Table V Sensitivity and specificity of CEP markers and genomic markers

Sensitivity and specificity were determined from ROC curves to maximize the sum of the two values and constructed from the output of logistic regression analysis fit with either CEP adduct concentrations plus autoantibody titers, homozygous risk genotype, or the combination of the CEP biomarkers and the risk genotype. c-statistics, 95% CI, and p values derived from single and joint markers were determined with SAS 9.1 based on log-transformed CEP marker concentrations. Verification of c-statistics and 95% CI was performed by bootstrap resampling and 10-fold cross-validation. The c-statistic is a measure of the area under the ROC curve and the accuracy of the markers to discriminate between AMD cases and controls with 1.0 equivalent to 100% accuracy and 0.5 equal to no discrimination. Combining the CEP and genomic markers significantly improved the c-statistics for all the genomic markers.

Markers aloneCEPARMS2HTRA1CFHC3
Sensitivity (%)7331376036
Specificity (%)6594897786
c-statistic0.760.620.630.690.62
95% CI0.73, 0.790.58, 0.670.58, 0.680.64, 0.730.57, 0.66
c-statistic (bootstrap)0.760.620.630.690.61
95% CI (bootstrap)0.73, 0.790.60, 0.650.60, 0.660.65, 0.720.58, 0.62
c-statistic (cross-validation)0.760.580.550.640.58
95% CI (cross-validation)0.73, 0.780.54, 0.620.50, 0.600.60, 0.690.54, 0.62
Joint effect of markersCEP + ARMS2CEP + HTRA1CEP + CFHCEP + C3
Sensitivity (%)63717571
Specificity (%)81677674
c-statistic0.790.760.800.80
95% CI0.76, 0.820.72, 0.800.77, 0.840.77, 0.84
c-statistic (bootstrap)0.790.760.810.80
95% CI (bootstrap)0.76, 0.820.71, 0.800.77, 0.840.77, 0.84
c-statistic (cross-validation)0.790.750.800.80
95% CI (cross-validation)0.75, 0.820.70, 0.790.76, 0.830.76, 0.83