Aims: Sentinel lymph node biopsy is an increasingly established procedure in the primary staging of high-risk melanoma patients. However, the laboratory evaluation of sentinel lymph node biopsies is a matter of controversy. The aim of this study was to determine the specificity of polymerase chain reaction (PCR) techniques for the evaluation of lymph nodes with regard to melanoma metastases in comparison with histology and immunohistology.
Methods and results: Sentinel lymph nodes (n = 41) from 29 melanoma patients and 29 lymph nodes from 27 patients without melanoma were analysed by histology (H&E) and immunohistology (Melan A, HMB45). cDNA of these lymph nodes was subjected to LightCycler PCR amplification using primers specific for tyrosinase and HMB45. Two melanoma sentinel lymph nodes contained naevus cells by histology and immunohistology and were therefore excluded from further evaluation. Eight (20.5%) of the remaining 39 melanoma sentinel lymph nodes were positive by histology and immunohistology and tyrosinase PCR, 15.4% (6/39) were positive only by tyrosinase PCR, 2.6% (1/39) were positive only by histology and immunohistology. HMB45 PCR revealed positive results in 7.7% (3/39) sentinel lymph nodes, which were also positive by tyrosinase PCR and histology and immunohistology. Of non-melanoma lymph nodes 13.8% (4/29) and 14.8% (4/27) of non-melanoma patients were positive by tyrosinase PCR but negative by histology and immunohistology and HMB45 PCR. Thus, tyrosinase PCR had a specificity of only 85.2%.
Conclusions: The specificity of tyrosinase PCR and the sensitivity of HMB45 PCR are too low to recommend these PCR examinations for the guidance of therapy, in particular complete regional lymph node dissection.