Multiparametric Analysis of Cell-Free DNA in Melanoma Patients

Cell-free DNA in blood (cfDNA) represents a promising biomarker for cancer diagnosis. Total cfDNA concentration showed a scarce discriminatory power between patients and controls. A higher specificity in cancer diagnosis can be achieved by detecting tumor specific alterations in cfDNA, such as DNA integrity, genetic and epigenetic modifications. The aim of the present study was to identify a sequential multi-marker panel in cfDNA able to increase the predictive capability in the diagnosis of cutaneous melanoma in comparison with each single marker alone. To this purpose, we tested total cfDNA concentration, cfDNA integrity, BRAF^V600E mutation and RASSF1A promoter methylation associated to cfDNA in a series of 76 melanoma patients and 63 healthy controls. The chosen biomarkers were assayed in cfDNA samples by qPCR. Comparison of biomarkers distribution in cases and controls was performed by a logistic regression model in both univariate and multivariate analysis. The predictive capability of each logistic model was investigated by means of the area under the ROC curve (AUC). To aid the reader to interpret the value of the AUC, values between 0.6 and 0.7, between 0.71 and 0.8 and greater than 0.8 were considered as indicating a weak predictive, satisfactory and good predictive capacity, respectively. The AUC value for each biomarker (univariate logistic model) was weak/satisfactory ranging between 0.64 (BRAF^V600E) to 0.85 (total cfDNA). A good overall predictive capability for the final logistic model was found with an AUC of 0.95. The highest predictive capability was given by total cfDNA (AUC:0.86) followed by integrity index 180/67 (AUC:0.90) and methylated RASSF1A (AUC:0.89). An approach based on the simultaneous determination of three biomarkers (total cfDNA, integrity index 180/67 and methylated RASSF1A) could improve the diagnostic performance in melanoma.