Co-expression of cancer stem cell markers CD24 and CD133 in gastric cancer tissues: Clinicopathological and prognostic significance

Abstract

BACKGROUND

Gastric cancer (GC) is one of the most common malignant tumors of the digestive system worldwide, the prognosis of patients with advanced GC remains poor.

AIM

To evaluate the combined expression characteristics of cancer stem cell markers CD24 and CD133 in GC pathological tissues, and to explore their association with patients’ clinicopathological parameters and postoperative survival outcomes.

METHODS

A total of 304 GC patients who underwent surgical treatment in our hospital from January 2018 to January 2020 were retrospectively included. Immunohistochemistry was used to detect the protein expression of CD24 and CD133 in tumor tissues, adjacent tissues, and normal gastric mucosa tissues. Based on staining intensity and the proportion of positive cells, expression levels were classified into low and high expression, while clinicopathological parameters were recorded. χ² test was used to evaluate the correlation between expression and categorical variables, Spearman rank correlation analysis was performed to assess the correlation between the expression intensities of the two markers, and multivariate regression models were applied to identify independent risk factors influencing co-expression. Kaplan-Meier survival curves and Log-rank test were used to compare survival differences among groups with different expression patterns.

RESULTS

Among the 304 patients, 155 cases (50.99%) were CD24 positive, including 91 low-expression and 64 high-expression; 133 cases (43.75%) were CD133 positive, including 81 low-expression and 52 high-expression. There were 74 cases (24.34%) with double positivity and 81 cases (26.64%) with double negativity. Compared with tumor tissues, the positive rates of CD24 and CD133 in normal gastric tissues and adjacent tissues were significantly lower (P < 0.05). Univariate analysis showed that co-expression of CD24 and CD133 in GC tissues was significantly correlated with tumor size, Lauren classification, T stage, N stage, and vascular invasion (P < 0.05), but not with patient age, gender, tumor site, World Health Organization histological classification, or M stage (P > 0.05). Further multivariate regression analysis suggested that tumor size, T stage, N stage, and vascular invasion were independent risk factors promoting CD24 and CD133 double positivity. Spearman rank correlation analysis indicated a moderate positive correlation between their expression intensities (r = 0.420, P < 0.001). During follow-up, 29 of 304 patients were lost (loss rate 9.54%); 146 deaths occurred. According to expression combination, there were 89 cases of CD24 single positivity (39 deaths), 68 cases of CD133 single positivity (31 deaths), 81 cases of double negativity (25 deaths), and 66 cases of double positivity (51 deaths). Log-rank test showed significant differences in overall survival among the four groups (χ² = 20.89, P < 0.001), with CD24+/CD133+ group showing the worst prognosis.

CONCLUSION

CD24 and CD133 exhibit high positive detection rates in GC tissues, and their co-positivity is closely associated with tumor stage progression and significantly indicates unfavorable survival outcomes. The co-expression of CD24/CD133 may reflect higher aggressiveness and metastatic potential of GC, serving as a potential prognostic marker and a direction for targeted therapeutic strategies. However, as this is a single-center retrospective study with limitations such as patient loss to follow-up and sample size, further prospective, multicenter, and mechanistic studies are required to validate its clinical applicability and biological role.

Key Words: Gastric cancer; CD24; CD133; Cancer stem cell; Co-expression; Prognosis; Immunohistochemistry

Core Tip: CD24 and CD133 exhibit high positive detection rates in gastric cancer tissues. The co-expression of CD24 and CD133 was closely related to the advanced tumor stage. Co-positivity of CD24 and CD133 predicts poor survival. The co-expression of CD24/CD133 may reflect the higher invasive and metastatic potential of gastric cancer. This combined indicator can be used as a potential prognostic marker and direction for targeted therapeutic strategies.

INTRODUCTION

Gastric cancer (GC) is one of the most common malignant tumors of the digestive system worldwide. According to data released by the International Agency for Research on Cancer, GC ranks among the top five in the global spectrum of malignant tumor incidence, with a particularly high prevalence in East Asia, where Chinese patients account for a significant proportion of global cases[1]. Although in recent years the overall survival (OS) rate of some patients has improved due to the widespread use of endoscopic screening, advances in surgical techniques, and the introduction of emerging therapies such as targeted therapy and immunotherapy, the prognosis of patients with advanced GC remains poor. Factors such as tumor heterogeneity, lack of typical clinical manifestations in the early stages, and the high risk of recurrence and metastasis seriously affect treatment outcomes and long-term survival[2-4]. Therefore, exploring novel biomarkers to assist in assessing disease progression and predicting patient prognosis is of great significance for optimizing individualized treatment strategies.

In recent years, research on cancer stem cells has gradually become a hotspot in the field of oncology. Cancer stem cells are considered a small subset of cells with self-renewal, unlimited proliferation, and multidirectional differentiation potential, capable of driving tumor initiation, progression, metastasis, and recurrence[5,6]. Increasing evidence[7-9] indicates that cancer stem cells are not only closely related to chemo- and radio-resistance but may also represent an important biological basis for poor tumor prognosis. Hence, identifying surface molecules that can accurately mark cancer stem cells and further investigating their clinical implications are critical for elucidating the mechanisms of GC development and improving clinical management strategies.

Among the reported cancer stem cell markers, CD24 and CD133 have attracted considerable attention. CD24 is a glycosylphosphatidylinositol-anchored protein widely distributed on the surface of hematopoietic cells and various solid tumor cells, and studies have shown that it plays roles in promoting cell adhesion, migration, signal transduction, and immune evasion[10,11]. CD133, also known as prominin-1, is a pentaspan transmembrane glycoprotein first identified on the surface of hematopoietic stem cells. It has since been widely confirmed to be highly expressed in multiple solid tumors, such as colorectal cancer and hepatocellular carcinoma, and is closely associated with tumor aggressiveness and resistance to therapy[12,13]. Previous studies have suggested that the individual expression of CD24 or CD133 may be related to the malignant behavior of GC, but the combined expression patterns of the two markers in tumor tissues and their predictive value for clinical outcomes remain insufficiently explored[14,15].

Based on this background, the present study retrospectively analyzed pathological tissue specimens from 304 GC patients, examining the expression of CD24 and CD133 in tumor tissues, adjacent tissues, and normal gastric mucosa. Combined with clinicopathological parameters and follow-up data, the study further investigated the relationship between their co-expression and tumor stage, infiltration, metastasis, and prognosis. The aim was to elucidate the role of CD24/CD133 double positivity in the development and progression of GC, assess its potential value in prognostic evaluation and treatment decision-making, and provide reference for future clinical applications.

MATERIALS AND METHODS

Study design and subjects

This study was a single-center retrospective study. The study population consisted of patients who underwent resection in the Department of Gastrointestinal Surgery of our hospital between October 2018 and June 2020 and were pathologically confirmed with primary GC. A total of 304 patients were included and finally analyzed. Inclusion criteria: (1) Patients were newly diagnosed with GC and underwent surgery in our hospital; (2) Pathology confirmed primary GC (based on postoperative paraffin-embedded sections); (3) Adequate paraffin-embedded tissue blocks (tumor tissue and adjacent/normal gastric mucosa samples available) were completely preserved; and (4) Outpatient and inpatient medical records, intraoperative and pathology reports, and follow-up data were relatively complete, allowing for acquisition of main clinicopathological variables and survival outcomes. Exclusion criteria: (1) Patients with a history of or concurrent malignant tumors in other systems; (2) Patients who received systemic therapy before surgery (such as neoadjuvant chemotherapy/radiotherapy/targeted or immunotherapy) that could significantly alter the expression of tissue biomarkers, and with unclear treatment course information; (3) Tissue specimens with poor fixation, degeneration, severe shedding, or unable to complete immunohistochemistry staining; and (4) Cases with severely missing clinicopathological data or follow-up information, making them unsuitable for inclusion. All eligible cases within the cohort were included as the study sample, and no prior sample size estimation was performed.

Ethics statement

This study was approved by the Ethics Committee of Guangxi Medical University Cancer Hospital [Approval No. KJCS2018(13)] and conducted in accordance with the Declaration of Helsinki. All participants signed written informed consent forms.

Collection and processing of tissue specimens

Tissue source: All specimens were obtained from routine postoperative formalin-fixed paraffin-embedded samples. For each case, one representative tumor tissue block was prioritized; when possible, paired adjacent tissue and normal gastric mucosa samples distant from the lesion were collected as controls. To ensure standardization, in this study, “adjacent tissue” was defined as tissue ≥ 2 cm away from the gross tumor margin without visible lesions; “normal gastric tissue” was defined as tissue ≥ 5 cm away from the tumor, pathologically confirmed to be free of tumor infiltration or dysplasia. Tissue sections: 4-μm consecutive slices were cut from paraffin blocks, mounted on slides, and baked for later use.

Immunohistochemistry staining method

To ensure consistency across specimens, all staining procedures were carried out in the same laboratory by the same personnel, strictly following the steps below: Dewaxing and rehydration: 4 μm paraffin-embedded tissue sections were mounted on slides, dewaxed in xylene twice for 5-10 minutes each, then immersed sequentially in 100%, 95%, 80%, and 70% ethanol for 3-5 minutes, followed by rehydration with distilled water until fully hydrated. Antigen retrieval: Sections were placed in citrate buffer (pH = 6.0) and underwent microwave heat-induced epitope retrieval: High heat for 8 minutes, low heat for 10 minutes; then cooled naturally to room temperature. Blocking endogenous peroxidase: Sections were incubated in 3% H₂O₂ at room temperature for 10 minutes to eliminate endogenous peroxidase activity, followed by phosphate buffered saline washes (3 times, 3-5 minutes each). Blocking nonspecific binding: Sections were incubated with 5% normal goat serum at room temperature for 20-30 minutes to reduce nonspecific background staining. Primary antibody incubation: CD24 was detected using an antibody from Proteintech (IL, United States); CD133 was detected using a rabbit recombinant monoclonal antibody from Abcam (United Kingdom). Both antibodies were diluted at 1:100 and incubated overnight at 4 °C. Secondary antibody incubation and chromogenic detection: HRP-labeled secondary antibody was incubated at room temperature for 30-60 minutes; DAB chromogenic reaction was applied for 3-10 minutes (microscopically monitored and terminated promptly to avoid excessive background staining). Controls: Each batch included positive control sections (tumor tissues known to express CD24 or CD133) and negative controls (omitting the primary antibody step) to monitor staining specificity and inter-batch quality. Counterstaining, dehydration, clearing, and mounting: After chromogenic detection, sections were counterstained with hematoxylin (about 1-2 minutes), dehydrated through graded alcohols, treated with clearing reagent, and finally mounted with neutral resin for preservation.

Immunohistochemical semi-quantitative scoring and grouping

To achieve quantitative comparison, this study adopted the semi-quantitative immunoreactive score (IRS) method[16] for each section: Staining intensity score (I): 0 (no staining), 1 (weak staining), 2 (moderate staining), 3 (strong staining). Percentage of positive cells score (P): 0 (< 5% positive cells), 1 (5%-25%), 2 (26%-50%), 3 (51%-75%), 4 (> 75%). IRS was calculated as IRS = I × P, ranging from 0-12. Expression grading: IRS = 0 defined as negative (-); IRS 1-8 defined as low expression (+); IRS 9-12 defined as high expression (++). According to the staining results of the two markers, patients were divided into four groups: CD24+/CD133+ (double-positive group); CD24+/CD133- (CD24 single-positive group); CD24-/CD133+ (CD133 single-positive group); CD24-/CD133- (double-negative group).

Pathological data and clinical variable collection

Clinicopathological variables were extracted item by item from inpatient records, operative notes, and pathology reports, including but not limited to: Patient age, sex, maximum tumor diameter (cm), tumor location (upper, middle, lower, ≥ 2 regions, or remnant stomach), World Health Organization histological classification, Lauren classification, vascular invasion (including blood vessel/Lymphatic invasion), primary tumor infiltration depth (T stage), lymph node metastasis (N stage), distant metastasis (M stage), and vascular invasion (including blood vessel/Lymphatic invasion). Tumor staging was performed and stratified according to the 8^th edition of American Joint Committee on Cancer/tumor-node-metastasis[17].

Follow-up method and endpoint definition

Follow-up method: Based on the hospital follow-up system, survival status, recurrence/metastasis, and death information were obtained through outpatient follow-up, telephone follow-up, and electronic medical record review. For patients who could not be contacted, the last contact date was recorded as the date of loss to follow-up, and such cases were censored in survival analysis (right-censored). In this study, “OS” was defined as the time from the date of surgery to death from any cause (death date as endpoint). Patients without death events were censored at the last follow-up date. Loss to follow-up handling: Patients lost to follow-up were right-censored at the last known follow-up time.

Quality control and interpretation consistency

All immunohistochemical sections were independently scored in a blinded manner by two experienced pathologists (each with ≥ 10 years of experience in gastrointestinal tumor pathology) without knowledge of clinical outcomes or other marker results. In case of disagreement, the two reviewers jointly re-examined the slides to reach consensus. Positive and negative controls were included in each daily/batch experiment, and inter-batch variations were recorded; if staining deviation occurred in a batch, restaining was performed or the batch was excluded and retested.

Statistical analysis

All data were entered by dedicated researchers and double-checked before statistical analysis using SPSS (version 25.0); graphs were prepared using GraphPad Prism and other software. Categorical data were expressed as n (%); intergroup comparisons were performed using the χ² test or Fisher’s exact test (when expected counts < 5). Spearman rank correlation analysis was used to assess the correlation between CD24 and CD133 expression intensity. Multivariate logistic regression analysis was applied to identify independent risk factors influencing CD24/CD133 co-expression, with results presented as odds ratios and 95% confidence intervals. Survival curves were estimated using the Kaplan-Meier method, and survival differences between groups were compared using the Log-rank test. A two-sided P value < 0.05 was considered statistically significant.

RESULTS

Expression of CD24 and CD133 in GC tissues

CD24 was commonly expressed in the cell membrane with mild cytoplasmic staining, while CD133 was mainly expressed in the cell membrane (also observed as punctate cytoplasmic staining). In interpretation, specific staining of tumor cells was regarded as a positive signal, whereas nonspecific staining of inflammatory cells and stroma was excluded. Among 304 patients, 155 cases (50.99%) were positive for CD24, including 91 cases with low expression and 64 cases with high expression; 133 cases (43.75%) were positive for CD133, including 81 cases with low expression and 52 cases with high expression. Double-positive cases were 74 (24.34%), and double-negative cases were 81 (26.64%), as shown in Figure 1.

Figure 1 Expression of CD24 and CD133 in gastric cancer tissues. A: Negative expression of CD24 in gastric cancer (GC) tissues; B: Low expression of CD24 in GC tissues; C: High expression of CD24 in GC tissues; D: Negative expression of CD133 in GC tissues; E: Low expression of CD133 in GC tissues; F: High expression of CD133 in GC tissues.

Comparison of CD24 and CD133 positivity rates in normal gastric tissue, adjacent tissue, and cancer tissue

Compared with cancer tissues, the positivity rates of CD24 and CD133 were significantly lower in normal gastric tissue and adjacent tissue (P < 0.05). However, no significant difference in positivity rates of CD24 and CD133 was found between adjacent tissue and normal gastric tissue (P > 0.05), as shown in Table 1.

Table 1 Comparison of CD24, CD133 positivity rates in normal gastric tissue, adjacent tissue, and cancer tissue, n (%).

	CD24 (n = 304)		CD133 (n = 304)
	Positive	Negative	Positive	Negative
Cancer tissue	155 (50.99)a,c	149 (49.01)	133 (43.75)d,f	171 (56.25)
Adjacent tissue	19 (6.25)b	285 (93.75)	15 (4.93)e	289 (95.07)
Normal tissue	10 (3.29)	294 (96.71)	8 (2.63)	296 (97.37)

^aP < 0.05 for comparison between cancer tissue and adjacent tissue.

^bP > 0.05 for comparison between adjacent tissue and normal gastric tissue.

^cP < 0.05 for comparison between cancer tissue and normal gastric tissue.

^dP < 0.05 for comparison between cancer tissue and adjacent tissue.

^eP > 0.05 for comparison between adjacent tissue and normal gastric tissue.

^fP < 0.05 for comparison between cancer tissue and normal gastric tissue.

Univariate analysis of the relationship between CD24/CD133 double-positive expression and clinicopathological parameters in GC tissues

Univariate analysis showed that co-expression of CD24 and CD133 in GC tissues was significantly associated with tumor diameter, Lauren classification, T stage, N stage, and vascular invasion (P < 0.05), but not with age, sex, tumor site, World Health Organization pathological type, or M stage (P > 0.05), as shown in Table 2.

Table 2 Univariate analysis of the relationship between CD24/CD133 double-positive expression and clinicopathological parameters in gastric cancer tissues, n (%).

Clinicopathological parameter	CD24/CD133 double-positive		χ2	P value
	Yes (n = 74)	No (n = 230)
Sex			0.493	0.482
Male	40 (54.05)	135 (58.70)
Female	34 (45.95)	95 (41.30)
Age (years)			0.047	0.827
< 60	13 (17.57)	43 (18.70)
≥ 60	61 (82.43)	187 (81.30)
Tumor site			0.448	0.503
Upper	4 (5.41)	11 (4.78)
Middle	8 (10.81)	36 (15.65)
Lower	60 (81.08)	178 (77.39)
≥ 2 regions or remnant stomach	2 (2.70)	5 (2.17)
Tumor size (cm)			35.494	< 0.001
< 5	23 (31.08)	161 (70.00)
≥ 5	51 (68.92)	69 (30.00)
WHO classification			3.022	0.082
Papillary adenocarcinoma	4 (5.41)	8 (3.48)
Tubular adenocarcinoma	59 (79.73)	202 (87.83)
Mucinous adenocarcinoma	2 (2.70)	6 (2.61)
Signet-ring cell carcinoma	9 (12.16)	14 (6.09)
Lauren classification			5.469	0.019
Intestinal type	46 (62.16)	175 (76.09)
Diffuse type	28 (37.84)	55 (23.91)
T stage			41.651	< 0.001
T1	0 (0.00)	24 (10.43)
T2	4 (5.41)	55 (23.91)
T3	15 (20.27)	78 (33.91)
T4	55 (74.32)	73 (31.74)
N stage			12.854	< 0.001
N0	4 (5.41)	27 (11.74)
N1	10 (13.51)	64 (27.83)
N2	25 (33.78)	81 (35.22)
N3	35 (47.30)	58 (25.22)
M stage			2.012	0.156
M0	8 (10.81)	12 (5.22)
M1	66 (89.19)	218 (94.78)
Vascular invasion			16.925	< 0.001
Present	13 (17.57)	7 (3.04)
Absent	61 (82.43)	223 (96.96)

WHO: World Health Organization.

Multivariate analysis of the relationship between CD24/CD133 double-positive expression and clinicopathological parameters in GC tissues

Using CD24/CD133 double expression in GC tissue as the dependent variable (double positive = 1, double negative = 0), variables with significant differences in Table 3 were entered as independent variables and coded. Logistic regression analysis showed that tumor diameter, T stage, N stage, and vascular invasion were independent risk factors promoting CD24/CD133 double-positive expression, as shown in Table 4.

Table 3 Variable assignment table.

Independent variable	Assignment method
Tumor size	< 5 cm = 0; ≥ 5 cm = 1
Lauren classification	Intestinal type = 0; diffuse type = 1
T stage	T1-T2 = 0; T3-T4 = 1
N stage	N0-N1 = 0; N2-N3 = 1
Vascular invasion	Absent = 0; present = 1

Table 4 Multivariate logistic regression analysis of CD24/CD133 double-positive expression in gastric cancer tissues.

Variable	β	SE	Wald χ2	P value	OR	95%CI
Tumor size	1.215	0.382	10.104	0.001	3.37	1.60-7.08
Lauren classification	0.467	0.345	1.835	0.176	1.61	0.81-3.16
T stage	1.524	0.431	12.509	< 0.001	4.59	1.98-10.64
N stage	1.002	0.397	6.363	0.012	2.72	1.25-5.90
Vascular invasion	1.876	0.562	11.143	< 0.001	6.53	2.17-19.67
Constant	-2.037	0.529	14.829	< 0.001

OR: Odds ratio; CI: Confidence interval.

Correlation analysis of CD24 and CD133 expression intensity

Cross-distribution analysis of CD24 and CD133 expression grading (negative/Low/high) in 304 GC tissues was performed. Based on marginal totals, CD24 was negative in 149, low in 91, high in 64; CD133 was negative in 171, low in 81, high in 52. Both markers were simultaneously positive (double positive) in 74 cases. Spearman rank correlation analysis indicated a moderate positive correlation between their expression intensities (r = 0.420, P < 0.001), as shown in Table 5.

Table 5 Spearman rank correlation analysis of CD24 and CD133 expression intensity.

CD133 expression	CD24 expression
	-	+	++
-	90	81	0
+	59	0	22
++	0	10	42

-: Negative; +: Low positive expression; ++: High positive expression.

Effect of CD24 and CD133 expression on patient prognosis

During follow-up, 29 of 304 patients were lost to follow-up (loss rate 9.54%); a total of 146 deaths occurred during the follow-up period. Grouped by expression patterns: CD24 single-positive group, 89 cases (39 deaths); CD133 single-positive group, 68 cases (31 deaths); double-negative group, 81 cases (25 deaths); double-positive group, 66 cases (51 deaths). Log-rank test showed significant differences in OS among the four groups (χ² = 20.89, P < 0.001), as shown in Figure 2.

Figure 2  Survival curves of patients with different CD24/CD133 expression patterns.

DISCUSSION

The significance of GC stem cell research lies in offering a novel perspective to re-understand the origin and essence of GC. Stem cells possess self-renewal and multidirectional differentiation potential. Although they exist in very small numbers within cancer tissues, they play crucial roles in the initiation, progression, recurrence, and metastasis of malignant tumors[18,19]. At present, research on GC stem cells remains exploratory. Identifying specific GC stem cell markers, isolating and purifying these cells, and understanding their impact on the biological behavior of GC form the foundation for all cell therapy-related studies.

In this retrospective analysis of 304 surgically resected GC patients, we found that the cancer stem cell markers CD24 and CD133 were highly detected in GC tissues (50.99% and 43.75%, respectively), with a certain degree of co-expression (double-positive, 24.34%). Statistical analysis showed that CD24/CD133 co-expression was closely associated with increased tumor diameter, Lauren diffuse type, advanced T stage, severe lymph node metastasis (N stage), and vascular invasion. In multivariate logistic regression, tumor diameter, T stage, N stage, and vascular invasion were identified as independent risk factors for double positivity of the two markers. Survival analysis revealed that patients in the CD24+/CD133+ group had significantly worse OS, suggesting that dual positivity may represent a more aggressive biological subtype of GC with poorer prognosis. These findings hold important clinical and biological significance.

Individually, the relationships between CD24 or CD133 expression and clinicopathological parameters or prognosis in GC have been reported in multiple studies, and our results are largely consistent with these. Previous systematic reviews and meta-analyses indicated that CD24 expression is associated with adverse pathological features of GC (such as vascular/Lymphatic invasion, lymph node metastasis) and worse OS, supporting CD24 as an unfavorable prognostic biomarker[20]. Similarly, multiple systematic reviews/meta-analyses have demonstrated that CD133 positivity is correlated with poor prognosis, advanced stage, and increased risk of metastasis, reinforcing its role as a cancer stem cell-related prognostic marker[21,22]. Compared with studies focusing on single markers, our study further emphasized the association of combined CD24/CD133 expression with tumor invasiveness, metastatic potential, and poor prognosis: In this cohort, the double-positive group exhibited larger tumor size, deeper invasion, more extensive lymph node metastasis, and significantly worse survival. These findings align with the growing view that combined detection of multiple markers may better capture tumor stem cell heterogeneity and predict invasive/metastatic potential[23].

From a molecular and cellular biology perspective, CD24 and CD133 participate in multiple tumor progression-related pathways, which may explain their correlation with poorer clinical outcomes. Previous studies have shown that CD24 promotes tumor cell adhesion, migration, and metastasis, and is involved in immune regulation and chemoresistance[24]. Under tumor microenvironmental stressors such as hypoxia, CD24 expression can be upregulated, thereby enhancing invasive phenotypes[25]. CD133 is commonly regarded as a marker of subpopulations with self-renewal and tumorigenic potential, linked to stemness maintenance, signaling pathways (e.g., Wnt, Notch, phosphatidylinositol 3-kinase/protein kinase B), and drug resistance, with high expression often indicating higher recurrence rates and poor survival[26-28]. When both CD24 and CD133 are co-expressed, they may synergistically identify or sustain a more “stemness-enriched/invasive” subpopulation: CD133 confers self-renewal capacity that underlies recurrence and resistance[29], while CD24-mediated adhesion, migration, and immune evasion mechanisms may enhance distant colonization and growth[30,31]. This combined effect could account for larger tumor volume, deeper invasion, and higher rates of lymph node metastasis, ultimately resulting in worse survival outcomes. In our study, Spearman’s rank correlation analysis (r = 0.420, P < 0.001) revealed a moderate positive correlation between the expression intensities of CD24 and CD133, supporting the notion that they may be co-regulated or co-enriched in specific GC subtypes.

From a clinical perspective, the combined detection of CD24 and CD133 may serve as a valuable reference for postoperative high-risk stratification, guiding follow-up frequency and adjuvant therapy decisions. For double-positive patients, clinicians may consider adopting more aggressive management strategies, such as intensified adjuvant treatment or participation in clinical trials. The significant association between CD24+/CD133+ co-expression and aggressive clinicopathological features - such as deeper tumor invasion, lymph node metastasis, and vascular invasion - suggests that this cellular subset may represent a biologically distinct population with enhanced invasive and stem-like properties. Beyond the clinical correlations, the observed moderate positive relationship between CD24 and CD133 expression intensities implies potential biological synergy rather than independent expression, supporting the hypothesis that their co-expression delineates a cancer stem cell-enriched niche within gastric tumors. However, while immunohistochemical data strongly link the co-expression status to adverse clinical outcomes, functional validation of stemness characteristics - such as self-renewal, differentiation potential, and tumorigenicity - was not performed in the present study. Future mechanistic research employing in vitro sphere formation and in vivo xenotransplantation assays using sorted CD24+/CD133+ cells would be essential to confirm their cancer stem cell-like functional capacity.

The strengths of this study include a relatively large sample size, simultaneous analysis of tumor, adjacent, and normal gastric tissues, and integration of detailed clinicopathological parameters with long-term survival data, offering a comprehensive perspective on the clinical significance of CD24 and CD133. Nonetheless, several limitations should be acknowledged. First, this was a single-center retrospective study, which may introduce selection bias and limit the generalizability of the findings. Second, immunohistochemistry is subject to variability in antibody specificity and scoring methods, which could affect reproducibility across studies. Third, the absence of stratified analysis for postoperative treatment regimens and lack of experimental validation constrain interpretation of causality. Future research should include multicenter, prospective cohorts to confirm these findings and promote standardization of immunohistochemistry protocols. Furthermore, mechanistic investigations into CD24/CD133-related signaling pathways may not only elucidate their roles in GC progression but also reveal novel therapeutic targets for this high-risk patient population.

CONCLUSION

This study demonstrated that CD24 and CD133 are significantly expressed in GC tissues and frequently co-expressed. Their combined positivity was associated with larger tumor size, advanced stage, vascular invasion, and poorer survival outcomes. These findings support the use of combined assessment of cancer stem cell markers as an important complement to molecular pathological evaluation of GC, suggesting that the CD24/CD133-positive subpopulation may represent a key biological basis for tumor aggressiveness and unfavorable prognosis. Given the retrospective and single-center limitations of this study, larger multicenter prospective studies and mechanistic investigations are required to confirm their clinical value in risk stratification and individualized treatment of GC.