Predicted Prognosis of Patients with Pancreatic Cancer by Machine Learning

Prediction of 12-month recurrence of pancreatic cancer using machine learning and prognostic factors

BACKGROUND AND AIM

Pancreatic cancer is lethal and prevalent among other cancer types. The recurrence of this tumor is high, especially in patients who did not receive adjuvant therapies. Early prediction of PC recurrence has a significant role in enhancing patients' prognosis and survival. So far, machine learning techniques have given us insight into favorable performance efficiency in various medical domains. So, this study aims to establish a prediction model based on machine learning to achieve better prediction on this topic.

MATERIALS AND METHODS

In this retrospective research, we used data from 585 PC patient cases from January 2019 to November 2023 from three clinical centers in Tehran City. Ten chosen ensemble and non-ensemble algorithms were used to establish prediction models on this topic.

RESULTS

Random forest and support vector machine with an AU-ROC of approximately 0.9 obtained more performance efficiency regarding PC recurrence. Lymph node metastasis, tumor size, tumor grade, radiotherapy, and chemotherapy were the best factors influencing PC recurrence.

CONCLUSION

Random forest and support vector machine algorithms demonstrated high-performance ability and clinical usability to improve doctors' decisions in achieving different therapeutic and diagnostic measures.

One novel transfer learning-based CLIP model combined with self-attention mechanism for differentiating the tumor-stroma ratio in pancreatic ductal adenocarcinoma

PURPOSE

To develop a contrastive language-image pretraining (CLIP) model based on transfer learning and combined with self-attention mechanism to predict the tumor-stroma ratio (TSR) in pancreatic ductal adenocarcinoma on preoperative enhanced CT images, in order to understand the biological characteristics of tumors for risk stratification and guiding feature fusion during artificial intelligence-based model representation.

MATERIAL AND METHODS

This retrospective study collected a total of 207 PDAC patients from three hospitals. TSR assessments were performed on surgical specimens by pathologists and divided into high TSR and low TSR groups. This study developed one novel CLIP-adapter model that integrates the CLIP paradigm with a self-attention mechanism for better utilizing features from multi-phase imaging, thereby enhancing the accuracy and reliability of tumor-stroma ratio predictions. Additionally, clinical variables, traditional radiomics model and deep learning models (ResNet50, ResNet101, ViT_Base_32, ViT_Base_16) were constructed for comparison.

RESULTS

The models showed significant efficacy in predicting TSR in PDAC. The performance of the CLIP-adapter model based on multi-phase feature fusion was superior to that based on any single phase (arterial or venous phase). The CLIP-adapter model outperformed traditional radiomics models and deep learning models, with CLIP-adapter_ViT_Base_32 performing the best, achieving the highest AUC (0.978) and accuracy (0.921) in the test set. Kaplan-Meier survival analysis showed longer overall survival in patients with low TSR compared to those with high TSR.

CONCLUSION

The CLIP-adapter model designed in this study provides a safe and accurate method for predicting the TSR in PDAC. The feature fusion module based on multi-modal (image and text) and multi-phase (arterial and venous phase) significantly improves model performance.

Machine learning algorithms and biomarkers identification for pancreatic cancer diagnosis using multi-omics data integration

PURPOSE

Pancreatic cancer is a lethal type of cancer with most of the cases being diagnosed in an advanced stage and poor prognosis. Developing new diagnostic and prognostic markers for pancreatic cancer can significantly improve early detection and patient outcomes. These biomarkers can potentially revolutionize medical practice by enabling personalized, more effective, targeted treatments, ultimately improving patient outcomes.

METHODS

The search strategy was developed following PRISMA guidelines. A comprehensive search was performed across four electronic databases: PubMed, Scopus, EMBASE, and Web of Science, covering all English publications up to September 2022. The Newcastle-Ottawa Scale (NOS) was utilized to assess bias, categorizing studies as "good," "fair," or "poor" quality based on their NOS scores. Descriptive statistics for all included studies were compiled and reviewed, along with the NOS scores for each study to indicate their quality assessment.

RESULTS

Our results showed that SVM and RF are the most widely used algorithms in machine learning and data analysis, particularly for biomarker identification. SVM, a supervised learning algorithm, is employed for both classification and regression by mapping data points in high-dimensional space to identify the optimal separating hyperplane between classes.

CONCLUSIONS

The application of machine-learning algorithms in the search for novel biomarkers in pancreatic cancer represents a significant advancement in the field. By harnessing the power of artificial intelligence, researchers are poised to make strides towards earlier detection and more effective treatment, ultimately improving patient outcomes in this challenging disease.

Machine learning-based prediction of pulmonary embolism to reduce unnecessary computed tomography scans in gastrointestinal cancer patients: a retrospective multicenter study

This study aimed to develop a machine learning (ML) model for predicting pulmonary embolism (PE) in patients with gastrointestinal cancers, a group at increased risk for PE. We conducted a retrospective, multicenter study analyzing patients who underwent computed tomographic pulmonary angiography (CTPA) between 2010 and 2020. The study utilized demographic and clinical data, including the Wells score and D-dimer levels, to train a random forest ML model. The model's effectiveness was assessed using the area under the receiver operating curve (AUROC). In total, 446 patients from hospital A and 139 from hospital B were included. The training set consisted of 356 patients from hospital A, with internal validation on 90 and external validation on 139 patients from hospital B. The model achieved an AUROC of 0.736 in hospital A and 0.669 in hospital B. The ML model significantly reduced the number of patients recommended for CTPA compared to the conventional diagnostic strategy (hospital A; 100.0% vs. 91.1%, P < 0.001, hospital B; 100.0% vs. 93.5%, P = 0.003). The results indicate that an ML-based prediction model can reduce unnecessary CTPA procedures in gastrointestinal cancer patients, highlighting its potential to enhance diagnostic efficiency and reduce patient burden.

Noninvasive prediction of lymph node metastasis in pancreatic cancer using an ultrasound-based clinicoradiomics machine learning model

OBJECTIVES

This study was designed to explore and validate the value of different machine learning models based on ultrasound image-omics features in the preoperative diagnosis of lymph node metastasis in pancreatic cancer (PC).

METHODS

This research involved 189 individuals diagnosed with PC confirmed by surgical pathology (training cohort: n = 151; test cohort: n = 38), including 50 cases of lymph node metastasis. Image-omics features were extracted from ultrasound images. After dimensionality reduction and screening, eight machine learning algorithms, including logistic regression (LR), support vector machine (SVM), K-nearest neighbors (KNN), random forest (RF), extra trees (ET), extreme gradient boosting (XGBoost), light gradient boosting machine (LightGBM), and multilayer perceptron (MLP), were used to establish image-omics models to predict lymph node metastasis in PC. The best omics prediction model was selected through ROC curve analysis. Machine learning models were used to analyze clinical features and determine variables to establish a clinical model. A combined model was constructed by combining ultrasound image-omics and clinical features. Decision curve analysis (DCA) and a nomogram were used to evaluate the clinical application value of the model.

RESULTS

A total of 1561 image-omics features were extracted from ultrasound images. 15 valuable image-omics features were determined by regularization, dimension reduction, and algorithm selection. In the image-omics model, the LR model showed higher prediction efficiency and robustness, with an area under the ROC curve (AUC) of 0.773 in the training set and an AUC of 0.850 in the test set. The clinical model constructed by the boundary of lesions in ultrasound images and the clinical feature CA199 (AUC = 0.875). The combined model had the best prediction performance, with an AUC of 0.872 in the training set and 0.918 in the test set. The combined model showed better clinical benefit according to DCA, and the nomogram score provided clinical prediction solutions.

CONCLUSION

The combined model established with clinical features has good diagnostic ability and can be used to predict lymph node metastasis in patients with PC. It is expected to provide an effective noninvasive method for clinical decision-making, thereby improving the diagnosis and treatment of PC.

Studying the impact of marital status on diagnosis and survival prediction in pancreatic ductal carcinoma using machine learning methods

Pancreatic cancer is a commonly occurring malignant tumor, with pancreatic ductal carcinoma (PDAC) accounting for approximately 95% of cases. According of its poor prognosis, identifying prognostic factors of pancreatic ductal carcinoma can provide physicians with a reliable theoretical foundation when predicting patient survival. This study aimed to analyze the impact of marital status on survival outcomes of PDAC patients using propensity score matching and machine learning. The goal was to develop a prognosis prediction model specific to married patients with PDAC. We extracted a total of 206,968 patient records of pancreatic cancer from the SEER database. To ensure the baseline characteristics of married and unmarried individuals were balanced, we used a 1:1 propensity matching score. We then conducted Kaplan-Meier analysis and Cox proportional-hazards regression to examine the impact of marital status on PDAC survival before and after matching. Additionally, we developed machine learning models to predict 5-year CSS and OS for married patients with PDAC specifically. In total, 24,044 PDAC patients were included in this study. After 1:1 propensity matching, 8043 married patients and 8,043 unmarried patients were successfully enrolled. Multivariate analysis and the Kaplan-Meier curves demonstrated that unmarried individuals had a poorer survival rate than their married counterparts. Among the algorithms tested, the random forest performed the best, with 0.734 5-year CSS and 0.795 5-year OS AUC. This study found a significant association between marital status and survival in PDAC patients. Married patients had the best prognosis, while widowed patients had the worst. The random forest is a reliable model for predicting survival in married patients with PDAC.

Methylation-Specific Electrophoresis

Methylation of CpG sites in the promoter region of genomic DNA is an important epigenetic modification that plays a critical role in gene regulation, particularly in gene silencing. Epigenetic abnormalities, along with genetic alterations, are implicated in carcinogenesis and cancer progression. Numerous studies have investigated the role of epigenetics in cancer using various tools to assess DNA methylation. However, conventional analysis methods for DNA methylation require a large amount of DNA but lack higher sensitivity, making them unsuitable for analysis of samples with high heterogeneity, such as tumor tissues. In this study, we introduce a novel electrophoresis method named "methylation-specific electrophoresis (MSE)," which utilizes a denaturing gradient acrylamide gel. We demonstrate the applicability of the MSE method for DNA methylation analysis of the mucin gene as an example.

Using machine learning models to predict the surgical risk of children with pancreaticobiliary maljunction and biliary dilatation

PURPOSE

To develop machine learning (ML) models to predict the surgical risk of children with pancreaticobiliary maljunction (PBM) and biliary dilatation.

METHODS

The subjects of this study were 157 pediatric patients who underwent surgery for PBM with biliary dilatation between January, 2015 and August, 2022. Using preoperative data, four ML models were developed, including logistic regression (LR), random forest (RF), support vector machine classifier (SVC), and extreme gradient boosting (XGBoost). The performance of each model was assessed via the area under the receiver operator characteristic curve (AUC). Model interpretations were generated by Shapley Additive Explanations. A nomogram was used to validate the best-performing model.

RESULTS

Sixty-eight patients (43.3%) were classified as the high-risk surgery group. The XGBoost model (AUC = 0.822) outperformed the LR (AUC = 0.798), RF (AUC = 0.802) and SVC (AUC = 0.804) models. In all four models, enhancement of the choledochal cystic wall and an abnormal position of the right hepatic artery were the two most important features. Moreover, the diameter of the choledochal cyst, bile duct variation, and serum amylase were selected as key predictive factors by all four models.

CONCLUSIONS

Using preoperative data, the ML models, especially XGBoost, have the potential to predict the surgical risk of children with PBM and biliary dilatation. The nomogram may provide surgeons early warning to avoid intraoperative iatrogenic injury.

Pancreatic Ductal Adenocarcinoma: Update of CT-Based Radiomics Applications in the Pre-Surgical Prediction of the Risk of Post-Operative Fistula, Resectability Status and Prognosis

BACKGROUND

Pancreatic ductal adenocarcinoma (PDAC) is the seventh leading cause of cancer-related deaths worldwide. Surgical resection is the main driver to improving survival in resectable tumors, while neoadjuvant treatment based on chemotherapy (and radiotherapy) is the best option-treatment for a non-primally resectable disease. CT-based imaging has a central role in detecting, staging, and managing PDAC. As several authors have proposed radiomics for risk stratification in patients undergoing surgery for PADC, in this narrative review, we have explored the actual fields of interest of radiomics tools in PDAC built on pre-surgical imaging and clinical variables, to obtain more objective and reliable predictors.

METHODS

The PubMed database was searched for papers published in the English language no earlier than January 2018.

RESULTS

We found 301 studies, and 11 satisfied our research criteria. Of those included, four were on resectability status prediction, three on preoperative pancreatic fistula (POPF) prediction, and four on survival prediction. Most of the studies were retrospective.

CONCLUSIONS

It is possible to conclude that many performing models have been developed to get predictive information in pre-surgical evaluation. However, all the studies were retrospective, lacking further external validation in prospective and multicentric cohorts. Furthermore, the radiomics models and the expression of results should be standardized and automatized to be applicable in clinical practice.

The Utility of Artificial Intelligence in the Diagnosis and Management of Pancreatic Cancer

Artificial intelligence (AI) has made significant advancements in the medical domain in recent years. AI, an expansive field comprising Machine Learning (ML) and, within it, Deep Learning (DL), seeks to emulate the intricate operations of the human brain. It examines vast amounts of data and plays a crucial role in decision-making, overcoming limitations related to human evaluation. DL utilizes complex algorithms to analyze data. ML and DL are subsets of AI that utilize hard statistical techniques that help machines consistently improve at tasks with experience. Pancreatic cancer is more common in developed countries and is one of the leading causes of cancer-related mortality worldwide. Managing pancreatic cancer remains a challenge despite significant advancements in diagnosis and treatment. AI has secured an almost ubiquitous presence in the field of oncological workup and management, especially in gastroenterology malignancies. AI is particularly useful for various investigations of pancreatic carcinoma because it has specific radiological features that enable diagnostic procedures without the requirement of a histological study. However, interpreting and evaluating resulting images is not always simple since images vary as the disease progresses. Secondly, a number of factors may impact prognosis and response to the treatment process. Currently, AI models have been created for diagnosing, grading, staging, and predicting prognosis and treatment response. This review presents the most up-to-date knowledge on the use of AI in the diagnosis and treatment of pancreatic carcinoma.

A pancreatic adenocarcinoma mimicking hepatoid carcinoma of uncertain histogenesis: A case report and literature review

In rare cases, metastatic adenocarcinomas of different origin may exhibit the features of hepatoid carcinoma (HC), a rare malignant epithelial tumor, most commonly occurring in the ovaries and stomach, as well as in the pancreas and biliary ducts. A case of a 72-year-old female patient who developed a highly aggressive, poorly differentiated pancreatic ductal adenocarcinoma with peritoneal carcinomatosis, demonstrating hepatoid differentiation upon conventional hematoxylin and eosin staining is reported in the present study. The patient presented with severe abdominal pain, and the radiological investigations performed revealed ovarian and hepatic tumor masses and peritoneal lesions, which were surgically removed. The gross examination of the peritoneum and omentum revealed multiple solid, firm, grey-white nodules, diffusely infiltrating the adipose tissue. The microscopic examination revealed a malignant epithelial proliferation, composed of polygonal cells with abundant eosinophilic cytoplasm and irregular, pleomorphic nuclei. Certain cells presented with intracytoplasmic mucus inclusions, raising suspicion of a HC with an uncertain histogenesis. Immunohistochemical staining was performed, and the tumor cells were found to be positive for cytokeratin (CK)7, CK18 and mucin 5AC, whereas negative staining for CK20, caudal-type homeobox transcription factor 2, α-fetoprotein, paired box gene 8, GATA-binding protein 3 and Wilms tumor 1 were documented. Thus, the diagnosis of metastatic pancreatic adenocarcinoma was established. The main aim of the present study was to provide further knowledge concerning poorly differentiated metastatic adenocarcinoma resembling HC, emphasizing the histopathological and immunohistochemical features of these malignant lesions and raising awareness of the diagnostic difficulties that may arise, as well as the importance of the use immunohistochemistry in differentiating carcinomas of uncertain histogenesis.

Organoid: Bridging the gap between basic research and clinical practice

Nowadays, the diagnosis and treatment system of malignant tumors has increasingly tended to be more precise and personalized while the existing tumor models are still unable to fully meet the needs of clinical practice. Notably, the emerging organoid platform has been proven to have huge potential in the field of basic-translational medicine, which is expected to promote a paradigm shift in personalized medicine. Here, given the unique advantages of organoid platform, we mainly explore the prominent role of organoid models in basic research and clinical practice from perspectives of tumor biology, tumorigenic microbes-host interaction, clinical decision-making, and regenerative strategy. In addition, we also put forward some practical suggestions on how to construct a new generation of organoid platform, which is destined to vigorously promote the reform of basic-translational medicine.

Deoxyribonucleic acid methylation driven aberrations in pancreatic cancer-related pathways

Pancreatic cancer (PanCa) presents a catastrophic disease with poor overall survival at advanced stages, with immediate requirement of new and effective treatment options. Besides genetic mutations, epigenetic dysregulation of signaling pathway-associated enriched genes are considered as novel therapeutic target. Mechanisms beneath the deoxyribonucleic acid methylation and its utility in developing of epi-drugs in PanCa are under trails. Combinations of epigenetic medicines with conventional cytotoxic treatments or targeted therapy are promising options to improving the dismal response and survival rate of PanCa patients. Recent studies have identified potentially valid pathways that support the prediction that future PanCa clinical trials will include vigorous testing of epigenomic therapies. Epigenetics thus promises to generate a significant amount of new knowledge of biological and medical importance. Our review could identify various components of epigenetic mechanisms known to be involved in the initiation and development of pancreatic ductal adenocarcinoma and related precancerous lesions, and novel pharmacological strategies that target these components could potentially lead to breakthroughs. We aim to highlight the possibilities that exist and the potential therapeutic interventions.

Application of machine learning approaches in predicting clinical outcomes in older adults - a systematic review and meta-analysis

BACKGROUND

Machine learning-based prediction models have the potential to have a considerable positive impact on geriatric care.

DESIGN

Systematic review and meta-analyses.

PARTICIPANTS

Older adults (≥ 65 years) in any setting.

INTERVENTION

Machine learning models for predicting clinical outcomes in older adults were evaluated. A random-effects meta-analysis was conducted in two grouped cohorts, where the predictive models were compared based on their performance in predicting mortality i) under and including 6 months ii) over 6 months.

OUTCOME MEASURES

Studies were grouped into two groups by the clinical outcome, and the models were compared based on the area under the receiver operating characteristic curve metric.

RESULTS

Thirty-seven studies that satisfied the systematic review criteria were appraised, and eight studies predicting a mortality outcome were included in the meta-analyses. We could only pool studies by mortality as there were inconsistent definitions and sparse data to pool studies for other clinical outcomes. The area under the receiver operating characteristic curve from the meta-analysis yielded a summary estimate of 0.80 (95% CI: 0.76 - 0.84) for mortality within 6 months and 0.81 (95% CI: 0.76 - 0.86) for mortality over 6 months, signifying good discriminatory power.

CONCLUSION

The meta-analysis indicates that machine learning models display good discriminatory power in predicting mortality. However, more large-scale validation studies are necessary. As electronic healthcare databases grow larger and more comprehensive, the available computational power increases and machine learning models become more sophisticated; there should be an effort to integrate these models into a larger research setting to predict various clinical outcomes.

Mucins 3A and 3B Are Expressed in the Epithelium of Human Large Airway

Aberrant mucus secretion is a hallmark of chronic obstructive pulmonary disease (COPD). Expression of the membrane-tethered mucins 3A and 3B (MUC3A, MUC3B) in human lung is largely unknown. In this observational cross-sectional study, we recruited subjects 45-65 years old from the general population of Stockholm, Sweden, during the years 2007-2011. Bronchial mucosal biopsies, bronchial brushings, and bronchoalveolar lavage fluid (BALF) were retrieved from COPD patients (n = 38), healthy never-smokers (n = 40), and smokers with normal lung function (n = 40). Protein expression of MUC3A and MUC3B in bronchial mucosal biopsies was assessed by immunohistochemical staining. In a subgroup of subjects (n = 28), MUC3A and MUC3B mRNAs were quantified in bronchial brushings using microarray. Non-parametric tests were used to perform correlation and group comparison analyses. A value of p < 0.05 was considered statistically significant. MUC3A and MUC3B immunohistochemical expression was localized to ciliated cells. MUC3B was also expressed in basal cells. MUC3A and MUC3B immunohistochemical expression was equal in all study groups but subjects with emphysema had higher MUC3A expression, compared to those without emphysema. Smokers had higher mRNA levels of MUC3A and MUC3B than non-smokers. MUC3A and MUC3B mRNA were higher in male subjects and correlated negatively with expiratory air flows. MUC3B mRNA correlated positively with total cell concentration and macrophage percentage, and negatively with CD4/CD8 T cell ratio in BALF. We concluded that MUC3A and MUC3B in large airways may be a marker of disease or may play a role in the pathophysiology of airway obstruction.

Deep Learning for Fully Automated Prediction of Overall Survival in Patients Undergoing Resection for Pancreatic Cancer: A Retrospective Multicenter Study

OBJECTIVE

To develop an imaging-derived biomarker for prediction of overall survival (OS) of pancreatic cancer by analyzing preoperative multiphase contrast-enhanced computed topography (CECT) using deep learning.

BACKGROUND

Exploiting prognostic biomarkers for guiding neoadjuvant and adjuvant treatment decisions may potentially improve outcomes in patients with resectable pancreatic cancer.

METHODS

This multicenter, retrospective study included 1516 patients with resected pancreatic ductal adenocarcinoma (PDAC) from 5 centers located in China. The discovery cohort (n=763), which included preoperative multiphase CECT scans and OS data from 2 centers, was used to construct a fully automated imaging-derived prognostic biomarker-DeepCT-PDAC-by training scalable deep segmentation and prognostic models (via self-learning) to comprehensively model the tumor-anatomy spatial relations and their appearance dynamics in multiphase CECT for OS prediction. The marker was independently tested using internal (n=574) and external validation cohorts (n=179, 3 centers) to evaluate its performance, robustness, and clinical usefulness.

RESULTS

Preoperatively, DeepCT-PDAC was the strongest predictor of OS in both internal and external validation cohorts [hazard ratio (HR) for high versus low risk 2.03, 95% confidence interval (CI): 1.50-2.75; HR: 2.47, CI: 1.35-4.53] in a multivariable analysis. Postoperatively, DeepCT-PDAC remained significant in both cohorts (HR: 2.49, CI: 1.89-3.28; HR: 2.15, CI: 1.14-4.05) after adjustment for potential confounders. For margin-negative patients, adjuvant chemoradiotherapy was associated with improved OS in the subgroup with DeepCT-PDAC low risk (HR: 0.35, CI: 0.19-0.64), but did not affect OS in the subgroup with high risk.

CONCLUSIONS

Deep learning-based CT imaging-derived biomarker enabled the objective and unbiased OS prediction for patients with resectable PDAC. This marker is applicable across hospitals, imaging protocols, and treatments, and has the potential to tailor neoadjuvant and adjuvant treatments at the individual level.

Diagnostic ability of deep learning in detection of pancreatic tumour

Pancreatic cancer is associated with higher mortality rates due to insufficient diagnosis techniques, often diagnosed at an advanced stage when effective treatment is no longer possible. Therefore, automated systems that can detect cancer early are crucial to improve diagnosis and treatment outcomes. In the medical field, several algorithms have been put into use. Valid and interpretable data are essential for effective diagnosis and therapy. There is much room for cutting-edge computer systems to develop. The main objective of this research is to predict pancreatic cancer early using deep learning and metaheuristic techniques. This research aims to create a deep learning and metaheuristic techniques-based system to predict pancreatic cancer early by analyzing medical imaging data, mainly CT scans, and identifying vital features and cancerous growths in the pancreas using Convolutional Neural Network (CNN) and YOLO model-based CNN (YCNN) models. Once diagnosed, the disease cannot be effectively treated, and its progression is unpredictable. That's why there's been a push in recent years to implement fully automated systems that can sense cancer at a prior stage and improve diagnosis and treatment. The paper aims to evaluate the effectiveness of the novel YCNN approach compared to other modern methods in predicting pancreatic cancer. To predict the vital features from the CT scan and the proportion of cancer feasts in the pancreas using the threshold parameters booked as markers. This paper employs a deep learning approach called a Convolutional Neural network (CNN) model to predict pancreatic cancer images. In addition, we use the YOLO model-based CNN (YCNN) to aid in the categorization process. Both biomarkers and CT image dataset is used for testing. The YCNN method was shown to perform well by a cent percent of accuracy compared to other modern techniques in a thorough review of comparative findings.

Machine learning in pancreas surgery, what is new? literature review

Background

Machine learning (ML) is an inquiry domain that aims to establish methodologies that leverage information to enhance performance of various applications. In the healthcare domain, the ML concept has gained prominence over the years. As a result, the adoption of ML algorithms has become expansive. The aim of this scoping review is to evaluate the application of ML in pancreatic surgery.

Methods

We integrated the preferred reporting items for systematic reviews and meta-analyses for scoping reviews. Articles that contained relevant data specializing in ML in pancreas surgery were included.

Results

A search of the following four databases PubMed, Cochrane, EMBASE, and IEEE and files adopted from Google and Google Scholar was 21. The main features of included studies revolved around the year of publication, the country, and the type of article. Additionally, all the included articles were published within January 2019 to May 2022.

Conclusion

The integration of ML in pancreas surgery has gained much attention in previous years. The outcomes derived from this study indicate an extensive literature gap on the topic despite efforts by various researchers. Hence, future studies exploring how pancreas surgeons can apply different learning algorithms to perform essential practices may ultimately improve patient outcomes.

Preoperative Risk Factors in Patients With Pancreatic Cancer

Background

Pancreatic cancer is gastrointestinal cancer with a poor prognosis. Although surgical techniques and chemotherapy have improved treatment outcomes, the 5-year survival rate for pancreatic cancer is less than 10%. In addition, resection of pancreatic cancer is highly invasive and is associated with high rates of postoperative complications and hospital mortality. The Japanese Pancreatic Association states that preoperative body composition assessment may predict postoperative complications. However, although impaired physical function is also a risk factor, few studies have examined it in combination with body composition. We examined preoperative nutritional status and physical function as risk factors for postoperative complications in pancreatic cancer patients.

Methods

Fifty-nine patients with pancreatic cancer who underwent surgical treatment and were discharged alive from January 1, 2018, to March 31, 2021, at the Japanese Red Cross Medical Center. This retrospective study was conducted using electronic medical records and a database of departments. Body composition and physical function were evaluated before and after surgery, and the risk factors between patients with and without complications were compared.

Results

Fifty-nine patients were analyzed: 14 and 45 patients in the uncomplicated and complicated groups, respectively. The major complications were pancreatic fistulas (33%) and infections (22%). There were significant differences in: age, 74.0 (44 - 88) (P = 0.02); walking speed, 0.93 m/s (0.3 - 2.2) (P = 0.01); and fat mass, 16.50 kg (4.7 - 46.2) (P = 0.02), in the patients with complications. On Multivariable logistic regression analysis, age (odds ratio: 2.28; confidence interval (CI): 1.3400 - 569.00; P = 0.03), preoperative fat mass (odds ratio: 2.28; CI: 1.4900 - 168.00; P = 0.02), and walking speed (odds ratio: 0.119; CI: 0.0134 - 1.07; P = 0.05) were identified as risk factors. Walking speed (odds ratio: 0.119; CI: 0.0134 - 1.07; P = 0.05) was the risk factor that was extracted.

Conclusions

Older age, more preoperative fat mass, and decreased walking speed were possible risk factors for postoperative complications.

Supervised learning applied to classifying fallers versus non-fallers among older adults with cancer

INTRODUCTION

Supervised machine learning approaches are increasingly used to analyze clinical data, including in geriatric oncology. This study presents a machine learning approach to understand falls in a cohort of older adults with advanced cancer starting chemotherapy, including fall prediction and identification of contributing factors.

MATERIALS AND METHODS

This secondary analysis of prospectively collected data from the GAP 70+ Trial (NCT02054741; PI: Mohile) enrolled patients aged ≥70 with advanced cancer and ≥ 1 geriatric assessment domain impairment who planned to start a new cancer treatment regimen. Of ≥2000 baseline variables ("features") collected, 73 were selected based on clinical judgment. Machine learning models to predict falls at three months were developed, optimized, and tested using data from 522 patients. A custom data preprocessing pipeline was implemented to prepare data for analysis. Both undersampling and oversampling techniques were applied to balance the outcome measure. Ensemble feature selection was applied to identify and select the most relevant features. Four models (logistic regression [LR], k-nearest neighbor [kNN], random forest [RF], and MultiLayer Perceptron [MLP]) were trained and subsequently tested on a holdout set. Receiver operating characteristic (ROC) curves were generated and area under the curve (AUC) was calculated for each model. SHapley Additive exPlanations (SHAP) values were utilized to further understand individual feature contributions to observed predictions.

RESULTS

Based on the ensemble feature selection algorithm, the top eight features were selected for inclusion in the final models. Selected features aligned with clinical intuition and prior literature. The LR, kNN, and RF models performed equivalently well in predicting falls in the test set, with AUC values 0.66-0.67, and the MLP model showed AUC 0.75. Ensemble feature selection resulted in improved AUC values compared to using LASSO alone. SHAP values, a model-agnostic technique, revealed logical associations between selected features and model predictions.

DISCUSSION

Machine learning techniques can augment hypothesis-driven research, including in older adults for whom randomized trial data are limited. Interpretable machine learning is particularly important, as understanding which features impact predictions is a critical aspect of decision-making and intervention. Clinicians should understand the philosophy, strengths, and limitations of a machine learning approach applied to patient data.

Automated classification of urine biomarkers to diagnose pancreatic cancer using 1-D convolutional neural networks

BACKGROUND

Early diagnosis of Pancreatic Ductal Adenocarcinoma (PDAC) is the main key to surviving cancer patients. Urine proteomic biomarkers which are creatinine, LYVE1, REG1B, and TFF1 present a promising non-invasive and inexpensive diagnostic method of the PDAC. Recent utilization of both microfluidics technology and artificial intelligence techniques enables accurate detection and analysis of these biomarkers. This paper proposes a new deep-learning model to identify urine biomarkers for the automated diagnosis of pancreatic cancers. The proposed model is composed of one-dimensional convolutional neural networks (1D-CNNs) and long short-term memory (LSTM). It can categorize patients into healthy pancreas, benign hepatobiliary disease, and PDAC cases automatically.

RESULTS

Experiments and evaluations have been successfully done on a public dataset of 590 urine samples of three classes, which are 183 healthy pancreas samples, 208 benign hepatobiliary disease samples, and 199 PDAC samples. The results demonstrated that our proposed 1-D CNN + LSTM model achieved the best accuracy score of 97% and the area under curve (AUC) of 98% versus the state-of-the-art models to diagnose pancreatic cancers using urine biomarkers.

CONCLUSION

A new efficient 1D CNN-LSTM model has been successfully developed for early PDAC diagnosis using four proteomic urine biomarkers of creatinine, LYVE1, REG1B, and TFF1. This developed model showed superior performance on other machine learning classifiers in previous studies. The main prospect of this study is the laboratory realization of our proposed deep classifier on urinary biomarker panels for assisting diagnostic procedures of pancreatic cancer patients.

Whole-exome mutational landscape and molecular marker study in mucinous and clear cell ovarian cancer cell lines 3AO and ES2

BACKGROUND

Ovarian cancer is one of the most lethal cancers in women because it is often diagnosed at an advanced stage. The molecular markers investigated thus far have been unsatisfactory.

METHODS

We performed whole-exome sequencing on the human ovarian cancer cell lines 3AO and ES2 and the normal ovarian epithelial cell line IOSE-80. Molecular markers of ovarian cancer were screened from shared mutation genes and copy number variation genes in the 6q21-qter region.

RESULTS

We found that missense mutations were the most common mutations in the gene (93%). The MUC12, FLG and MUC16 genes were highly mutated in 3AO and ES2 cells. Copy number amplification occurred mainly in 4p16.1 and 11q14.3, and copy number deletions occurred in 4q34.3 and 18p11.21. A total of 23 hub genes were screened, of which 16 were closely related to the survival of ovarian cancer patients. The three genes CCDC170, THBS2 and COL14A1 are most significantly correlated with the survival and prognosis of ovarian cancer. In particular, the overall survival of ovarian cancer patients with high CCDC170 gene expression was significantly prolonged (P < 0.001). The expression of CCDC170 in normal tissues was significantly higher than that in ovarian cancer tissues (P < 0.05), and its expression was significantly decreased in advanced ovarian cancer. Western blotting and immunofluorescence assays also showed that the expression of CCDC170 in ovarian cancer cells was significantly lower than that in normal cells (P < 0.001, P < 0.01).

CONCLUSIONS

CCDC170 is expected to become a new diagnostic molecular target and prognostic indicator for ovarian cancer patients, which can provide new ideas for the design of antitumor drugs.

Robotic pancreaticoduodenectomy for pancreatic ductal adenocarcinoma: Analysis of surgical outcomes and long-term prognosis in a high-volume center

BACKGROUND

Robotic pancreaticoduodenectomy (RPD) has been reported to be safe and feasible for patients with pancreatic ductal adenocarcinoma (PDAC) of the pancreatic head. This study aimed to analyze the surgical outcomes and risk factors for poor long-term prognosis of these patients.

METHODS

Data from patients who underwent RPD for PDAC of pancreatic head were retrospectively analyzed. Multivariate Cox regression analysis was used to seek the independent prognostic factors for overall survival (OS), and an online nomogram calculator was developed based on the independent prognostic factors.

RESULTS

Of the 273 patients who met the inclusion criteria, the median operative time was 280.0 minutes, the estimated blood loss was 100.0 mL, the median OS was 23.6 months, and the median recurrence-free survival (RFS) was 14.4 months. Multivariate analysis showed that preoperative carbohydrate antigen 19-9 (CA19-9) [hazard ratio (HR) = 2.607, 95% confidence interval (CI): 1.560-4.354, P < 0.001], lymph node metastasis (HR = 1.429, 95% CI: 1.005-2.034, P = 0.047), tumor moderately (HR = 3.190, 95% CI: 1.813-5.614, P < 0.001) or poorly differentiated (HR = 5.114, 95% CI: 2.839-9.212, P < 0.001), and Clavien-Dindo grade ≥ III (HR = 1.657, 95% CI: 1.079-2.546, P = 0.021) were independent prognostic factors for OS. The concordance index (C-index) of the nomogram constructed based on the above four independent prognostic factors was 0.685 (95% CI: 0.640-0.729), which was significantly higher than that of the AJCC staging (8th edition): 0.541 (95% CI: 0.493-0.589) (P < 0.001).

CONCLUSIONS

This large-scale study indicated that RPD was feasible for PDAC of pancreatic head. Preoperative CA19-9, lymph node metastasis, tumor poorly differentiated, and Clavien-Dindo grade ≥ III were independent prognostic factors for OS. The online nomogram calculator could predict the OS of these patients in a simple and convenient manner.

Construction of a cuproptosis-associated long non-coding RNA risk prediction model for pancreatic adenocarcinoma based on the TCGA database

Cuproptosis is a recently identified controlled process of cell death that functions in tumor development and treatment. Long non-coding RNAs (lncRNAs) are RNA molecules longer than 200 nucleotides that bind to transcription factors and regulate tumor invasion, penetration, metastasis, and prognosis. However, there are limited data on the function of cuproptosis-associated lncRNAs in pancreatic adenocarcinoma. Utilizing data retrieved from the cancer genome atlas database, we devised a risk prediction model of cuproptosis-associated lncRNAs in pancreatic adenocarcinoma, determined their prognostic significance and relationship with tumor immunity, and screened potential therapeutic drugs. Overall, 178 patients were randomized to a training or test group. We then obtained 6 characteristic cuproptosis-associated lncRNAs from the training group, based on which we constructed the risk prediction model, calculated the risk score, and verified the test group results. Subsequently, we performed differential gene analysis, tumor immunoassays, functional enrichment analysis, and potential drug screening. Finally, we found that the prediction model was highly reliable for the prognostic assessment of pancreatic adenocarcinoma patients. Generally, low risk patients had better outcomes than high risk patients. A tumor immunoassay showed that immunotherapy may benefit high risk patients more as there is a greater likelihood that the tumors could escape the immune system in low-risk patients. Through drug screening, we identified ten drugs that may have therapeutic effects on patients with pancreatic adenocarcinoma. In conclusion, this study constructed a risk prediction model of cuproptosis-associated lncRNAs, which can reliably predict the prognosis of pancreatic adenocarcinoma patients, provided a clinical reference for determining treatment approach, and provided some insights into the associations between lncRNAs and cuproptosis. This provides useful insight to aid in the development of therapeutic drugs for pancreatic adenocarcinoma.

Artificial Intelligence for the Prediction and Early Diagnosis of Pancreatic Cancer: Scoping Review (Preprint).. [DOI: 10.2196/preprints.44248] [Citation(s) in RCA: 0] [Impact Index Per Article: 0]

BACKGROUND

Pancreatic cancer is the 12th most common cancer worldwide, with an overall survival rate of 4.9%. Early diagnosis of pancreatic cancer is essential for timely treatment and survival. Artificial intelligence (AI) provides advanced models and algorithms for better diagnosis of pancreatic cancer.

OBJECTIVE

This study aims to explore AI models used for the prediction and early diagnosis of pancreatic cancers as reported in the literature.

METHODS

A scoping review was conducted and reported in line with the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews) guidelines. PubMed, Google Scholar, Science Direct, BioRXiv, and MedRxiv were explored to identify relevant articles. Study selection and data extraction were independently conducted by 2 reviewers. Data extracted from the included studies were synthesized narratively.

RESULTS

Of the 1185 publications, 30 studies were included in the scoping review. The included articles reported the use of AI for 6 different purposes. Of these included articles, AI techniques were mostly used for the diagnosis of pancreatic cancer (14/30, 47%). Radiological images (14/30, 47%) were the most frequently used data in the included articles. Most of the included articles used data sets with a size of &lt;1000 samples (11/30, 37%). Deep learning models were the most prominent branch of AI used for pancreatic cancer diagnosis in the studies, and the convolutional neural network was the most used algorithm (18/30, 60%). Six validation approaches were used in the included studies, of which the most frequently used approaches were k-fold cross-validation (10/30, 33%) and external validation (10/30, 33%). A higher level of accuracy (99%) was found in studies that used support vector machine, decision trees, and k-means clustering algorithms.

CONCLUSIONS

This review presents an overview of studies based on AI models and algorithms used to predict and diagnose pancreatic cancer patients. AI is expected to play a vital role in advancing pancreatic cancer prediction and diagnosis. Further research is required to provide data that support clinical decisions in health care.

Comprehensive machine-learning survival framework develops a consensus model in large-scale multicenter cohorts for pancreatic cancer

As the most aggressive tumor, the outcome of pancreatic cancer (PACA) has not improved observably over the last decade. Anatomy-based TNM staging does not exactly identify treatment-sensitive patients, and an ideal biomarker is urgently needed for precision medicine. Based on expression files of 1280 patients from 10 multicenter cohorts, we screened 32 consensus prognostic genes. Ten machine-learning algorithms were transformed into 76 combinations, of which we selected the optimal algorithm to construct an artificial intelligence-derived prognostic signature (AIDPS) according to the average C-index in the nine testing cohorts. The results of the training cohort, nine testing cohorts, Meta-Cohort, and three external validation cohorts (290 patients) consistently indicated that AIDPS could accurately predict the prognosis of PACA. After incorporating several vital clinicopathological features and 86 published signatures, AIDPS exhibited robust and dramatically superior predictive capability. Moreover, in other prevalent digestive system tumors, the nine-gene AIDPS could still accurately stratify the prognosis. Of note, our AIDPS had important clinical implications for PACA, and patients with low AIDPS owned a dismal prognosis, higher genomic alterations, and denser immune cell infiltrates as well as were more sensitive to immunotherapy. Meanwhile, the high AIDPS group possessed observably prolonged survival, and panobinostat may be a potential agent for patients with high AIDPS. Overall, our study provides an attractive tool to further guide the clinical management and individualized treatment of PACA.

A 70‑RNA model based on SVR and RFE for predicting the pancreatic cancer clinical prognosis

Researches on the prognosis of pancreatic cancer is of great significance to improve the patient treatment effect and survival. Current researches mainly focus on the prediction of the survival status and the determination of prognostic markers. Each patient has its own characteristics, there is no report about the prediction of survival time. However, accurate prediction of survival time is critical for personalized medicine. In this paper, a hybrid algorithm of Support Vector Regression (SVR) and Recursive Feature Elimination (RFE) was used to construct a quantitative prediction model of Overall Survival (OS) for pancreatic cancer patients, 70 RNAs related to OS were determined, including 33 mRNAs, 28 lncRNAs, and 9 miRNAs. The results of 10-fold cross-validation (R2 is 0.9693) and the generalization ability (R2 is 0.9666) showed that the model has reliable predictive performance and these 70 RNAs are important factors influencing the OS of pancreatic cancer patients. To further study the relationship between RNA-RNA interaction and the survival, competitive endogenous RNA (ceRNA) regulation network was constructed. Degree centrality, betweenness centrality and closeness centrality of nodes in the ceRNA network showed that hsa-mir-570, hsa-mir-944, hsa-mir-6506, hsa-mir-3136, MMP16, PLGLB2, HPGD, FUT1, MFSD2A, SULT1E1, SLC13A5, ZNF488, F2RL2, TNFRSF8, TNFSF11, FHDC1, ISLR2 and THSD7B are hub nodes, which are key RNAs closely determining the OS of pancreatic cancer patients.

Atypical Mucin Expression Predicts Worse Overall Survival in Resectable Pancreatic Ductal Adenocarcinoma

Background. Pancreatic ductal adenocarcinoma (PDAC) displays a typical mucin expression pattern which is characterized by MUC1 positive, MUC2 negative, and MUC5AC positive. More and more evidences show that mucins are involved in the development of pancreatic diseases. However, the relationship between mucin expression and prognosis of PDAC patients has been controversial in the past decades; therefore, we aim to figure out the association of mucin expression with survival in PDAC patients who underwent radical resection. Methods. We performed immunohistochemistry (IHC) to detect the expression of MUC1, MUC2, and MUC5AC in resected PDAC specimens from Shanghai Cancer Center, Fudan University (FUSCC, $n=427$ ) and obtained the corresponding clinical statistical data for retrospective study. Kaplan-Meier methods and Mantel-Cox tests were used to compare the survival curves, and the Cox regression model was employed for multivariate analyses to determine the independent risk factors. Survival analysis was also performed in the Queensland Centre for Medical Genomics (QCMG, $n=70$ ) PDAC cohort to verify the conclusion. Results. Both the FUSCC cohort and the QCMG cohort demonstrated that MUC1 absence was significantly correlated with worse overall survival (OS). The presence of MUC2 showed marginal significance in predicting shorter OS of PDAC patients, while MUC5AC had no prognostic value. In the FUSCC cohort, MUC1 absence was associated with increased proportion of stage III PDAC ( $p=0.011$ ), and MUC1 absence and MUC2 presence were associated with tumour perineural aggression ( $p=0.011$ and $p=0.030$ , respectively). Multivariable adjusted hazard ratios (HRs) for mortality of MUC1 and MUC2 were 0.492 (95% CI: 0.274-0.883, $p=0.017$ ) and 1.596 (95% CI: 1.061-2.401, $p=0.025$ ), respectively. Conclusions. MUC1 absence or MUC2 presence is independently associated with poor OS among patients with resectable PDAC. MUC5AC absence tended to be associated with short-term death.

Decreased complement 4d increases poor prognosis in patients with non‑small cell lung cancer combined with gastrointestinal lymph node metastasis

Lung cancer is a common malignancy that is difficult to treat and has a high risk of mortality. Although gastrointestinal lymph node metastasis has long been known to exert major impact on the prognosis of lung cancer, the mechanism of its occurrence and potential biological markers remain elusive. Therefore, the present study retrospectively analyzed data from 132 patients with non-small cell lung cancer (NSCLC) combined with lymph node metastasis between February 2010 and April 2019 from the First Affiliated Hospital of Soochow University (Suzhou, China) and Sichuan Cancer Hospital (Chengdu, China). Overall survival was assessed using Kaplan-Meier analysis and Cox logistic regression model. In addition, a prediction model was constructed based on immune indicators such as complement C3b and C4d (measured by ELISA), before the accuracy of this model was validated using calibration curves for 5-year OS. Among the 132 included patients, a total of 92 (70.0%) succumbed to the disease within 5 years. Multifactorial analysis revealed that complement C3b deficiency increased the risk of mortality by nearly two-fold [hazard ratio (HR)=2.23; 95% CI=1.20-4.14; P=0.017], whilst complement C4d deficiency similarly increased the risk of mortality by two-fold (HR=2.14; 95% CI=1.14-4.00; P=0.012). The variables were subsequently screened using Cox model to construct a prediction model based on complement C3b and C4d levels before a Nomogram plotted. By internal validation for the 132 patients, the Nomogram accurately estimated the risk of mortality, with a corrected C-index of 0.810. External validation of the model in another 50 patients from Sichuan Cancer Hospital revealed an accuracy of 77.0%. Overall, this mortality risk prediction model constructed based on complement levels showed accuracy in assessing the prognosis of patients with metastatic NSCLC. Therefore, complement C3b and C4d have potential for use as biomarkers to predict the risk of mortality in such patients.

Deep Learning-Based Pathology Image Analysis Enhances Magee Feature Correlation With Oncotype DX Breast Recurrence Score

Background

Oncotype DX Recurrence Score (RS) has been widely used to predict chemotherapy benefits in patients with estrogen receptor-positive breast cancer. Studies showed that the features used in Magee equations correlate with RS. We aimed to examine whether deep learning (DL)-based histology image analyses can enhance such correlations.

Methods

We retrieved 382 cases with RS diagnosed between 2011 and 2015 from the Emory University and the Ohio State University. All patients received surgery. DL models were developed to detect nuclei of tumor cells and tumor-infiltrating lymphocytes (TILs) and segment tumor cell nuclei in hematoxylin and eosin (H&E) stained histopathology whole slide images (WSIs). Based on the DL-based analysis, we derived image features from WSIs, such as tumor cell number, TIL number variance, and nuclear grades. The entire patient cohorts were divided into one training set (125 cases) and two validation sets (82 and 175 cases) based on the data sources and WSI resolutions. The training set was used to train the linear regression models to predict RS. For prediction performance comparison, we used independent variables from Magee features alone or the combination of WSI-derived image and Magee features.

Results

The Pearson's correlation coefficients between the actual RS and predicted RS by DL-based analysis were 0.7058 (p-value = 1.32 × 10-13) and 0.5041 (p-value = 1.15 × 10-12) for the validation sets 1 and 2, respectively. The adjusted R 2 values using Magee features alone are 0.3442 and 0.2167 in the two validation sets, respectively. In contrast, the adjusted R 2 values were enhanced to 0.4431 and 0.2182 when WSI-derived imaging features were jointly used with Magee features.

Conclusion

Our results suggest that DL-based digital pathological features can enhance Magee feature correlation with RS.

A Novel Prognostic Score Based on Artificial Intelligence in Hepatocellular Carcinoma: A Long-Term Follow-Up Analysis

Objective

T cell immunity plays an important role in anti-tumor effects and immunosuppression often leads to the development and relapse of cancer. This study aimed to investigate the effect of T cell numbers on the long-term prognosis of patients with hepatocellular carcinoma (HCC) and construct an artificial neural network (ANN) model to evaluate its prognostic value.

Methods

We enrolled 3,427 patients with HCC at Beijing Ditan Hospital, Capital Medical University, and randomly divided them into two groups of 1,861 and 809 patients as the training and validation sets, respectively. Cox regression analysis was used to screen for independent risk factors of survival in patients with HCC. These factors were used to build an ANN model using Python. Concordance index, calibration curve, and decision curve analysis were used to evaluate the model performance.

Results

The 1-year, 3-year, 5-year, and 10-year cumulative overall survival (OS) rates were 66.9%, 45.7%, 34.9%, and 22.6%, respectively. Cox multivariate regression analysis showed that age, white blood cell count, creatinine, total bilirubin, γ-GGT, LDH, tumor size ≥ 5 cm, tumor number ≥ 2, portal vein tumor thrombus, and AFP ≥ 400 ng/ml were independent risk factors for long-term survival in HCC. Antiviral therapy, albumin, T cell, and CD8 T cell counts were independent protective factors. An ANN model was developed for long-term survival. The areas under the receiver operating characteristic (ROC) curve of 1-year, 3-year, and 5-year OS rates by ANNs were 0.838, 0.833, and 0.843, respectively, which were higher than those of the Barcelona Clinic Liver Cancer (BCLC), tumor node metastasis (TNM), Okuda, Chinese University Prognostic Index (CUPI), Cancer of the Liver Italian Program (CLIP), Japan Integrated Staging (JIS), and albumin–bilirubin (ALBI) models (P < 0.0001). According to the ANN model scores, all patients were divided into high-, middle-, and low-risk groups. Compared with low-risk patients, the hazard ratios of 5-year OS of the high-risk group were 8.11 (95% CI: 7.0-9.4) and 6.13 (95% CI: 4.28-8.79) (P<0.0001) in the training and validation sets, respectively.

Conclusion

High levels of circulating T cells and CD8 + T cells in peripheral blood may benefit the long-term survival of patients with HCC. The ANN model has a good individual prediction performance, which can be used to assess the prognosis of HCC and lay the foundation for the implementation of precision treatment in the future.

Identification and Validation of Constructing the Prognostic Model With Four DNA Methylation-Driven Genes in Pancreatic Cancer

Background: Pancreatic cancer (PC) is a highly aggressive gastrointestinal tumor and has a poor prognosis. Evaluating the prognosis validly is urgent for PC patients. In this study, we utilized the RNA-sequencing (RNA-seq) profiles and DNA methylation expression data comprehensively to develop and validate a prognostic signature in patients with PC.

Methods: The integrated analysis of RNA-seq, DNA methylation expression profiles, and relevant clinical information was performed to select four DNA methylation-driven genes. Then, a prognostic signature was established by the univariate, multivariate Cox, and least absolute shrinkage and selection operator (LASSO) regression analyses in The Cancer Genome Atlas (TCGA) dataset. GSE62452 cohort was utilized for external validation. Finally, a nomogram model was set up and evaluated by calibration curves.

Results: Nine DNA methylation-driven genes that were related to overall survival (OS) were identified. After multivariate Cox and LASSO regression analyses, four of these genes (RIC3, MBOAT2, SEZ6L, and OAS2) were selected to establish the predictive signature. The PC patients were stratified into two groups according to the median risk score, of which the low-risk group displayed a prominently favorable OS compared with the high-risk group, whether in the training (p < 0.001) or validation (p < 0.01) cohort. Then, the univariate and multivariate Cox regression analyses showed that age, grade, risk score, and the number of positive lymph nodes were significantly associated with OS in PC patients. Therefore, we used these clinical variables to construct a nomogram; and its performance in predicting the 1-, 2-, and 3-year OS of patients with PC was assessed via calibration curves.

Conclusion: A prognostic risk score signature was built with the four alternative DNA methylation-driven genes. Furthermore, in combination with the risk score, age, grade, and the number of positive lymph nodes, a nomogram was established for conveniently predicting the individualized prognosis of PC patients.

Emerging Role of Epigenetic Alterations as Biomarkers and Novel Targets for Treatments in Pancreatic Ductal Adenocarcinoma

Simple Summary

Epigenetic alterations cause changes in gene expression without affecting the DNA sequence and are found to affect several molecular pathways in pancreatic tumors. Such changes are reversible, making them potential drug targets. Furthermore, epigenetic alterations occur early in the disease course and may thus be explored for early detection. Hence, a deeper understanding of epigenetics in pancreatic cancer may lead to improved diagnostics, treatments, and prognostication.

Abstract

Pancreatic ductal adenocarcinoma (PDAC) is a lethal disease with limited treatment options. Emerging evidence shows that epigenetic alterations are present in PDAC. The changes are potentially reversible and therefore promising therapeutic targets. Epigenetic aberrations also influence the tumor microenvironment with the potential to modulate and possibly enhance immune-based treatments. Epigenetic marks can also serve as diagnostic screening tools, as epigenetic changes occur at early stages of the disease. Further, epigenetics can be used in prognostication. The field is evolving, and this review seeks to provide an updated overview of the emerging role of epigenetics in the diagnosis, treatment, and prognostication of PDAC.

Random survival forest model identifies novel biomarkers of event-free survival in high-risk pediatric acute lymphoblastic leukemia

High-risk pediatric B-ALL patients experience 5-year negative event rates up to 25%. Although some biomarkers of relapse are utilized in the clinic, their ability to predict outcomes in high-risk patients is limited. Here, we propose a random survival forest (RSF) machine learning model utilizing interpretable genomic inputs to predict relapse/death in high-risk pediatric B-ALL patients. We utilized whole exome sequencing profiles from 156 patients in the TARGET-ALL study (with samples collected at presentation) further stratified into training and test cohorts (109 and 47 patients, respectively). To avoid overfitting and facilitate the interpretation of machine learning results, input genomic variables were engineered using a stepwise approach involving univariable Cox models to select variables directly associated with outcomes, genomic coordinate-based analysis to select mutational hotspots, and correlation analysis to eliminate feature co-linearity. Model training identified 7 genomic regions most predictive of relapse/death-free survival. The test cohort error rate was 12.47%, and a polygenic score based on the sum of the top 7 variables effectively stratified patients into two groups, with significant differences in time to relapse/death (log-rank P = 0.001, hazard ratio = 5.41). Our model outperformed other EFS modeling approaches including an RSF using gold-standard prognostic variables (error rate = 24.35%). Validation in 174 standard-risk patients and 3 patients who failed to respond to induction therapy confirmed that our RSF model and polygenic score were specific to high-risk disease. We propose that our feature selection/engineering approach can increase the clinical interpretability of RSF, and our polygenic score could be utilized for enhance clinical decision-making in high-risk B-ALL.

Application of artificial intelligence to pancreatic adenocarcinoma

BACKGROUND AND OBJECTIVES

Pancreatic cancer (PC) is one of the deadliest cancers worldwide although substantial advancement has been made in its comprehensive treatment. The development of artificial intelligence (AI) technology has allowed its clinical applications to expand remarkably in recent years. Diverse methods and algorithms are employed by AI to extrapolate new data from clinical records to aid in the treatment of PC. In this review, we will summarize AI's use in several aspects of PC diagnosis and therapy, as well as its limits and potential future research avenues.

METHODS

We examine the most recent research on the use of AI in PC. The articles are categorized and examined according to the medical task of their algorithm. Two search engines, PubMed and Google Scholar, were used to screen the articles.

RESULTS

Overall, 66 papers published in 2001 and after were selected. Of the four medical tasks (risk assessment, diagnosis, treatment, and prognosis prediction), diagnosis was the most frequently researched, and retrospective single-center studies were the most prevalent. We found that the different medical tasks and algorithms included in the reviewed studies caused the performance of their models to vary greatly. Deep learning algorithms, on the other hand, produced excellent results in all of the subdivisions studied.

CONCLUSIONS

AI is a promising tool for helping PC patients and may contribute to improved patient outcomes. The integration of humans and AI in clinical medicine is still in its infancy and requires the in-depth cooperation of multidisciplinary personnel.

Information extraction for prognostic stage prediction from breast cancer medical records using NLP and ML

For cancer prediction, the prognostic stage is the main factor that helps medical experts to decide the optimal treatment for a patient. Specialists study prognostic stage information from medical reports, often in an unstructured form, and take a larger review time. The main objective of this study is to suggest a generic clinical decision-unifying staging method to extract the most reliable prognostic stage information of breast cancer from medical records of various health institutions. Additional prognostic elements should be extracted from medical reports to identify the cancer stage for getting an exact measure of cancer and improving care quality. This study has collected 465 pathological and clinical reports of breast cancer sufferers from India's reputed medical institutions. The unstructured records were found distinct from each institute. Anatomic and biologic factors are extracted from medical records using the natural language processing, machine learning and rule-based method for prognostic stage detection. This study has extracted anatomic stage, grade, estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2) from medical reports with high accuracy and predicted prognostic stage for both regions. The prognostic stage prediction's average accuracy is found 92% and 82% in rural and urban areas, respectively. It was essential to combine biological and anatomical elements under a single prognostic staging method. A generic clinical decision-unifying staging method for prognostic stage detection with great accuracy in various institutions of different regional areas suggests that the proposed research improves the prognosis of breast cancer.

Advances in the epidemiology of pancreatic cancer: Trends, risk factors, screening, and prognosis

Pancreatic cancer is a malignancy with poor prognosis and high mortality. The recent increase in pancreatic cancer incidence and mortality has resulted in an increased number of studies on its epidemiology. This comprehensive and systematic literature review summarizes the advances in the epidemiology of pancreatic cancer, including its epidemiological trends, risk factors, risk prediction models, screening modalities, and prognosis. The risk factors for pancreatic cancers can be categorized as those related to individual characteristics, lifestyle and environment, and disease status. Several prediction models for pancreatic cancer have been developed in populations with new-onset diabetes or a family history of pancreatic cancer; however, these models require further validation. Despite recent progress in pancreatic cancer screening, the quantity and quality of related studies are also unsatisfactory, especially with respect to the identification of high-risk populations and development of effective screening modality. Apart from the populations with familial genetic risk and those at a high risk of sporadic pancreatic cancer, risk factors such as new-onset diabetes may be a new direction for timely intervention. We hope this work will provide new ideas for further prevention and treatment of pancreatic cancer.

A Nomogram Based on a Collagen Feature Support Vector Machine for Predicting the Treatment Response to Neoadjuvant Chemoradiotherapy in Rectal Cancer Patients

BACKGROUND

The relationship between collagen features (CFs) in the tumor microenvironment and the treatment response to neoadjuvant chemoradiotherapy (nCRT) is still unknown. This study aimed to develop and validate a perdition model based on the CFs and clinicopathological characteristics to predict the treatment response to nCRT among locally advanced rectal cancer (LARC) patients.

METHODS

In this multicenter, retrospective analysis, 428 patients were included and randomly divided into a training cohort (299 patients) and validation cohort (129 patients) [7:3 ratio]. A total of 11 CFs were extracted from a multiphoton image of pretreatment biopsy, and a support vector machine (SVM) was then used to construct a CFs-SVM classifier. A prediction model was developed and presented with a nomogram using multivariable analysis. Further validation of the nomogram was performed in the validation cohort.

RESULTS

The CFs-SVM classifier, which integrated collagen area, straightness, and crosslink density, was significantly associated with treatment response. Predictors contained in the nomogram included the CFs-SVM classifier and clinicopathological characteristics by multivariable analysis. The CFs nomogram demonstrated good discrimination, with area under the receiver operating characteristic curves (AUROCs) of 0.834 in the training cohort and 0.854 in the validation cohort. Decision curve analysis indicated that the CFs nomogram was clinically useful. Moreover, compared with the traditional clinicopathological model, the CFs nomogram showed more powerful discrimination in determining the response to nCRT.

CONCLUSIONS

The CFs-SVM classifier based on CFs in the tumor microenvironment is associated with treatment response, and the CFs nomogram integrating the CFs-SVM classifier and clinicopathological characteristics is useful for individualized prediction of the treatment response to nCRT among LARC patients.

Pancreatic Ductal Adenocarcinoma: The Dawn of the Era of Nuclear Medicine?

Pancreatic ductal adenocarcinoma (PDAC), accounting for 90-95% of all pancreatic tumors, is a highly devastating disease associated with poor prognosis. The lack of accurate diagnostic tests and failure of conventional therapies contribute to this pejorative issue. Over the last decade, the advent of theranostics in nuclear medicine has opened great opportunities for the diagnosis and treatment of several solid tumors. Several radiotracers dedicated to PDAC imaging or internal vectorized radiotherapy have been developed and some of them are currently under clinical consideration. The functional information provided by Positron Emission Tomography (PET) or Single Photon Emission Computed Tomography (SPECT) could indeed provide an additive diagnostic value and thus help in the selection of patients for targeted therapies. Moreover, the therapeutic potential of β-- and α-emitter-radiolabeled agents could also overcome the resistance to conventional therapies. This review summarizes the current knowledge concerning the recent developments in the nuclear medicine field for the management of PDAC patients.

Mucin expression, epigenetic regulation and patient survival: A toolkit of prognostic biomarkers in epithelial cancers

Twenty mucin genes have been identified and classified in two groups (encoding secreted and membrane-bound proteins). Secreted mucins participate in mucus formation by assembling a 3-dimensional network via oligomerization, whereas membrane-bound mucins are anchored to the outer membrane mediating extracellular interactions and cell signaling. Both groups have been associated with carcinogenesis progression in epithelial cancers, and are therefore considered as potential therapeutic targets. In the present review, we discuss the link between mucin expression patterns and patient survival and propose mucins as prognosis biomarkers of epithelial cancers (esophagus, gastric, pancreatic, colorectal, lung, breast or ovarian cancers). We also investigate the relationship between mucin expression and overall survival in the TCGA dataset. In particular, epigenetic mechanisms regulating mucin gene expression, such as aberrant DNA methylation and histone modification, are interesting as they are also associated with diagnosis or prognosis significance. Indeed, mucin hypomethylation has been shown to be associated with carcinogenesis progression and was linked to prognosis in colon cancer or pancreatic cancer patients. Finally we describe the relationship between mucin expression and non-coding RNAs that also may serve as biomarkers. Altogether the concomitant knowledge of specific mucin-pattern expression and epigenetic regulation could be translated as biomarkers with a better specificity/sensitivity performance in several epithelial cancers.

Prognostic value of Glypican family genes in early-stage pancreatic ductal adenocarcinoma after pancreaticoduodenectomy and possible mechanisms

Background

This study explored the prognostic significance of Glypican (GPC) family genes in patients with pancreatic ductal adenocarcinoma (PDAC) after pancreaticoduodenectomy using data from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO).

Methods

A total of 112 PDAC patients from TCGA and 48 patients from GEO were included in the analysis. The relationship between overall survival and the expression of GPC family genes as well as basic clinical characteristics was analyzed using the Kaplan-Meier method with the log-rank test. Joint effects survival analysis was performed to further examine the relationship between GPC genes and prognosis. A prognosis nomogram was established based on clinical characteristics and prognosis-related genes. Prognosis-related genes were investigated by genome-wide co-expression analysis and gene set enrichment analysis (GSEA) was carried out to identify potential mechanisms of these genes affecting prognosis.

Results

In TCGA database, high expression of GPC2, GPC3, and GPC5 was significantly associated with favorable survival (log-rank P = 0.031, 0.021, and 0.028, respectively; adjusted P value = 0.005, 0.022, and 0.020, respectively), and joint effects analysis of these genes was effective for prognosis prediction. The prognosis nomogram was applied to predict the survival probability using the total scores calculated. Genome-wide co-expression and GSEA analysis suggested that the GPC2 may affect prognosis through sequence-specific DNA binding, protein transport, cell differentiation and oncogenic signatures (KRAS, RAF, STK33, and VEGFA). GPC3 may be related to cell adhesion, angiogenesis, inflammatory response, signaling pathways like Ras, Rap1, PI3K-Akt, chemokine, GPCR, and signatures like cyclin D1, p53, PTEN. GPC5 may be involved in transcription factor complex, TFRC1, oncogenic signatures (HOXA9 and BMI1), gene methylation, phospholipid metabolic process, glycerophospholipid metabolism, cell cycle, and EGFR pathway.

Conclusion

GPC2, GPC3, and GPC5 expression may serve as prognostic indicators in PDAC, and combination of these genes showed a higher efficiency for prognosis prediction.

Complement sC5b-9 and CH50 increase the risk of cancer-related mortality in patients with non-small cell lung cancer

Objectives: Immunologic dysfunction occurred in most of patients with non-small cell lung cancer (NSCLC), which worsened the overall survival (OS) of patients. Complement activation plays a significant role in abnormal activation of immune system. However, the prognostic value of complement components such as CH50 and sC5b-9 in NSCLC patients remains unclear. This study evaluated the risk factors of NSCLC and created a prediction model.

Methods: A real-world study was conducted including data from 928 patients with NSCLC between April 1, 2005 and June 1, 2015. CH50 and sC5b-9 were recorded during the admission. Cox proportional hazard model was applied for survival analyses and for assessing risk factors of cancer-related mortality and to create a nomogram for prediction. The accuracy of the model was evaluated by C-index and calibration curve.

Results: In this study, the mortality in group with high CH50 level (≥ 480.56 umol/L) was 92.0%. Based on univariate analysis, we put factors (P <0.05) into a multivariate regression model, patients with high CH50 level (P <0.001, HR=1.59) and sC5b-9 >1422.18 μmol/L (P <0.001, HR=2.28) remained statistically factors for worsened OS and regarded as independent risk factors. These independently associated risk factors were applied to establish an OS estimation nomogram. Nomogram revealed good accuracy in estimating the risk, with a bootstrap-corrected C index of 0.741.

Conclusion: sC5b-9 and CH50 increased the risk of cancer-related mortality in patients with NSCLC. Nomogram based on multivariate analysis demonstrated good accuracy in estimating the risk of overall mortality.

Unsupervised Hierarchical Clustering of Pancreatic Adenocarcinoma Dataset from TCGA Defines a Mucin Expression Profile that Impacts Overall Survival

Simple Summary

Pancreatic cancer has a dramatic outcome (survival curve < 6 months) that is the consequence of late diagnosis and the lack of efficient therapy. We investigated the relationship between the 22 mucin gene expression and the patient survival in pancreatic cancer datasets that provide a comprehensive mapping of transcriptomic alterations occurring during carcinogenesis. Using unsupervised hierarchical clustering analysis of mucin gene expression patterns, we identified two major clusters of patients: atypical mucin signature (#1; MUC15, MUC14/EMCN, and MUC18/MCAM) and membrane-bound mucin signature (#2; MUC1, -4, -16, -17, -20, and -21). The signature #2 is associated with shorter overall survival, suggesting that the pattern of membrane-bound mucin expression could be a new prognostic marker for PDAC patients.

Abstract

Mucins are commonly associated with pancreatic ductal adenocarcinoma (PDAC) that is a deadly disease because of the lack of early diagnosis and efficient therapies. There are 22 mucin genes encoding large O-glycoproteins divided into two major subgroups: membrane-bound and secreted mucins. We investigated mucin expression and their impact on patient survival in the PDAC dataset from The Cancer Genome Atlas (PAAD-TCGA). We observed a statistically significant increased messenger RNA (mRNA) relative level of most of the membrane-bound mucins (MUC1/3A/4/12/13/16/17/20), secreted mucins (MUC5AC/5B), and atypical mucins (MUC14/18) compared to normal pancreas. We show that MUC1/4/5B/14/17/20/21 mRNA levels are associated with poorer survival in the high-expression group compared to the low-expression group. Using unsupervised clustering analysis of mucin gene expression patterns, we identified two major clusters of patients. Cluster #1 harbors a higher expression of MUC15 and atypical MUC14/MUC18, whereas cluster #2 is characterized by a global overexpression of membrane-bound mucins (MUC1/4/16/17/20/21). Cluster #2 is associated with shorter overall survival. The patient stratification appears to be independent of usual clinical features (tumor stage, differentiation grade, lymph node invasion) suggesting that the pattern of membrane-bound mucin expression could be a new prognostic marker for PDAC patients.

Microdissected pancreatic cancer proteomes reveal tumor heterogeneity and therapeutic targets

Pancreatic ductal adenocarcinoma (PDAC) is characterized by a relative paucity of cancer cells that are surrounded by an abundance of nontumor cells and extracellular matrix, known as stroma. The interaction between stroma and cancer cells contributes to poor outcome, but how proteins from these individual compartments drive aggressive tumor behavior is not known. Here, we report the proteomic analysis of laser-capture microdissected (LCM) PDAC samples. We isolated stroma, tumor, and bulk samples from a cohort with long- and short-term survivors. Compartment-specific proteins were measured by mass spectrometry, yielding what we believe to be the largest PDAC proteome landscape to date. These analyses revealed that, in bulk analysis, tumor-derived proteins were typically masked and that LCM was required to reveal biology and prognostic markers. We validated tumor CALB2 and stromal COL11A1 expression as compartment-specific prognostic markers. We identified and functionally addressed the contributions of the tumor cell receptor EPHA2 to tumor cell viability and motility, underscoring the value of compartment-specific protein analysis in PDAC.

Application of artificial intelligence for the diagnosis, treatment, and prognosis of pancreatic cancer

Pancreatic cancer is a complex cancer of the digestive tract. Diagnosis and treatment can be very difficult because of unclear early symptoms, the deep anatomical location of cancer tissues, and the high degree of cancer cell invasion. The prognosis is extremely poor; the 5-year survival rate of patients with pancreatic cancer is less than 1%. Artificial intelligence (AI) has great potential for application in the medical field. In addition to AI-based applications, such as disease data processing, imaging, and pathological image recognition, robotic surgery has revolutionized surgical procedures. To better understand the current role of AI in pancreatic cancer and predict future development trends, this article comprehensively reports the application of AI to the diagnosis, treatment, and prognosis of pancreatic cancer.