Identifying the Best Marker Combination in CEA, CA125, CY211, NSE, and SCC for Lung Cancer Screening by Combining ROC Curve and Logistic Regression Analyses: Is It Feasible?

Table 6

AUCs of various combinations of markers for screening lung cancer in different populations with different sexes and ages.

AUC	Healthy subjects		Pneumonia patients
	Male	Female	Male	Female
			<65	≥65	<65	≥65
CEA	0.829	0.846^b3	0.646	0.681	0.718^e2	0.681^f3
CA125	0.766	0.738	0.666	0.547	0.665	0.615
CY211	0.763	0.625	0.680^c3	0.689	0.637	0.620
NSE	0.624	0.617	0.598	0.640	0.639	0.635
CEA + CA125	0.873^a1	0.858^b1	0.674	0.652	0.700^e3	0.680
CEA + CY211	0.863^a2	0.846	0.692^c2	0.727^d1	0.651	0.713^f2
CEA + NSE	0.839^a3	0.847^b2	0.623	0.700^d2	0.728^e1	0.731^f1
CA125 + CY211	0.832	0.725	0.698^c1	0.681	0.665	0.627
CA125 + NSE	0.773	0.745	0.671	0.624	0.684	0.664
CY211 + NSE	0.772	0.645	0.680	0.697^d3	0.645	0.652

a1 vs. a2: , ; a2 vs. a3: , ; a1 vs. a3: , ; b1 vs. b2: , ; b2 vs. b3: , ; b1 vs. b3: , ; c1 vs. c2: , ; c1 vs. c3: , ; c2 vs. c3: , ; d1 vs. d2: , ; d1 vs. d3: , ; d2 vs. d3: , ; e1 vs. e2: , ; e1 vs. e3: , ; e2 vs. e3: , ; f1 vs. f2: , ; f1 vs. f3: , ; f2 vs. f3: , .