Background: The survival benefit of sentinel node biopsy is still controversial. The aim of our study was to assess the overall survival (OS; calculated both from the date of primary tumor excision and lymph node dissection) data from two large groups of AJCC 2002 stage-III cutaneous melanoma patients-after completion lymph node dissection (CLND after positive sentinel node biopsy) and after therapeutic LND (TLND for clinically/cytologically detected regional lymph node metastases).
Materials and methods: We analyzed the outcomes for 544 consecutive patients, who underwent CLND (47.4%; 258 patients) or TLND (52.6%; 286 patients) at one institution between December 1994 and January 2005. There were no significant differences between the two groups in terms of age and gender distribution and in the parameters of the primary tumor. Median follow-up time was 36 months (range 6-110 months).
Results: We found no significant differences in OS (from the date of primary tumor excision) between CLND and TLND patients in the groups with primary tumor thicknesses of 1.0 mm or less or greater than 4.0 mm (pT1 and pT4); however, in patients with thicknesses greater than 1.0 mm and 4.0 mm or less (in subgroups pT2 and pT3), we found significantly better OS for CLND than for TLND patients-CLND: median OS not reached, 5-year OS was 57.2% (95%CI: 44.4-70.1%); TLND: median OS 42.1 months, 5-year OS was 37.9% (95%CI: 26.5-49.2%) (P = 0.0006). In the entire CLND and TLND groups, the median OS and 5-year OS rates were 60.5 months and 52.5% (95%CI: 45.6-61.5%) and 38.2 months and 39.5% (95%CI: 32.7-46.5%), respectively. Based on multivariate analysis, we have found that in the CLND group the important factors negatively influencing OS (from the date of lymphadenectomy) are: male gender, features of primary tumor (higher Breslow thickness and presence of ulceration) and features of nodal metastases (extracapsular invasion and number of involved nodes). In the TLND group, however, the negative prognostic factors are: male gender and features of nodal metastases (extracapsular invasion and number of involved nodes) without the impact of primary tumor characteristics.
Conclusion: The results of the study demonstrate that the survival benefit after positive sentinel node biopsy with subsequent CLND is probably limited only to the subgroup of patients with primary tumor thicknesses not larger than 4 mm and not less than 1 mm when compared with lymph node dissection of palpable nodes. The primary tumor features have no impact on survival after lymphadenectomy performed for clinically involved nodes.