Table 1.
Baseline characteristics of patients.
Table 2.
HCV quantification with Abbott RealTime standard method (ART) and ultrasensitive method (US) in all analyzed samples.
Fig 1.
The temporal trend of HCV RNA for 14 patients with available daily serum samples; values are expressed in Log IU/ml.
We also reported values <12 IU/ml determined by ultrasensitive assay. B: Average of Log (HCV RNA) measurements; the red arrows indicate that three different slopes appear within the first month (T1m). Blue dashed lines correspond to the undetectable limits with the standard measurement of HCV RNA. The first slope falls within the first day (T1) and the second slope falls within the fourth day (T4). C: mean slopes of HCV RNA measurements observed during the first three months of treatment. T0, T1, and T4 correspond to day 0, 1, 4 days of treatment; T1m, T2m, T3m correspond to months 1, 2, 3 of treatment, respectively.
Fig 2.
Box plot for EVR (Panel A) and RVR (Panel B) of the first two slopes of HCV RNA measurements for the 14 patients. The top and bottom of the blue boxes are the first (25th percentile) and the third (75th percentile) quartiles; the red band inside the boxes is the second quartile (i.e., the median); the dashed lines extending vertically from the boxes (called whiskers) indicate data outside the interquartile range that are not outliers. Points are drawn as outliers if they are larger than q3 + w(q3 –q1) or smaller than q1 –w(q3 –q1), where q1 and q3 are the 25th and 75th percentiles, respectively. We used w = 1.5 that corresponds to approximately +/–2.7σ and 99.3 coverage if the data are normally distributed. The plotted whisker extends to the extreme values, which are the most extreme data values that are not outliers. The resulting p-values of a two-sided Wilcoxon rank sum test performed between the first two slopes within each group of patients are shown in black.
Fig 3.
Panel A: This plot displays the percent variability explained by each principal component and is called Pareto chart. It is a type of chart that contains both bars and a line graph, where individual values are represented in descending order by bars, and the line represents the cumulative total. In particular, the y-axis represents the percentage of the data variance explained by each principal component, whereas the x-axis represents the principal components that are able to explain the first 100% of the cumulative distribution. The principal component analysis is performed using the slopes of the HCV RNA values for the 14 patients. The first principal component is able to explain more than the 40% of the variance explained and the first three components are able to take into account more than 80% of the cumulative distribution. Looking at the PCAs’ loadings we found that the first principal component shows high correlation with the second and the third slope, while the second principal component correlates with the fifth slope and the third component correlates with the fourth slope. This indicates that the first slope, i.e., one day after the enrolment, does not contribute to the total variance. Inside each bar, the white writings represent which slopes correlate with each principal component. Panel B: This plot represents a scatter plot (score plot) of the projection of the data (i.e. the HCV RNA values for the 14 patients) onto the first two PCs; the x-axis contains the first PC while the y-axis contains the second PC. In this plot it is possible to group patients as EVR or RVR. This means that the first two principal components are able to classify patients as EVR or RVR. Since the first principal component correlates with the second and third slope (see panel A) and the second principal component correlates with the fifth slope (see panel A), we can conclude that these three slopes are necessary to classify patients as EVR or RVR.
Fig 4.
Panel A: Score plot where the x-axis contains the second PC while the y-axis contains third PC. Panel B: Box plot of the third score–that is the projection of the data on the third PC- for detected and undetected patients at month 1 after therapy initiation as results from the ultrasensitive assay. The resulting p-values of a two-sided Wilcoxon rank sum test are shown. This plot and the p-value show that the second and third PC shown in panel A, are able to separate detected and undetected patients at month 1 after therapy initiation in a statistically significant way.
Fig 5.
Linear discriminant analysis of viral Log (HCV RNA) for the classes EVR and RVR (Panel A) and class SVR (Panel B). The black line represents the decision region separating data; red region corresponds to EVR (or SVR) patients whereas green region corresponds to the RVR (or no-SVR) patients. Red circles and green squares represent Log (HCV RNA) measurements at days T1 and T4. Misclassified patients are x-marked.
Fig 6.
Receiver-operating characteristics ROC curve on the 20 patients’ HCV RNA measurements at month 1 after starting therapy.
AUC = area under the ROC curve = 0.7. The contingency table for the two-tailed Fisher’s exact test is reported: the p-value = 0.07 indicates that the SVR response is quite statistically and significantly different in patients with detectable HCV RNA at the first month compared to patients with undetectable HCV RNA.