Six patients under the age of 12 were included in the study, but these patients did not undergo sentinel lymph node biopsy. All survived, but there were several late recurrences – after 5 years – even in the node-negative patients, she said.
When the investigators looked at recurrence-free survival, they found that node-positive patients had significantly worse recurrence-free survival and melanoma-specific survival than that of node-negative patients (60% vs.94% and 78% vs. 97%, respectively) at a median follow-up of 5 years (Ann. Surg. Oncol. 2012;19:3888-95).
A study by investigators at the University of California, San Francisco, focused more on historical data, finding that many of the 70 patients included in the study had putative risk factors for melanoma. For example, 20% had numerous nevi, 27% had a positive family history, and 25% had a history of sunburn. Also, this study was the only one of the three to address the presence of LCMN, and only three patients had melanoma arising in a nevus, providing further evidence of a low risk of melanoma in LCMN.
The diagnosis of pediatric melanoma was delayed by about a year in more than 60% of the patients.
The investigators noted that primary lesion characteristics differed from those seen in adults, and they concluded that the conventional ABCDE criteria used to help in the diagnosis of melanoma did not capture melanoma in about 60% of the childhood cases and 40% of the adolescent cases.
Lesions in this study were much more likely to be amelanotic in children, and of uniform color in adolescents. Lesional evolution was nearly universal, and bleeding, bumps, variable diameter, and de novo development were common.
On histopathology, a majority of tumors were not superficial spreading type; more were unclassified spitzoid and other histopathologic subtypes, she noted.
Ten patients (14%) died from their melanoma, and of these, 7 had amelanotic melanoma. Only 1 patient under age 10 years died, and that was in the setting of a large congenital melanocytic nevus.
Based on their findings, the investigators suggested pediatric-specific ABCD criteria (A = amelanotic, B = bleeding, bumps, C = color uniformity, and D = de novo and any diameter) to be used along with the conventional ABCDE criteria to facilitate earlier recognition and treatment of pediatric melanoma (J. Am. Acad. Dermatol. 2013 [doi:10.1016/j.jaad.2012.12.953]).
Although there are some conflicting findings in these three studies – including differing conclusions with respect to the value of sentinel lymph node biopsy for predicting outcomes – there also are some consistent findings, Dr. Maguiness said.
Prepubertal melanoma tends to involve thicker tumors, and darker skin types are overrepresented. Also, lesions in all ages in the pediatric population tend to be amelanotic with spitzoid histology, and tend to have higher rates of positive sentinel lymph node biopsies, compared with adult cases. Prepubertal cases have the highest rates of node positivity, she said.
"So, in conclusion, the risk of malignant melanoma within large congenital nevi seems to be lower than we thought, and the diagnosis of malignant melanoma of childhood has excellent prognosis – speaking to the unique natural history and biology of these tumors, which we probably don’t fully understand," she said, adding that adolescent presentations of melanoma seem to be similar to those in adults, with a slightly better overall prognosis.
Dr. Maguiness reported having no disclosures.