Identification of genes with high heterogeneity of expression as a predictor of different prognosis and therapeutic responses in colorectal cancer: a challenge and a strategy

Table 1 Cox regression results for candidate genes with HHE based on TCGA clinical data

	Univariate			Multivariate
	HR	P value	95% CI	HR	P value	95% CI
Age (60 > vs. < 60)	1.8	0.12	0.92–3.39
Gender (Female vs. male)	1.4	0.23	0.72–2.63
Pathological stage (P1-P2 vs. P3-P4)	11.9	2.4E-09	4.2–33.6	11.3	0.0008	3.37–18.23
TNM T stage (T1-T2 vs. T3-T4)	9.3	0.007	1.27–21.32	1.76	0.59	0.21–14.67
BLACAT1 expression (High vs. low)	1.37	0.03	1.01–1.86	1.19	0.34	0.82–1.73
SFTA2 expression (High vs. low)	1.43	0.007	1.09–1.86	1.53	0.01	1.1–2.13
S100A2 expression (High vs. low)	1.35	0.03	1.03–1.77	1.36	0.09	0.94–1.97
BGN expression (High vs. low)	1.39	0.03	1.02–1.9	0.69	0.38	0.3–1.58
SERPINE1 expression (High vs. low)	1.52	0.002	1.16–2.1	1.09	0.73	0.64–1.87
COMP expression (High vs. low)	1.37	0.02	1.03–1.83	0.80	0.54	0.4–1.64
THBS2 expression (High vs. low)	1.38	0.04	1–1.89	0.65	0.37	0.25–1.66
SELE expression (High vs. low)	1.33	0.01	1.04–1.69	2.1	0.001	1.32–3.48
SACS expression (High vs. low)	1.39	0.02	1.03–1.87	1.6	0.01	1.11–2.29
PBK expression (High vs. low)	0.67	0.006	0.52–0.89	0.76	0.24	0.49–1.19
SLC38A5 expression (High vs. low)	0.72	0.01	0.53–0.93	0.75	0.11	0.53–1.06
CXCL1 expression (High vs. low)	0.73	0.03	0.54–1	0.58	0.007	0.40–0.87

The expression matrix of genes in the Z-score scale was used to analyze the relationship between gene expression and survival

ISSN: 1475-2867