Utility of tests for circulating melanoma cells in identifying patients who develop recurrent melanoma

Prospective studies were carried out on 186 patients with AJCC stage I (13), II (76) and III (97 patients) melanoma before and after surgical removal of their tumour. The goal was to determine if PCR tests on blood samples for MART-1 and tyrosinase were predictive of recurrence of melanoma in a 2-year follow-up period. (PCR assays for MUC-18, p97 and gp100 were positive in blood samples from normal subjects and excluded from the study.) PCR tests for MART-1 and tyrosinase were most commonly positive in the first 3 months following surgical removal of melanoma, and three tests over this period gave maximum sensitivity in the identification of patients who subsequently developed recurrences. Positive tests for the first time in the second year of follow-up had similar predictive power. The tests identified 68.5% of patients who developed recurrences in the 2-year follow-up period. Assays for MART-1 were mainly positive in patients with locoregional recurrences, whereas tyrosinase was detected in blood samples from patients with both locoregional and disseminated recurrences. (Positivity rate for tyrosinase in 48 patients with disseminated melanoma was 60.4% compared to 14.6% for MART-1.) Tests for MART-1 and tyrosinase were strongly predictive of disease-free survival (DFS) and were more powerful predictors of DFS than lymph node status or thickness of the primary melanoma. (Hazard ratios by Cox analysis were 2.97 in patients with disseminated recurrences and 2.93 in those with locoregional recurrences.) These results indicate that PCR tests for MART-1 and tyrosinase are powerful prognostic indicators, but their practical utility for selecting patients for adjuvant therapy is limited by the high false-negative rate of approximately 30%. Intermittent shedding of melanoma cells into the circulation would appear to be the most likely explanation for the latter.