asymptomatic or symptomatic NCM. Number-One Predictor of NCM The single factor that is the most important predictive factor for NCM is the number of satellite lesions.7 Patients with NCM had significantly more satellite lesions (68.5) versus patients without NCM (18 satellites). Marghoob showed that having >20 satellites was associated with a 5-fold increased risk of NCM.7 Another association to consider in patients with a CMN over the lumbosacral area in the midline is spinal cord/column abnormalities. Skin lesions in this location may be a marker for deeper abnormalities.8 An MRI can rule out this association. Risk of Melanoma Overall, melanoma is very rare in childhood. One of the known risk factors is a large congenital nevus. The incidence of melanoma has been reported to be between 3.3% and 6.0% 9,10,11 in several prospective and retrospective studies of melanomas arising within large congenital nevi. A systemic review of published data found that overall 2.8% of patients with LCMN develop melanoma.12 Melanoma has been reported to occur in the actual CMN as well as other sites. DeDavid found 12% of patients with LCMN had primary melanoma within a CMN, and all patients had their CMN in an axial location.13 Patients with CMN on the extremities did not develop melanoma. Melanomas have also occurred at cutaneous sites remote from the CMN. It is difficult to know if these melanomas are de novo or whether the patients were at higher risk by virtue of their CMN. No study has found melanomas occurring in satellite lesions. Noncutaneous melanoma has been reported as well. It has occurred in the context of NCM (malignant neurocutaneous melanosis). One study found 21 of 289 patients with NCM had malignant growths. A review of the NYU-LCMN registry in 2000 found eight melanomas of which 3/8 occurred in LCMN; 1/8 occurred in normal skin, 1/8 were in the retroperitoneum and 2/8 were in the CNS. The origin of one melanoma was unknown. At least half of all melanomas that arise within CMN occur at an early age, often before age 5.13 This knowledge has led to relatively aggressive attempts by some centers to remove the nevus in infancy in an attempt to prevent melanoma development. CNS melanoma also tends to occur at an early age. Management of CMN is very difficult, and one often needs to weigh aggressive surgery against watchful waiting in the face of a small, but real, risk of melanoma. Predictive factors to isolate which patients most likely to develop melanoma would be very useful. As with the risk of NCM, the axial location and larger number of satellites are indicators of increased risk of melanoma. Bittencourt et al found that in eight patients with melanoma, all had more than 20 satellites.4 There are several reasons that melanoma can be difficult to detect in CMN. These lesions often have variegated texture and color, and subtle change is difficult to detect. The depth of the lesion can also obscure a subtle change. Just as the benign nevus cells can be quite deep in a CMN so can the malignant cells. This may lead to a deep nodule that is not easy to detect until it is large and palpable. CMN that occur in the scalp may be difficult to observe and follow. Finally, benign nodular growths that simulate melanoma can arise within CMN. These often regress. Pathologic distinction can often be made between proliferative nodules and melanoma. One study showed that a PET scan was useful in identifying malignant nodules.14 This may be a useful tool in the future. Evaluating the Size of CMN in Relation to Melanoma Risk Many dermatologists question the risk of melanoma in small- and medium-sized nevi as they are more commonly encountered in clinical practice. Swerdlow11 examined 232 medium CMN and found no melanomas. Sahin et al, in a short-term follow-up, found no melanomas arising in medium CMN.15 There were however three patients (all over age 35) who developed melanoma at other cutaneous sites. There have also been reports of melanoma in adulthood arising from a CMN.16,17 It’s worth considering the morphology of the smaller CNM in decision-making. It may be much harder to observe a dark, irregular, nodular CMN, and earlier excision may be favored. There is clearly not enough prospective data to definitively assess the risk of melanoma in small and small-medium congenital nevi. The risk appears very low and has not been reported to occur prior to puberty.16,18 This would support a “wait-and-watch” approach with removal when the child is old enough to have a local anaesthetic. Cosmesis Parents and older patients are often concerned about the cosmetic appearance of the nevus. This is especially so for the large and large-medium nevi as well as ones located in such critical areas as the face, regardless of size. Larger nevi are often difficult to remove and may require staged excision, grafting or skin expanders. Despite these techniques, some of the very large nevi cannot be entirely removed. Many of the nevi that are successfully removed leave heavy scarring. If some excised nevus cells are left behind, the scar may repigment which can be cosmetically unappealing. Managing CMN Large CMN often require a team approach that includes dermatologists, plastic surgeons and pediatricians. Neurologists, oncologists and psychiatrists may need to get involved in certain cases. Parents need to be made aware of the issues and participate in decisions. Whether or not an MRI should be done early on can be a difficult decision. Clearly if a patient is symptomatic then imaging should be performed. A patient whose LCMN is located over the posterior axis or who has many satellite lesions should be considered strongly for MRI. It should be made clear, however, if the MRI shows melanosis and the patient is asymptomatic, then the results may not help direct management and may add significant distress for the family. However, if the child remains asymptomatic, then usually the family can be cautiously optimistic about outcome. It is unclear if and when a follow-up MRI should be done if an initial MRI suggests NCM in an asymptomatic child. Timing of an MRI may also be an issue. Some authors suggest that over time it may be harder to detect neurocutaneous melanosis. It is unclear whether this happens because melanin decreases or the myelin obscures the melanosis.19,20 Some clinicians suggest MRIs as early as possible to pick up possible changes. A child whose CMN is on an extremity or who has few or no satellite nevi may not require imaging. Also, no evidence suggests that patients with small or intermediate nevi face increased risk of NCM. They do not require routine MRIs. The next major issue is how to manage the risk of melanoma within the LCMN. Many authors suggest early excision within the first year of life.21,22 The goal is to remove the at-risk lesion as early as possible because melanomas within LCMN tend to occur at a young age. For those lesions that cannot be completely excised, many surgeons will still debulk the lesion believing that it will still help in reducing the risk. There is no clear data to support this contention. Other centers use clinical observation until children are pre-school age, and then provide surgical intervention. Clinical observation can be a challenge, and repeat photographs and reassessments are helpful. Any suspicious change or nodules should be biopsied. Children who have undergone complete excision should also have follow-up for re-pigmentation of the scar, new satellites and melanomas that may occur elsewhere. For small and small-intermediate nevi, it is reasonable to wait until the child can undergo local anaesthesia for surgical removal. Melanoma has not been reported in childhood in these lesions so waiting until late childhood or adolescence shouldn’t be problematic. For a small nevus in a cosmetically-sensitive area, the decision may be made for removal prior to school entry. Other treatment options include cosmetic cover-up, curettage and laser therapy. Laser therapy may be of some benefit in selected patients, but the lesions can repigment. It is presumed that lightening of the nevus does not necessarily alter the risk of melanoma. There is also the unproven concern that laser therapy may alter the biology of the nevus cells and in some way induce cellular proliferation. A Round-Up of the Facts LCMN can be associated with risks of melanoma and neurocutaneous melanosis. The risks for melanoma include: large size of the congenital nevus (>20 cm); multiple satellite nevi and a posterior axis location. Smaller nevi likely have a much smaller risk. Neurocutaneous melanosis may accompany large CMN and again, multiple satellite nevi and posterior axis are risk factors for its development. When NCM manifests symptomatically, the outcome is very poor. Asymptomatic NCM appears to have a better outcome. Management of LCMN includes excision, debulking procedures and careful lifelong follow-up. Timing of removal should take into account the surgical complexity, presence of NCM and likelihood of complete excision. Intermediate and smaller nevi can be followed and removed when the child reaches puberty. Back to the case . . . The family was reassured that the nevus would unlikely achieve an adult size >20 cm. The risk of melanoma was discussed as being extremely low. An MRI showed no underlying neurocutaneous melanosis as would be expected with a single medium CMN. The family was counselled that the nevus could be removed by serial excision when the child was older and closer to puberty. Digital photos were taken, and the patient will be followed yearly.
Case Presentation LM, a healthy 2-year-old boy, presented with a nevus noted at birth. The nevus has a darker central area and measures 4 cm by 4 cm (see below). It’s located over the occipital area to the left of midline. LM’s parents have heard from friends that there’s a chance this will turn into skin cancer. The family is referred to a dermatologist who is concerned that the posterior location may indicate potential associated problems. C ongenital melanocytic nevi (CMN) are benign proliferations of nevus cells present from birth. They vary considerably in morphology with respect to color, shape, size and border regularity (see photos throughout). On histopathology, the nevus cells tend to be deep in the dermis and may be present within dermal appendages such as hair follicles and nerves. This is in contrast to acquired nevi in which the nevus cells are typically more superficial. So-called “tardive” congenital nevi are comparable to congenital melanocytic nevi in their clinical and pathologic features but are not present at birth. They may appear up to about 2 years of age. Overall, CMN occur in about 1% of children. CMN can generate concern amongst parents and physicians regarding the risk of malignancy and cosmesis. A third concern, of which dermatologists should be aware, is the issue of neurocutaneous melanosis (NCM), which can accompany certain congenital melanocytic nevi. These three concerns will be reviewed in this article. The risks of melanoma or NCM and the ease of cosmetic repair are often approached by considering the size of the nevus. The system most commonly used to categorize the size of a CMN is based on the predicted final adult size of the lesion as follows: • <1.5 cm is a small CMN • 1.5 cm to 19.9 cm is a medium or intermediate nevus • >20cm is considered large (LCMN).1 A rough estimate suggests a 9-cm CMN on the head or 6-cm CMN on the body in childhood will achieve an adult size of >20 cm.2 Ruiz-Moldanado3 has recently suggested a further modification to the “small, medium, large” system that would allow for more groupings. If adopted, this classification may enable better prognostication by narrowing the range of sizes incorporated in any one group. Some patients with large CMN also have multiple satellite lesions. These are smaller melanocytic nevi that often accompany a LCMN, and they may or may not be numerous. Their presence may also have some predictive value. Additional nevi can also appear after birth. Neurocutaneous Melanosis Nevus cells in the skin and elements of the central nervous system share a common origin from the neural crest. Dysregulated growth of melanoblasts can lead to melanocytic proliferations in the central nervous system as well as the skin. This phenomenon is called neurocutaneous melanosis. It has been more formally defined by Kadonaga and Frieden as melanocytic proliferations in the central nervous system in association with one LCMN or at least three smaller CMN.2 These proliferations are usually benign but can become malignant. NCM can be asymptomatic or manifest with neurologic symptoms. Manifest or symptomatic NCM often presents in the first few years.4 Symptoms can include hydrocephalus, seizures, increased head circumference or developmental delay. Manifest NCM typically has a very poor prognosis with a very high mortality rate.4 The importance of establishing the diagnosis of symptomatic NCM is in its ability to help direct management. A decision to undergo complicated surgical excision of a LCNM may be withheld if the overall prognosis for the patient is extremely poor. Asymptomatic NCM is a diagnosis made by imaging or autopsy as the patient will not clinically manifest the neurological signs. It is unclear how prevalent this problem is because not all patients with CMN undergo neuroimaging. Frieden et al found that 30% of patients with a LCMN had asymptomatic NCM.5 Routine MRIs will likely pick up an increasing number of cases of asymptomatic NCM in the context of a LCMN or multiple smaller CMN. However, this approach will also raise new issues: • What follow-up should these patients receive? • Should they have repeat imaging, and if so, when? • Does the information gained justify the risk of anesthesia for the MRI? Some might argue that a normal MRI might be very reassuring to parents. It might also be argued that knowing in advance that the CNS is involved would allow doctors and parents to closely monitor for symptomatic neurologic involvement. However, most cases are found incidentally, and outcomes appear to be good. Are certain patients more at risk for developing NCM? Certain features do increase the likelihood. Nevus size is a risk factor as most reported cases occur in patients with large or multiple smaller congenital nevi. There is no evidence that single small CMN (the majority of congenital nevi) have any association with NCM. In respect to large CMN, one study showed a 5-year cumulative risk of NCM of 2.5%.4 The risk may be higher in those with lesions >50 cm. Location also appears to be a risk factor. In most reported cases, the LCMN has been in a posterior axial location.6 This holds true for the cases of asymptomatic or symptomatic NCM. Number-One Predictor of NCM The single factor that is the most important predictive factor for NCM is the number of satellite lesions.7 Patients with NCM had significantly more satellite lesions (68.5) versus patients without NCM (18 satellites). Marghoob showed that having >20 satellites was associated with a 5-fold increased risk of NCM.7 Another association to consider in patients with a CMN over the lumbosacral area in the midline is spinal cord/column abnormalities. Skin lesions in this location may be a marker for deeper abnormalities.8 An MRI can rule out this association. Risk of Melanoma Overall, melanoma is very rare in childhood. One of the known risk factors is a large congenital nevus. The incidence of melanoma has been reported to be between 3.3% and 6.0% 9,10,11 in several prospective and retrospective studies of melanomas arising within large congenital nevi. A systemic review of published data found that overall 2.8% of patients with LCMN develop melanoma.12 Melanoma has been reported to occur in the actual CMN as well as other sites. DeDavid found 12% of patients with LCMN had primary melanoma within a CMN, and all patients had their CMN in an axial location.13 Patients with CMN on the extremities did not develop melanoma. Melanomas have also occurred at cutaneous sites remote from the CMN. It is difficult to know if these melanomas are de novo or whether the patients were at higher risk by virtue of their CMN. No study has found melanomas occurring in satellite lesions. Noncutaneous melanoma has been reported as well. It has occurred in the context of NCM (malignant neurocutaneous melanosis). One study found 21 of 289 patients with NCM had malignant growths. A review of the NYU-LCMN registry in 2000 found eight melanomas of which 3/8 occurred in LCMN; 1/8 occurred in normal skin, 1/8 were in the retroperitoneum and 2/8 were in the CNS. The origin of one melanoma was unknown. At least half of all melanomas that arise within CMN occur at an early age, often before age 5.13 This knowledge has led to relatively aggressive attempts by some centers to remove the nevus in infancy in an attempt to prevent melanoma development. CNS melanoma also tends to occur at an early age. Management of CMN is very difficult, and one often needs to weigh aggressive surgery against watchful waiting in the face of a small, but real, risk of melanoma. Predictive factors to isolate which patients most likely to develop melanoma would be very useful. As with the risk of NCM, the axial location and larger number of satellites are indicators of increased risk of melanoma. Bittencourt et al found that in eight patients with melanoma, all had more than 20 satellites.4 There are several reasons that melanoma can be difficult to detect in CMN. These lesions often have variegated texture and color, and subtle change is difficult to detect. The depth of the lesion can also obscure a subtle change. Just as the benign nevus cells can be quite deep in a CMN so can the malignant cells. This may lead to a deep nodule that is not easy to detect until it is large and palpable. CMN that occur in the scalp may be difficult to observe and follow. Finally, benign nodular growths that simulate melanoma can arise within CMN. These often regress. Pathologic distinction can often be made between proliferative nodules and melanoma. One study showed that a PET scan was useful in identifying malignant nodules.14 This may be a useful tool in the future. Evaluating the Size of CMN in Relation to Melanoma Risk Many dermatologists question the risk of melanoma in small- and medium-sized nevi as they are more commonly encountered in clinical practice. Swerdlow11 examined 232 medium CMN and found no melanomas. Sahin et al, in a short-term follow-up, found no melanomas arising in medium CMN.15 There were however three patients (all over age 35) who developed melanoma at other cutaneous sites. There have also been reports of melanoma in adulthood arising from a CMN.16,17 It’s worth considering the morphology of the smaller CNM in decision-making. It may be much harder to observe a dark, irregular, nodular CMN, and earlier excision may be favored. There is clearly not enough prospective data to definitively assess the risk of melanoma in small and small-medium congenital nevi. The risk appears very low and has not been reported to occur prior to puberty.16,18 This would support a “wait-and-watch” approach with removal when the child is old enough to have a local anaesthetic. Cosmesis Parents and older patients are often concerned about the cosmetic appearance of the nevus. This is especially so for the large and large-medium nevi as well as ones located in such critical areas as the face, regardless of size. Larger nevi are often difficult to remove and may require staged excision, grafting or skin expanders. Despite these techniques, some of the very large nevi cannot be entirely removed. Many of the nevi that are successfully removed leave heavy scarring. If some excised nevus cells are left behind, the scar may repigment which can be cosmetically unappealing. Managing CMN Large CMN often require a team approach that includes dermatologists, plastic surgeons and pediatricians. Neurologists, oncologists and psychiatrists may need to get involved in certain cases. Parents need to be made aware of the issues and participate in decisions. Whether or not an MRI should be done early on can be a difficult decision. Clearly if a patient is symptomatic then imaging should be performed. A patient whose LCMN is located over the posterior axis or who has many satellite lesions should be considered strongly for MRI. It should be made clear, however, if the MRI shows melanosis and the patient is asymptomatic, then the results may not help direct management and may add significant distress for the family. However, if the child remains asymptomatic, then usually the family can be cautiously optimistic about outcome. It is unclear if and when a follow-up MRI should be done if an initial MRI suggests NCM in an asymptomatic child. Timing of an MRI may also be an issue. Some authors suggest that over time it may be harder to detect neurocutaneous melanosis. It is unclear whether this happens because melanin decreases or the myelin obscures the melanosis.19,20 Some clinicians suggest MRIs as early as possible to pick up possible changes. A child whose CMN is on an extremity or who has few or no satellite nevi may not require imaging. Also, no evidence suggests that patients with small or intermediate nevi face increased risk of NCM. They do not require routine MRIs. The next major issue is how to manage the risk of melanoma within the LCMN. Many authors suggest early excision within the first year of life.21,22 The goal is to remove the at-risk lesion as early as possible because melanomas within LCMN tend to occur at a young age. For those lesions that cannot be completely excised, many surgeons will still debulk the lesion believing that it will still help in reducing the risk. There is no clear data to support this contention. Other centers use clinical observation until children are pre-school age, and then provide surgical intervention. Clinical observation can be a challenge, and repeat photographs and reassessments are helpful. Any suspicious change or nodules should be biopsied. Children who have undergone complete excision should also have follow-up for re-pigmentation of the scar, new satellites and melanomas that may occur elsewhere. For small and small-intermediate nevi, it is reasonable to wait until the child can undergo local anaesthesia for surgical removal. Melanoma has not been reported in childhood in these lesions so waiting until late childhood or adolescence shouldn’t be problematic. For a small nevus in a cosmetically-sensitive area, the decision may be made for removal prior to school entry. Other treatment options include cosmetic cover-up, curettage and laser therapy. Laser therapy may be of some benefit in selected patients, but the lesions can repigment. It is presumed that lightening of the nevus does not necessarily alter the risk of melanoma. There is also the unproven concern that laser therapy may alter the biology of the nevus cells and in some way induce cellular proliferation. A Round-Up of the Facts LCMN can be associated with risks of melanoma and neurocutaneous melanosis. The risks for melanoma include: large size of the congenital nevus (>20 cm); multiple satellite nevi and a posterior axis location. Smaller nevi likely have a much smaller risk. Neurocutaneous melanosis may accompany large CMN and again, multiple satellite nevi and posterior axis are risk factors for its development. When NCM manifests symptomatically, the outcome is very poor. Asymptomatic NCM appears to have a better outcome. Management of LCMN includes excision, debulking procedures and careful lifelong follow-up. Timing of removal should take into account the surgical complexity, presence of NCM and likelihood of complete excision. Intermediate and smaller nevi can be followed and removed when the child reaches puberty. Back to the case . . . The family was reassured that the nevus would unlikely achieve an adult size >20 cm. The risk of melanoma was discussed as being extremely low. An MRI showed no underlying neurocutaneous melanosis as would be expected with a single medium CMN. The family was counselled that the nevus could be removed by serial excision when the child was older and closer to puberty. Digital photos were taken, and the patient will be followed yearly.
