Patients and Methods: Different techniques of blood collection, RNA isolation, and RT-PCR were compared, and the detectability of tyrosinase mRNA was tested using nine different melanoma cell lines. The sensitivity of the method was confirmed by blood spiking experiments and the specificity by restriction enzyme analysis. Subsequently, a total of 153 blood samples from 137 individuals (30 healthy subjects, five basal cell carcinoma, and 102 melanoma patients) were investigated.

Results: The detection level of melanoma cells differed between the cell lines tested. However, we could reproducibly detect single melanoma cells by spiking whole blood samples from healthy volunteers. One of 43 patients with primary melanoma (2.3%), zero of 15 patients with regional metastasis (0%), and 12 of 44 patients with advanced disease (27.3%) were found to be RT-PCR positive. All blood samples obtained from controls and patients with basal cell carcinoma were tyrosinase mRNA negative.

Conclusion: Our data support the recent doubts that the detection of circulating tumor cells in melanoma patients using the tyrosinase mRNA RT-PCR is not sensitive enough to be used either as a melanoma progression marker in early stages of the disease or to monitor therapy in advanced stages of the disease. (C) 1997 by American Society of Clinical Oncology.