Revisiting C.G. Moertel's land of small tumors

Leveraging machine learning techniques for predicting pancreatic neuroendocrine tumor grades using biochemical and tumor markers

BACKGROUND

The incidence of pancreatic neuroendocrine tumors (PNETs) is now increasing rapidly. The tumor grade of PNETs significantly affects the treatment strategy and prognosis. However, there is still no effective way to non-invasively classify PNET grades. Machine learning (ML) algorithms have shown potential in improving the prediction accuracy using comprehensive data.

AIM

To provide a ML approach to predict PNET tumor grade using clinical data.

METHODS

The clinical data of histologically confirmed PNET cases between 2012 and 2018 were collected. A method of minimum P for the Chi-square test was used to divide the continuous variables into binary variables. The continuous variables were transformed into binary variables according to the cutoff value, while the P value was minimum. Four classical supervised ML models, including logistic regression, support vector machine (SVM), linear discriminant analysis (LDA) and multi-layer perceptron (MLP) were trained by clinical data, and the models were labeled with the pathological tumor grade of each PNET patient. The performance of each model, including the weight of the different parameters, were evaluated.

RESULTS

In total, 91 PNET cases were included in this study, in which 32 were G1, 48 were G2 and 11 were G3. The results showed that there were significant differences among the clinical parameters of patients with different grades. Patients with higher grades tended to have higher values of total bilirubin, alpha fetoprotein, carcinoembryonic antigen, carbohydrate antigen 19-9 and carbohydrate antigen 72-4. Among the models we used, LDA performed best in predicting the PNET tumor grade. Meanwhile, MLP had the highest recall rate for G3 cases. All of the models stabilized when the sample size was over 70 percent of the total, except for SVM. Different parameters varied in affecting the outcomes of the models. Overall, alanine transaminase, total bilirubin, carcinoembryonic antigen, carbohydrate antigen 19-9 and carbohydrate antigen 72-4 affected the outcome greater than other parameters.

CONCLUSION

ML could be a simple and effective method in non-invasively predicting PNET grades by using the routine data obtained from the results of biochemical and tumor markers.

Animal models of spontaneous pancreatic neuroendocrine tumors

Pancreatic neuroendocrine tumors (PNETs) are usually low-grade neoplasms derived from the endocrine pancreas. PNETs can be functioning and cause well-described hormonal hypersecretion syndromes or non-functioning and cause only tumor mass effect. PNETs appear to be more common recently likely due to incidental detection by imaging. Although the diagnosis and management of PNETs have been evolving rapidly, much remains to be studied in the areas of molecular pathogenesis, molecular markers of tumor behavior, early detection, and targeted drug therapy. Unique challenges facing PNETs studies are long disease course, the deep location of pancreas and difficult access to pancreatic tissue, and the variety of tumors, which make animal models valuable tools for PNETs studies. Existing animal models of PNETs have provided insights into the pathogenesis and natural history of human PNETs. Future studies on animal models of PNETs should address early tumor detection, molecular markers of tumor behavior, and novel targeted therapies.

Pancreatic neuroendocrine tumors: current opinions on a rare, but potentially curable neoplasm

Pancreatic neuroendocrine tumors (PNETs) share a unique genetic identity, functional behavior, and clinical course. Compared with tumors of the exocrine pancreas, they are rare and show a different biologic behavior and prognosis. On the basis of data from recent studies, all PNETs, outside of small insulinomas, should be considered potentially malignant and treated accordingly. Untreated tumors have a high possibility to grow locally into adjacent structures or spread to distant organs. Although surgical excision irrespective of tumor functioning or nonfunctioning state remains the cornerstone of therapy, providing the best disease-free and survival rates to date, the understanding of the genetic nature of the disease yields new 'targets' to consider in drug development. The aim of this review is to summarize all recent advances of genetic research and new drug development in terms of PNETs, especially their genetic identity and subsequent alterations leading to the development of near or total malignant activity, and the new medical treatment strategies of this potentially curable disease on the basis of therapeutical agents acting, where possible, at the genetic level.

Update on the management of gastroenteropancreatic neuroendocrine tumors with emphasis on the role of imaging

OBJECTIVE

The purposes of this article are to review the current management of gastroenteropancreatic neuroendocrine tumors (GEP-NETs) based on the 2012 National Comprehensive Cancer Network guidelines and to describe the role of imaging in a multidisciplinary approach.

CONCLUSION

The management of GEP-NETs has become complex, requiring a multidisciplinary approach. The World Health Organization classification of GEP-NETs has been revised; the U.S. Food and Drug Administration has approved molecular targeted agents (sunitinib, everolimus) for the treatment of pancreatic NETs; and the National Comprehensive Cancer Network clinical practice guidelines have been updated.

Animal models and cell lines of pancreatic neuroendocrine tumors

Pancreatic neuroendocrine tumors (PNETs), also known as islet cell tumors, are mostly indolent neoplasms that probably arise from a network of endocrine cells that includes islet cells and pluripotent precursors in the pancreatic ductal epithelium. The incidence and prevalence of PNETs continue to rise in recent years because of more sensitive detection. The molecular pathogenesis, early detection, molecular predictors of tumor behavior, and targeted drug therapy of PNETs are not well understood and require additional basic and translational research. The rarity and indolent nature of these tumors, difficulty of access to appropriate patient tissue samples, and varying histopathology and secreted hormones pose particular challenges to PNET researchers. Animal models and cell lines are indispensable tools for investigating the pathogenesis, pathophysiology, mechanisms for tumor invasion and metastasis, and therapeutics of PNETs. This review summarizes currently available animal models and cell lines of PNETs, which have provided valuable insights into the pathogenesis and natural history of human PNETs. In the future, animal models and cell lines of PNETs should also be used to study early tumor detection and molecular predictors of tumor behavior and to test the responses to, and mechanisms for, novel targeted drug therapies.

Pancreatic neuroendocrine tumors: biology, diagnosis,and treatment

Pancreatic neuroendocrine tumors (PNETs), a group of endocrine tumors arising in the pancreas, are among the most common neuroendocrine tumors. The genetic causes of familial and sporadic PNETs are somewhat understood, but their molecular pathogenesis remains unknown. Most PNETs are indolent but have malignant potential. The biological behavior of an individual PNET is unpredictable; higher tumor grade, lymph node and liver metastasis, and larger tumor size generally indicate a less favorable prognosis. Endocrine testing, imaging, and histological evidence are necessary to accurately diagnose PNETs. A 4-pronged aggressive treatment approach consisting of surgery, locoregional therapy, systemic therapy, and complication control has become popular in academic centers around the world. The optimal application of the multiple systemic therapeutic modalities is under development; efficacy, safety, availability, and cost should be considered when treating a specific patient. The clinical presentation, diagnosis, and treatment of specific types of PNETs and familial PNET syndromes, including the novel Mahvash disease, are summarized.

Multimodal liver-directed management of neuroendocrine hepatic metastases

A preponderance of patients with neuroendocrine tumors (NETs) will experience hepatic metastases during the course of their disease. Many diagnoses of NETs are made only after the neoplasms have spread from their primary gastroenteropancreatic sites to the liver. This paper reviews current evidence-based treatments for neuroendocrine hepatic metastases, encompassing surgery, hepatic artery embolization (HAE) and chemoembolization (HACE), radioembolization, hepatic artery infusion (HAI), thermal ablation (radiofrequency, microwave, and cryoablation), alcohol ablation, and liver transplantation as therapeutic modalities. Consideration of a multidisciplinary approach to liver-directed therapy is strongly encouraged to limit morbidity and mortality in this patient population.