Introduction

Gastric cancer (GC) ranked fifth in incidence and fourth in cancer-related mortality worldwide in 2020.1 The prognosis of GC is closely related to the clinical stage. Early diagnosis and treatment are crucial.2,3 Gastroscopy is the primary method for GC screening; however, patient intolerance to the procedure and the high cost make it unsuitable for routine screening. Several studies have shown that serum pepsinogen I (PG-I), PG-II, and the PG-I/II ratio (PGR) are closely related to the state of the gastric mucosa.4,5 Abnormal serum PGs (including PG-I, PG-II and PGR) results indicate potential presence of atrophic gastritis, which is considered a significant risk factor for GC.5,6 Studies have shown that combining serum concentrations of PG-I, PG-II and PGR with Helicobacter pylori (H. pylori) antibody titers can stratify populations into low-risk and high-risk populations for GC. Individuals who test positive for atrophic gastritis by serum pepsinogen need to undergo endoscopic examination to detect the presence of GC.7,8 The key advantages of pepsinogen in GC screening are its non-invasive nature, making it a convenient and cost-effective tool for population-based screening programs. Thus, establishing appropriate reference ranges of PGs is of substantial significance for GC screening.

Several studies have reported the reference ranges of PGs.9,10 However, in these studies the enrolled individuals were absent from prior gastroscopy evaluation. Thus, the “healthy people” were apparent healthy people. Besides, these studies used only one method to assess PGs, but there are two main commercial methods for detecting PGs: chemiluminescence immunoassay (CLIA) and enzyme-linked immunosorbent assay (ELISA). However, no studies have compared these two methods. How to interpret the results obtained from the two methods is of educational significance to physicians and laboratory technicians.

Therefore, in this study, we detected serum PG-I and PG-II levels by CLIA and ELISA in healthy volunteers excluding abnormal gastroscopy and urea breath test (UBT) results. Our study aimed to (1) establish the reference ranges of PG-I, PG-II, and the PGR using two different methods and (2) compare the results of PGs on two platforms.

Participants and Methods Participants

The Participants were randomly selected from the Chinese Adult Digestive Diseases Surveillance program in 2016, a nationally representative cross-sectional survey. The survey was conducted by random cluster sampling, and village/resident groups were randomly selected by the national working group according to probability proportionate to size sampling guidelines. Six representative regions, Beijing (1140), Guiyang (862), Harbin (294), Nanchang (288), Yichun (288), and Beihai (32), were selected, covering northern (Beijing, Harbin) to southern (Guiyang, Nanchang, Yichun, and Beihai) China. 2904 participants were recruited in strict accordance with the principle of random stratified sampling. Trained medical personnel collected data via questionnaires, anthropometric measurements, gastroscopy, UBT, abdominal ultrasound, and serologic tests. The samples were also used to measure PGs levels for this study.

The exclusion criteria were as follows: (1) Obvious abnormalities in gastroscopy: GC, gastric ulcer, and chronic atrophic gastritis. (2) Obvious abnormalities in UBT: H. pylori infection. (3) Obvious abnormalities in abdominal ultrasound: abdominal tumors. (4) Obvious abnormalities in serologic tests: alanine aminotransferase (ALT)>100 U/L (2-fold of upper normal limit), total cholesterol (TC)>11.4 mmol/L (2-fold of upper normal limit), triglycerides (TG)>6.8 mmol/L (4-fold of upper normal limit). (5) Absence of these examinations. The flowchart illustrating participants’ inclusion is shown in Figure 1.

Figure 1 Flowchart for participants’ inclusion.

Abbreviations: UBT, urea breath test.

Notes: The method of D/R rule to exclude outliers was adapted from Horowitz et al.11

Collection of Blood Samples

Nine milliliters of venous fasting blood were collected from each participant. The samples were centrifuged at 1500 ×g for 10 min to obtain serum samples. All serum samples were frozen at −80°C immediately and then sent to the central laboratory in Beijing via cold-chain transportation for subsequent detection.

Examination Methods UBT

The UBT is a method for H. pylori detection that was established based on the ability of the H. pylori urease enzyme to break down an isotope-labeled urea solution ingested by the patient into carbon dioxide and ammonia.12

The exhaled gas was collected from the enrolled individuals on an empty stomach for the13 C breath test. The breath was stored at room temperature and sent to the medical inspection institution designated by the national project team by cold-chain transportation for detection on the second day.

Abdominal Ultrasound Test

The centers for disease control of each district summoned the participants to the local hospital for abdominal ultrasound examination (bile duct, gallbladder, and liver) (15–20 persons per batch). All participants were examined on an empty stomach (no eating or drinking for 6 h; patients with high blood pressure could take their medication with a small amount of water) by qualified ultrasound doctors. The results were reported via the information collection and management system.

Gastroscopy

Gastroscopy was performed after the participants completed an abdominal ultrasound examination. All participants were examined on an empty stomach (no eating or drinking for 6 h; patients with high blood pressure could take their medication with a small amount of water) by qualified gastroenterologists, and the results were reported via the information collection and management system.

Laboratory Analyses

Serum PG-I and PG-II levels were measured using an automated CLIA analyzer (Abbott ARCHITECT i2000) and commercial ELISA tests (Biohit, Finland) with the appropriate reagents, calibrators, and quality controls. All tests were performed according to the manufacturers’ guidelines.

Carbohydrate antigen 72–4 was measured using Abbott ARCHITECT i2000. ALT, TC, and TG were measured using a Beckman AU5800 instrument with Beckman and Sekisui Medical reagents.

Statistics

95% reference ranges were determined according to the document CLSI-EP28-A3.11 The outlying observations were deleted by D/R rule.11 The 2.5th percentile (P2.5) and the 97.5th percentile (P97.5) of the PGs results were listed as bilateral 95% reference ranges. The 5th percentile (P5) of the PGs results were also listed as unilateral 95% reference ranges. The Kolmogorov–Smirnov test was used to assess the distribution of variables. The Kruskal–Wallis H or Mann–Whitney U-tests were used to compare groups in non-Gaussian distributed data. Spearman correlation analysis was utilized to evaluate the correlation between two methods. The closer the absolute value of the correlation coefficient is to 1, the stronger the correlation between the two variables. Linear regression analysis was employed to derive regression equation. Allowable total error (ATE) zone analysis, based on a graphical technique and simple calculation, was applied to assess the consistency between two methods. 80% of the data fall in the ATE zone and cluster around the bisector, indicating that the two methods agree well. IBM SPSS Statistics 26.0 and GraphPad Prism 8.0 were used for the statistical analyses and data plotting. Two-sided p-values < 0.05 were considered to indicate significance. We divided the participants into three age groups according to the World Health Organization’s age classification criteria: 18–44 years old, 45–59 years old, and ≥60 years old.

Results The Baseline Characteristics of Participants

After applying the strict exclusion criteria, the total number of enrolled participants was 747 (age: 19–64 years; 298 male participants, 449 female participants). The baseline characteristics of participants were shown in Table 1.

Table 1 The Baseline Characteristics of Participants

Concentration of PGs in Healthy Individuals and Affecting Factors

The concentrations of PG-I measured by CLIA, PG-II measured by CLIA, PG-I measured by ELISA, and PG-II measured by ELISA were higher in males than in females (p<0.05), whereas PGR measured by CLIA and PGR measured by ELISA showed no significant differences between males and females (p>0.05) (Table 2).

Table 2 Comparison of Serum PG-I Measured by CLIA, PG-II Measured by CLIA, PGR Measured by CLIA, PG-I Measured by ELISA, PG-II Measured by ELISA, and PGR Measured by ELISA by Sex, Age, and Geographical Location

As shown in Table 2 and Figure 2, the levels of PG-I measured by CLIA increased with age but gradually decreased after age 60 (Figure 2A). PG-II measured by CLIA, PG-I measured by ELISA, and PG-II measured by ELISA consistently increased with age (Figure 2B, D, and E), while PGR measured by CLIA decreased with age (Figure 2C). There was no difference in PGR measured by ELISA among the different age groups (Figure 2F). Moreover, the Mann–Whitney U-test revealed no significant difference in all PGs Results between the 45–59 and >60 age groups (p>0.05). Therefore, the participants were divided into 18–44 and ≥ 45-year groups.

Figure 2 Comparison of serum PGs among different age groups. (A) Comparison of serum PG-I measured by CLIA among different age groups. (B) Comparison of serum PG-II measured by CLIA among different age groups. (C) Comparison of serum PGR measured by CLIA among different age groups. (D) Comparison of serum PG-I measured by ELISA among different age groups. (E) Comparison of serum PG-II measured by ELISA among different age groups. (F) Comparison of serum PGR measured by ELISA among different age groups.

Abbreviations: CLIA, chemiluminescence immunoassay; ELISA, enzyme-linked immunosorbent assay; PGs, pepsinogens (including PG-I, PG-II and PGR); PG-I, pepsinogen I; PG-II, pepsinogen II; PGR, the PG-I/II ratio.

The six regions in this study (Beijing, Harbin, Guiyang, Nanchang, Yichun, and Beihai) were divided into two groups, southern China and northern China, according to the country’s geographic dividing line. The Kruskal–Wallis H-test was used to assess the contribution of these regions to the PGs results. PG-I measured by ELISA was significantly higher in northern China than in southern China (p<0.05), while other PGs results were not affected by geographical location (p>0.05) (Table 2).

Comparison of the Results of Serum PG-I, PG-II, and the PGR Between CLIA and ELISA

Linear regression analysis revealed that the linear regression equation for PG-I is PG-I measured by ELISA = 1.815 * PG-I measured by CLIA + 0.827. Spearman correlation analysis revealed that the correlation coefficient is 0.899 (p<0.001). The linear regression equation for PG-II is PG-II measured by ELISA = 0.968 * PG-II measured by CLIA − 1.249, and the correlation coefficient is 0.887 (p<0.001). The linear regression equation for PGR is PGR measured by ELISA = 2.213 * PGR measured by CLIA + 1.918, and the correlation coefficient is 0.777 (p<0.001).

To compare the consistency of CLIA and ELISA methods, we also refer to another graphical method—allowable total error and limits for erroneous results (ATE-LER) zones—as recommended by the United States Food and Drug Administration (FDA).13 According to the criteria for external quality control established by the National Center for Clinical Laboratories in China, the allowable total errors of PG-I, PG-II, and PGR were target value±30%, and 80% of the data should be within the allowable total errors. Based on this standard, the ATE zone is the gray area between the lines y=−30%*x and y=30%*x in this study (Figure 3). However, only fewer PG-I (2.54%) (Figure 3A), PG-II (62.12%) (Figure 3B), and PGR (0.80%) (Figure 3C) data points fell in the ATE zones. Thus, CLIA and ELISA exhibit poor agreement. Due to the lack of an LER cut-off value, we did not set the LER zones.

Figure 3 Consistencies between PGs measured by CLIA and ELISA using allowable total error zone analysis. (A) Consistencies between serum PG-I measured by CLIA and ELISA using allowable total error zone analysis. (B) Consistencies between serum PG-II measured by CLIA and ELISA using allowable total error zone analysis. (C) Consistencies between serum PGR measured by CLIA and ELISA using allowable total error zone analysis.

Abbreviations: CLIA, chemiluminescence immunoassay; ELISA, enzyme-linked immunosorbent assay; PGs, pepsinogens (including PG-I, PG-II and PGR); PG-I, pepsinogen I; PG-II, pepsinogen II; PGR, the PG-I/II ratio.

Establishment of the Reference Ranges for PGs on Two Platforms

Bilateral reference ranges and unilateral reference ranges of serum PG-I, PG-II, and the PGR by ELISA and CLIA were established for the Chinese population. The reference ranges were stratified by sex and age group (Table 3).

Table 3 Reference Ranges of PG-I Measured by CLIA, PG-II Measured by CLIA, PGR Measured by CLIA, PG-I Measured by ELISA, PG-II Measured by ELISA, and PGR Measured by ELISA by Sex and Age

Discussion

The gold standard for determining gastric mucosal lesions is histological examination. Considering the difficulty in obtaining gastric mucosal tissue samples, more convenient serum samples are a valuable alternative for atrophic gastritis and GC screening. PG-I and PG-II secreted by gastric mucosa cells are proenzymes for pepsin. PG-I is secreted by glands in the gastric fundus and body, and PG-II is secreted in the gastric antrum and duodenum.14 Both PG-I and PG-II decrease in gastric atrophy. PG-I usually shows a more marked decrease than PG-II. Thus, a low PG-I level, a low PGR, or both are good indicators of atrophic changes in the gastric mucosa.14,15 Moreover, atrophic gastritis are associated with elevated GC risk.16,17 It is also well known that the majority of GC is caused by H. pylori infection.18 Therefore, Japanese researchers developed the ABC (D) screening program that combines PGs (PG-I < 70 ng/mL and PGR < 3.0 as positive) atrophic markers and the anti-H. Pylori IgG etiological marker to stratify high-risk patients for further follow-up; however, the sensitivity and specificity of this approach need to be further improved.19 A meta-analysis showed that the cut-off values to predict atrophy vary across countries and detection methods.20 Therefore, it is important to establish appropriate PGs reference ranges according to different countries and detection methods.

In this study, we enrolled 747 healthy participants from a nationally representative population following strict exclusion criteria. Then, we set up the respective reference ranges of PG-I, PG-II, and the PGR stratified by age, sex, and method (Table 3). Several studies on the reference ranges of PGs have been reported.9,10 The strengths of this study are as follows: (1) this study is part of the national Chinese Adult Digestive Diseases Surveillance program. Participants were recruited by random stratified sampling from six representative regions of China, covering the north to south regions. Thus, the conclusion could represent the whole Chinese population. Other reports were performed in a single region of China or other countries. (2) The exclusion criteria in this study were strict. Since all the 747 participants included in this study performed gastroscopy and UBT, those with gastric disease were ruled out accurately. The reference ranges from previous studies may be inappropriate for the enrollment of patients with gastric diseases. (3) The various methods for PGs level analysis may lead to differences in results. At least four PGs methods have been reported, namely CLIA, ELISA, time-resolved fluorescence analysis, and latex-enhanced immunoturbidimetry. However, no studies have compared PGs results between the two mainstream commercial Methods: CLIA and ELISA.

Consistent with previous studies, this study showed that the PGs results were affected by sex and age.21–23 As shown in Table 2, the serum PG-I and PG-II levels of males were significantly higher than those of females, while there was no difference in PGR between males and females. Iijima et al24 demonstrated that the level of gastric acid secretion was higher in males than in females due to the inhibitory effect of estradiol on acid secretion.25 Moreover, in H. pylori-negative participants, the serum pepsinogen concentrations correlated well with acid secretion, but there was no relationship between the PGR and acid secretion.26 This finding may explain the PGs differences between the different sexes.

As shown in Table 2 and Figure 2, the levels of PG-I measured by CLIA increased with age but gradually decreased after age 60, similar to Sun’s findings.21 However, the levels of PG-I measured by ELISA consistently increased with age, similar to Hong’s results.23 Although serum pepsinogen concentrations were well correlated with acid secretion,26 the effect of age on gastric acid secretion is controversial. Iijima24 showed that aging is positively associated with gastric acid secretion in H. pylori-negative individuals, while K Haruma27 found that advancing age does not influence gastric acid secretion in such individuals. In our study, PG-I measured by CLIA and PG-I measured by ELISA showed different trends with age, which requires further research. Besides, with both methods, PG-II levels gradually increased with age, and the PG-II levels were more significant than the PG-I levels, resulting in the gradual decline of the PGR with age.

The correlation coefficients of PG-I, PG-II, and the PGR between CLIA and ELISA were 0.899, 0.887, and 0.777, respectively, indicating a certain correlation between the two methods. Similar conclusions have been obtained in previous studies.28,29 Chiang et al28 compared the ELISA method and latex-enhanced turbidimetric immunoassay produced by GastroPanel (Helsinki, Finland) and LZ-Test (Tokyo, Japan) in the Taiwan population. The coefficient factor of PGR neared 0.7. Leja et al29 studied a Latvian population using two ELISA assays and a latex agglutination assay produced by Biohit (Finland), Vector Best (Russia) and Eiken (Japan), respectively. The correlation of PGs showed a wide range of 0.79–0.90.

However, a high correlation coefficient can only indicate that the two methods are linearly related rather than in agreement with each other.30 Considering the drawbacks of the correlation coefficient, the FDA suggested ATE-LER zones to obtain more accurate results.29,30 This study refers to the approach. As shown in Figure 3A and C, most red points are above the ATE zone, meaning that the ELISA method for PG-I and the PGR deviates over 30% from the CLIA method. As shown in Figure 3B, although the systematic deviation is slight, only 62.12% of the red points are in the ATE zone, far below 80%. Overall, the two methods exhibit a strong linear relationship but less consistency. This finding can provide valuable insights for clinicians interpreting laboratory results. (1) Due to the poor consistency, it is advisable to establish distinct reference ranges for PGs detected by different methods. Clinicians can choose appropriate reference ranges based on the testing platform. (2) When PGs are used for continuous monitoring, it is recommended to consistently use the same testing platform. In cases where historical data includes results from both ELISA and CLIA, linear regression equations can be utilized by clinicians to convert results between the two methods, given their strong linear correlation.

Our study has several limitations. First, the reference range of PG-I measured by ELISA for southern China and northern China respectively could not be established due to the limited enrolled individuals. PG-I measured by ELISA was affected by geographical location, while the other PGs results were not. Second, Due to the lack of a gold standard for comparison, it is unknown which method is closer to the true value. Third, participants were recruited using a method of random stratified sampling, with the aim of reducing sampling bias. However, potential sampling bias may still exist. Fourth, due to limited funding support and personnel, as well as considerations for minimizing patients’ safety, gastroscopic biopsy was not performed in this study.

Notwithstanding these limitations, this study provides a reliable reference range investigation and method comparison for PGs using two platforms. To the best of our knowledge, this is the first study to use gastroscopy to screen apparently healthy individuals to establish PGs reference ranges.

Data Sharing Statement

The datasets used and/or analyzed during this study are available from the corresponding author on reasonable request.

Ethics Approval and Consent to Participate

This study was approved by the Ethical Committee of Peking Union Medical College Hospital (S-T272) and was conducted according to the principles of the Declaration of Helsinki. Informed consent was obtained from all subjects and/or their legal guardian(s).

Acknowledgments

The authors are grateful to all the patients, researchers, and institutions that participated in this study. This paper has been uploaded to ResearchSquare as a preprint: https://europepmc.org/article/ppr/ppr691189

Funding

This study was funded by 1: National Natural Science Foundation of China, Natural Science Foundation of Tibet Autonomous Region (XZ2021ZR-ZY13(Z)). 2: National High Level Hospital Clinical Research Funding(2022-PUMCH-A-059).

Disclosure

The authors declare that they have no competing interests.

References

1. Sung H, Ferlay J, Siegel RL, et al. Global cancer statistics 2020: Globocan estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2021;71(3):209–249. doi:10.3322/caac.21660

2. Machlowska J, Baj J, Sitarz M, Maciejewski R, Sitarz R. Gastric Cancer: Epidemiology, risk factors, classification, genomic characteristics and treatment strategies. Int J Mol Sci. 2020;21(11). doi:10.3390/ijms21114012

3. Smyth EC, Nilsson M, Grabsch HI, van Grieken NC, Lordick F. Gastric cancer. Lancet. 2020;396(10251):635–648. doi:10.1016/s0140-6736(20)31288-5

4. Kiyohira K, Yoshihara M, Ito M, Haruma K, Tanaka S, Chayama K. Serum pepsinogen concentration as a marker of Helicobacter pylori infection and the histologic grade of gastritis; evaluation of gastric mucosa by serum pepsinogen levels. J Gastroenterol. 2003;38(4):332–338. doi:10.1007/s005350300060

5. Zhou MJ, Huang RJ. Catching Up with the World: Pepsinogen screening for gastric cancer in the United States. Can Epidemiol Biom Prev. 2022;31(7):1257–1258. doi:10.1158/1055-9965.Epi-22-0372

6. Adamu MA, Weck MN, Gao L, Brenner H. Incidence of chronic atrophic gastritis: systematic review and meta-analysis of follow-up studies. Eur J Epidemiol. 2010;25(7):439–448. doi:10.1007/s10654-010-9482-0

7. Lomba-Viana R, Dinis-Ribeiro M, Fonseca F, Vieira AS, Bento MJ, Lomba-Viana H. Serum pepsinogen test for early detection of gastric cancer in a European country. Eur J Gastroenterol Hepatol. 2012;24(1):37–41. doi:10.1097/MEG.0b013e32834d0a0a

8. Miki K. Gastric cancer screening using the serum pepsinogen test method. Gastric Can. 2006;9(4):245–253. doi:10.1007/s10120-006-0397-0

9. Zhang L, Niu Y, Lv YJ, et al. Preliminary study on reference interval of serum pepsinogen in healthy subjects. Patient Prefer Adherence. 2021;15:2725–2730. doi:10.2147/ppa.S330656

10. Huang M, Tang AG, Mu S, et al. Serum pepsinogen reference intervals in apparently healthy Chinese population with latex enhanced turbidimetric immunoassay. J Clin Pathol. 2014;67(4):350–354. doi:10.1136/jclinpath-2013-201942

11. Horowitz GL, Altaie S, Boyd JC, Ceriotti F, Garg U, Defining HP. Establishing, and Verifying Reference Intervals in the Clinical Laboratory. Approved Guideline—Third ed. Clinical Laboratory and Standards Institute; 2008:28.

12. O’Connor A. The urea breath test for the noninvasive detection of helicobacter pylori. Methods Mol Biol. 2021;2283:15–20. doi:10.1007/978-1-0716-1302-3_2

13. Center for Devices and Radiological Health UFaDA. Guidance for industry and FDA staff: Recommendations for clinical laboratory improvement amendments of 1988 (CLIA) waiver applications for manufacturers of in vitro diagnostic devices; 2008.

14. Kuipers EJ. In through the out door: serology for atrophic gastritis. Eur J Gastroenterol Hepatol. 2003;15(8):877–879. doi:10.1097/00042737-200308000-00007

15. Gantuya B, Oyuntsetseg K, Bolor D, et al. Evaluation of serum markers for gastric cancer and its precursor diseases among high incidence and mortality rate of gastric cancer area. Gastric Can. 2019;22(1):104–112. doi:10.1007/s10120-018-0844-8

16. Chiang TH, Chiu SY, Chen SL, et al. Serum pepsinogen as a predictor for gastric cancer death: A 16-year community-based cohort study. J Clin Gastroenterol. 2019;53(5):e186–e193. doi:10.1097/mcg.0000000000000992

17. Kudo T, Kakizaki S, Sohara N, et al. Analysis of ABC (D) stratification for screening patients with gastric cancer. World J Gastroenterol. 2011;17(43):4793–4798. doi:10.3748/wjg.v17.i43.4793

18. Correa P. Helicobacter pylori and gastric carcinogenesis. Am J Surg Pathol. 1995;19(Suppl 1):S37–43.

19. Miki K. Gastric cancer screening by combined assay for serum anti-Helicobacter pylori IgG antibody and serum pepsinogen levels – ”ABC method”. Proc Jpn Acad Ser B Phys Biol Sci. 2011;87(7):405–414. doi:10.2183/pjab.87.405

20. Huang YK, Yu JC, Kang WM, et al. Significance of serum pepsinogens as a biomarker for gastric cancer and atrophic gastritis screening: a systematic review and meta-analysis. PLoS One. 2015;10(11):e0142080. doi:10.1371/journal.pone.0142080

21. Sun LP, Gong YH, Wang L, Yuan Y. Serum pepsinogen levels and their influencing factors: a population-based study in 6990 Chinese from North China. World J Gastroenterol. 2007;13(48):6562–6567. doi:10.3748/wjg.v13.i48.6562

22. Huang RG, Xiao HL, Zhou B, et al. Serum pepsinogen levels are correlated with age, sex and the level of helicobacter pylori infection in healthy individuals. Am J Med Sci. 2016;352(5):481–486. doi:10.1016/j.amjms.2016.08.005

23. Yu H, Liu Y, Jiang S, et al. Serum pepsinogen II levels are doubled with Helicobacter pylori infection in an asymptomatic population of 40,383 Chinese subjects. Medicine. 2021;100(27):e26562. doi:10.1097/md.0000000000026562

24. Iijima K, Ohara S, Koike T, Sekine H, Shimosegawa T. Gastric acid secretion of normal Japanese subjects in relation to Helicobacter pylori infection, aging, and gender. Scand J Gastroenterol. 2004;39(8):709–716. doi:10.1080/00365520410005911

25. Adeniyi KO. Gastric acid secretion and parietal cell mass: effect of sex hormones. Gastroenterology. 1991;101(1):66–69. doi:10.1016/0016-5085(91)90460-3

26. Iijima K, Sekine H, Koike T, Imatani A, Ohara S, Shimosegawa T. Serum pepsinogen concentrations as a measure of gastric acid secretion in Helicobacter pylori-negative and -positive Japanese subjects. J Gastroenterol. 2005;40(10):938–944. doi:10.1007/s00535-005-1677-x

27. Haruma K, Kamada T, Kawaguchi H, et al. Effect of age and Helicobacter pylori infection on gastric acid secretion. J Gastroenterol Hepatol. 2000;15(3):277–283. doi:10.1046/j.1440-1746.2000.02131.x

28. Chiang TH, Chen YN, Chen YR, et al. Brand interchangeability of pepsinogen tests in the real-world setting after eradication of Helicobacter pylori: a community-based study. BMC Gastroenterol. 2022;22(1):69. doi:10.1186/s12876-022-02155-7

29. Leja M, Camargo MC, Polaka I, et al. Detection of gastric atrophy by circulating pepsinogens: a comparison of three assays. Helicobacter. 2017;22(4). doi:10.1111/hel.12393

30. Zhou YH, Zang JJ, Wu MJ, Xu JF, He J. Allowable Total Error and Limits for Erroneous Results (ATE/LER) zones for agreement measurement. J Clin Lab Anal. 2011;25(2):83–89. doi:10.1002/jcla.20437