Early Diagnosis Of Pancreatic Cancer Research Articles

Early diagnosis of pancreatic cancer: Shedding light on an unresolved challenge.

Diagnosing early-stage pancreatic cancer (PC) remains a clinical challenge. Hence, studying novel imaging aspects that could enhance the diagnostic accuracy of malignant pancreatic precursor lesions is imperative. This article aims to underscore the promising role of emerging imaging aspects that may facilitate the earlier diagnosis of PC, thereby improving its management and prognosis.

Pancreatic cancer: from early detection to personalized treatment approaches.

Pancreatic cancer is notorious for its persistently poor prognosis and health outcomes, so some of the questions that may be begged are "Why is it mostly diagnosed at end stage?", "What could we possibly do with the advancing technology in today's world to detect early pancreatic cancer and intervene?", and "Are there any implementation of the existing novel imaging technologies?". Well, to start with, this is in part because the majority of patients presented would already have reached a locally advanced or metastatic stage at the time of diagnosis due to its highly aggressive characteristics and lack of symptoms. Due to this striking disparity in survival, advancements in early detection and intervention are likely to significantly increase patients' survival. Presently, screening is frequently used in high-risk individuals in order to obtain an early pancreatic cancer diagnosis. Having a thorough understanding of the pathogenesis and risk factors of pancreatic cancer may enable us to identify individuals at high risk, diagnose the disease early, and begin treatment promptly. In this review, the authors outline the clinical hurdles to early pancreatic cancer detection, describe high-risk populations, and discuss current screening initiatives for high-risk individuals. The ultimate goal of this current review is to study the roles of both traditional and novel imaging modalities for early pancreatic cancer detection. A lot of the novel imaging techniques mentioned seem promising, but they need to be put to the test on a large scale and may need to be combined with other non-invasive biomarkers before they can be widely used.

Morphology and optical properties of Au–Ag hybrid nanoparticles regulation and its ultra-sensitive SERS immunoassay detection in carbohydrate antigen 19-9

The development of an ultra-sensitive detection method for carbohydrate antigen 19-9 (CA19-9) is very important for the early diagnosis of pancreatic cancer. In this work, we developed a new strategy to achieve a variety of Au–Ag hybrid nanoparticles from janus to core-satellite which is controlled by the volume of AgNO3 and the concentration of benzimidazolecarboxylic acid (MBIA). With the volume of AgNO3 increased, Au–Ag hybrid nanoparticles changed from janus to core-satellite and the characteristic absorption peak showed two opposite trends. The size and number of Ag islands were determined by the concentration of MBIA. Au–Ag core-satellites nanoparticles with a large number of small-sized Ag have the highest SERS intensity. Then we used them as SERS nanotags and Au-Polystyrene nanospheres modified by captured anti-CA19-9 antibody as solid substrates to realize the ultra-sensitive detection of CA19-9 with a low limit of detection of 1.25 × 10−6 IU/mL and a wide linear range of 1.00 × 10−5 -1.00 × 104 IU/mL. This work not only demonstrates that MBIA and AgNO3 were the key factors in the growth of Au–Ag hybrid nanoparticles from 2D to 3D structure but also supplies an ultra-sensitive detection method for CA19-9 which has a potential practicability in the clinical early diagnoses of pancreatic cancer.

Evaluations of the combined use of blood- and tissue-based protein biomarkers for pancreatic cancer.

Pancreatic ductal adenocarcinoma (PDAC) is a highly aggressive malignancy with a low 5-year survival rate. Biomarkers may be of value for the early diagnosis of pancreatic cancer. This study assessed blood- and tumour tissue-based biomarkers associated with pancreatic cancer. We studied 61 patients who underwent pancreatic resection. Of these 61 patients, 46 patients had PDAC, and 15 patients had inflammatory tumours. Blood and tumour tissue levels of VEGF, hypoxia-inducible factor 1α (HIF-1α) and glucose transporter 1 (GLUT1) were measured. Blood concentrations of VEGF (p < 0.000001) and HIF-1α (p = 0.000002) were significantly higher in the PDAC group than in the inflammatory tumour group. Tumour tissue concentrations of VEGF (p < 0.000001), HIF-1α (p = 0.000005) and GLUT1 (0.000002) were also significantly higher in the PDAC group. Univariate analyses revealed that age, BMI, and blood levels of CA19-9, VEGF, and HIF-1α were potential predictors of PDAC. Potential predictors of PDAC in tumour tissue were VEGF, HIF-1α and GLUT1. Multivariate analyses found that VEGF was the most powerful independent predictor of PDAC in blood (OR = 1.016; 95% CI: 1.007-1.025; 0.001) and tumour tissue (OR = 1.02; 95% CI: 1.008-1.032, p = 0.001). The cut-off point for blood VEGF was 134.56 pg/ml, with a sensitivity of 97.8%, specificity of 86.7%, PPV of 95.7%, and NPV of 92.9%. The cut-off point for tissue tumour VEGF in PDAC was 208.59 pg/mg, with a sensitivity, specificity, PPV and NPV of 97.7%, 92.9%, 97.7%, and 92.9%, respectively. There are significant differences in blood-based biomarkers for differentiating between PDAC and inflammatory tumours of the pancreas. VEGF was an independent predictor of PDAC independent of its addition to the routinely used tumour marker CA19-9 antigen.

A causality-inspired generalized model for automated pancreatic cancer diagnosis

Pancreatic cancer (PC) is a severely malignant cancer variant with high mortality. Since PC has no obvious symptoms, most PC patients are belatedly diagnosed at advanced disease stages. Recently, artificial intelligence (AI) approaches have demonstrated promising prospects for early diagnosis of pancreatic cancer. However, certain non-causal factors (such as intensity and texture appearance variations, also called confounders) tend to induce spurious correlation with PC diagnosis. This undermines the generalization performance and the clinical applicability of the AI-based PC diagnosis approaches. Therefore, we propose a causal intervention based automated method for pancreatic cancer diagnosis with contrast-enhanced computerized tomography (CT) images, where a confounding effects reduction scheme is developed for alleviating spurious correlations to achieve unbiased learning, thereby improving the generalization performance. Specifically, a continuous image generation strategy was developed to simulate wide variations of intensity differences caused by imaging heterogeneities, where Monte Carlo sampling is added to further enhance the continuity of simulated images. Then, to enhance the pancreatic texture variability, a texture diversification method was introduced in conjunction with gradient-based data augmentation. Finally, a causal intervention strategy was proposed to alleviate the adverse confounding effects by decoupling the causal and non-causal factors and combining them randomly. Extensive experiments showed remarkable diagnosis performance on a cross-validation dataset. Also, promising generalization performance with an average accuracy of 0.87 was attained on three independent test sets of a total of 782 subjects. Therefore, the proposed method shows high clinical feasibility and applicability for pancreatic cancer diagnosis.

Novel serum protein biomarker panel for early diagnosis of pancreatic cancer.

Pancreatic ductal adenocarcinoma (PDAC) is one of the deadliest cancers. Late presentation of disease at the time of diagnosis is one of the major reasons for dismal prognostic outcomes for PDAC patients. Currently, there is a lack of clinical biomarkers, which can be used to diagnose PDAC patients at an early resectable stage. This study performed proteomic mass spectrometry to identify novel blood-based biomarkers for early diagnosis of PDAC. Serum specimens from 88 PDAC patients and 88 healthy controls (60 discovery cohort and 28 validation cohort) were analyzed using data independent acquisition high resolution mass spectrometry to identify candidate biomarker proteins. A total of 249 proteins were identified and quantified by the mass spectrometric analysis. Six proteins were markedly (>1.5 fold) and significantly (p < .05; q < 0.1) increased in PDAC patients compared to healthy controls in discovery cohort. Notably, four of these six proteins were significantly upregulated in an independent validation cohort. The top three upregulated proteins (i.e., Polymeric Immunoglobulin Receptor [PIGR], von Willebrand Factor [vWF], and Fibrinogen) were validated using enzyme linked immunosorbent assay, which led to selection of PIGR and vWF as a diagnostic biomarker panel for PDAC. The panel showed high ability to diagnose early stage (stage I and II) PDAC patients (area under the curve [AUC]: 0.8926), which was further improved after the addition of clinically used prognostic biomarker (Ca 19-9) to the panel (AUC: 0.9798). In conclusion, a novel serum protein biomarker panel for early diagnosis of PDAC was identified.

Endoscopic Ultrasound-Guided Needle-Based Confocal Endomicroscopy as a Diagnostic Imaging Biomarker for Intraductal Papillary Mucinous Neoplasms.

Pancreatic cancer is on track to become the second leading cause of cancer-related deaths by 2030, yet there is a lack of accurate diagnostic tests for early detection. Intraductal papillary mucinous neoplasms (IPMNs) are precursors to pancreatic cancer and are increasingly being detected. Despite the development and refinement of multiple guidelines, diagnosing high-grade dysplasia or cancer in IPMNs using clinical, radiologic, endosonographic, and cyst fluid features still falls short in terms of accuracy, leading to both under- and overtreatment. EUS-guided needle-based confocal laser endomicroscopy (nCLE) is a novel technology that allows real-time optical biopsies of pancreatic cystic lesions. Emerging data has demonstrated that EUS-nCLE can diagnose and risk stratify IPMNs more accurately than conventional diagnostic tools. Implementing EUS-nCLE in clinical practice can potentially improve early diagnosis of pancreatic cancer, reduce unnecessary surgeries of IPMNs with low-grade dysplasia, and advance the field of digital pathomics. In this review, we summarize the current evidence that supports using EUS-nCLE as a diagnostic imaging biomarker for diagnosing IPMNs and for risk stratifying their degree of neoplasia. Moreover, we will present emerging data on the role of adding artificial intelligence (AI) algorithms to nCLE and integrating novel fluid biomarkers into nCLE.

How a significant increase in survival in pancreatic cancer is achieved. The role of nutritional status and supportive care: A review

Pancreatic cancer (PC) is a serious public health problem. The mortality rate of patients with PC remains one of the highest among cancers. Early diagnosis of PC is challenging, so it is often diagnosed in the later stages. Current treatment approaches, including surgery, neoadjuvant and adjuvant chemotherapy, chemoradiotherapy, and supportive care, have demonstrated improved outcomes. A significant problem remains exocrine pancreatic insufficiency (EPI) in patients with PC, which requires enzyme replacement therapy. However, this is not given due attention in the Russian literature. This review addresses the survival trends of patients with PC, current therapies, and enzyme replacement therapy as an integral part of supportive care and improvement of nutritional status; also, the issues of routing patients with PC are addressed. It is emphasized that the diagnosis and treatment of EPI are mandatory to improve and maintain the nutritional status and quality of life; failure to treat EPI renders antitumor treatment ineffective.

APR-246 enhances the anticancer effect of doxorubicin against p53-mutant AsPC-1 pancreatic cancer cells

Pancreatic cancer (PC) is a disease with a high mortality rate, and the early diagnosis of PC is still challenging. The 5-year relative survival rate remains below 8%, and therapeutic strategies are ineffective at increasing patient survival rates. In PC cells, resistance to treatment has been associated with genetic changes that initiate aberrant developmental signaling pathways; therefore, new strategies for treating PC are warranted. Here, we sought to explore the small molecule p53 reactivator, APR-246, either as a mono-drug or combined with doxorubicin (Doxo) in the p53 mutant (p53-mut) AsPC-1 PC cell line. Cytotoxicity of APR-246 and Doxo was studied alone or in combination with the AsPC-1 cell line with MTT assay. CalcuSyn software was used to calculate the combination index (CI), which was based on the Chou and Talalay technique. Apoptosis was assessed after staining with Acridine orange/ethidium bromide (AO/EB). Morphological alterations were examined under a 200x magnification by an inverted microscope. Gene expression was quantified using quantitative reverse transcription PCR assay (RT-qPCR). APR-246 alone exhibited modest anti-proliferative effects, while Doxo combined with APR-246 showed synergistic effects associated with morphological change; no more enhanced apoptosis was observed, as supported by NOXA gene levels. In conclusion, APR-246, whether monotherapy or combined with Doxo, was hindering the proliferation of p53-mut AsPC-1 PC cells, and further in vitro and in vivo research is warranted.

Editorial: Pancreatic cystic lesions: aiding in the early diagnosis of pancreatic cancer

Editorial: Pancreatic cystic lesions: aiding in the early diagnosis of pancreatic cancer

Pancreatic cancer with multiple liver metastasis complicating multi organ infarcts from Marantic endocarditis and Trousseau’s syndrome.

Background Marantic endocarditis and Trousseau’s syndrome are historically linked with pancreatic cancer. The patient had catastrophic embolic events which caused multi organ infarct due to underlying advanced pancreatic cancer. Doctors should be aware of early signs and symptoms of pancreatic cancer and conduct necessary clinical assessment and investigations which can prevent any severe complications. Case presentation A middle-aged healthy gentleman who presented with left sided weakness, slurred speech, left sided sensory neglect, mild headache and fever. CT head showed acute right sided temporal ischaemic changes with subsequent MRI head showed multiple infarcts in the brain. Transoesophageal echocardiogram reported possible MV vegetation. MRI of the liver was done due to deranged LFTs which showed multiple liver metastasis with primary mass in the tail of pancreas. Left leg ultrasound doppler showed a large left leg deep vein thrombosis involving femoral vein as he complained of left leg pain for 2 months and went to GP with right leg pain 3 months ago which was treated as right ankle sprain. The clinical events explained that most likely he developed Trousseau&amp;apos;s syndrome 3 months ago and the embolic phenomena was due to rare Marantic endocarditis (Non-bacterial thrombotic endocarditis) secondary to pancreatic cancer. Ultrasound guided biopsy of liver metastasis was planned, but sadly the patient passed away the next day. Conclusion The diagnosis of early pancreatic cancer with imaging poses a challenge sometimes, therefore the pancreatic protocol of CT or MRI is the first line investigation. Moreover, the cause of unprovoked thrombo-embolism should be investigated to rule out any underlying malignancy.

Advancements in SERS‐based biological detection and its application and perspectives in pancreatic cancer

AbstractSurface‐enhanced Raman spectroscopy (SERS) has become an essential bio‐detection technique. Due to its high sensitivity, good signal specificity, and resistance to photobleaching, SERS has been widely used in biomedical research fields such as molecular imaging, tumor diagnosis, and drug monitoring. This review focuses on the progress of SERS in biomedical applications. We first introduce the basic principle of SERS and the progress of substrate research. Then, we summarize the latest research progress on SERS in drug monitoring, cell and exosome detection, tumor imaging, and detection platforms combining microfluidic and lateral flow technologies. Subsequently, the applied research of SERS in early diagnosis of pancreatic cancer and drug efficacy monitoring is described. Finally, the future development direction and possible challenges of SERS in tumor diagnosis and treatment are proposed.

Non-invasive biomarkers for early diagnosis of pancreatic cancer risk: metabolite genomewide association study based on the KCPS-II cohort

BackgroundPancreatic cancer is a lethal disease with a high mortality rate. The difficulty of early diagnosis is one of its primary causes. Therefore, we aimed to discover non-invasive biomarkers that facilitate the early diagnosis of pancreatic cancer risk.MethodsThe study subjects were randomly selected from the Korean Cancer Prevention Study-II and matched by age, sex, and blood collection point [pancreatic cancer incidence (n = 128) vs. control (n = 256)]. The baseline serum samples were analyzed by non-targeted metabolomics, and XGBoost was used to select significant metabolites related to pancreatic cancer incidence. Genomewide association study for the selected metabolites discovered valuable single nucleotide polymorphisms (SNPs). Moderation and mediation analysis were conducted to explore the variables related to pancreatic cancer risk.ResultsEleven discriminant metabolites were selected by applying a cut-off of 4.0 in XGBoost. Five SNP presented significance in metabolite-GWAS (p ≤ 5 × 10–6) and logistic regression analysis. Among them, the pair metabolite of rs2370981, rs55870181, and rs72805402 displayed a different network pattern with clinical/biochemical indicators on comparison with allelic carrier and non-carrier. In addition, we demonstrated the indirect effect of rs59519100 on pancreatic cancer risk mediated by γ-glutamyl tyrosine, which affects the smoking status. The predictive ability for pancreatic cancer on the model using five SNPs and four pair metabolites with the conventional risk factors was the highest (AUC: 0.738 [0.661–0.815]).ConclusionsSignatures involving metabolites and SNPs discovered in the present research may be closely associated with the pathogenesis of pancreatic cancer and for use as predictive biomarkers allowing early pancreatic cancer diagnosis and therapy.

Predicting Pancreatic Cancer using AIML Techniques

Abstract: The late diagnosis and low survival rates associated with pancreatic cancer make it a substantial global health challenge to detect at an early stage. The aim of this study is to explore the potential of deep learning methods in enhancing the early identification of pancreatic cancer. To build a convolutional neural network (CNN) model to detect early indications of pancreatic malignancies, we employed medical images from a computed tomography (CT) scan dataset. By illuminating the potential of deep learning as a tool, our discoveries bring hope for enhanced patient survival rates by aiding in the early diagnosis of pancreatic cancer. The importance of artificial intelligence in medical imaging and its revolutionary effects on cancer diagnosis are emphasized by this research. Pancreatic Tumor stands as a global leader in responsible death rates caused by cancer. Curing pancreatic cancer is possible when detected early

Genomic insights and prognostic significance of novel biomarkers in pancreatic ductal adenocarcinoma: A comprehensive analysis

Pancreatic ductal adenocarcinoma (PDAC) is a highly prevalent digestive system malignancy, with a significant impact on public health, especially in the elderly population. The advent of the Human Genome Project has opened new avenues for precision medicine, allowing researchers to explore genetic markers and molecular targets for cancer diagnosis and treatment. Despite significant advances in genomic research, early diagnosis of pancreatic cancer remains elusive due to the lack of highly sensitive and specific markers. Therefore, there is a need for in-depth research to identify more precise and reliable diagnostic markers for pancreatic cancer. In this study, we utilized a combination of public databases from different sources to meticulously screen genes associated with prognosis in pancreatic cancer. We used gene differential analysis, univariate cox regression analysis, least absolute selection and shrinkage operator (LASSO) regression, and multivariate cox regression analysis to identify genes associated with prognosis. Subsequently, we constructed a scoring system, validated its validity using survival analysis and ROC analysis, and further confirmed its reliability by nomogram and decision curve analysis (DCA). We evaluated the diagnostic value of this scoring system for pancreatic cancer prognosis and validated the function of the genes using single cell data analysis. Our analysis identifies six genes, including GABRA3, IL20RB, CDK1, GPR87, TTYH3, and KCNA2, that were strongly associated with PDAC prognosis. Clinical prognostic models based on these genes showed strong predictive power not only in the training set but also in external datasets. Functional enrichment analysis revealed significant differences between high- and low-risk groups mainly in immune-related functions. Additionally, we explored the potential of the risk score as a marker for immunotherapy response and identified key factors within the tumor microenvironment. The single-cell RNA sequencing analysis further enriched our understanding of cell clusters and six hub genes expressions. This comprehensive investigation provides valuable insights into pancreatic PDAC and its intricate immune landscape. The identified genes and their functional significance underscore the importance of continued research into improving diagnosis and treatment strategies for PDAC.

Integrating a microRNA signature as a liquid biopsy-based tool for the early diagnosis and prediction of potential therapeutic targets in pancreatic cancer

IntroductionPancreatic cancer is a highly aggressive cancer, and early diagnosis significantly improves patient prognosis due to the early implementation of curative-intent surgery. Our study aimed to implement machine-learning algorithms to aid in early pancreatic cancer diagnosis based on minimally invasive liquid biopsies.Materials and methodsThe analysis data were derived from nine public pancreatic cancer miRNA datasets and two sequencing datasets from 26 pancreatic cancer patients treated in our medical center, featuring small RNAseq data for patient-matched tumor and non-tumor samples and serum. Upon batch-effect removal, systematic analyses for differences between paired tissue and serum samples were performed. The robust rank aggregation (RRA) algorithm was used to reveal feature markers that were co-expressed by both sample types. The repeatability and real-world significance of the enriched markers were then determined by validating their expression in our patients’ serum. The top candidate markers were used to assess the accuracy of predicting pancreatic cancer through four machine learning methods. Notably, these markers were also applied for the identification of pancreatic cancer and pancreatitis. Finally, we explored the clinical prognostic value, candidate targets and predict possible regulatory cell biology mechanisms involved.ResultsOur multicenter analysis identified hsa-miR-1246, hsa-miR-205-5p, and hsa-miR-191-5p as promising candidate serum biomarkers to identify pancreatic cancer. In the test dataset, the accuracy values of the prediction model applied via four methods were 94.4%, 84.9%, 82.3%, and 83.3%, respectively. In the real-world study, the accuracy values of this miRNA signatures were 82.3%, 83.5%, 79.0%, and 82.2. Moreover, elevated levels of these miRNAs were significant indicators of advanced disease stage and allowed the discrimination of pancreatitis from pancreatic cancer with an accuracy rate of 91.5%. Elevated expression of hsa-miR-205-5p, a previously undescribed blood marker for pancreatic cancer, is associated with negative clinical outcomes in patients.ConclusionA panel of three miRNAs was developed with satisfactory statistical and computational performance in real-world data. Circulating hsa-miRNA 205-5p serum levels serve as a minimally invasive, early detection tool for pancreatic cancer diagnosis and disease staging and might help monitor therapy success.

Addressing the unmet clinical need for low-volume assays in early diagnosis of pancreatic cancer

The incidental detection of pancreatic cysts, an opportunity for the early detection of pancreatic cancer, is increasing, owing to an aging population and improvements in imaging technology. The classification of pancreatic cystic precursors currently relies on imaging and cyst fluid evaluations, including cytology and protein and genomic analyses. However, there are persistent limitations that obstruct the accuracy and quality of information for clinicians, including the limited volume of the complex, often acellular, and proteinaceous milieu that comprises pancreatic cyst fluid. The constraints of currently available clinical assays lead clinicians to the subjective and inconsistent application of diagnostic tools, which can contribute to unnecessary surgery and missed pancreatic cancers. Herein, we describe the pathway toward pancreatic cyst classification and diagnosis, the volume requirements for several clinically available diagnostic tools, and some analytical and diagnostic limitations for each assay. We then discuss current and future work on novel markers and methods, and how to expand the utility of clinical pancreatic cyst fluid samples. Results of ongoing studies applying SERS as a detection mode suggest that 50 µL of pancreatic cyst fluid is more than sufficient to accurately rule out non-mucinous pancreatic cysts with no malignant potential from further evaluation. This process is expected to leave sufficient fluid to analyze a follow-up, rule-in panel of markers currently in development that can stratify grades of dysplasia in mucinous pancreatic cysts and improve clinical decision-making.

Earlier Diagnosis of Pancreatic Cancer: Is It Possible?

Pancreatic ductal adenocarcinoma has a very high mortality rate which has been only minimally improved in the last 30 years. This high mortality is closely related to late diagnosis, which is usually made when the tumor is large and has extensively infiltrated neighboring tissues or distant metastases are already present. This is a paradoxical situation for a tumor that requires nearly 15 years to develop since the first founding mutation. Response to chemotherapy under such late circumstances is poor, resistance is frequent, and prolongation of survival is almost negligible. Early surgery has been, and still is, the only approach with a slightly better outcome. Unfortunately, the relapse percentage after surgery is still very high. In fact, early surgery clearly requires early diagnosis. Despite all the advances in diagnostic methods, the available tools for improving these results are scarce. Serum tumor markers permit a late diagnosis, but their contribution to an improved therapeutic result is very limited. On the other hand, effective screening methods for high-risk populations have not been fully developed as yet. This paper discusses the difficulties of early diagnosis, evaluates whether the available diagnostic tools are adequate, and proposes some simple and not-so-simple measures to improve it.

Early Diagnosis of Pancreatic Cancer via Machine Learning Analysis of a National Electronic Medical Record Database.

Pancreatic cancer (PaC) is often diagnosed at advanced stages, resulting in one of the lowest survival rates among patients with cancer. The purpose of this study was to investigate whether machine learning (ML) models can predict with high sensitivity and specificity an increased risk for PaC ahead of clinical diagnosis. Optum deidentified electronic health record (EHR) data set was used to extract 1-year data for each patient and to sample for PaC diagnosis, the number of interactions with the health care system, and unique demographic and clinical features. Data for patients with PaC diagnosis were collected between 1 and 2 years before the diagnosis. Standard binary classification ML models were used on training and testing data sets. Data analyses were performed using the scikit-learn package version 1.0.1. The data set consisted of 18,987 patient EHRs collected between December 31, 2007, and December 31, 2017. EHRs with 10 unique features and at least three health care interactions were used for model training (N = 15,189; n = 8,438 [56%] with PaC) and testing (N = 3,798; n = 2,127 [56%] with PaC). The ensemble model achieved an AUC of 0.89, a sensitivity of 85.61%, and a specificity of 76.18% on the testing data set and produced superior results compared with other binary classifiers. Increasing unique health care interactions to nine failed to improve the AUC score. When the testing data set was enlarged to 5,696 patients, the ensemble model achieved an AUC of 0.92 and a specificity of 93.21%, but the sensitivity was compromised. The ensemble model exceeded the state-of-the-art level of performance for prediction of PaC ahead of clinical diagnosis with a minimal clinically guided input, providing a potential strategy for selection of high-risk patients for further screening.

Proposing new early detection indicators for pancreatic cancer: Combining machine learning and neural networks for serum miRNA-based diagnostic model.

Pancreatic cancer (PC) is a lethal malignancy that ranks seventh in terms of global cancer-related mortality. Despite advancements in treatment, the five-year survival rate remains low, emphasizing the urgent need for reliable early detection methods. MicroRNAs (miRNAs), a group of non-coding RNAs involved in critical gene regulatory mechanisms, have garnered significant attention as potential diagnostic and prognostic biomarkers for pancreatic cancer (PC). Their suitability stems from their accessibility and stability in blood, making them particularly appealing for clinical applications. In this study, we analyzed serum miRNA expression profiles from three independent PC datasets obtained from the Gene Expression Omnibus (GEO) database. To identify serum miRNAs associated with PC incidence, we employed three machine learning algorithms: Support Vector Machine-Recursive Feature Elimination (SVM-RFE), Least Absolute Shrinkage and Selection Operator (LASSO), and Random Forest. We developed an artificial neural network model to assess the accuracy of the identified PC-related serum miRNAs (PCRSMs) and create a nomogram. These findings were further validated through qPCR experiments. Additionally, patient samples with PC were classified using the consensus clustering method. Our analysis revealed three PCRSMs, namely hsa-miR-4648, hsa-miR-125b-1-3p, and hsa-miR-3201, using the three machine learning algorithms. The artificial neural network model demonstrated high accuracy in distinguishing between normal and pancreatic cancer samples, with verification and training groups exhibiting AUC values of 0.935 and 0.926, respectively. We also utilized the consensus clustering method to classify PC samples into two optimal subtypes. Furthermore, our investigation into the expression of PCRSMs unveiled a significant negative correlation between the expression of hsa-miR-125b-1-3p and age. Our study introduces a novel artificial neural network model for early diagnosis of pancreatic cancer, carrying significant clinical implications. Furthermore, our findings provide valuable insights into the pathogenesis of pancreatic cancer and offer potential avenues for drug screening, personalized treatment, and immunotherapy against this lethal disease.