Distinct subgroups of pediatric melanocytic lesions, which were identified by an integrated clinicopathologic and genomic analysis, were found to have different clinical behaviors, suggesting that this combined diagnostic modality could inform individualized diagnoses and treatments for patients with these rare malignancies.
Distinct subgroups of pediatric melanocytic lesions, which were identified by an integrated clinicopathologic and genomic analysis, were found to have different clinical behaviors, suggesting that this combined diagnostic modality could inform individualized diagnoses and treatments for patients with these rare malignancies, according to findings from a registry analysis published in Cancer.1
“What we're seeing from the registry’s data is that these are incredibly diverse tumors,” said senior author Armita Bahrami, MD, an acting professor in the Department of Pathology and Laboratory Medicine at Emory University School of Medicine and a member of the Cell and Molecular Biology research program at Winship Cancer Institute of Emory University.2 “Some patients need drastic measures. For many others, masterful watching is the best option. It is critical for the care team to consider all available clinical, pathologic and genomic data in order to assign patients to the appropriate risk category.”
Pediatric melanoma is a rare disease that has a diagnostic and therapeutic challenge in that the natural history of these tumors is not well defined with available data. Moreover, diagnosing pediatric melanoma requires an integrated approach to optimize therapeutic management.
To characterize clinicopathologic features of atypical pediatric melanocytic lesions and validate predefined disease subgroups, investigators developed the Molecular Analysis of Childhood Melanocytic Tumors (MACMEL) registry that prospectively enrolled pediatric and adolescent patients with melanocytic lesions.
“What is different about the MACMEL registry is that it is prospective,” said lead study author Alberto Pappo, MD, director of the Solid Tumor Division, co-leader of the Development Biology and Solid Tumor Program, and the Alvin Mauer Endowed Chair at St. Jude Children’s Research Hospital.2 “We’re seeing the vast majority of enrolled patients as part of the melanoma clinic at St. Jude. We can follow these patients and conduct detailed pathology and molecular analysis.”
Eligible patients were under the age of 19 years at initial diagnosis and had centrally reviewed, confirmed diagnoses of atypical Spitz tumor/Spitz melanoma (AST/SM; n = 37), adult-type melanoma (CM; 17), melanoma arising in a giant congenital nevus (MCM; n = 4), or atypical melanocytic proliferation of other types (OT; n = 12).
Select tumors underwent fluorescence in situ hybridization (FISH) testing for CDKN2A deletions, as well as BRAF, ALK, ROS1, and NTRK gene rearrangements. Additionally, RNA sequencing was performed on available tumors. Targeted capture next-generation sequencing (NGS) was conducted on selected samples using a custom, 301 cancer-associated gene panel.
Overall, patients (n = 70) were a median age of 10.5 years at enrollment (range, 0.0-18.0) and the majority were White (n = 64; 91.4%) females (n = 39; 55.7%). Patients were a median age of 9.5 years at diagnosis (range, 0.0-18.0). The majority of patients were previously treated (n = 41; 58.6%) compared with those with newly diagnosed disease (n = 29; 41.4%).
American Joint Committee on Cancer Staging 8th edition guidelines defined patients as having stage 0 (n = 1; 1.4%), stage IA (n = 4; 5.7%), stage IB (n = 8; 11.4%), stage IIA (n = 12; 17.2%), stage IIB (n = 4; 5.7%), stage IIC (n = 4; 5.7%), stage IIIA (n = 7; 10%), stage IIIB (n = 9; 12.8%), stage IIIC (n = 11; 15.7%), stage IIID (n = 2; 2.9%), or stage IV (n = 3; 4.3%) disease. Five patients (7.2%) had missing staging data.
Sites of primary tumors included acral (palmar/plantar; n = 2; 2.9%), central nervous system (n = 2; 2.9%), ear (n = 3; 4.3%), face (n = 6; 8.6%), head (n = 1; 1.4%), neck (n = 3; 4.3%), scalp (n = 8; 11.5%), trunk (n = 18; 25.7%), lower extremity (n = 11; 15.7%), and upper extremity (n = 16; 22.7%).
The median follow-up time from enrollment was 2.13 years (range, 0.17-4.50) and most patients (n = 57; 81.43%) were alive and free of disease. Ten percent of patients (n = 7) were alive with persistent disease or relapse and 8.57% of patients (n = 6) were deceased.
In the AST/SM cohort, TERT promoter analysis did not identify any mutations in the 33 patients who underwent testing. Copy aberrations were identified by FISH in 16 of 20 patients tested.
In 19 patients, NGS testing done with whole genome sequencing (n = 1) identified a BRIP1 Q685* mutation. Transcriptome analysis (n = 16) identified MAP3K8 fusions (n = 6), ALK fusions (n = 4), and NTRK1 fusions (n = 1), as well as FLOT2-ARAF and GPNMB-ERBB2 mutations. Whole exome sequencing (n = 10) identified 1 patient with a RAC N92K mutation and 1 patient with co-occurring ATM and SDHAF2 mutations.
In the CM group, 7 of 16 patients had positive TERT promoter mutation analysis results. Of all patients, 11 harbored a BRAF V600E mutation by immunohistochemistry, targeted capture, or exome sequencing. Transcriptome sequencing (n = 4) and targeted sequencing analysis (n = 3) identified NRAS, ATM, CTNNB1, MAP2K1, and JAK3 mutations, as well as a novel ZNF38-MARS rearrangement.
In the MCM group, all lesions analyzed harbored a detectable NRASQ61 mutation by targeted capture NGS, and other genomic findings included MC1R, MAP2K1, and TP53 mutations.
In the OT group, 6 patients had pigment synthesizing melanoma/pigmented epithelioid melanocytoma, 2 had nevoid melanoma, 2 had atypical melanocytic proliferation in an acquired nevus, 1 had congenital nevus with a blue nevus and peripheral nerve sheath differentiation, 1 had atypical compound spindle cell melanocytic tumor in an acquired nevus, and 1 had atypical melanocytic proliferation not otherwise specified. Notable genomic findings in this subgroup included MLL3 rearrangements (n = 1) and BRAF V600E mutations (n = 4).
“Now that we have all this information available, we can be significantly more selective and integrate this information into treatment decision making. It will help us provide the best information possible for the patient as far as prognosis, recurrence and the need for additional surgery or therapy,” Pappo concluded.2