Secondary Logo

Institutional members access full text with Ovid®

Clinicopathologic features correlated with paradoxical outcomes in stage IIC versus IIIA melanoma patients

Tan, Sally Y.a; Najita, Julieg; Li, Xiaoxueg; Strazzulla, Lauren C.k; Dunbar, Hailii; Lee, Mee-youngj; Seery, Virginia J.b,e; Buchbinder, Elizabeth I.h; Tawa, Nicholas E.c,e; McDermott, David F.b,e; Lee, Sandra J.f; Atkins, Michael B.l,*; Kim, Caroline C.d,e,*

doi: 10.1097/CMR.0000000000000483
ORIGINAL ARTICLES :Epidemiology of melanoma

Under current AJCC staging criteria, stage IIC patients paradoxically have worse outcomes than IIIA patients despite the lack of nodal metastatic disease. This study sought to identify additional clinicopathologic characteristics correlated with worse patient outcomes. Retrospective chart review of stage IIC and IIIA melanoma patients were evaluated between 1995 and 2011 with clinical follow-up through 2015. Records were reviewed for demographics, clinical characteristics, and tumor pathology. Fisher’s exact test and Wilcoxon’s rank-sum test were used to assess group differences. Clinicopathologic features were evaluated relative to overall survival (OS), time to distant metastases, and local/regional recurrence. Overall, 128 patients were included (45 stage IIC and 83 stage IIIA) with a median follow-up time of 5.7 years. Compared with stage IIIA patients, stage IIC patients were older, and their melanomas were more likely to be nodular, amelanotic, thicker, have higher mitotic rate, tumor lymphocytic infiltrate, no radial growth phase, and less likely to have associated precursor lesions. Stage IIC patients had shorter OS and time to distant metastases; multivariate regression revealed that older age (>55 years) and mitotic rate (>5 mitoses/mm2) were independent predictors of OS. Melanomas in stage IIC disease may be biologically distinct from those that are seen in stage IIIA. While sentinel node biopsies remain the standard-of-care, these results suggest that clinicians may want to assess the clinicopathologic characteristics described above to aggressively counsel, screen for distant disease, and consider adjuvant therapy, in particular for older patients and higher mitotic rates in thicker primary tumors, regardless of nodal status.

aDepartment of Medicine, Brigham and Women’s Hospital, Harvard Medical School

Departments of bMedicine, Division of Hematology and Oncology


dDermatology, Harvard Medical School

eMedicine, Cutaneous Oncology Multidisciplinary Program, Beth Israel Deaconess Medical Center

Departments of fBiostatistics and Computational Biology, Harvard Medical School

gBiostatistics and Computational Biology, Harvard School of Public Health

hHematology and Oncology, Dana-Farber Cancer Institute

iDepartment of Medicine, Tufts Medical School, Boston, Massachusetts

jNorthwell Health Monter Cancer Center, Lake Success

kRonald O. Perelman Department of Dermatology, New York University School of Medicine, New York City, New York

lDepartment of Medical Oncology, Georgetown-Lombardi Comprehensive Cancer Center, Washington, District of Columbia, USA

*Michael B. Atkins and Caroline C. Kim contributed equally to the writing of this article.

Correspondence to Caroline C. Kim, MD, Pigmented Lesion Clinic, Cutaneous Oncology Program, Beth Israel Deaconess Medical Center, 330 Brookline Avenue, Boston, MA 02215, USA Tel: +1 617 667 3753; fax: +1 617 975 5033; e-mail:

Received February 12, 2018

Accepted June 17, 2018

Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved.