Standardized clinical management of pancreatic adenocarcinoma (PDAC) remains challenging and high-volume centers provide essential insights for establishing effective multimodal treatment approaches. This retrospective observational study evaluated the impact of standardized, multimodal clinical management on survival outcomes in patients with PDAC across all stages, based on NCCN guidelines resectability criteria, at a high-volume center. From 2019, 4143 patients were diagnosed with PDAC, with 3268 patients receiving further treatment, including surgical resection and/or systemic therapy. The median overall survival (OS) was 18.5 (95 %CI 17.5-19.4) months for the treated cohort and the 5-year survival rate reached 23.3 %. Patients who underwent surgical resection had significantly improved median OS compared to those who received non-surgical treatments (28.4 months vs. 13.0 months, P < 0.001), with corresponding 5-year survival rates of 31.6 % vs. 15.0 %. Moreover, the patients who received NAT followed by surgical resection had improved survival outcomes compared to those who underwent upfront surgical resection in both resectable (median OS: 37.5 months vs. 28.9 months, P < 0.01) and borderline resectable group (median OS: 31.8 months vs. 18.4 months, P < 0.001). This study demonstrated a 5-year survival rate exceeding 20 % for PDAC across all stages at our center. The application of evidence-based treatment strategies through the multidisciplinary team, accompanying with standardized and comprehensive therapeutic managements, high patient adherence, have been considered as critical determinants in enhancing therapeutic efficacy and improving long-term prognosis for patients with PDAC.

Pancreatic ductal adenocarcinoma (PDAC) is a highly aggressive malignancy associated with poor prognosis. Baicalein, a flavonoid extracted from the roots of Scutellaria baicalensis, traditionally used in Chinese medicine, has demonstrated potential in inhibiting cancer development and progression. However, its mechanism of action remains poorly understood, particularly regarding epigenetic gene regulation through m6A RNA methylation. In this study, three human PDAC cell lines and one nonmalignant cell line were employed. The effects of baicalein were examined using multiple assays, including RT-qPCR, MeRIP-qPCR, Western blotting, spheroid formation, RNA stability, and MTT, to evaluate cellular functions and m6A regulation. Baicalein significantly reduced cell viability, migration, invasion, and colony formation. It also downregulated FTO, an enzyme critical for m6A RNA demethylation. Knockdown of FTO replicated the effects of baicalein, underscoring its oncogenic role in PDAC. Bioinformatic analysis identified ZEB1-a key transcription factor in epithelial-to-mesenchymal transition-as an m6A-modified target regulated by FTO. Both baicalein treatment and FTO knockdown enhanced m6A modification and decreased ZEB1 mRNA stability, thereby suppressing stemness-related features. Rescue experiments further confirmed that baicalein disrupts the TGF-β/FTO/ZEB1 signaling axis, highlighting its therapeutic potential in PDAC. This study offers fundamental insights for the development of novel therapeutic strategies targeting PDAC.

Upper gastrointestinal (UGI) cancers, comprising malignancies of the esophagus, stomach, duodenum, pancreas, liver, biliary tract, and gallbladder, are the second leading cause of cancer-related mortality in the United States and are associated with significant comorbidities. Recent studies show a disproportionate rise in pancreatic and stomach cancer among young adults. This study aims to use a nationwide, population-based cohort to (i) evaluate the trend of all UGI cancer as an aggregate and (ii) examine the role of demographics, histology, and tumor stage in UGI cancer incidence among young adults.

Individuals diagnosed with UGI cancer in the United States from 2001 to 2019 were identified and obtained from the Surveillance, Epidemiology, and End Results-National Program of Cancer Registries database. The primary outcomes were incidence rates of UGI cancer (calculated per 100,000, age-adjusted to the year 2000 US population), stratified by sex and age (< 55 years for young adults and ≥ 55 years for older adults). Trends, annual percentage change, and average annual percentage change were calculated using the parametric method. Sensitivity analysis was performed according to primary site and histology; further analysis examining race and cancer stage was performed in the young adult subgroup.

A total of 2,333,161 patients with UGI cancer were identified. Most cases were male, and 14.3% were < 55 years of age. Incidence of UGI cancer increased most in women younger than 55 years, driven primarily by pancreatic and stomach cancers, as well as neuroendocrine tumor and gastrointestinal stromal tumor histology. African American race and localized tumors and malignancy with distant spread are also contributing to the disparate increase among young women. UGI mortality rates have not changed significantly in young adults.

The overall incidence rate of upper gastrointestinal cancer is increasing significantly in young women compared with men. Increased endoscopic procedures and disparate exposure to risk factors are likely contributing to these trends.

To assess the probability of being cured of pancreatic ductal adenocarcinoma (PDAC) by pancreatic surgery.

Statistical cure implies that a patient treated for a specific disease will have the same life expectancy as if he/she never had that disease.

Patients who underwent pancreatic resection for PDAC between 2010 and 2021 were retrospectively identified using a multi-institutional database. A nonmixture statistical cure model was applied to compare disease-free survival to the survival expected for a matched general population.

Among 2554 patients, either in the setting of upfront (n=1691) or neoadjuvant strategy (n=863), the cure model showed that the probability that surgery would offer the same life expectancy (and tumor-free) as the matched general population was 20.4% (95% CI: 18.3, 22.5). Cure likelihood reached the 95% of certainty (time to cure) after 5.3 years (95% CI: 4.7, 6.0). A preoperative model was developed based on tumor stage at diagnosis ( P =0.001), radiologic size ( P =0.001), response to chemotherapy ( P =0.007), American Society of Anesthesiology class ( P =0.001), and preoperative Ca19-9 ( P =0.001). A postoperative model with the addition of surgery type ( P =0.015), pathologic size ( P =0.001), tumor grading ( P =0.001), resection margin ( P =0.001), positive lymph node ratio ( P =0.001), and the receipt of adjuvant therapy ( P =0.001) was also developed.

Patients operated for PDAC can achieve a life expectancy similar to that of the general population, and the likelihood of cure increases with the passage of recurrence-free time. An online calculator was developed and available at https://aicep.website/?cff-form=15 .

Growing gastrointestinal cancers in the United States necessitate further research due to substantial health care and economic impacts. The aim of this study was to analyze trends and future projections for 5 major gastrointestinal cancers (colorectal, pancreatic, liver, stomach, and esophageal).

Data were sourced from the Surveillance, Epidemiology, and End Results database; National Center for Health Statistics; and Global Burden of Diseases databases. An age-period-cohort model using the Bayesian Information Criterion method was applied to project incidence and mortality rates to 2040.

Men consistently exhibited higher incidence and mortality rates across all gastrointestinal cancers, with significant variation across the 51 US states. From 2000 to 2020, colorectal cancer incidence and mortality rates declined across all racial groups, except for the incidence rates of American Indian and Alaska Native (AIAN) men, Hispanic men, and Hispanic women, which remained stable. Pancreatic cancer incidence increased across all groups except for AIAN men, while mortality rates rose only for White men and Hispanic women. Liver cancer incidence rose among AIAN men and White, AIAN, and Hispanic women, while mortality rates declined for most groups. Stomach cancer incidence and mortality either declined or stabilized, and esophageal cancer rates showed a general decline. By 2040, increases in incidence and mortality are projected for most gastrointestinal cancers, particularly in men.

Despite varied trends over the past 2 decades, an overall increase in gastrointestinal cancer incidence and mortality rates is anticipated in the next 20 years in the United States, underscoring the need for effective prevention and intervention strategies.

This study evaluated the impact of renal function impairment on long-term survival outcomes and adjuvant therapy in patients with pancreatic cancer undergoing pancreaticoduodenectomy (PD).

In this study, 264 patients who underwent PD for pancreatic head cancer between 2011 and 2021 were retrospectively analyzed. The patients were subsequently categorized into three groups according to the estimated glomerular filtration rate: normal group (> 90 mL/min/1.73 m2, n = 73), moderate group (45-90 mL/min/1.73 m2, n = 176), and severe group (< 45 mL/min/1.73 m2, n = 15). The primary outcomes evaluated were postoperative complications, overall survival (OS), and relapse-free survival (RFS). Additionally, the completion of adjuvant therapy and risk factors for adjuvant therapy discontinuation were analyzed.

The total proportion of patients with complications was significantly higher in the severe group (p = 0.008). The proportion of patients with severe complications (Clavien-Dindo classification ≥ IIIa) did not significantly differ between the chronic kidney disease (CKD) groups (p = 0.730). The proportion of patients in whom adjuvant therapy was completed was notably lower in the severe group (p = 0.011). Multiple logistic regression analysis revealed that CKD groups and hemoglobin levels ≤ 11.5 g/dL were independent predictors of adjuvant therapy completion failure (p = 0.016 and p = 0.016). There was no significant difference in the OS and RFS rates between the CKD groups (p = 0.499, p = 0.688).

In patients with pancreatic cancer and CKD, performing PD safely may be feasible; however, adjuvant therapy completion is challenging.

This is a case of a 67-year-old woman diagnosed with a 35-mm pancreatic body cancer with a chief complaint of epigastric discomfort. Computed tomography demonstrated invasion of the common hepatic artery, portal vein, and stomach, and chemotherapy was initiated for locally advanced pancreatic cancer. After 9 months of chemotherapy, the tumor remained stable on imaging, and the tumor markers were within the normal range. After additional chemoradiotherapy, the patient underwent a conversion surgery, a pancreaticoduodenectomy. Magnetic resonance cholangiopancreatography (MRCP) at the time of diagnosis demonstrated main pancreatic duct (MPD) dilatation on the tail side of the tumor; however, most of the MPD signal disappeared on MRCP after chemotherapy. Surgical findings failed to identify MPD on the first pancreatic resection plane, and additional resection was conducted; however, no MPD was found. As a pancreatic duct anastomosis was not available, pancreatic reconstruction was selected for pancreaticogastric anastomosis using the invagination method. Pathologically, the pancreatic tissue on the tail side of the tumor was replaced by fibrotic tissue, and MPD could not be identified. To the best of our knowledge, this is the first case report of the disappearance of a dilated pancreatic duct on the tail side accompanied by exocrine tissue loss during preoperative treatment for pancreatic cancer.

Our previous studies revealed a novel link between gemcitabine (GEM) chemotherapy and elevated glutamine-fructose-6-phosphate transaminase 2 (GFPT2) expression in pancreatic cancer (PaCa) cells. GFPT2 is a rate-limiting enzyme in the hexosamine biosynthesis pathway (HBP). HBP can enhance metastatic potential by regulating epithelial-mesenchymal transition (EMT). The aim of this study was to further evaluate the effect of chemotherapy-induced GFPT2 expression on metastatic potential. GFPT2 expression was evaluated in a mouse xenograft model following GEM exposure and in clinical specimens of patients after chemotherapy using immunohistochemical analysis. The roles of GFPT2 in HBP activation, downstream pathways, and cellular functions in PaCa cells with regulated GFPT2 expression were investigated. GEM exposure increased GFPT2 expression in tumors resected from a mouse xenograft model and in patients treated with neoadjuvant chemotherapy (NAC). GFPT2 expression was correlated with post-operative liver metastasis after NAC. Its expression activated the HBP, promoting migration and invasion. Treatment with HBP inhibitors reversed these effects. Additionally, GFPT2 upregulated ZEB1 and vimentin expression and downregulated E-cadherin expression. GEM induction upregulated GFPT2 expression. Elevated GFPT2 levels promoted invasion by activating the HBP, suggesting the potential role of this mechanism in promoting chemotherapy-induced metastasis.

This study aimed to investigate the prognostic significance of PET/CT radiomics to predict overall survival (OS) in patients with resectable pancreatic ductal adenocarcinoma (PDAC).

We enrolled 627 patients with resectable PDAC who underwent preoperative 18 F-FDG PET/CT and subsequent curative surgery. Radiomics analysis of the PET/CT images for the primary tumor was performed using the Chang-Gung Image Texture Analysis toolbox. Radiomics features were subjected to least absolute shrinkage and selection operator (LASSO) regression to select the most valuable imaging features of OS. The prognostic significance was evaluated by Cox proportional hazards regression analysis. Conventional PET parameters and LASSO score were assessed as predictive factors for OS by time-dependent receiver operating characteristic curve analysis.

During a mean follow-up of 28.8 months, 378 patients (60.3%) died. In the multivariable Cox regression analysis, tumor differentiation, resection margin status, tumor stage, and LASSO score were independent prognostic factors for OS (HR, 1.753, 1.669, 2.655, and 2.946; all P < 0.001, respectively). The time-dependent receiver operating characteristic curve analysis showed that the LASSO score had better predictive performance for OS than conventional PET parameters.

The LASSO score using the 18 F-FDG PET/CT radiomics of the primary tumor was the independent prognostic factor for predicting OS in patients with resectable PDAC and may be helpful in determining therapeutic and follow-up plans for these patients.

To investigate the values of apparent diffusion coefficient (ADC) for the treatment response evaluation in pancreatic cancer (PC) patients receiving neoadjuvant therapy (NAT).

This study included 103 NAT patients with histologically proven PC. ADC maps were generated using monoexponential diffusion-weighted imaging (b values: 50, 800 s/mm2). Tumors' minimum, maximum, and mean ADCs were measured and compared pre- and post-NAT. Variations in ADC values measured between pre- and post-NAT completion for NAT methods (chemotherapy, chemoradiotherapy), tumor locations (head/neck, body/tail), tumor regression grade (TRG) levels (0-2, 3), N stages (N0, N1/N2) and tumor resection margin status (R0, R1), were further analyzed.

The minimum, maximum, and mean ADC values all increased dramatically after NAT, rising from 23.4 to 25.4% (all p < 0.001): mean (average: 1.626 × 10- 3 mm2/s vs. 1.315 × 10- 3 mm2/s), minimum (median: 1.274 × 10- 3 mm2/s vs. 1.034 × 10- 3 mm2/s), and maximum (average: 1.981 × 10- 3 mm2/s vs. 1.580 × 10- 3 mm2/s). The ADCs between the subgroups of all the criteria under investigation did not differ significantly for the minimum, maximum, or mean values pre- or post-NAT (P = 0.08 to 1.00). In the patients with borderline resectable PC (n = 47), the rate of tumor size changes after NAT was correlated with the pre-NAT mean ADC values (Spearman's coefficient: 0.288, P = 0.049).

The ADC values of PC increased significantly following NAT; however, the percentage increases failed to provide any predictive value for the resection margin status or TRG levels.

To investigate the utilization of an end-to-end multimodal convolutional model in the rapid and accurate diagnosis of pancreatic diseases using abdominal CT images.

In this study, a novel lightweight label-free end-to-end multimodal network (eeMulNet) model was proposed for the rapid and precise diagnosis of abnormal pancreas. The eeMulNet consists of two steps: pancreatic region localization and multimodal CT diagnosis integrating textual and image data. A research dataset comprising 715 CT scans with various types of pancreas diseases and 228 CT scans from a control group was collected. The training set and independent test set for the multimodal classification network were randomly divided in an 8:2 ratio (755 for training and 188 for testing).

The eeMulNet model demonstrated outstanding performance on an independent test set of 188 CT scans (Normal: 45, Abnormal: 143), with an area under the curve (AUC) of 1.0, accuracy of 100%, and sensitivity of 100%. The average testing duration per patient was 41.04 seconds, while the classification network took only 0.04 seconds.

The proposed eeMulNet model offers a promising approach for the diagnosis of pancreatic diseases. It can support the identification of suspicious cases during daily radiology work and enhance the accuracy of pancreatic disease diagnosis. The codes and models of eeMulNet are publicly available at Rudeguy1/eeMulNet (github.com).

Mutated Kirsten rat sarcoma viral oncogene homolog (KRAS) is the most common oncogene alteration in pancreatic ductal adenocarcinoma, and KRAS glycine to cystine substitution at codon 12 (G12C) mutations (KRAS G12Cmut) are observed in 1%-2%. Several inhibitors of KRAS G12C have recently demonstrated promise in solid tumors, including pancreatic cancer. Little is known regarding clinical, genomics, and outcome data of this population.

Patients with pancreatic cancer and KRAS G12Cmut were identified at Memorial Sloan Kettering Cancer Center and via the American Association of Cancer Research Project Genomics, Evidence, Neoplasia, Information, Exchange database. Clinical, treatment, genomic, and outcomes data were analyzed. A cohort of patients at Memorial Sloan Kettering Cancer Center with non-G12C KRAS pancreatic cancer was included for comparison.

Among 3571 patients with pancreatic ductal adenocarcinoma, 39 (1.1%) with KRAS G12Cmut were identified. Median age was 67 years, and 56% were female. Median body mass index was 29.2 kg/m2, and 67% had a smoking history. Median overall survival was 13 months (95% CI: 9.4 months, not reached) for stage IV and 26 months (95% CI: 23 months, not reached) for stage I-III. Complete genomic data (via American Association of Cancer Research Project Genomics, Evidence, Neoplasia, Information, Exchange database) was available for 74 patients. Most common co-alterations included TP53 (73%), CDKN2A (33%), SMAD4 (28%), and ARID1A (21%). Compared with a large cohort (n = 2931) of non-G12C KRAS-mutated pancreatic ductal adenocarcinoma, ARID1A co-mutations were more frequent in KRAS G12Cmut (P < .05). Overall survival did not differ between KRAS G12Cmut and non-G12C KRAS pancreatic ductal adenocarcinoma. Germline pathogenic variants were identified in 17% of patients; 2 patients received KRAS G12C-directed therapy.

Pancreatic cancer and KRAS G12Cmut may be associated with a distinct clinical phenotype. Genomic features are similar to non-G12C KRAS-mutated pancreatic cancer, although enrichment of ARID1A co-mutations was observed. Targeting of KRAS G12C in pancreatic cancer provides a precedent for broader KRAS targeting in pancreatic cancer.

Despite advances made in pancreatic cancer treatment, the extent of progress made in pancreatic cancer mortality at the population level remains unclear. Our cross-sectional study sought to measure trends in pancreatic cancer mortality in the United States in the last 2 decades.

Patients with pancreatic cancer mortality from 1999 to 2020 were analyzed from the Centers for Disease Control and Prevention's Wide-Ranging Online Data for Epidemiologic Research (CDC WONDER). Age-adjusted mortality rates (AAMRs) per 100,000 individuals were measured. We used joinpoint trend analysis to determine average annual percent change (AAPC) in AAMR trends.

From 1999 to 2020, pancreatic cancer accounted for 809,197 deaths. Overall, the AAMRs of pancreatic cancer increased from 20.74 per 100,000 individuals in 1999 to 21.60 per 100,000 individuals in 2020. The highest AAMR was recorded in non-Hispanic Black males (30.11 per 100,000 individuals), and the lowest, in non-Hispanic White females (18.51 per 100,000 individuals). Patients aged 75-84 years had the highest AAMR (6.87 per 100,000 individuals) compared to the younger patients. The highest AAMR was observed in the Northeast region (22.07 per 100,000 individuals) and rural regions (21.29 per 100,000 individuals).

There was no improvement in pancreatic cancer mortality in the last two decades. These findings emphasize the importance of efforts to increase access to multidisciplinary cancer care with the realization that without it, improvements in treatment standards will not translate to lower cancer mortality at the population level.

To comprehensively analyze the clinical significance of Immune Checkpoint-Related Genes (ICRGs) in Pancreatic Adenocarcinoma (PAAD).

PAAD tissues and normal pancreatic tissues were obtained from The Cancer Genome Atlas (TCGA) and Genotype-Tissue Expression (GTEx) databases, and 283 ICRGs were integrated by the Kyoto Encyclopedia of Genes and Genomes (KEGG) and Reactome datasets. Unsupervised clustering was used to obtain potential ICRGs-based PAAD subtypes. Wilcoxon test was performed to screen Differentially Expressed ICRGs (DEICRGs), while cox regression analyses were utilized to identify prognosis-related ICRGs and clinicopathological factors, and construct the corresponding models. The Tumor Immune Microenvironment (TIME) was evaluated. Moreover, the authors performed enrichment analysis, Gene Set Enrichment Analysis (GSEA), and transcription factor regulatory networks to realize underlying mechanisms.

Three ICRGs-based PAAD subtypes were identified, and they were associated with three ESTIMATE scores, a Tumor Microenvironment (TMB) score, 14 therapeutic immune checkpoints, and infiltration levels of seven immune cells. On top of that, the authors constructed two signatures based on DEICRGs to predict the Overall Survival (OS) (Area Under the ROC Curve [AUC: 0.741∼0.778]) and Progression-Free Survival (PFS) (AUC: 0.746∼0.831) of patients. Two nomograms were established by combining clinical variables and signatures. In addition, the authors found higher infiltration of naïve B cells and CD8+ T-cells in low-risk PAAD patients, and higher infiltration of suppressive immune cells and cancer-related signaling pathways in high-risk PAAD patients.

The present study suggested that ICRGs were associated with TIME formation and prognosis of PAAD patients, which may serve as novel clinical biomarkers and therapeutic targets.

Pancreatic cancer is one of the deadliest cancers in the world. In recent years, the incidence and mortality rates of pancreatic cancer have shown an increasing trend year by year. This study investigates the independent effects of age, period, and cohort on the global incidence, mortality, and disability-adjusted life years (DALYs) of pancreatic cancer from 1990 to 2019, and evaluates the differences in the burden of pancreatic cancer across regions with different Sociodemographic Index (SDI) levels.

Estimating the impact of age, period, and cohort on pancreatic cancer disease burden in different SDI regions using age-period-cohort modeling with data (with 95% uncertainty intervals [UI]) from the Global Burden of Disease (GBD) Study 2019 and net drift of age-standardized incidence rates (ASIR), age-standardized mortality rates (ASMR), and age-standardized DALY rates (ASDR) for pancreatic cancer in 120 countries.

The number of new cases of pancreatic cancer worldwide increased from 197,348 (95% UI: 188,604,203,971) in 1990 to 530,297 (486,175,573,635) in 2019, the number of deaths increased from 198,051 (189,329 to 204,763) in 1990 to 531,107 (491,948 to 566,537) in 2019, and the number of DALY increased from 4,647,207 (4,465,440 to 4,812,129) in 1990 to 11,549,016 (10,777,405 to 1,238,912) in 2019. The ASIR of the average levels in global pancreatic cancer increased from 5.22 (4.97 to 5.40) per 100,000 population to 6.57 (6.00 to 7.09) per 100,000 population, the ASMR increased from 5.34 (5.07 to 5.52) per 100,000 population to 6.62 (6.11 to 7.06) per 100,000 population, and the ASDR increased from 115.47 (110.82 to 119.60) per 100,000 population to 139.61 (130.18 to 149.14) per 100,000 population. The incidence, mortality, and DALY rates of pancreatic cancer increase with age globally and across all SDI regions, peaking in the 85-89 age group. In high and high-middle SDI regions, the growth rate for males is higher than for females before the age of 85, while females have a higher growth rate after 85. The 75-79 age group exhibits the highest DALY rate in high and high-middle SDI regions, significantly higher than the global and other SDI regions. From 1990 to 2019, the period effects of pancreatic cancer incidence, mortality, and DALY rates have increased significantly worldwide, while remaining almost unchanged in high and high-middle SDI regions. In contrast, period effects have significantly increased in middle, low-middle, and low SDI regions. Cohort effects are more pronounced in middle, low-middle, and low SDI regions.

With the aggravation of population aging, the incidence and mortality rates of pancreatic cancer in the world are increasing, and effective prevention and control measures can be achieved by reducing the exposure of risk factors. The APC model used in our analysis provides a novel approach to understanding the complex trends in the incidence, mortality, and disability-adjusted life years of pancreatic cancer. It can inform the development of targeted interventions to reduce the severe disease burden caused by pancreatic cancer.

BACKGROUND Adenocarcinomas and pancreatic neuroendocrine tumors (pNETs) display some similarities and differences. The aim of this study was to compare preoperative data and morphological parameters, and to assess postoperative complications after resection. MATERIAL AND METHODS Data of 162 patients who underwent distal pancreatic resection for neuroendocrine or adenocarcinoma tumor were retrospectively analyzed. After applying inclusion and exclusion criteria, 131 patients were included in the study. The preoperative data analyzed included age, sex, and ASA-PS (American Society of Anesthesiologists Physical Status) grade. The diameter of the pancreatic duct and the texture of the pancreas were analyzed. Postoperative data included grading (G1-G3), the presence of PanIN (pancreatic intraepithelial neoplasia), infiltration of structures, and postoperative complications. RESULTS Patients with adenocarcinoma were statistically older and had a higher ASA-PS class than patients with NET (P<0.001). Statistically significantly more patients with adenocarcinoma had a histopathological diagnosis of G3 (p<0.001). In patients with adenocarcinomas infiltration of structures occurred more frequently. Pancreatic duct diameter ≥3 mm was more common in patients with adenocarcinoma (P=0.045). Clinically significant pancreatic fistulas were more frequent in patients with neuroendocrine tumors (P=0.044). CONCLUSIONS Adenocarcinomas in the pancreatic body and tail are more aggressive, they cause more frequent infiltration of structures, and more often metastasize to lymph nodes compared to NETs. NETs tend to have softer pancreatic texture and higher incidence of clinically significant pancreatic fistulas, but postoperative complications of Clavien-Dindo grade ≥III occur at a similar rate in both groups.

Pancreatic cancer (PC) poses a substantial global health challenge, ranking as the fourth leading cause of cancer-related deaths due to its high mortality rate. Late-stage diagnoses are common due to the absence of specific symptoms. Pancreatic ductal adenocarcinoma (PDAC) accounts for the majority of PC cases. Recent research has suggested a potential link between elevated serum levels of bile acids (BAs) and tumorigenesis of PDAC. This study aims to understand how taurochenodeoxycholic acid (TCDCA), a secondary BA, influences PDAC using RNA sequencing techniques on the Capan-1 cell line. We identified 2,950 differentially expressed genes (DEGs) following TCDCA treatment, with 1,597 upregulated and 1,353 downregulated genes. These DEGs were associated with critical PDAC pathways, including coagulation, angiogenesis, cell migration, and signaling regulation. Furthermore, we reviewed relevant literature highlighting genes like DKK-1, KRT80, UPLA, and SerpinB2, known for their roles in PDAC tumorigenesis and metastasis. Our study sheds light on the complex relationship between BAs and PDAC, offering insights into potential diagnostic markers and therapeutic targets. Further research is needed to unravel these findings' precise mechanisms and clinical implications, potentially improving PDAC diagnosis and treatment.

A woman in her 40s underwent evaluation for abdominal pain, jaundice and acholic stools and was diagnosed with metastatic pancreatic head adenocarcinoma. She was enrolled in a clinical trial investigating the benefits of ibrutinib with nab-paclitaxel and gemcitabine, and subsequently received modified FOLFIRINOX. Over the course of 6 years on chemotherapy, she experienced complete regression of the pancreatic and liver lesions, as well as normalisation of her tumour markers. She has been off chemotherapy for 6 months with no evidence of disease and normal tumour markers. Despite advances in chemotherapy and surgical options, metastatic pancreatic adenocarcinoma continues to carry a grim prognosis. This case report demonstrates a rare case of a long-term survivor of unresectable metastatic pancreatic adenocarcinoma treated with chemotherapy alone.

Neoadjuvant chemotherapy (NAC) is becoming favored for all pancreatic adenocarcinoma (PDAC). Patients with seemingly resectable disease infrequently still display vascular involvement intraoperatively. Outcomes following NAC versus upfront surgery in patients undergoing pancreaticoduodenectomy (PD) with vascular resection are unknown.

We performed a retrospective cohort study of PDAC patients who underwent PD with vascular resection between January 1, 2013, to December 31, 2020, within a single academic center. Clinicopathologic characteristics and disease-free survival (DFS) were compared between NAC versus upfront surgery cohorts using the Kaplan-Meier estimate and Cox proportional-hazards regression model.

Eighty-one patients who underwent PD with vascular resection for PDAC were included. Forty-six patients (56%) received NAC. The NAC cohort more often had pathologic N0 status (47.8% vs. 8.6%, p < 0.001), had decreased vascular invasion (11% vs. 40%, p = 0.002), and completed chemotherapy (80% vs. 40%, p < 0.01). The NAC cohort demonstrated improved DFS (40.5 vs. 14.3 months, p = 0.007). In multivariable analysis, NAC remained independently associated with increased DFS (HR = 0.48, p = 0.02).

NAC was associated with improved clinicopathologic outcomes and DFS in PD with vascular resection. These findings demonstrate the advantage of NAC in PDAC patients undergoing PD with vascular resection.

Pancreatic cancer is the third leading cause of cancer deaths in the United States. Despite decreasing cancer mortality rates as a whole, pancreatic cancer death rates in the United States remain steady and demonstrate racial/ethnic disparities. Divergent cancer mortality trends have also been observed between metro and nonmetro populations. We therefore aimed to compare metro and nonmetro trends in pancreatic cancer mortality rates in the United States from 1999 to 2020 and investigate potential sex and racial/ethnic differences.

We analyzed National Center for Health Statistics data for all pancreatic cancer deaths among individuals aged 25 years or older in the United States. We estimated the average annual percent change (AAPC) in age-standardized pancreatic cancer mortality rates in metro versus nonmetro areas by sex and race/ethnicity.

Of the total 810,425 pancreatic cancer-related deaths identified from 1999 to 2020, 668,547 occurred in metro areas and 141,878 in nonmetro areas. Non-Hispanic Black individuals had the highest rates of pancreatic cancer mortality regardless of metropolitan status. In both metro and nonmetro areas, pancreatic cancer mortality rates among non-Hispanic White individuals increased over the study period (AAPC: metro, males, 0.32%; females, 0.27%; nonmetro, males, 0.77%; females, 0.62%). Non-Hispanic Black individuals in metro areas had a decrease in pancreatic cancer mortality (AAPC: males, -0.25%; females, -0.29%), but rates among non-Hispanic Black women in nonmetro areas increased (AAPC, 0.49%).

There are variations not only in pancreatic cancer mortality by metro and nonmetro status but also by sex and race/ethnicity within these areas. Individuals who live in nonmetro areas have higher pancreatic cancer mortality rates and increasing death rates compared with their metro counterparts. These findings highlight the need for targeted cancer prevention strategies that are specific to metro or nonmetro populations.

It remains unclear today whether the poor prognosis of pancreatic ductal adenocarcinoma (PDAC) was further worsened by the COVID-19 pandemic and whether this may affect providers and patients, today. Hence, this study aimed to investigate the effect of COVID-19 on care delivery and outcomes of patients with PDAC in the United States.

The National Cancer Database was queried for PDAC, between 2017 and 2020. Changes in the number of diagnoses and treatment patterns were compared annually for the entire cohort. Changes in surgical outcomes and median time from diagnosis to treatment were compared and analyzed. Chi-square, Mann-Whitney U, and Kruskal-Wallis tests were performed.

Of 127,613 patients with PDAC, PDAC diagnoses from 2017 (30,573) to 2019 (33,465) increased but decreased in 2020 (31,218). The number of patients receiving surgery or radiotherapy was stable between 2017 to 2019 (21.75% ± 0.05% and 13.9% ± 0.3%, respectively) but decreased in 2020 (20.7% and 12.4% respectively). Although patients received chemotherapy with increasing frequently from 2016 (60.7%) to 2019 (63.5%), this trend stopped in 2020 (63%). Of 27,490 patients undergoing surgery, the mean time from diagnosis to surgery increased from 2017 (34 days) to 2019 (56 days), with an increase in delay in 2020 (81 days). Moreover, patients who were tested for COVID-19, had a longer median time from diagnosis to surgery even if tested negative (COVID+, 140 days; COVID-, 112 days; P < .001).

Although the oncologic quality of PDAC surgery remained the same during the pandemic, not only did the pandemic lead to an underdiagnosis of PDAC and care delays, but even the suspicion of COVID-19 in patients with a negative test adversely affected their care.

Acute cholangitis (AC) is a common complication of pancreatic ductal adenocarcinoma (PDAC). Herein, we evaluated outcomes after the first AC episode and predictors of mortality and AC recurrence in patients with stage IV PDAC.

We conducted a single-center, retrospective observational study using institutional databases. Clinical data and outcomes for patients with stage IV PDAC and at least one documented episode of AC, were assessed. Overall survival (OS) was estimated using the Kaplan-Meier method, and Cox regression model was employed to identify predictors of AC recurrence and mortality.

One hundred and twenty-four patients with stage IV PDAC and AC identified between January 01, 2014 and October 31, 2020 were included. Median OS after first episode of AC was 4.1 months (95 % CI, 4.0-5.5), and 30-day, 6, and 12-month survival was 86.2 % (95 % CI, 80.3-92.5), 37 % (95 % CI, 29.3-46.6 %) and 18.9 % (95 % CI, 13.1-27.3 %), respectively. Primary tumor in pancreatic body/tail (HR 2.29, 95 % CI: 1.26 to 4.18, p = 0.011), concomitant metastases to liver and other sites (HR 1.96, 95 % CI: 1.16 to 3.31, p = 0.003) and grade 3 AC (HR 2.26, 95 % CI: 1.45 to 3.52, p < 0.001), predicted worse outcomes. Intensive care unit admission, sepsis, systemic therapy, treatment regimen, and time to intervention did not predict survival or risk of recurrence of AC.

AC confers significant morbidity and mortality in advanced PDAC. Worse outcomes are associated with higher grade AC, primary tumor location in pancreatic body/tail, and metastases to liver and other sites.

Pancreatic cancer, a highly fatal malignancy, has shown a global rise in the incidence and mortality rates. However, these rates vary significantly across different regions worldwide. This study aims to assess the incidence and mortality of pancreatic cancer in Saudi Arabia. We collected the data from 16 annual cancer incidence reports in Saudi Arabia for the study period (2005-2020) and from the WHO's IARC Global Cancer Observatory website. Although the burden of pancreatic cancer in Saudi Arabia is relatively lower compared to global rates, the disease incidence has shown a steady increase over the study period, in addition to regional variations within the country. The disease predominantly affects the elderly population, aged 50 years and above in both genders, with males exhibiting higher rates than females. Further studies are required to identify the potential risk factors for pancreatic cancer in the Saudi population.

Pancreatic ductal adenocarcinoma (PDAC) is poised to become the second leading cause of cancer-related death by 2030, necessitating innovative therapeutic strategies. Genetic and epigenetic alterations, including those involving the COMPASS-like complex genes, have emerged as critical drivers of PDAC progression. This review explores the genetic and epigenetic landscape of PDAC, focusing on the role of the COMPASS-like complex in regulating chromatin accessibility and gene expression. Specifically, we delve into the functions of key components such as KDM6A, KMT2D, KMT2C, KMT2A, and KMT2B, highlighting their significance as potential therapeutic targets. Furthermore, we discuss the implications of these findings for developing novel treatment modalities for PDAC.

Patients with paraquat poisoning (PP) have a mortality rate comparable to that of advanced malignancies, yet palliative care is seldom considered in these patients. This audit aimed to identify triggers for early palliative care referral in critically ill patients with PP.

Medical records of patients with PP were audited. Predictors of mortality within 48 hours of hospitalization and 24 hours of intensive care unit (ICU) admission were considered as triggers for palliative care referral.

Among 108 patients, 84 complete records were analyzed, and 53 out of 84 (63.1%) expired. Within 48 hours after hospitalization, the lowest oxygen partial pressure in arterial blood to a fraction of inspired oxygen [the ratio of partial pressure of oxygen in arterial blood (PaO2) to the fraction of inspiratory oxygen concentration (FiO2) (PaO2/FiO2)] was the independent predictor of mortality, cut-off ≤ 197; the area under the curve (AUC), 0.924; sensitivity, 97%; specificity, 78%; p <0.001; and 95% confidence interval (CI): 0.878-0.978. Kaplan-Meier survival plot showed that the mean survival time of patients with the lowest PaO2/FiO2, ≤197, was 4.64 days vs 17.20 days with PaO2/FiO2 >197 (log-rank p < 0.001). Sequential organ failure assessment (SOFA) score within 24 hours of ICU admission had a cut-off ≥9; AUC, 0.980; p < 0.001; 95% CI: 0.955-1.000; 91% sensitivity; and 90% specificity for mortality prediction. Out of the total of 84 patients with PP analyzed, there were 11 patients admitted to the high dependency units (13.1%) and 73 patients admitted to the ICU (86.9%). Out of the total of 84 patients of PP in whom data was analyzed, 53 (63.1%) patients required ventilator support. All the 53 patients who required ventilator support due to worsening hypoxemia, eventually expired.

The lowest PaO2/FiO2 ≤ 197 within 48 hours of hospitalization, SOFA score ≥9 within 24 hours of ICU admission or need for mechanical ventilation are predictors of mortality in PP patients, who might benefit from early palliative care.

Rao S, Maddani SS, Chaudhuri S, Bhatt MT, Karanth S, Damani A, et al. Utility of Clinical Variables for Deciding Palliative Care in Paraquat Poisoning: A Retrospective Study. Indian J Crit Care Med 2024;28(5):453-460.

With an overall 5-year survival rate of 12%, pancreas ductal adenocarcinoma (PDAC) is an aggressive cancer that claims more than 50,000 patient lives each year in the United States alone. Even those few patients who undergo curative-intent resection with favorable pathology reports are likely to experience recurrence within the first two years after surgery and ultimately die from their cancer. We hypothesize that risk factors for these early recurrences can be identified with thorough preoperative staging, thus enabling proper patient selection for surgical resection and avoiding unnecessary harm. Herein, we review evidence supporting multidisciplinary and multimodality staging, comprehensive neoadjuvant treatment strategies, and optimal patient selection for curative-intent surgical resections. We further review data generated from our standardized approach at the Mayo Clinic and extrapolate to inform potential future investigations.

Pancreatic ductal adenocarcinoma (PADC) still has nowadays a very impaired long-term survival. Most studies are focused on overall survival; however, local recurrence occurs about up to 50% of cases and seems to be highly related with margin resection status. We aim to analyze the impact of vascular resection margins on local recurrence (LR) and to assess its impact on overall and disease-free survival.

Eighty out of 191 patients who underwent pancreatoduodenectomy in a university hospital between 2006 and 2021 with PDAC diagnosis were analyzed and vascular margin status specifically addressed. Univariate and multivariate were performed. Time to LR was compared by using the Kaplan-Meier method and prognostic factors assessed using Cox regression hazards model.

LR appeared in 10 (50%) of the overall R1 resections in the venous margin and 9 (60%) in the arterial one. Time to LR was significantly shorter when any margin was overall affected (23.2 vs 44.7 months, p = 0.01) and specifically in the arterial margin involvement (13.7 vs 32.1 months, p = 0.009). Overall R1 resections (HR 2.61, p = 0.013) and a positive arterial margin (HR 2.84, p = 0.012) were associated with local recurrence on univariate analysis, whereas arterial positive margin remained significant on multivariate analysis (HR 2.70, p = 0.031).

Arterial margin invasion is correlated in our cohort with local recurrence. Given the limited ability to modify this margin intraoperatively, preoperative therapies should be considered to improve local margin clearance.

The International Cancer of the Pancreas Screening Consortium recommended annual imaging for individuals at increased risk for developing a pancreatic ductal adenocarcinoma (PDAC) who did not have concerning pancreatic findings or a cyst <3 cm without worrisome features. We aimed to determine if 3-cm cyst size accurately predicted advanced precursor lesions in high-risk individuals undergoing surveillance.

Imaging for high-risk individuals (HRIs) undergoing PDAC surveillance from 2007 to 2021 was reviewed and pancreatic abnormalities were recorded including dominant cyst size and number of cysts. Subjects were excluded if they had the following: (1) no follow-up imaging after baseline, (2) solid lesion at baseline, or (3) development of solid lesion without having cyst on prior imaging.

Five of the 77 HRIs found to have a cystic lesion on surveillance developed a PDAC: 3 with cystic lesion >1 cm as compared with only 2 of 67 HRIs with cystic lesions <1 cm (P < 0.05). None of these cysts developed worrisome findings and 4/5 PDACs did not arise from visualized cystic precursor lesion.

Patients with a cyst ≥1 cm were at increased risk for developing PDAC compared with patients with cyst <1 cm. Pancreatic ductal adenocarcinoma usually did not arise from a recognized cystic lesion.

Cancer is one of the leading causes of death worldwide. Early cancer detection is critical because it can significantly improve treatment outcomes, thus saving lives, reducing suffering, and lessening psychological and economic burdens. Cancer biomarkers provide varied information about cancer, from early detection of malignancy to decisions on treatment and subsequent monitoring. A large variety of molecular, histologic, radiographic, or physiological entities or features are among the common types of cancer biomarkers. Sizeable recent methodological progress and insights have promoted significant developments in the field of early cancer detection biomarkers. Here we provide an overview of recent advances in the knowledge related to biomolecules and cellular entities used for early cancer detection. We examine data from the CAS Content Collection, the largest human-curated collection of published scientific information, as well as from the biomarker datasets at Excelra, and analyze the publication landscape of recent research. We also discuss the evolution of key concepts and cancer biomarkers development pipelines, with a particular focus on pancreatic and liver cancers, which are known to be remarkably difficult to detect early and to have particularly high morbidity and mortality. The objective of the paper is to provide a broad overview of the evolving landscape of current knowledge on cancer biomarkers and to outline challenges and evaluate growth opportunities, in order to further efforts in solving the problems that remain. The merit of this review stems from the extensive, wide-ranging coverage of the most up-to-date scientific information, allowing unique, unmatched breadth of landscape analysis and in-depth insights.

(1) Background: This study aimed to assess the global burden of pancreatic cancer attributable to a high BMI in 1990-2019. (2) Methods: An ecological study was carried out. Data about deaths and Disability-Adjusted Life Years (DALYs) for pancreatic cancer were extracted from the Global Burden of Disease (GBD) study. The age-standardized rates (ASRs, per 100,000) were presented. In order to determine trends of pancreatic cancer burden, joinpoint regression analysis was used to calculate the average annual percent change (AAPC). (3) Results: The highest ASRs of DALYs of pancreatic cancer were found in the United Arab Emirates (47.5 per 100,000), followed by countries with about 25.0 per 100,000 (such as Hungary, Czechia, and Montenegro). From 1990 to 2019, the ASRs of deaths and DALYs of pancreatic cancer attributable to a high BMI significantly increased (p < 0.001) for both sexes in all ages, and across all SDI quintiles and all GBD regions. The highest fraction of DALYs attributable to a high BMI was found in the United States of America and China (equally about 15.0%), followed by the Russian Federation, India, Germany, and Brazil (about 5.0%, equally). (4) Conclusions: Further analytical epidemiological studies are necessary to elucidate the relationship between pancreatic cancer and a high BMI.

Palliative care consultation (PCC) has been shown to improve quality of life and reduce costs for various chronic life-threatening diseases. Despite PCC incorporation into modern pancreatic cancer care guidelines, limited data regarding its specific utilization and impact on resource use is available.

The 2016-2020 Nationwide Readmissions Database was used to identify all adult hospitalizations entailing pancreatic cancer. Only patients with at least one readmission within 90 days were included to account for uncaptured out-of-hospital mortality. Multivariable regression models were used to ascertain the relationship between inpatient PCC during initial hospitalization and index as well as cumulative costs, overall length of stay (LOS), readmission rate, and number of repeat hospitalizations.

Of an estimated 175,805 patients with pancreatic cancer, 11.1% had inpatient PCC during the index admission. PCC utilization significantly increased from 10.5% in 2016 to 11.6% in 2020 (nptrend < 0.001). After adjustment, PCC was associated with reduced index hospitalization costs [β: - $1100; 95% confidence interval (CI) - 1500, - 800; P < 0.001] and cumulative 90-day costs (β: - $11,700; 95% CI - 12,700, - 10,000; P < 0.001). PCC was associated with longer index LOS (β: + 1.12 days, 95% CI 0.92-1.31, P < 0.001) but significantly reduced cumulative LOS (β: - 3.16 days; 95% CI - 3.67, - 2.65; P < 0.001). Finally, PCC was linked with decreased odds of 30-day nonelective readmission (AOR: 0.48, 95% CI 0.45-0.50, P < 0.001).

PCC was associated with decreased costs, readmission rates, and number of hospitalizations among patients with pancreatic cancer. Directed strategies to increase utilization and reduce barriers to consultation should be implemented to encourage practitioners to maximize inpatient PCC referral rates.

The diagnosis and monitoring of disease through the detection of circulating protein biomarkers is a growing field in the practice of oncology. The search for more effective protein biomarkers to aid in the diagnosis and treatment of patients with pancreatic ductal adenocarcinoma (PDAC) remains a valuable area of study, given the aggressive and often occult nature of this malignancy. Liquid biopsies are attractive, as they offer a minimally invasive and cost-effective approach when compared to traditional biopsy methods and imaging modalities used for diagnosis and surveillance. Carbohydrate antigen (CA) 19-9 is currently the most commonly used serum protein biomarker for the diagnosis and monitoring of patients with PDAC, but due to its sensitivity and specificity, its utility remains limited. In this review, we examine how circulating protein biomarkers are used in the diagnosis, prognostication, and surveillance of PDAC. We also highlight protein biomarkers that are currently under investigation that have the potential to enhance our ability to detect early-stage malignancies, predict response to therapy, and monitor for recurrence, but these markers require larger prospective validation studies before they can be widely implemented. Continued efforts to identify and validate novel biomarkers will be crucial for improving the management and outcomes of patients with this challenging disease.

Pancreatic cancer is a highly lethal disease with limited treatment options. Hence, there is a considerable medical need for novel treatment strategies. Monoclonal antibodies (mAbs) have significantly improved cancer therapy, primarily due to their ability to stimulate antibody-dependent cellular cytotoxicity (ADCC), which plays a crucial role in their therapeutic efficacy. As a result, significant effort has been focused on improving this critical function by engineering mAbs with Fc regions that have increased affinity for the Fc receptor CD16 expressed on natural killer (NK) cells, the major cell population that mediates ADCC in humans. Here we report on the preclinical characterization of a mAb directed to the target antigen B7-H3 (CD276) containing an Fc part with the amino acid substitutions S239D/I332E to increase affinity for CD16 (B7-H3-SDIE) for the treatment of pancreatic cancer. B7-H3 (CD276) is highly expressed in many tumor entities, whereas expression on healthy tissues is more limited. Our findings confirm high expression of B7-H3 on pancreatic cancer cells. Furthermore, our study shows that B7-H3-SDIE effectively activates NK cells against pancreatic cancer cells in an antigen-dependent manner, as demonstrated by the analysis of NK cell activation, degranulation and cytokine release. The activation of NK cells resulted in significant tumor cell lysis in both short-term and long-term cytotoxicity assays. In conclusion, B7-H3-SDIE constitutes a promising agent for the treatment of pancreatic cancer.

(1) Background: Pancreatic adenocarcinoma (PAC) is one of the most lethal types of cancer. Most cases of PAC occur in the head of the pancreas. Given the proximity of the pancreatic head to the bile duct, most patients present clinically during early stages of the disease, while distally located PAC could have delayed clinical presentation. (2) Aims: To assess predictors of non-head PAC. (3) Methods: A retrospective multicenter study was conducted, including all patients who had endoscopic ultrasound (EUS) for pancreatic masses and who had histologic confirmation of PAC. (4) Results: Of the 151 patients included, 92 (60.9%) had pancreatic head cancer, and 59 (39.1%) had distal pancreatic cancer. PAC at body was the most common location in the distal PAC group (31 patients (52.5%)). Logistic regression analysis demonstrated a significant association of obesity with distal migration of PAC (OR 4.44, 95% CI 1.15-17.19, p = 0.03), while none of the other assessed parameters showed a significant association. Notably, abdominal pain was more significantly associated with distal PAC vs. head location (OR 2.85, 95% CI 1.32-6.16, p = 0.008). (5) Conclusions: Obesity shows a significant association as a clinical predictor of distal PAC. Further studies are needed to better explore this association.

Peritoneal metastases (PM) from pancreatic ductal adenocarcinoma (PDAC) are currently treated with palliative systemic chemotherapy alone, with unsatisfactory results. Cytoreductive surgery (CRS) and hyperthermic intraperitoneal chemotherapy (HIPEC) may provide an oncologic benefit for highly selected patients.

Patients with PDAC and isolated PM who completed ≥ 6 months of systemic chemotherapy with objective response between 2017 and 2022 were retrospectively reviewed. All patients met the inclusion/exclusion criteria as per our previously published PDAC CRS/HIPEC protocol. Patients who underwent CRS/HIPEC were compared with matched patients who underwent systemic therapy alone. Overall survival (OS) from diagnosis of PM and progression-free survival (PFS) from CRS/HIPEC was evaluated.

In total, 61 patients met the inclusion criteria: 38 underwent systemic therapy alone and 23 CRS/HIPEC. There were no differences in baseline prognostic factors, including age, sex, tumor size, tumor location, anatomic resectability, or serum cancer antigen (CA) 19-9 (p > 0.05). Median OS from PM diagnosis in patients who underwent systemic therapy alone was 19 months with 1, 2, and 3 year OS of 81%, 31%, and 8%, respectively. In contrast, median OS from PM diagnosis in patients who underwent CRS/HIPEC was 41 months with improved 1, 2, and 3 year OS of 91%, 66%, and 59%, respectively (p = 0.002). In the 21 patients who achieved complete cytoreduction (CC-0), no adjuvant therapy was administered and the median PFS was 17 months.

CRS/HIPEC in highly selected patients with PDAC and PM results in promising oncologic outcomes that are unlikely to be achieved with systemic chemotherapy alone. Further investigation is warranted and ongoing (NCT04858009).

Pancreatic ductal adenocarcinoma has some of the worst prognostic outcomes among various cancer types. Detection of histologic patterns of pancreatic tumors is essential to predict prognosis and decide the treatment for patients. This histologic classification can have a large degree of variability even among expert pathologists.

To detect aggressive adenocarcinoma and less aggressive pancreatic tumors from nonneoplasm cases using a graph convolutional network-based deep learning model.

Our model uses a convolutional neural network to extract detailed information from every small region in a whole slide image. Then, we use a graph architecture to aggregate the extracted features from these regions and their positional information to capture the whole slide-level structure and make the final prediction.

We evaluated our model on an independent test set and achieved an F1 score of 0.85 for detecting neoplastic cells and ductal adenocarcinoma, significantly outperforming other baseline methods.

If validated in prospective studies, this approach has a great potential to assist pathologists in identifying adenocarcinoma and other types of pancreatic tumors in clinical settings.