Sage Journals: Discover world-class research

Abstract

Background:

An ongoing debate exists on the relationship between fibrinogen levels, urinary albumin excretion (UAE), and pancreatic ductal adenocarcinoma (PDAC), as demonstrated by clinical data of patients with benign and malignant pancreatic tumors.

Methods:

FinnGen and EMBL-EBI provided data on single-nucleotide polymorphisms associated with fibrinogen levels, UAE, and PDAC. To examine exposure-outcome relationships, weighted median, MR-Egger, and inverse variance weighted methods were employed. Heterogeneity and horizontal pleiotropy were evaluated using Cochran’s Q test and MR-Egger intercept analysis, respectively. The robustness of the Mendelian randomization (MR) analysis results was assessed using a leave-one-out sensitivity analysis. Clinical information, including sex, age, tumor location, pathological diagnosis, UAE level, fibrinogen level, and serum carbohydrate antigen 19-9 (CA19-9) level, was collected from patients with pancreatic malignant tumors at Peking Union Medical College Hospital from 2019 to 2024.

Results:

Fibrinogen levels and PDAC exhibited a strong positive causal association according to MR analysis (95% confidence interval [CI]: 1.630 to 8.753; P = .004). The UAE and PDAC exhibited a substantial beneficial causal association (95% CI: 0.055 to 3.831; P = .044) in the intercept analysis, and Cochran’s Q test did not show horizontal pleiotropy or heterogeneity (P > .05). Sensitivity analysis demonstrated the robustness of the findings. The inclusion of both fibrinogen and UAE levels increased the prediction rate for pancreatic malignancy by 6.36%. Genetic variations at SLC22A4 (rs12777) and FMO4 (rs1227104) may influence systemic inflammation, fibrinogen production, renal/metabolic homeostasis, and immunometabolic reprogramming, thereby shaping the tumor microenvironment and contributing to PDAC development.

Conclusion:

This study demonstrated the potential of fibrinogen levels and UAE as biomarkers of pancreatic malignancy through MR and clinical data validation. These metrics significantly improved the prediction of pancreatic cancer when used in combination with CA19-9 levels. In addition, this study proposes a mechanism for fibrinogen levels and UAE in PDAC involving SLC22A4 and FMO4.

Keywords

Fibrinogen level urinary albumin excretion pancreatic ductal adenocarcinoma pancreatic cancer Mendelian randomization

Introduction

Pancreatic ductal adenocarcinoma (PDAC) is a very aggressive disease with poor prognosis and is expected to become the second leading cause of cancer-related mortality by 2030.¹ Currently, early diagnosis of pancreatic cancer is difficult, and no effective treatment options exist for advanced pancreatic cancer. Most patients with PDAC do not develop symptoms until the disease progresses, and when symptoms appear, they are usually vague and nonspecific.^2,3 In addition, immunotherapy for PDAC has shown limited effectiveness.⁴

A popular serum biomarker for PDAC is the tumor marker sialyl Lewis A blood group antigen (sLea) carbohydrate antigen 19-9 (CA 19-9)⁵ which exhibits 80% to 90% sensitivity and specificity in symptomatic individuals.^6,7 However, 10% of patients do not express sLea,⁸ and thus, the CA 19-9 levels cannot be used for the diagnosis of this group of patients. In addition, biliary obstruction and benign pancreaticobiliary illness may also cause an increase in CA 19-9 levels.^9-11

Therefore, to assess the extent of disease more accurately before treatment, new potential diagnostic markers for PDAC should be explored.

In the present study, we analyzed blood samples from patients with PDAC using matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS).¹² We found that the plasma fibrinogen levels of patients with PDAC correlated with their overall survival (OS), and this correlation was statistically significant. In particular, the mean fibrinogen level was significantly higher in the PDAC group than in the control group (3.08 ± 0.565 vs 2.54 ± 0.249 log10 ng/mL, P < .001); moreover, the serum fibrinogen levels were significantly higher in patients with PDAC with distant metastasis than in those without distant metastasis (3.38 ± 0.581 vs 2.93 ± 0.499 log10 ng/mL, P = .002). The median OS was significantly longer in patients with PDAC with low fibrinogen levels (< 1000 ng/mL) than in those with high fibrinogen levels (⩾ 1000 ng/mL).^12,13 Therefore, we hypothesized that serum fibrinogen could be a potential biomarker of PDAC. Fibrinogen is an acute phase protein that regulates coagulation and fibrinolysis, with hyperfibrinogenemia being frequently associated with many malignancies.^14,15 Moreover, as an important coagulation factor, fibrinogen plays a key role in the inflammatory response and activation of the coagulation system in the body.^16,17 Preston et al reported that elevated fibrinogen production in patients with pancreatic cancer is associated with 2 main mechanisms: direct activation of the coagulation response for the production of procoagulant factors, including fibrinogen, coagulation factor (F) VIII, and tissue factor (TF) and indirect stimulation of procoagulant secretion from monocytes through inflammatory mediators, thereby upregulating fibrinogen levels.^18-21 In summary, we demonstrated that the serum fibrinogen level can be used as a potential biomarker for the diagnosis and prediction of PDAC prognosis. However, further large-scale studies are warranted.

Urine albumin excretion (UAE) is another potential PDAC biomarker. Abnormal UAE as an early marker of diabetic nephropathy is defined as a urinary albumin excretion rate (UAER) in the range of 20 to 200 μg/min (or 30-300 mg/24 h) or a urinary albumin/creatinine ratio (ACR) of 30 to 300 mg/g.²² At this stage, the lesions remain reversible, and timely intervention can prevent progression to clinical proteinuria.²² In the hyperglycemic state of diabetes, which is highly susceptible to diabetic nephropathy, the glomerular filtration membrane normally prevents the excretion of most albumin to the urine.^23,24 In the early stages of diabetic nephropathy, the glomerular filtration membrane charge and molecular barriers are impaired, leading to increased albumin filtration and impaired tubular reabsorption, resulting in microalbuminuria.²⁵ New-onset diabetes and insulin abnormalities are risk factors for PDAC.²⁶ Approximately 25% of patients with PDAC have diabetes and 40% have prediabetes, which is predominant in patients with new-onset diabetes mellitus.²⁷ Hence, new-onset diabetes may be associated with early diagnosis of pancreatic cancer, and consequently, detecting UAE abnormalities may help identify this group of patients. A large number of studies have shown that an increase in UAE is closely related to the onset and progression of many diseases. Some studies have shown that the higher the UAE, the higher the risk of developing systemic cancer, lung cancer, and mortality.²⁸

Herein, we demonstrated the clinical utility of serum fibrinogen and urinary albumin excretion (UAE) as indicators of PDAC using Mendelian randomization (MR). The MR uses genetic variations (eg, single-nucleotide polymorphisms [SNPs]) to assess the causal effect of exposure (eg, fibrinogen levels, UAE) on a disease (eg, PDAC).²⁹ The proposed method has several advantages. First, MR enhances the reliability of causal inferences by reducing the influence of confounding factors through the random assignment of genetic variants.³⁰ Second, MR effectively avoids the problem of reverse causation because genetic variants are determined before an individual is born.³¹ In addition, MR can reveal the directionality between exposure and outcome, providing evidence of true causality. As it is usually based on existing data from genome-wide association studies, MR is particularly suitable for factors that are difficult to study through randomized controlled trials. Based on the clinical information of patients with pancreatic cancer at the Peking Union Medical College Hospital from 2019 to 2024, we constructed a model for a more complete prediction of pancreatic cancer.

Methods

Mendelian randomization

Study design and source of data

A 2-sample MR study was conducted using data from a large-scale genome-wide association study (GWAS) of fibrinogen levels, UAE, and PDAC. The MR relies on 3 main assumptions³²: (1) the association assumption: SNPs are strongly associated with exposure factors; (2) the independence assumption: SNPs are independent of the confounding variables; and (3) the exclusivity assumption: the SNP does not act on the outcome variable through pathways other than the exposure factor. The design is illustrated in Figure 1.

Figure 1.

Mendelian randomization design for causal analysis of fibrinogen level, urinary albumin excretion (UAE), and pancreatic ductal adenocarcinoma (PDAC).

Data sources

Data on fibrinogen levels, UAE, and PDAC were obtained from the EMBL-EBI (https://www.ebi.ac.uk/) and FinnGen (https://www.finngen.fi/en) databases. As all data were obtained from freely accessible databases, no further ethical clearance was required. EMBL-EBI provided data on fibrinogen levels (data set GCST003194), which included information on 120 246 individuals of European ancestry. The UAE data were obtained from EMBL-EBI and included information on 382 500 individuals of European ancestry (data sets GCST006586, GCST002382, and GCST006463). EMBL-EBI also identified 20 SNPs that were strongly associated with fibrinogen levels and 74 SNPs that were strongly associated with UAE. Following the removal of confounding variables and linkage disequilibrium, 19 SNPs were associated with fibrinogen levels, whereas 26 were associated with UAE. In accordance with the Effect Allele Frequency (EAF) findings, duplications and mismatches were eliminated by coordinating the allelic orientation with respect to the exposure and outcome SNPs.FinnGen provided PDAC data, which included 314 193 individuals from Europe and 731 other cases (Data set No. finngen_R10_C3_PANCREAS_ADENO_DUCTAL_EXALLC).

Choosing genetic instrument variables

To verify Hypothesis 1, SNPs that were highly correlated with the exposure factors in the GWAS were first filtered using a cutoff value of P < 5 × 10⁻⁶. To prevent bias resulting from linkage disequilibrium, independent SNPs were filtered using cut-offs of R2 < 0.001 and kb = 10 000. R2 is given by the formula:

$R 2 = 2 \times [(1 - MAF) \times MAF \times β 2]$

where R2 is the overall explained variation in a few selected exposure-related instrumental variables (IVs); MAF is the influence of small allele frequency; and β is the predicted effect value of SNP.

Data analysis

The present study adhered to the STROBE-MR guidelines.³³ Two-sample MR analysis was conducted using the “TwoSampleMR (0.6.6)” program package for R 4.4.0. The fundamental techniques used for determining causation are weighted median, MR-Egger, and inverse variance weighting (IVW). Among these approaches, the primary analytical technique for achieving unbiased causal estimation without horizontal pleiotropy is IVW. The weighted median and MR-Egger are complementary techniques for MR analysis. The latter is less susceptible to measurement errors and outliers, whereas the former occasionally provides reliable causal inference even in the presence of pleiotropy.

The MR-pleiotropy residual sums and outliers approach (MRPRESSO) was used to rectify and evaluate the MR analyses following the removal of outlier SNPs (P < 1). MR-Egger intercept analysis was used to evaluate horizontal pleiotropy for meeting Hypothesis 3, with P ⩾ .05 denoting lack of horizontal pleiotropy. Heterogeneity was assessed using the Cochran’s Q test; with P ⩾ .05 indicating the lack of heterogeneity. A leave-one-out sensitivity assessment was conducted to assess the robustness of the data and to identify individual SNPs that substantially affected the overall outcomes.

Summary of Mendelian randomization workflow

In this study, 2-sample MR analysis was performed. First, significant genome-wide SNPs associated with fibrinogen levels and UAE were identified from published GWAS data sets. Independent instruments were selected after linkage disequilibrium clumping and harmonization of alleles between the exposure and outcome datasets. Second, causal estimates of the effects of fibrinogen and UAE on PDAC risk were obtained using IVW MR as the primary method, complemented by MR-Egger and weighted median approaches. Third, sensitivity analyses were conducted, including heterogeneity tests, pleiotropy assessments, and leave-one-out analyses, to evaluate robustness. Finally, significant loci were annotated to candidate genes and subjected to functional enrichment analyses to explore their biological mechanisms.

Pancreatic cancer diagnosis and prognosis evaluation model

Pancreatic cancer samples

This study was conducted according to the 1975 Declaration of Helsinki (revised in 2024).³⁴ The clinical data of patients with pancreatic cancer at Peking Union Medical College Hospital from 2019 to 2024 were collected. In addition, the clinical data of 150 patients with benign pancreatic tumors were collected. A total of 169 patients with malignant pancreatic tumors were included in the study, including 114 patients with PDAC, as confirmed by pathological diagnosis. Patients who lacked complete clinical data and those who did not sign the informed consent form or whose legal representatives did not provide consent were excluded. The completeness of the clinical information (sex, age, pathological diagnosis, etc) of patients included in the study was ensured to be consistent with the description of the same data in different systems for patients included in the study; moreover, the values were within the range of clinical logic or practical needs. In this study, data on sex, age, tumor location, pathological diagnosis, as well as UAE, fibrinogen, and serum CA19-9 levels were collected. This study was approved by the Ethics Committee of the Peking Union Medical College Hospital, Chinese Academy of Medical Sciences, under approval number I-24PJ1452. All participants or their legal representatives provided informed consent to participate in this study and agreed to its publication.

Statistical analysis

The statistical method of chi-square test was used for dichotomous variables (sex, UAE level), whereas the normality test, analysis of variance (ANOVA), independent samples Kruskal-Wallis test, and independent samples median test were performed for continuous variables (fibrinogen level, age). All tests were performed using the IBM SPSS Statistics version 27 (SPSS Inc, Chicago, Illinois). The accuracy, sensitivity, and specificity of UAE, fibrinogen, and CA19-9 levels in distinguishing patients with pancreatic cancer from those with benign tumors were evaluated using the receiver operating characteristic (ROC) curve analysis.

Gene Ontology and Kyoto Encyclopedia of Genes and Genomes pathway analysis, functional enrichment analysis, and protein-protein interaction

Protein-protein interactions were analyzed using STRING 12.0 (https://cn.string-db.org/), followed by the conduct of Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses with a screening condition of FDR ⩽ 0.05.

Results

We analyzed the causal effects between exposure factors (UAE and fibrinogen levels) and outcome variable (PDAC) using 2-sample MR. Forest plots of MR analyses are shown in Figure 2, whereas scatter plots of effect estimates for each SNP are shown in Supplemental Figure S1. For SNPs with significant effects, we revealed the potential mechanisms of fibrinogen levels and UAE as biomarkers of PDAC using GO and KEGG pathway analyses, functional enrichment analysis, and protein-protein interactions. We also proposed an assessment model for pancreatic cancer diagnosis and prediction based on MR analysis results using clinical samples.

Figure 2.

MR analysis of the causal relationship between fibrinogen level, UAE, and PDAC. (A) Forest plot of the estimated effect value of the causal relationship between fibrinogen level and PDAC, the estimated effect value of rs12777 had a more significant effect on the total estimated effect value. (B) Forest plot of the estimated effect value of the causal relationship between UAE and PDAC, the estimated effect value of rs12727104 had a more significant effect on the total estimated effect value.

Figure 3.

Statistical table of clinical data of patients with benign and malignant tumors of the pancreas and ROC curves of subjects with fibrinogen level, UAE, and CA19-9 in patients with benign and malignant tumors of the pancreas. (A) Point-box plot of fibrinogen levels in benign and malignant tumors of the pancreas. Fibrinogen levels in the group of malignant tumors of the pancreas were significantly higher than those in the group of benign tumors of the pancreas. (B) Dot-box plots of age statistics of patients with benign and malignant tumors of the pancreas. The pancreatic malignant tumors of the patients were significantly older than patients with benign tumors of the pancreas. (C) Working characteristic curve of subjects with CA19-9 in patients with benign and malignant tumors of the pancreas, with a specificity of 0.867, a sensitivity of 0.848, and an AUC of 0.861. (D) Working characteristic curves of subjects with CA19-9, UAE, and CA19-9 in patients with benign and malignant tumors of the pancreas. Working characterization curve with specificity of 0.967, sensitivity of 0.818, and AUC of 0.924.

Mendelian randomization

Mendelian randomization analysis results of fibrinogen levels and pancreatic ductal adenocarcinoma

In fibrinogen levels, all 3 analytical methods showed a positive correlation between fibrinogen levels and PDAC: IVW (95% confidence interval [CI]: 1.630 to 8.753; P = .004), MR-Egger (95% CI: −1.095 to 12.044; P = .125), and weighted median (95% CI: −0.777 to 7.230; P = 0.114); a significant positive correlation was shown only for IVW. We identified no horizontal pleiotropy when using intercept analysis (P = .920). Cochran’s Q test indicated no heterogeneity (P = .105). Sensitivity analysis revealed that the conclusions were robust. The 95% CIs for the combined effect values of MR-Egger were −1.095 to 12.044, whereas those for the combined effect values of IVW were 1.630 to 8.753, with the estimated effect values of each SNP falling within the −5 to 18 range (Figure 2A). However, rs12777 showed an estimated effect value of 20 to 20.5 (95% CI: 6 to 22), which was notably higher than the combined effect estimates (MR-Egger = 5.475; IVW = 5.191), suggesting this variant may be an outlier or have a disproportionate influence on the analysis. This speculation was confirmed via sensitivity analysis, which excluded each IV individually and then assessed the combined effect of the remaining variables. This facilitated the detection of oversensitivity to a particular variable. As shown in Supplemental Figure S3a, after removing rs12777, the combined effect values decreased to 4.3 to 4.4 (εMR Egger: 1.075-1.175; εIVW: 0.791-0.891). The 95% CIs of the sensitivity analysis (1.75-8.8) overlapped with the 95% CIs of the original analysis (MR-Egger and IVW), although the substantial change in point estimates confirmed that rs12777 was indeed influential in the original analysis. The estimated effect values for individual SNPs on the scatterplot (Supplemental Figure S1a) roughly fell above the baseline (y = 0), suggesting that elevated fibrinogen levels increase PDAC risk. Concomitantly, the MR-Egger and IVW regression intercept were not significantly different from zero, indicating no evidence of horizontal pleiotropy. The funnel plot in Supplemental Figure S2a shows the standard error (or a function thereof, such as the square root of the variance) in the horizontal axis and the effect size (eg, beta coefficient) in the vertical axis. The roughly symmetrical distribution of data points suggested no major publication bias, with most SNPs showing standard errors below 0.15, indicating reasonably precise effect estimates.

Mendelian randomization analysis results of urinary albumin excretion and pancreatic ductal adenocarcinoma

In UAE rate, all 3 analytical methods showed a positive correlation between UAE and PDAC: IVW (odd ratio [OR] = 6.978; 95% CI: 0.055 to 3.831; P = .044), MR-Egger (OR = 53.332; 95% CI: −3.638 to 11.591; P = .317), and weighted median (OR = 33.743; 95% CI = 0.854 to 6.184; P = .009) (Figure 2B). We identified no horizontal pleiotropy when using intercept analysis (P = .594). Cochran’s Q test indicated no heterogeneity (P = .542). Sensitivity analysis showed that the conclusions were robust. The 95% CIs for the combined effect values of MR-Egger were −.638 to 11.591, whereas those of IVW were 0.055 to 3.831, with the estimated effect values of each SNP falling within the −5 to 15 range. However, rs12727104 showed an estimated effect value of −8 to −10 (95% CI: −22 to 3), which was notably higher than the combined effect estimates (MR-Egger = 3.977; IVW = 1.943), suggesting this variant may be an outlier or have a disproportionate influence on the analysis. This speculation was confirmed via sensitivity analysis (Supplemental Figure S3b). After removing rs12727104, the combined effect values were 2.2 to 2.3 (εMR Egger: 1.677-1.777; εIVW: 0.257-0.357). The 95% CIs of the sensitivity analysis (0.15-3.85) overlapped with the 95% CIs of the original analysis (MR-Egger and IVW), indicating the reliability of the conclusion of this MR analysis. The estimated effect values for individual SNPs on the scatterplot (Supplemental Figure S1b) roughly fell above the baseline (y = 0), suggesting that an increase in UAE increases PDAC risk. In addition, the MR-Egger IVW regression intercept were not significant (P = .594), suggesting no directional pleiotropy. The funnel plot (Supplemental Figure S2b) shows the standard error (or a function thereof, eg, the square root of the variance) in the horizontal axis and the effect size (eg, the beta coefficient) in the vertical axis, with rs12727104 clearly identified as an influential outlier.

Gene Ontology and Kyoto Encyclopedia of Genes and Genomes pathway analysis, functional enrichment analysis, and protein-protein interaction

Effect of the SLC22A4 (rs12777) on pancreatic ductal adenocarcinoma

To explore the potential biological functions of SLC22A4, we performed GO, KEGG pathway, and functional enrichment analyses. The GO annotation consisted of 3 parts: biological processes, molecular functions, and cellular components (Figure 4). Biological process and molecular function analyses showed that SLC22A4 and related genes were enriched in biological processes related to drug transport, metabolic processes, and immune responses, and in molecular functions related to transporter activity, enzyme binding, and catalytic activity. In addition, SLC22A4 and related genes were closely associated with cellular components such as cell membranes, cytoplasm, and ribosomes (Figure 5). The KEGG pathway enrichment analyses indicated that SLC22A4 and related genes were involved in pathways critical to cancer biology. These pathways included ABC transporters, bile secretion, and antifolate resistance pathways (Figure 6B). SLC22A4 is associated with the ABC transporter, which contributes to chemoresistance in cancer cells. The involvement of the bile secretory pathway may reflect metabolic dysregulation in PDAC. The role of SLC22A4 in the antifolate resistance pathway highlights its potential role in the development of therapeutic resistance. Protein interaction network analysis revealed that SLC22A4 interacts with various transporters and immune-related proteins. Notably, SLC22A4 interacts with ABC transporters, such as ABCC1, ABCC3, and ABCC5, which have been implicated in drug resistance and metabolic processes. In addition, interactions with immune-related proteins such as RUNX1 and PTPN22 suggested that SLC22A4 plays a role in immune regulation of the tumor microenvironment (Figure 7B). Overall, SLC22A4 appears to play a multifaceted role in PDAC, participating in immune regulation, metabolic control, and drug resistance mechanisms. The interactions of SLC22A4 with key transporters and immune-associated proteins, as well as its involvement in key pathways, indicate SLC22A4 as a potential therapeutic target for PDAC. Further experimental validation is required to elucidate the exact mechanisms by which SLC22A4 affects tumor biology and therapeutic efficacy. In summary, our data support a plausible mechanism in which rs12777 alters SLC22A4 function or expression, leading to systemic inflammation and changes in immune cell behavior, which in turn increase fibrinogen levels and ultimately modify the PDAC microenvironment.

Figure 4.

The bubble plots showing GO pathway enrichment data for FMO4 and related genes. (A) Molecular function (gene ontology) enrichment. (B) Biological process (gene ontology) enrichment. (C) Cellular component (gene ontology) enrichment.

Figure 5.

The bubble plots showing GO pathway enrichment data for SLC22A4 and related genes. (A) Molecular function (gene ontology) enrichment. (B) Biological process (gene ontology) enrichment. (C) Cellular component (gene ontology) enrichment.

Figure 6.

The bubble plots showing KEGG pathway enrichment data. (A) The bubble plots showing KEGG pathway enrichment data for FMO4 and related genes. (B) The bubble plots showing KEGG pathway enrichment data for SLC22A4 and related genes.

Figure 7.

Protein-protein interaction network. (A) Protein-protein interaction network for FMO4. (B) Protein-protein interaction network for SLC22A4.

Effect of the FMO4 (rs12727104) on pancreatic ductal adenocarcinoma

Biological process and molecular function analyses revealed that FMO4 and its related genes were enriched in biological processes related to drug metabolism, detoxification, and oxidative stress and molecular functions related to oxidoreductase activity, flavin adenine dinucleotide (FAD) binding, and xenobiotic metabolism. In addition, FMO4 and related genes were closely associated with cellular components, such as the endoplasmic reticulum and cytoplasm (Figure 4). The KEGG pathway enrichment analysis revealed that FMO4 and its related genes were enriched in cytochrome P450, P450-to-xenobiotic metabolism, and chemocarcinogenesis (Figure 6A). The role of FMO4 in chemocarcinogenic pathways highlights its potential contribution to tumorigenesis. Protein interaction network analysis revealed that FMO4 interacts with various metabolic enzymes and transporters (Figure 7A). Notably, FMO4 interacts with cytochrome P450 enzymes (eg, CYP3A4, CYP2C9, and CYP2C19), which are implicated in drug metabolism and detoxification processes. In addition, interactions with transporters such as UGT1A1 and UGT1A6 suggested that FMO4 plays a role in metabolite conjugation and excretion. These results suggested that FMO4 plays a multifaceted role in PDAC by participating in metabolic detoxification, drug resistance, and the oxidative stress response. Its interactions with key metabolic enzymes and transporters as well as its involvement in key pathways indicate FMO4 as a potential therapeutic target for PDAC. However, further experimental validation is required to elucidate the exact mechanism by which FMO4 affects tumor biology and therapeutic efficacy. These findings suggested that rs1227104 may alter FMO4 activity or expression, leading to disturbed metabolic and oxidative homeostasis. This contributes to increased UAE and altered immune metabolism, thereby promoting changes in the PDAC microenvironment.

Pancreatic cancer diagnosis and prognosis evaluation model

Patient characteristics

Table 1 and Figure 3 present the demographic data of patients with benign and malignant pancreatic tumors. Patients with pancreatic malignant tumors were significantly older than those with benign pancreatic tumors (P < .001), while the number of male patients with pancreatic malignant tumors was higher than that of female patients (P < .001). Among the malignant pancreatic tumors, the pancreatic head was the most common tumor location, with PDAC being the most common malignant tumor in the pathological diagnosis of patients.

Table 1.

Demographics and tumor status of patients with pancreatic cancer.

Categorization	Malignant	Benign	P/X
Sex (X = 35.87; P < .001)
Male	97	38	X = 35.87; P < .001
Female	72	112
Age (P < .001)			P < .001
Mean ± SD (53.82 ± 15.28)
> 53	127	51
⩽ 53	42	99
Indicator
Urinary albumin excretion
T(X = 8.29, P = .004)	27 (16%)	9 (6%)	X = 8.29, P = .004
N	142 (84%)	141 (94%)
Fibrinogen levels(P < .001)	3.16 ± 0.90	2.56 ± 0.55	P < .001
CA19-9	180.65	15.67

Statistical data and tumor status of patients with benign and malignant pancreatic tumors. Patients with pancreatic malignant tumors were significantly older than those with benign pancreatic tumors; the number of male patients with pancreatic malignant tumors was higher than that of female patients; fibrinogen levels in the pancreatic malignant tumor group (3.16 ± 0.90) were significantly higher than those in the pancreatic benign tumor group (2.56 ± 0.55); the urine protein positivity rate of patients with malignant pancreatic tumors was significantly greater than that of patients with benign pancreatic tumors. The age group “<53” means “9-52,” whereas the age group “⩾ 53 (53-87)” means “53-87.”

Fibrinogen levels and urinary protein excretion rate in benign and malignant pancreatic tumors

The fibrinogen level in the pancreatic malignant tumor group (3.16 ± 0.90) was significantly higher than that in the pancreatic benign tumor group (2.56 ± 0.55, P < .001) (Figure 3A). The urine protein positivity rate in the pancreatic malignancy group (27%) was significantly higher than that in the pancreatic benign tumor group (9%, P < .001) (Table 1).

Receiver operating characteristic curve analyses in benign and malignant pancreatic tumors

The prediction model for pancreatic malignancy constructed from CA19-9 levels of patients with benign and malignant pancreatic cancer showed an area under the curve (AUC) of 0.861, specificity of 0.867, and sensitivity of 0.848, whereas the prediction model constructed from CA19-9-assisted fibrinogen levels and urinary protein excretion rate showed an AUC of 0.924, specificity of 0.967, and sensitivity of 0.818 (Figure 3C and D). The inclusion of both fibrinogen and UAE levels increased the prediction rate of pancreatic malignancy by 6.36%.

Discussion

Pancreatic cancer, especially PDAC, is a lethal malignant tumor of unknown etiology.³⁵ The worldwide PDAC incidence is predicted to reach 18.6 cases per 100 000 people by 2050, with a global average annual growth rate of 1.1%. Treating PDAC is a formidable challenge, with a mortality/morbidity rate of 98% and a 5-year OS of 2% to 9%. Cancer type, serum albumin level, tumor size, treatment choice, access to healthcare, and individual characteristics (eg, age, sex, health, and lifestyle) influence OS rates.³⁶ In conclusion, continued research on risk factors, early detection methods, and treatment strategies for PDAC is essential. Morbidity and mortality rates have remained relatively stable or have slightly increased.³⁷ In certain regions, PDAC may surpass breast cancer as the third most prevalent cause of death by 2025.³⁸ Therefore, conducting further studies on PDAC risk factors, early detection techniques, and treatment strategies is critical.

Elevated fibrinogen levels are an independent poor prognostic factor in various cancers, including lung, gastric, colorectal, liver, and breast cancers.^39-42 Elevated fibrinogen levels are associated with a poor prognosis in a wide range of cancers, and the specific role and mechanism of action of fibrinogen in PDAC deserve in-depth investigation. Fibrinogen creates an immunosuppressive microenvironment by recruiting immunosuppressive cells (eg, M2-type tumor-associated macrophages, myeloid-derived suppressor cells) and suppressing T-cell activity. Its degradation products (eg, fibrinopeptides) can directly activate proinflammatory factors such as transforming growth factor-beta (TGF-β) and interleukin 6 (IL-6), promote Th17/Treg imbalance, and weaken the antitumor immune response.^14,43,44

The MR analysis of fibrinogen levels in PDAC showed that the estimated effect size of SLC22A4 (rs12777) had a more significant effect on the overall combined effect size. Several SLCs have been shown to be involved in glucose uptake and lactate release, as well as promote the phagocytosis of apoptotic cells, and as cellular autophagy and apoptosis are critical processes in tumor immunotherapy responses, SLC-mediated metabolic regulation may influence therapeutic outcomes.⁴⁵ Moreover, the interaction of SLC22A4 with RUNX1 and PTPN22 suggests that SLC22A4 is involved in immune-related processes, potentially through regulating the tumor microenvironment. This regulation may affect immune cell infiltration and function, thereby influencing tumor monitoring and progression. The GO and KEGG analyses further confirmed the relevance of this gene in PDAC by linking it to hallmark pathways of cancer biology, such as cell proliferation, survival, and treatment tolerance pathways. By analyzing the GSE49641 data set, a previous study showed that the expression level of SLC22A4 in the peripheral blood of patients with PDAC was significantly higher than that in the normal group (P < .001).⁴⁶ Therefore, SLCs are potential targets for PDAC immunotherapy. Studies have shown that the tumor microenvironment of PDAC is highly heterogeneous, and PD-1+ cells and Foxp 3+ T-cells can be assessed for postoperative prognosis in patients with PDAC.^47,48 In addition, fibrinogen induces T-cell depletion and enhances the immune escape of tumor cells by synergizing with immune checkpoint molecules such as PD-L1.⁴⁹ Notably, some studies confirmed that SLC1A5, SLC7A5, and SLC3A2 are correlated with the expression of PD1 and PD-L1 and may be associated with the subtype of immune cell infiltration.⁵⁰ Based on these findings, elevated fibrinogen levels in patients with PDAC may affect immune cell infiltration and function via SLC22A4, thereby affecting the PDAC tumor microenvironment, weakening the antitumor immune response, and leading to a poor prognosis. However, the role of fibrinogen levels and the SLC family in PDAC pathogenesis, as well as the relationship between these genes and PDAC prognosis, remains unclear and needs to be verified by further experiments.

Each UAE doubling was associated with a 7% higher risk of overall cancer (hazard ratio [HR] = 1.06) and a 15% higher risk of urological cancer (HR = 1.14).²⁸ Elevated UAE levels have been associated with poor prognosis in a variety of cancers^51,52 including PDAC; however, the specific mechanisms deserve in-depth investigation. In the MR analysis examining the relationship between UAE and PDAC, the SNP rs12727104 (mapped to FMO4) showed a disproportionately large influence on the overall effect estimate. The FMO is a class of NADPH- and oxygen-dependent metabolizing enzymes, including 6 isoforms, FMO1-6, in humans, which are involved in the oxidative metabolism of drugs and toxicants.^53-55 In recent years, FMO has been found to play key roles in tumorigenesis, metastasis, and immune microenvironment regulation.^56,57 FMO4 promotes fatty acid oxidation (FAO) and induces M2 tumor-associated macrophage (TAM) polarization by activating the AMPK-ACC1 pathway in hepatocellular carcinoma.⁵⁸ By analyzing clinical samples from patients with hepatocellular carcinoma, Luo et al⁵⁸ found that high expression of FMO4 correlated with a decrease in CD8+ T-cell infiltration (P < .001), demonstrating that FMO4 affects the tumor microenvironment of hepatocellular carcinoma. The interaction of FMO4 with proteins such as CYP3A4 and UGT1A1 observed in the present study suggested that FMO4 is involved in oxidative stress. Moreover, GO and KEGG analyses confirmed the relevance of this gene in PDAC, as it was found to be associated with signature pathways in cancer biology (eg, pathways involved in drug resistance and chemical carcinogenesis). Therefore, we hypothesized that elevated UAE in patients with PDAC may alter metabolism and oxidative stress levels in vivo by affecting FMO4, which in turn affects the tumor microenvironment of PDAC, leading to poor prognosis. Currently, limited studies exist on the role of UAE and FMO in PDAC pathogenesis and the relationship between FMO4 and PDAC prognosis. The relevant pathways and molecular mechanisms have not yet been clarified.

Taken together, our MR and annotation results suggested concrete mechanistic hypotheses linking genetic variation to PDAC risk via systemic biomarkers. Specifically, SLC22A4 variations may influence the cellular uptake of redox-active substrates and immune cell function, increasing systemic inflammation and fibrinogen production; FMO4 variation may perturb xenobiotic/lipid metabolism and oxidative stress responses, contributing to renal endothelial/tubular dysfunction (higher UAE) and immunometabolic shifts in the tumor microenvironment. Our MR analyses provided evidence for causal directionality between exposure (fibrinogen/UAE) and PDAC but did not demonstrate the molecular steps—functional follow-up (eQTL/resident-tissue expression, CRISPR perturbation, metabolomics, and immune profiling) that are required to confirm whether the implicated SNPs alter gene expression or protein function in relevant tissues (liver, kidney, pancreas, and immune cells).

Based on these findings, we evaluated the clinical data of patients with benign and malignant pancreatic tumors for fibrinogen levels and urinary protein excretion rate to determine whether these are biomarkers for the prediction and diagnosis of pancreatic cancer and PDAC. The statistical results showed that patients with malignant pancreatic tumors had significantly higher fibrinogen levels and urinary protein excretion rates than those with benign pancreatic tumors. In addition, patients with elevated fibrinogen levels, urinary protein excretion rate, and CA19-9 levels had a worse prognosis than other patients.

Currently, the most common biomarker for detecting pancreatic malignancy and PDAC is CA19-9, which has a sensitivity, specificity, and AUC of 72% (95% CI: 71% to 73%), 86% (95% CI: 85% to 86%), and 0.8474 (95% CI: 0.8227 to 0.8676), respectively, for differentiating between benign and malignant pancreatic disease.⁵⁹ When CA19-9 was combined with fibrinogen levels and UAE, the sensitivity, specificity, and AUC for distinguishing benign and malignant pancreatic tumors were 0.818, 0.967, and 0.924, respectively. This combination improved the prediction of pancreatic malignant tumors by 6.36% compared with using CA19-9 alone. Moreover, the positive predictive value of CA19-9 for pancreatic malignant tumors was extremely low (< 1%).^60,61 Therefore, the development of novel biomarkers for pancreatic cancer or complementary biomarkers for CA19-9 is crucial. Our findings can assist in establishing evidence for the potential of fibrinogen levels and urinary protein excretion rates as diagnostic and prognostic biomarkers for pancreatic malignancy and PDAC, especially because these 2 metrics complement CA19-9 to greatly enhance the prediction of pancreatic malignancy.

Although this study validated the potential role of fibrinogen levels and UAE in pancreatic cancer through an MR analysis and clinical data, some limitations remain. First, the existing GWAS cohort for pancreatic cancer had a limited sample size, which might have affected the statistical power and reliability of the results. A population overlap may exist between exposure and outcome, leading to biased results. Second, horizontal pleiotropy was observed in MR analyses when using the validation data set. Despite the use of the MRPRESSO and MR-Egger methods to assess and correct horizontal pleiotropy, undetected horizontal pleiotropy may have still occurred. Third, MR analyses can only confirm the direction of the association and cannot provide quantitative causal effects. Additional quantitative studies are required to validate these associations. Fourth, the results were limited to a European population; therefore, the findings may not apply to other ethnic groups. Genetic backgrounds and environmental factors may lead to different results in different ethnic groups. Fifth, several statistical methods (eg, weighted median, MR-Egger, and IVW methods) were used, each with its own strengths and weaknesses, which may affect the interpretation of the results. Cochran’s Q test and MR-Egger intercept analysis were used to assess heterogeneity and horizontal multiplicity; however, the statistical power of the tests for these methods may be limited, especially in the case of small sample sizes. Sixth, clinical data were obtained from patients with pancreatic cancer at Peking Union Medical College Hospital from 2019 to 2024, which may introduce selection bias, and long-term follow-up data are lacking, which may affect the assessment of prognostic indicators. Seventh, the sample size was not calculated in accordance with the standards developed by Zhang and Hartmann, and the limited sample size may have affected the statistical significance of the results.

Beyond supporting a causal role in PDAC, our findings highlighted the potential translational applications of fibrinogen and UAE. Fibrinogen is routinely measured in clinical practice, and elevated levels may not only reflect systemic inflammation but also serve as a prognostic marker of tumor aggressiveness, response to therapy, and the risk of thromboembolic complications in patients with PDAC. The UAE, a marker of renal endothelial dysfunction and metabolic stress, could provide additional prognostic information by identifying subgroups of patients with higher systemic metabolic burden, and potentially worse outcomes. In combination with established markers such as CA19-9, both fibrinogen and UAE may contribute to a more refined risk stratification, early identification of high-risk individuals, and monitoring of treatment response. These findings suggest that integrating genetically informed biomarkers into clinical workflows could improve PDAC management and prognosis; however, further validation in prospective clinical cohorts is required.

Conclusion

This study demonstrated the potential of fibrinogen level and UAE as biomarkers of pancreatic malignancy through MR analysis and clinical data validation. These metrics significantly improved the prediction of pancreatic cancer when used in combination with CA19-9 levels. In addition, this study proposes a potential mechanism for fibrinogen levels and urinary protein excretion rate in PDAC involving SLC22A4 and FMO4.

Supplemental Material

sj-tif-1-onc-10.1177_11795549251392490 – Supplemental material for Proposing an Optimized Prediction Model for Pancreatic Cancer Based on Mendelian Randomization Analysis of Fibrinogen Levels and Urinary Protein Excretion Rate

Supplemental material, sj-tif-1-onc-10.1177_11795549251392490 for Proposing an Optimized Prediction Model for Pancreatic Cancer Based on Mendelian Randomization Analysis of Fibrinogen Levels and Urinary Protein Excretion Rate by Qian Wu, Mengqing Sun, Yan Lin, Shujin Lin, Ting Lin, Xiao Han and Xianlin Han in Clinical Medicine Insights: Oncology

Supplemental Material

sj-tif-2-onc-10.1177_11795549251392490 – Supplemental material for Proposing an Optimized Prediction Model for Pancreatic Cancer Based on Mendelian Randomization Analysis of Fibrinogen Levels and Urinary Protein Excretion Rate

Supplemental material, sj-tif-2-onc-10.1177_11795549251392490 for Proposing an Optimized Prediction Model for Pancreatic Cancer Based on Mendelian Randomization Analysis of Fibrinogen Levels and Urinary Protein Excretion Rate by Qian Wu, Mengqing Sun, Yan Lin, Shujin Lin, Ting Lin, Xiao Han and Xianlin Han in Clinical Medicine Insights: Oncology

Supplemental Material

sj-tif-3-onc-10.1177_11795549251392490 – Supplemental material for Proposing an Optimized Prediction Model for Pancreatic Cancer Based on Mendelian Randomization Analysis of Fibrinogen Levels and Urinary Protein Excretion Rate

Supplemental material, sj-tif-3-onc-10.1177_11795549251392490 for Proposing an Optimized Prediction Model for Pancreatic Cancer Based on Mendelian Randomization Analysis of Fibrinogen Levels and Urinary Protein Excretion Rate by Qian Wu, Mengqing Sun, Yan Lin, Shujin Lin, Ting Lin, Xiao Han and Xianlin Han in Clinical Medicine Insights: Oncology

Footnotes

Acknowledgements

N/A.

ORCID iD

Xiao Han

Ethical considerations

This study was approved by the Ethics Committee of Peking Union Medical College Hospital (PUMCH), Chinese Academy of Medical Sciences, under approval number I-24PJ1452 on July 12, 2024. The study included information on deceased participants, which was used with the explicit consent of legally authorized representatives (LARs). All participants (including LARs of the deceased participants) or their legal representatives provided written informed consent to participate in the study and agreed to its publication.

Author contributions

Qian Wu, Mengqing Sun, Xiao Han, PhD, and Xianlin Han, MD, conceived and designed the work on this topic and contributed substantially to the acquisition, analysis, and interpretation of the data. Qian Wu and Mengqing Sun drafted the article. Yan Lin, Shujin Lin, and Ting Lin performed the data analysis and experiments. All listed authors contributed to the conception, design, collection, analysis, or interpretation of the data and contributed to the writing and intellectual content of the article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The study was supported by funds from the National High-Level Hospital Clinical Research Funding No. 2022-PUMCH-B-003; the Chinese Academy of Medical Sciences Innovation Fund for Medical Sciences (CAMS) Initiative for Innovative Medicine Grant No. 2021-I2M-1-022; and the Natural Science Foundation of Fujian Province (2023J01411).

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data availability statement

The data supporting this study are openly available from the EMBL-EBI (https://www.ebi.ac.uk/) and FinnGen databases ().

Supplemental material

Supplemental material for this article is available online.

References

Delle Cave

. Emerging therapeutic options in pancreatic cancer management. Int J Mol Sci. 2024;25:1929.

Shi

, et al. Prognostic significance of tissue factor in patients with pancreatic cancer: a systematic review protocol. BMJ Open. 2020;10:e037431.

O’Reilly

EM.

Therapeutic developments in pancreatic cancer. Nat Rev Gastroenterol Hepatol. 2024;21:7-24.

Hosein

Dougan

Aguirre

Maitra

Translational advances in pancreatic ductal adenocarcinoma therapy. Nat Cancer. 2022;3:272-286.

Zhang

Cui

Zheng

, et al. Single-cell RNA sequencing reveals the process of CA19-9 production and dynamics of the immune microenvironment between CA19-9 (+) and CA19-9 (-) PDAC. Chin Med J. 2024.

Lee

Teng

TZJ

Shelat

VG.

Carbohydrate antigen 19-9-tumor marker: past, present, and future. World J Gastrointest Surg. 2020;12:468-490.

Janga

LSN

Sambe

Yasir

, et al. Holistic understanding of the role of carbohydrate antigen 19-9 in pancreatic cancer screening, early diagnosis, and prognosis: a systematic review. Cureus. 2023;15:e44382.

Coppola

La Vaccara

Farolfi

, et al. Role of CA 19.9 in the management of resectable pancreatic cancer: state of the art and future perspectives. Biomedicines. 2022;10.

Cemal Kizilarslanoglu

Civelek

Kemal Kilic

Kerem

Gunaydin

Ulger

. Extremely high level of CA 19-9 in a patient with metastatic pancreatic cancer: second highest level of CA 19-9 in literature. J Buon. 2014;19:583-584.

10.

Kawamura

Honda

Takano

, et al. Prognostic role of carcinoembryonic antigen and carbohydrate antigen 19-9 in stage IV colorectal cancer. Anticancer Res. 2022;42:3921-3928.

11.

Jelski

Mroczko

Biochemical markers of colorectal cancer-present and future. Cancer Manag Res. 2020;12:4789-4797.

12.

Chung

Lee

, et al. Serum fibrinogen as a diagnostic and prognostic biomarker for pancreatic ductal adenocarcinoma. Pancreatology. 2020;20:1465-1471.

13.

Salmiheimo

Mustonen

Stenman

, et al. Systemic inflammatory response and elevated tumour markers predict worse survival in resectable pancreatic ductal adenocarcinoma. PLoS ONE. 2016;11:e0163064.

14.

Chen

, et al. Fibrinogen and tumors. Front Oncol. 2024;14:1393599.

15.

Luyendyk

Schoenecker

Flick

MJ.

The multifaceted role of fibrinogen in tissue injury and inflammation. Blood. 2019;133:511-520.

16.

Xie

Wan

Liu

Wang

Zhou

Combination of C-reactive protein and fibrinogen is useful for diagnosing periprosthetic joint infection in patients with inflammatory diseases. Chin Med J. 2022;135:1986-1992.

17.

Davalos

Akassoglou

Fibrinogen as a key regulator of inflammation in disease. Semin Immunopathol. 2012;34:43-62.

18.

Yao

Zhang

Xue

Serum fibrinogen levels are positively correlated with advanced tumor stage and poor survival in patients with gastric cancer undergoing gastrectomy: a large cohort retrospective study. BMC Cancer. 2016;16:480.

19.

Falanga

Marchetti

Cancer-associated thrombosis: enhanced awareness and pathophysiologic complexity. J Thromb Haemost. 2023;21:1397-1408.

20.

Riaz

Fuentes

Deng

, et al. Comparative effectiveness of anticoagulants in patients with cancer-associated thrombosis. JAMA Netw Open. 2023;6:e2325283.

21.

Ikezoe

Cancer-associated thrombosis and bleeding. Int J Hematol. 2024;119:493-494.

22.

Bennett

Aditya

BS.

An overview of diabetic nephropathy: epidemiology, pathophysiology and treatment. J Diabetes Nurs. 2015;18:61-67.

23.

Samsu

Diabetic nephropathy: challenges in pathogenesis, diagnosis, and treatment. Biomed Res Int. 2021;2021:1497449.

24.

Daehn

Duffield

JS.

The glomerular filtration barrier: a structural target for novel kidney therapies. Nat Rev Drug Discov. 2021;20:770-788.

25.

Zoja

Xinaris

Macconi

Diabetic nephropathy: novel molecular mechanisms and therapeutic targets. Front Pharmacol. 2020;11:586892.

26.

Mellenthin

Balaban

Dugic

Cullati

Risk factors for pancreatic cancer in patients with new-onset diabetes: a systematic review and meta-analysis. Cancers. 2022;14:4684.

27.

Yang

Wen

Guo

Chai

Sun

Cheng

Targeting early diagnosis and treatment of pancreatic cancer among the diabetic population: a comprehensive review of biomarker screening strategies. Diabetol Metab Syndr. 2025;17:176.

28.

Luo

Kieneker

van der Vegt

, et al. Urinary albumin excretion and cancer risk: the PREVEND cohort study. Nephrol Dial Transplant. 2023;38:2723-2732.

29.

Gao

Sang

Jia

, et al. Association between sleep traits and rheumatoid arthritis: a Mendelian randomization study. Front Public Health. 2022;10:940161.

30.

Julian

Boddy

Islam

, et al. A review of Mendelian randomization in amyotrophic lateral sclerosis. Brain. 2022;145:832-842.

31.

Ference

Holmes

Smith

GD.

Using Mendelian randomization to improve the design of randomized trials. Cold Spring Harb Perspect Med. 2021;11.

32.

Davies

Holmes

Davey Smith

Reading Mendelian randomisation studies: a guide, glossary, and checklist for clinicians. BMJ. 2018;362:k601.

33.

Burgess

Butterworth

Thompson

SG.

Mendelian randomization analysis with multiple genetic variants using summarized data. Genet Epidemiol. 2013;37:658-665.

34.

Parums

DV.

Editorial: the 2024 revision of the declaration of Helsinki and its continued role as a code of ethics to guide medical research. Med Sci Monit. 2024;30:e947428.

35.

Zhao

Yang

Wang

, et al. An overview of epigenetic methylation in pancreatic cancer progression. Front Oncol. 2022;12:854773.

36.

Latenstein

AEJ

van der Geest

LGM

Bonsing

, et al. Nationwide trends in incidence, treatment and survival of pancreatic ductal adenocarcinoma. Eur J Cancer. 2020.

37.

Afghani

Klein

AP.

Pancreatic adenocarcinoma: trends in epidemiology, risk factors, and outcomes. Hematol Oncol Clin North Am. 2022;36:879-895.

38.

Zhao

Chen

, et al. Pancreatic cancer: a review of epidemiology, trend, and risk factors. World J Gastroenterol. 2021;27:4298-4321.

39.

Zhang

Tan

Wang

Liu

The association between plasma fibrinogen levels and lung cancer: a meta-analysis. J Thorac Dis. 2019;11:4492-4500.

40.

Arigami

Uenosono

Matsushita

, et al. Combined fibrinogen concentration and neutrophil-lymphocyte ratio as a prognostic marker of gastric cancer. Oncol Lett. 2016;11:1537-1544.

41.

Mei

Zhao

Liu

Clinical and prognostic significance of preoperative plasma fibrinogen levels in patients with operable breast cancer. PLoS ONE. 2016;11:e0146233.

42.

Han

, et al. Prognostic value of combined preoperative fibrinogen and neutrophil-lymphocyte ratio in patients with hepatocellular carcinoma after liver transplantation. Oncotarget. 2017;8:4301-4312.

43.

Wolberg

AS.

Fibrinogen and fibrin: synthesis, structure, and function in health and disease. J Thromb Haemost. 2023;21:3005-3015.

44.

Zhong

Fibrinogen-to-albumin ratios are related to mortality in patients with secondary malignant neoplasm of bone and bone marrow. Heliyon. 2023;9:e18797.

45.

Morioka

Perry

JSA

Raymond

, et al. Efferocytosis induces a novel SLC program to promote glucose uptake and lactate release. Nature. 2018;563:714-718.

46.

Meng

Fang

Huang

Identification of solute carrier family genes related to the prognosis and tumor-infiltrating immune cells of pancreatic ductal adenocarcinoma. Ann Transl Med. 2022;10:57.

47.

Group Young Researchers In Inflammatory Carcinogenesis

Wandmacher

Mehdorn

Sebens

The heterogeneity of the tumor microenvironment as essential determinant of development, progression and therapy response of pancreatic cancer. Cancers. 2021;13:4932.

48.

Kiryu

Ito

Suka

, et al. Prognostic value of immune factors in the tumor microenvironment of patients with pancreatic ductal adenocarcinoma. BMC Cancer. 2021;21:1197.

49.

Zheng

Quan

Dai

Zhang

The prognostic value of combination of plasma fibrinogen and CA19-9 in non-distant metastatic breast cancer patients undergoing surgery. Cancer Manag Res. 2020;12:8875-8886.

50.

Ansari

Craze

Althobiti

, et al. Enhanced glutamine uptake influences composition of immune cell infiltrates in breast cancer. Br J Cancer. 2020;122:94-101.

51.

Luo

Kieneker

Yang

, et al. An increase in albuminuria is associated with a higher incidence of malignancies. Clin Kidney J. 2024;17:sfae009.

52.

Mok

Ballew

Sang

, et al. Albuminuria, kidney function, and cancer risk in the community. Am J Epidemiol. 2020;189:942-950.

53.

Gong

Hou

Zhou

, et al. FMO family may serve as novel marker and potential therapeutic target for the peritoneal metastasis in gastric cancer. Front Oncol. 2023;13:1144775.

54.

Phillips

Shephard

EA.

Flavin-containing monooxygenase 3 (FMO3): genetic variants and their consequences for drug metabolism and disease. Xenobiotica. 2020;50:19-33.

55.

Kong

Chen

, et al. Alcoholic fatty liver disease inhibited the co-expression of Fmo5 and PPARα to activate the NF-κB signaling pathway, thereby reducing liver injury via inducing gut microbiota disturbance. J Exp Clin Cancer Res. 2021;40:18.

56.

Sha

Jin

Budczies

Kluck

Stenzinger

Sinicrope

FA.

Tumor mutational burden as a predictive biomarker in solid tumors. Cancer Discov. 2020;10:1808-1825.

57.

Thorsson

Gibbs

Brown

, et al. The immune landscape of cancer. Immunity. 2019;51:411-412.

58.

Luo

Chen

, et al. FMO4 shapes immuno-metabolic reconfiguration in hepatocellular carcinoma. Clin Transl Med. 2022;12:e740.

59.

Zhao

Chen

Diagnostic value of serum carbohydrate antigen 19-9 in pancreatic cancer: a systematic review and meta-analysis. Eur J Gastroenterol Hepatol. 2022;34:891-904.

60.

Kim

Lee

Paik

Rhee

Choi

KW.

Clinical usefulness of carbohydrate antigen 19-9 as a screening test for pancreatic cancer in an asymptomatic population. J Gastroenterol Hepatol. 2004;19:182-186.

61.

Nakisa

Sempere

Chen

, et al. Tumor-associated carbohydrate antigen 19-9 (CA 19-9), a promising target for antibody-based detection, diagnosis, and immunotherapy of cancer. ChemMedChem. 2024;19:e202400491.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

2.17 MB

0.00 MB

1.90 MB

2.33 MB