Identification and validation of the prognostic value of cyclic GMP-AMP synthase-stimulator of interferon (cGAS-STING) related genes in gastric cancer

Figures & data

Table 1. The clinical characteristics of the patients

Items	Databases
	TCGA		GEO
age	>65	204	150
	≤65	163	283
	Unknow	3	-
Gender	Female	133	137
	Man	237	296
M	M0	327	-
	M1	25	-
	Mx	18	-
N	N0	108	80
	N1	96	188
	N2	74	132
	N3	74	33
	Nx	16	-
Stage	I	49	11
	II	111	38
	III	149	92
	IV	38	292
Treatment	Pharmaceutical Therapy	185	-
	Radiation Therapy	185	-

Figure 1. Prognostic differentially expressed CSRs. a. Heatmap of differentially expressed CSRs. b. Volcano plots of differentially expressed CSRs. c. UCR analysis for the CSRs identification in the TCGA patient cohort

Figure 2. Biological function analysis of differentially expressed CSRs: GO analysis (a) and KEGG pathways analysis (b)

Table 2. The coefficients of each gene

id	coef	HR	HR.95 L	HR.95 H	pvalue
IFNB1	2.180261	8.848614	0.556624	140.6657	0.122384
IFNA4	4.871504	130.517	0.371338	45,873.81	0.103366
IL6	0.135794	1.145446	1.015096	1.292535	0.027592
NFKB2	−0.36677	0.692966	0.526914	0.911348	0.008687
TRIM25	−0.46663	0.627114	0.462248	0.850782	0.002715

Figure 3. Risk score, reflecting overall survival, based on the CSRs signature comprising five genes, in the training and validation cohorts. (a–b) Kaplan-Meier curve for OS of patients with high- and low risk scores in the training and validation cohorts. (c–d) Risk-score distribution in the training and validation cohorts. (e–f) The survival status plot associated with risk score in the training and validation cohorts. (g–h) Heatmap of the expression of the five CSRs in the high- and low-risk groups and the training and validation cohorts

Figure 4. The model was combined with a regression analysis of clinical indicators. (a-b) Assessment of the contribution of each factor to GC survival by UCR and MCR analysis in training cohort. (c, d) Assessment of the contribution of each factor to GC survival by UCR and MCR analysis in verification cohort

Figure 5. The ROC curves and nomograms for predicting survival rate of GC. (a, c) ROC curves (receiver operating characteristics). (b, d) Nomogram

Table 3. cGAS-STING pathway related gene sets that associated with high-risk group

NAME	NES	NOM p-val	FDR q-val	FWER p-val
KEGG_T_CELL_RECEPTOR_SIGNALING_PATHWAY	−1.904	0.006	0.069	0.163
KEGG_B_CELL_RECEPTOR_SIGNALING_PATHWAY	−1.760	0.031	0.113	0.404
KEGG_TOLL_LIKE_RECEPTOR_SIGNALING_PATHWAY	−1.736	0.025	0.130	0.458
KEGG_CHEMOKINE_SIGNALING_PATHWAY	−1.566	0.048	0.184	0.743
KEGG_APOPTOSIS	−2.238	0.000	0.003	0.002
KEGG_RIG_I_LIKE_RECEPTOR_SIGNALING_PATHWAY	−2.194	0.000	0.003	0.004

ES, enrichment score; NES, normalized enrichment score; NOM, nominal; FDR, false discovery rate.

Figure 6. The GSEA of possible pathways of high-risk groups in GC

Figure 7. Correlation analysis of the risk score model and immune cell infiltration. (a) B cells, (b) DCs, (c) CD4 + T cells, (d) macrophages, (e) CD8 + T cells, (f) Neutrophils

Figure 8. Immune landscape of patients from High- and Low-Riskgroups. Differences of 22 subtypes of immune cells between two groups of TCGA dataset(a), and GEO database (b)

Figure 9. Kaplan–Meier analysis of 22 kinds of immune cells. (a-b) The levels of T cells CD4 memory activated (a) and T cells follicular helper (b) were significantly associated with OS in patients with GC in TCGA database. (d-f) The levels of T cells CD4 memory activated (c), T cells follicular helper (d), B cells memory (e) and Mast cells resting (f) were significantly associated with OS in patients with GC in GEO database

Availability of data and materials

The datasets analyzed was acquired from The Cancer Genome Atlas (TCGA) database (https://portal.gdc.cancer.gov/) and GEO database (https://www.ncbi.nlm.nih.gov/geo/).