Novel mRNA biomarker-based liquid biopsy for the detection of resectable pancreatic cancer

Novel mRNA biomarker-based liquid biopsy for the detection of resectable pancreatic cancer | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Novel mRNA biomarker-based liquid biopsy for the detection of resectable pancreatic cancer Jong-chan Lee, Sung Won Kang, Eun-Jin Sim, Jin-Sik Bae, Seong-mo Koo, and 10 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5119465/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 23 Apr, 2025 Read the published version in BMC Cancer → Version 1 posted 4 You are reading this latest preprint version Abstract Background Pancreatic ductal adenocarcinoma (PDAC) is one of the most lethal malignancies and most often diagnosed at an advanced stage. Identification of markers for the early diagnosis of PDAC is crucial. In this study, we aimed to identify novel mRNA biomarkers for diagnosing PDAC, focusing on early-stage tumorigenesis and associated immunological changes. Methods Blood samples and clinical information from 1,963 individuals were obtained from a single tertiary hospital between 2015 and 2021. Candidate mRNA biomarkers were identified through literature review, and their expression levels in buffy coat samples were measured using reverse-transcription quantitative polymerase chain reaction. Machine learning-based feature selection confirmed the final biomarker panel, which was tested using an independent dataset for diagnostic performance. Results In total, 1,504 individuals (417 patients with PDAC and 1,087 non-diseased controls) were eligible for the study. Among the 55 candidate biomarkers identified, 15 mRNAs ( CCL5, CCR5, CLEC7A, CXCL8, CXCR2, CXCR4, FOXP3, IFNA1, IFNL1, PTGES, PTGES2, PTGS2, SLC27A2, TNF , and VEGFA) were selected based on their diagnostic performance in distinguishing PDAC from control groups. The final model, HELP-15 (Human Early Liquid biopsy for PDAC), identified all PDAC stages (area under the curve [AUC] = 0.956) in the test set. For resectable pancreatic cancer (RPC), the AUC was 0.968, compared to 0.910 for carbohydrate antigen 19 − 9 (CA19-9). The combination of the panel and CA19-9 levels had an AUC of 0.985 in patients with RPC. For all PDAC stages in patients with normal CA19-9 levels, the AUC of the panel was 0.967, whereas CA19-9 alone or in combination with the panel had AUCs of 0.658 and 0.885, respectively. Conclusion Compared to CA19-9, the mRNA biomarker panel, HELP-15, significantly improved diagnostic performance in patients with RPC, particularly in those with normal CA19-9 levels. liquid biopsy mRNA buffy coat pancreatic cancer biomarker Figures Figure 1 Figure 2 Figure 3 Introduction Pancreatic ductal adenocarcinoma (PDAC) is one of the most lethal malignancies among all cancer types, with the lowest 5-year survival rate, and its incidence is projected to increase continuously [ 1 ]. The high mortality observed in PDAC results from the aggressive and metastatic progression of cancer, owing to the characteristics of the innate tumor microenvironment (TME) [ 2 ]. Although more than 87% of patients are diagnosed at an advanced stage, PDAC has high recurrence rates even after applicable treatments are performed [ 3 , 4 ]. Despite the evaluation of several treatment options in recent decades, surgical resection remains the only available curative option [ 5 ]. This highlights the importance of developing effective biomarkers for the detection of early-stage PDAC. Currently, the only FDA-approved clinical biomarker of PDAC is serum carbohydrate antigen 19 − 9 (CA19-9) [ 6 ]. Although CA19-9 is useful for determining treatment response, it is not suitable for early-stage diagnosis [ 7 , 8 ]. CA19-9 expression is genetically absent in Lewis antigen-negative individuals, comprising approximately 10% of the population [ 9 ]. Consequently, various biomarkers have been evaluated as alternatives to CA19-9 [ 10 – 13 ]. However, these alternative markers have neither demonstrated diagnostic performance superior to that of CA19-9 nor have they established their clinical reproducibility. PDAC is immunosuppressive, similar to other cancer types, in which tumor cells modify the surrounding tumor microenvironment (TME) to evade host immune surveillance. Although the exact mechanism has not yet been defined, studies have shown that this process is highly correlated with the formation of tumor-associated mesenchymal stem cells, granulocyte macrophage colony-stimulating factor, and polymorphonuclear myeloid-derived suppressor cells, which lead to alterations in immune cell responses and relevant signaling pathways [ 14 – 18 ]. We hypothesized that immune reprogramming by tumor cells influences immune system-related protein expression at the mRNA level during early tumorigenesis. Thus, detecting these biomolecular changes through liquid biopsy may facilitate early-stage PDAC diagnosis. In this study, we aimed to address the limitations of CA19-9 by focusing on its diagnostic performance for early-stage PDAC in patients with normal CA19-9 levels (< 37.0 U/mL). Specifically, we sought to identify novel mRNA biomarkers derived from the immune system to diagnose PDAC, emphasizing the early tumorigenesis associated with immune reprogramming. Material and methods Samples were obtained from the Human Bioresource Center of the Seoul National University Bundang Hospital (SNUBH) between September 2015 and December 2021. The study was reviewed and approved by the Institutional Review Board (IRB approval number: X-2011-651-903) and was conducted in accordance with the Declaration of Helsinki. Private patient information was anonymized and de-identified prior to analysis. Patient recruitment The clinical information of the patients and buffy coat samples were obtained from the electronic medical data and SNUBH biobank using the following inclusion criteria [ 1 ]. PDAC group: individuals who were histologically diagnosed with PDAC, including resectable pancreatic cancer (RPC), borderline RPC (BRPC), locally advanced pancreatic cancer (LAPC), and metastatic pancreatic cancer (MPC). Tumor stage was determined according to clinical staging. In this study, patients with PDAC were divided into two groups: RPC and advanced pancreatic cancer (APC), which included BRPC, LAPC, and MPC [ 2 ]. Non-disease control group: individuals with no prior diagnosis of malignancy. Additionally, all samples were divided into CA19-9 low (< 37.0 U/mL) and high (≥ 37.0 U/mL) groups, according to the patients’ medical records, to further analyze the efficacy of candidate markers against individuals with normal levels of CA19-9. Some patient data were excluded during medical record validation and reverse-transcription quantitative polymerase chain reaction (RT-qPCR) analysis. The exclusion criteria were as follows: (1) samples without complete medical records (e.g., CA19-9 value); (2) more than one month gap between the initial diagnosis date and the sample collection date; (3) samples collected after surgery or chemotherapy; and (4) samples of insufficient quality for RT-qPCR analysis (e.g., insufficient concentration, volume, or purity). Blood collection and RNA preparation Prior to buffy coat extraction, whole blood was collected in EDTA tubes. Whole blood samples were then centrifuged at 1,800 × g for 10 min at 4°C within 4 h of collection. Buffy coat samples were separated from the plasma and red blood cell layers and stored at − 80°C immediately after separation. Total RNA was isolated from buffy coat samples using a NucleoSpin RNA Blood kit (MACHEREY-NAGEL, Düren, Germany). For cDNA synthesis, 1 µg of total RNA was reverse-transcribed using the GoScript Reverse Transcription System (Promega, Madison, WI, USA). The cDNA product obtained was stored at − 80°C. RT-qPCR assay RT-qPCR was performed using a probe-based multiplex assay. Each probe was labeled at the 5′-end with a reporter dye (FAM or HEX). The primers and probes used in this study were obtained from Integrated DNA Technologies Inc. (Coralville, IA, USA). The mRNA expression levels were measured using GoTaq Probe qPCR Master Mix (Promega) in a 20 µL final volume. qPCR was performed using the QuantStudio 3 and QuantStudio 5 Real-Time PCR systems (Applied Biosystems, Foster City, CA, USA) under standard cycling conditions. The relative gene expression values were calculated by subtracting the GAPDH Ct value from the target gene Ct values, which were denoted as ΔCt: ΔCt = Ct target gene – Ct GAPDH Candidate marker selection A PRISMA-based literature search was conducted in the PubMed and MEDLINE databases using the following query: (pancreatic cancer OR pancreatic ductal adenocarcinoma) AND (liquid biopsy OR early detection OR biomarker) AND (immune OR immunological reprogramming) [ 19 ]. Through an extensive literature search and review, we screened, identified, and selected candidate markers that were highly correlated with immunological pathways during the development of PDAC from PanINs. The marker-selection scheme is illustrated in Fig. 1 . Feature and panel selection Feature selection was implemented to minimize the number of biomarkers and generate an optimal machine learning (M/L) model with the best diagnostic performance. The data of all eligible patients were split into training (64%), validation (16%), and test (20%) sets with controlled randomization, considering the demographic distribution of clinical cancer staging and CA19-9 values, to prevent skewed data ( Supplementary Figure S1 ). Using only the training and validation sets with dichotomous labeling of the PDAC and control groups, all possible combinations of markers were analyzed using the LR, XGB, LGBM, RF, and SVM algorithms. Fivefold cross-validation was conducted to avoid overfitting specific data subsets during the training phase. The performance was averaged over five training iterations. Twenty models per algorithm with the highest area under the curve (AUC) values were selected for further analysis, and 100 models were subjected to hyperparameter fine-tuning while being cross-paired with the remaining algorithms. Consequently, 500 models with optimal hyperparameters were subjected to performance evaluation. For performance evaluation, the sum of the training and validation sets was used to train the models, and an independent test set was used to analyze the performance metrics. The test set AUC of the 500 M/L models were compared with the AUC of CA19-9 alone in control vs. PDAC, control vs. RPC, control vs. PDAC (< 37.0 U/mL), and control vs. RPC (< 37.0 U/mL) groups. Machine learning-based feature selection Using ΔCt values as feature variables, candidate biomarkers were optimized to produce the best classification performance for distinguishing non-disease controls from the PDAC group. First, feature selection was performed by analyzing all possible biomarker combinations. Using only the training and validation datasets, five-fold cross-validation was conducted using five distinct algorithms suitable for classification: logistic regression (LR), random forest (RF), light gradient boost machine (LGBM), extreme gradient boost (XGB), and support vector machine (SVM). Second, the candidate models were subjected to fine-tuning of the hyperparameters, with all possible combinations unique to each algorithm. Various metrics were employed to evaluate the model performance, including sensitivity, specificity, Youden index, and AUC, based on the values of True Positive (TP), True Negative (TN), False Positive (FP), False Negative (FN), and receiver operating characteristic (ROC) curve. Sensitivity, specificity, and Youden index were calculated as follows: Sensitivity = TP / (TP + FN) Specificity = TN / (TN + FP) Youden index = Sensitivity + Specificity − 1 Python programming language (version 3.11.4) was used in the feature selection and modeling steps. The Pandas and NumPy libraries were used for data processing, and a sklearn library was used for machine-learning modeling and evaluation. Statistical analysis The mRNA expression levels (ΔCt) of potential candidate biomarkers were subjected to statistical analyses. Nonparametric tests (Mann–Whitney U and Kolmogorov–Smirnov tests) were performed to select primary biomarker candidates using GraphPad Prism (version 9.5.1; San Diego, CA, USA). Statistical significance was set at p < 0.05. Data availability The data generated in this study are available upon reasonable request from the corresponding author. Results Patient characteristics In total, 1,963 individuals were identified in this study. After the first exclusion, 1,504 subjects were eligible for the study and divided into two groups: the PDAC group (n = 417; 28%) and non-disease control group (n = 1, 087; 72%). The PDAC group was further divided into two groups according to clinical staging: RPC (n = 76; 5%) and APC (n = 341; 22%). The APC group included patients with BRPC/LAPC (n = 143; 9%) and MPC (n = 198; 13%). The median evaluated CA19-9 levels were significantly different between patients with PDAC and non-diseased controls (PDAC, 350.0 U/mL; control, 6.9 U/mL). The demographic data are summarized in Table 1 . Table 1 Patient characteristics PDAC Control Total RPC BRPC /LAPC MPC Number of patients 417 (28%) 76 (5%) 143 (9%) 198 (13%) 1087 (72%) 1504 (100%) Age 66 (37 – 88) 65 (47–88) 66 (38–86) 66 (37–86) 45 (20–76) 49 (20 – 88) Sex male 219 (53%) 48 (63%) 66 (46%) 105 (53%) 689 (63%) 908 (60%) female 198 (48%) 28 (37%) 77 (54%) 93 (47%) 398 (37%) 596 (40%) Baseline tumor marker CEA 3.3 (1 – 1790) 2.6 (1–142.9) 2.5 (1–118) 5.2 (1–1790) 1.5 (0.3–16) 1.7 (0.3 – 1790) CA19-9 350 (1.9 – 20001) 109.5 (1.9–7100) 207 (1.9–20001) 1435 (1.9–20001) 6.9 (0.6–84.1) 8.9 (0.6 – 20001) Data are presented as median (min–max) or n (%). PDAC, pancreatic ductal adenocarcinoma; RPC, resectable pancreatic cancer; BRPC, borderline resectable pancreatic cancer; LAPC, locally advanced pancreatic cancer; MPC, metastatic pancreatic cancer; CA 19 − 9, carbohydrate antigen 19 − 9; CEA, carcinoembryonic antigen. Candidate marker screening To search for immunologically relevant markers, a PRISMA-based literature search was conducted to identify 1,755 articles and screen 55 biomarkers ( Supplementary Table S1 ). Candidate markers were screened through a series of nonparametric tests to assess the significance of their expression between the PDAC and control groups. Markers with low expression levels (Ct GAPDH > 35) during RT-qPCR were excluded from the list of candidates. Consequently, 19 markers remained after screening ( Supplementary Figure S2 ). Feature and panel selection We identified 112 M/L models paired with LR, SVM, XGB, and LGBM, but not RF, which surpassed the performance of CA19-9 alone in all four comparisons. Among them, the SVM algorithm with 15 markers (referred to as HELP-15) was selected as the optimal model for classification, with an AUC value of > 0.950 in all comparisons and the highest AUC value of 0.967 in control vs. PDAC (< 37.0 U/mL; Figs. 2 and 3 ). The optimal model included CCL5, CCR5, CLEC7A, CXCL8, CXCR2, CXCR4, FOXP3, IFNA1, FNL1, PTGES, PTGES2, PTGS2, SLC27A2, TNF , and VEGFA as features ( Supplementary Table S2 ). Diagnostic performance discriminating PDAC and non-diseased control The AUC of HELP-15 was 0.956 for all PDAC patients, which was significantly higher than that of CA19-9 alone (AUC = 0.927) (Table 2 ). The sensitivity was 82.4%, which was higher than that of CA19-9 alone (76.5%), whereas the specificity was 93.6%, which was lower than that of CA19-9 alone (99.5%). Table 2 Genetic functions in biomarker panel Gene Description Reference sequence accession ID Function Mapping * CCL5 C-C Motif Chemokine Ligand 5 NM_002985 Chemokine related to immune cell migration and inflammation maintenance. C, D CCR5 C-C Motif Chemokine Receptor 5 NM_000579 Chemokine receptor that regulates inflammatory reaction. C CLEC7A C-Type Lectin Domain Containing 7A NM_022570 A gene that codes for Dectin-1, which is a pattern recognition receptor for innate immune cells. C CXCL8 C-X-C motif chemokine ligand 8 NM_000584 Chemokine that recruits immune cells to the tumor microenvironment. A, C, D CXCR2 C-X-C Motif Chemokine Receptor 2 NM_001557 Chemokine receptor for ligands such as CXCL8 that is expressed on the surface of immune cells, in general. C CXCR4 C-X-C Motif Chemokine Receptor 4 NM_003467 Chemokine receptor for ligands such as CXCL12 that is expressed on the surface of immune cells, in general. B, C FOXP3 Forkhead Box P3 NM_014009 Transcription factor essential for regulatory T-cell development and maintenance of immune tolerance. A IFNA1 Interferon Alpha 1 NM_024015 Cytokine that exhibits anti-tumor, and immunomodulatory functions including enhancement of natural killer cell cytotoxicity and macrophage activation. A IFNL1 Interferon Lambda 1 NM_172140 Cytokine that exhibits anti-tumor and immunomodulatory functions predominantly in the epithelial tissues. C PTGES Prostaglandin E synthase NM_004878 Isomerase that catalyzes the conversion of prostaglandin H2 (PGH2) into more stable prostaglandin E2 (PGE2) A PTGES2 Prostaglandin E synthase 2 NM_198938 Isomerase that catalyzes the conversion of PGH2 into more stable prostaglandin E2 (PGE2). A PTGS2 Prostaglandin-Endoperoxide Synthase 2 NM_000963 Also known as cyclooxygenase-2 (COX2) that converts arachidonic acid into PGH2. A, D SLC27A2 Solute carrier family 27 member 2 NM_003645 A gene that codes for fatty acid transport protein 2 (FATP2) that is crucial for lipid and fatty acid metabolism. A TNF Tumor Necrosis Factor NM_000594 Cytokine that regulates wide spectrum of biological processes including immune system signaling, apoptosis, and inflammation. A, D VEGFA Vascular Endothelial Growth Factor A NM_003376 Growth factor that promotes angiogenesis and vascular permeability which can be initiated by tumor cells. B, D * Mapping in the immune system reprogramming: A, immune suppression; B, M2 polarization; C, pro-tumoral activity; D, angiogenesis. (Refer to the Figure S3 ) For patients with RPC, the biomarker panel showed an AUC of 0.968, which was higher than that of CA19-9 (AUC = 0.910). The sensitivity was 93.8%, which was significantly higher than that of CA19-9 alone (62.5%), although the specificity (93.6%) was lower than that of CA19-9 alone (99.5%). HELP-15 showed great improvements over CA19-9 in distinguishing PDAC from the control group, especially in patients with RPC. Additional analysis of patients with normal Ca19-9 levels The general clinical cutoff value for the CA19-9 test was 37.0 U/mL. It is recommended that individuals with CA19-9 levels ≥ 37.0 U/mL are recommended to proceed with imaging techniques and continuous observation. In this study, 21% of patients with PDAC had normal CA19-9 levels (< 37.0 U/mL). Considering the genetic absence and delayed expression of CA19-9 in the population, patients with PDAC and normal CA19-9 levels are important targets for improving the overall PDAC diagnosis. The sensitivity and specificity of CA19-9 alone for patients with CA19-9 levels < 37.0 U/mL were not calculated because all data fell below the clinical standard. Only the AUC values were compared with those of the final model. When comparing patients between the control and PDAC groups with normal CA19-9 levels, the sensitivity, specificity, and AUC of HELP-15 were 88.9%, 93.5%, and 0.967, respectively (Table 3 ). The AUC of CA19-9 alone was 0.658, which was significantly lower than that of the final model. For patients with RPC and normal CA19-9 levels, the sensitivity, specificity, and AUC of the final model were 100.0%, 93.5%, and 0.980, respectively. Using the same test set, CA19-9 alone showed an AUC of 0.716, which was significantly lower than that of the final model. Table 3 Diagnostic performance of the marker panel Training set Test set Marker panel Sensitivity (95% CI) Specificity (95% CI) AUC (95% CI) p-value (Area = 0.5) Sensitivity (95% CI) Specificity (95% CI) AUC (95% CI) p-value (Area = 0.5) PDAC CA19-9 only 78.9 (78.8–79.0) 99.4 (99.4–99.4) 0.928 (0.914–0.943) < 0.0001 76.5 (76.0–77.0) 99.5 (99.5–99.6) 0.927 (0.897–0.956) < 0.0001 15 markers 91.3 (91.2–91.4) 95.6 (95.6–95.7) 0.980 (0.973–0.988) < 0.0001 82.4 (81.9–82.8) 93.6 (93.4–93.8) 0.956 (0.933–0.979) < 0.001 15 markers + CA19-9 98.2 (98.2–98.2) 98.8 (98.8–98.9) 0.999 (0.998-1.000) < 0.0001 92.9 (92.6–93.3) 92.2 (92.0-92.4) 0.971 (0.953–0.990) < 0.1 RPC CA19-9 only 70.0 (69.6–70.4) 99.4 (99.4–99.4) 0.917 (0.899–0.934) < 0.0001 62.5 (60.9–64.1) 99.5 (99.5–99.6) 0.910 (0.873–0.946) < 0.0001 15 markers 91.7 (91.4–91.9) 95.6 (95.6–95.7) 0.981 (0.972–0.990) < 0.0001 93.8 (93.0-94.5) 93.6 (93.4–93.8) 0.968 (0.946–0.991) < 0.0001 15 markers + CA19-9 98.3 (98.2–98.4) 98.8 (0.988–0.989) 0.999 (0.998-1.000) < 0.0001 93.8 (93.0-94.5) 92.2 (92.0-92.4) 0.985 (0.970-1.000) < 0.0001 PDAC (CA19-9 < 37.0 U/ml) CA19-9 only N/A N/A 0.660 (0.630–0.690) < 0.0001 N/A N/A 0.658 (0.597–0.719) < 0.0001 15 markers 95.7 (95.5–95.8) 95.6 (95.6–95.6) 0.987 (0.979–0.994) < 0.0001 88.9 (87.9–89.8) 93.5 (93.3–93.8) 0.967 (0.944–0.990) < 0.0001 15 markers + CA19-9 91.3 (91.1–91.5) 99.4 (99.4–99.4) 0.998 (0.995-1.000) < 0.0001 66.7 (65.2–68.1) 92.6 (92.4–92.9) 0.885 (0.844–0.926) < 0.0001 RPC (CA19-9 < 37.0 U/ml) CA19-9 only N/A N/A 0.727 (0.697–0.756) < 0.0001 N/A N/A 0.716 (0.657–0.775) < 0.0001 15 markers 94.4 (94.1–94.8) 95.6 (95.6–95.6) 0.991 (0.985–0.997) < 0.0001 100.0 (100.0-100.0) 93.5 (93.3–93.8) 0.980 (0.961–0.998) < 0.0001 15 markers + CA19-9 94.4 (94.1–94.8) 99.4 (0.994–0.994) 0.999 (0.997-1.000) < 0.0001 80.0 (77.6–82.4) 92.6 (92.4–92.9) 0.967 (0.943–0.990) < 0.0001 AUC, area under curve; PDAC, pancreatic ductal adenocarcinoma; RPC, resectable pancreatic cancer; CA 19 − 9, carbohydrate antigen 19 − 9. Discussion We developed an immune system-derived mRNA biomarker panel accompanied by an M/L algorithm that displayed effective diagnostic performance against PDAC. The panel included CCL5, CCR5, CLEC7A, CXCL8, CXCR2, CXCR4, FOXP3, IFNA1, IFNL1, PTGES, PTGES2, PTGS2, SLC27A2, TNF , and VEGFA , which were selected from 55 candidate biomarkers using a series of nonparametric tests, M/L-based feature selections, and performance evaluations. Based on these results, we constructed an optimal biomarker model that showed significant improvements in distinguishing between the RPC and PDAC groups with normal CA19-9 levels. In the last two decades, several studies have focused on combining multiple proteins, miRNAs, and cfDNAs to achieve meaningful PDAC diagnostic performance [ 10 – 13 , 20 – 22 ]. Klein et al. demonstrated a reasonable performance for various cancer types. However, the sensitivity for PDAC was measured at 83.7%, with a lower stage I sensitivity of 61.9% [ 23 ]. Lee et al. reported a sensitivity of 92.5% for detecting RPC; however, the sensitivity was 64.3% in the group with a normal range of CA19-9 when using triple protein marker panels, including LRG1, TTR, and CA19-9 [ 24 ]. Recently, Nakamura et al. showed excellent diagnostic performance with their transcriptomic signature, with an AUC of 0.930 and sensitivity in the early stages of PDAC (stages I and II) [ 25 ]. However, exosome-based and cell-free miRNAs require complicated analytical procedures that use unstable exosome-based miRNAs, making signatures less accessible to the general population [ 26 ]. Therefore, clinically feasible biomarkers are needed for PDAC detection. In this context, our biomarker panel demonstrated several notable strengths compared to those in previous studies. First, the final model consistently outperformed CA19-9 in diagnosing RPC (Fig. 2 c, 2 d), even in patients with normal CA19-9 levels (Fig. 3 ). Second, the use of RT-qPCR, a widely available and cost-effective analytical method, enhances the accessibility and affordability of the HELP-15 panel for the general population. Third, we utilized machine-learning models to optimize the diagnostic performance of the final model by leveraging extensive clinical data. By partitioning the dataset into training, validation, and test sets, we minimized the risk of overfitting and ensured robust performance. The efficacy of the biomarker panel was particularly promising for patients with RPC and normal CA19-9 levels. Although the CA19-9 test is limited by its delayed expression in early-stage PDAC and susceptibility to false-negative results, the HELP-15 panel, when used in conjunction with the CA19-9 test, shows significant potential as a screening tool for PDAC. Our study had some limitations. First, there was a disproportionate age distribution between individuals in the PDAC and control groups. To ensure that HELP-15 was not biased by age, we performed a correlation test between age and model probability, which indicated that the correlation was not significant ( Supplementary Figure S4 ). Second, some molecular biological mechanisms have not been fully elucidated. However, we implemented a structured representation method that included a comprehensive literature review and multifaceted comparison of the experimental data. Finally, the study was conducted at a single center. Although multicenter research has advantages, a homogeneous population might be more effective for identifying novel biomarkers. Further large-scale and multicenter clinical validation is necessary to address these limitations. In conclusion, compared with CA19-9, the immune system-derived mRNA biomarker panel, HELP-15, significantly improved the diagnostic performance in patients with RPC, particularly in those with PDAC and normal CA19-9 levels. Abbreviations APC advanced pancreatic cancer AUC area under the curve BRPC borderline resectable pancreatic cancer CA19-9 carbohydrate antigen 19 − 9 FN false negative FP false positive LAPC locally advanced pancreatic cancer LGBM light gradient boost machine LR logistic regression M/L machine learning MPC metastatic pancreatic cancer PDAC pancreatic ductal adenocarcinoma RF random forest RPC resectable pancreatic cancer SVM support vector machine TME tumor microenvironment TN true negative TP true positive XGB extreme gradient boost Declarations Ethics approval and consent to participate We obtained ethics approvals from the Human Bioresource Center of the Seoul National University Bundang Hospital (Affiliated College of Medicine, Seoul National University). The study was reviewed and approved by the Institutional Review Board (IRB approval number: X-2011-651-903) and was conducted in accordance with the Declaration of Helsinki. Private patient information was anonymized and de-identified prior to analysis. Consent for publication Not applicable. Availability of data and material The data that support the findings of this study are not openly available due to reasons of sensitivity and are available from the corresponding author upon reasonable request. Data are located in controlled access data storage at the Human Bioresource Center of Seoul National University Bundang Hospital. Competing interests Jin-Hyeok Hwang and Jihie Kim declare conflict of interest including advisory committee of HuVet bio Inc. Hyoung-Hwa Jeong is a CEO of HuVet bio Inc. Other authors have no COI to declare. Funding Not applicable. Authors' contributions Conceptualization - Jong-chan Lee, Hyoung Hwa Jeong, Jihie Kim, Jin-Hyeok Hwang Data curation - Eun-Jin Sim, Yuna Youn, Jin-Sik Bae, Seong-mo Koo, Mun-sub Byoun Formal analysis - Sung Won Kang, Eun-Jin Sim, Yuna Youn Funding acquisition - Jihie Kim, Jin-Hyeok Hwang Investigation - Sung Won Kang, Serin Kwon, Seoi Hong, Yunji Kim Methodology - Jong-chan Lee, Serin Kwon, Seoi Hong, Yunji Kim Project administration - Hyoung Hwa Jeong, Jihie Kim, Jin-Hyeok Hwang Resources - Eun-Jin Sim, Serin Kwon, Seoi Hong, Yunji Kim Software - Sung Won Kang, Serin Kwon, Seoi Hong, Yunji Kim Supervision - Kwangrok Jung, Jaihwan Kim, Jihie Kim, Jin-Hyeok Hwang Validation - Serin Kwon, Seoi Hong, Yunji Kim, Yuna Youn, Kwangrok Jung, Jaihwan Kim Visualization - Serin Kwon, Seoi Hong, Yunji Kim, Yuna Youn Writing – original draft - Jong-chan Lee, Sung Won Kang, Eun-Jin Sim, Serin Kwon, Seoi Hong, Yunji Kim, Kwangrok Jung, Jaihwan Kim Writing – review & editing - Jong-chan Lee, Sung Won Kang, Eun-Jin Sim, Serin Kwon, Seoi Hong, Yunji Kim, Kwangrok Jung, Jaihwan Kim, Hyoung Hwa Jeong, Jihie Kim, Jin-Hyeok Hwang Acknowledgements We thank Editage (www.editage.co.kr) for the English proofreading. References Siegel RL, Miller KD, Wagle NS, Jemal A, Cancer statistics. 2023. CA Cancer J Clin. 2023;73:17–48. https://doi.org/10.3322/caac.21763 Li J, Wientjes MG, Au JL-S. Pancreatic cancer: Pathobiology, treatment options, and drug delivery. AAPS J. 2010;12:223–32. https://doi.org/10.1208/s12248-010-9181-5 . Kim VM, Ahuja N. Early detection of pancreatic cancer. Chin J Cancer Res. 2015;27:321–31. https://doi.org/10.3978/j.issn.1000-9604.2015.07.03 . Jones RP, Psarelli E-E, Jackson R, Ghaneh P, Halloran CM, Palmer DH, et al. Patterns of recurrence after resection of pancreatic ductal adenocarcinoma: A secondary analysis of the ESPAC-4 randomized adjuvant chemotherapy trial. JAMA Surg. 2019;154:1038–48. https://doi.org/10.1001/jamasurg.2019.3337 . Grossberg AJ, Chu LC, Deig CR, Fishman EK, Hwang WL, Maitra A, et al. Multidisciplinary standards of care and recent progress in pancreatic ductal adenocarcinoma. CA Cancer J Clin. 2020;70:375–403. https://doi.org/10.3322/caac.21626 . McGuigan A, Kelly P, Turkington RC, Jones C, Coleman HG, McCain RS. Pancreatic cancer: A review of clinical diagnosis, epidemiology, treatment and outcomes. World J Gastroenterol. 2018;24:4846–61. https://doi.org/10.3748/wjg.v24.i43.4846 . Ryan DP, Hong TS, Bardeesy N. Pancreatic adenocarcinoma. N Engl J Med. 2014;371:1039–49. https://doi.org/10.1056/NEJMra1404198 . Liu J, Gao J, Du Y, Li Z, Ren Y, Gu J, et al. Combination of plasma microRNAs with serum CA19-9 for early detection of pancreatic cancer. Int J Cancer. 2012;131:683–91. https://doi.org/10.1002/ijc.26422 . Epub November 19, 2011. Guo M, Luo G, Lu R, Shi W, Cheng H, Lu Y, et al. Distribution of Lewis and secretor polymorphisms and corresponding CA19-9 antigen expression in a Chinese population. FEBS Open Bio. 2017;7:1660–71. https://doi.org/10.1002/2211-5463.12278. eCollection . Macdonald S, Mair F. Tackling cancers of unmet need: The pancreatic cancer pathway. Lancet Gastroenterol Hepatol. 2016;1:266–7. https://doi.org/10.1016/S2468-1253(16)30113-3 . Cohen JD, Li L, Wang Y, Thoburn C, Afsari B, Danilova L, et al. Detection and localization of surgically resectable cancers with a multi-analyte blood test. Science. 2018;359:926–30. https://doi.org/10.1126/science.aar3247 . Liu MC, Oxnard GR, Klein EA, Swanton C, Seiden MV, CCGA Consortium. Sensitive and specific multi-cancer detection and localization using methylation signatures in cell-free DNA. Ann Oncol. 2020;31:745–59. https://doi.org/10.1016/j.annonc.2020.02.011 . Killock D, Diagnosis. CancerSEEK and destroy – A blood test for early cancer detection. Nat Rev Clin Oncol. 2018;15:133. https://doi.org/10.1038/nrclinonc.2018.21 . Shi Y, Du L, Lin L, Wang Y. Tumour-associated mesenchymal stem/stromal cells: Emerging therapeutic targets. Nat Rev Drug Discov. 2017;16:35–52. https://doi.org/10.1038/nrd.2016.193 . Sai B, Dai Y, Fan S, Wang F, Wang L, Li Z, et al. Cancer-educated mesenchymal stem cells promote the survival of cancer cells at primary and distant metastatic sites via the expansion of bone marrow-derived-PMN-MDSCs. Cell Death Dis. 2019;10:941. https://doi.org/10.1038/s41419-019-2149-1 . Mathew E, Brannon AL, Del Vecchio AC, Garcia PE, Penny MK, Kane KT, et al. Mesenchymal stem cells promote pancreatic tumor growth by inducing alternative polarization of macrophages. Neoplasia. 2016;18:142–51. https://doi.org/10.1016/j.neo.2016.01.005 . Waghray M, Yalamanchili M, Dziubinski M, Zeinali M, Erkkinen M, Yang H, et al. GM-CSF mediates mesenchymal-epithelial cross-talk in pancreatic cancer. Cancer Discov. 2016;6:886–99. https://doi.org/10.1158/2159-8290.CD-15-0947 . Veglia F, Tyurin VA, Blasi M, De Leo A, Kossenkov AV, Donthireddy L, et al. Fatty acid transport Protein 2 reprograms neutrophils in cancer. Nature. 2019;569:73–8. https://doi.org/10.1038/s41586-019-1118-2 . Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. BMJ. 2021;372:n71. https://doi.org/10.1136/bmj.n71 . Wei J, Yang L, Wu Y-N, Xu J. Serum miR-1290 and miR-1246 as potential diagnostic biomarkers of human pancreatic cancer. J Cancer. 2020;11:1325–33. https://doi.org/10.7150/jca.38048 . Guler GD, Ning Y, Ku C-J, Phillips T, McCarthy E, Ellison CK, et al. Detection of early stage pancreatic cancer using 5-hydroxymethylcytosine signatures in circulating cell free DNA. Nat Commun. 2020;11:5270. https://doi.org/10.1038/s41467-020-18965-w . Borrebaeck CAK. Precision diagnostics: Moving towards protein biomarker signatures of clinical utility in cancer. Nat Rev Cancer. 2017;17:199–204. https://doi.org/10.1038/nrc.2016.153 . Klein EA, Richards D, Cohn A, Tummala M, Tummala M, Lapham R, Cosgrove D, et al. Clinical validation of a targeted methylation-based multi-cancer early detection test using an independent validation set. Ann Oncol. 2021;32:1167–77. https://doi.org/10.1016/j.annonc.2021.05.806 . Lee D-H, Yoon W, Lee A, Han Y, Byun Y, Kang JS, et al. Multi-biomarker panel prediction model for diagnosis of pancreatic cancer. J Hepatobiliary Pancreat Sci. 2023;30:122–32. https://doi.org/10.1002/jhbp.986 . Nakamura K, Zhu Z, Roy S, Jun E, Han H, Munoz RM, et al. An exosome-based transcriptomic signature for noninvasive, early detection of patients with pancreatic ductal adenocarcinoma: A multicenter cohort study. Gastroenterology. 2022;163:1252–66. .e2. Soria F, Droller MJ, Lotan Y, Gontero P, D’Andrea D, Gust KM, et al. An up-to-date catalog of available urinary biomarkers for the surveillance of non-muscle invasive bladder cancer. World J Urol. 2018;36:1981–95. https://doi.org/10.1007/s00345-018-2380-x . Additional Declarations Competing interest reported. Jin-Hyeok Hwang and Jihie Kim declare conflict of interest including advisory committee of HuVet bio Inc. Hyoung-Hwa Jeong is a CEO of HuVet bio Inc. Other authors have no COI to declare. Supplementary Files 05Supplementarydocumentsv2.docx 06supplementarytablev2.xlsx 07Suppfigure1.tif 07Suppfigure2.tif 07Suppfigure3.tif 07Suppfigure4.tif Cite Share Download PDF Status: Published Journal Publication published 23 Apr, 2025 Read the published version in BMC Cancer → Version 1 posted Editorial decision: Revision requested 03 Oct, 2024 Editor assigned by journal 01 Oct, 2024 Submission checks completed at journal 01 Oct, 2024 First submitted to journal 19 Sep, 2024 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-5119465","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":361737097,"identity":"236e5faa-2ea0-4d3f-a0ac-d62c6a1a71d1","order_by":0,"name":"Jong-chan Lee","email":"","orcid":"","institution":"Seoul National University Bundang Hospital","correspondingAuthor":false,"prefix":"","firstName":"Jong-chan","middleName":"","lastName":"Lee","suffix":""},{"id":361737100,"identity":"4f78c2e3-83b3-4f46-8554-e5b87e0b8231","order_by":1,"name":"Sung Won Kang","email":"","orcid":"","institution":"HuVet bio Inc","correspondingAuthor":false,"prefix":"","firstName":"Sung","middleName":"Won","lastName":"Kang","suffix":""},{"id":361737101,"identity":"f1b90d29-9f1e-479c-8264-4f5943f0411b","order_by":2,"name":"Eun-Jin Sim","email":"","orcid":"","institution":"HuVet bio Inc","correspondingAuthor":false,"prefix":"","firstName":"Eun-Jin","middleName":"","lastName":"Sim","suffix":""},{"id":361737102,"identity":"f1b20053-3fc1-4ff1-b04e-c7e71d6c36c8","order_by":3,"name":"Jin-Sik Bae","email":"","orcid":"","institution":"HuVet bio Inc","correspondingAuthor":false,"prefix":"","firstName":"Jin-Sik","middleName":"","lastName":"Bae","suffix":""},{"id":361737106,"identity":"c6d3996f-5f45-4c71-8c10-d784670287fd","order_by":4,"name":"Seong-mo Koo","email":"","orcid":"","institution":"HuVet bio Inc","correspondingAuthor":false,"prefix":"","firstName":"Seong-mo","middleName":"","lastName":"Koo","suffix":""},{"id":361737107,"identity":"6bedd04b-1f05-43a0-9923-ff6bac179254","order_by":5,"name":"Mun-sub Byoun","email":"","orcid":"","institution":"HuVet bio Inc","correspondingAuthor":false,"prefix":"","firstName":"Mun-sub","middleName":"","lastName":"Byoun","suffix":""},{"id":361737108,"identity":"64d3e352-3c77-48e2-b3bb-831a144dbf8e","order_by":6,"name":"Serin Kwon","email":"","orcid":"","institution":"Dongguk University","correspondingAuthor":false,"prefix":"","firstName":"Serin","middleName":"","lastName":"Kwon","suffix":""},{"id":361737110,"identity":"b5eb09c1-6cc0-4bcf-b145-492b90ecd006","order_by":7,"name":"Seoi Hong","email":"","orcid":"","institution":"Dongguk University","correspondingAuthor":false,"prefix":"","firstName":"Seoi","middleName":"","lastName":"Hong","suffix":""},{"id":361737111,"identity":"1949ae6b-1da9-4bb3-9d92-7d08e7a03441","order_by":8,"name":"Yunji Kim","email":"","orcid":"","institution":"Dongguk University","correspondingAuthor":false,"prefix":"","firstName":"Yunji","middleName":"","lastName":"Kim","suffix":""},{"id":361737113,"identity":"2dd48eb0-3aff-46f3-907d-c1de84734306","order_by":9,"name":"Yuna Youn","email":"","orcid":"","institution":"Seoul National University Bundang Hospital","correspondingAuthor":false,"prefix":"","firstName":"Yuna","middleName":"","lastName":"Youn","suffix":""},{"id":361737114,"identity":"2836a98f-d6ba-47ae-a697-d3be7c302e75","order_by":10,"name":"Kwangrok Jung","email":"","orcid":"","institution":"Seoul National University Bundang Hospital","correspondingAuthor":false,"prefix":"","firstName":"Kwangrok","middleName":"","lastName":"Jung","suffix":""},{"id":361737115,"identity":"751b7e9c-12bb-4e78-80f2-a91689f3f748","order_by":11,"name":"Jaihwan Kim","email":"","orcid":"","institution":"Seoul National University Bundang Hospital","correspondingAuthor":false,"prefix":"","firstName":"Jaihwan","middleName":"","lastName":"Kim","suffix":""},{"id":361737116,"identity":"596a468c-6970-4e6e-9042-1f49784df057","order_by":12,"name":"Hyoung Hwa Jeong","email":"","orcid":"","institution":"HuVet bio Inc","correspondingAuthor":false,"prefix":"","firstName":"Hyoung","middleName":"Hwa","lastName":"Jeong","suffix":""},{"id":361737117,"identity":"408c65b5-5e4a-4b9a-b78b-5a02bdf1a054","order_by":13,"name":"Jihie Kim","email":"","orcid":"","institution":"Dongguk University","correspondingAuthor":false,"prefix":"","firstName":"Jihie","middleName":"","lastName":"Kim","suffix":""},{"id":361737118,"identity":"6d396683-788b-4325-993f-61cca7dd591c","order_by":14,"name":"Jin-Hyeok Hwang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAyklEQVRIiWNgGAWjYFAC5gYwxQ/jSxDWwtgI1iPZQLIWgwPEajE43tj+4EPFHbvN588ek2CosWOQnH2AgJYzBxsbZ5x5lrztRl6aBMOxZAZpvgT8WsxuJDY287YdTja7wWMmwcB2gEGOh4DD4FqM+88AtfwjQYudAUOOmQRj2wEGaUJa7IF+mTnjzOEEiRs5xhaJfck8kj0EtEi2Nx/48KHisD1//xnDGx++2clJnCGgBQYSG0BkAgMDIWchOZBolaNgFIyCUTDyAAAchEMo/vsotQAAAABJRU5ErkJggg==","orcid":"","institution":"Seoul National University Bundang Hospital","correspondingAuthor":true,"prefix":"","firstName":"Jin-Hyeok","middleName":"","lastName":"Hwang","suffix":""}],"badges":[],"createdAt":"2024-09-19 22:10:19","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5119465/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5119465/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s12885-025-14124-w","type":"published","date":"2025-04-23T15:58:08+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":68716226,"identity":"0edf67d5-1708-4f74-8fec-bd21abe84fb6","added_by":"auto","created_at":"2024-11-11 09:56:21","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":210955,"visible":true,"origin":"","legend":"\u003cp\u003eSchematic representation of marker selection and model construction. A PRISMA-based literature search was conducted and 55 candidate markers were identified. Nineteen markers were identified using RT-qPCR and statistical analyses. Machine learning-based feature selection was conducted to generate optimal diagnostic models. Multi-marker models were evaluated in four comparisons based on sensitivity, specificity, and AUC. *n, the number of markers remaining after each process. AUC, area under the curve; RT-qPCR, reverse-transcription quantitative polymerase chain reaction.\u003c/p\u003e","description":"","filename":"03Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-5119465/v1/aad4a1c037b378d0a60d3ea0.png"},{"id":68717548,"identity":"356f1d14-9abd-404f-8fa5-5320bfe6cd4a","added_by":"auto","created_at":"2024-11-11 10:12:21","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":347779,"visible":true,"origin":"","legend":"\u003cp\u003eROC curve for training and test sets of CA19-9 alone, HELP-15, and HELP-15 with CA19-9. \u003cstrong\u003e(a)\u003c/strong\u003e Control vs. PDAC training set. \u003cstrong\u003e(b)\u003c/strong\u003e Control vs. PDAC test set. \u003cstrong\u003e(c)\u003c/strong\u003e Control vs. RPC training set. \u003cstrong\u003e(d)\u003c/strong\u003e Control vs. RPC test set. CA19-9, carbohydrate antigen 19-9; PDAC, pancreatic ductal adenocarcinoma; ROC, receiver operating characteristic; RPC, resectable pancreatic cancer.\u003c/p\u003e","description":"","filename":"03Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-5119465/v1/cbf75b31f5db7ff9472deb2c.png"},{"id":68716472,"identity":"b1ba99f8-ffbd-4905-86fc-be9ce4634a1f","added_by":"auto","created_at":"2024-11-11 10:04:21","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":397051,"visible":true,"origin":"","legend":"\u003cp\u003eAdditional analysis in patients with a normal range of CA19-9 (\u0026lt;37.0 U/mL). Receiver operating characteristic curve for the training and test sets of CA19-9 alone, HELP-15, and HELP-15 with CA19-9. \u003cstrong\u003e(a)\u003c/strong\u003e Control vs. PDAC training set. \u003cstrong\u003e(b)\u003c/strong\u003e Control vs. PDAC test set. \u003cstrong\u003e(c)\u003c/strong\u003e Control vs. RPC training set. \u003cstrong\u003e(d)\u003c/strong\u003e Control vs. RPC test set. CA19-9, carbohydrate antigen 19-9; PDAC, pancreatic ductal adenocarcinoma; RPC, resectable pancreatic cancer.\u003c/p\u003e","description":"","filename":"03Figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-5119465/v1/e175d421d4dc5f5598f5ab7d.png"},{"id":81569704,"identity":"9178a53b-06a5-4da6-862d-479763c16938","added_by":"auto","created_at":"2025-04-28 16:10:22","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2164530,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5119465/v1/6754840f-34c6-4dc6-8416-29683bd30013.pdf"},{"id":68716231,"identity":"c5cf1116-eb05-4a76-967b-c0f91931857e","added_by":"auto","created_at":"2024-11-11 09:56:21","extension":"docx","order_by":6,"title":"","display":"","copyAsset":false,"role":"supplement","size":23428,"visible":true,"origin":"","legend":"","description":"","filename":"05Supplementarydocumentsv2.docx","url":"https://assets-eu.researchsquare.com/files/rs-5119465/v1/0a51c6cbdd081095fd7a14a0.docx"},{"id":68716229,"identity":"ce9c678b-8e82-4e67-be3c-e010c93d02d9","added_by":"auto","created_at":"2024-11-11 09:56:21","extension":"xlsx","order_by":7,"title":"","display":"","copyAsset":false,"role":"supplement","size":19284,"visible":true,"origin":"","legend":"","description":"","filename":"06supplementarytablev2.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-5119465/v1/dba7f27b09e241072f247a90.xlsx"},{"id":68716230,"identity":"9fc68183-9542-43de-b254-e906888a782d","added_by":"auto","created_at":"2024-11-11 09:56:21","extension":"tif","order_by":8,"title":"","display":"","copyAsset":false,"role":"supplement","size":605996,"visible":true,"origin":"","legend":"","description":"","filename":"07Suppfigure1.tif","url":"https://assets-eu.researchsquare.com/files/rs-5119465/v1/ba347904f8df404f16348c51.tif"},{"id":68716474,"identity":"46e5aa93-b509-408d-aba1-de97a152bdcb","added_by":"auto","created_at":"2024-11-11 10:04:21","extension":"tif","order_by":9,"title":"","display":"","copyAsset":false,"role":"supplement","size":908470,"visible":true,"origin":"","legend":"","description":"","filename":"07Suppfigure2.tif","url":"https://assets-eu.researchsquare.com/files/rs-5119465/v1/03f85e90fc433f118945556d.tif"},{"id":68716475,"identity":"2ec9711a-d946-4391-83ac-cb192600cee2","added_by":"auto","created_at":"2024-11-11 10:04:21","extension":"tif","order_by":10,"title":"","display":"","copyAsset":false,"role":"supplement","size":3134266,"visible":true,"origin":"","legend":"","description":"","filename":"07Suppfigure3.tif","url":"https://assets-eu.researchsquare.com/files/rs-5119465/v1/3c2a42f3f1a2f9d66855827f.tif"},{"id":68716234,"identity":"22f2db7a-dbfc-4397-a65c-e5de353df28a","added_by":"auto","created_at":"2024-11-11 09:56:21","extension":"tif","order_by":11,"title":"","display":"","copyAsset":false,"role":"supplement","size":579556,"visible":true,"origin":"","legend":"","description":"","filename":"07Suppfigure4.tif","url":"https://assets-eu.researchsquare.com/files/rs-5119465/v1/62f77ae8b098903f3bb2e49c.tif"}],"financialInterests":"Competing interest reported. Jin-Hyeok Hwang and Jihie Kim declare conflict of interest including advisory committee of HuVet bio Inc. Hyoung-Hwa Jeong is a CEO of HuVet bio Inc. Other authors have no COI to declare.","formattedTitle":"Novel mRNA biomarker-based liquid biopsy for the detection of resectable pancreatic cancer","fulltext":[{"header":"Introduction","content":"\u003cp\u003ePancreatic ductal adenocarcinoma (PDAC) is one of the most lethal malignancies among all cancer types, with the lowest 5-year survival rate, and its incidence is projected to increase continuously [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. The high mortality observed in PDAC results from the aggressive and metastatic progression of cancer, owing to the characteristics of the innate tumor microenvironment (TME) [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. Although more than 87% of patients are diagnosed at an advanced stage, PDAC has high recurrence rates even after applicable treatments are performed [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. Despite the evaluation of several treatment options in recent decades, surgical resection remains the only available curative option [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. This highlights the importance of developing effective biomarkers for the detection of early-stage PDAC.\u003c/p\u003e \u003cp\u003eCurrently, the only FDA-approved clinical biomarker of PDAC is serum carbohydrate antigen 19\u0026thinsp;\u0026minus;\u0026thinsp;9 (CA19-9) [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. Although CA19-9 is useful for determining treatment response, it is not suitable for early-stage diagnosis [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. CA19-9 expression is genetically absent in Lewis antigen-negative individuals, comprising approximately 10% of the population [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. Consequently, various biomarkers have been evaluated as alternatives to CA19-9 [\u003cspan additionalcitationids=\"CR11 CR12\" citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. However, these alternative markers have neither demonstrated diagnostic performance superior to that of CA19-9 nor have they established their clinical reproducibility.\u003c/p\u003e \u003cp\u003ePDAC is immunosuppressive, similar to other cancer types, in which tumor cells modify the surrounding tumor microenvironment (TME) to evade host immune surveillance. Although the exact mechanism has not yet been defined, studies have shown that this process is highly correlated with the formation of tumor-associated mesenchymal stem cells, granulocyte macrophage colony-stimulating factor, and polymorphonuclear myeloid-derived suppressor cells, which lead to alterations in immune cell responses and relevant signaling pathways [\u003cspan additionalcitationids=\"CR15 CR16 CR17\" citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. We hypothesized that immune reprogramming by tumor cells influences immune system-related protein expression at the mRNA level during early tumorigenesis. Thus, detecting these biomolecular changes through liquid biopsy may facilitate early-stage PDAC diagnosis.\u003c/p\u003e \u003cp\u003eIn this study, we aimed to address the limitations of CA19-9 by focusing on its diagnostic performance for early-stage PDAC in patients with normal CA19-9 levels (\u0026lt;\u0026thinsp;37.0 U/mL). Specifically, we sought to identify novel mRNA biomarkers derived from the immune system to diagnose PDAC, emphasizing the early tumorigenesis associated with immune reprogramming.\u003c/p\u003e"},{"header":"Material and methods","content":"\u003cp\u003e Samples were obtained from the Human Bioresource Center of the Seoul National University Bundang Hospital (SNUBH) between September 2015 and December 2021. The study was reviewed and approved by the Institutional Review Board (IRB approval number: X-2011-651-903) and was conducted in accordance with the Declaration of Helsinki. Private patient information was anonymized and de-identified prior to analysis.\u003c/p\u003e \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003ePatient recruitment\u003c/h2\u003e \u003cp\u003eThe clinical information of the patients and buffy coat samples were obtained from the electronic medical data and SNUBH biobank using the following inclusion criteria [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. PDAC group: individuals who were histologically diagnosed with PDAC, including resectable pancreatic cancer (RPC), borderline RPC (BRPC), locally advanced pancreatic cancer (LAPC), and metastatic pancreatic cancer (MPC). Tumor stage was determined according to clinical staging. In this study, patients with PDAC were divided into two groups: RPC and advanced pancreatic cancer (APC), which included BRPC, LAPC, and MPC [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. Non-disease control group: individuals with no prior diagnosis of malignancy.\u003c/p\u003e \u003cp\u003eAdditionally, all samples were divided into CA19-9 low (\u0026lt;\u0026thinsp;37.0 U/mL) and high (\u0026ge;\u0026thinsp;37.0 U/mL) groups, according to the patients\u0026rsquo; medical records, to further analyze the efficacy of candidate markers against individuals with normal levels of CA19-9.\u003c/p\u003e \u003cp\u003eSome patient data were excluded during medical record validation and reverse-transcription quantitative polymerase chain reaction (RT-qPCR) analysis. The exclusion criteria were as follows: (1) samples without complete medical records (e.g., CA19-9 value); (2) more than one month gap between the initial diagnosis date and the sample collection date; (3) samples collected after surgery or chemotherapy; and (4) samples of insufficient quality for RT-qPCR analysis (e.g., insufficient concentration, volume, or purity).\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eBlood collection and RNA preparation\u003c/h3\u003e\n\u003cp\u003ePrior to buffy coat extraction, whole blood was collected in EDTA tubes. Whole blood samples were then centrifuged at 1,800 \u0026times; \u003cem\u003eg\u003c/em\u003e for 10 min at 4\u0026deg;C within 4 h of collection. Buffy coat samples were separated from the plasma and red blood cell layers and stored at \u0026minus;\u0026thinsp;80\u0026deg;C immediately after separation. Total RNA was isolated from buffy coat samples using a NucleoSpin RNA Blood kit (MACHEREY-NAGEL, D\u0026uuml;ren, Germany). For cDNA synthesis, 1 \u0026micro;g of total RNA was reverse-transcribed using the GoScript Reverse Transcription System (Promega, Madison, WI, USA). The cDNA product obtained was stored at \u0026minus;\u0026thinsp;80\u0026deg;C.\u003c/p\u003e\n\u003ch3\u003eRT-qPCR assay\u003c/h3\u003e\n\u003cp\u003eRT-qPCR was performed using a probe-based multiplex assay. Each probe was labeled at the 5\u0026prime;-end with a reporter dye (FAM or HEX). The primers and probes used in this study were obtained from Integrated DNA Technologies Inc. (Coralville, IA, USA). The mRNA expression levels were measured using GoTaq Probe qPCR Master Mix (Promega) in a 20 \u0026micro;L final volume. qPCR was performed using the QuantStudio 3 and QuantStudio 5 Real-Time PCR systems (Applied Biosystems, Foster City, CA, USA) under standard cycling conditions. The relative gene expression values were calculated by subtracting the \u003cem\u003eGAPDH\u003c/em\u003e Ct value from the target gene Ct values, which were denoted as ΔCt:\u003c/p\u003e \u003cp\u003eΔCt\u0026thinsp;=\u0026thinsp;Ct\u003csub\u003etarget gene\u003c/sub\u003e \u0026ndash; Ct\u003csub\u003e\u003cem\u003eGAPDH\u003c/em\u003e\u003c/sub\u003e\u003c/p\u003e\n\u003ch3\u003eCandidate marker selection\u003c/h3\u003e\n\u003cp\u003eA PRISMA-based literature search was conducted in the PubMed and MEDLINE databases using the following query: (pancreatic cancer OR pancreatic ductal adenocarcinoma) AND (liquid biopsy OR early detection OR biomarker) AND (immune OR immunological reprogramming) [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. Through an extensive literature search and review, we screened, identified, and selected candidate markers that were highly correlated with immunological pathways during the development of PDAC from PanINs. The marker-selection scheme is illustrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\n\u003ch3\u003eFeature and panel selection\u003c/h3\u003e\n\u003cp\u003eFeature selection was implemented to minimize the number of biomarkers and generate an optimal machine learning (M/L) model with the best diagnostic performance. The data of all eligible patients were split into training (64%), validation (16%), and test (20%) sets with controlled randomization, considering the demographic distribution of clinical cancer staging and CA19-9 values, to prevent skewed data (\u003cb\u003eSupplementary Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e\u003c/b\u003e). Using only the training and validation sets with dichotomous labeling of the PDAC and control groups, all possible combinations of markers were analyzed using the LR, XGB, LGBM, RF, and SVM algorithms. Fivefold cross-validation was conducted to avoid overfitting specific data subsets during the training phase. The performance was averaged over five training iterations. Twenty models per algorithm with the highest area under the curve (AUC) values were selected for further analysis, and 100 models were subjected to hyperparameter fine-tuning while being cross-paired with the remaining algorithms. Consequently, 500 models with optimal hyperparameters were subjected to performance evaluation.\u003c/p\u003e \u003cp\u003eFor performance evaluation, the sum of the training and validation sets was used to train the models, and an independent test set was used to analyze the performance metrics. The test set AUC of the 500 M/L models were compared with the AUC of CA19-9 alone in control vs. PDAC, control vs. RPC, control vs. PDAC (\u0026lt;\u0026thinsp;37.0 U/mL), and control vs. RPC (\u0026lt;\u0026thinsp;37.0 U/mL) groups.\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eMachine learning-based feature selection\u003c/h2\u003e \u003cp\u003eUsing ΔCt values as feature variables, candidate biomarkers were optimized to produce the best classification performance for distinguishing non-disease controls from the PDAC group. First, feature selection was performed by analyzing all possible biomarker combinations. Using only the training and validation datasets, five-fold cross-validation was conducted using five distinct algorithms suitable for classification: logistic regression (LR), random forest (RF), light gradient boost machine (LGBM), extreme gradient boost (XGB), and support vector machine (SVM). Second, the candidate models were subjected to fine-tuning of the hyperparameters, with all possible combinations unique to each algorithm.\u003c/p\u003e \u003cp\u003eVarious metrics were employed to evaluate the model performance, including sensitivity, specificity, Youden index, and AUC, based on the values of True Positive (TP), True Negative (TN), False Positive (FP), False Negative (FN), and receiver operating characteristic (ROC) curve. Sensitivity, specificity, and Youden index were calculated as follows:\u003c/p\u003e \u003cp\u003eSensitivity\u0026thinsp;=\u0026thinsp;TP / (TP\u0026thinsp;+\u0026thinsp;FN)\u003c/p\u003e \u003cp\u003eSpecificity\u0026thinsp;=\u0026thinsp;TN / (TN\u0026thinsp;+\u0026thinsp;FP)\u003c/p\u003e \u003cp\u003eYouden index\u0026thinsp;=\u0026thinsp;Sensitivity\u0026thinsp;+\u0026thinsp;Specificity \u0026minus;\u0026thinsp;1\u003c/p\u003e \u003cp\u003ePython programming language (version 3.11.4) was used in the feature selection and modeling steps. The Pandas and NumPy libraries were used for data processing, and a sklearn library was used for machine-learning modeling and evaluation.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003eStatistical analysis\u003c/h2\u003e \u003cp\u003eThe mRNA expression levels (ΔCt) of potential candidate biomarkers were subjected to statistical analyses. Nonparametric tests (Mann\u0026ndash;Whitney U and Kolmogorov\u0026ndash;Smirnov tests) were performed to select primary biomarker candidates using GraphPad Prism (version 9.5.1; San Diego, CA, USA). Statistical significance was set at p\u0026thinsp;\u0026lt;\u0026thinsp;0.05.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eData availability\u003c/h3\u003e\n\u003cp\u003eThe data generated in this study are available upon reasonable request from the corresponding author.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003ePatient characteristics\u003c/h2\u003e \u003cp\u003eIn total, 1,963 individuals were identified in this study. After the first exclusion, 1,504 subjects were eligible for the study and divided into two groups: the PDAC group (n\u0026thinsp;=\u0026thinsp;417; 28%) and non-disease control group (n\u0026thinsp;=\u0026thinsp;1, 087; 72%). The PDAC group was further divided into two groups according to clinical staging: RPC (n\u0026thinsp;=\u0026thinsp;76; 5%) and APC (n\u0026thinsp;=\u0026thinsp;341; 22%). The APC group included patients with BRPC/LAPC (n\u0026thinsp;=\u0026thinsp;143; 9%) and MPC (n\u0026thinsp;=\u0026thinsp;198; 13%). The median evaluated CA19-9 levels were significantly different between patients with PDAC and non-diseased controls (PDAC, 350.0 U/mL; control, 6.9 U/mL). The demographic data are summarized in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003ePatient characteristics\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"8\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colspan=\"4\" nameend=\"c5\" namest=\"c2\"\u003e \u003cp\u003ePDAC\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eControl\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eTotal\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eRPC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eBRPC\u003c/p\u003e \u003cp\u003e/LAPC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eMPC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c8\" namest=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNumber of patients\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e417\u003c/b\u003e\u003c/p\u003e \u003cp\u003e\u003cb\u003e(28%)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e76\u003c/p\u003e \u003cp\u003e(5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e143\u003c/p\u003e \u003cp\u003e(9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e198\u003c/p\u003e \u003cp\u003e(13%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1087\u003c/p\u003e \u003cp\u003e(72%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c8\" namest=\"c7\"\u003e \u003cp\u003e\u003cb\u003e1504\u003c/b\u003e\u003c/p\u003e \u003cp\u003e\u003cb\u003e(100%)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e66\u003c/b\u003e\u003c/p\u003e \u003cp\u003e\u003cb\u003e(37\u003c/b\u003e\u0026ndash;\u003cb\u003e88)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e65\u003c/p\u003e \u003cp\u003e(47\u0026ndash;88)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e66\u003c/p\u003e \u003cp\u003e(38\u0026ndash;86)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e66\u003c/p\u003e \u003cp\u003e(37\u0026ndash;86)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e45\u003c/p\u003e \u003cp\u003e(20\u0026ndash;76)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c8\" namest=\"c7\"\u003e \u003cp\u003e\u003cb\u003e49\u003c/b\u003e\u003c/p\u003e \u003cp\u003e\u003cb\u003e(20\u003c/b\u003e\u0026ndash;\u003cb\u003e88)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSex\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c8\" namest=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003emale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e219\u003c/b\u003e\u003c/p\u003e \u003cp\u003e\u003cb\u003e(53%)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e48\u003c/p\u003e \u003cp\u003e(63%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e66\u003c/p\u003e \u003cp\u003e(46%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e105\u003c/p\u003e \u003cp\u003e(53%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e689\u003c/p\u003e \u003cp\u003e(63%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c8\" namest=\"c7\"\u003e \u003cp\u003e\u003cb\u003e908\u003c/b\u003e\u003c/p\u003e \u003cp\u003e\u003cb\u003e(60%)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003efemale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e198\u003c/b\u003e\u003c/p\u003e \u003cp\u003e\u003cb\u003e(48%)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e28\u003c/p\u003e \u003cp\u003e(37%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e77\u003c/p\u003e \u003cp\u003e(54%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e93\u003c/p\u003e \u003cp\u003e(47%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e398\u003c/p\u003e \u003cp\u003e(37%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c8\" namest=\"c7\"\u003e \u003cp\u003e\u003cb\u003e596\u003c/b\u003e\u003c/p\u003e \u003cp\u003e\u003cb\u003e(40%)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBaseline\u003c/p\u003e \u003cp\u003etumor marker\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c8\" namest=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCEA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e3.3\u003c/b\u003e\u003c/p\u003e \u003cp\u003e\u003cb\u003e(1\u003c/b\u003e\u0026ndash;\u003cb\u003e1790)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2.6\u003c/p\u003e \u003cp\u003e(1\u0026ndash;142.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2.5\u003c/p\u003e \u003cp\u003e(1\u0026ndash;118)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e5.2\u003c/p\u003e \u003cp\u003e(1\u0026ndash;1790)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1.5\u003c/p\u003e \u003cp\u003e(0.3\u0026ndash;16)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c8\" namest=\"c7\"\u003e \u003cp\u003e\u003cb\u003e1.7\u003c/b\u003e\u003c/p\u003e \u003cp\u003e\u003cb\u003e(0.3\u003c/b\u003e\u0026ndash;\u003cb\u003e1790)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCA19-9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e350\u003c/b\u003e\u003c/p\u003e \u003cp\u003e\u003cb\u003e(1.9\u003c/b\u003e\u0026ndash;\u003cb\u003e20001)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e109.5\u003c/p\u003e \u003cp\u003e(1.9\u0026ndash;7100)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e207\u003c/p\u003e \u003cp\u003e(1.9\u0026ndash;20001)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1435\u003c/p\u003e \u003cp\u003e(1.9\u0026ndash;20001)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e6.9\u003c/p\u003e \u003cp\u003e(0.6\u0026ndash;84.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c8\" namest=\"c7\"\u003e \u003cp\u003e\u003cb\u003e8.9\u003c/b\u003e\u003c/p\u003e \u003cp\u003e\u003cb\u003e(0.6\u003c/b\u003e\u0026ndash;\u003cb\u003e20001)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eData are presented as median (min\u0026ndash;max) or n (%). PDAC, pancreatic ductal adenocarcinoma; RPC, resectable pancreatic cancer; BRPC, borderline resectable pancreatic cancer; LAPC, locally advanced pancreatic cancer; MPC, metastatic pancreatic cancer; CA 19\u0026thinsp;\u0026minus;\u0026thinsp;9, carbohydrate antigen 19\u0026thinsp;\u0026minus;\u0026thinsp;9; CEA, carcinoembryonic antigen.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eCandidate marker screening\u003c/h2\u003e \u003cp\u003eTo search for immunologically relevant markers, a PRISMA-based literature search was conducted to identify 1,755 articles and screen 55 biomarkers (\u003cb\u003eSupplementary Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e\u003c/b\u003e). Candidate markers were screened through a series of nonparametric tests to assess the significance of their expression between the PDAC and control groups. Markers with low expression levels (Ct\u003csub\u003e\u003cem\u003eGAPDH\u003c/em\u003e\u003c/sub\u003e \u0026gt; 35) during RT-qPCR were excluded from the list of candidates. Consequently, 19 markers remained after screening (\u003cb\u003eSupplementary Figure \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e\u003c/b\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eFeature and panel selection\u003c/h2\u003e \u003cp\u003eWe identified 112 M/L models paired with LR, SVM, XGB, and LGBM, but not RF, which surpassed the performance of CA19-9 alone in all four comparisons. Among them, the SVM algorithm with 15 markers (referred to as HELP-15) was selected as the optimal model for classification, with an AUC value of \u0026gt;\u0026thinsp;0.950 in all comparisons and the highest AUC value of 0.967 in control vs. PDAC (\u0026lt;\u0026thinsp;37.0 U/mL; Figs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e2\u003c/span\u003e and \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). The optimal model included \u003cem\u003eCCL5, CCR5, CLEC7A, CXCL8, CXCR2, CXCR4, FOXP3, IFNA1, FNL1, PTGES, PTGES2, PTGS2, SLC27A2, TNF\u003c/em\u003e, and \u003cem\u003eVEGFA\u003c/em\u003e as features (\u003cb\u003eSupplementary Table \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e\u003c/b\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eDiagnostic performance discriminating PDAC and non-diseased control\u003c/h2\u003e \u003cp\u003eThe AUC of HELP-15 was 0.956 for all PDAC patients, which was significantly higher than that of CA19-9 alone (AUC\u0026thinsp;=\u0026thinsp;0.927) (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). The sensitivity was 82.4%, which was higher than that of CA19-9 alone (76.5%), whereas the specificity was 93.6%, which was lower than that of CA19-9 alone (99.5%).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eGenetic functions in biomarker panel\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGene\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDescription\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eReference\u003c/p\u003e \u003cp\u003esequence\u003c/p\u003e \u003cp\u003eaccession ID\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eFunction\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eMapping\u003csup\u003e*\u003c/sup\u003e\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eCCL5\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eC-C Motif Chemokine Ligand 5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNM_002985\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eChemokine related to immune cell migration and inflammation maintenance.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eC, D\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eCCR5\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eC-C Motif Chemokine Receptor 5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNM_000579\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eChemokine receptor that regulates inflammatory reaction.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eC\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eCLEC7A\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eC-Type Lectin Domain Containing 7A\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNM_022570\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eA gene that codes for Dectin-1, which is a pattern recognition receptor for innate immune cells.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eC\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eCXCL8\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eC-X-C motif chemokine ligand 8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNM_000584\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eChemokine that recruits immune cells to the tumor microenvironment.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eA, C, D\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eCXCR2\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eC-X-C Motif Chemokine Receptor 2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNM_001557\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eChemokine receptor for ligands such as CXCL8 that is expressed on the surface of immune cells, in general.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eC\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eCXCR4\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eC-X-C Motif Chemokine Receptor 4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNM_003467\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eChemokine receptor for ligands such as CXCL12 that is expressed on the surface of immune cells, in general.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eB, C\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eFOXP3\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eForkhead Box P3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNM_014009\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eTranscription factor essential for regulatory T-cell development and maintenance of immune tolerance.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eA\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eIFNA1\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eInterferon Alpha 1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNM_024015\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eCytokine that exhibits anti-tumor, and immunomodulatory functions including enhancement of natural killer cell cytotoxicity and macrophage activation.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eA\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eIFNL1\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eInterferon Lambda 1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNM_172140\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eCytokine that exhibits anti-tumor and immunomodulatory functions predominantly in the epithelial tissues.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eC\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003ePTGES\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eProstaglandin E synthase\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNM_004878\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eIsomerase that catalyzes the conversion of prostaglandin H2 (PGH2) into more stable prostaglandin E2 (PGE2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eA\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003ePTGES2\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eProstaglandin E synthase 2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNM_198938\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eIsomerase that catalyzes the conversion of PGH2 into more stable prostaglandin E2 (PGE2).\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eA\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003ePTGS2\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eProstaglandin-Endoperoxide Synthase 2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNM_000963\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAlso known as cyclooxygenase-2 (COX2) that converts arachidonic acid into PGH2.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eA, D\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eSLC27A2\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSolute carrier family 27 member 2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNM_003645\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eA gene that codes for fatty acid transport protein 2 (FATP2) that is crucial for lipid and fatty acid metabolism.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eA\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eTNF\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTumor Necrosis Factor\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNM_000594\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eCytokine that regulates wide spectrum of biological processes including immune system signaling, apoptosis, and inflammation.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eA, D\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eVEGFA\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eVascular Endothelial Growth Factor A\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNM_003376\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eGrowth factor that promotes angiogenesis and vascular permeability which can be initiated by tumor cells.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eB, D\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e* Mapping in the immune system reprogramming: A, immune suppression; B, M2 polarization; C, pro-tumoral activity; D, angiogenesis. (Refer to the Figure \u003cspan refid=\"MOESM3\" class=\"InternalRef\"\u003eS3\u003c/span\u003e)\u003c/p\u003e \u003cp\u003eFor patients with RPC, the biomarker panel showed an AUC of 0.968, which was higher than that of CA19-9 (AUC\u0026thinsp;=\u0026thinsp;0.910). The sensitivity was 93.8%, which was significantly higher than that of CA19-9 alone (62.5%), although the specificity (93.6%) was lower than that of CA19-9 alone (99.5%). HELP-15 showed great improvements over CA19-9 in distinguishing PDAC from the control group, especially in patients with RPC.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003eAdditional analysis of patients with normal Ca19-9 levels\u003c/h2\u003e \u003cp\u003eThe general clinical cutoff value for the CA19-9 test was 37.0 U/mL. It is recommended that individuals with CA19-9 levels\u0026thinsp;\u0026ge;\u0026thinsp;37.0 U/mL are recommended to proceed with imaging techniques and continuous observation. In this study, 21% of patients with PDAC had normal CA19-9 levels (\u0026lt;\u0026thinsp;37.0 U/mL). Considering the genetic absence and delayed expression of CA19-9 in the population, patients with PDAC and normal CA19-9 levels are important targets for improving the overall PDAC diagnosis. The sensitivity and specificity of CA19-9 alone for patients with CA19-9 levels\u0026thinsp;\u0026lt;\u0026thinsp;37.0 U/mL were not calculated because all data fell below the clinical standard. Only the AUC values were compared with those of the final model.\u003c/p\u003e \u003cp\u003eWhen comparing patients between the control and PDAC groups with normal CA19-9 levels, the sensitivity, specificity, and AUC of HELP-15 were 88.9%, 93.5%, and 0.967, respectively (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). The AUC of CA19-9 alone was 0.658, which was significantly lower than that of the final model. For patients with RPC and normal CA19-9 levels, the sensitivity, specificity, and AUC of the final model were 100.0%, 93.5%, and 0.980, respectively. Using the same test set, CA19-9 alone showed an AUC of 0.716, which was significantly lower than that of the final model.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eDiagnostic performance of the marker panel\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"11\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c10\" colnum=\"10\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c11\" colnum=\"11\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colspan=\"4\" nameend=\"c7\" namest=\"c4\"\u003e \u003cp\u003eTraining set\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"4\" nameend=\"c11\" namest=\"c8\"\u003e \u003cp\u003eTest set\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMarker panel\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eSensitivity (95% CI)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSpecificity (95% CI)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAUC\u003c/p\u003e \u003cp\u003e(95% CI)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u003cem\u003ep-value\u003c/em\u003e\u003c/p\u003e \u003cp\u003e\u003cem\u003e(Area\u0026thinsp;=\u0026thinsp;0.5)\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eSensitivity (95% CI)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eSpecificity (95% CI)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c10\"\u003e \u003cp\u003eAUC\u003c/p\u003e \u003cp\u003e(95% CI)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c11\"\u003e \u003cp\u003e\u003cem\u003ep-value\u003c/em\u003e\u003c/p\u003e \u003cp\u003e\u003cem\u003e(Area\u0026thinsp;=\u0026thinsp;0.5)\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003ePDAC\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003eCA19-9 only\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003e78.9\u003c/p\u003e \u003cp\u003e(78.8\u0026ndash;79.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e99.4\u003c/p\u003e \u003cp\u003e(99.4\u0026ndash;99.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.928\u003c/p\u003e \u003cp\u003e(0.914\u0026ndash;0.943)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u003cem\u003e\u0026lt;\u0026thinsp;0.0001\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e76.5\u003c/p\u003e \u003cp\u003e(76.0\u0026ndash;77.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e99.5\u003c/p\u003e \u003cp\u003e(99.5\u0026ndash;99.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e0.927\u003c/p\u003e \u003cp\u003e(0.897\u0026ndash;0.956)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e\u003cem\u003e\u0026lt;\u0026thinsp;0.0001\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e15 markers\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003e91.3\u003c/p\u003e \u003cp\u003e(91.2\u0026ndash;91.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e95.6\u003c/p\u003e \u003cp\u003e(95.6\u0026ndash;95.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.980\u003c/p\u003e \u003cp\u003e(0.973\u0026ndash;0.988)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u003cem\u003e\u0026lt;\u0026thinsp;0.0001\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e82.4\u003c/p\u003e \u003cp\u003e(81.9\u0026ndash;82.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e93.6\u003c/p\u003e \u003cp\u003e(93.4\u0026ndash;93.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e0.956\u003c/p\u003e \u003cp\u003e(0.933\u0026ndash;0.979)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e\u003cem\u003e\u0026lt;\u0026thinsp;0.001\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e15 markers\u0026thinsp;+\u0026thinsp;CA19-9\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003e98.2\u003c/p\u003e \u003cp\u003e(98.2\u0026ndash;98.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e98.8\u003c/p\u003e \u003cp\u003e(98.8\u0026ndash;98.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.999\u003c/p\u003e \u003cp\u003e(0.998-1.000)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u003cem\u003e\u0026lt;\u0026thinsp;0.0001\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e92.9\u003c/p\u003e \u003cp\u003e(92.6\u0026ndash;93.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e92.2\u003c/p\u003e \u003cp\u003e(92.0-92.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e0.971\u003c/p\u003e \u003cp\u003e(0.953\u0026ndash;0.990)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e\u003cem\u003e\u0026lt;\u0026thinsp;0.1\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eRPC\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003eCA19-9 only\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003e70.0\u003c/p\u003e \u003cp\u003e(69.6\u0026ndash;70.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e99.4\u003c/p\u003e \u003cp\u003e(99.4\u0026ndash;99.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.917\u003c/p\u003e \u003cp\u003e(0.899\u0026ndash;0.934)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u003cem\u003e\u0026lt;\u0026thinsp;0.0001\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e62.5\u003c/p\u003e \u003cp\u003e(60.9\u0026ndash;64.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e99.5\u003c/p\u003e \u003cp\u003e(99.5\u0026ndash;99.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e0.910\u003c/p\u003e \u003cp\u003e(0.873\u0026ndash;0.946)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e\u003cem\u003e\u0026lt;\u0026thinsp;0.0001\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e15 markers\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003e91.7\u003c/p\u003e \u003cp\u003e(91.4\u0026ndash;91.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e95.6\u003c/p\u003e \u003cp\u003e(95.6\u0026ndash;95.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.981\u003c/p\u003e \u003cp\u003e(0.972\u0026ndash;0.990)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u003cem\u003e\u0026lt;\u0026thinsp;0.0001\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e93.8\u003c/p\u003e \u003cp\u003e(93.0-94.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e93.6\u003c/p\u003e \u003cp\u003e(93.4\u0026ndash;93.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e0.968\u003c/p\u003e \u003cp\u003e(0.946\u0026ndash;0.991)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e\u003cem\u003e\u0026lt;\u0026thinsp;0.0001\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e15 markers\u0026thinsp;+\u0026thinsp;CA19-9\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003e98.3\u003c/p\u003e \u003cp\u003e(98.2\u0026ndash;98.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e98.8\u003c/p\u003e \u003cp\u003e(0.988\u0026ndash;0.989)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.999\u003c/p\u003e \u003cp\u003e(0.998-1.000)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u003cem\u003e\u0026lt;\u0026thinsp;0.0001\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e93.8\u003c/p\u003e \u003cp\u003e(93.0-94.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e92.2\u003c/p\u003e \u003cp\u003e(92.0-92.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e0.985\u003c/p\u003e \u003cp\u003e(0.970-1.000)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e\u003cem\u003e\u0026lt;\u0026thinsp;0.0001\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003ePDAC\u003c/b\u003e\u003c/p\u003e \u003cp\u003e\u003cb\u003e(CA19-9\u0026thinsp;\u0026lt;\u0026thinsp;37.0 U/ml)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003eCA19-9 only\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eN/A\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eN/A\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.660\u003c/p\u003e \u003cp\u003e(0.630\u0026ndash;0.690)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u003cem\u003e\u0026lt;\u0026thinsp;0.0001\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eN/A\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003eN/A\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e0.658\u003c/p\u003e \u003cp\u003e(0.597\u0026ndash;0.719)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e\u003cem\u003e\u0026lt;\u0026thinsp;0.0001\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e15 markers\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003e95.7\u003c/p\u003e \u003cp\u003e(95.5\u0026ndash;95.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e95.6\u003c/p\u003e \u003cp\u003e(95.6\u0026ndash;95.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.987\u003c/p\u003e \u003cp\u003e(0.979\u0026ndash;0.994)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u003cem\u003e\u0026lt;\u0026thinsp;0.0001\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e88.9\u003c/p\u003e \u003cp\u003e(87.9\u0026ndash;89.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e93.5\u003c/p\u003e \u003cp\u003e(93.3\u0026ndash;93.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e0.967\u003c/p\u003e \u003cp\u003e(0.944\u0026ndash;0.990)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e\u003cem\u003e\u0026lt;\u0026thinsp;0.0001\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e15 markers\u0026thinsp;+\u0026thinsp;CA19-9\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003e91.3\u003c/p\u003e \u003cp\u003e(91.1\u0026ndash;91.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e99.4\u003c/p\u003e \u003cp\u003e(99.4\u0026ndash;99.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.998\u003c/p\u003e \u003cp\u003e(0.995-1.000)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u003cem\u003e\u0026lt;\u0026thinsp;0.0001\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e66.7\u003c/p\u003e \u003cp\u003e(65.2\u0026ndash;68.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e92.6\u003c/p\u003e \u003cp\u003e(92.4\u0026ndash;92.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e0.885\u003c/p\u003e \u003cp\u003e(0.844\u0026ndash;0.926)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e\u003cem\u003e\u0026lt;\u0026thinsp;0.0001\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eRPC\u003c/b\u003e\u003c/p\u003e \u003cp\u003e\u003cb\u003e(CA19-9\u0026thinsp;\u0026lt;\u0026thinsp;37.0 U/ml)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003eCA19-9 only\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eN/A\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eN/A\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.727\u003c/p\u003e \u003cp\u003e(0.697\u0026ndash;0.756)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u003cem\u003e\u0026lt;\u0026thinsp;0.0001\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eN/A\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003eN/A\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e0.716\u003c/p\u003e \u003cp\u003e(0.657\u0026ndash;0.775)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e\u003cem\u003e\u0026lt;\u0026thinsp;0.0001\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e15 markers\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003e94.4\u003c/p\u003e \u003cp\u003e(94.1\u0026ndash;94.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e95.6\u003c/p\u003e \u003cp\u003e(95.6\u0026ndash;95.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.991\u003c/p\u003e \u003cp\u003e(0.985\u0026ndash;0.997)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u003cem\u003e\u0026lt;\u0026thinsp;0.0001\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e100.0\u003c/p\u003e \u003cp\u003e(100.0-100.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e93.5\u003c/p\u003e \u003cp\u003e(93.3\u0026ndash;93.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e0.980\u003c/p\u003e \u003cp\u003e(0.961\u0026ndash;0.998)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e\u003cem\u003e\u0026lt;\u0026thinsp;0.0001\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e15 markers\u0026thinsp;+\u0026thinsp;CA19-9\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003e94.4\u003c/p\u003e \u003cp\u003e(94.1\u0026ndash;94.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e99.4\u003c/p\u003e \u003cp\u003e(0.994\u0026ndash;0.994)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.999\u003c/p\u003e \u003cp\u003e(0.997-1.000)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u003cem\u003e\u0026lt;\u0026thinsp;0.0001\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e80.0\u003c/p\u003e \u003cp\u003e(77.6\u0026ndash;82.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e92.6\u003c/p\u003e \u003cp\u003e(92.4\u0026ndash;92.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e0.967\u003c/p\u003e \u003cp\u003e(0.943\u0026ndash;0.990)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e\u003cem\u003e\u0026lt;\u0026thinsp;0.0001\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eAUC, area under curve; PDAC, pancreatic ductal adenocarcinoma; RPC, resectable pancreatic cancer; CA 19\u0026thinsp;\u0026minus;\u0026thinsp;9, carbohydrate antigen 19\u0026thinsp;\u0026minus;\u0026thinsp;9.\u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eWe developed an immune system-derived mRNA biomarker panel accompanied by an M/L algorithm that displayed effective diagnostic performance against PDAC. The panel included \u003cem\u003eCCL5, CCR5, CLEC7A, CXCL8, CXCR2, CXCR4, FOXP3, IFNA1, IFNL1, PTGES, PTGES2, PTGS2, SLC27A2, TNF\u003c/em\u003e, and \u003cem\u003eVEGFA\u003c/em\u003e, which were selected from 55 candidate biomarkers using a series of nonparametric tests, M/L-based feature selections, and performance evaluations. Based on these results, we constructed an optimal biomarker model that showed significant improvements in distinguishing between the RPC and PDAC groups with normal CA19-9 levels.\u003c/p\u003e \u003cp\u003eIn the last two decades, several studies have focused on combining multiple proteins, miRNAs, and cfDNAs to achieve meaningful PDAC diagnostic performance [\u003cspan additionalcitationids=\"CR11 CR12\" citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan additionalcitationids=\"CR21\" citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. Klein et al. demonstrated a reasonable performance for various cancer types. However, the sensitivity for PDAC was measured at 83.7%, with a lower stage I sensitivity of 61.9% [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]. Lee et al. reported a sensitivity of 92.5% for detecting RPC; however, the sensitivity was 64.3% in the group with a normal range of CA19-9 when using triple protein marker panels, including LRG1, TTR, and CA19-9 [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]. Recently, Nakamura et al. showed excellent diagnostic performance with their transcriptomic signature, with an AUC of 0.930 and sensitivity in the early stages of PDAC (stages I and II) [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]. However, exosome-based and cell-free miRNAs require complicated analytical procedures that use unstable exosome-based miRNAs, making signatures less accessible to the general population [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]. Therefore, clinically feasible biomarkers are needed for PDAC detection.\u003c/p\u003e \u003cp\u003eIn this context, our biomarker panel demonstrated several notable strengths compared to those in previous studies. First, the final model consistently outperformed CA19-9 in diagnosing RPC (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e2\u003c/span\u003ec, \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e2\u003c/span\u003ed), even in patients with normal CA19-9 levels (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Second, the use of RT-qPCR, a widely available and cost-effective analytical method, enhances the accessibility and affordability of the HELP-15 panel for the general population. Third, we utilized machine-learning models to optimize the diagnostic performance of the final model by leveraging extensive clinical data. By partitioning the dataset into training, validation, and test sets, we minimized the risk of overfitting and ensured robust performance. The efficacy of the biomarker panel was particularly promising for patients with RPC and normal CA19-9 levels. Although the CA19-9 test is limited by its delayed expression in early-stage PDAC and susceptibility to false-negative results, the HELP-15 panel, when used in conjunction with the CA19-9 test, shows significant potential as a screening tool for PDAC.\u003c/p\u003e \u003cp\u003eOur study had some limitations. First, there was a disproportionate age distribution between individuals in the PDAC and control groups. To ensure that HELP-15 was not biased by age, we performed a correlation test between age and model probability, which indicated that the correlation was not significant (\u003cb\u003eSupplementary Figure \u003cspan refid=\"MOESM4\" class=\"InternalRef\"\u003eS4\u003c/span\u003e\u003c/b\u003e). Second, some molecular biological mechanisms have not been fully elucidated. However, we implemented a structured representation method that included a comprehensive literature review and multifaceted comparison of the experimental data. Finally, the study was conducted at a single center. Although multicenter research has advantages, a homogeneous population might be more effective for identifying novel biomarkers. Further large-scale and multicenter clinical validation is necessary to address these limitations.\u003c/p\u003e \u003cp\u003eIn conclusion, compared with CA19-9, the immune system-derived mRNA biomarker panel, HELP-15, significantly improved the diagnostic performance in patients with RPC, particularly in those with PDAC and normal CA19-9 levels.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cdiv class=\"DefinitionList\"\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eAPC\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eadvanced pancreatic cancer\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eAUC\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003earea under the curve\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eBRPC\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eborderline resectable pancreatic cancer\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eCA19-9\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003ecarbohydrate antigen 19\u0026thinsp;\u0026minus;\u0026thinsp;9\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eFN\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003efalse negative\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eFP\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003efalse positive\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eLAPC\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003elocally advanced pancreatic cancer\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eLGBM\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003elight gradient boost machine\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eLR\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003elogistic regression\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eM/L\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003emachine learning\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eMPC\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003emetastatic pancreatic cancer\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003ePDAC\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003epancreatic ductal adenocarcinoma\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eRF\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003erandom forest\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eRPC\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eresectable pancreatic cancer\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eSVM\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003esupport vector machine\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eTME\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003etumor microenvironment\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eTN\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003etrue negative\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eTP\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003etrue positive\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eXGB\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eextreme gradient boost\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe obtained ethics approvals from the Human Bioresource Center of the Seoul National University Bundang Hospital (Affiliated College of Medicine, Seoul National University).\u0026nbsp;The study was reviewed and approved by the Institutional Review Board (IRB approval number: X-2011-651-903) and was conducted in accordance with the Declaration of Helsinki. Private patient information was anonymized and de-identified prior to analysis.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and material\u003c/strong\u003e \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe data that support the findings of this study are not openly available due to reasons of sensitivity and are available from the corresponding author upon reasonable request. Data are located in controlled access data storage at the Human Bioresource Center of Seoul National University Bundang Hospital.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eJin-Hyeok Hwang and Jihie Kim declare conflict of interest including advisory committee of HuVet bio Inc. Hyoung-Hwa Jeong is a CEO of HuVet bio Inc. Other authors have no COI to declare.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026apos; contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eConceptualization - Jong-chan Lee, Hyoung Hwa Jeong, Jihie Kim, Jin-Hyeok Hwang\u003c/p\u003e\n\u003cp\u003eData curation - Eun-Jin Sim, Yuna Youn, Jin-Sik Bae, Seong-mo Koo, Mun-sub Byoun\u003c/p\u003e\n\u003cp\u003eFormal analysis - Sung Won Kang, Eun-Jin Sim, Yuna Youn\u003c/p\u003e\n\u003cp\u003eFunding acquisition - Jihie Kim, Jin-Hyeok Hwang\u003c/p\u003e\n\u003cp\u003eInvestigation - Sung Won Kang, Serin Kwon, Seoi Hong, Yunji Kim\u003c/p\u003e\n\u003cp\u003eMethodology - Jong-chan Lee, Serin Kwon, Seoi Hong, Yunji Kim\u003c/p\u003e\n\u003cp\u003eProject administration - Hyoung Hwa Jeong, Jihie Kim, Jin-Hyeok Hwang\u003c/p\u003e\n\u003cp\u003eResources - Eun-Jin Sim, Serin Kwon, Seoi Hong, Yunji Kim\u003c/p\u003e\n\u003cp\u003eSoftware - Sung Won Kang, Serin Kwon, Seoi Hong, Yunji Kim\u003c/p\u003e\n\u003cp\u003eSupervision - Kwangrok Jung, Jaihwan Kim, Jihie Kim, Jin-Hyeok Hwang\u003c/p\u003e\n\u003cp\u003eValidation - Serin Kwon, Seoi Hong, Yunji Kim, Yuna Youn, Kwangrok Jung, Jaihwan Kim\u003c/p\u003e\n\u003cp\u003eVisualization - Serin Kwon, Seoi Hong, Yunji Kim, Yuna Youn\u003c/p\u003e\n\u003cp\u003eWriting \u0026ndash; original draft - Jong-chan Lee, Sung Won Kang, Eun-Jin Sim, Serin Kwon, Seoi Hong, Yunji Kim, Kwangrok Jung, Jaihwan Kim\u003c/p\u003e\n\u003cp\u003eWriting \u0026ndash; review \u0026amp; editing - Jong-chan Lee, Sung Won Kang, Eun-Jin Sim, Serin Kwon, Seoi Hong, Yunji Kim, Kwangrok Jung, Jaihwan Kim, Hyoung Hwa Jeong, Jihie Kim, Jin-Hyeok Hwang\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe thank Editage (www.editage.co.kr) for the English proofreading.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eSiegel RL, Miller KD, Wagle NS, Jemal A, Cancer statistics. 2023. CA Cancer J Clin. 2023;73:17\u0026ndash;48. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3322/caac.21763\u003c/span\u003e\u003cspan address=\"10.3322/caac.21763\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi J, Wientjes MG, Au JL-S. Pancreatic cancer: Pathobiology, treatment options, and drug delivery. AAPS J. 2010;12:223\u0026ndash;32. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1208/s12248-010-9181-5\u003c/span\u003e\u003cspan address=\"10.1208/s12248-010-9181-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKim VM, Ahuja N. Early detection of pancreatic cancer. Chin J Cancer Res. 2015;27:321\u0026ndash;31. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3978/j.issn.1000-9604.2015.07.03\u003c/span\u003e\u003cspan address=\"10.3978/j.issn.1000-9604.2015.07.03\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJones RP, Psarelli E-E, Jackson R, Ghaneh P, Halloran CM, Palmer DH, et al. Patterns of recurrence after resection of pancreatic ductal adenocarcinoma: A secondary analysis of the ESPAC-4 randomized adjuvant chemotherapy trial. JAMA Surg. 2019;154:1038\u0026ndash;48. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1001/jamasurg.2019.3337\u003c/span\u003e\u003cspan address=\"10.1001/jamasurg.2019.3337\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGrossberg AJ, Chu LC, Deig CR, Fishman EK, Hwang WL, Maitra A, et al. Multidisciplinary standards of care and recent progress in pancreatic ductal adenocarcinoma. CA Cancer J Clin. 2020;70:375\u0026ndash;403. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3322/caac.21626\u003c/span\u003e\u003cspan address=\"10.3322/caac.21626\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcGuigan A, Kelly P, Turkington RC, Jones C, Coleman HG, McCain RS. Pancreatic cancer: A review of clinical diagnosis, epidemiology, treatment and outcomes. World J Gastroenterol. 2018;24:4846\u0026ndash;61. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3748/wjg.v24.i43.4846\u003c/span\u003e\u003cspan address=\"10.3748/wjg.v24.i43.4846\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRyan DP, Hong TS, Bardeesy N. Pancreatic adenocarcinoma. N Engl J Med. 2014;371:1039\u0026ndash;49. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1056/NEJMra1404198\u003c/span\u003e\u003cspan address=\"10.1056/NEJMra1404198\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu J, Gao J, Du Y, Li Z, Ren Y, Gu J, et al. Combination of plasma microRNAs with serum CA19-9 for early detection of pancreatic cancer. Int J Cancer. 2012;131:683\u0026ndash;91. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/ijc.26422\u003c/span\u003e\u003cspan address=\"10.1002/ijc.26422\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Epub November 19, 2011.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGuo M, Luo G, Lu R, Shi W, Cheng H, Lu Y, et al. Distribution of Lewis and secretor polymorphisms and corresponding CA19-9 antigen expression in a Chinese population. FEBS Open Bio. 2017;7:1660\u0026ndash;71. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/2211-5463.12278. eCollection\u003c/span\u003e\u003cspan address=\"10.1002/2211-5463.12278. eCollection\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMacdonald S, Mair F. Tackling cancers of unmet need: The pancreatic cancer pathway. Lancet Gastroenterol Hepatol. 2016;1:266\u0026ndash;7. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/S2468-1253(16)30113-3\u003c/span\u003e\u003cspan address=\"10.1016/S2468-1253(16)30113-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCohen JD, Li L, Wang Y, Thoburn C, Afsari B, Danilova L, et al. Detection and localization of surgically resectable cancers with a multi-analyte blood test. Science. 2018;359:926\u0026ndash;30. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1126/science.aar3247\u003c/span\u003e\u003cspan address=\"10.1126/science.aar3247\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu MC, Oxnard GR, Klein EA, Swanton C, Seiden MV, CCGA Consortium. Sensitive and specific multi-cancer detection and localization using methylation signatures in cell-free DNA. Ann Oncol. 2020;31:745\u0026ndash;59. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.annonc.2020.02.011\u003c/span\u003e\u003cspan address=\"10.1016/j.annonc.2020.02.011\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKillock D, Diagnosis. CancerSEEK and destroy \u0026ndash; A blood test for early cancer detection. Nat Rev Clin Oncol. 2018;15:133. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/nrclinonc.2018.21\u003c/span\u003e\u003cspan address=\"10.1038/nrclinonc.2018.21\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShi Y, Du L, Lin L, Wang Y. Tumour-associated mesenchymal stem/stromal cells: Emerging therapeutic targets. Nat Rev Drug Discov. 2017;16:35\u0026ndash;52. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/nrd.2016.193\u003c/span\u003e\u003cspan address=\"10.1038/nrd.2016.193\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSai B, Dai Y, Fan S, Wang F, Wang L, Li Z, et al. Cancer-educated mesenchymal stem cells promote the survival of cancer cells at primary and distant metastatic sites via the expansion of bone marrow-derived-PMN-MDSCs. Cell Death Dis. 2019;10:941. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41419-019-2149-1\u003c/span\u003e\u003cspan address=\"10.1038/s41419-019-2149-1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMathew E, Brannon AL, Del Vecchio AC, Garcia PE, Penny MK, Kane KT, et al. Mesenchymal stem cells promote pancreatic tumor growth by inducing alternative polarization of macrophages. Neoplasia. 2016;18:142\u0026ndash;51. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.neo.2016.01.005\u003c/span\u003e\u003cspan address=\"10.1016/j.neo.2016.01.005\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWaghray M, Yalamanchili M, Dziubinski M, Zeinali M, Erkkinen M, Yang H, et al. GM-CSF mediates mesenchymal-epithelial cross-talk in pancreatic cancer. Cancer Discov. 2016;6:886\u0026ndash;99. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1158/2159-8290.CD-15-0947\u003c/span\u003e\u003cspan address=\"10.1158/2159-8290.CD-15-0947\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVeglia F, Tyurin VA, Blasi M, De Leo A, Kossenkov AV, Donthireddy L, et al. Fatty acid transport Protein 2 reprograms neutrophils in cancer. Nature. 2019;569:73\u0026ndash;8. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41586-019-1118-2\u003c/span\u003e\u003cspan address=\"10.1038/s41586-019-1118-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePage MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. BMJ. 2021;372:n71. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1136/bmj.n71\u003c/span\u003e\u003cspan address=\"10.1136/bmj.n71\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWei J, Yang L, Wu Y-N, Xu J. Serum miR-1290 and miR-1246 as potential diagnostic biomarkers of human pancreatic cancer. J Cancer. 2020;11:1325\u0026ndash;33. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.7150/jca.38048\u003c/span\u003e\u003cspan address=\"10.7150/jca.38048\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGuler GD, Ning Y, Ku C-J, Phillips T, McCarthy E, Ellison CK, et al. Detection of early stage pancreatic cancer using 5-hydroxymethylcytosine signatures in circulating cell free DNA. Nat Commun. 2020;11:5270. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41467-020-18965-w\u003c/span\u003e\u003cspan address=\"10.1038/s41467-020-18965-w\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBorrebaeck CAK. Precision diagnostics: Moving towards protein biomarker signatures of clinical utility in cancer. Nat Rev Cancer. 2017;17:199\u0026ndash;204. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/nrc.2016.153\u003c/span\u003e\u003cspan address=\"10.1038/nrc.2016.153\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKlein EA, Richards D, Cohn A, Tummala M, Tummala M, Lapham R, Cosgrove D, et al. Clinical validation of a targeted methylation-based multi-cancer early detection test using an independent validation set. Ann Oncol. 2021;32:1167\u0026ndash;77. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.annonc.2021.05.806\u003c/span\u003e\u003cspan address=\"10.1016/j.annonc.2021.05.806\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLee D-H, Yoon W, Lee A, Han Y, Byun Y, Kang JS, et al. Multi-biomarker panel prediction model for diagnosis of pancreatic cancer. J Hepatobiliary Pancreat Sci. 2023;30:122\u0026ndash;32. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/jhbp.986\u003c/span\u003e\u003cspan address=\"10.1002/jhbp.986\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNakamura K, Zhu Z, Roy S, Jun E, Han H, Munoz RM, et al. An exosome-based transcriptomic signature for noninvasive, early detection of patients with pancreatic ductal adenocarcinoma: A multicenter cohort study. Gastroenterology. 2022;163:1252\u0026ndash;66. .e2.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSoria F, Droller MJ, Lotan Y, Gontero P, D\u0026rsquo;Andrea D, Gust KM, et al. An up-to-date catalog of available urinary biomarkers for the surveillance of non-muscle invasive bladder cancer. World J Urol. 2018;36:1981\u0026ndash;95. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00345-018-2380-x\u003c/span\u003e\u003cspan address=\"10.1007/s00345-018-2380-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-cancer","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bcan","sideBox":"Learn more about [BMC Cancer](http://bmccancer.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bcan/default.aspx","title":"BMC Cancer","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"liquid biopsy, mRNA, buffy coat, pancreatic cancer, biomarker","lastPublishedDoi":"10.21203/rs.3.rs-5119465/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5119465/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003ePancreatic ductal adenocarcinoma (PDAC) is one of the most lethal malignancies and most often diagnosed at an advanced stage. Identification of markers for the early diagnosis of PDAC is crucial. In this study, we aimed to identify novel mRNA biomarkers for diagnosing PDAC, focusing on early-stage tumorigenesis and associated immunological changes.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003e Blood samples and clinical information from 1,963 individuals were obtained from a single tertiary hospital between 2015 and 2021. Candidate mRNA biomarkers were identified through literature review, and their expression levels in buffy coat samples were measured using reverse-transcription quantitative polymerase chain reaction. Machine learning-based feature selection confirmed the final biomarker panel, which was tested using an independent dataset for diagnostic performance.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eIn total, 1,504 individuals (417 patients with PDAC and 1,087 non-diseased controls) were eligible for the study. Among the 55 candidate biomarkers identified, 15 mRNAs (\u003cem\u003eCCL5, CCR5, CLEC7A, CXCL8, CXCR2, CXCR4, FOXP3, IFNA1, IFNL1, PTGES, PTGES2, PTGS2, SLC27A2, TNF\u003c/em\u003e, and \u003cem\u003eVEGFA)\u003c/em\u003e were selected based on their diagnostic performance in distinguishing PDAC from control groups. The final model, HELP-15 (Human Early Liquid biopsy for PDAC), identified all PDAC stages (area under the curve [AUC]\u0026thinsp;=\u0026thinsp;0.956) in the test set. For resectable pancreatic cancer (RPC), the AUC was 0.968, compared to 0.910 for carbohydrate antigen 19\u0026thinsp;\u0026minus;\u0026thinsp;9 (CA19-9). The combination of the panel and CA19-9 levels had an AUC of 0.985 in patients with RPC. For all PDAC stages in patients with normal CA19-9 levels, the AUC of the panel was 0.967, whereas CA19-9 alone or in combination with the panel had AUCs of 0.658 and 0.885, respectively.\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e \u003cp\u003eCompared to CA19-9, the mRNA biomarker panel, HELP-15, significantly improved diagnostic performance in patients with RPC, particularly in those with normal CA19-9 levels.\u003c/p\u003e","manuscriptTitle":"Novel mRNA biomarker-based liquid biopsy for the detection of resectable pancreatic cancer","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-11-11 09:56:16","doi":"10.21203/rs.3.rs-5119465/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2024-10-03T06:33:56+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-10-01T12:44:55+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-10-01T12:42:56+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Cancer","date":"2024-09-19T22:09:01+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-cancer","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bcan","sideBox":"Learn more about [BMC Cancer](http://bmccancer.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bcan/default.aspx","title":"BMC Cancer","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"dc8c587b-936f-4434-9a48-5d2c0fdded8e","owner":[],"postedDate":"November 11th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-04-28T16:03:13+00:00","versionOfRecord":{"articleIdentity":"rs-5119465","link":"https://doi.org/10.1186/s12885-025-14124-w","journal":{"identity":"bmc-cancer","isVorOnly":false,"title":"BMC Cancer"},"publishedOn":"2025-04-23 15:58:08","publishedOnDateReadable":"April 23rd, 2025"},"versionCreatedAt":"2024-11-11 09:56:16","video":"","vorDoi":"10.1186/s12885-025-14124-w","vorDoiUrl":"https://doi.org/10.1186/s12885-025-14124-w","workflowStages":[]},"version":"v1","identity":"rs-5119465","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5119465","identity":"rs-5119465","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}