Skip to main content
. 2018 Mar 23;7(5):1697–1705. doi: 10.1002/cam4.1398

Table 2.

Diagnostic performance of miRNAs for colorectal cancer detection

Median log10‐expression P a Adjusted P b Apparent AUC .632+ AUC .632+ sensitivity
Controls CRC cases At 80% specificity At 90% specificity
miR.21 −8.0 −6.8 <0.001 <0.001 0.84 0.84 78 62
miR.25 −8.9 −8.5 <0.001 <0.001 0.65 0.63 46 31
miR.18a −10.2 −9.8 0.003 0.036 0.59 0.56 29 12
miR.22 −9.7 −10.2 0.004 0.046 0.59 0.55 26 12
miR.15b −8.7 −8.4 0.005 0.054 0.58 0.57 28 20
miR.142.5p −10.1 −9.7 0.015 0.176 0.57 0.54 22 13
miR.29c −9.7 −9.6 0.116 1.392 0.53 0.5 22 13
miR.19a −9.3 −9.1 0.595 7.144 0.49 0.45 14 6
miR.16 −8.4 −8.4 0.541 6.486 0.49 0.44 14 7
miR.192 −10.3 −10.4 0.555 6.658 0.49 0.46 19 9
miR.26a −7.8 −8.0 0.806 9.674 0.48 0.43 17 8
miR.150 −9.0 −9.0 0.752 9.029 0.46 0.43 15 7
a

Wilcoxon signed‐rank test for difference in miRNA expression distribution between CRC cases and controls.

b

The P‐value was adjusted for multiple testing by the Bonferroni method by multiplying the nominal P‐value with 12 (the number of miRNAs analyzed).