Background There is a current need for a reliable prognostic marker for melanoma patients, particularly those with stage 2 and stage 3 disease, so that adjuvant therapies can be directed appropriately.
Objectives To establish whether or not the use of tyrosinase-specific or melanA/MART-1-specific reverse transcriptase-coupled–polymerase chain reaction (RT–PCR) of peripheral blood cells detects preclinical disease progression in patients with malignant melanoma.
Methods Two hundred and ninety-nine patients with melanoma in clinical stages 1–4 were observed in this study. Samples were obtained sequentially from 153 of these patients at 4-week intervals over a period of up to 2 years and correlated with clinical evidence of disease activity. Tyrosinase and melanA/MART-1 amplicons were analysed by agarose gel electrophoresis and Southern blot hybridization subsequent to a single round of amplification.
Results We demonstrated a statistically significant increase in tyrosinase RT–PCR positivity with advancing stage of melanoma progression. The percentage tyrosinase positivity in 910 samples tested was: stage 1, 135 samples, 34% positive; stage 2, 196 samples, 51% positive; stage 3, 423 samples, 50% positive; and stage 4, 156 samples, 65% positive. The positivity rate for individual patients tested sequentially was higher if only one positive test was required to label a patient positive, at 42%, 65%, 82% and 81% for patients in stages 1–4, respectively. However, we did not find a clear pattern of conversion from negativity to positivity in patients who progressed during the study from stage 2 to stage 3 or stage 3 to stage 4, and found no clear evidence of increased positivity rates in the 6-week period following melanoma-related surgery in patients with stage 3 and 4 disease. The positivity rate for melanA/MART-1 was lower for both patients and samples, and no melanA/MART-1-positive sample was negative for tyrosinase.
Conclusions We conclude that the presence of circulating tyrosinase-positive cells as detected by this method appears to be a discontinuous rather than a continuous phenomenon, even in patients with stage 4 disease. For this reason the assay cannot be recommended as a method of sequentially monitoring individual patients in a clinical setting.