Does surgery provide a survival advantage in non-disseminated poorly differentiated gastroenteropancreatic neuroendocrine neoplasms?

Development and Validation of a Novel Machine Learning Model to Predict the Survival of Patients with Gastrointestinal Neuroendocrine Neoplasms

INTRODUCTION

Well-calibrated models for personalized prognostication of patients with gastrointestinal neuroendocrine neoplasms (GINENs) are limited. This study aimed to develop and validate a machine-learning model to predict the survival of patients with GINENs.

METHODS

Oblique random survival forest (ORSF) model, Cox proportional hazard risk model, Cox model with least absolute shrinkage and selection operator penalization, CoxBoost, Survival Gradient Boosting Machine, Extreme Gradient Boosting survival regression, DeepHit, DeepSurv, DNNSurv, logistic-hazard model, and PC-hazard model were compared. We further tuned hyperparameters and selected variables for the best-performing ORSF. Then, the final ORSF model was validated.

RESULTS

A total of 43,444 patients with GINENs were included. The median (interquartile range) survival time was 53 (19-102) months. The ORSF model performed best, in which age, histology, M stage, tumor size, primary tumor site, sex, tumor number, surgery, lymph nodes removed, N stage, race, and grade were ranked as important variables. However, chemotherapy and radiotherapy were not necessary for the ORSF model. The ORSF model had an overall C index of 0.86 (95% confidence interval, 0.85-0.87). The area under the receiver operation curves at 1, 3, 5, and 10 years were 0.91, 0.89, 0.87, and 0.80, respectively. The decision curve analysis showed superior clinical usefulness of the ORSF model than the American Joint Committee on Cancer Stage. A nomogram and an online tool were given.

CONCLUSION

The machine learning ORSF model could precisely predict the survival of patients with GINENs, with the ability to identify patients at high risk for death and probably guide clinical practice.

Management of Gastric Neuroendocrine Tumors: A Review

Gastric neuroendocrine tumors (G-NET) are rare tumors arising from enterochromaffin-like cells of the gastric mucosa. They belong to a larger group called gastroenteropancreatic neuroendocrine tumors and are classified as low, intermediate, or high-grade tumors based on their proliferative indices. They are further categorized into three subtypes based on their morphologic characteristics, pathogenesis, and behavior. Types 1 and 2 tumors are characterized by elevated serum gastrin and are usually multifocal. They typically occur in the setting of atrophic gastritis or MEN1/Zollinger Ellison syndrome, respectively. Type 2 tumors are associated with the most symptoms, such as abdominal pain and diarrhea. Type 3 tumors are associated with normal serum gastrin, are usually solitary, and occur sporadically. This type has the most aggressive phenotype and metastatic potential. Treatment and prognosis for G-NET is dependent on their type, size, and stage. Type 1 has the best prognosis, and Type 3 has the worst. This review discusses the presentation, workup, and surgical management of these tumors.

European Neuroendocrine Tumor Society (ENETS) 2023 guidance paper for digestive neuroendocrine carcinoma

This ENETS guidance paper, developed by a multidisciplinary working group, provides up-to-date and practical advice on the diagnosis and management of digestive neuroendocrine carcinoma, based on recent developments and study results. These recommendations aim to pave the road for more standardized care for our patients resulting in improved outcomes. Prognosis is generally poor for digestive NEC, most are advanced at diagnosis and median survival in metastatic disease is 11-12 months. Surgery can be of benefit for localized disease after extensive preoperative imaging. Carboplatin in combination with etoposide is recommended as first-line treatment for metastatic disease. Irinotecan with fluoropyrimidines has the best evidence as second-line treatment. Immunotherapy plays a minor role in biomarker-unselected patients. Molecular profiling if available is encouraged to identify new targets. More prospective clinical trials are highly needed to fulfil the unmet needs in this field, especially on new predictive and prognostic biomarkers and to improve survival of patients with advanced disease.

Surgical Management of G3 Gastroenteropancreatic Neuroendocrine Neoplasms: A Systematic Review and Meta-analysis

BACKGROUND

Grade 3 (G3) gastroenteropancreatic (GEP) neuroendocrine neoplasms (NENs) are rare, aggressive tumors with poor prognosis. The World Health Organization 2017 and 2019 classifications further subdivided G3 NENs into G3 neuroendocrine tumors (NETs) and neuroendocrine carcinomas (NECs). Current guidelines favor medical management in most of these patients, and the role of surgical management is not well defined. We performed a systematic literature review and meta-analysis of surgical management versus nonsurgical management for G3 GEP NENs.

MATERIALS AND METHODS

A PRISMA-compliant systematic review of the MEDLINE, Embase, Scopus, and Cochrane Library databases (end-of-search date: 16 July 2021) was conducted. Individual patient survival data were reconstructed, and random-effects meta-analyses were performed.

RESULTS

Fourteen studies comprising 1810 surgical and 910 nonsurgical patients were systematically reviewed. Publication bias adjusted meta-analysis of 12 studies (1788 surgical and 857 nonsurgical patients) showed increased overall survival (OS) after surgical compared with nonsurgical management for G3 GEP NENs [hazard ratio (HR) 0.40, 95% confidence interval (CI) 0.31-0.53]. Subgroup meta-analyses showed increased OS after surgical management for both pancreatic and gastrointestinal primary sites separately. In another subgroup meta-analysis of G3 GEP NETs (not NECs), surgical management was associated with increased OS compared with nonsurgical management (HR 0.26, 95% CI 0.11-0.61).

CONCLUSIONS

Surgical management of G3 GEP NENs may provide a potential survival benefit in well-selected cases. Further research is needed to define which patients will benefit most from surgical versus nonsurgical management. The current literature is limited by inconsistent reporting of survival outcomes in surgical versus nonsurgical groups, tumor grade, differentiation, primary tumor site, and selection criteria for surgical and nonsurgical management.

Is There a Role for Surgical Resection of Grade 3 Neuroendocrine Neoplasms?

BACKGROUND

Grade 3 (G3) gastroenteropancreatic (GEP) neuroendocrine neoplasms (NENs) are aggressive tumors with poor survival outcomes for which medical management is generally recommended. This study sought to evaluate outcomes of surgically treated G3 GEP-NEN patients.

METHODS

A single-institutional prospective NEN database was reviewed. Patients with G3 GEP-NENs based on World Health Organization (WHO) 2019 definitions included well-differentiated neuroendocrine tumors (G3NET) and poorly differentiated neuroendocrine carcinomas (G3NEC). Clinicopathologic factors were compared between groups. Overall survival from G3 diagnosis was assessed by the Kaplan-Meier method.

RESULTS

Surgical resection was performed for 463 patients (211 G1, 208 G2, 44 G3). Most had metastatic disease at presentation (54% G1, 69% G2, 91% G3; p < 0.001). The G3 cohort included 39 G3NETs and 5 G3NECs, 22 of pancreatic and 22 of midgut origin. Median overall survival (mOS; in months) was 268.1 for G1NETs, 129.9 for G2NETs, 50.5 for G3NETs, and 28.5 for G3NECs (p < 0.001). Over the same period, 31 G3 patients (12 G3NETs, 19 G3NECs) were treated non-surgically, with mOS of 19.0 for G3NETs and 12.4 for G3NECs.

CONCLUSIONS

Surgical resection of G3 GEP-NENs remains controversial due to poor prognosis, and surgical series are rare. This large, single-institutional study found significantly lower mOS in patients with resected G3NENs than those with G1/G2 tumors, reflecting more aggressive tumor biology and a higher proportion with metastatic disease. The mOS for resected G3NETs and G3NECs exceeded historical non-surgical G3NEN series (mOS 11-19 months), suggesting surgery should be considered in carefully selected patients with G3NENs, especially those with well-differentiated tumors.

Role of surgical treatments in high-grade or advanced gastroenteropancreatic neuroendocrine neoplasms

Over the last 40 years, the incidence and prevalence of gastroenteropancreatic neuroendocrine neoplasms (GEP-NENs) have continued to increase. Compared to other epithelial neoplasms in the same organ, GEP-NENs exhibit indolent biological behavior, resulting in more chances to undergo surgery. However, the role of surgery in high-grade or advanced GEP-NENs is still controversial. Surgery is associated with survival improvement of well-differentiated high-grade GEP-NENs, whereas poorly differentiated GEP-NENs that may benefit from resection require careful selection based on Ki67 and other tissue biomarkers. Additionally, surgery also plays an important role in locally advanced and metastatic disease. For locally advanced GEP-NENs, isolated major vascular involvement is no longer an absolute contraindication. In the setting of metastatic GEP-NENs, radical intended surgery is recommended for patients with low-grade and resectable metastases. For unresectable metastatic disease, a variety of surgical approaches, including cytoreduction of liver metastasis, liver transplantation, and surgery after neoadjuvant treatment, show survival benefits. Primary tumor resection in GEP-NENs with unresectable metastatic disease is associated with symptom control, prolonged survival, and improved sensitivity toward systemic therapies. Although there is no established neoadjuvant or adjuvant strategy, increasing attention has been given to this emerging research area. Some studies have reported that neoadjuvant therapy effectively reduces tumor burden, improves the effectiveness of subsequent surgery, and decreases surgical complications.

Surgical Treatment of Patients with Poorly Differentiated Pancreatic Neuroendocrine Carcinoma: An NCDB Analysis

BACKGROUND

Consensus guidelines discourage resection of poorly differentiated pancreatic neuroendocrine carcinoma (panNEC) given its association with poor long-term survival. This study assessed treatment patterns and outcomes for this rare malignancy using the National Cancer Database (NCDB).

METHODS

Patients with non-functional pancreatic neuroendocrine tumors in the NCDB (2004-2016) were categorized based on pathologic differentiation. Logistic and Cox proportional hazard regressions identified associations with resection and overall survival (OS). Survival was compared using Kaplan-Meier and log-rank tests.

RESULTS

Most patients (83%) in the cohort of 8560 patients had well-differentiated tumors (panNET). The median OS was 47 months (panNET, 63 months vs panNEC, 17 months; p < 0.001). Surgery was less likely for older patients (odds ratio [OR], 0.97), patients with panNEC (OR, 0.27), and patients with metastasis at diagnosis (OR, 0.08) (all p < 0.001). After propensity score-matching of these factors, surgical resection was associated with longer OS (82 vs 29 months; p < 0.001) and a decreased hazard of mortality (hazard ratio [HR], 0.37; p < 0.001). Surgery remained associated with longer OS when stratified by differentiation (98 vs 41 months for patients with panNET and 36 vs 8 months for patients with panNEC). Overall survival did not differ between patients with panNEC who underwent surgery and patients with panNET who did not (both 39 months; p = 0.294).

CONCLUSIONS

Poorly differentiated panNEC exhibits poorer survival than well-differentiated panNET. In the current cohort, surgical resection was strongly and independently associated with improved OS, suggesting that patients with panNEC who are suitable operative candidates should be considered for multimodality therapy, including surgery.

Progress in the Management of Pancreatic Neuroendocrine Tumors

Pancreatic neuroendocrine tumors (PNETs) are a heterogeneous and orphan group of neoplasms that vary in their histology, clinical features, prognosis, and management. The treatment of PNETs is highly dependent on the stage at presentation, tumor grade and differentiation, presence of symptoms from hormonal overproduction or from local growth, tumor burden, and rate of progression. The US Food and Drug Administration has recently approved many novel treatments, which have altered decision making and positively impacted the care and prognosis of these patients. In this review, we focus on the significant progress made in the management of PNETs over the past decade, as well as the active areas of research. Expected final online publication date for the Annual Review of Medicine, Volume 73 is January 2022. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.