Survival Analysis and Clinicopathological Factors Associated With False-Negative Sentinel Lymph Node Biopsy Findings in Patients with Cutaneous Melanoma

Background We analyzed the outcomes and factors associated with false-negative (FN) results of sentinel lymph node (SLN) biopsy findings in patients with cutaneous melanoma. SLN biopsy failure rate was defined as nodal recurrence in the biopsied regional basin without previous local or in-transit recurrence.Methods Between April 1997 and December 2004, a total of 1207 patients with cutaneous melanoma with a median Breslow thickness of 2.4 mm underwent SLN biopsy by preoperative and intraoperative lymphoscintigraphy combined with dye injection. In 228 cases, we found positive SLNs; of these, 220 underwent completion lymph node dissection (CLND). Median follow-up was 3 years.Results The SLN biopsy failure rate was 5.8% (57 of 979 SLN negative). Median time to occurrence of FN relapse after SLN biopsy was 16 months (range, 3–74 months). The FN SLN biopsy results correlated with primary tumor thickness >4 mm (P = .0012), primary tumor ulceration (P = .0002), primary tumor level of invasion Clark stage IV/V (P = .0005), and nodular melanoma histological type (P = .0375). Five-year overall survival, calculated from the date of primary tumor excision, in the FN group was 53.7%, which was not statistically significantly worse than the CLND group (56.8%; P = .9). The FN group was characterized by a higher ratio of two or more metastatic nodes and extracapsular involvement of lymph nodes after LND compared with the CLND group (P < .0001 and P < .0001, respectively). Additional detailed pathological review of FN SLN revealed metastatic disease in 14 patients, which decreased the SLN biopsy failure rate to 4.4% (43 of 979).Conclusions Survival of patients with FN results of SLN biopsy does not differ statistically significantly from that of patients undergoing CLND, although it is slightly lower. The SLN biopsy failure rate is approximately 5.0% in long-term follow-up and is associated mainly with the same factors that indicate a poor prognosis in primary melanoma.Preliminary results of this study were presented as an oral presentation during the Melanoma Session on the 59th Annual Cancer Symposium of the Society of Surgical Oncology, San Diego, CA, March 23–25, 2006.