Is complete lymph node dissection after a positive sentinel lymph node biopsy for cutaneous melanoma always necessary? A meta-analysis.

BACKGROUND

The current recommendation for patients with cutaneous melanoma and a positive sentinel lymph node (SLN) biopsy is a complete lymph node dissection (CLND). However, metastatic melanoma is not present in approximately 80% of CLND specimens. A meta-analysis was performed to identify the clinicopathological variables most predictive of non-sentinel node (NSN) metastases when the sentinel node is positive in patients with melanoma.

METHODS

A systematic search was conducted using MEDLINE, PubMed, EMBASE, Current Contents Connect, Cochrane library, Google scholar, Science Direct, and Web of Science. The search identified 54 relevant articles reporting the frequency of NSN metastases in melanoma. Original data was abstracted from each study and used to calculate a pooled odds ratio (OR) and 95% confidence interval (95% CI).

FINDINGS

The pooled estimates that were found to be significantly associated with the high likelihood of NSN metastases were: ulceration (OR: 1.88, 95% CI: 1.53-2.31), satellitosis (OR: 3.25, 95% CI: 1.86-5.66), neurotropism (OR: 2.51, 95% CI: 1.39-4.53), >1 positive SLN (OR: 1.77, 95% CI: 1.2-2.62), Starz 3 (old) (OR: 1.83, 95% CI: 0.89-3.76), Angiolymphatic invasion (OR: 2.46, 95% CI: 1.34-4.54), extensive location (OR: 2.22, 95% CI: 1.74-2.81), macrometastases >2 mm (OR: 1.95, 95% CI: 1.61-2.35), extranodal extension (OR: 3.38, 95% CI: 1.79-6.40) and capsular involvement (OR: 3.16, 95% CI: 1.37-7.27). There were 3 characteristics not associated with NSN metastases: subcapsular location (OR: 0.51, 95% CI: 0.38-0.67), Rotterdam Criteria <0.1 mm (OR: 0.29, 95% CI: 0.17-0.50) and Starz I (new) (OR: 0.44, 95% CI: 0.22-0.91). Other variables including gender, Breslow thickness 2-4 mm and extremity as primary site were found to be equivocal.

INTERPRETATION

This meta-analysis provides evidence that patients with low SLN tumor burden could probably be spared the morbidity associated with CLND. We identified 9 factors predictive of non-SLN metastases that should be recorded and evaluated routinely in SLN databases. However, further studies are needed to confirm the standard criteria for not performing CLND.