Value of serum pepsinogen ratio screening for early gastric cancer and precancerous lesions in Youcheng area

Abstract

BACKGROUND

The 5-year survival rate of patients with advanced gastric cancer remains extremely low (< 15%), whereas the 5-year survival rate of patients with early gastric cancer (EGC) is > 90%. Consequently, strengthening the screening of patients with EGC and precancerous lesions (PCLs) is essential.

AIM

To identify the value of serum pepsinogen ratio (PGR) screening for EGC and PCLs in the Shengli Oilfield Central Hospital.

METHODS

We first selected 385 patients with gastric lesions in the Youcheng area, determining benign lesions, PCLs, and EGC in 135, 123, and 127 cases, respectively, based on endoscopy and case diagnosis. The positive rates of pepsinogen I, pepsinogen II and Helicobacter pylori (H. pylori) in the three groups were detected, and the PGR was calculated. Subsequently, we plotted receiver operating characteristic curves to analyze the screening value of PGR and H. pylori-positive rates for PCLs and EGC.

RESULTS

PGR expression demonstrated a decreasing trend in patients with benign lesions, PCLs, and EGC successively according to the detection results, whereas the H. pylori-positive rate was notably increased in patients with PCLs and EGC compared to those with benign lesions. The area under the curves (AUCs) of PGR, H. pylori, and their combination in differentiating patients with benign lesions from those with PCLs were 0.611, 0.582, and 0.689, respectively; PGR, H. pylori, and their combination had an AUC of 0.618, 0.502, and 0.618 in distinguishing PCL patients from EGC patients, respectively; the AUCs of PGR, H. pylori, and their combination in discriminating patients with benign lesions from those with EGC were 0.708, 0.581, and 0.750, respectively.

CONCLUSION

PGR has great screening potential for patients with EGC and PCLs in the Youcheng area, and the screening efficiency is further improved by combining the H. pylori-positive rate.

Key Words: Pepsinogen; Pepsinogen ratio; Early gastric cancer; Precancerous lesions; Screening value

Core Tip: To improve the 5-year survival rate of patients with gastric cancer and the detection rate of early gastric cancer (EGC), this study gathered the screening of patients with EGC and precancerous lesion (PCL) as a breakthrough for relevant research. We included 385 patients with gastric lesions in the Youcheng area for clinical analysis and revealed that pepsinogen ratio has the highest diagnostic efficiency in differentiating patients with benign lesions from those with EGC, followed by differentiating patients with PCL from those with EGC and finally distinguishing patients with benign lesions from those with PCL. The combined pepsinogen ratio and Helicobacter pylori detection exhibited a significant efficacy improvement effect in distinguishing patients with benign lesions from those with PCL, as well as differentiating patients with benign lesions from those with EGC.

INTRODUCTION

Gastric cancer (GC) ranks fifth in prevalence among all tumors and fourth in cancer-related death globally[1,2]. The risk of developing this disease is starting to decline but remains high in Asia, Eastern Europe, and South America regions[3]. Early GC (EGC) is a condition in which tumor infiltration involves only the gastric mucosa and submucosa[4]. Patients at this stage demonstrate no specific identifiable symptoms, making it extremely easy to miss the optimal time for diagnosis and treatment[5]. Neoadjuvant chemotherapy, molecularly targeted therapy, immunotherapy, and combination therapy are currently used to intervene in GC, in addition to radical surgery[6]. However, the 5-year survival rate remains extremely low, with statistics indicating a 5-year survival rate of < 15% in patients with advanced GC, compared to 90% in those with EGC[7,8]. Therefore, strengthening EGC and precancerous lesion (PCL) screening has important clinical value for saving the lives of patients with GC.

Helicobacter pylori (H. pylori) generally causes stem cell proliferation disorders in the normal proliferative area of the gastric mucosa, which hinders normal stem cell differentiation, causing intrinsic gland atrophy in the gastric mucosa and GC progression[9]. Moreover, GC may still develop due to gastric mucosal atrophy lesions despite H. pylori eradication[9]. This may partially explain why single screening for H. pylori is not satisfactory and frequently requires joint screening to further improve screening efficiency[10]. Pepsinogen (PG), as a pepsin precursor, is secreted by gastric mucosal cells, including PG I and PG II[11]. PG I primarily comes from the secretion of gastric fundus chief cells, reflecting the function of gastric acid secretory gland cells[12]. PG II primarily originates from the mucous cells in the antral and pyloric glands, as well as the duodenal gland in the upper part of the duodenum, which is associated with many pathological changes, including fundus gland atrophy, epithelial metaplasia, and dysplasia[13]. The PG ratio (PGR), also known as PG I/PG II, reflects the entire pathological process of gastric tissue, from inflammation to cancer, with the advantages of non-invasiveness, simple operation, and low cost as a serum biological indicator compared with gastroscopy[14]. Additionally, PGR has demonstrated high sensitivity and specificity in EGC and great potential for EGC screening[15]. Han et al[16] reported that PGR can be used for GC screening, prevention, and diagnosis. This study mainly analyzes the value of serum PGR in screening patients with EGC and PCLs in the Youcheng area to provide more useful clues for screening patients with EGC and PCLs.

MATERIALS AND METHODS

General data

This retrospective study was conducted on 385 patients with gastric lesions in the Oil City area who were admitted to Shengli Oilfield Central Hospital, with benign lesions determined by endoscopy and case diagnosis in 135 patients, PCLs in 123 cases, and EGC in 127 cases. The positive rates of PGI, PGII, and H. pylori in the three groups were measured, and the PGR was calculated. The three groups demonstrated no statistical difference in baseline data (P > 0.05), which was clinically comparable. This study established strict inclusion and exclusion criteria to ensure the similarity of the research subjects, and the specific criteria are presented below. Sensitivity and specificity: (1) Sensitivity = number of true positives/(number of true positives + number of false negatives) × 100%; and (2) Specificity = number of true negatives/(number of true negatives + number of false positives) × 100%.

Eligibility and exclusion criteria

Inclusion criteria: All patients who underwent endoscopic examination and pathological biopsy, meeting the diagnostic criteria for benign lesions, PCLs, or EGC[17]: (1) Benign lesions mainly include chronic atrophic gastritis, gastric ulcers, gastric polyps, postoperative gastritis, and hypertrophic gastritis; (2) PCLs primarily include dysplasia (intraepithelial neoplasia); and (3) EGC is mainly manifested as the cancer tissue being only confined to the gastric mucosa layer or submucosa, regardless of lymph node metastasis; age was of 30-80 years; not taking any medication (proton pump inhibitors, histamine 2 receptor antagonists, etc.) that affects outcome measures within two weeks before examinations; no history of precancerous diseases, such as chronic atrophic gastritis, gastric ulcers, gastric polyps, remnant stomach after surgery, hypertrophic gastritis, and pernicious anemia; other risk factors for GC (such as high salt intake, pickled diet, smoking, and heavy alcohol consumption); PCLs, including dysplasia (intraepithelial neoplasia); other malignant tumors.

Exclusion criteria: History of benign and malignant digestive tract diseases, including gastroesophageal reflux, benign and malignant intestinal tumors, intestinal tuberculosis, and Crohn’s disease; oral administration of aspirin, warfarin, non-steroidal anti-inflammatory drugs, and other drugs in the past four weeks; trauma and surgery in the past six months; serious cardiovascular, lung, brain, and kidney dysfunctions and other diseases; mental illness and disturbances.

Outcome measures

The PGR and positive rate of H. pylori were recorded. Fasting venous blood (5 mL) was collected from all subjects early in the morning for serological PG I and PG II concentration determination. The serum was separated by centrifugation for immediate enzyme-linked immunosorbent assays after standing for 30 minutes. Two drops of serum were then sucked up with a disposable plastic straw and added to the well of the test strip, and the results were observed and recorded after 5 and 15 minutes. All specimens were operated strictly following laboratory standard operating procedures.

H. pylori was detected with the 13C-urea breath test. The baseline expiratory samples (0 minute) of all subjects on an empty stomach in the morning or 2 hours after meals were given a 13C-urea capsule with 80-100 mL of cold drinking water after collection, and the exhalation was collected for measurement after sitting quietly for 30 minutes. The breath samples before and after taking the 13C-urea capsule were measured with a breath test. The subject was judged to be H. pylori-positive when the positive judgment value was ≥ 4.0 ± 0.4. The testing process was conducted in strict accordance with the standard operating procedures of the instrument. All of the above testing operations followed standardized operating protocols.

Statistical analysis

Statistical Package for the Social Sciences version 23.0 was used for statistical analysis. Quantitative data, expressed by mean ± SD, were analyzed between groups by independent sample t-tests and among groups with one-way analysis of variance, followed by pairwise comparisons with LSD-t tests. Categorical data were presented as the number of cases (percentage), and the comparison between groups was made by the χ² test. Receiver operating characteristic (ROC) curves were utilized to calculate the effectiveness of PGR and H. pylori in diagnosing EGC. A P-value of < 0.05 denoted statistical significance.

RESULTS

General information of three groups of patients

The differences in sex, age, body mass index, family history of GC, smoking history, and drinking history were not statistically different among patients with benign lesions, PCLs, and EGC (P > 0.05) (Table 1).

Table 1 General information of three groups of patients.

Data	Benign lesions (n = 135)	Precancerous lesions (n = 123)	Early GC (n = 127)	χ2/F value	P value
Sex, n (%)				3.819	0.148
Male	75 (55.56)	81 (65.85)	70 (55.12)
Female	60 (44.44)	42 (34.15)	57 (44.88)
Age (years), mean ± SD	56.34 ± 8.07	55.11 ± 7.26	56.50 ± 8.80	1.668	0.190
Body mass index (kg/m2), mean ± SD	23.03 ± 3.36	23.28 ± 3.07	22.67 ± 3.36	1.102	0.333
Family history of GC, n (%)				0.601	0.741
Without	127 (94.07)	116 (94.31)	117 (92.13)
With	8 (5.93)	7 (5.69)	10 (7.87)
History of smoking, n (%)				2.458	0.293
Without	81 (60.00)	62 (50.41)	72 (56.69)
With	54 (40.00)	61 (49.59)	55 (43.31)
History of drinking, n (%)				2.440	0.295
Without	101 (74.81)	85 (69.11)	84 (66.14)
With	34 (25.19)	38 (30.89)	43 (33.86)

GC: Gastric cancer.

PGR and H. pylori-positive rates in three groups

Patients with benign lesions exhibited markedly higher PGR levels than those with PCLs (P < 0.05) and EGC (P < 0.01). Additionally, higher H. pylori-positive rates were identified in patients with GPL and EGC compared to those with benign lesions (P < 0.05) (Figure 1).

Figure 1 Pepsinogen ratio and Helicobacter pylori-positive rates of three groups of patients. A: Pepsinogen ratio of three groups of patients; B: Helicobacter pylori-positive rates of three groups of patients. ^aP < 0.05 and ^bP < 0.01 in the inter-group comparison or in the comparison with benign lesions. PGR: Pepsinogen ratio; GC: Gastric cancer; HP: Helicobacter pylori.

Value of PGR screening for PCLs

ROC curves analysis revealed that the AUC, specificity, and sensitivity for differentiating benign lesions from PCLs were 0.611 [95% confidence interval (CI): 0.542-0.680], 38.52%, and 96.75% in PGR, 0.582 (95%CI: 0.512-0.652), 57.04%, and 59.35% in H. pylori, 0.689 (95%CI: 0.623-0.754), 71.85%, and 66.67% in PGR + H. pylori, respectively (Figure 2 and Table 2).

Figure 2 Receiver operating characteristic curves analysis of the effectiveness of pepsinogen ratio and Helicobacter pylori in distinguishing benign lesions from precancerous lesions. A: Receiver operating characteristic (ROC) curves analysis of the effectiveness of pepsinogen ratio in differentiating benign lesions from precancerous lesions (PCLs); B: ROC curves analysis of the effectiveness of Helicobacter pylori in distinguishing benign lesions from PCLs; C: ROC curves analysis of the effectiveness of pepsinogen ratio + Helicobacter pylori in distinguishing benign lesions from PCLs. PGR: Pepsinogen ratio; HP: Helicobacter pylori.

Table 2 The value of pepsinogen ratio screening for precancerous lesions.

Indicators	AUC	SE	95%CI	Cut-off value	Specificity (%)	Sensitivity (%)
PGR	0.611	0.035	0.542-0.680	34.17	38.52	96.75
H. pylori	0.582	0.036	0.512-0.652	/	57.04	59.35
Combined detection	0.689	0.034	0.623-0.754	0.50	71.85	66.67

PGR: Pepsinogen ratio; AUC: Area under the curve; H. pylori: Helicobacter pylori; CI: Confidence interval.

Value of PGR screening for EGC

ROC curves analysis revealed that the AUC, specificity, and sensitivity for differentiating PCLs from EGC were 0.618 (95%CI: 0.547-0.690), 36.59%, and 99.21% in PGR, 0.502 (95%CI: 0.430-0.573), 59.35%, and 40.94% in H. pylori, and 0.618 (95%CI: 0.547-0.689), 98.43%, and 36.59% in PGR + H. pylori, respectively. Furthermore, ROC curves analysis demonstrated that the AUC, specificity, and sensitivity for differentiating benign lesions from EGC were 0.708 (95%CI: 0.644-0.771), 50.37%, and 99.21% in PGR, 0.581 (95%CI: 0.511-0.650), 57.04%, and 59.06% in H. pylori, and 0.750 (95%CI: 0.690-0.809), 44.44%, and 99.21% in PGR + H. pylori, respectively (Figure 3 and Table 3).

Figure 3 Receiver operating characteristic curves. A: The receiver operating characteristic (ROC) curve of pepsinogen ratio in differentiating precancerous lesions from early gastric cancer (EGC); B: The ROC curve of Helicobacter pylori in distinguishing precancerous lesions from EGC; C: The ROC curve of the combined detection in differentiating precancerous lesions from EGC; D: The ROC curve of pepsinogen ratio in distinguishing benign lesions from EGC; E: The ROC curve of Helicobacter pylori in differentiating benign lesions from EGC; F: The ROC curve of the combined detection in distinguishing benign lesions from EGC. PGR: Pepsinogen ratio; HP: Helicobacter pylori.

Table 3 Value of pepsinogen ratio screening for early gastric cancer.

Screening indicator	Indicators	AUC	SE	95%CI	Cut-off value	Specificity (%)	Sensitivity (%)
Precancerous lesions and early GC	PGR	0.618	0.036	0.547-0.690	20.48	36.59	99.21
	H. pylori	0.502	0.037	0.430-0.573	/	59.35	40.94
	Combined detection	0.618	0.036	0.547-0.689	0.39	98.43	36.59
Benign lesions and early GC	PGR	0.708	0.032	0.644-0.771	20.68	50.37	99.21
	H. pylori	0.581	0.035	0.511-0.650	/	57.04	59.06
	Combined detection	0.750	0.030	0.690-0.809	0.38	44.44	99.21

GC: Gastric cancer; PGR: Pepsinogen ratio; AUC: Area under the curve; H. pylori: Helicobacter pylori; CI: Confidence interval.

DISCUSSION

Upper gastrointestinal endoscopy combined with tissue biopsy is the gold standard for EGC screening. This screening method exhibited sensitivity and specificity of 69% and 96%, respectively, but it has disadvantages, including invasiveness, high cost, and time-consuming[18]. Additionally, it is extremely low in cost-effectiveness as a large-scale screening strategy[19]. Therefore, other methods, such as serum biomarker screening, need to be adopted to improve the screening efficiency and enhance the early diagnosis rate and survival rate of GC[20].

In our study, PGR levels indicated a significant stepwise decreasing trend in patients with benign lesions, PCLs, and EGC successively. However, the H. pylori-positive rate was notably higher in patients with PCLs and EGC vs those with benign lesions, indicating that PGR and H. pylori may both mediate the early pathological procession of GC. Previous studies revealed that PGR levels were abnormally reduced in patients with atrophic gastritis and GC, similar to our results[21,22]. To a certain extent, the above changes in PGR levels are associated with the fact that serum PG I and PG II concentrations reflect the pathological condition of the gastric mucosa. PG I and PG II level alterations are associated with inflammation, atrophy, and intestinal metaplasia in the gastric mucosa[23]. Both PG I and PG II levels demonstrate an upward trend during GC lesion formation, but the degree of increase in PG II levels is faster with the lesion progression (whereas PG I is relatively gentle and not statistically significant), thereby decreasing PGR levels[24]. Studies reported that PG I gradually decreases with GC lesion progression, whereas PG II levels gradually increase, causing a gradual decrease in PGR levels with GC lesion progression[25]. The development process from benign lesions, PCLs to EGC will be accompanied by progressive atrophy of gastric mucosal glands, a decrease in the number of chief cells, and impaired secretory function, causing a decrease in PG I levels from the perspective of molecular and cellular mechanisms[26]. Atrophy of gastric mucosal glands with glandular metaplasia increases PG II levels, and PG I levels are down-regulated when the body is in a state of intestinal metaplasia and GC[27,28]. Another large prospective cohort study indicates that serum PG levels can be used to predict GC risk, especially for non-cardiac GC[29]. Patients with GC are at a higher risk of H. pylori infection, reaching up to 88%, and type I H. pylori affected PGR levels, causing lower PGR levels[30]. This partly helps to explain the higher H. pylori-positive rate and the significantly lower PGR levels in patients with EGC. ROC curves were utilized to analyze the screening value of PGR and the H. pylori-positive rate for PCLs first. We revealed that the AUC, sensitivity, and specificity for distinguishing benign lesions from PCLs were 0.611, 96.75%, and 38.52% in PGR, respectively. A PGR value of < 34.17 indicates that patients with benign lesions demonstrate a higher risk of developing PCLs. The screening data of H. pylori revealed AUC, specificity, and sensitivity for distinguishing benign lesions from PCLs of 0.582, 57.04%, and 59.35%, respectively. The AUC, specificity, and sensitivity were increased to 0.689, 71.85%, and 66.67% under the combined detection of the two, respectively. The above data indicates that PGR exhibits a higher diagnostic efficiency for PCLs compared to H. pylori, and the combination of the two can further improve the screening efficiency for PCLs. Subsequently, the analysis of the diagnostic efficacy of both for EGC revealed that PGR demonstrated an AUC of 0.618 for distinguishing between PCLs and EGC, with excellent sensitivity (99.21%) and low specificity (36.59%). A PGR value of < 20.48 indicates that patients with PCL are at greater risk of developing EGC. However, H. pylori had a lower AUC (AUC = 0.502) than PGR for distinguishing PCLs from EGC, with specificity and sensitivity of 59.35% and 40.94%, respectively. The AUC only increased to be equivalent to PGR (0.618) under their joint testing. The AUC of PGR was as high as 0.708, with a sensitivity of up to 99.21% and a specificity of only 50.37% for differentiating benign lesions from EGC. A PGR value of < 20.68 indicates a relatively high risk for patients with benign lesions to develop into EGC. The AUC of H. pylori was far less than that of PGR (0.581), with specificity and sensitivity of 55%-60%. The AUC of their joint testing could reach as high as 0.750 (the specificity and sensitivity were 44.44% and 99.21%, respectively). In the study by Trivanovic et al[31] showed the AUC of PGR for distinguishing GC from non-malignant lesions is 0.700, similar to our research results. Liang et al[32] proposed a combination of serum carcinoembryonic antigen, cancer antigen 19-9, and cancer antigen 72-4 for GC diagnosis, with a combined AUC of only 0.632 for predicting GC. Fang et al[33] reported the AUCs of neutrophil-to-lymphocyte ratio and platelet-to-lymphocyte ratio detection of GC at 0.715 and 0.707, respectively, whereas the AUC of their combined detection of GC only increased to 0.739. Additionally, Sasakabe et al[34] demonstrated that PGR + H. pylori for diagnosing GC reduces the risk of GC misdiagnosis cases in low-risk patients, indicating that the combination of the two, to some extent, helps improve the diagnostic efficiency of GC. Low PGR is strongly associated with the severity of gastric atrophy in gastric neoplastic lesions and serves as a sensitive biomarker of GC PCLs or EGC, similar to our results[35]. Baek et al[36] emphasized that both PGR and H. pylori positivity indicate early diffuse GC, especially for Korean women younger than 40 years old.

This study has confirmed PGR as a potential screening tool for EGC and PCLs, especially when combined with H. pylori detection, thereby providing a screening method that is cost-effective and non-invasive for the clinical practice of EGC. However, this study has several limitations that need to be resolved in the future: First, a sample size of 385 patients was included for statistical analysis, but it may not entirely represent the broader population, especially those outside the Youcheng area. Second, this study did not consider all potential confounding factors that may affect PG levels, including dietary habits, age, and genetic predisposition. Third, further research on a larger and more diverse population (especially those outside the Youcheng area) is warranted, and emphasis should be placed on formulating explicit clinical guidelines to confirm these results and optimize their practical application in GC screening programs. Fourth, personalized medicine and targeted intervention strategies need to be supplemented to further improve GC-related treatments while supplementing longitudinal studies to track changes in PGR over time and its correlation with cancer progression. This study will be continuously refined based on the above in the future.

CONCLUSION

In summary, PGR demonstrates the highest diagnostic efficacy in distinguishing benign lesions from EGC, between patients with PCLs and those with EGC, and between benign lesions and PCLs. The joint detection by PGR and H. pylori greatly improves the efficiency of distinguishing between benign lesions and PCLs, as well as between benign lesions and EGC, whereas the efficiency of distinguishing PCLs from EGC is equivalent to PGR. The results of this study help guide the screening of EGC in our hospital, thereby improving the early diagnosis and treatment rate of EGC among people in the Youcheng area.