Monitoring with plasma apolipoprotein A2-isoforms can predict developing new-onset steatotic liver disease caused by pancreatic exocrine insufficiency following pancreatectomy

Introduction

Development of steatotic liver disease (SLD) is one of the most important postoperative outcomes in patients undergoing pancreatectomy (1-17). Recently, instead of non-alcoholic fatty liver disease (NAFLD), SLD was chosen as an overarching term to encompass the various etiologies of steatosis because the terms “non-alcoholic” and “fatty” were considered stigmatizing (18). Although SLD is primarily linked to obesity and metabolic syndrome, early exposure to nutrient deficiency in developing countries reportedly leads to the development of malnutrition syndromes, such as kwashiorkor, which is associated with hepatomegaly and SLD as a consequence of profound protein deficiency, whereas anorexia nervosa, one of the most common forms of undernutrition in developed countries, is also associated with SLD (19).

New-onset SLD following pancreatectomy is presumably associated with malnutrition, pancreatic exocrine insufficiency (PEI), and increased response to pancreatic enzyme replacement therapy (5,6). However, the pathophysiology of post-operative SLD remains poorly understood. Moreover, an effective strategy for administering pancreatic enzymes is crucial for preventing postoperative SLD. However, no reliable and convenient indicators exist for predicting the development of post-pancreatectomy SLD, and the development of liquid biomarkers reflecting pancreatic exocrine function is urgently required. Currently, the most used indirect pancreatic exocrine function test is the fecal elastase test, because it is simple, noninvasive, and relatively inexpensive. Although the fecal elastase test is the most appropriate initial test, it must be performed on a semi-solid or solid stool specimen (20,21). It is difficult to perform a fecal elastase test in patients who previously underwent pancreatectomy because they often have diarrhea postoperatively. The collection of stool samples is not as versatile as blood tests, which is considered a physical burden to the patient. In addition, fecal elastase test is not currently approved as an in vitro diagnostic (IVD) in Japan; thus, increasing research has been conducted on methods for evaluating pancreatic exocrine function using blood biomarkers as an alternative to stool tests.

We recently identified unique alterations of apolipoprotein A2-isoforms (apoA2-i) as a promising biomarker in serum to detect patients with early stages of pancreatic cancer through different proteomic approaches (22-25). ApoA2-i identify patients with pancreatic cancer based on the evaluation of the organ microenvironment with susceptibility to developing pancreatic cancer (26). Circulating apoA2-i consist of three isoforms with homo-or heterodimers: heavy isoforms with C-terminal amino acids with ATQ/ATQ, light isoforms with AT/AT, and intermediate isoforms with ATQ/AT (23). Carboxypeptidases A secreted from the pancreatic parenchyma aberrantly cleave the C-terminal amino acids and those release leads to a reduction of apoA2-ATQ/AT plasma levels, the major intermediate isoform of apoA2-i, which is predominantly found in patients of pancreatic cancer (23,26,27). On the other hand, in the case of developing PEI, apoA2-AT/AT as a light isoform is significantly decreased compared to normal exocrine function (28-31).

In this study, we aimed to estimate whether plasma apoA2-AT/AT can be used as a predictive biomarker for post-pancreatectomy SLD and identify the possibility of its association with other risk factors for post-pancreatectomy SLD development. We present this article in accordance with the STROBE reporting checklist (available at https://hbsn.amegroups.com/article/view/10.21037/hbsn-2025-325/rc).

Methods

Study design

We conducted a retrospective cohort study at a tertiary care center to investigate the risk of new-onset SLD following pancreatectomy and estimate disease predictors. The study population comprised patients who underwent pancreatectomy between March 2021 and March 2024 at Nippon Medical School Hospital.

After receiving approval from the institutional review board, two independent investigators reviewed the medical records of all patients who underwent pancreatectomy within the database. The demographic and perioperative variables extracted from the database and chart review included age, sex, body mass index, history of tobacco use, alcohol intake, diagnosis, type of pancreatic surgery performed [distal pancreatectomy (DP) or pancreaticoduodenectomy (PD)], diagnostics, and laboratory data (including pancreatic function, SLD, PEI, lipid, and tumor marker panels). PEI was diagnosed clinically by the treating surgeons based on patient-reported symptoms and the need for pancreatic enzyme replacement therapy by pancrelipase; some patients received pre-onset administration of pancrelipase.

Patients

Patients who underwent pancreatectomy, including DP and PD, at Nippon Medical School Hospital between March 2021 and March 2024 were eligible for this study. This clinical study was approved by the Ethics Committees of Nippon Medical School (No. A-2020-044) and the Central Committee of Nippon Medical School (No. M-2022-073) (32). All the participants agreed to sample collection and provided written informed consent. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. Patients who had preoperative diagnosis of SLD, postoperative diagnosis of SLD caused by drug-induced liver injury, were lack of adequate imaging studies for review, lack of observation time more than 2 months after surgery were excluded.

Plasma sample collection and measurement of apoA2-i by enzyme-linked immunosorbent assay (ELISA) methods

Plasma samples were collected 1, 3, 6, and 12 months after pancreatectomy and at the time when SLD was observed. We measured the concentration of the apoA2 isoforms in the plasma samples, heavier homodimers (ATQ/ATQ) and lighter homodimers (AT/AT), using an ELISA kit for IVD (Toray APOA2-iTQ; Toray Inc., Tokyo, Japan) (26). The concentration of apoA2-ATQ/AT heterodimers was calculated as previously described (26,27).

Diagnosis of SLD

Non-contrast-enhanced computed tomography (CT) images were manually reviewed by one of the two independent investigators under the supervision of a radiologist. For each pre-operative study and all available post-operative non-contrast CT studies, investigators drew three separate approximately 2 cm circular regions of interest (ROI) in the right and left liver, while taking care to avoid the liver margin and large hepatic vessels in two-slice CT images (see Figure 1). The mean hepatic attenuation of each ROI was averaged to obtain the average hepatic attenuation. SLD was defined as an average hepatic attenuation of <50 Hounsfield units (33). This approach was selected over the liver-to-spleen ratio because a considerable number of participants underwent DP with splenectomy.

Figure 1 Case presentation of SLD after PD. Non-contrast enhanced CT images of before surgery (A), 2 months after surgery (B), 6 months after surgery (C). Mean CT values were 53.9, 5.6, and 38.3 HU, respectively. (D) The apoA2-AT/AT values 1, 3, and 6 months after surgery were 2.6, 5.8, and 16 μg/mL, respectively. The apoA2-ATQ/ATQ values 1, 3, and 6 months after surgery were 117.2, 259.2, and 214 μg/mL, respectively. This patient received pre-onset administration of pancrelipase. After developing new-onset SLD, it improved with increased administration of pancrelipase. −, negative; +, positive. ApoA2, apolipoprotein A2; CT, computed tomography; HU, Hounsfield units; PD, pancreaticoduodenectomy; SLD, steatotic liver disease.

Outcome measures

The primary outcome measures included the incidence of new-onset SLD following pancreatectomy, identification of perioperative risk factors for new-onset SLD, and time to SLD diagnosis, defined as days from surgery to imaging diagnosis.

Statistical analyses

To evaluate the overall performance of apoA2-AT/AT, apoA2-ATQ/AT, and apoA2-ATQ/ATQ in classifying SLD, a receiver operating characteristic (ROC) curve analyses were performed, and the areas under the ROC curve (AUCs) were calculated. Furthermore, differences in AUCs between apoA2-AT/AT, each of apoA2-ATQ/ATQ, and apoA2-ATQ/AT were compared using Venkatraman’s test to evaluate differences in overall classification performance of each isoform.

Patients were divided into two groups based on their apoA2-AT/AT levels: high (≥9.58 µg/mL) and low (<9.58 µg/mL). This cutoff value was the median of the apoA2-AT/AT levels. For the baseline variables, summary statistics were calculated using frequencies and proportions for categorical data, and means and standard deviations (SDs) for continuous data. The Chi-squared test or Fisher’s exact test was employed to compare categorical variables, and Welch’s t-test or Wilcoxon rank-sum test was used to compare continuous variables, as appropriate. Univariate and multivariate logistic regression models were used to assess the relationship between high and low apoA2-AT/AT levels and the presence or absence of SLD. In the multivariable logistic regression model, two additional factors (i.e., pre-onset administration of pancrelipase and surgical method) were included to address differences between the apoA2-AT/AT groups. Survival analysis was conducted to explore the association between the time from surgery to SLD onset and the apoA2-AT/AT levels. A stratified log-rank test and Cox regression model, adjusted for the pre-onset administration of pancrelipase, were used in the survival analysis. The cumulative incidence function was estimated using the Kaplan-Meier method. Missing values in patient characteristics were presented. Since other analyses had no missing values, imputation was not performed. A two-sided P value of less than 0.05 was considered statistically. Statistical analyses were performed using R [version 4.4.2 (34); R Foundation for Statistical Computing, Vienna, Austria].

Results

Patient characteristics and ROC curve analysis of apoA2-i

Although the primary enrolled 87 patients underwent pancreatectomy performed at the Nippon Medical School Hospital between March 2021 and March 2024, eight patients were excluded owing to preoperative diagnosis of SLD (n=5), lack of adequate imaging studies for review (n=1), lack of observation time more than 2 months after surgery (n=1), and diagnosis of SLD caused by drug-induced liver injury (n=1) (Figure 2). The median follow-up time was 8.4 (interquartile range, 3.4–12.2) months.

Figure 2 Study participants selection flow chart. ApoA2, apolipoprotein A2; SLD, steatotic liver disease.

The minimum value of apoA2-AT/AT, apoA2-ATQ/AT, and the maximum value of apoA2-ATQ/ATQ after operation during the duration of follow-up between the SLD (n=32) and non-SLD group (n=47) were compared. The median value of apoA2-AT/AT in the SLD group was significantly lower than that in the non-SLD group (median 2.23 vs. 18.1 µg/mL, Wilcoxon rank sum test P<0.001) (Figure 3A). On the other hand, the median value of apoA2-ATQ/ATQ in the SLD group was significantly higher than that in the non-SLD group (median 181.9 vs. 154.4 µg/mL, Wilcoxon rank sum test P=0.02) (Figure 3B). The median value of apoA2-ATQ/AT in the SLD group was significantly lower than that in the non-SLD group (median 0.0 vs. 44.7 µg/mL, Wilcoxon rank sum test P<0.001) (Figure 3C). Significant differences in the apoA2-AT/AT, apoA2-ATQ/ATQ, and apoA2-ATQ/AT levels were observed between the development of SLD and non-SLD groups. Moreover, the ROC curve analysis revealed AUC of apoA2-AT/AT, apoA2-ATQ/ATQ, and apoA2-ATQ/AT to detect SLD were 0.843, 0.658, and 0.810, respectively (Figure 3D). In the ROC curve analysis, significant differences between apoA2-AT/AT and apoA2-ATQ/AT (P=0.02), apoA2-AT/AT and apoA2-ATQ/ATQ to detect SLD (P<0.001) were also observed. We determined that apoA2-AT/AT was statistically the most effective of the three isoforms for detecting SLD.

Figure 3 Comparison of apoA2-i values between the SLD group and non-SLD group and ROC curve analysis to detect SLD based on three apoA2-i levels. (A) ApoA2-AT/AT distribution. Upon comparing the minimum value of apoA2-AT/AT after operation during the duration of follow up between the SLD (n=32) and non-SLD (n=47) groups, the median values of both groups were 2.23 and 18.1 µg/mL (Wilcoxon rank sum test P<0.001). (B) ApoA2-ATQ/ATQ distribution. Comparing the maximum value of apoA2-ATQ/ATQ between the SLD and non-SLD groups, the median values of both groups were 181.9 and 154.4 µg/mL (Wilcoxon rank sum test P=0.02). (C) ApoA2-ATQ/AT distribution. Comparing the minimum value of apoA2-ATQ/AT between the SLD and non-SLD groups, the median values of both groups were 0.0 and 44.7 µg/mL (Wilcoxon rank sum test P<0.001). (D) ROC curve analysis to detect SLD based on the three apoA2-i levels. AUC of apoA2-AT/AT, apoA2-ATQ/ATQ, and apoA2-ATQ/AT to detect SLD were 0.843, 0.658, and 0.810, respectively. ApoA2, apolipoprotein A2; apoA2-i, apolipoprotein A2-isoforms; AUC, area under the ROC curve; ROC, receiver operating characteristic; SLD, steatotic liver disease.

Patients who underwent pancreatectomy were classified into two groups: high- and low-expression of apoA2-AT/AT. Because median value of apoA2-AT/AT in 79 patients was 9.58 µg/mL, the cut-off value of apoA2-AT/AT was defined according to 9.58 µg/mL of minimum concentration during follow-up after pancreatectomy, and then 79 patients were classified into 40 patients of the high expression group (≥9.58 µg/mL), and 39 patients of the low apoA2-AT/AT expression group (<9.58 µg/mL). The patient characteristics of the high- and low-apoA2-AT/AT expression groups are shown in Table 1. The statistical significance of age (P=0.02), T-stage (P=0.007), histological types (P<0.001), diagnostics (P=0.03), tumor locations (P<0.001), and surgical methods (P<0.001) was recognized between the high- and low-apoA2-AT/AT expression groups (P<0.05) (Table 1).

Differences of developing risk for post-operative SLD between high and low apoA2-AT/AT expression groups

The rate of SLD development in the low apoA2-AT/AT expression group was 72%, which was significantly higher than that in the high expression group (10%) (2×2 contingency table, Pearson’s Chi-squared test, P<0.001) (Table 2). ROC curve analysis to detect SLD by apoA2-AT/AT is shown in Figure 4A,4B. The AUC was 0.843. When the cut-off value was 9.58 µg/mL, the sensitivity and specificity to detect SLD were 87.5% and 76.6%, respectively (Figure 4A). On the other hand, the ROC curve analysis revealed that the best cut-off value calculated by Youden’s index was 6.15 µg/mL (Figure 4B). When the cut-off value was 6.15 µg/mL, the sensitivity and specificity were 81.3 % and 85.1%, respectively.

Figure 4 ROC curve analysis to detect SLD based on the apoA2-AT/AT levels. (A) AUC was 0.843. When the cut-off value was 9.58 µg/mL, the sensitivity and specificity to detect SLD were 87.5% and 76.6%, respectively. (B) On the other hand, the ROC curve analysis revealed that the best cut-off value calculated by Youden’s index was 6.15 µg/mL. When the cut-off value was 6.15 µg/mL, the sensitivity and specificity were 81.3 % and 85.1%, respectively. ApoA2, apolipoprotein A2; AUC, area under the ROC curve; CI, confidence interval; ROC, receiver operating characteristic; SLD, steatotic liver disease.

The cumulative incidence function estimate of the low-expression group was also significantly higher than that of the high-expression group (stratified log-rank P<0.001) (Figure 5). Although 61.4% of the patients in the low apoA2-AT/AT expression group developed SLD, only 11.0% in the high expression group developed within 1 year after surgery. Adjusted hazard ratio of developing SLD for comparison with the high expression group was 9.15 [95% confidence interval (CI): 3.13–26.77].

Figure 5 The cumulative incidence function estimate of developing SLD cumulative incidence function estimate (stratified log-rank P<0.001). ApoA2, apolipoprotein A2; SLD, steatotic liver disease.

Univariable and multivariable logistic regression analysis of the risk for developing SLD

As shown in Table 3, plasma apoA2-AT/AT levels <9.58 µg/mL and surgical methods were significant risk factors of new-onset SLD according to univariable logistic regression analysis. Plasma apoA2-AT/AT levels, pre-onset administration of pancrelipase, and surgical methods were significant independent risk factors in the multivariate analysis. Particularly, the odds ratios (ORs) of apoA2-AT/AT levels <9.58 µg/mL in the univariate and multivariate analysis were 20.1 (95% CI: 6.59–74.5) and 13.6 (95% CI: 2.68–134.2), respectively.

Discussion

This is the first study to show that low levels of apoA2-AT/AT in the plasma correlate with newly developed SLD in patients after pancreatectomy by measuring the plasma apoA2-i levels for assessing pancreatic exocrine function.

In clinical practice, the diagnostic approach to PEI can be used to evaluate the maldigestion of nutrients or specifically quantify exocrine pancreatic secretions. The fecal elastase-1 test is the most commonly employed indirect test for pancreatic exocrine function; however, its main limitation is that the measurement must be performed on solid stools (20). In contrast, liquid stool can be associated with false-positive results. The ¹³C-mixed triglyceride breath test is also an indirect test that monitors the digestion of an isotope-labeled fat meal, thus quantifying fat malabsorption; however, the test is relatively time-consuming, requires specific instruments and reagents, is only available in a few referral centers, and is not approved in Japan (6,20,21).

To overcome these problems, there is a high demand for blood biomarkers to predict the exocrine function of the pancreas. We determined that apoA2-AT/AT was statistically the most effective of the three isoforms for detecting SLD. ApoA2-AT/AT is significantly decreased in patients with PEI because carboxypeptidase A involved in cleaving C-terminal amino acids are depleted in patients with PEI (29,30). Therefore, potential patients who will develop SLD do not have the ability to secrete carboxypeptidase A according to developing PEI. In this case, even if apoA2-AT/AT decreases significantly in the bloodstream, there is no contradiction. Moreover, because pancrelipase is not absorbed into the blood, the plasma levels of apoA2-AT/AT are not affected by the presence or absence of pancrelipase (Figure 1) (35). In fact, apoA2-AT/AT concentration in the SLD group was significantly lower than that in the non-SLD group. In addition, we defined the cut-off value as 9.58 µg/mL between the high- and low-expression groups into median of the apoA2-AT/AT concentration in this cohort; the rate of SLD development in the low-expression group was significantly higher than that in the high-expression group. Moreover, the cumulative incidence function estimated by the Kaplan-Meier method demonstrated that the adjusted hazard ratio of developing SLD compared to the high-expression group was 9.15, indicating a significantly higher susceptibility to SLD Youden’s index was 6.15 µg/mL, it was very close to 9.58 µg/mL that was used in this study as median of this cohort. Multivariable logistic regression analysis also revealed that the apoA2-AT/AT expression levels remained an independent risk factor for the pre-onset administration of pancrelipase and surgical methods that are already known generally (16). These data suggest that apoA2-AT/AT is a potential blood biomarker for the prediction of post-operative SLD. However, even in the high apoA2-AT/AT expression group, few patients had SLD. All those patients underwent PD with superior mesenteric artery (SMA) plexus dissection. SMA plexus dissection in PD may also cause fat malabsorption, which may result in undernutrition and SLD due to severe diarrhea with or without PEI (2,5). Post-operative SLD is probably associated with malnutrition, similar to kwashiorkor or anorexia nervosa accompanied by SLD (19).

The present study had certain limitations. This was a single-center, retrospective study; therefore, it was an exploratory study. Hence, a large-scale multicenter study is warranted to demonstrate the reproducibility of the results. Furthermore, a prospective study to explore the required dose for preventing or treating SLD with pancrelipase indicated that apoA2-AT/AT monitoring should be carried out.

Conclusions

In conclusion, our findings suggest that postoperative low plasma apoA2-AT/AT levels are associated with a higher risk of new-onset SLD following pancreatectomy, which may result from PEI. Plasma apoA2-AT/AT levels are not only a novel non-invasive biomarker for pancreatic exocrine function, but also a predictive biomarker for developing post-pancreatectomy SLD.

Acknowledgments

We would like to thank Editage (https://www.editage.jp/) for English language editing.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://hbsn.amegroups.com/article/view/10.21037/hbsn-2025-325/rc

Data Sharing Statement: Available at https://hbsn.amegroups.com/article/view/10.21037/hbsn-2025-325/dss

Peer Review File: Available at https://hbsn.amegroups.com/article/view/10.21037/hbsn-2025-325/prf

Funding: This study was supported by a Grant-in-Aid for Scientific Research (C) from the Japan Society for the Promotion of Science (No. 23K08141 to A.M.), and grants from the Japan Agency for Medical Research and Development (AMED): AMED P-PROMOTE (No. 25ama221431h0002 to K.H.) and AMED Practical Research for Innovative Cancer Control (No. 25ck0106922h0002 to K.H.).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://hbsn.amegroups.com/article/view/10.21037/hbsn-2025-325/coif). K.N. received consulting fees from Toray Industries, Inc. K.H. received royalties or licenses and consulting fees from Toray Industries, Inc. and participated on a Data Safety Monitoring Board or Advisory Board of Toray Industries, Inc. The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was carried out in accordance with the Declaration of Helsinki and its subsequent amendments. This study was approved by the Ethics Committees of Nippon Medical School (No. A-2020-044) and the Central Committee of Nippon Medical School (No. M-2022-073). All the participants agreed to the sample collection and provided written informed consent.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Nomura R, Ishizaki Y, Suzuki K, et al. Development of hepatic steatosis after pancreatoduodenectomy. AJR Am J Roentgenol 2007;189:1484-8. [Crossref] [PubMed]
2. Kato H, Isaji S, Azumi Y, et al. Development of nonalcoholic fatty liver disease (NAFLD) and nonalcoholic steatohepatitis (NASH) after pancreaticoduodenectomy: proposal of a postoperative NAFLD scoring system. J Hepatobiliary Pancreat Sci 2010;17:296-304. [Crossref] [PubMed]
3. Tanaka N, Horiuchi A, Yokoyama T, et al. Clinical characteristics of de novo nonalcoholic fatty liver disease following pancreaticoduodenectomy. J Gastroenterol 2011;46:758-68. [Crossref] [PubMed]
4. Okamura Y, Sugimoto H, Yamada S, et al. Risk factors for hepatic steatosis after pancreatectomy: a retrospective observational cohort study of the importance of nutritional management. Pancreas 2012;41:1067-72. [Crossref] [PubMed]
5. Nagai M, Sho M, Satoi S, et al. Effects of pancrelipase on nonalcoholic fatty liver disease after pancreaticoduodenectomy. J Hepatobiliary Pancreat Sci 2014;21:186-92. [Crossref] [PubMed]
6. Nakagawa N, Murakami Y, Uemura K, et al. Nonalcoholic fatty liver disease after pancreatoduodenectomy is closely associated with postoperative pancreatic exocrine insufficiency. J Surg Oncol 2014;110:720-6. [Crossref] [PubMed]
7. Sato R, Kishiwada M, Kuriyama N, et al. Paradoxical impact of the remnant pancreatic volume and infectious complications on the development of nonalcoholic fatty liver disease after pancreaticoduodenectomy. J Hepatobiliary Pancreat Sci 2014;21:562-72. [Crossref] [PubMed]
8. Nagaya T, Tanaka N, Kimura T, et al. Mechanism of the development of nonalcoholic steatohepatitis after pancreaticoduodenectomy. BBA Clin 2015;3:168-74. [Crossref] [PubMed]
9. Ohgi K, Okamura Y, Yamamoto Y, et al. Perioperative Computed Tomography Assessments of the Pancreas Predict Nonalcoholic Fatty Liver Disease After Pancreaticoduodenectomy. Medicine (Baltimore) 2016;95:e2535. [Crossref] [PubMed]
10. Sato T, Matsuo Y, Shiga K, et al. Factors that predict the occurrence of and recovery from non-alcoholic fatty liver disease after pancreatoduodenectomy. Surgery 2016;160:318-30. [Crossref] [PubMed]
11. Nakamura M, Nakata K, Matsumoto H, et al. Acyl/free carnitine ratio is a risk factor for hepatic steatosis after pancreatoduodenectomy and total pancreatectomy. Pancreatology 2017;17:135-8. [Crossref] [PubMed]
12. Okabe H, Yamashita YI, Inoue R, et al. Postoperative nonalcoholic fatty liver disease is correlated with malnutrition leading to an unpreferable clinical course for pancreatic cancer patients undergoing pancreaticoduodenectomy. Surg Today 2020;50:193-9. [Crossref] [PubMed]
13. Beger HG, Mayer B, Vasilescu C, et al. Long-term Metabolic Morbidity and Steatohepatosis Following Standard Pancreatic Resections and Parenchyma-sparing, Local Extirpations for Benign Tumor: A Systematic Review and Meta-analysis. Ann Surg 2022;275:54-66. [Crossref] [PubMed]
14. Tsunematsu M, Gocho T, Yanagaki M, et al. The impact of postoperative exocrine index on non-alcoholic fatty liver disease following pancreaticoduodenectomy. Ann Gastroenterol Surg 2022;6:704-11. [Crossref] [PubMed]
15. Yamamura K, Yamashita YI, Yamao T, et al. Clinical impact of atrophic changes in remnant pancreas on the development of nonalcoholic fatty liver disease after pancreaticoduodenectomy. Ann Gastroenterol Surg 2022;6:555-61. [Crossref] [PubMed]
16. Patel V, Shah P, Ludwig DR, et al. Development of de novo nonalcoholic fatty liver disease following pancreatectomy. Medicine (Baltimore) 2023;102:e32782. [Crossref] [PubMed]
17. Izumi H, Yoshii H, Fujino R, et al. Factors contributing to nonalcoholic fatty liver disease (NAFLD) and fat deposition after pancreaticoduodenectomy: A retrospective analysis. Ann Gastroenterol Surg 2023;7:793-9. [Crossref] [PubMed]
18. Rinella ME, Lazarus JV, Ratziu V, et al. A multisociety Delphi consensus statement on new fatty liver disease nomenclature. J Hepatol 2023;79:1542-56. [Crossref] [PubMed]
19. Risi R, Tuccinardi D, Mariani S, et al. Liver disease in obesity and underweight: the two sides of the coin. A narrative review. Eat Weight Disord 2021;26:2097-107. [Crossref] [PubMed]
20. Whitcomb DC, Buchner AM, Forsmark CE. AGA Clinical Practice Update on the Epidemiology, Evaluation, and Management of Exocrine Pancreatic Insufficiency: Expert Review. Gastroenterology 2023;165:1292-301. [Crossref] [PubMed]
21. Capurso G, Traini M, Piciucchi M, et al. Exocrine pancreatic insufficiency: prevalence, diagnosis, and management. Clin Exp Gastroenterol 2019;12:129-39. [Crossref] [PubMed]
22. Honda K, Okusaka T, Felix K, et al. Altered plasma apolipoprotein modifications in patients with pancreatic cancer: protein characterization and multi-institutional validation. PLoS One 2012;7:e46908. [Crossref] [PubMed]
23. Honda K, Srivastava S. Potential usefulness of apolipoprotein A2 isoforms for screening and risk stratification of pancreatic cancer. Biomark Med 2016;10:1197-207. [Crossref] [PubMed]
24. Sato Y, Kobayashi T, Nishiumi S, et al. Prospective Study Using Plasma Apolipoprotein A2-Isoforms to Screen for High-Risk Status of Pancreatic Cancer. Cancers (Basel) 2020;12:2625. [Crossref] [PubMed]
25. Kashiro A, Kobayashi M, Oh T, et al. Clinical development of a blood biomarker using apolipoprotein-A2 isoforms for early detection of pancreatic cancer. J Gastroenterol 2024;59:263-78. [Crossref] [PubMed]
26. Honda K, Kobayashi M, Okusaka T, et al. Plasma biomarker for detection of early stage pancreatic cancer and risk factors for pancreatic malignancy using antibodies for apolipoprotein-AII isoforms. Sci Rep 2015;5:15921. [Crossref] [PubMed]
27. Honda K, Katzke VA, Hüsing A, et al. CA19-9 and apolipoprotein-A2 isoforms as detection markers for pancreatic cancer: a prospective evaluation. Int J Cancer 2019;144:1877-87. [Crossref] [PubMed]
28. Kato S, Honda K. Use of Biomarkers and Imaging for Early Detection of Pancreatic Cancer. Cancers (Basel) 2020;12:1965. [Crossref] [PubMed]
29. Kobayashi T, Sato Y, Nishiumi S, et al. Serum apolipoprotein A2 isoforms in autoimmune pancreatitis. Biochem Biophys Res Commun 2018;497:903-7. [Crossref] [PubMed]
30. Hayasaki A, Murata Y, Usui M, et al. Clinical Significance of Plasma Apolipoprotein-AII Isoforms as a Marker of Pancreatic Exocrine Disorder for Patients with Pancreatic Adenocarcinoma Undergoing Chemoradiotherapy, Paying Attention to Pancreatic Morphological Changes. Biomed Res Int 2019;2019:5738614. [Crossref] [PubMed]
31. Futagami S, Agawa S, Nakamura K, et al. Apolipoprotein A2 isoforms associated with exocrine pancreatic insufficiency in early chronic pancreatitis. J Gastroenterol Hepatol 2023;38:1949-57. [Crossref] [PubMed]
32. Otsuka T, Matsuyama K. Nippon Medical School's Ethical Review Processes for Studies Involving Human Subjects. J Nippon Med Sch 2024;91:136-9. [Crossref] [PubMed]
33. Starekova J, Hernando D, Pickhardt PJ, et al. Quantification of Liver Fat Content with CT and MRI: State of the Art. Radiology 2021;301:250-62. [Crossref] [PubMed]
34. Kanda Y. Investigation of the freely available easy-to-use software 'EZR' for medical statistics. Bone Marrow Transplant 2013;48:452-8. [Crossref] [PubMed]
35. Gewert K, Holowachuk SA, Rippe C, et al. The enzyme levels in blood are not affected by oral administration of a pancreatic enzyme preparation (Creon 10,000) in pancreas-insufficient pigs. Pancreas 2004;28:80-8. [Crossref] [PubMed]