Correlations between immune response and vascularization qRT-PCR gene expression clusters in squamous cervical cancer

Table 2 Cox regression analyses

	Univariate Cox regression		Multivariate Cox regression
Variable	Hazard ratio (95% CI)	p value	Hazard ratio (95% CI)	p value	Hazard ratio (95% CI)	p value
Lymph node status	1.20 (0.35-4.11)	0.769	1.56 (0.30-8.07)	0.599	1.35 (0.27-6.77)	0.718
Tumour size	1.05 (1.02-1.09)	0.006	1.05 (0.99-1.11)	0.121	1.06 1.00-1.12)	0.073
Vaso-invasion	1.22 (0.36-4.15)	0.756
Infiltration depth	1.06 (0.99-1.13)	0.096	1.01 (0.92-1.11)	0.785	0.97 (0.89-1.06)	0.556
IL6	2.21 (1.34-3.66)	0.002	2.29 (1.21-4.34)	0.011
IL5	0.96 (0.61-1.51)	0.853
ANGPT2	1.12 (0.56-2.24)	0.746
TBX21	0.997 (0.53-1.87)	0.993
CD14	0.45 (0.17-1.18)	0.104
PECAM1	0.90 (0.46-1.76)	0.754
IL6/IL17 + IL5	4.66 (1.90-11.41)	0.001			4.17 (1.41-12.40)	0.010

Univariate Cox regression hazard ratios are shown for the critical prognostic categorical clinico-pathological parameter lymph node tumour positivity and the continuous variables tumour size (per mm) and infiltration depth (per mm), as well as for the expression of genes representative for the different clusters. Normalized Cq values were converted to expression values to obtain hazard ratio’s corresponding with increased presence of the gene product. Multivariate Cox regression analyses are shown for the genes significant in the univariate analysis combined with the most critical clinico-pathological parameters (restricted by the number of patients). The combination of the IL6/IL17 ratio and IL5 is divided in three categories: a low ratio, a high ratio and high IL5 levels and a high ratio combined with low IL5 levels.

ISSN: 1476-4598