E level of alpha = 0.05.ResultsThe box-plots reported in Figure 1, panel A , describe the distribution of each Epigenetic Reader Domain biomarker in case and controls.Table 2 reports some descriptive statistics of these distributions. Using the Kolmogorov mirnov test, we found that the difference of the distributions of each biomarker in cases and controls was statistically significant (p-value ,0.05). As reported in supplemental Table S1, the same results were observed when this comparison was performed according to the stage of disease for cfDNA and integrity index 180/67. Conversely these findings were notFigure 4. Contribution of each biomarker to the final model – ROC Curves. ROC 12926553 curves corresponding to the contribution of each biomarker in the final multivariate logistic model. Without total cfDNA (AUC = 0.86), without integrity index 180/67 (AUC = 0.90), without methylated RASSF1A (AUC = 0.89). doi:10.1371/journal.pone.0049843.gCell-Free DNA Biomarkers in MelanomaTable 5. Contribution of each biomarker of the final model.AUC Final model Without the following variables: total cfDNA (ng/ml plasma) integrity index 180/67 methylated RASSF1A (GE/ml plasma) 0.862 0.903 0.894 0.AUC 95 CI 0.910?.p-value ,0.0.801?.923 0.854?.952 0.839?.,0.0001 ,0.0001 ,0.and Figure 3). The contribution of each variable of the final model to the diagnostic performance is shown in Table 5 and graphically described in Figure 4. The highest predictive capability was given by total cfDNA 
(AUC:0.86, 95 CI: 0.80?.92) followed by integrity index 180/67 (AUC:0.90, 95 CI: 0.85?.95) and methylated RASSF1A (AUC:0.89, 95 CI: 0.84?.95). As shown in the supplemental figure S1 a comparable predictive capability was observed for each considered biomarker (univariate analysis) according to the stage of disease. Only for BRAFV600E within the stage 0 and stage III V the 95 CI of the AUC includes the 0.5 value.Abbreviations: AUC, area under the ROC curve; CI, Confidence Interval. doi:10.1371/journal.pone.0049843.tDiscussionThe analysis of cfDNA may have the potential to complement or 23727046 replace the existing cancer tissue and blood biomarkers in the future [35]. In order to reach this goal, specific and sensitive analytical procedures must be developed and optimized to compute proper circulating target molecules showing differences between patients and healthy subjects. It is now widely accepted that a 
single biomarker cannot fully distinguish between controls and patients and inhibitor consequently an approach based on different markers would be preferable in order to achieve a stronger predictive ability [36]. It has been demonstrated that in prenatal screening, a combination of multiple markers, each with limited sensitivity and/or specificity, can lead to a more powerful screening test [37]. Similarly, Schneider and Mizejewski [38] suggest to develop a multi-marker screening approach for cancer diagnosis. Unfortunately this strategy has been proven unsuccessful, notwithstanding the high number of new biomarkers reported in the literature, even if some examples on prostate ovarian and colorectal cancer clearly showed that multi-marker screening can have its place in early cancer detection [38?9]. The study presented here tests the diagnostic potential of four markers associated to cfDNA in identifying melanoma patients.observed within stage I I for methylated RASSF1A and within stage 0 and stage III V for BRAFV600E. For all the biomarkers considered in the logistic regression model we found that a linear relat.E level of alpha = 0.05.ResultsThe box-plots reported in Figure 1, panel A , describe the distribution of each biomarker in case and controls.Table 2 reports some descriptive statistics of these distributions. Using the Kolmogorov mirnov test, we found that the difference of the distributions of each biomarker in cases and controls was statistically significant (p-value ,0.05). As reported in supplemental Table S1, the same results were observed when this comparison was performed according to the stage of disease for cfDNA and integrity index 180/67. Conversely these findings were notFigure 4. Contribution of each biomarker to the final model – ROC Curves. ROC 12926553 curves corresponding to the contribution of each biomarker in the final multivariate logistic model. Without total cfDNA (AUC = 0.86), without integrity index 180/67 (AUC = 0.90), without methylated RASSF1A (AUC = 0.89). doi:10.1371/journal.pone.0049843.gCell-Free DNA Biomarkers in MelanomaTable 5. Contribution of each biomarker of the final model.AUC Final model Without the following variables: total cfDNA (ng/ml plasma) integrity index 180/67 methylated RASSF1A (GE/ml plasma) 0.862 0.903 0.894 0.AUC 95 CI 0.910?.p-value ,0.0.801?.923 0.854?.952 0.839?.,0.0001 ,0.0001 ,0.and Figure 3). The contribution of each variable of the final model to the diagnostic performance is shown in Table 5 and graphically described in Figure 4. The highest predictive capability was given by total cfDNA (AUC:0.86, 95 CI: 0.80?.92) followed by integrity index 180/67 (AUC:0.90, 95 CI: 0.85?.95) and methylated RASSF1A (AUC:0.89, 95 CI: 0.84?.95). As shown in the supplemental figure S1 a comparable predictive capability was observed for each considered biomarker (univariate analysis) according to the stage of disease. Only for BRAFV600E within the stage 0 and stage III V the 95 CI of the AUC includes the 0.5 value.Abbreviations: AUC, area under the ROC curve; CI, Confidence Interval. doi:10.1371/journal.pone.0049843.tDiscussionThe analysis of cfDNA may have the potential to complement or 23727046 replace the existing cancer tissue and blood biomarkers in the future [35]. In order to reach this goal, specific and sensitive analytical procedures must be developed and optimized to compute proper circulating target molecules showing differences between patients and healthy subjects. It is now widely accepted that a single biomarker cannot fully distinguish between controls and patients and consequently an approach based on different markers would be preferable in order to achieve a stronger predictive ability [36]. It has been demonstrated that in prenatal screening, a combination of multiple markers, each with limited sensitivity and/or specificity, can lead to a more powerful screening test [37]. Similarly, Schneider and Mizejewski [38] suggest to develop a multi-marker screening approach for cancer diagnosis. Unfortunately this strategy has been proven unsuccessful, notwithstanding the high number of new biomarkers reported in the literature, even if some examples on prostate ovarian and colorectal cancer clearly showed that multi-marker screening can have its place in early cancer detection [38?9]. The study presented here tests the diagnostic potential of four markers associated to cfDNA in identifying melanoma patients.observed within stage I I for methylated RASSF1A and within stage 0 and stage III V for BRAFV600E. For all the biomarkers considered in the logistic regression model we found that a linear relat.
