Back to Authors

Panneerselvam, A.

Mayo Clinic

Mayo Clinic

Citations 1

First analysis of an international pediatric melanoma and atypical melanocytic neoplasm database
Averbook, Bruce J., Jukic, D.M., Delman, Keith A. , Pittelkow, Mark R. , Rao, Jia‐Sheng , Panneerselvam, A., Zager, Jonathan S. , Sabel, Michael S. , Swetter, Susan M. , Kirkwood, John M.

Journal of Clinical Oncology | August 25, 2020 | Volume: 27 | Issue: 15_suppl | Pages: 9013-9013

10.1200/jco.2009.27.15_suppl.9013 Get It

9013 Background: Pediatric melanoma (PM) care has been extrapolated from adult melanoma data. PM and atypical melanocytic neoplasms (AMNs) appear to have different biology. An international database (DB) was developed to clarify their behavior. Methods: IRB approval was obtained at 12 institutions. An SQL-DB was developed for web entry of de-identified demographic and pathologic data for PM and AMN patients (pts) < 21yr through an honest broker system at the University of Pittsburgh. Institutions retained a key of pts entered with assigned numbers for quality assurance and updates. Statistical analysis used Kaplan-Meier survival curves, univariate linear trends and log rank tests. In situ melanoma was excluded from PM survival analysis. Results: 828 pts were registered as of 31 Oct 2008 (ages 11mo-23; median 15yr). 34 pts 21–23yr entered were left in the DB for statistical comparison. Diagnosis years ranged from 1936–2008. 455 pts had complete follow-up. Too few AMN pts had complete follow-up for analysis (18/208). After excluding 32 in situ and 40 with other incomplete data, 365 PM out of 415 total PM were evaluable for OS and 351 for DFS (Stage IV removed). Mean/median age for evaluable PM pts was 16.44/17-yrs (range 1–21yr). 591 pts were age 10–20 while 203 were < 10. Sentinel lymph node (SLN) biopsy showed spread in 30.1 % PM pts (compared to 50% [4/8] of AMN SLN pts). 10-yr PM OS was 80.6%, and pts 0–10 yr had 100% 10-yr OS compared to 69.6% for pts age 10–15 and 79.49% for age 15–20 (p= 0.1473). OS did not differ significantly by gender. Stage predicted OS (p<0.0001). 10-yr OS was 94.13% for Stage I (n=174), 79.62% for stage II (n=67) & 77.14% for stage III (n=75). Thickness affected 10-yr OS: 0–1mm=97% (n=147), 1.1–2mm 70% (n=84), 2.01–4mm 78% (n=71) & >4mm 81% (n=25), p= 0.0099. Survival was similar for pts with PM > 1mm of the several T stage groupings. Ulceration adversely affected OS (p=0.022). Mitosis, defined as present/absent did not alter survival. Nodal metastasis correlated with worse OS (p= 0.170). Conclusions: Stage, thickness, ulceration, and nodal status are significant predictors of OS for PM. Further study will focus on multivariable analysis of PM and AMNs after updating pts, increasing accrual, and cleaning data. No significant financial relationships to disclose.