Trends, clinicopathological features, and prognosis of 687 cases of early-onset gastric cancer: A single-center, retrospective study

Abstract

BACKGROUND

Early-onset gastric cancer (EOGC) presents distinct clinicopathological features compared with late-onset gastric cancer. However, data on temporal trends and prognosis of EOGC in certain regions remain limited.

AIM

To investigate the temporal trends, clinicopathological features, and prognosis of EOGC in a regional cancer center over a 10-year period.

METHODS

This single-center, retrospective study included newly diagnosed patients with EOGC (age ≤ 45 years and histopathologically confirmed gastric adenocarcinoma) admitted to Jiangxi Cancer Hospital between January 1, 2013, and December 31, 2022. Clinicopathological data were collected from medical records. Temporal trends were analyzed using the Cochran-Armitage trend χ² test. Overall survival (OS) was estimated using the Kaplan-Meier survival curve, and survival differences were compared using the log-rank test.

RESULTS

A total of 687 patients were included, with a female predominance (53.28%). The most common primary location was the gastric body (67.83%), and diffuse-type histology predominated (90.15%); 35.92% of the patients tested positive for Helicobacter pylori. Common metastatic sites included lymph nodes (14.70%), ovaries (11.20%), and peritoneum (8.44%). From 2013 to 2022, the proportion of EOGC among all gastric cancers declined significantly from 12.05% to 6.53% (P = 0.000). However, within the EOGC population, the proportion of patients ≤ 35 years increased (P = 0.042). The proportion of overweight patients increased significantly (P = 0.010). The median OS of the entire cohort was 47.0 months, with 5-year and 10-year survival rates of 43.8% and 21.8%, respectively. Survival improved significantly over the 10-year period (P = 0.000), with median OS extending from 20-27 months in earlier years to 61-74 months in later years. Higher body mass index and radical gastrectomy were associated with better survival (both P < 0.01).

CONCLUSION

EOGC accounts for a decreasing proportion of all gastric cancers but presents at a younger age. The disease is characterized by female predominance, diffuse histology, and high propensity for peritoneal and ovarian metastasis. OS has improved markedly over time, likely due to increased rates of radical surgery, better nutritional status, and advances in systemic therapies.

Key Words: Early-onset gastric cancer; Clinicopathological features; Overall survival; Prognosis; Trend

Core Tip: Early-onset gastric cancer accounts for a decreasing proportion of all gastric cancers but presents at a younger age. The disease is characterized by female predominance, diffuse histology, and high propensity for peritoneal and ovarian metastasis. Overall survival has improved markedly over time, likely due to increased rates of radical surgery, better nutritional status, and advances in systemic therapies.

INTRODUCTION

Globally, 968350 new cases and 659853 deaths from gastric cancer were reported in 2022[1]. Gastric cancer is more common in men, and the median age at diagnosis is 68 years[2]. The incidence and mortality of gastric cancer have shown a decreasing trend; however, there are still some countries showing an increasing trend, especially among populations younger than 45 years[3]. Early-onset gastric cancer (EOGC), also known as young gastric cancer, is characterized by unique clinical and pathological features in contrast to late-onset gastric cancer (LOGC). In recent decades, significant changes have occurred in the dietary habits, lifestyles, and economic levels of the population in southeastern China, which may affect the incidence and clinicopathological features of EOGC. Therefore, we systematically reviewed the trends, clinicopathological features, and prognosis of EOGC in a single center.

MATERIALS AND METHODS

Study design

In this study, EOGC was defined as gastric cancer with an age of diagnosis of less than 45 years. The main objectives were to investigate the clinicopathological features of EOGC in a province in southeastern China, with particular emphasis on the changing trends of these characteristics over a 10-year period. Overall survival (OS) time was followed up, and a univariate prognostic analysis was performed (Figure 1). This study was approved by the Medical Ethics Committee of Jiangxi Cancer Hospital (Approval No. 2025ky328).

Figure 1 Flowchart of the patient selection process. EOGC: Early-onset gastric cancer.

Study population

A retrospective study was conducted on newly diagnosed patients with EOGC admitted to Jiangxi Cancer Hospital from January 1, 2013, to December 31, 2022. Inclusion criteria were age ≤ 45 years, newly diagnosed, and histopathologically or cytologically confirmed as gastric adenocarcinoma by gastroscopic biopsy. Exclusion criteria were other pathological types (e.g., neuroendocrine carcinoma, gastrointestinal stromal tumor, lymphoma, etc.), concurrent with other cancers, and postoperative recurrence or metastasis.

Data collection

Patient data and information were collected through the medical record system and included the following: Gender; age at diagnosis; height, weight, and body mass index (BMI); smoking and alcohol consumption history; history of benign gastric disease, family history, whether gastric cancer was detected by health examination, tumor location, and Lauren classification of pathological type; expression of human epidermal growth factor receptor 2, mismatch repair, and programmed cell death ligand 1; baseline tumor-node-metastasis stage and sites and organs of metastasis; and whether radical surgery (RS) was performed.

Follow-up

Follow-up was conducted via telephone and outpatient reexaminations. The last follow-up date was May 31, 2025. The median follow-up time was 61.0 months (range: 30.0 months to 148.0 months). Patients lost to follow-up were not included in the survival analysis. The endpoint of this study was OS, defined as the time from diagnosis of gastric adenocarcinoma to gastric cancer-related death or the follow-up deadline.

Statistical analysis

Statistical analyses were performed using SPSS version 27. The Cochran-Armitage trend χ² test was used to analyze temporal trends. The Kaplan-Meier (product-limit) method was used to estimate survival functions for different groups, and the log-rank test was used for between-group comparisons. A significance level of P = 0.05 was adopted.

RESULTS

The clinicopathological features of EOGC

A total of 687 patients were included, of whom 366 (53.28%) were female. The median age of onset was 40 years (range: 17-45 years). Eighteen patients had a family history of gastric cancer, and six patients were diagnosed by health examination. The most common primary tumor site was the gastric body (67.83%), followed by the pylorus (26.78%). The diffuse type was the predominant pathological type, accounting for 90.15% of cases. At the time of diagnosis, 20.67% of patients were underweight, and 13.39% were overweight or obese. In addition, 35.92% of the patients tested positive for Helicobacter pylori (H. pylori). The most common metastatic sites at initial treatment were lymph nodes (14.70%), ovaries (11.20%), peritoneum (8.44%), liver (5.82%), and bone (2.77%). A total of 484 patients (70.45%) underwent RS (Table 1).

Table 1 The clinicopathological features of 687 cases of early-onset gastric cancer.

Variables	n (%)
Gender
Male	321 (46.72)
Female	366 (53.28)
Age
≤ 24	16 (2.33)
25-35	164 (23.87)
≥ 36	507 (73.80)
Smoking history
Yes	138 (20.09)
No	549 (79.91)
Alcohol history
Yes	111 (16.16)
No	576 (83.84)
Stomach disease history
Yes	15 (2.18)
No	672 (97.82)
Family history
Yes	18 (2.62)
No	669 (97.38)
BMI
< 18.5	142 (21.10)
18.5-24.9	439 (65.23)
25.0-29.9	84 (12.48)
≥ 30	8 (1.19)
Unknown	14
Health check-up
Yes	6 (0.87)
No	681 (99.13)
H. pylori infection
Yes	213 (35.92)
No	380 (64.02)
Unknown	94
Tumor location
Fundus cardia	31 (4.51)
Body	466 (67.83)
Antrum pylorus	184 (26.78)
Others	7 (1.02)
Lauren classification
Intestinal	59 (8.81)
Diffuse	604 (90.15)
Mixed	7 (1.04)
Unknown	17
HER-2
Positive	28 (4.94)
Negative	539 (95.06)
Unknown	120
MMR
dMMR	12 (2.27)
pMMR	516 (97.73)
Unknown	159
PD-L1
CPS ≥ 5	54 (26.73)
CPS < 5	148 (73.27)
Unknown	485
Radical surgery
Yes	484 (70.45)
No	203 (29.55)
TNM stage
I	97 (14.18)
II	103 (15.06)
III	271 (39.62)
IV	213 (31.14)
Unknown	3
Metastatic site
Lymph node	101 (14.70)
Ovary1	41 (11.20)
Peritoneum/malignant ascites	58 (8.44)
Liver	40 (5.82)
Bone	19 (2.77)
Others	8 (1.16)

¹The formula for calculating ovarian metastasis is based on female patients, not the entire group of patients.

Unknown patients are not included in the total number. BMI: Body mass index; H. pylori: Helicobacter pylori; HER-2: Human epidermal growth factor receptor 2; MMR: Mismatch repair; dMMR: Deficient mismatch repair; pMMR: Proficient mismatch repair; PD-L1: Programmed cell death ligand 1; CPS: Combined positive score; TNM: Tumor-node-metastasis.

The trends of clinicopathological features among EOGC

From 2013 to 2022, the absolute number of EOGC cases showed a trend of decreasing year by year, and the proportion of EOGC among all gastric cancer cases showed a gradual year-by-year decline, from 12.05% in 2013 to 6.53% in 2022 (Table 2). There was a significant linear trend in the composition ratio between EOGC and LOGC over the years (χ² = 34.482, P = 0.000), suggesting that the age structure of gastric cancer onset is shifting toward later onset (Figure 2A). Furthermore, within the EOGC population, there was a linear trend in the composition ratio among the three age subgroups over time (χ² = 4.141, P = 0.042), with an increasing proportion of patients aged ≤ 35 years (Figure 2B), indicating that the onset of EOGC is becoming younger.

Figure 2 The trends of early-onset gastric cancer. A: Patient group; B: Age group; C: Gender group; D: Body mass index group; E: Lauren classification; F: Tumor location. BMI: Body mass index; EOGC: Early-onset gastric cancer; LOGC: Late-onset gastric cancer.

Table 2 The proportion of early-onset gastric cancer among gastric cancer each year.

Year	EOGC (n)	LOGC (n)	GC (n)	The proportion of EOGC/GC
2013	73	533	606	12.05%
2014	104	795	899	11.57%
2015	96	760	856	11.21%
2016	85	885	970	8.86%
2017	68	742	810	8.40%
2018	69	670	729	8.09%
2019	66	737	803	8.22%
2020	41	708	749	5.47%
2021	52	678	730	7.12%
2022	43	615	658	6.53%

EOGC: Early-onset gastric cancer; LOGC: Late-onset gastric cancer; GC: Gastric cancer.

There was no significant linear trend in the composition ratio between female and male patients (χ² = 0.006, P = 0.940) (Figure 2C). Among patients with EOGC, there was a significant linear trend in the composition ratio across BMI groups (χ² = 6.712, P = 0.010), with a steadily increasing proportion of overweight individuals and a corresponding decrease in the proportion of underweight individuals (Figure 2D). The predominant pathological type of EOGC was diffuse-type gastric cancer. Over the years, there was no significant linear trend in the composition ratio of different Lauren subtypes (χ² = 0.061, P = 0.804) (Figure 2E). Over the 10-year period, the distribution of primary tumor sites in EOGC remained relatively stable, with no significant linear trend (χ² = 2.475, P = 0.116) (Figure 2F).

The OS of patients with EOGC

A total of 674 patients were included in the survival analysis. The median OS of the entire cohort was 47.0 months [95% confidence interval (CI): 37.9-56.1 months]. The 1-year, 5-year, and 10-year survival rates were 76.1%, 43.8%, and 21.8%, respectively (Figure 3).

Figure 3 The overall survival of 674 early-onset gastric cancer patients.

The trends of OS among EOGC by year of diagnosis

The Kaplan-Meier survival analysis showed a significant difference in OS among EOGC patients diagnosed in different years between 2013 and 2022 (log-rank P = 0.000) (Figure 4). Patient survival prognosis showed a significant year-by-year improvement, with median OS time increasing from 20.0-27.0 months to 61.0-74.0 months.

Figure 4 The survival curve for early-onset gastric cancer patients by year of diagnosis (2013-2022).

Univariate analysis of OS among EOGC

The OS of EOGC patients was similar between genders. The median OS was 42.0 months (95%CI: 30.3-53.7 months) for females and 55.0 months (95%CI: 42.6-67.4 months) for males. The log-rank test indicated no statistically significant difference in survival between them (P = 0.157) (Figure 5A).

Figure 5 Survival curves for early-onset gastric cancer patients by different clinical features. A: Sex; B: Age; C: Body mass index categories; D: With and without undergo radical surgery. EOGC: Early-onset gastric cancer; BMI: Body mass index; RS: Radical surgery; NRS: Not undergo radical surgery.

EOGC patients were divided into three groups according to age of onset: ≤ 24 years, 25-35 years, and 36-45 years. The median OS was 33.0 months (95%CI: 1.0-70.1 months), 44.0 months (95%CI: 22.4-65.6 months), and 51.0 months (95%CI: 40.8-61.2 months), respectively. The log-rank test showed no statistically significant difference in survival among the three groups (P = 0.860) (Figure 5B).

EOGC patients were divided into four groups based on BMI: Underweight (BMI < 18.5), normal weight (BMI: 18.5-24.9), overweight (BMI: 25-29.9), and obese (BMI > 30). The median OS for the underweight group was 35.0 months (95%CI: 16.0-54.0 months); for the normal weight group, 42.0 months (95%CI: 30.6-53.4 months); for the overweight group, 83.0 months (95%CI: 57.8-108.2 months); and the median OS was not reached for the obese group. The Kaplan-Meier survival analysis showed a significant difference in OS among EOGC patients across BMI groups (P = 0.003), with greater survival benefit in the higher BMI groups (Figure 5C).

Furthermore, survival analysis showed that patients who underwent radical gastrectomy (RS) had significantly better OS than those who did not undergo RS. The 5-year survival rates were 54.4% and 17.1% in the two groups, respectively. The median survival times were 71.0 months (95%CI: 60.6-81.4 months) and 13.0 months (95%CI: 10.3-15.7 months), respectively. The log-rank test indicated that the difference was statistically significant (P = 0.000) (Figure 5D).

DISCUSSION

EOGC, also referred to as young gastric cancer, lacks a universally accepted age cutoff, although most studies define it as age < 45 years[4-6]. In this single-center, retrospective study of 687 patients with EOGC in a province in southeastern China, we observed several important findings over a 10-year period.

The data showed that the proportion of EOGC among all gastric cancer cases in our center decreased significantly from 12.05% in 2013 to 6.53% in 2022, with a significant linear trend (χ² = 34.482, P = 0.000). This suggests that the age structure of gastric cancer is shifting toward later onset. Similar trends have been observed globally. Lin et al[3] reported that although gastric cancer incidence and mortality are declining overall, certain populations under 45 years of age show increasing trends. In the United States, Bergquist et al[7] found that the incidence of non-cardia gastric cancer among individuals aged 25-39 years increased by 2.6% annually between 1977 and 2011. However, our finding of a declining proportion of EOGC differs from some Western reports, possibly reflecting regional differences in H. pylori prevalence, dietary patterns, and socioeconomic development in southeastern China.

Notably, within the EOGC population, the proportion of patients aged ≤ 35 years increased significantly over time (χ² = 4.141, P = 0.042), indicating that EOGC is presenting at a younger age. This “younger among the young” phenomenon has been observed elsewhere. Rona et al[5] reported that patients with EOGC often present at an advanced stage and at a younger age within the young cohort. Similarly, a Korean study by Mun et al[6] using proteogenomic characterization identified distinct molecular features in very young EOGC patients (≤ 40 years). These findings suggest that the biological behavior of EOGC may differ across age subgroups, warranting further investigation.

Unlike traditional gastric cancer, which predominates in males with a male-to-female ratio of approximately 2:1[2], in our cohort, EOGC showed a female predominance (53.28%). The female predominance in EOGC has been consistently reported in the literature. Liu et al[4] found that among EOGC patients, 52.8% were female, whereas among LOGC patients, only 32.4% were female.

Notably, there was no significant linear trend in the female-to-male composition ratio over our 10-year period, suggesting that the sex predominance in EOGC has remained stable over time. Furthermore, univariate analysis revealed no significant difference in OS between female and male patients. This finding is consistent with Liu et al[4], who also reported no sex-based survival difference in EOGC, but contrasts with traditional gastric cancer, where female sex has been associated with better prognosis in some studies[8]. The lack of a sex-related survival difference in EOGC may be explained by the overriding impact of aggressive tumor biology (e.g., diffuse histology, signet-ring cells) that diminishes the protective effect typically seen in older female patients.

Over the 10-year period, the distribution of Lauren classification and primary tumor location remained stable. The diffuse type was the predominant pathological type, accounting for 90.15% of cases, which aligns with well-established literature indicating that EOGC is more frequently diffuse than LOGC[4-6,9]. The gastric body was the most common primary site, followed by the pylorus, with no significant linear trend over time. This stability suggests that the biological phenotype of EOGC has remained consistent over the past decade.

H. pylori stands as the predominant infectious agent linked to the onset of gastric cancer. The infection rate of H. pylori has been decreasing over the past decades. The prevalence significantly decreased from 58.3% (95%CI: 50.7%-65.5%) in the period 1983-1994 to 40.0% (95%CI: 38.2%-41.8%) in the period 2015-2019. The prevalence increased with age, ranging from 28.0% (95%CI: 23.9%-32.5%) in children and adolescents to 46.1% (95%CI: 44.5%-47.6%) in adults[10]. In this study, the infection rate of H. pylori in EOGC was 35.9%, suggesting a reduced role of H. pylori in the development of EOGC.

In contrast, BMI composition changed significantly, with a continuously increasing proportion of overweight individuals and a corresponding decrease in the proportion of underweight individuals. This likely reflects the rapid economic development and Westernization of dietary habits in southeastern China over the past decades, characterized by increased intake of high-calorie, high-fat foods and reduced physical activity[11]. Similar trends in rising BMI among young Asian populations have been documented[12].

Higher BMI was associated with better survival in our cohort: The overweight group had a median OS of 83.0 months, while the underweight group had only 35.0 months (P = 0.003). This “obesity paradox” has been observed in several gastric cancer studies. A meta-analysis of 23 studies by Park et al[13] found that overweight and obese patients with gastric cancer had better OS compared with normal-weight patients. Similarly, Liu et al[4] reported that higher BMI was associated with improved prognosis in EOGC. Potential explanations include: (1) Better nutritional reserve to tolerate aggressive treatments such as chemotherapy and surgery; (2) Lower likelihood of cachexia; and (3) Possible biological differences in tumor behavior among patients with different metabolic profiles. However, the obesity paradox remains controversial, and some studies have reported opposite findings[14], highlighting the need for prospective studies.

The entire cohort had a median OS of 47.0 months, with 5-year and 10-year survival rates of 43.8% and 21.8%, respectively. Compared with historical data, these survival figures are relatively favorable. For context, Rona et al[5] reported a median OS of only 17.6 months for EOGC patients in a Western cohort, likely due to a more advanced stage at presentation. The better survival in our cohort may be partly explained by the high proportion of patients (70.45%) who underwent radical gastrectomy.

More importantly, survival improved significantly year by year from 2013 to 2022 (log-rank P = 0.000), with median OS extending from 20-27 months in earlier years (2013-2015) to 61-74 months in later years (2020-2022). Several factors may explain this substantial improvement: Firstly, the increasing rate of RS with curative intent. In our cohort, patients who underwent radical gastrectomy had significantly better survival (median OS: 71.0 months vs 13.0 months, P = 0.000). Over the 10-year period, surgical techniques have evolved, with wider adoption of D2 lymphadenectomy and minimally invasive approaches (laparoscopic and robotic gastrectomy), which have been shown to improve outcomes with reduced morbidity[15,16]. Secondly, advances in perioperative and systemic therapies. The introduction of perioperative chemotherapy (e.g., FLOT regimen) has improved survival in locally advanced gastric cancer[17]. Additionally, targeted therapies and immunotherapies have become available: Trastuzumab for human epidermal growth factor receptor 2-positive gastric cancer[18], and immune checkpoint inhibitors (e.g., nivolumab, pembrolizumab) for programmed cell death ligand 1-positive advanced gastric cancer[19]. Although our study did not collect detailed treatment data beyond surgery, the temporal improvement likely reflects the progressive implementation of these therapies. Thirdly, improvements in supportive care, including better nutritional support, management of treatment-related toxicities, and multidisciplinary team approaches.

Similar temporal improvements in EOGC prognosis have been reported. Liu et al[4] demonstrated that patients diagnosed in more recent years (2013-2018) had significantly better survival than those diagnosed earlier (2008-2012) in a Chinese cohort. A large population-based study from the Netherlands by van der Kaaij et al[20] also showed improved survival for both EOGC and LOGC over time, attributed to centralization of care and standardized treatment protocols.

Limitations

Several limitations of this study should be acknowledged. First, EOGC has distinct genomic alterations, such as CDH1 mutations[21], as a single-center retrospective study, no patient received CDH1 testing or genetic counseling, selection bias may be present, and the findings may not be generalizable to other regions of China. Second, detailed data on specific chemotherapeutic regimens, targeted therapy, immunotherapy, and radiation therapy were not collected, limiting our ability to attribute survival improvements to specific treatments. Third, the sample size for certain subgroups (e.g., the obese group) was relatively small, limiting statistical power for some comparisons. Finally, changes in diagnostic and staging modalities over the 10-year period may have introduced bias.

CONCLUSION

EOGC shows female predominance and a high propensity for peritoneal and ovarian metastasis. From 2013 to 2022, the proportion of EOGC among all gastric cancers declined, but within the EOGC population, the proportion of patients aged ≤ 35 years increased, indicating a trend toward younger age of onset. Meanwhile, BMI at diagnosis has risen significantly over time. In future, prospective studies should be recommended to include genetic assessment of EOGC. Encouragingly, OS has improved markedly; this improvement is likely attributable to increased RS rates, improved nutritional status, advances in systemic therapies, and comprehensive disease management.