Introduction
High-grade (G3) neuroendocrine tumors (NETs) are rare and extremely aggressive forms of cancer (citation). An estimated 8,000 NETs are diagnosed in the United States annually (citation). Of these estimated 8,000 cases, approximately 60% originate from the gastrointestinal (GI) tract (citation), and 20% of patients present with high-grade disease (citation) The incidence of NETs seems to be increasing, with one report showing a 2.5-fold increase over the past 15 years (citation). Nevertheless, due to the rarity of G3 NETs, few studies establish an ideal therapeutic approach (citation). Current literature demonstrates limited stage disease (G1 and G2) NETs of the GI tract are amendable with surgical resection (citation), but few studies explore the benefits of surgical resection with adjuvant chemotherapy for G3 NETs (citation). Furthermore, Heetfeld et al. (citation) noted multiple studies fail to differentiate between morphologically distinct NETs of advanced stage: G3 NETs and neuroendocrine carcinomas (NECs) (citation).
The goal of our study is to investigate the therapeutic impact of surgical resection with adjuvant chemotherapy compared to surgical resection alone for GI tract G3 NETs and NECs. To do so, we will examine data from the National Cancer Data Base (NCDB) and retrospectively compare 5-month survival (citation) of patients who undergo treatment for GI tract G3 NETs and NECs. We hypothesize that adjuvant chemotherapy with surgical resection will increase survival for patients with GI tract G3 NETs and NECs compared to patients who are treated with surgical resection alone. Our findings may warrant further study into adjuvant chemotherapy as an optimal therapeutic treatment for G3 NETs and NECs, and could ultimately refine recommendations for World Health Organization (WHO) NET classification (citation).
Background
High-grade (G3) neuroendocrine tumors (NETs) and neuroendocrine carcinomas (NECs) are rare and extremely aggressive forms of cancer (citation). An estimated 8,000 NETs are diagnosed in the United States annually (citation). Of these estimated 8,000 cases, approximately 60% originate from the gastrointestinal (GI) tract (citation), and 20% of patients present with high-grade disease (citation). Of the G3 NETs and NECs that arise from the GI tract, the most common primary sites include the esophagus and colon (citation). Incidence of GI tract G3 NETs and NECs has increased over the past few decades (citation), but there has been little impact on patient survival (citation). Without treatment, G3 NETs and NECs are typically fatal within weeks of diagnosis (citation). Even with treatment, the median survival for patients with G3 NETs and NECs is between 4 to 16 months (citation). Limited stage (G1 and G2) disease, although a better prognosis, reports a median overall survival range from 8 to 20 months with some form of treatment (citation).
The most recent classification of NETs by the WHO classifies G3 NETs as poorly differentiated (citation). G3 NETs are defined as having a mitotic count of greater than 20 per 10 high powered fields (hpf) (citation), and having a Ki-67 proliferative index of greater than 20 percent (citation). Additionally, the WHO classification system does not differentiate histologically distinct NECs, small cell G3 NETs, and large cell G3 NETs (citation). Our proposed study will investigate histologically distinct G3 NETs and NECs separately.
Due to the rarity of G3 NETs and NECs, few studies establish an ideal therapeutic approach (citation). Current literature demonstrates limited stage disease (G1 and G2) NETs of the GI tract are amendable with surgical resection (citation), while survivorship of G3 NETs and NECs treated by surgical resection remains unchanged. However, few studies explore the benefits of adjuvant chemotherapy with surgical resection for G3 NETs (citation).
Surgical resection for G1 and G2 NETs provides a promising prognosis (citation), with median survival reported to be as much as 5 years (citation). Literature on surgical resection for G3 NETs and NECs report conflicting findings (citation). Unfortunately, as Joseph et al.(citation) accounts, some patients with G3 NETs and NECs have been considered unresectable by physicians who are neither surgeons nor familiar with the existing collection of treatment strategies.
Surgical resection combined with chemotherapy has been shown to increase G3 NET and NEC survival (citation). Specifically, chemoembolization before surgery and intraoperative chemotherapy are reported to be effective treatments for systemic NETs, but few reports differentiate limited stage NETs and G3 NETs (citation). Casas et al. (citation) reported that patients with small cell G3 NET of the esophagus had an improved overall survival of 20 months when treated with surgery and chemotherapy compared to an improved overall survival of 5 months for surgical resection only. Heetfeld et al. (citation) noted that patients with NECs experienced an increase in overall survival by 16.4 months (95% CI 13.4-19.5) for patients treated with adjuvant platinum-based etoposide chemotherapy, but noted no increase in overall survival for patients with G3 NETs. Our proposed study seeks to retrospectively explore adjuvant chemotherapy with surgical resection as an optimal treatment strategy for G3 NETs and NECs.
GOALS AND OBJECTIVES
Specific Aims. Despite being rare cancers, NET and NEC incidence has been increasing over the past 15 years. Unfortunately, there has been little impact on patient survival in the same timeframe (citation). Most NETs and NECs arise from the GI tract (citation), and about 20 percent of patients present with high-grade disease (citation). Estimated 5-month survival of patients with G3 NETs and NECs is about __ percent with treatment (citation). Currently, an optimal treatment strategy for these cancers has not been identified, but surgical resection has been shown to be promising for G1 and G2 NETs of the gastrointestinal tract (citation). Few studies have investigated the therapeutic impact of surgical resection with adjuvant chemotherapy for G3NETs and NECs (citation). The overarching goal of our study is to investigate an optimal treatment strategy for G3NETs and NECs. To achieve our goal, we propose the following specific aims:
Aim 1. Investigate surgical resection with adjuvant chemotherapy as an optimal treatment associated with survival outcomes among patients with GI tract G3NETs and NECs. We hypothesize that surgical resection with adjuvant chemotherapy is associated with increased survival compared to surgical resection alone.
CONCEPTUAL MODEL
Figure 1. Overview of Study Design
METHODS
Overview. By investigating data collected from 2004-2012 in the NCDB, which comprises about 70 percent of newly diagnosed cancers in the United States, and 30 million historical records (citation), we propose to retrospectively compare 5-month survivorship of patients with GI tract G3 NETs and NECs who undergo surgical resection with adjuvant chemotherapy to surgical resection alone. An overview of our approach is illustrated in Figure 1.
Target Population. The proposed study seeks to determine an optimal treatment for all patients with G3NETs and NECs of the GI tract.
Study Population. The data applied to the proposed study will be derived from the NCDB from 2004-2012. A search of the NCDB will be performed for the diagnosis of G3 NET (International Classification of Diseases for Oncology, Third Edition [ICD-O-3]: 8246/3) from the years 2004 to 2012 (citation).
Data Collection. Secondary data on G3 NETs and NECs will be obtained from the National Cancer Database (NCDB). The NCDB is jointly sponsored by the American College of Surgeons and the American Cancer Society (citation). This database collects data from over 1,500 Commission on Cancer (CoC) accredited institutions on demographics, treatments, and outcomes of patients with malignant neoplasms (citation). Furthermore, the NCDB contains information on primary tumor site, tumor morphology, stage at diagnosis, and first line of treatment, and provides information on cancer incidence and survival for patients in the US. Data from the NCDB represents about 70% of newly diagnosed cancers in the US population, and contains around 30 million records (citation)
Data Analysis. A retrospective analysis of patients with G3 NETs and NECs will be performed with 5-month survival (outcome) as the primary endpoint of this study. Patients will be divided into two groups: surgical resection (non-exposure) and adjuvant chemotherapy (exposure). Descriptive statistics will be reported as means and standard deviations to summarize continuous variables or frequencies. Descriptive statistics will be reported as proportions for categorical data. Univariate analyses of descriptive statistics will be conducted using a student’s t test or chi-squared test when suitable. The Kaplan-Meier method will be used to calculate univariate survival. Significance of univariate survival will be determined by a log-rank test. Statistically significant variables on univariate analysis will be used to run a Cox regression analysis to identify independent associations with post-operative G3NET survival. Statistical analysis will be conducted using the SAS platform JMP® Pro, version 12.0, 2015 (SAS Institute Inc., Cary, NC, USA).
Variables. Patient demographics, tumor characteristics, and treatment methods will be assessed. Patient demographics for age at diagnosis, race, gender, post-operative survival time in days, and vital statistics will be included. Tumor characteristics will include primary site, histology, behavior, grade, and tumor size. Tumor characteristic coding will use ICD-O-3 classification terms, Traditional American Joint Committee on Cancer (AJCC) Staging System terms, or Collaborative Stage (CS) Data Collection Systems terms. Treatment data will include surgery to the primary site (NAACCR 1290), surgery to distant sites (NAACCR 1294), and systemic surgery sequence (NAACCR 1639). For systemic surgery sequence, adjuvant chemotherapy (NAACCR 1292) is coded as 2-7 (citation).