Management of advanced high grade gastroenteropancreatic neuroendocrine neoplasms (GEP-NENs): comprehensive review of the current literature.

The classification and management of neuroendocrine neoplasms (NENs) arising in the tubular gastrointestinal (GI) tract and pancreas have significantly evolved over the last decades. In the latest WHO classification published in 2022, NENs are separated regardless of their primary origin into two main groups: well-differentiated neuroendocrine tumors (NETs) and poorly differentiated neuroendocrine carcinomas (NECs). The substantial changes in the grading system changed the definition of grade 3 to include high-grade well-differentiated NETs (G3-NETs), and poorly differentiated NECs (-NECs). Although these two subgroups are considered high grades with Ki-67 >20%, they have different genomic profiles, prognosis, and clinical behavior, which critically influence their treatment strategies. The available clinical trial data to guide therapy of these high-grade subgroups are extremely limited, which impacts their management. In this review, we will summarize the current advances in the multidisciplinary approach for the management of high-grade gastroenteropancreatic NENs (GEP-NENs) including G3-NETs and NECs.

Pancreatic carcinoma, its variants and precursors: Overview of the current WHO classification.

Pancreatic carcinoma is a relatively common malignant tumor with increasing incidence and mortality. The tumor is usually diagnosed at an advanced stage and generally has a poor prognosis, with only 5% of patients surviving 5 years from the time of diagnosis. The stage of the disease at the time of diagnosis is a crucial factor for the prognosis; 25% of patients with localized tumors survive 3 years from diagnosis, compared to only 1% of those with generalized tumors. Radical surgical removal of the tumor (partial or total pancreatectomy) is a key factor in improving survival. Therefore, the topic is highly relevant to surgeons. Statistics on pancreatic carcinoma mainly focus on ductal adenocarcinoma, which is the most common and least favorable malignant tumor of the pancreas. This review focuses on ductal adenocarcinoma, its variants, and precancerous lesions. The article summarizes information from the latest WHO classification of 2019, which was released 11 years after the previous edition. Compared to the previous version, this new WHO classification introduced rather minor changes in the field of ductal adenocarcinoma. The delineation of rare variants of ductal adenocarcinoma is justified based on genetic and morphological similarities and clinical relevance, as individual subtypes significantly differ in prognosis. The article also includes a description of macroscopic and microscopic precursors of ductal adenocarcinoma and their definitions. Genetic and immunohistochemical differential diagnostic aspects are briefly discussed, as these are more relevant to pathologists than to surgeons.

PanIN or IPMN? Redefining Lesion Size in 3 Dimensions.

Pancreatic ductal adenocarcinoma (PDAC) develops from 2 known precursor lesions: a majority (∼85%) develops from pancreatic intraepithelial neoplasia (PanIN), and a minority develops from intraductal papillary mucinous neoplasms (IPMNs). Clinical classification of PanIN and IPMN relies on a combination of low-resolution, 3-dimensional (D) imaging (computed tomography, CT), and high-resolution, 2D imaging (histology). The definitions of PanIN and IPMN currently rely heavily on size. IPMNs are defined as macroscopic: generally >1.0 cm and visible in CT, and PanINs are defined as microscopic: generally <0.5 cm and not identifiable in CT. As 2D evaluation fails to take into account 3D structures, we hypothesized that this classification would fail in evaluation of high-resolution, 3D images. To characterize the size and prevalence of PanINs in 3D, 47 thick slabs of pancreas were harvested from grossly normal areas of pancreatic resections, excluding samples from individuals with a diagnosis of an IPMN. All patients but one underwent preoperative CT scans. Through construction of cellular resolution 3D maps, we identified >1400 ductal precursor lesions that met the 2D histologic size criteria of PanINs. We show that, when 3D space is considered, 25 of these lesions can be digitally sectioned to meet the 2D histologic size criterion of IPMN. Re-evaluation of the preoperative CT images of individuals found to possess these large precursor lesions showed that nearly half are visible on imaging. These findings demonstrate that the clinical classification of PanIN and IPMN fails in evaluation of high-resolution, 3D images, emphasizing the need for re-evaluation of classification guidelines that place significant weight on 2D assessment of 3D structures.

Molecular Classification of Gastrointestinal and Pancreatic Neuroendocrine Neoplasms: Are We Ready for That?

In the last two decades, the increasing availability of technologies for molecular analyses has allowed an insight in the genomic alterations of neuroendocrine neoplasms (NEN) of the gastrointestinal tract and pancreas. This knowledge has confirmed, supported, and informed the pathological classification of NEN, clarifying the differences between neuroendocrine carcinomas (NEC) and neuroendocrine tumors (NET) and helping to define the G3 NET category. At the same time, the identification genomic alterations, in terms of gene mutation, structural abnormalities, and epigenetic changes differentially involved in the pathogenesis of NEC and NET has identified potential molecular targets for precision therapy. This review critically recapitulates the available molecular features of digestive NEC and NET, highlighting their correlates with pathological aspects and clinical characteristics of these neoplasms and revising their role as predictive biomarkers for targeted therapy. In this context, the feasibility and applicability of a molecular classification of gastrointestinal and pancreatic NEN will be explored.

An Illustrated Review of the Recent 2019 World Health Organization Classification of Neuroendocrine Neoplasms: A Radiologic and Pathologic Correlation.

ABSTRACT: Recent advances in molecular pathology and an improved understanding of the etiology of neuroendocrine neoplasms (NENs) have given rise to an updated World Health Organization classification. Since gastroenteropancreatic NENs (GEP-NENs) are the most common forms of NENs and their incidence has been increasing constantly, they will be the focus of our attention. Here, we review the findings at the foundation of the new classification system, discuss how it impacts imaging research and radiological practice, and illustrate typical and atypical imaging and pathological findings. Gastroenteropancreatic NENs have a highly variable clinical course, which existing classification schemes based on proliferation rate were unable to fully capture. While well- and poorly differentiated NENs both express neuroendocrine markers, they are fundamentally different diseases, which may show similar proliferation rates. Genetic alterations specific to well-differentiated neuroendocrine tumors graded 1 to 3 and poorly differentiated neuroendocrine cancers of small cell and large-cell subtype have been identified. The new tumor classification places new demands and creates opportunities for radiologists to continue providing the clinically most relevant report and on researchers to design projects, which continue to be clinically applicable.

Is it resectable? A survey regarding tumor classification and vascular involvement.

INTRODUCTION: The anatomic location of the pancreas can result in involvement of major vasculature, which may act as a contraindication to resection. Several classification systems have been developed. We sought to discover the variations in the HPB community determining PDAC resectability. METHODS: The multiple-choice survey was distributed to all full members of the IHPBA. Questions were asked regarding demographics and clinical scenarios regarding tumor resectability. RESULTS: 164 responses were submitted. Most of the respondents were male and had been in practice for over 10 years. The median age range was 40-50 years old. Most practiced in either Asia (n = 57,35.9%), North America (n = 52,32.7%), or Europe (n = 32,20.1%). Classification systems used to determine resectability were: NCCN (n = 42,26.3%), JPS (n = 35,21.9%), International consensus (n = 33,20.6%), AHPBA/SSO (n = 23,14.4%), Alliance (n = 3,1.9%), and other/no-classification (n = 23,14.5%). There was significant variation in the frequency of the most common answer within the scenarios (84.7%-33.5%). Participant concordance with their stated classification system found a median rate of 62.5%. Participant decision of tumor resectability was not dependent on their adopted classification system. CONCLUSION: When classifying PDAC resectability, there is significant variation between surgeons as to how they would classify a specific tumour, independent of the classification system they use. In addition, surgeons do not show high concordance with the definitions within that classification system.

DNA Methylation Profiling Enables Accurate Classification of Nonductal Primary Pancreatic Neoplasms.

BACKGROUND & AIMS: Cytologic and histopathologic diagnosis of non-ductal pancreatic neoplasms can be challenging in daily clinical practice, whereas it is crucial for therapy and prognosis. The cancer methylome is successfully used as a diagnostic tool in other cancer entities. Here, we investigate if methylation profiling can improve the diagnostic work-up of pancreatic neoplasms. METHODS: DNA methylation data were obtained for 301 primary tumors spanning 6 primary pancreatic neoplasms and 20 normal pancreas controls. Neural Network, Random Forest, and extreme gradient boosting machine learning models were trained to distinguish between tumor types. Methylation data of 29 nonpancreatic neoplasms (n = 3708) were used to develop an algorithm capable of detecting neoplasms of non-pancreatic origin. RESULTS: After benchmarking 3 state-of-the-art machine learning models, the random forest model emerged as the best classifier with 96.9% accuracy. All classifications received a probability score reflecting the confidence of the prediction. Increasing the score threshold improved the random forest classifier performance up to 100% with 87% of samples with scores surpassing the cutoff. Using a logistic regression model, detection of nonpancreatic neoplasms achieved an area under the curve of >0.99. Analysis of biopsy specimens showed concordant classification with their paired resection sample. CONCLUSIONS: Pancreatic neoplasms can be classified with high accuracy based on DNA methylation signatures. Additionally, non-pancreatic neoplasms are identified with near perfect precision. In summary, methylation profiling can serve as a valuable adjunct in the diagnosis of pancreatic neoplasms with minimal risk for misdiagnosis, even in the pre-operative setting.

Evidence for molecular subtyping in pancreatic ductal adenocarcinoma: a systematic review.

BACKGROUND: Pancreatic Ductal Adenocarcinoma (PDAC) patients exhibit varied responses to multimodal therapy. RNA gene sequencing has unravelled distinct tumour biology subtypes, forming the focus of this review exploring its impact on survival outcomes. METHODS: A systematic search across PubMed, Medline, Embase, and CINAHL databases targeted studies assessing long-term overall and disease-free survival in PDAC patients with molecular subtyping. RESULTS: Fifteen studies including 2731 patients were identified. Molecular subtyping was performed by RNA sequencing and Immunohistochemistry in 14 studies and by Mass Spectrometry in 1 study. Two main tumour subtypes were identified (classical and basal-like or squamous) with basal like associated with poorer outcomes. Further subtypes were identified in individual studies. Superior survival was seen with classical subtype in all other analyses that compared the classical and basal subtypes. High risk stromal subtypes were identified on further analysis of the stroma and were associated with a worse survival independent of the tumour subtype. CONCLUSION: Molecular subtyping of PDAC specimens can identify patients with high-risk tumour biology and poor survival outcomes. Routine subtyping is limited by the cost of RNA sequencing and the volume of raw data generated which has made its translation into routine clinical practice difficult.

Risk factors for gastroenteropancreatic neuroendocrine neoplasms (GEP-NENs): a three-centric case-control study.

PURPOSE: Risk factors for sporadic GEP-NENs are still not well defined. To identify the main clinical risk factors represents the aim of this study performed by three Italian referral centers for NENs. METHODS: We performed a retrospective case-control study including 148 consecutive sporadic GEP-NENs and 210 age- and sex-matched controls. We collected data on clinical features, cancer family history and other potential risk factors. RESULTS: Mean age was 58.3 ± 15.8 years; 50% males, primary site was pancreas (50.7%), followed by ileum (22.3%). The 62.8% and 29.1% of cases were G1 and G2, respectively; the 40% had locally advanced or metastatic disease at diagnosis. Independent risk factors for GEP-NENs were: family history of non-neuroendocrine GEP cancer (OR 2.16, 95% CI 1.31-3.55, p = 0.003), type 2 diabetes mellitus (T2DM) (OR 2.5, 95% CI 1.39-4.51, p = 0.002) and obesity (OR 1.88, 95% CI 1.18-2.99, p = 0.007). In the T2DM subjects, metformin use was a protective factor (OR 0.28, 95% CI 0.08-0.93, p = 0.049). T2DM was also associated with a more advanced (OR 2.39, 95% CI 1.05-5.46, p = 0.035) and progressive disease (OR 2.47, 95% CI 1.08-5.34, p = 0.03). Stratifying cases by primary site, independent risk factors for pancreatic NENs were T2DM (OR 2.57, 95% CI 1.28-5.15, p = 0.008) and obesity (OR 1.98, 95% CI 1.11-3.52, p = 0.020), while for intestinal NENs family history of non-neuroendocrine GEP cancer (OR 2.46, 95% CI 1.38-4.38, p = 0.003) and obesity (OR 1.90, 95% CI 1.08-3.33, p = 0.026). CONCLUSION: This study reinforces a role for family history of non-neuroendocrine GEP cancer, T2DM and obesity as independent risk factors for GEP-NENs and suggests a role of metformin as a protective factor in T2DM subjects. If confirmed, these findings could have a significant impact on prevention strategies for GEP-NENs.

Proteogenomic analysis of pancreatic cancer subtypes.

Pancreatic cancer remains a significant public health problem with an ever-rising incidence of disease. Cancers of the pancreas are characterised by various molecular aberrations, including changes in the proteomics and genomics landscape of the tumour cells. Therefore, there is a need to identify the proteomic landscape of pancreatic cancer and the specific genomic and molecular alterations associated with disease subtypes. Here, we carry out an integrative bioinformatics analysis of The Cancer Genome Atlas dataset, including proteomics and whole-exome sequencing data collected from pancreatic cancer patients. We apply unsupervised clustering on the proteomics dataset to reveal the two distinct subtypes of pancreatic cancer. Using functional and pathway analysis based on the proteomics data, we demonstrate the different molecular processes and signalling aberrations of the pancreatic cancer subtypes. In addition, we explore the clinical characteristics of these subtypes to show differences in disease outcome. Using datasets of mutations and copy number alterations, we show that various signalling pathways previously associated with pancreatic cancer are altered among both subtypes of pancreatic tumours, including the Wnt pathway, Notch pathway and PI3K-mTOR pathways. Altogether, we reveal the proteogenomic landscape of pancreatic cancer subtypes and the altered molecular processes that can be leveraged to devise more effective treatments.

Intraductal Papillary Mucinous Carcinoma Versus Conventional Pancreatic Ductal Adenocarcinoma: A Comprehensive Review of Clinical-Pathological Features, Outcomes, and Molecular Insights.

Intraductal papillary mucinous neoplasms (IPMN) are common and one of the main precursor lesions of pancreatic ductal adenocarcinoma (PDAC). PDAC derived from an IPMN is called intraductal papillary mucinous carcinoma (IPMC) and defines a subgroup of patients with ill-defined specificities. As compared to conventional PDAC, IPMCs have been associated to clinical particularities and favorable pathological features, as well as debated outcomes. However, IPMNs and IPMCs include distinct subtypes of precursor (gastric, pancreato-biliary, intestinal) and invasive (tubular, colloid) lesions, also associated to specific characteristics. Notably, consistent data have shown intestinal IPMNs and associated colloid carcinomas, defining the "intestinal pathway", to be associated with less aggressive features. Genomic specificities have also been uncovered, such as mutations of the GNAS gene, and recent data provide more insights into the mechanisms involved in IPMCs carcinogenesis. This review synthetizes available data on clinical-pathological features and outcomes associated with IPMCs and their subtypes. We also describe known genomic hallmarks of these lesions and summarize the latest data about molecular processes involved in IPMNs initiation and progression to IPMCs. Finally, potential implications for clinical practice and future research strategies are discussed.

Construction of a novel risk model based on the random forest algorithm to distinguish pancreatic cancers with different prognoses and immune microenvironment features.

Immune-related long noncoding RNAs (irlncRNAs) are actively involved in regulating the immune status. This study aimed to establish a risk model of irlncRNAs and further investigate the roles of irlncRNAs in predicting prognosis and the immune landscape in pancreatic cancer. The transcriptome profiles and clinical information of 176 pancreatic cancer patients were retrieved from The Cancer Genome Atlas (TCGA). Immune-related genes (irgenes) downloaded from ImmPort were used to screen 1903 immune-related lncRNAs (irlncRNAs) using Pearson's correlation analysis (R > 0.5; p < 0.001). Random survival forest (RSF) and survival tree analysis showed that 9 irlncRNAs were highly correlated with overall survival (OS) according to the variable importance (VIMP) and minimal depth. Next, Cox regression analysis was used to establish a risk model with 3 irlncRNAs (LINC00462, LINC01887, RP11-706C16.8) that was evaluated by Kaplan-Meier analysis, the areas under the curve (AUCs) of the receiver operating characteristics and the C-index. Additionally, we performed Cox regression analysis to establish the clinical prognostic model, which showed that the risk score was an independent prognostic factor (p < 0.001). A nomogram and calibration plots were drawn to visualize the clinical features. The Wilcoxon signed-rank test and Pearson's correlation analysis further explored the irlncRNA signatures and immune cell infiltration, as well as the immunotherapy response.

Pancreatic collision tumor of ductal adenocarcinoma and neuroendocrine tumor-A rare case report.

A collision tumor is composed of two adjacent histological distinct neoplasms without the histological admixture of cell types in the same organ or tissue. It is rare in pancreas. Herein we report an unusual case of a mixed malignant neuroendocrine tumor (NET) and ductal adenocarcinoma of pancreas in a 24 year old male who presented with history abdomen pain. A clinicoradiological diagnosis of chronic calcific pancreatitis with carcinoma body of pancreas was made. Distal pancreaticosplenectomy specimen showed a grey white, nodular growth measuring 2 x 2 x 1.2 cm on the cut surface of pancreas. Histopathology revealed a composite tumor consisting of ductal and neuroendocrine origin. Immunohistochemistry showed complementary staining for CK7 in adenocarcinoma and chromogranin A in NET areas confirming a collision tumor. Accurate evaluation of the radiologic pointers, histomorphologic evaluation to recognize and quantitate the individual components, appropriate immunohistochemical evaluation and correlation is essential for diagnosis.

De novo mutational signature discovery in tumor genomes using SparseSignatures.

Cancer is the result of mutagenic processes that can be inferred from tumor genomes by analyzing rate spectra of point mutations, or "mutational signatures". Here we present SparseSignatures, a novel framework to extract signatures from somatic point mutation data. Our approach incorporates a user-specified background signature, employs regularization to reduce noise in non-background signatures, uses cross-validation to identify the number of signatures, and is scalable to large datasets. We show that SparseSignatures outperforms current state-of-the-art methods on simulated data using a variety of standard metrics. We then apply SparseSignatures to whole genome sequences of pancreatic and breast tumors, discovering well-differentiated signatures that are linked to known mutagenic mechanisms and are strongly associated with patient clinical features.

High-Grade Gastroenteropancreatic Neuroendocrine Neoplasms and Improved Prognostic Stratification With the New World Health Organization 2019 Classification: A Validation Study From a Single-Institution Retrospective Analysis.

OBJECTIVES: There is a pressing need to develop clinical management pathways for grade 3 (G3) gastroenteropancreatic neuroendocrine neoplasms (GEP NEN). METHODS: We performed a retrospective study on patients with metastatic G3 GEP NEN. The relationship between baseline characteristics and progression-free survival and overall survival was analyzed using the Kaplan-Meier method. Univariate and multivariate analyses were performed using the Cox proportional hazards model. RESULTS: We included 142 patients (74 well-differentiated neuroendocrine tumors [WDNETs], 68 poorly differentiated neuroendocrine carcinomas [PDNECs]). Patients with WDNET had prolonged survival compared with PDNEC (median, 24 vs 15 months, P = 0.0001), which persisted in both pancreatic and nonpancreatic cohorts. Well-differentiated morphology, Ki-67 <50% and positive somatostatin receptor imaging were independently associated with prolonged survival. Of the subgroup treated with first-line platinum-based chemotherapy, response rates were favorable (partial response, 47%; stable disease, 30%); there was no significant difference in response rates nor progression-free survival between WDNET and PDNEC despite significantly prolonged overall survival in the WDNET cohort. CONCLUSIONS: Our study corroborates the knowledge of 2 prognostically distinct subgroups within the World Health Organization 2019 G3 GEP NEN population, observed in both pancreatic and nonpancreatic gastrointestinal cohorts. Definitive management pathways are needed to reflect the differences between G3 WDNET and PDNEC.

Genomic analysis of pancreatic cancer reveals 3 molecular subtypes with different clinical outcomes.

ABSTRACT: Pancreatic cancer has a very high mortality with a 5-year survival of <5%. The purpose of this study was to classify specific molecular subtypes associated with prognosis of pancreatic cancer using The Cancer Genome Atlas (TCGA) multiplatform genomic data.Multiplatform genomic data (N = 178), including gene expression, copy number alteration, and somatic mutation data, were obtained from cancer browser (https://genome-cancer.ucsc.edu, cohort: TCGA Pancreatic Cancer). Clinical data including survival results were analyzed. We also used validation cohort (GSE50827) to confirm the robustness of these molecular subtypes in pancreatic cancer.When we performed unsupervised clustering using TCGA gene expression data, we found three distinct molecular subtypes associated with different survival results. Copy number alteration and somatic mutation data showed different genomic patterns for these three subtypes. Ingenuity pathway analysis revealed that each subtype showed differentially altered pathways. Using each subtype-specific genes (200 were selected), we could predict molecular subtype in another cohort, confirming the robustness of these molecular subtypes of pancreatic cancer. Cox regression analysis revealed that molecular subtype is the only significant prognostic factor for pancreatic cancer (P = .042, 95% confidence interval 0.523-0.98).Genomic analysis of pancreatic cancer revealed 3 distinct molecular subtypes associated with different survival results. Using these subtype-specific genes and prediction model, we could predict molecular subtype associated with prognosis of pancreatic cancer.

Pancreatic Neuroendocrine Neoplasms in Multiple Endocrine Neoplasia Type 1.

Pancreatic neuroendocrine tumors (pNETs) are a rare group of cancers accounting for about 1-2% of all pancreatic neoplasms. About 10% of pNETs arise within endocrine tumor syndromes, such as Multiple Endocrine Neoplasia type 1 (MEN1). pNETs affect 30-80% of MEN1 patients, manifesting prevalently as multiple microadenomas. pNETs in patients with MEN1 are particularly difficult to treat due to differences in their growth potential, their multiplicity, the frequent requirement of extensive surgery, the high rate of post-operative recurrences, and the concomitant development of other tumors. MEN1 syndrome is caused by germinal heterozygote inactivating mutation of the MEN1 gene, encoding the menin tumor suppressor protein. MEN1-related pNETs develop following the complete loss of function of wild-type menin. Menin is a key regulator of endocrine cell plasticity and its loss in these cells is sufficient for tumor initiation. Somatic biallelic loss of wild-type menin in the neuroendocrine pancreas presumably alters the epigenetic control of gene expression, mediated by histone modifications and DNA hypermethylation, as a driver of MEN1-associated pNET tumorigenesis. In this light, epigenetic-based therapies aimed to correct the altered DNA methylation, and/or histone modifications might be a possible therapeutic strategy for MEN1 pNETs, for whom standard treatments fail.

Artificial neural networks for multi-omics classifications of hepato-pancreato-biliary cancers: towards the clinical application of genetic data.

PURPOSE: Several multi-omics classifications have been proposed for hepato-pancreato-biliary (HPB) cancers, but these classifications have not proven their role in the clinical practice and been validated in external cohorts. PATIENTS AND METHODS: Data from whole-exome sequencing (WES) of The Cancer Genome Atlas (TCGA) patients were used as an input for the artificial neural network (ANN) to predict the anatomical site, iClusters (cell-of-origin patterns) and molecular subtype classifications. The Ohio State University (OSU) and the International Cancer Genome Consortium (ICGC) patients with HPB cancer were included in external validation cohorts. TCGA, OSU and ICGC data were merged, and survival analyses were performed using both the 'classic' survival analysis and a machine learning algorithm (random survival forest). RESULTS: Although the ANN predicting the anatomical site of the tumour (i.e. cholangiocarcinoma, hepatocellular carcinoma of the liver, pancreatic ductal adenocarcinoma) demonstrated a low accuracy in TCGA test cohort, the ANNs predicting the iClusters (cell-of-origin patterns) and molecular subtype classifications demonstrated a good accuracy of 75% and 82% in TCGA test cohort, respectively. The random survival forest analysis and Cox' multivariable survival models demonstrated that models for HPB cancers that integrated clinical data with molecular classifications (iClusters, molecular subtypes) had an increased prognostic accuracy compared with standard staging systems. CONCLUSION: The analyses of genetic status (i.e. WES, gene panels) of patients with HPB cancers might predict the classifications proposed by TCGA project and help to select patients suitable to targeted therapies. The molecular classifications of HPB cancers when integrated with clinical information could improve the ability to predict the prognosis of patients with HPB cancer.

Methylation patterns partition pancreatic cancer into distinct prognostic subtypes.

In the initiation and progression of pancreatic cancer, DNA methylation plays a critical role. The present study attempts to explore specific prognosis subtypes based on DNA methylation data and develop an epigenetic signature to predict the overall survival (OS) of patients with pancreatic cancer.147 samples were included in the training cohort, whereas the validation cohort included 226 samples. The 298 OS-related methylation sites in the training cohort were selected for consensus clustering, and the authors identified three subtypes with a significant difference in prognosis. Cluster1 was associated with poor OS, low-grade disease and high lymph node involvement. In addition, we identified 33 specific methylation sites in Cluster1. Subsequently, we developed a robust qualitative signature consisting of 14 methylation sites to individually predict OS in the training cohort, and the predictive accuracy of this model was confirmed in the validation cohort. Functional enrichment analysis showed that the selected genes in the model were mainly enriched in known cancer-related pathways. Patients were divided into high- and low-risk groups by the model, and a significant difference in OS was observed between these groups. Classification based on the modeling of a specific DNA methylation site can reveal the heterogeneity of pancreatic cancer and provide guidance for clinicians in predicting the prognosis of pancreatic cancer and providing personalized treatment.