Characteristics Associated with Pathologic Nodal Burden in Patients Presenting with Clinical Melanoma Nodal Metastasis
Nodal observation is safe for patients with microscopic melanoma metastasis in a sentinel lymph node (LN). Complete LN dissection (CLND) remains the standard of care for patients with clinically evident LN (cLN) metastases, even though about 40% have only one pathologic LN (pLN). We sought to identify clinical features associated with having one pLN among patients with cLNs.
Patients at three melanoma centers who underwent CLND for cLNs were identified. Clinicopathologic and imaging characteristics associated with one pLN were determined by multivariable logistic regression and classification tree analysis.
Of 190 patients, 90 (47.4%) had one pLN and 100 (52.6%) had more than one pLN. By multivariable logistic regression, extremity versus truncal primary (odds ratio [OR] 2.15, p = 0.012), axillary versus superficial inguinal location (OR 3.89, p = 0.009), and preoperative cross-sectional imaging demonstrating more than one versus one cLN (OR 17.1, p < 0.001) were associated with more than one pLN. The negative predictive value for additional pathologic nodal disease of preoperative imaging was 70.9%, increasing to 74.4% for positron emission tomography/computed tomography. In the subgroup of patients with one cLN, the classification tree identified two groups with < 10% risk of additional pLNs: (1) Breslow thickness > 6.55 mm (n = 17); and (2) if unknown primary or Breslow thickness ≤ 6.55 mm, then LN diameter > 1.8 cm in the inguinal location (n = 22).
The majority of patients with one cLN from melanoma by preoperative imaging will harbor no additional pathologic nodes on CLND. Safety of nodal observation after clinical nodal excision in these patients, particularly in an era of effective adjuvant therapies, deserves prospective evaluation.
Partial funding was provided by National Cancer Institute (Grant Nos. T32-CA009109-41 and T32-CA009111), by a Farrow Fellowship, and by the Rebecca Clay Harris Memorial Fellowship.
Minyoung Kwak, Yun Song, Phyllis A. Gimotty, Norma E. Farrow, Andrew D. Tieniber, Jonathan G. Davick, Gabriella N. Tortorello, Georgia M. Beasley, Craig L. Slingluff Jr, and Giorgos C. Karakousis have no relevant conflicts of interest or disclosures to declare.
- 7.Coit DG, Thompson JA, Albertini MR, et al. National Comprehensive Cancer Network (NCCN) clinical practice guidelines in oncology, cutaneous melanoma, version 2.2019. https://www.nccn.org/professionals/physician_gls/#melanoma.. Accessed 1 Apr 2019.
- 10.R Core Team (2019). R: a language and environment for statistical computing. Vienna: R Foundation for Statistical Computing. https://www.R-project.org/. Accessed 1 July 2018.
- 11.Salford Predictive Model (SPM). San Diego, CA: Salford Systems. https://www.salford-systems.com/products/spm.
- 12.SAS/STAT Software®. Cary, NC: SAS Institute Inc.Google Scholar
- 13.Balch CM, Soong S-J, Thompson JF, et al. Age as a predictor of sentinel node metastasis among patients with localized melanoma: an inverse correlation of melanoma mortality and incidence of sentinel node metastasis among young and old patients. Ann Surg Oncol. 2014;21(4):1075–81.CrossRefGoogle Scholar
- 18.Gershenwald JE, Scolyer RA, Hess KR, et al. Melanoma staging: evidence-based changes in the American Joint Committee on Cancer eight edition cancer staging manual. CA Cancer J Clin. 2017;67(6):472–92.Google Scholar