88 Chapter 4 from 34% to 87% and 84% to 98%, respectively, and the marker combination yielded a sensitivity and specificity of both 90%. For cervicovaginal self-samples, the sensitivity and specificity of single genes ranged from 28% to 78% and 85% to 95%, respectively, while the marker combination revealed a sensitivity of 89% and specificity of 92%. Similarly, for the clinician-taken cervical scrapes, the sensitivity and specificity of the single genes ranged from 44% to 87% and 67% to 93%, respectively, while the marker combination sensitivity was 93% with a specificity of 90%. Sensitivities and specificities were calculated based on the maximal Youden’s Index (J) threshold (Supplementary Table 1). 1 − specificity Urine 0.00 0.25 0.50 0.75 1.00 ADCYAP1 AUC=0.68 BHLHE22 AUC=0.76 CDH13 AUC=0.69 CDO1 AUC=0.91 GALR1 AUC=0.70 GHSR AUC=0.85 HAND2 AUC=0.65 SST AUC=0.62 ZIC1 AUC=0.62 CDO1 + GHSR + ZIC1 AUC=0.94 Self-sample 0.00 0.25 0.50 0.75 1.00 ADCYAP1 AUC=0.78 BHLHE22 AUC=0.84 CDH13 AUC=0.65 CDO1 AUC=0.95 GALR1 AUC=0.63 GHSR AUC=0.89 HAND2 AUC=0.61 SST AUC=0.74 ZIC1 AUC=0.76 CDH13 + CDO1 + ZIC1 AUC=0.97 Scrape 0.00 0.25 0.50 0.75 1.00 sensitivity ADCYAP1 AUC=0.83 BHLHE22 AUC=0.85 CDH13 AUC=0.90 CDO1 AUC=0.90 GALR1 AUC=0.79 GHSR AUC=0.93 HAND2 AUC=0.71 SST AUC=0.61 ZIC1 AUC=0.78 CDH13 + GHSR + SST AUC=0.95 0.00 0.25 0.50 0.75 1.00 Figure 2: Diagnostic performance of individual markers (ADCYAP1, BHLHE22, CDH13, CDO1, GALR1, GHSR, HAND2, SST, ZIC1) and the optimal three-marker panels (based on multivariable logistic regression) for endometrial cancer detection in urine, cervicovaginal self-samples and clinician-taken cervical scrapes. Non-cross-validated ROC curves are shown and quantified by AUC values. Individual genes and marker panels with the highest performance per sample type are depicted in bold. AUC, area under the ROC curve; ROC, receiver operating characteristic. The diagnostic performance of individual markers and marker panels were validated by LOOCV, which yielded virtually equal AUC values (Table 3), sensitivities, and specificities (Supplementary Table 2) for the single markers and optimal three marker combinations. Additionally, the performance for early-stage endometrial cancer detection was assessed by performing a sub-analysis including only stage I endometrial cancers (n = 72). This revealed nearly equal diagnostic performances for both the individual markers and marker panels in urine, cervicovaginal self-samples and clinician-taken cervical scrapes, which were also validated by LOOCV (Supplementary Table 3).
RkJQdWJsaXNoZXIy MjY0ODMw