B7-H3 and CD39 Co-Localization in Gastric Cancer: A Potential Prognostic Biomarker and Potential Dual-Target for Immunotherapy

Abstract

Introduction

Gastric cancer (GC) is a highly heterogeneous malignancy, necessitating novel therapeutic targets. B7-H3 and CD39, as immune checkpoints, are potential modulators of the tumor microenvironment and may influence the efficacy of immunotherapies.

Methods

B7-H3, CD39, and CD8 expression was assessed via immunohistochemistry (IHC) in 268 GC tissues and 80 gastric precancerous lesions. The correlation between B7-H3 and CD39 expression was analyzed using Spearman's correlation. Multiplex immunohistochemistry (m-IHC) was employed to determine the co-localization of B7-H3 and CD39 in GC tissues. Kaplan-Meier survival analysis and Cox regression models were utilized to evaluate clinical outcomes in different patient subgroups.

Results

Both B7-H3 and CD39 expression showed a stepwise increase during gastric carcinogenesis including chronic superficial gastritis (CSG), chronic atrophic gastritis (CAG), low-grade intraepithelial neoplasia (LGIN), high-grade intraepithelial neoplasia (HGIN) to GC, with significantly higher expression levels in GC tissues compared to all precancerous lesions (P < .05). A significant positive correlation was observed between B7-H3 and CD39 expression (r = 0.2398, P < .001). Co-localization of B7-H3 and CD39 was detected within tumor nests and peritumoral regions and was significantly correlated with tumor volume (P = .017), tumor stage (P = .001), tumor depth (P = .002), lymph node metastasis (P = .005), lymph node involvement (P = .004) and distant metastasis (P = .028). Kaplan-Meier analysis revealed that patients with co-localized B7-H3 and CD39 expression exhibited significantly poorer prognosis (P = .0055). Cox regression analysis confirmed that this co-localization was a significant predictor of survival (P = .007) and an independent prognostic factor in multivariate analysis (P = .027).

Conclusion

The co-localized expression of B7-H3 and CD39 in GC patients is strongly associated with poor prognosis. This dual-target expression pattern provides novel insights and a theoretical foundation for the development of dual-target immune checkpoint inhibitors as potential therapeutic strategies.

Keywords

B7-H3 CD39 gastric cancer dual-target immunotherapy

Introduction

Gastric cancer (GC) ranks as the third leading cause of cancer-related mortality worldwide, with approximately 1.5 million new cases diagnosed annually. The disease predominantly affects middle-aged and older men, especially in East Asia.^1,2 Advances in tumor microenvironment (TME) research have catalyzed progress in GC immunotherapy, particularly with the development of PD-1/PD-L1 inhibitors. While these therapies show promise in improving patient survival, they are hindered by low response rates and immune-related adverse events.^3,4 Consequently, alternative immunotherapeutic targets, such as B7-H3 and CD39, are gaining increased attention.

B7-H3 (CD276) is a co-stimulatory molecule characterized by low expression in normal tissues but significantly elevated levels in tumor tissues.^5–7 It facilitates immune evasion by impairing cytotoxic T cells and natural killer cells,⁸ and contributes to dysregulated bio-signaling and altered energy metabolism in tumor cells.^9,10 B7-H3 is a promising therapeutic target, with monoclonal antibodies and CAR-T cell therapies in clinical development showing encouraging results. Notably, DS-7300 and ¹³¹I-omburtamab have demonstrated safety, with DS-7300 progressing to Phase II trials following positive Phase I outcomes.^11,12

CD39, first identified in Epstein-Barr virus-infected B-lymphocytes in 1982, is another critical immune molecule in the TME. As an ectonucleoside triphosphate diphosphohydrolase, CD39 regulates the conversion of extracellular ATP (eATP) into adenosine, balancing the pro-inflammatory immune response and the immunosuppressive effects of adenosine.^13,14 CD39 is frequently overexpressed in various solid tumors, with elevated levels correlating with poor prognosis and disease progression.^15–17 Targeting the adenosine signaling pathway—specifically through CD39 inhibition to prevent adenosine accumulation—represents an emerging therapeutic strategy.^16,18 Combining CD39 inhibitors with anti-PD1/PD-L1 therapies has demonstrated superior efficacy compared to monotherapy, although resistance to PD1 inhibitors in some patients underscores the need for innovative treatment approaches.

Despite their distinct roles in cancer progression and immune evasion, the synergistic effects of B7-H3 and CD39 in GC remain poorly understood. Our previous research identified B7-H3 as a critical prognostic marker, with elevated expression in the TME strongly associated with poorer survival outcomes.^19,20 Given the pivotal roles of B7-H3 and CD39 in shaping the TME and modulating immune responses, as well as the promising potential of dual-target therapies in oncology,^21,22 further investigation into their interplay in GC is warranted. To clarify the terminology used in this study, we define “both high expression” as the concurrent high expression of both B7-H3 and CD39 within the same patient tissue sample. In contrast, “co-localization” refers to the specific spatial overlap where both molecules are expressed at the exact same physical location within the tissue architecture, as determined by techniques like multiplex immunohistochemistry. This study investigates both the phenomenon of concurrent high expression of B7-H3 and CD39 at the sample level and their precise spatial co-localization at the architectural level, as these distinct aspects provide complementary insights into the roles of these molecules in GC. In this study, we assessed B7-H3 and CD39 expression at distinct development stages of gastric carcinogenesis. We further analyze the relationship between B7-H3 and CD39 expression in GC. This study aims to address this gap by analyzing the co-expression of B7-H3 and CD39 in GC, evaluating their combined prognostic significance, and exploring their potential as biomarkers to inform immunotherapeutic strategies.

Materials and Methods

Patients and Specimens

This retrospective cohort study was conducted in accordance with the REMARK guidelines for tumor marker prognostic studies.²³ We collected specimens in different pathological stages of the precancerous gastric disease acquired from Pathology of the hospital from 2016 to 2021, obtaining 20 cases of chronic superficial gastritis (CSG), chronic atrophic gastritis (CAG), low-grade intraepithelial neoplasia (LGIN), high-grade intraepithelial neoplasia (HGIN) respectively. Additoonally, a total of 268 gastric cancer samples were collected from patients diagnosed in 2011 at a medical institution. Consecutive eligible patients meeting the inclusion criteria during the study period were enrolled.

Gastroscopic biopsy specimens were obtained from all cases except GC, which were collected through surgical resection.

All tissue samples, regardless of collection method, were harvested from the gastric antrum. All diagnoses were independently confirmed by two certified pathologists. Post-surgical resection specimens were fixed in neutral-buffered formalin and subsequently embedded in paraffin for histological analysis. This retrospective study incorporated a comprehensive dataset of clinicopathological characteristics, including age, gender, tumor volume, tumor differentiation, stage, and depth (as detailed in Table 1). Follow-up data for all 268 patients were collected from medical records and family contacts. Ethical approval for the study was granted by the Clinical Research Ethics Committee of the hospital. All GC patients underwent radical gastrectomy without neoadjuvant therapy.Based on prior GC biomarker studies anticipating Hazard Ratio (HR) ≥ 1.8 for co-expression markers (α = 0.05, power = 80%), 250 cases were required. Our cohort (n = 268) exceeded this threshold. All patient data were de-identified prior to analysis to ensure complete anonymity, with no possibility of tracing any individual's identity. The inclusion and exclusion criteria were defined as follows:

Table 1.

Correlations Between B7-H3 and CD39 Expression Status and Clinicopathological Features.

Clinical or pathological characteristics	Total	B7-H3		P	CD39		P	B7-H3-CD39		P	B7-H3-CD39		P
Clinical or pathological characteristics	Total	Low	High	P	Low	High	P	Both-high	Others	P	co-localization	Others	P
All case	268	122	146		100	168		108	160		195	73
Sex				.083			.024			.227			.947
Male	210	94	116		71	139		89	121		153	57
Female	58	28	30		29	29		19	39		42	16
Age(years)				.901			<.001			.250			.574
<70	164	71	93		42	106		71	93		117	47
≥70	104	51	53		58	62		37	67		78	26
Tumor volumes (cm²)				.943			.477			.499			.017
<5	186	87	99		72	114		72	114		127	59
≥5	82	35	47		28	54		36	46		68	14
Tumor stage				.041			.036			.110			.001
0	11	6	5		2	9		4	7		3	8
1	32	18	14		16	16		8	24		19	13
2	66	38	28		31	35		22	44		47	19
3	123	48	75		36	87		59	64		97	26
4	36	12	24		15	21		15	21		29	7
Tumor depth				.037			.172			.015			.002
T1	36	23	13		14	22		10	26		17	19
T2	34	19	15		18	16		7	27		24	10
T3	169	68	101		60	109		78	91		132	37
T4	29	12	17		8	21		13	16		22	7
Degree of tumor differentiation				.960			.504			.453			.441
1	6	4	2		3	3		1	5		3	3
2	121	55	66		41	80		51	70		88	33
3	141	63	78		56	85		56	85		104	37
Distant metastasis				.155			.261			.699			.028
M0		112	126		86	152		97	141		168	70
M1		10	20		14	16		11	19		27	3
Lymph metastasis				.373			.245			.0612			.005
N0	85	46	39		36	49		27	58		52	33
N1	183	76	107		64	119		81	102		143	40
Lymph node involvement				.006			.342			.020			.004
N0	85	46	39		36	49		27	58		33	52
N1	62	35	27		26	36		20	42		17	45
N2	56	17	39		17	39		30	26		15	41
N3	65	24	41		21	44		31	34		8	57
N3	65	24	41		21	44		31	34		8	57

Inclusion Criteria

Samples meeting all of the following conditions were included:

Histopathologically confirmed gastric adenocarcinoma treated with radical gastrectomy;

Tumor cellularity > 20% with preserved tissue integrity (no severe degradation or inadequate fixation) in archival FFPE blocks;

Availability of complete clinicopathological data (including age, sex, tumor location, histological grade, and pTNM stage [AJCC eighth edition]);

Absence of preoperative neoadjuvant therapy (chemotherapy or radiotherapy).

Exclusion Criteria

Samples were excluded if they met any of the following criteria:

Non-adenocarcinoma histology (eg, neuroendocrine carcinoma, lymphoma, or gastrointestinal stromal tumor [GIST]);

Non-radical surgical procedures (palliative resection, exploratory laparotomy, or endoscopic submucosal dissection/mucosal resection [ESD/EMR] specimens);

Inadequate tumor content or compromised sample quality (eg, excessive decalcification, autolysis);

Concurrent active autoimmune disorders (eg, systemic lupus erythematosus, rheumatoid arthritis), HIV infection, or active syphilis;

Missing essential clinical information (particularly pTNM staging data).

Tissue Microarray (TMA) Construction

Tissue Microarray (TMA) was constructed using an automated tissue array instrument. Each paraffin-embedded specimen was divided into two 1 mm cores, each containing tumor cell tissue and infiltrating immune cell tissue. Subsequently, 4-μm-thick TMA sections were created.

Immunohistochemistry and Scoring

4-μm-thick sections were cut from paraffin-embedded tissue blocks, mounted on polylysine-coated slides, and then deparaffinized and rehydrated using xylene and graded alcohol. The sections were treated by autoclaving in sodium citrate retrieval solution for 3 min. Endogenous peroxidase was blocked with 3% hydrogen peroxide for 15 min at room temperature. After washing twice in PBST, sections were blocked with 3% bovine serum albumin for 45 min, and then incubated overnight at 4 °C with primary antibodies. The following primary antibodies were used according to the manufacturer's instructions: mouse anti-human B7-H3 monoclonal antibody, 1/200 dilution (Cat No: 66 481-1-Ig, Proteintech, China), rabbit anti-human CD39 polyclonal antibody, 1/1000 dilution (Cat No: 14211-1-AP, Proteintech, China) and mouse anti-human CD8 monoclonal antibody, 1/10 000 dilution (Cat No: 66868-1-Ig, Proteintech, China). After primary antibody binding, slides were washed with PBST and then incubated with labeled polymer horseradish peroxidase rabbit/mouse antibody for 15 min (GP016129; Gene Technology, China). Immunoreactivity was detected using DAB reagent, after which the slides were counterstained with hematoxylin and sealed with neutral gum. Immunohistochemical sections were examined blindly and independently by two pathologists and scanned simultaneously by a Dmetrix imaging system (Dmetrix, Tucson, AZ, USA) with a K-Viewer imaging system (Jiang feng Bio, China). Sections were examined by staining at low magnification (×100). B7-H3 staining was scored by immunostaining intensity and ratio. Intensity was scored at high magnification (×200) and defined as follows: 0, 1+, 2+, and 3+. The percentage of B7-H3-positive cells in diseased glands or total tumor cells was estimated at four levels: 1 (<10%), 2 (10%-40%), 3 (40%-70%), and 4 (>70%). The immunoreactivity score was calculated as intensity multiplied by the percentage of stained cells. Scores of 0 were classified as no expression, 1-3 as weak, 4-7 as moderate, and 8 or more as strong. Patients with no and weak expression were categorized into the low expression group, and others into the high expression group. CD39 staining was scored by mean optical density intensity, defined as follows: 0, 1 (<10%), 2 (10%-50%), 3 (>50%). Scores of 0 were classified as no expression, 1 as weak expression, 2 as moderate expression, and 3 as high expression. Similarly, patients with no expression and weak expression were categorized into the low expression group, and others into the high expression group.

Multiple Immunohistochemistry and Scoring

Multiplex immunohistochemistry (mIHC) and antibodies specific for B7-H3, CD39, and CD8 were performed on gastric cancer TMA tissues as described above. Briefly, slides were dewaxed in xylene and then rehydrated with ethanol. Antigen retrieval was performed at boiling point in sodium citrate buffer (pH 6.0) for 15 min and endogenous peroxidase activity was blocked with 3% hydrogen peroxide for 15 min at room temperature. Nonspecific antigen binding was blocked with goat serum solution for 30 min. Slides were incubated with the primary antibody B7-H3 (1:200, Cat No: 66481-1-Ig, Proteintech, China) overnight at 4 °C, and then horseradish peroxidase (HRP)-coupled secondary antibody was added for 30 min at room temperature. Next, slides were incubated with tyramide signal amplification (TSA) Fluorescein (Cat No: NEL701A001KT, Akoya) for 10 min at room temperature. Between subsequent staining runs, slides were microwaved to remove the Ab-TSA complex and blocked with goat serum solution, and the CD39 primary antibody was incubated between this cycle using TSA Cyanine 3 (Cat No: NEL704A001KT, Akoya) to correspond to the label. Similarly, in the last run, CD8 was visualized using TSA Cyanine 5 (Cat No: NEL705A001KT, Akoya), and finally, nuclei were visualized with DAPI (DAPI Fluoromount-G, Southern Biotech, USA). Representative images of each sample were captured and analyzed by Mantra and Inform software (Akoya Biosciences). Pathologic tissues from each patient were randomized in 4 multispectral regions (200x magnification field of view), the co-localization of B7-H3 and CD39 was quantified using a scoring system based on the percentage of double positive area, ranging from 0 to 12. A score of 0 indicates no co-localization expression, while scores of 1-4 represent low co-localization expression, 5-8 indicate moderate co-localization expression, and 8-12 denote strong co-localization expression. For the purpose of this study, moderate and strong co-localization expressions were considered as positive B7-H3 and CD39 co-localization. Additionally, CD8⁺ T cells, CD39⁺ CD8⁺ T cells were automatically quantified in each field of view region. The average distribution density of cell types was calculated for each patient's tissue locus. The median density of immune cells of interest in all tumors was used to classify tumors into high and low infiltration groups.

Statistical Analysis

Statistical analyses were performed using GraphPad prism 9.0 and SPSS software. Overall survival (OS) was defined as the time from the start of treatment to the date of death. It should be emphasized that OS was recorded in only 198 patients. The chi-square test was used to analyze the correlation between categorical variables. Correlations between variables were analyzed using Spearman rank correlation analysis. Survival analysis was performed using Kaplan-Meier curves, and Cox proportional risk models were used for univariate and multivariate analyses. P < .05 was defined as statistically significant difference.

Results

B7-H3 and CD39 Expression at Different Pathological Stages of Precancerous Diseases of the Stomach

To investigate the expression and correlation of B7-H3 and CD39 in gastric cancer tissues, we first examined the expression levels of these two markers in gastric cancer tissues and non-neoplastic control specimens. During the progression of gastric precancerous lesions, B7-H3 and CD39 expression levels demonstrated significant dynamic changes. As shown in Figure 1A, immunohistochemical (IHC) results revealed representative expression patterns of both proteins across different pathological stages (CSG, CAG, LGIN, HGIN, and GC). Quantitative analysis indicated that the expression intensities of B7-H3 and CD39 in GC tissues were significantly higher than those in other precancerous stages (P < .05). Multiplex immunohistochemistry (m-IHC) further identified co-localized expression of B7-H3 and CD39 within tumor nests and peritumoral areas in GC (Figure 1D and E). B7-H3 was primarily expressed in tumor and stromal cells, while CD39 exhibited a broader distribution across tumor cells, stromal cells, and immune cells.

Figure 1.

Expression Patterns of B7-H3 and CD39 in Gastric Precancerous Lesions and Gastric Cancer Tissues. (A) Representative IHC Expression Patterns of B7-H3 and CD39 Across Different Pathological Stages (CSG, CAG, LGIN, HGIN, and GC). (B) Quantitative Analysis of Positive Expression Areas of CD39. (C) Quantitative Analysis of Positive Expression Areas of B7-H3. (D) Representative Multiplex IHC Expression Patterns of B7-H3 and CD39 Across Different Pathological Stages (CSG, CAG, LGIN, HGIN, and GC). (E) Quantitative Analysis of Multiplex IHC Expression of B7-H3 and CD39. (F) Representative Co-Localization Expression of B7-H3 and CD39 in GC. IHC Scale Bars: 50 µm; mIHC Scale Bars: 100 µm.

Co-Localized Expression of B7-H3 and CD39 in Gastric Cancer: Correlation with Clinicopathologic Factors and Prognosis

Furthermore, we analyzed B7-H3 and CD39 expression in 268 gastric cancer tissue samples using IHC (Figure 2A and B). High B7-H3 expression was observed in 146 samples, while high CD39 expression was found in 168. Spearman correlation analysis revealed a positive relationship between B7-H3 and CD39 expression in gastric cancer tissues (r = 0.2398, P < .0001; Figure 2C).

Figure 2.

Co-Localization of B7-H3 and CD39 in Gastric Cancer Cells Indicates Poor Prognosis. (A, B) Representative Immunohistochemical Images Showing Low and High Expression of B7-H3 (A) and CD39 (B) in Gastric Cancer (GC) Specimens. (C) Correlation Analysis of B7-H3 and CD39 Expression. (F, G, H, I) Kaplan-Meier Survival Curves for Overall Survival of GC Patients Based on the Expression Status of B7-H3 (F), CD39 (G), Dual High Expression of B7-H3 and CD39 (H), and Co-Localized Expression Status of B7-H3-CD39 (I). Scale Bars: 100 µm.

Correlation analysis showed significant correlations between B7-H3 expression and tumor stage (P = .041), tumor depth (P = .037), and lymph node involvement (P = .006). The CD39 expression correlated only with tumor stage (P = .036). However, combined high expression of B7-H3 and CD39 was significantly associated with tumor depth (P = .015) and lymph node involvement (P = .020). Co-localized expression of B7-H3 and CD39 was strongly correlated with clinicopathological features, including tumor volume (P = .017), tumor stage (P = .001), tumor depth (P = .002), distant metastasis (P = .028), lymph metastasis (P = .005), and lymph node involvement (P = .004) (Table 1).

Prognostic analysis revealed that patients with high B7-H3 expression had significantly poorer OS compared to those with low expression (P = .0130; Figure 2D). CD39 expression alone was not significantly associated with OS (P = .1771; Figure 2E). Notably, patients with combined high B7-H3 and CD39 expression exhibited worse OS (P = .0432; Figure 2F), with co-localization of high expression marking the poorest prognosis (P = .0055; Figure 1G).

Correlation of B7-H3 and CD39 Expression with CD8⁺ T Cell Infiltration in Gastric Cancer

Given the critical role of CD8⁺ T cells in antitumor immunity, we explored the correlation between B7-H3 and CD39 expression and CD8⁺ T cell infiltration. IHC revealed no significant correlation between B7-H3 expression and CD8 (r = 0.0756, P = .2371) or between CD39 and CD8 (r = 0.0948, P = .1365; Figure 3A-C). Similarly, m-IHC showed no significant correlation between co-localized B7-H3 and CD39 expression and CD8⁺ T cell infiltration (r = 0.0208, P = .7344; Figure 3D and E). However, a notable enrichment of CD39⁺ CD8⁺ T cells was observed in tumor stromal regions, comprising 66% of the total CD8⁺ T cell population (Figure 3F and G).

Figure 3.

Absence of Correlation Between Co-localization of B7-H3 and CD39 Expression and CD8⁺ T Cell Infiltration in Gastric Cancer. (A, B) Representative Immunohistochemical Images Showing the Expression of B7-H3 with CD8 (A) or CD39 with CD8 (B) in the Same Patient. (C) Correlation Between B7-H3 Expression and CD8 Expression, and Between CD39 Expression and CD8 Expression. (D, E) Representative Multiplex Immunohistochemistry Images of B7-H3, CD39, and CD8 (D), and Correlation Analysis Between the Co-Localization Expression Score of B7-H3 and CD39 and the Extent of CD8 Infiltration (E). (F, G) Representative Multiplex Immunohistochemistry Images of CD39 and CD8, with Arrows Indicating CD39⁺ CD8⁺ T Cells (F), and the Proportion of CD39⁺ CD8⁺ T Cells within the CD8⁺ T Cell Population (G). Scale Bars: 100 µm.

Prognostic Significance of Co-Localized B7-H3 and CD39 Expression with CD8⁺ T Cell Infiltration

We analyzed the prognostic impact of B7-H3 and CD39 expression in relation to CD8⁺ T cells. Patients with high B7-H3 but low CD8 expression exhibited significantly reduced survival rates (P = .0159; Figure 4A), while high CD39 with low CD8 expression showed no significant impact (P = .3883; Figure 4B). Combined high expression of B7-H3 and CD39 with low CD8 expression also showed no significant survival difference (P = .1994; Figure 4C). However, co-localization of high B7-H3 and CD39 expression with low CD8 was associated with the poorest survival outcomes (P = .0074; Figure 4D). Although CD8⁺ T cell infiltration alone was not significantly associated with survival (P = .7059; Figure 4E), patients with low CD39⁺ CD8⁺ T cell expression had markedly poorer survival compared to those with high infiltration (P = .0060; Figure 4F).

Figure 4.

Kaplan-Meier Survival Curves for Gastric Cancer Patients Based on Specific Expression Statuses and Immune Cell Infiltration Levels. (A, B, C, D) Kaplan-Meier Survival Curves for GC Patients Stratified by B7-H3^HighCD8^Low (A) Expression Status 、 CD39^HighCD8^Low (B) Expression Status 、 B7-H3-CD39^{Both high}CD8^Low Expression Status (C) and B7-H3-CD39^{Co-localization}CD8^low Expression Status (D). (E, F) Kaplan-Meier Survival Curves for GC Patients Based on CD8⁺ T Cell Infiltration Levels (E) and CD39⁺ CD8⁺ T Cell Infiltration Status (F).

Univariate Cox regression analysis identified tumor size, stage, depth, distant metastasis, high B7-H3 expression, and co-localization of B7-H3 and CD39 as significant predictors of survival (P < .05; Table 2). Multivariate Cox regression confirmed that co-localized B7-H3 and CD39 expression remained an independent prognostic factor for survival (P = .027; Table 2).

Table 2.

Cox Regression Analysis of Risk Factors for Death.

Variables	Univariate analysis			P	Multivariate Analysis			P
	HR	95% CI			HR	95% CI
	HR	Lower	Upper		HR	Lower	Upper
Sex (female vs male)	0.901	0.583	1.391	.638
Age (<70 years vs ≥70 years)	0.832	0.580	1.192	.315
Tumor volume (<5 cm² vs ≥5 cm²)	0.547	0.379	0.789	.001	0.699	0.476	1.027	.068
Tumor differentiation (well/moderate vs poor)	0.994	0.831	1.189	.948
Tumor stage (I, II, II vs IV)	0.403	0.252	0.644	<.001	0.321	0.116	0.893	.029
Tumor depth (T1, T2 vs T3, T4)	0.260	0.151	0.447	<.001	0.229	0.698	4.487	.288
Distant metastasis (M0 vs M1)	0.472	0.285	0.782	.004	1.709	0.547	5.339	.357
Lymph metastasis (N0 vs N1)	0.296	0.184	0.475	<.001	0.356	0.219	0.578	<.001
B7-H3 (Low vs High)	1.584	1.096	2.290	.014
CD39 (Low vs High)	1.308	0.882	1.940	.182
B7-H3-CD39 (Others vs Both High)	1.441	1.007	2.064	.046
B7-H3-CD39 (Others vs Co-localization expression)	0.543	0.349	0.844	.007	0.569	0.346	0.938	.027

Discussion

In this study, we identified a significant positive correlation between the expression of B7-H3 and CD39 in gastric cancer tissues, with notable co-localization in both tumor nests and peritumoral regions. While both molecules showed limited expression in gastric precancerous lesions, their coordinated upregulation peaked in GC tissues, suggesting a stage-specific synergy during malignant transformation. Multiplex immunohistochemistry confirmed that B7-H3 and CD39 colocalized in tumor and stromal compartments, yet exhibited distinct cellular tropism: B7-H3 was predominantly expressed in tumor and stromal cells, whereas CD39 additionally marked immune infiltrates. This tumor-enriched co-expression pattern implies a potential functional crosstalk between B7-H3 and CD39 in GC progression, challenging the conventional focus on their isolated roles in tumor biology. While previous studies have predominantly examined the individual roles of B7-H3 or CD39 in tumor biology, our findings based on data from 268 GC patients suggest that there may be a clinically relevant functional interplay between these two molecules.

CD39 primarily functions to hydrolyze Adenosine Triphosphate (ATP) into adenosine, which has dual effects: inhibiting the activity of T cells, Natural Killer (NK) cells, and dendritic cells (DCs) via A2A and A2B receptors, while simultaneously enhancing the function of regulatory T cells (Tregs), myeloid-derived suppressor cells (MDSCs), and M2 macrophages, thereby facilitating tumor immune evasion.²⁴ In contrast, B7-H3 suppresses T cell activity and enhances the secretion of immunosuppressive factors through interactions with unknown receptors on T or NK cells, promoting immune escape.²⁵ Given that patients with high co-localized expression of B7-H3 and CD39 exhibit the poorest prognosis, it is plausible that the synergistic immune evasion mechanisms mediated by these molecules play a critical role. Furthermore, multivariate analysis confirmed that co-localized high expression of B7-H3 and CD39 is an independent prognostic risk factor, supporting the potential of dual-targeted therapies against these molecules.

Beyond their immunosuppressive roles, B7-H3 and CD39 intersect in various oncogenic signaling pathways. B7-H3 activates the JAK/STAT signaling pathway, promotes epithelial-mesenchymal transition (EMT) through increased expression of matrix metalloproteinases (MMPs), and drives tumor migration, invasion, and metastasis.^26,27 The activation of STAT3 further regulates cytokines such as IL-6 and TGF-β, enhancing immune evasion and malignant phenotypes.^28–30 Additionally, B7-H3 contributes to metabolic reprograming by enhancing glycolytic capacity through increased Reactive Oxygen Species (ROS) production and HIF-1α stabilization.⁹ CD39 expression, in turn, is regulated by stimuli including hypoxia, TGF-β, EMT, oxidative stress, and inflammatory cytokines—many of which are influenced by B7-H3.^31,32 Hypoxia and TGF-β are particularly critical inducers of CD39 expression, suggesting a regulatory link between B7-H3 and CD39 signaling. Together, these molecules influence the metabolic landscape of the TME, with B7-H3 enhancing energy metabolism and ATP production during glycolysis,^33,34 while CD39 directly modulates ATP depletion and adenosine accumulation.^14,35 This metabolic synergy fosters a TME conducive to tumor proliferation and metastasis.

These findings underscore the therapeutic potential of targeting B7-H3 and CD39 through dual CAR-T cell therapies or bispecific antibodies. Such strategies could inhibit immune escape mechanisms, remodel the metabolic properties of the TME, and enhance the efficacy of antitumor immunity.

As key effectors in antitumor immunity, CD8⁺ T cells are critical targets of the immunosuppressive effects of B7-H3 and CD39. Our analysis revealed no significant correlation between B7-H3 or CD39 expression and CD8⁺ T cell infiltration, suggesting that these molecules may regulate T cells through mechanisms related to functional exhaustion rather than recruitment or proliferation.

Interestingly, we observed a substantial presence of CD39⁺ CD8⁺ T cells in gastric cancer tissues, particularly within the tumor stroma. While CD39⁺ CD8⁺ T cells are often considered bystanders associated with poor prognosis,^36,37 our findings demonstrated that high infiltration of CD39⁺ CD8⁺ T cells correlated with improved overall survival. This aligns with studies in other cancers, such as head and neck tumors, where CD103⁺ CD39⁺ CD8⁺ tumor-infiltrating lymphocytes (TILs) were linked to better survival outcomes.^38,39 These TILs exhibit a tissue-resident memory phenotype, enabling sustained immune surveillance and antitumor activity. This highlights the complexity of CD39's role in the TME, where its effects may vary depending on the cellular context. High expression of CD39 on CD8⁺ T cells may enhance immune surveillance and function rather than uniformly contributing to immune suppression. Further research is essential to delineate the specific regulatory roles of CD39 in different cell populations within the TME. This study also has several limitations. While we revealed the association between B7-H3/CD39 co-expression and poor prognosis and explored potential underlying mechanisms, direct experimental evidence elucidating the specific molecular interactions driving gastric cancer progression is lacking; moreover, the multiplex IHC approach employed cannot precisely resolve their subcellular co-localization or demonstrate direct molecular binding. As a single-center retrospective study, the generalizability of our findings requires validation in prospective multi-center cohorts. Finally, the therapeutic potential of dual targeting B7-H3 and CD39 remains to be supported by preclinical or clinical data.

Our study demonstrates the co-localized expression of B7-H3 and CD39 in gastric cancer tissues and establishes their association with poor prognosis. These findings provide a foundation for further exploration of their roles in tumor immunity and potential as therapeutic targets. Dual-targeting strategies that disrupt the interplay between B7-H3 and CD39 could reshape the TME, enhance immune recognition, and improve therapeutic outcomes in gastric cancer.

Conclusion

The present study observed that the expression of the immune-checkpoint molecules B7-H3 and CD39 increased progressively across the gastric carcinogenesis sequence, with the highest levels detected in established GC. Co-localization of these molecules was found to be associated with larger tumor volume, more advanced stage, deeper invasion, lymph-node involvement, distant metastasis, and reduced overall survival. These findings suggest this co-localization pattern may have potential as both a prognostic biomarker and a possible dual-target (B7-H3-CD39) for immunotherapy consideration. Multivariate analysis indicated this co-localization might serve as an independent prognostic biomarker. The observed lack of correlation with overall CD8⁺ T cell infiltration, combined with the prognostic associations, raises the possibility that this co-localization could contribute to an immunosuppressive microenvironment. Together, these results provide a basis for further investigation of dual B7-H3/CD39-targeted approaches as a potential therapeutic strategy in GC.

Footnotes

Abbreviations

ORCID iDs

Qiange Zhang

Lei Cao

Ethical Approval and Consent to Participate

This retrospective study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Ethics Committee of Soochow University (approval number: 2018121；2022030).

Informed Consent

Since this study was a retrospective study, the Ethics Committee of The First Affiliated Hospital of Soochow University has confirmed that no informed consent is required.

Author Contribution

All authors contributed to the study conception and design. Qiange Zhang and Ying Liu: Writing-original draft, Statistical analysis. Shenghua Zhan: Data curation, Resources. Cuiping Liu, Sisi Ding, Yu Shen, Lili Huang and Qi Ma: Resources, Funding acquisition. Siji Chen: Data collection. Hanqin Xuan and Ruipeng Wang: Material preparation. TingWang Jiang: Supervision. Lei Cao: Supervision, Data analysis, Writing-review & editing, Funding acquisition. All authors read and approved of the final manuscript.

Funding

This study was funded by grants from Science and Technology Project Plan of Suzhou (SKY2022129, SKY2022130), National Natural Science Foundation of China (81874163, 82270556), Suzhou Municipal Science and Technology Bureau (NO. SKJY2021086), the Project of State Key Laboratory of Radiation Medicine and Protection, Soochow University (NO. GZK1202218), Children's Hospital of Soochow University Key Research Program Incubation Project (2023ZDPY05) and Gusu Talent Project of Suzhou (GSWS2020106, GSWS2020011).

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Availability of Data and Material

All data used in this study are included in the article. Please contact the corresponding author for data requests.

References

Siegel

Miller

Wagle

Jemal

. Cancer statistics, 2023. CA Cancer J Clin. 2023;73(1):17–48. doi:10.3322/caac.21763

Kocarnik

Compton

Dean

, et al. Cancer incidence, mortality, years of life lost, years lived with disability, and disability-adjusted life years for 29 cancer groups from 2010 to 2019: a systematic analysis for the global burden of disease study 2019. JAMA Oncol. 2022;8(3):420. doi:10.1001/jamaoncol.2021.6987

Bejarano

Jordāo

MJC

Joyce

. Therapeutic targeting of the tumor microenvironment. Cancer Discov. 2021;11(4):933–959. doi:10.1158/2159-8290.CD-20-1808

Guan

. Gastric cancer treatment: recent progress and future perspectives. J Hematol Oncol. 2023;16(1):57. doi:10.1186/s13045-023-01451-3

Pan

Vieth

Gerhard

Mejías-Luque

. JAK-STAT1 signaling pathway is an early response to helicobacter pylori infection and contributes to immune escape and gastric carcinogenesis. Int J Mol Sci. 2022;23(8):4147. doi:10.3390/ijms23084147

Zhou

Jin

. B7-H3/CD276: an emerging cancer immunotherapy. Front Immunol. 2021;12:701006. doi:10.3389/fimmu.2021.701006

Amori

Sugawara

Shigematsu

, et al. Tumor B7-H3 expression in diagnostic biopsy specimens and survival in patients with metastatic prostate cancer. Prostate Cancer Prostatic Dis. 2021;24(3):767–774. doi:10.1038/s41391-021-00331-6

Grote

Ureña-Bailén

Chan

KCH

, et al. In vitro evaluation of CD276-CAR NK-92 functionality, migration and invasion potential in the presence of immune inhibitory factors of the tumor microenvironment. Cells. 2021;10(5):1020. doi:10.3390/cells10051020

Lim

Liu

Da Silva

, et al. Immunoregulatory protein B7-H3 reprograms glucose metabolism in cancer cells by ROS-mediated stabilization of HIF1α. Cancer Res. 2016;76(8):2231–2242. doi:10.1158/0008-5472.CAN-15-1538

10.

Shi

Cao

, et al. B7-H3 promotes aerobic glycolysis and chemoresistance in colorectal cancer cells by regulating HK2. Cell Death Dis. 2019;10(4):308. doi:10.1038/s41419-019-1549-6

11.

Kramer

Pandit-Taskar

Kushner

, et al. Phase 1 study of intraventricular 131I-omburtamab targeting B7H3 (CD276)-expressing CNS malignancies. J Hematol Oncol. 2022;15(1):165. doi:10.1186/s13045-022-01383-4

12.

Bendell

Doi

Patel

, et al. A phase I/II, two-part, multicenter, first-in-human study of DS-7300a in patients with advanced solid malignant tumors. J Clin Oncol. 2020;38(15_suppl):TPS3646–TPS3646. doi:10.1200/JCO.2020.38.15_suppl.TPS3646

13.

Van Den Bulk

Van Der Ploeg

Ijsselsteijn

, et al. CD103 And CD39 coexpression identifies neoantigen-specific cytotoxic T cells in colorectal cancers with low mutation burden. J Immunother Cancer. 2023;11(2):e005887. doi:10.1136/jitc-2022-005887

14.

Yegutkin

Boison

. ATP and adenosine metabolism in cancer: exploitation for therapeutic gain. Pharmacol Rev. 2022;74(3):799–824. doi:10.1124/pharmrev.121.000528

15.

Lin

Moscarelli

Zhu

Lin

Ratner

. CXCL10-induced regulatory T cells and adenosine signaling promote immunosuppression and progression of epithelial ovarian cancer. Sci Rep. 2025;15(1). doi:10.1038/s41598-025-06812-1

16.

Jacoberger-Foissac

Cousineau

Bareche

, et al. CD73 inhibits cGAS–STING and cooperates with CD39 to promote pancreatic cancer. Cancer Immunol Res. 2023;11(1):56–71. doi:10.1158/2326-6066.CIR-22-0260

17.

Zou

Tan

Liu

, et al. The CD39+ HBV surface protein-targeted CAR-T and personalized tumor-reactive CD8+ T cells exhibit potent anti-HCC activity. Mol Ther. 2021;29(5):1794–1807. doi:10.1016/j.ymthe.2021.01.021

18.

Maity

Ganguly

Deb

. Therapeutic potential of adenosine receptor modulators in cancer treatment. RSC Adv. 2025;15(26):20418–20445. doi:10.1039/d5ra02235e

19.

Chen

Zhan

Ding

, et al. B7-H3 and CD47 co-expression in gastric cancer is a predictor of poor prognosis and potential targets for future dual-targeting immunotherapy. J Cancer Res Clin Oncol. 2023;149(18):16609–16621. doi:10.1007/s00432-023-05408-4

20.

Zhan

Liu

Zhang

, et al. Overexpression of B7-H3 in α-SMA-positive fibroblasts is associated with cancer progression and survival in gastric adenocarcinomas. Front Oncol. 2020;9:1466. doi:10.3389/fonc.2019.01466

21.

Hirabayashi

, et al. Dual-targeting CAR-T cells with optimal co-stimulation and metabolic fitness enhance antitumor activity and prevent escape in solid tumors. Nat Cancer. 2021;2(9):904–918. doi:10.1038/s43018-021-00244-2

22.

Yang

Mou

Wan

, et al. Rethinking therapeutic strategies of dual-target drugs: an update on pharmacological small-molecule compounds in cancer. Med Res Rev. 2024;44(6):2600–2623. doi:10.1002/med.22057

23.

McShane

Altman

Sauerbrei

Taube

Gion

Clark

. REporting recommendations for tumour MARKer prognostic studies (REMARK). Br J Cancer. 2005;93(4):387–391. doi:10.1038/sj.bjc.6602678

24.

Moesta

Xiao

, et al. Targeting CD39 in cancer reveals an extracellular ATP- and inflammasome-driven tumor immunity. Cancer Discov. 2019;9(12):1754–1773. doi:10.1158/2159-8290.CD-19-0541

25.

Shi

Wang

, et al. B7-H3 inhibits the IFN-γ-dependent cytotoxicity of Vγ9Vδ2T cells against colon cancer cells. OncoImmunology. 2020;9(1):1748991. doi:10.1080/2162402X.2020.1748991

26.

Emaldi

Rey-Iborra

Marín

, et al. Impact of B7-H3 expression on metastasis, immune exhaustion and JAK/STAT and PI3K/AKT pathways in clear cell renal cell carcinoma. OncoImmunology. 2024;13(1):2419686. doi:10.1080/2162402X.2024.2419686

27.

Zhang

Huang

Gao

, et al. High expression of B7-H3 on monocyte/macrophages in tumor microenvironment promotes lung cancer progression by inhibiting apoptosis. Transl Oncol. 2024;41:101874. doi:10.1016/j.tranon.2023.101874

28.

Zhou

Mao

Zhu

, et al. TGF-β1 promotes colorectal cancer immune escape by elevating B7-H3 and B7-H4 via the miR-155/miR-143 axis. Oncotarget. 2016;7(41):67196–67211. doi:10.18632/oncotarget.11950

29.

Han

Wang

Shi

, et al. Negative roles of B7-H3 and B7-H4 in the microenvironment of cervical cancer. Exp Cell Res. 2018;371(1):222–230. doi:10.1016/j.yexcr.2018.08.014

30.

Lee

Park

Lee

, et al. MSN/STAT3 drives cancer stemness and chemoresistance via IL-6/LPAR1 ligand receptor complex in triple-negative breast cancer. Breast Cancer Res. 2025;27(1). doi:10.1186/s13058-025-02072-z

31.

Wang

Chen

, et al. CD39/CD73 upregulation on myeloid-derived suppressor cells via TGF-β-mTOR-HIF-1 signaling in patients with non-small cell lung cancer. OncoImmunology. 2017;6(6):e1320011. doi:10.1080/2162402X.2017.1320011

32.

Peres

Donate

Talbot

, et al. TGF-β signalling defect is linked to low CD39 expression on regulatory T cells and methotrexate resistance in rheumatoid arthritis. J Autoimmun. 2018;90:49–58. doi:10.1016/j.jaut.2018.01.004

33.

Huang

Zhang

, et al. FUT8-mediated aberrant N-glycosylation of B7H3 suppresses the immune response in triple-negative breast cancer. Nat Commun. 2021;12(1). doi:10.1038/s41467-021-22618-x

34.

Yang

Jia

, et al. Histone lactylation-driven B7-H3 expression promotes tumor immune evasion. Theranostics. 2025;15(6):2338–2359. doi:10.7150/thno.105947

35.

Sun

Han

Seth

, et al. Disordered purinergic signaling and abnormal cellular metabolism are associated with development of liver cancer in Cd39/Entpd1 null mice. Hepatology. 2013;57(1):205–216. doi:10.1002/hep.25989

36.

Canale

Ramello

Núñez

, et al. CD39 Expression defines cell exhaustion in tumor-infiltrating CD8+ T cells. Cancer Res. 2018;78(1):115–128. doi:10.1158/0008-5472.CAN-16-2684

37.

Simoni

Becht

Fehlings

, et al. Bystander CD8+ T cells are abundant and phenotypically distinct in human tumour infiltrates. Nature. 2018;557(7706):575–579. doi:10.1038/s41586-018-0130-2

38.

Talhouni

Fadhil

Mongan

, et al. Activated tissue resident memory T-cells (CD8+CD103+CD39+) uniquely predict survival in left sided “immune-hot” colorectal cancers. Front Immunol. 2023;14:1057292. doi:10.3389/fimmu.2023.1057292

39.

Duhen

Montler

, et al. Co-expression of CD39 and CD103 identifies tumor-reactive CD8T cells in human solid tumors. Nat Commun. 2018;9(1):2724. doi:10.1038/s41467-018-05072-0

B7-H3 and CD39 Co-Localization in Gastric Cancer: A Potential Prognostic Biomarker and Potential Dual-Target for Immunotherapy

Abstract

Introduction

Methods

Results

Conclusion

Keywords

Introduction

Materials and Methods

Patients and Specimens

Tissue Microarray (TMA) Construction

Immunohistochemistry and Scoring

Multiple Immunohistochemistry and Scoring

Statistical Analysis

Results

B7-H3 and CD39 Expression at Different Pathological Stages of Precancerous Diseases of the Stomach

Co-Localized Expression of B7-H3 and CD39 in Gastric Cancer: Correlation with Clinicopathologic Factors and Prognosis

Correlation of B7-H3 and CD39 Expression with CD8+ T Cell Infiltration in Gastric Cancer

Prognostic Significance of Co-Localized B7-H3 and CD39 Expression with CD8+ T Cell Infiltration

Discussion

Conclusion

Footnotes

Abbreviations

ORCID iDs

Ethical Approval and Consent to Participate

Informed Consent

Author Contribution

Funding

Declaration of Conflicting Interests

Availability of Data and Material

References

Correlation of B7-H3 and CD39 Expression with CD8⁺ T Cell Infiltration in Gastric Cancer

Prognostic Significance of Co-Localized B7-H3 and CD39 Expression with CD8⁺ T Cell Infiltration