ABSTRACT
OBJECTIVE: The aim of this study was to evaluate if our molecular algorithm, based on tumor circulating transcripts, may predict relapse risk in cutaneous malignant melanoma (CMM). RESULTS: The multi-marker panel was able to differentiate patients with CMM from HC with high diagnostic sensitivity and specificity, especially for MITF-m and TGFB2 (91-100%) whose levels decreased during follow-up of recurrence-free patients, and remained stable in the case of relapse. PAX3d higher than 2.76 copies/µL emerged as a promising biomarker [specificity = 75-93% and negative predictive value = 75-98%] to stratify subjects at high risk of CMM recurrence independently of age, gender and AJCC staging [OD = 9.5(3.2-28.0), p < 0.001]. The survival analysis confirmed PAX3d performance in relapse prediction with significant differences in recurrence risk 12 months after the basal time-point (p = 0.008). MATERIALS AND METHODS: Peripheral blood was collected from 111 CMM patients and from 87 healthy controls (HC) randomly selected. Each specimen was examined by qRT-PCR analysis for the expression of 3 tumor-related transcripts (PAX3d, MITF-m and TGFB2) at diagnosis, and at the following 6 and 12 months during clinical monitoring. CONCLUSIONS: We demonstrated the usefulness of our molecular algorithm to indirectly detect circulating melanoma cells in blood, along with PAX3d capability to assess patients' progression and relapse prediction.