Identification and Validation of Biglycan as Prognosis and Therapy Markers for Patients with Stomach Adenocarcinoma

Figures & data

Figure 1 The mRNA expression of BGN in STAD. (A) Expression profile across tumor samples and paired normal tissues. (B) mRNA Expression profile in cancer cell lines. (C) Differential expression of BGN in normal and STAD tissues. *P<0.05 and ***P<0.001.

Abbreviations: BGN, biglycan; STAD, stomach adenocarcinoma.

Figure 2 The association of BGN mRNA expression with clinical phenotype of patients with STAD. ***P<0.001.

Abbreviations: BGN, biglycan; STAD, stomach adenocarcinoma.

Figure 3 The BGN protein expression. (A) Immunohistochemical images in normal and STAD tissues. (B) Protein location of BGN in cancer cell. (C) The immunofluorescence staining of BGN in cancer cell.

Abbreviations: BGN, biglycan; STAD, stomach adenocarcinoma.

Figure 4 The genetic alteration of BGN in STAD. (A) Mutation sites. (B) Genetic alteration based on cancer type. (C) Distribution of different types of mutations for BGN in stomach cancer.

Abbreviation: BGN, biglycan.

Figure 5 Prognosis analysis on immune cells in stomach adenocarcinoma.

Figure 6 Spearman correlations between abundance of cytokine and BGN expression in STAD. (A) Tumor-infiltrating lymphocytes. (B) Immunoinhibitor. (C) Immunostimulator. (D) Chemokine.

Abbreviations: BGN, iglycan; STAD, stomach adenocarcinoma.

Figure 7 Comparison analysis on BGN expression. (A) Expression difference between responders and non-responders. Points in the above scatter plot represent the expression difference of BGN in various data sets. Oval marker: dataset showing statistical significance. (B) Mutation difference between responders and non-responders. Points in the above scatter plot represent the mutation difference of BGN in various data sets. (C) Expression difference based on immune subtypes in STAD. (D) Expression difference based on molecular subtypes in STAD.

Abbreviations: BGN, biglycan; STAD, stomach adenocarcinoma.

Figure 8 The influence of BGN expression on the overall survival of patients.

Abbreviations: HR, hazard ratio; BGN, biglycan.

Table 1 The Restrict Prognosis Analysis Between BGN High and Low Expression Groups

Characteristics			HR (95% CI)	P-value
Sub-type	Clinical stage	Stage 1	1.52 (0.43–5.39)	0.510
		Stage 2	2.24 (0.94–5.31)	0.061
		Stage 3	2.14 (1.47–3.12)	5.00E-05
		Stage 4	1.65 (1.04–2.63)	0.032
	Stage T	T1	/	/
		T2	1.83 (1.05–3.19)	0.031
		T3	1.32 (0.88–1.99)	0.180
		T4	1.99 (0.66–5.98)	0.210
	Stage N	N0	1.26 (0.42–3.78)	0.680
		N1	3.07 (1.63–5.77)	2.40E-04
		N2	2.4 (1.32–4.38)	3.10E-03
		N3	2.09 (1.08–4.07)	0.0260
	Stage M	M0	2.22 (1.51–3.26)	2.90E-05
		M1	1.19 (0.6–2.35)	0.610
	Lauren classification	Intestinal	3.11 (1.94–4.98)	6.40E-07
		Diffuse	2.15 (1.37–3.37)	6.60E-04
		Mixed	1.38 (0.44–4.38)	0.580
	Differentiation	Poorly differentiated	1.79 (1.03–3.1)	0.036
		Moderately differentiated	1.26 (0.6–2.63)	0.540
		Well differentiated	2.5 (0.73–8.5)	0.130
Clinical cohorts	Gender	Female	1.94 (1.24–3.06)	3.40E-04
		Male	1.78 (1.36–2.34)	2.30E-05
	Perforation	No	1.54 (0.94–2.53)	0.085
		Yes	/	/
	Treatment	Surgery alone	1.9 (1.31–2.75)	6.20E-04
		5 FU based adjuvant	0.91 (0.61–1.35)	0.620
		Others adjuvant	1.98 (0.58–6.75)	0.270
	HER2 status	Negative	1.71 (1.28–2.28)	2.00E-04
		Positive	1.97 (1.41–2.74)	4.30E-05

Abbreviations: T, topography; N, lymph node; M, metastasis; HER2, human epidermal growth factor receptor-2.

Table 2 Cox Regression Results for BGN in Human Cancers

Cancer	Cox Coefficient	P-value	FDR Corrected	Median Expression
BLCA	0.173	2.80E-02	1.71E-01	4840.94
BRCA	0.029	7.50E-01	9.12E-01	13534.75
CESC	0.145	2.70E-01	6.16E-01	3519.3
COAD	0.227	2.70E-02	3.36E-01	3668.32
ESCA	0.002	9.90E-01	9.95E-01	7597.77
GBM	0.083	3.90E-01	8.97E-01	5275.09
HNSC	0.052	4.60E-01	7.59E-01	7745.8
KIRC	0.154	5.30E-02	1.15E-01	18792.74
KIRP	0.501	1.30E-03	1.21E-02	3734.77
LAML	−0.183	1.10E-01	4.86E-01	20.07
LGG	0.262	5.90E-03	1.56E-02	1158.44
LIHC	−0.047	6.20E-01	8.15E-01	3959.11
LUAD	0.067	3.60E-01	6.00E-01	11285.2
LUSC	0.084	2.00E-01	7.13E-01	8774.25
OV	0.126	1.00E-01	6.69E-01	6992.26
PAAD	−0.025	8.20E-01	9.06E-01	24473.83
READ	0.379	8.10E-02	9.32E-01	3985.44
SARC	0.022	8.30E-01	9.29E-01	19533.82
SKCM	0.144	3.20E-02	1.31E-01	4732.43
STAD	0.202	1.70E-02	2.36E-01	9300.53
UCEC	0.149	1.40E-01	9.64E-01	4548.51

Figure 9 Co-expressed genes analysis. (A) Whole protein–protein interaction (PPI) network. (B) MCODE network with highest score in PPI network. MXRA8: seed gene.

Table 3 The Top 5 Consistent KEGG Pathways Predicted in Three Databases

Pathway ID and Name	WebGestalt (P-value)	KOBAS (Corrected P-value)	String (False Discovery Rate)
hsa04974: Protein digestion and absorption	1.91E-13	2.35E-15	5.63E-13
hsa04512: ECM–receptor interaction	6.94E-06	1.03E-6	1.21E-05
hsa04510: Focal adhesion	2.27E-04	1.78E-5	2.40E-04
hsa05165: Human papillomavirus infection	1.71E-03	6.34E-5	7.30E-04
hsa04151: PI3K-Akt signaling pathway	2.00E-03	7.12E-5	7.30E-04

Abbreviation: KEGG, KEGG-Kyoto Encyclopedia of Genes and Genomes.

Figure 10 MXRA8 expression analysis. (A) Correlation of BGN with MXRA8 adjusted by tumor purity. (B) Expression comparison in normal/STAD tissues. (T: tumor; N: normal). (C) Survival analysis on MXRA8.

Abbreviations: BGN, biglycan; MXRA8, matrix remodeling associated 8; HR, hazard ratio.