Prognostic impact and reasons for variability by tumor location in gastric cancer

Abstract

BACKGROUND

Gastric cancer (GC) is a highly prevalent gastrointestinal tract tumor. Several trials have demonstrated that the location of GC can affect patient prognosis. However, the factors determining tumor location remain unclear.

AIM

To investigate the tumor location of patients, we went on to study the influencing factors that lead to changes in the location of GC.

METHODS

A retrospective evaluation was carried out on 3287 patients who underwent gastrectomy for GC in Zhejiang Cancer Hospital. The patients were followed up post-diagnosis and post-gastrectomy. The clinicopathological variables and overall survival of the patients were recorded. By analyzing the location of GC, the tumor location was divided into four categories: “Upper”, “middle”, “lower”, and “total”. Statistical software was utilized to analyze the relationship of each variable with the location of GC.

RESULTS

A total of 3287 patients were included in this study. The clinicopathological indices of gender, age, serum levels of carcinoembryonic antigen (CEA), carbohydrate antigen (CA19-9) and CA72-4 levels, were significantly associated with tumor location in patients with GC. In addition, there was a strong correlation between GC location and the prognosis of postoperative patients. Specifically, patients with “lower” and “middle” GC demonstrated a better prognosis than those with tumors in other categories.

CONCLUSION

The five clinicopathological indices of gender, age, CEA, CA19-9 and CA72-4 levels exhibit varying degrees of influence on the tumor location. The tumor location correlates with patient prognosis following surgery.

Key Words: Gastric cancer; Clinicopathologic characteristics; Tumor marker; Tumor location; Overall survival

Core Tip: This study investigates how the location of gastric cancer (GC) impacts prognosis by analyzing data from 3287 patients who underwent gastrectomy. It reveals that clinicopathological factors, including gender, age, and serum levels of carcinoembryonic antigen, carbohydrate antigen (CA) 19-9, and CA72-4, significantly influence tumor location. Notably, GC located in the “lower” and “middle” regions correlates with a better overall survival compared to other locations. These findings underscore the importance of tumor location in predicting patient outcomes and guiding postoperative management strategies.

INTRODUCTION

Gastric cancer (GC) refers to a type of malignant tumor originating from the epithelial cells of the gastric mucosa. It is the fifth most common cancer worldwide and has the third highest mortality rate among all malignant tumors[1]. Adenocarcinoma is the most prevalent type of GC, accounting for more than 95% of cases, and is one of the most common malignant tumors in China. Symptoms of early GC are not obvious, leading to late-stage diagnosis and poor prognosis for many patients. The incidence and mortality rates of GC are high in East Asian countries, with approximately half of the world’s cases occurring in these countries[2]. Although the mortality rate of GC in China has decreased significantly with improvements in economic level and technology, especially in rural areas[3], GC is still an important cancer burden at present and one of the key issues in China’s cancer prevention and control strategies. According to various clinicopathological indices, the location of GC has been shown to impact patient prognosis, with different tumor locations correlated with prognostic survival. However, the factors influencing the different locations of GC remain controversial.

In this study, we retrospectively reported the clinicopathological characteristics and survival of 3287 GC patients over a 12-year period. We analyzed the effect of multiple variables on the tumor location and identified gender, age, carcinoembryonic antigen (CEA), CA19-9, and CA72-4 as independent risk factors for the location of GC. Moreover, our findings revealed that the main factors affecting the location of GC differed between male and female patients. Notably, further research is required to understand why survival rates vary according to the location of GC. These findings can aid in studying the factors influencing different GC locations, predicting the incidence, and assessing the survival prognosis of GC patients. This not only improves the detection rate of early GC during screening but also has far-reaching implications for the clinical treatment of the disease.

MATERIALS AND METHODS

Patients

A total of 3287 patients diagnosed with GC who underwent gastrectomy at Zhejiang Cancer Hospital from 2008 to 2019 were included in this study. We collected and analyzed the clinicopathological characteristics, tumor location, tumor marker levels, postoperative survival, surgical methods and extent of resection of GC patients for retrospective evaluation. The patients ranged in age from 25 to 95 years, and all underwent radical surgical treatment. Patients with postoperative residual GC, unsuccessful surgery or malignant metastases from other organs were excluded from this study. We followed these postoperative GC patients at regular intervals of 3-6 months, and data were collected accordingly. Prior to analyzing these patient data, personal information records, such as patient names and medical record numbers, were deidentified. All patients signed informed consent forms. The study was approved by the Ethics Committee of Zhejiang Cancer Hospital.

Clinicopathological characteristics

We collected data on the following variables: Age, gender, body mass index (BMI), pathological type, tumor location, surgical method, type of resection, pathological tumor-node-metastasis (pTNM, AJCC-TNM 8^th), tumor differentiation grade, vascular tumor thrombus, nerve invasion, lymph node ratio, and preoperative tumor markers. Patients with abnormally elevated tumor markers were excluded from the analysis. Survival data were obtained by telephone follow-ups and medical records.

Division of gastric tumor location

In this study, the subsites of GC were classified as the cardia, gastric fundus, gastroesophageal junction, gastric body, greater curvature, lesser curvature, gastric antrum, pylorus, gastric horn, and distal stomach[4]. These sites were further categorized as being in the upper, middle, or lower third of the stomach. Tumors located primarily at the gastroesophageal junction, cardia or fundus were classified as being in the upper third of the stomach. Those located in the pylorus, sinus, horn of the stomach or distal stomach were classified as being in the lower third of the stomach. Tumors located in the middle of the stomach body, greater curvature, or lesser curvature were identified as being in the middle third of the stomach. Tumors involving two adjacent regions were classified according to their proximity: Those near the cardia were assigned to the upper third, while those near the pyloric antrum were assigned to the lower third. Tumors spanning non-adjacent regions or all three regions were categorized as “Total”. For statistical analysis and clarity, the classifications “Upper 1/3”, “Middle 1/3”, and “Lower 1/3” were simplified to “Upper”, “Middle”, and “Lower”, respectively, while “Total” remained unchanged.

Statistical analysis

The χ² test was used to evaluate between-group differences in categorical variables. Bivariate correlations between study variables were calculated using Spearman’s rank correlation coefficient. The interaction between the dependent and independent variables was analyzed using logistic regression. P values < 0.05 were considered to indicate statistical significance. The study flowchart is shown in Figure 1. Statistical analyses were performed using IBM SPSS Statistics, version 25.0 (IBM Corp, Armonk, NY, United States) and GraphPad Prism 8, version 8.3.0 (GraphPad Software, San Diego, CA, United States). A random forest model was constructed using R version 4.3.1 (randomForestSRC package 3.2.3). The minimum depth method and variable importance (VIMP) method were used to filter important features, and the ggplot2 package 3.5.0 was used to visualize the images. Survival analysis was performed using the Kaplan-Meier survival curve method.

Figure 1 Flow chart of the data selection and processing steps. AFP: Alpha-fetoprotein; CEA: Carcinoembryonic antigen; CA: Carbohydrate antigen; BMI: Body mass index; VIMP: Variable importance.

RESULTS

Location of GC affects postoperative survival

To assess the factors influencing the prognostic survival of GC patients, we constructed a random forest model to analyze the prognosis-related influencing factors. Patients surviving less than 12 months were excluded to refine the analysis. In this study, only patients who underwent laparoscopic gastrectomy or open gastrectomy were included, allowing us to exclude the effect of surgical modality on prognosis. Thus, we discovered a significant correlation between tumor location and patient prognosis (Figure 2).

Figure 2 Forest plot of factors influencing survival in patients with gastric cancer. The forest map displays hazard ratios (HRs) and 95% confidence intervals (95%CIs) for various factors that influence tumor prognosis according to multiple Cox regression analyses. Each dot represents the HR for a specific factor, the horizontal line represents the 95%CI, and the vertical blank line represents the reference line (HR = 1.0), indicating no impact. Factors with good HR and CI on the right side of the line were significant predictors of poor prognosis. Values in parentheses are the highest normal levels for that tumor marker. CEA: Carcinoembryonic antigen; AFP: Alpha-fetoprotein; CA: Carbohydrate antigen; BMI: Body mass index; 95%CI: 95% confidence interval; HR: Hazard ratio; pTNM: Pathological tumor-node-metastasis.

To further examine how the location of GC impacts postoperative survival, 3287 patients were divided into four groups according to their stomach cancer location: “upper”, “middle”, “lower”, and “total”. Our clinical data showed the distribution of these cases as follows: Upper (52.4%), middle (30.6%), lower (15.4%), and total (1.6%) (Table 1). Survival prognosis was relatively better for patients with middle and lower GC compared to those with upper and total GC (Figure 3). Regarding the 5-year survival rate, no significant difference was noted between middle and lower GC patients. However, the overall survival (OS) rate disparity between patients with upper GC and those with total GC was more evident (Table 2). Our findings confirm that tumors situated nearer to the stomach’s proximal end correlate with shorter survival periods and poorer outcomes, which is consistent with previous studies[5-7]. Overall, the evidence suggests that the location of GC is a significant factor in determining patient prognosis.

Figure 3 Kaplan-Meier survival curves for different locations of gastric cancer, where different colors represent different locations. Red: Lower 1/3; green: Middle 1/3; blue: Total; purple: Upper 1/3). The solid line represents the Kaplan-Meier estimates for each group. The shaded area around the survival curve represents the 95% confidence interval, representing the range of fluctuations in the true survival estimate. A P value less than 0.05 was considered to indicate statistical significance.

Table 1 Clinicopathological characteristics of gastric cancer patient samples, n = 3287, n (%).

Characteristics	Number of cases
Gender
Male	2541 (77.3)
Female	746 (22.7)
Age
< 60	1191 (36.2)
≥ 60	2096 (63.8)
BMI
≤ 18	296 (9.0)
18-24	2090 (63.6)
≥ 24	901 (27.4)
Family history
No	2102 (63.9)
Yes	1185 (36.1)
Smoking history
No	1767 (53.8)
Yes	1520 (46.2)
Drinking history
No	2261 (68.8)
Yes	1026 (31.2)
Surgery methods
Open	2274 (84.4)
Laparoscopy	513 (15.6)
Range of resection
Proximal	187 (5.7)
Distal	1719 (52.3)
Total	1381 (42.0)
Tumor location
Middle	1007 (30.6)
Lower	507 (15.4)
Upper	1721 (52.4)
Total	52 (1.6)
Differentiation
Well	22 (0.7)
Well-moderate	61 (1.9)
Moderate	771 (23.5)
Moderate-poor	1300 (39.5)
Poor	1133 (34.5)
Pathological type
Adenocarcinoma	2283 (87.7)
PMGC	206 (6.3)
MGC	198 (6.0)
Vascular tumor thrombus
No	1720 (52.3)
Yes	1567 (47.7)
Nerve invasion
No	1718 (52.3)
Yes	1569 (47.7)
Maximum tumor diameter (cm)
< 5	1982 (60.3)
≥ 5	1305 (39.7)
Tumor depth
T1	622 (18.9)
T2	447 (13.6)
T3	203 (6.2)
T4	2015 (61.3)
Lymph node metastasis
N0	1142 (34.7)
N1	622 (18.9)
N2	703 (21.4)
N3	820 (24.9)
pTNM
I	788 (24.0)
II	733 (22.3)
III	1766 (53.7)
CEA
< 5	2578 (78.4)
≥ 5	709 (21.6)
CA19-9
< 37	2573 (78.3)
≥ 37	714 (21.7)
CA125
< 35	3101 (94.3)
≥ 35	186 (5.7)
AFP
< 10	3110 (94.6)
≥ 10	177 (5.4)
CA242
< 15	2735 (83.2)
≥ 15	552 (16.8)
CA72-4
< 6.9	2661 (81.0)
≥ 6.9	626 (19.0)

AFP: Alpha-fetoprotein; CEA: Carcinoembryonic antigen; CA: Carbohydrate antigen; BMI: Body mass index; pTNM: Pathological tumor-node-metastasis; PMGC: Peritoneal metastases of gastric cancer; MGC: Mucinous gastric carcinoma.

Table 2 Five-year overall survival of patients with different tumor locations, %.

Tumor location	OS
	1-year	2-year	3-year	4-year	5-year
Upper	89.7	73.3	52.3	40.6	32.2
Middle	93.5	78.1	53.6	41.8	30.4
Lower	91.2	77.9	57.8	44.4	33.5
Total	78.8	53.8	36.5	26.9	25.0

OS: Overall survival.

Location of GC is associated with multiple factors

Subsequently, we examined the associations between various clinical variables and the anatomical location of GC. We identified the relevant factors contributing to the location of GC through the random forest model, the minimum depth method and the VIMP method (Figure 4 and Figure 5). This preliminary analysis paved the way for deeper exploration.

Figure 4 Variable importance of the predictors for the predictive models for gastric cancer location. A: The minimum depth method was used to calculate the influencing factors related to the location of gastric cancer (GC). The vertical dashed line represents the minimum depth threshold; B: The variable importance method was used to calculate the influencing factors related to the location of GC. The green dots indicate strong correlations among impact factors, and the orange dots indicate weak correlations among impact factors. CEA: Carcinoembryonic antigen; AFP: Alpha-fetoprotein; CA: Carbohydrate antigen; BMI: Body mass index; pTNM: Pathological tumor-node-metastasis.

Figure 5 Random forest plot of influencing factors related to gastric cancer location. According to the random forest model, the importance of each parameter to the degree of influence on the location of gastric cancer was calculated, and the histogram was drawn in order of importance. CEA: Carcinoembryonic antigen; CA: Carbohydrate antigen; BMI: Body mass index; pTNM: Pathological tumor-node-metastasis.

A series of one-way χ² tests examining the association between patient variables and the location of GC indicated that gender, age, differentiation, extent of resection, vascular tumor thrombus, nerve invasion, pTNM stage, and serum levels of CEA, CA19-9, CA242, and CA72-4 were significant factors (P < 0.05) (Table 3). Furthermore, Spearman’s correlation analysis of GC location with relevant clinicopathological features revealed that gender (P = 0.001), age (P < 0.001), CEA (P < 0.001), CA19-9 (P < 0.001), CA242 (P < 0.001), and CA72-4 (P = 0.006) were significantly correlated with high confidence levels (Table 4). However, there was no significant correlation between GC location and BMI (P = 0.354), family history (P = 0.177), smoking history (P = 0.140), drinking history (P = 0.409), pathological type (P = 0.131), or alpha-fetoprotein (AFP) level (P = 0.197).

Table 3 Correlations between tumor location and clinicopathologic characteristics of gastric cancer patients.

Characteristics	Tumor location
	Middle	Lower	Upper	Total	P value
Gender
Male	390	1285	826	40	< 0.001
Female	117	436	181	12
Age
< 60	228	692	248	23	< 0.001
≥ 60	279	1029	759	29
BMI
≤ 18	40	161	89	6	0.354
18-24	306	1107	645	32
≥ 24	161	453	273	14
Family history
No	320	1125	629	28	0.177
Yes	187	596	378	24
Smoking history
No	290	935	515	27	0.140
Yes	217	786	492	25
Drinking history
No	364	1178	682	37	0.409
Yes	143	543	325	15
Range of resection
Proximal	9	1	177	0	< 0.001
Distal	197	1515	5	2
Total	301	205	825	50
Differentiation
Well	2	15	5	0	< 0.001
Well-moderate	4	37	19	1
Moderate	101	363	300	7
Moderate-poor	193	693	397	17
Poor	207	613	286	27
Pathological type
Adenocarcinoma	460	1509	871	43	0.131
PMGC	20	111	72	3
MGC	27	101	64	6
Vascular tumor thrombus
No	281	923	501	15	0.001
Yes	226	798	506	37
Nerve invasion
No	266	1023	421	8	< 0.001
Yes	241	698	586	44
Maximum tumor diameter
< 5 cm	344	1129	505	4	< 0.001
≥ 5 cm	163	592	502	48
Tumor depth
T1	103	433	86	0	< 0.001
T2	80	265	100	2
T3	40	96	64	3
T4	284	927	757	47
Lymph node metastasis
N0	196	624	318	4	< 0.001
N1	101	331	186	4
N2	95	350	244	14
N3	115	416	259	30
pTNM
I	135	501	150	2	< 0.001
II	136	377	218	2
III	236	843	639	48
CEA
< 5	424	1369	753	32	< 0.001
≥ 5	83	352	254	20
CA19-9
< 37	423	1365	752	33	< 0.001
≥ 37	84	356	255	19
CA125
< 35	474	1635	950	42	< 0.001
≥ 35	33	86	57	10
AFP
< 10	484	1626	954	46	0.197
≥ 10	23	95	53	6
CA242
< 15	439	1448	812	36	0.001
≥ 15	68	273	195	16
CA72-4
< 6.9	432	1391	802	36	0.008
≥ 6.9	75	330	205	16

AFP: Alpha-fetoprotein; CEA: Carcinoembryonic antigen; CA: Carbohydrate antigen; BMI: Body mass index; pTNM: Pathological tumor-node-metastasis; PMGC: Peritoneal metastases of gastric cancer; MGC: Mucinous gastric carcinoma.

Table 4 Spearman analysis of the correlation between tumor location and clinicopathological data.

Characteristics	Tumor location
	Spearman correlation	P value
Gender	-0.056	0.001
Age	0.148	< 0.001
BMI	-0.022	0.197
Family history	0.019	0.279
Smoking history	0.040	0.020
Drinking history	0.022	0.202
Range of resection	0.256	< 0.001
Differentiation	-0.092	< 0.001
Pathological type	0.041	0.019
Vascular tumor thrombus	0.054	0.002
Nerve invasion	0.128	< 0.001
Maximum tumor diameter	0.169	< 0.001
Tumor depth	0.190	< 0.001
Lymph node metastasis	0.077	< 0.001
pTNM	0.159	< 0.001
CEA	0.083	< 0.001
CA19-9	0.080	< 0.001
CA125	0.013	0.452
AFP	0.014	0.410
CA242	0.064	< 0.001
CA72-4	0.048	0.006

CEA: Carcinoembryonic antigen; CA: Carbohydrate antigen; BMI: Body mass index; pTNM: Pathological tumor-node-metastasis.

This comprehensive analysis underscores the multifactorial nature of GC localization and highlights key clinicopathological variables that are significantly associated with the disease.

The main factors affecting the location of GC

To explore the main factors affecting the location of GC, we conducted multiple logistic regression analysis. The analysis demonstrated a high degree of statistical significance (P < 0.001), indicating a strong agreement between the model’s predictions and the observed data. This confirms the robustness of our model. Importantly, the CA242 index did not achieve statistical significance (P > 0.05), indicating it does not independently affect the location of GC but may interact with other variables. Thus, the CA242 index was excluded from the final analysis of major influencing factors.

Several types of cancer, including those affecting the stomach, liver, and colon, are significantly more prevalent in men than in women[8-10]. Therefore, to more accurately understand the relationship between the location of GC and its influencing factors, we divided 3287 patients into two groups according to gender: Males (2541) and females (746) (Table 5). Multiple logistic regression analysis was then performed for both groups. In female patients, the main influencing factors for gastric body cancer were CA19-9 [P = 0.041, odds ratio (OR) = 2.399] and CA72-4 (P < 0.001, OR = 7.052). The probability of developing carcinoma of the gastric body was 2.399-fold and 7.052-fold higher in patients with high CA19-9 and CA72-4 expression, respectively. The main influencing factors for distal GC were age (P = 0.003, OR = 0.387), CEA level (P = 0.034, OR = 0.312), and CA72-4 level (P < 0.001, OR = 7.104). High CA72-4 expression increased the probability of developing distal gastric carcinoma by 7.104-fold compared to that of total GC. In addition, high CEA expression and age older than 60 years were associated with a higher probability of total stomach cancer than distal stomach cancer. For proximal GC, the main influencing factors were age (P < 0.001, OR = 0.150), CEA level (P = 0.010, OR = 0.241), and CA72-4 level (P < 0.001, OR = 8.152). High CA72-4 expression was associated with an 8.152-fold higher probability of proximal gastric carcinoma than total GC. Conversely, a high CEA level and an age over 60 years were linked to a higher probability of total GC than proximal GC (Table 6).

Table 5 Tumor markers characteristics of gastric cancer patient samples (male and female, n = 3287), n (%).

Characteristics	Male	Female
Age
< 60	852 (33.5)	339 (45.4)
≥ 60	1689 (66.5)	407 (54.6)
CEA
< 5	1949 (76.7)	629 (84.3)
≥ 5	592 (23.3)	117 (15.7)
CA19-9
< 37	1988 (78.2)	585 (78.4)
≥ 37	553 (21.8)	161 (21.6)
CA242
< 15	2125 (83.6)	610 (81.8)
≥ 15	416 (16.4)	136 (18.2)
CA72-4
< 6.9	2051 (80.7)	610 (81.8)
≥ 6.9	490 (19.3)	136 (18.2)

CEA: Carcinoembryonic antigen; CA: Carbohydrate antigen.

Table 6 Logistic regression analysis of multiple clinicopathological characteristic variables related to the location of gastric cancer (female).

		P value	OR
Middle	CA19-9	0.041	2.339
	CA72-4	< 0.001	7.052
Lower	Age	0.003	0.387
	CEA	0.034	0.312
	CA72-4	< 0.001	7.104
Upper	Age	< 0.001	0.150
	CEA	0.010	0.241
	CA72-4	< 0.001	8.152

CEA: Carcinoembryonic antigen; CA: Carbohydrate antigen; OR: Odd ratio.

In male patients, the primary influencing factors for gastric body cancer were CEA (P < 0.001, OR = 3.545) and CA19-9 (P < 0.001, OR = 2.246). Elevated levels of CEA and CA19-9, increased the probability of carcinoma in the gastric body by 3.545-fold and 2.246-fold, respectively, compared to total stomach cancer. For distal GC, the main influencing factors were CEA (P < 0.001, OR = 2.869) and CA19-9 (P < 0.001, OR = 1.932). High expression of CEA and CA19-9 increased the probabilities of distal gastric carcinoma by 2.869-fold and 1.932-fold, respectively. The primary influencing factors for proximal GC were age (P < 0.001, OR = 0.501), CEA level (P < 0.001, OR = 2.473), and CA19-9 level (P = 0.015, OR = 1.539). High expression of CEA and CA19-9 increased the probability of proximal gastric carcinoma by 2.473-fold and 1.539-fold, respectively, compared to that of total gastric carcinoma. Additionally, increasing age, especially above 60 years, was associated with a higher probability of total stomach cancer than proximal stomach cancer (Table 7).

Table 7 Logistic regression analysis of multiple clinicopathological characteristic variables related to the location of gastric cancer (male).

		P value	OR
Middle	CEA	< 0.001	3.545
	CA19-9	< 0.001	2.246
Lower	CEA	< 0.001	2.869
	CA19-9	< 0.001	1.932
Upper	Age	< 0.001	0.501
	CEA	< 0.001	2.473
	CA19-9	0.015	1.539

CEA: Carcinoembryonic antigen; CA: Carbohydrate antigen; OR: Odd ratio.

Based on the foregoing analysis, CA72-4 has emerged as a pivotal tumor marker in female patients, particularly for assessing the incidence of GC subsites. In contrast, CA72-4 was not used to assess the incidence of GC sub-sites in male patients. Therefore, we believe that CA72-4 is an important tumor marker for studying gender difference in the location of GC. The CEA index is important for predicting total stomach cancer in females. In male patients, CEA and CA19-9 are prominent tumor markers for analyzing the incidence of GC.

DISCUSSION

In this study, we demonstrated that the location of GC was associated with patient survival: Patients with gastric body and distal GC exhibiting longer survival than those with proximal GC or total GC. Five variables [gender (P < 0.001), age (P < 0.001), CEA (P < 0.001), CA19-9 (P < 0.001), and CA72-4 (P = 0.006)] were found to influence the location of GC through univariate correlation analysis, Spearman correlation analysis, and logistic regression analysis. It was further inferred that age was an important factor in the development of total GC, CA72-4 was an important factor in the incidence of GC subsites in female patients, and CEA and CA19-9 were important factors in the incidence of GC subsites in male patients.

Differences in prognostic survival due to site of GC are associated with several factors. To further understand the causes of differences in postoperative patient survival based on GC location, we conducted a comprehensive search to summarize the physical, physiological, and genomic mechanisms affected by distal GC and proximal GC. This analysis was performed to elucidate the differences in survival prognosis attributable to tumor location. Some researchers have proposed that distal GC may have a better prognosis because of the early symptoms caused by pyloric obstruction, a complication that facilitates early diagnosis and treatment[11]. However, no conclusive evidence currently exists to support this hypothesis. A more reliable study demonstrated that gastric cardia cancer is highly invasive and more likely to involve blood vessels and lymph nodes than GC in other locations[12]. When GC invades large blood vessels, it may cause hemorrhage. Invasion of the esophagus can cause gastroesophageal stricture, resulting in dysphagia and subsequent malnutrition, both of which negatively impact survival. Moreover, vascular invasion and metastatic lymph node involvement have long been recognized as important factors in worsening the prognosis of patients with GC[13-15]. Reducing the incidence and mortality of GC remains a critical health objective. Our research proves that age, gender, and tumor markers influence the location of GC and patient survival. Previous studies have found that elevated serum levels of CEA and CA19-9 are closely related to tumor invasion, lymph node metastasis and TNM stage, while CA72-4 is more frequently observed in patients with vascular involvement[16-18]. Additionally, it has been postulated that elevated serum CEA levels may promote peritoneal tumor metastasis[19]. These results demonstrate that these tumor markers influence the prognostic survival of patients.

Contrary to our findings, some studies have shown that a smoking history is not only associated with an increased risk of GC overall[20], but specifically increases the risk of distal GC, without significantly affecting the risk of cardia or gastric body cancer[21]. In addition, an earlier case-control trial in Sweden identified BMI as an important factor in cardia cancer[22], with a higher prevalence of obesity among patients with this cancer type[23]. Furthermore, a Japanese study revealed that alcohol intake was associated with an increased risk of adenocarcinoma at the gastric cardia, possibly because alcohol promotes gastroesophageal reflux disease and thus increases the associated risk[24]. The variation in results may be attributed to difference in sample size. Previous studies on GC had sample sizes of fewer than 1000 patients, whereas our study included over 3000 patients. Therefore, our large sample size likely contributes to more accurate and objective findings. However, although factors such as family history[25] and AFP[26] have been reported to be associated with GC, there is limited conclusive evidence linking them to the tumor location. This is consistent with the results of our clinical data analysis.

This study has important clinical implications for the treatment of GC and addresses a gap in the existing literature. The identification of age, gender, and tumor markers (CEA, CA19-9, and CA72-4) as factors influencing GC location will contribute to the further development and standardization of clinical diagnostic methods with improved sensitivity and specificity. These findings may also enhance prognostic assessments for GC patients.

There are also some limitations in this study. First, the patient information collected was not comprehensive enough. Poor dietary habits are known to increase Helicobacter pylori infection, which causes persistent chronic inflammation of the gastric mucosa, and increases the risk of distal gastric adenocarcinoma[27,28]. Additionally, the incidence of GC varies greatly by geographic location, ethnicity, and socioeconomic status. Distal GC is more prevalent in developing countries, among Black populations, and in lower socioeconomic groups, whereas proximal tumors are more common in developed countries, among White populations, and in higher socioeconomic classes[29,30]. Unfortunately, these variables were not collected in our current study. Moreover, due to the vast geographic expanse of China, many post-surgical oncology patients return to local hospitals for follow-up treatment. We did not collect this data because of the lack of interoperability between medical record systems across hospitals in China, as well as patients’ limited ability to accurately report their treatment plans. Future research, if feasible, could benefit from sequencing tumor tissue specimens or conducting animal model experiments to further investigate the gene expression profiles of tumors at different gastric sites. Such studies could help explore the relationships between these differentially expressed genes and tumor markers and identify potential interconnected pathways that could guide subsequent evaluation of tumor location and prognosis.

CONCLUSION

In conclusion, this study illustrated that tumor location affects prognostic survival. In addition, we found that the five clinicopathological indices of gender, age, CEA, CA19-9, and CA72-4 levels exhibit varying degrees of influence on the location of GC. Therefore, early screening and early prevention are particularly important.

ACKNOWLEDGEMENTS

We thank all the laboratory members for insightful discussions.