Treatment of Gastroenteropancreatic Neuroendocrine Tumors
Treatment of Gastroenteropancreatic Neuroendocrine Tumors
Objective: To describe recent advances in the treatment of gastroenteropancreatic neuroendocrine tumors (GEP-NETs).
Methods: A review of the published English language literature on GEP-NET therapy with a focus on practice-changing clinical trials.
Results: Somatostatin analog (SSA) treatment remains a cornerstone of GEP-NET therapy, primarily for patients with hormonally functional tumors and midgut carcinoids. The biologic agents everolimus and sunitinib have similar tumor-stabilizing effects in pancreatic NETs and are both approved to treat progressive low-intermediate-grade tumors. Their role in nonpancreatic NETs remains controversial. Cytotoxic chemotherapy is effective against pancreatic NETs, but modern prospective data is lacking. Radiolabeled SSAs will likely become more widely available once phase III randomized studies are completed.
Conclusions: New treatment options for GEP-NETs have become available and highlight the necessity of developing predictive biomarkers that will allow for appropriate and individualized therapy selection.
Neuroendocrine tumors (NETs) arise in the diffuse neuroendocrine system and are characterized by their ability to synthesize, store, and secrete a variety of neuroamines and peptides. Gastroenteropancreatic neuroendocrine tumors (GEP-NETs) are subdivided into 2 categories: carcinoid tumors of the luminal gastrointestinal (GI) tract and pancreatic NETs. Carcinoid tumors originate in the enterochromaffin cells of the gut and have distinctive clinicopathologic features based on the site of origin. Tumors originating in the embryonic midgut (ileocecum) often secrete serotonin and other vasoactive substances, giving rise to the malignant carcinoid syndrome. In contrast, hindgut tumors (distal colon and rectum) rarely produce hormones. Pancreatic NETs are thought to arise in the pancreatic islets of Langerhans and can release a variety of peptide hormones, including gastrin, insulin, and glucagon; however, the majority of pancreatic NETs are hormonally silent.
Recent years have seen a significant expansion in treatment options available for patients with advanced GEP-NETs. This review focuses on major therapeutic categories including somatostatin analogs (SSAs), radiolabeled SSAs, cytotoxic drugs, angiogenesis inhibitors, and mammalian target of rapamycin (mTOR) inhibitors. Recent genetic sequencing studies have already identified new potential targets that will undoubtedly alter the future treatment landscape.
SSAs have had a significant impact on patients with hormonal symptoms associated with functioning GEP-NETs. In recent years, studies have established that SSAs can inhibit neuroendocrine tumor growth, primarily in patients with small-bowel carcinoid tumors (see "Antiproliferative Effects of SSAs" below). The original rationale for SSA development was the discovery that native human somatostatin acts as an endocrine inhibitor. Among other functions, somatostatin reduces GI secretion, inhibits peristalsis, and impedes the secretion of other GI hormones, such as serotonin. Somatostatin exerts its effects by binding with 5 somatostatin receptor subtypes (ssts1–5), which belong to a family of G protein-coupled receptors. The short half-life of native human somatostatin (approximately 2 minutes) limits its clinical utility. As a result, SSAs with an improved pharmacokinetic profile have been synthesized by eliminating enzymatic cleavage sites within the peptide chain.
The SSAs currently available in clinical practice are octreotide and lanreotide. Both compounds are cyclic octapeptides with very similar somatostatin receptor affinities; they bind avidly to sst2 and moderately to sst5. The first clinical trial of octreotide evaluated the drug in 25 patients with malignant carcinoid syndrome. Flushing and diarrhea were promptly palliated in 22 patients (88%), and major reductions in urine 5-hydroxyindoleacetic acid (5-HIAA) were observed in 18 cases (72%), leading to the approval of octreotide by the U.S. Food and Drug Administration (FDA) for the management of the carcinoid syndrome. A subsequent crossover trial comparing octreotide versus lanreotide in 33 carcinoid syndrome patients demonstrated no significant differences in symptom control or reduction of tumor markers. Other small trials and clinical series have demonstrated that both SSAs are effective at palliation of hormonal syndromes associated with pancreatic NETs. Rapid reversal of the watery diarrhea, hypokalemia, and achlorhydria (WDHA) syndrome associated with vasoactive intestinal peptide (VIP)omas has been observed with SSA use. Glucagonoma syndrome, particularly the necrolytic migratory erythema (NME) rash, is likewise treatable with SSAs. The insulinoma syndrome is more refractory to SSAs, likely due to the fact that most insulinomas express relatively low levels of the sst2 receptor. Caution must be exercised when treating advanced insulinomas with SSAs due to the potential for hypoglycemia exacerbation by inhibition of the counterregulatory hormone, glucagon. The gastrinoma syndrome can also be palliated with SSAs; however, high-dose proton-pump inhibitors (PPIs) exert even tighter control of acid secretion.
Octreotide was originally developed as an immediate-release subcutaneous (SC) formulation and was studied at doses of 100 to 500 mcg 2 to 3 times daily. More recently, long-acting depot formulations of octreotide and lanreotide have been developed, enabling monthly dosing. The initial clinical trial of depot-octreotide long-acting repeatable (LAR) for the treatment of the carcinoid syndrome investigated doses of 10, 20, and 30 mg given intramuscularly (IM) every 4 weeks. While there was no correlation between octreotide LAR dose and symptomatic improvement, suppression of urine 5-HIAA appeared to be inferior in the 10-mg arm compared to the 20- and 30-mg arms. As a result, the FDA label recommends a starting dose of 20 mg with titration to 30 mg in patients with suboptimal symptom control. Depot lanreotide is administered as a deep SC injection at doses ranging from 60 to 120 mg every 4 weeks. The formulation is available as a prefilled syringe and allows for the achievement of therapeutic plasma levels within 24 hours. Although approved within the European Union for the management of the carcinoid syndrome, the sole FDA-approved indication in the U.S. is for the management of acromegaly.
Both octreotide and lanreotide are exceptionally well-tolerated agents. Side effects are generally mild and may include nausea, steatorrhea, and bloating. Supplemental use of pancreatic digestive enzymes may be considered in patients complaining of gas or steatorrhea associated with SSA use. Long-term administration of SSAs can result in increased rates of biliary stone and sludge formation due to inhibitory effects of SSAs on gallbladder contractility. Maximally tolerated doses (MTD) of SSAs have not been identified in clinical trials. In routine practice, above-label doses are commonly administered to patients who experience suboptimal control of their hormonal syndromes. Patients who experience symptom exacerbation towards the final week of each treatment cycle may benefit from an increased frequency of administration (i.e., every 3 weeks). In other cases, supplemental dosing of SC short-acting octreotide may serve to control breakthrough symptoms, such as flushing or diarrhea. Evidence suggests that somatostatin receptors may be saturated at an octreotide LAR dose of 60 mg; thus, higher doses may be associated with a minimal marginal benefit.
The term "carcinoid crisis" refers to acute flushing and hypotension caused by the massive release of serotonin and other vasoactive substances into the circulation. Triggers of carcinoid crisis include general anesthesia, epinephrine, and physical manipulation of tumors; therefore, carcinoid crises most commonly occur in an intraoperative setting. Patients who have carcinoid syndrome or are prone to serotonin release should receive a prophylactic dose of octreotide 250 to 500 mcg SC or intravenously (IV) prior to a surgical or interventional procedure. Patients who develop intraoperative hypotension should receive bolus IV doses of 500 to 1000 mcg until symptom control is achieved. Alternatively, an IV infusion of octreotide may be delivered after a bolus dose.
While SSAs were initially developed to control hormonal syndromes, accumulating data suggest that they can also inhibit tumor growth. The putative mechanisms of growth inhibition fall into 2 categories: "direct" and "indirect." The direct antiproliferative effect involves an interaction between SSAs and somatostatin receptors on tumor cells, leading to inhibition of cellular proliferation and apoptosis. Although the precise signaling transduction pathways are not fully understood, the initial steps appear to involve activation of phosphotyrosine phosphatases (PTPs), as well as modulation of the mitogen-activated protein kinase (MAPK) pathway. The indirect antiproliferative effect occurs through suppression of circulating growth factors, such as vascular endothelial growth factor (VEGF), growth hormone (GH), and insulin-like growth factor (IGF), as well as the interaction of SSAs with somatostatin receptors on endothelial cells.
Until recently, evidence of the antiproliferative effects of SSAs derived from single-arm phase II trials documented disease stabilization in roughly 50% of patients with progressive GEP-NETs who were treated with octreotide or lanreotide Higher level evidence establishing the antiproliferative effect of octreotide emerged with the publication of the PROMID study. This randomized phase III trial compared octreotide LAR 30 mg versus placebo in 85 patients with advanced midgut NETs. The primary endpoint, time to tumor progression (TTP), increased from 6 months in the placebo arm to 14.3 months in the octreotide LAR arm (P = .000072). Serious adverse events were evenly balanced. Multivariate analysis suggested that patients with resected primary tumors and low tumor burden benefitted most significantly from treatment with octreotide versus placebo. Based on the results of the PROMID trial, octreotide LAR therapy is considered an appropriate first-line systemic therapy for patients with metastatic, unresectable midgut carcinoid tumors and is endorsed for this indication by the National Comprehensive Cancer Network (NCCN), as well as other guidelines.
As a general rule, tumors that are slow-growing, mitotically inactive, and express somatostatin receptors are most likely to be effectively controlled with SSAs. The majority of midgut carcinoid tumors meet these criteria. It is still unclear whether the growth of non-midgut GEP-NETs can be controlled with SSAs. The CLARINET study was designed to address this question by randomizing patients with hormonally nonfunctioning GEP-NETs to treatment with depot-lanreotide versus placebo. The target accrual has been achieved, and the results are pending.
Interferons (IFNs) can inhibit tumor growth through a variety of mechanisms including immune stimulation, inhibition of angiogenesis, and induction of cell cycle arrest. IFNs can also upregulate somatostatin receptor expression, thus potentiating the effect of SSAs. Alpha-IFN (also known as leukocyte IFN) has been studied extensively in NETs. Single-arm studies of alpha-IFN reported objective response rates in the 5 to 10% range, with higher rates of disease stabilization and palliation of hormonal syndromes. In vitro studies showing synergy between alpha-IFN and SSAs led to several randomized studies investigating SSA/IFN combinations. One trial of patients with carcinoid syndrome who had become refractory to octreotide reported symptomatic improvement in 49% of patients after the addition of alpha-IFN. Another study reported disease stabilization in 14 of 21 patients for whom alpha-IFN was added to pre-existing octreotide therapy.
Several randomized clinical trials have investigated SSAs alone versus in combination with alpha-IFN. One study of 68 patients with metastatic midgut carcinoid tumors who had undergone prior hepatic embolization evaluated octreotide alone versus octreotide combined with alpha-IFN (3 million units administered 5 times per week). The 5-year survival in the combination arm was 57% versus 37% in the octreotide monotherapy arm. Due to the small number of patients, the results were not statistically significant (P = .13). The hazard ratio (HR) for tumor progression was 0.28 (P = .008). Another randomized study of 109 GEP-NET patients compared octreotide alone versus octreotide combined with alpha-IFN (4.5 million units administered 3 times per week). The median survival was prolonged in the combination arm (51 vs. 35 months), but the results did not achieve statistical significance (P = .38). The response rates in both arms were <6%. A 3-arm trial of 80 therapy-naïve patients with advanced GEP-NETs compared subcutaneous lanreotide alone to alpha-IFN (5 million units 3 times weekly) or the combination of the 2 drugs. In this trial, TTP was similar in all 3arms. Among 11 patients who were treated with lanreotide monotherapy and crossed over to combination therapy, 1 patient showed measurable inhibition of tumor growth.
Given the underpowered nature of these randomized trials, it is difficult to draw any definitive conclusions regarding the effects of alpha-IFN on overall survival (OS) or PFS. Midgut carcinoid tumors appear to be most sensitive to the antisecretory and antiproliferative effects of alpha-IFN. Enthusiasm for this drug is somewhat tempered by potential side effects, which include flu-like symptoms, depression, and myelosuppression. However, it is important to note most GEP-NET trials have studied low doses of alpha-IFN, which are fairly tolerable.
The mTOR enzyme is a serine/threonine kinase that lies downstream of the phosphatidylinositide 3-kinase (PI3K)/protein kinase B (AKT) pathway and regulates cell growth, proliferation, and metabolism in response to environmental factors. The importance of mTOR in neuroendocrine cancers is highlighted by the fact that patients with germline mutations of the TSC2(tuberous sclerosis 2) gene, an endogenous inhibitor of mTOR, are prone to develop pancreatic NETs. Somatic mutations in mTOR pathway genes, including TSC2, PTEN(phosphatase and tensin homolog) and PIK3CA(phosphatidylinositol-4,5-bisphosphate 3-kinase) occur in roughly 15% of pancreatic NETs. Other alterations in PI3K/AKT/mTOR pathway genes, including amplifications of AKT1/2, are observed in nearly one-third of small bowel carcinoid tumors.
The oral mTOR inhibitor everolimus has been extensively studied in GEP-NETs. In the RADIANT 1 trial, 160 patients with advanced, progressive pancreatic NETs were evaluated in 2 strata: everolimus monotherapy (n = 115) or everolimus plus octreotide (n = 45). Response rates and median PFS were 9% and 9.7 months in the monotherapy arm versus 4% and 16.7 months in the combined therapy arm, respectively. The RADIANT 2 trial randomly assigned 429 patients with hormonally functional carcinoid tumors to treatment with everolimus (10 mg) plus octreotide versus placebo plus octreotide. On central radiographic review, median PFS increased from 11.3 months on the control arm to 16.4 months on the experimental arm (HR 0.77, P = .026). While clinically significant, the primary endpoint fell short of its prespecified statistical significance threshold. One potential factor contributing to the lack of statistical significance was loss of progression events due to discrepancies in central versus local radiographic review. Objective responses were observed in only 5 patients in the everolimus arm (2%) versus 4 patients in the placebo arm (2%). There was no significant trend toward improvement in OS, a fact that may be attributable to the crossover design of the study. Everolimus is not currently approved by the FDA for the treatment of carcinoid tumors.
The RADIANT 3 trial randomized assigned 410 patients with low- and intermediate-grade pancreatic NETs to treatment with everolimus (10 mg) versus placebo. Despite an objective response rate of only 5% in the everolimus arm, the study demonstrated a clinically and statistically significant improvement in PFS, the primary endpoint. Median PFS increased from 4.6 months on the placebo arm to 11 months on the everolimus arm (HR 0.35, P<.001). OS differences were not observed, possibly due to the crossover design. As a result of the significant improvement in PFS, everolimus was approved by the FDA for treatment of patients with advanced pancreatic NETs.
Side effects of everolimus include aphthous oral ulcers, skin rash, diarrhea, hyperglycemia and cytopenias. Pneumonitis is a relatively rare but potentially serious toxicity. As an analog of rapamycin, everolimus has immunosuppressive properties and can cause atypical infections. While most toxicities are mild, cumulative and chronic side effects may impact patient quality of life.
Neuroendocrine tumors are among the most highly vascular of solid tumors and frequently express the VEGF ligand and receptor (VEGFR), which are key drivers of angiogenesis. Increased levels of circulating VEGF have been correlated with tumor progression. Consequently, inhibition of the VEGF pathway has been a promising therapeutic target. Most VEGF-targeting agents can be divided into 2 categories: tyrosine kinase VEGFR inhibitors and circulating VEGF inhibitors.
The tyrosine kinase receptor inhibitor sunitinib targets VEGFR-1, -2, and -3, as well as stem-cell factor receptor (cKIT) and platelet-derived growth factor receptor (PDGFR). In a phase II trial of 109 patients, treatment with sunitinib yielded objective response rates of 2.4% and 16.7% in carcinoid and pancreatic NETs, respectively. Based on the relatively high response rates in the latter cohort, a multi-national phase III study was launched that randomly assigned patients with low- and intermediate-grade pancreatic NETs to sunitinib (37.5 mg daily) versus placebo. The study was discontinued on interim analysis after enrollment of 171 patients and demonstrated a statistically significant improvement in median PFS from 5.5 months on the placebo arm to 11.1 months on the sunitinib arm. There was a trend toward improvement in OS, which was not statistically significant in the mature analysis. The objective response rate associated with sunitinib was 9.3%. Side effects of sunitinib included diarrhea, nausea, fatigue, cytopenias, hypertension, and palmar-plantar erythrodysesthesia. Despite these toxicities, no overall differences were noted between study groups in global health-related quality of life. As a result of this study, sunitinib is FDA approved to treat pancreatic NETs.
Bevacizumab is a monoclonal antibody to circulating VEGF-A. In a randomized phase II trial, 44 patients with metastatic carcinoid tumors were randomly assigned to treatment with bevacizumab or pegylated alpha-IFN (PEG-IFN) for 18 weeks, followed by administration of both agents in combination. At the week 18 time point, the rate of PFS was 95% on the bevacizumab arm versus 68% on the PEG-IFN arm. Moreover, the objective radiographic response rate in the bevacizumab arm was 18% versus 0% with PEG-IFN. On functional computed tomography (CT) scans performed at baseline and on day 2 of therapy, bevacizumab treatment yielded an average reduction in tumor blood flow of 49%. An ongoing phase III study led by the South West Oncology Group (SWOG) is comparing bevacizumab to alpha-IFN in patients with metastatic carcinoid tumors with a primary endpoint of PFS. The study has completed accrual, and the results are pending.
Several factors predict the sensitivity of GEP-NETs to cytotoxic chemotherapy, including tumor grade, differentiation, and primary site. Poorly differentiated neuroendocrine cancers, which can be divided into small cell and nonsmall cell categories, are generally characterized by a high mitotic rate, a Ki-67 proliferation index exceeding 50%, and tumor necrosis. The treatment of these aggressive malignancies is generally similar to the treatment of small cell lung cancer. A study evaluating cisplatin and etoposide in GEP-NETs reported a response rate of 67% in poorly differentiated tumors versus 7% in well-differentiated tumors. Another study of cisplatin and etoposide in poorly differentiated GEP-NETs demonstrated a response rate of 42%. The durations of response in both studies were short (8–9 months) and with median survivals of only 15–19 months.
The chemosensitivity of low- to intermediate-grade tumors correlates with primary tumor site. Among GEP-NETs, pancreatic primary tumors appear to be most highly sensitive to cytotoxic agents, including streptozocin (STZ), dacarbazine, temozolomide, and fluoropyrimidines. One of the first cytotoxic drugs tested in pancreatic NETs was the nitrosourea STZ, an agent that is specifically toxic to pancreatic islet beta-cells. Based on promising phase II data, 2 randomized trials were conducted by the Eastern Cooperative Oncology Group (ECOG) in the 1970s and 1980s. The first of these studies reported response rates of 63% with STZ plus 5-fluorouracil (5-FU) versus 36% with STZ monotherapy. The second trial compared STZ plus doxorubicin versus STZ plus 5-FU and reported response rates and TTPs of 69% and 20 months versus 45% and 6.9 months, respectively. STZ plus doxorubicin also had a significant advantage in terms of survival (median OS 2.2 versus 1.4 years).
The high response rates associated with STZ-based regimens were subsequently questioned due to the partial reliance on nonradiographic response criteria, such as improvement in tumor markers or reduction in hepatomegaly. For example, in a small retrospective series of 16 patients, only 1 patient achieved an objective response by CT scan criteria. A more recent retrospective study investigating the combination of STZ, 5-FU, and doxorubicin in 84 pancreatic NETs reported a response rate of 39% using objective radiographic criteria and a median response duration of 9.3 months. Enthusiasm for STZ-based chemotherapy is somewhat tempered by toxicity concerns, including myelosuppression, nausea, and renal insufficiency.
In recent years, the oral alkylating agent temozolomide has emerged as an active agent in pancreatic NETs. Temozolomide shares a metabolite with dacarbazine and is converted to the active alkylator MTIC (3-methyl-[triazen-1-yl]-imidazole-4-carboxamide), which induces DNA methylation at the O-6 position of guanine. A phase II study investigating the combination of temozolomide and thalidomide reported an objective response rate of 45% in the pancreatic NET subset of patients (5 of 11 patients). A recent retrospective study of temozolomide combined with capecitabine in 30 chemo-naïve pancreatic NET patients reported an objective radiographic response rate of 70% and median progression-free survival (PFS) of 18 months. A phase II trial of temozolomide combined with bevacizumab reported a response rate of 33% and median PFS of 14.3 months in the pancreatic NET cohort, which consisted of 15 patients. Side effects of temozolomide are relatively tolerable when patients receive appropriate nausea prophylaxis and include thrombocytopenia and lymphopenia. The data supporting temozolomide-based regimens in pancreatic NETs consists, thus far, of retrospective series and small cohorts of prospective trials. An ECOG-sponsored prospective randomized trial of temozolomide monotherapy versus capecitabine plus temozolomide will provide much needed prospective data in a large cohort of patients.
Well-differentiated carcinoid tumors appear to be significantly more resistant to cytotoxic chemotherapy. A randomized clinical trial of doxorubicin plus 5-FU compared with STZ plus 5-FU reported response rates of 16% and median PFS of 5 months in both arms. In the phase II study of temozolomide plus thalidomide, the response rate in patients with carcinoid tumors was only 7% (compared to 45% in the pancreatic NET cohort). In the phase II study of temozolomide plus bevacizumab, the response rate in the carcinoid tumor cohort was 0% (compared to 33% in the pancreatic NET cohort). There are several potential explanations for the differences in chemosensitivity. In 1 study, carcinoid tumors expressed a much higher level of methyl-guanine-methyl-transferase (MGMT), a DNA repair enzyme, compared to pancreatic NETs. Additionally, the low mitotic activity of certain carcinoid tumors, particularly those originating in the midgut, may confer higher resistance to cytotoxic agents than pancreatic NETs, which tend to have a higher proliferative rate.
An emerging systemic treatment modality consists of radiolabeled SSAs. This targeted form of systemic radiation, also known as peptide receptor radiotherapy (PRRT), enables the delivery of radioactive isotopes to somatostatin receptor-expressing tumors. Nearly 80% of GEP-NETs express somatostatin receptors as evidenced by radiotracer uptake on somatostatin-receptor scintigraphy (SRS, OctreoScan®). Radiolabeled SSAs consist of a radionuclide isotope, an SSA (peptide), and a chelator used to bind them. Commonly used SSAs include octreotide and octreotate, an analog with enhanced binding to sst2. Chelators include DPTA (diethylenetriamine-penta-acetic acid) and DOTA (tetraazacyclododecane-tetra-acetic acid). The main selection criterion for PRRT is evidence of strong radiotracer expression (ideally higher than normal liver tissue) on OctreoScan or another somatostatin receptor imaging modality.
Early clinical trials of PRRT used octreotide labeled with high, cytotoxic doses of In-pentetreotide, the isotope used in OctreoScans. Although symptomatic responses were observed in some cases, objective radiographic responses were rare, possibly due to the short tissue penetration of Auger electrons emitted by the In isotope. The next generation of radiolabeled SSAs used yttrium (Y), a high-energy β-particle emitter with a maximum tissue penetration range of approximately 12 mm. In 1 study, objective responses associated with Y-DOTA-Tyr3-octreotide (Y-DOTATOC) were observed in 28% of 87 patients with advanced GEP-NETs. However, in a subsequent multicenter trial of GEP-NETs using doses of Y-DOTATOC up to 886 mCi in 4 cycles, a response rate of only 9% was reported. Another phase II study enrolled 90 patients with metastatic hormonally functioning carcinoid tumors who experienced symptom progression while on SSAs. Treatment consisted of Y-DOTATOC at a dose of 4.4 GBq times 3 cycles. While the majority of patients (74%) appeared to obtain clinical benefit, only 4 patients (4%) achieved a partial radiographic response. The median PFS and OS were 16.3 and 26.9 months, respectively. Toxicities associated with Y-DOTATOC include renal insufficiency, which is partially ameliorated by concurrent amino acid infusions. In 1 large institutional series of 1,109 patients, 102 patients (9%) had severe permanent renal toxicity. Acute GI toxicities, including nausea and vomiting, are primarily attributable to the amino acid infusions. Myelosuppression is another side effect of PRRT, with grade 3 and 4 hematological toxicities occurring in approximately 13% of cases.
A newer generation of radiolabeled SSAs employs lutetium-177 (Lu), a medium energy β emitter with a maximal tissue penetration of 2 mm and a half-life of 6.7 days. Lu also emits γ rays allowing for scintigraphy and dosimetry using the same compound. The most commonly used peptide-isotope compound is Lu-DOTA-Tyr3-octreotate (Lu-DOTATATE). In 1 series of 310 patients with GEP-NETs, an objective response rate of 30% was reported. Response rates were particularly high in pancreatic NETs ranging from 36% in nonfunctioning pancreatic NETs to approximately 40 to 60% in functional gastrinomas, insulinomas, and VIPomas. Median TTP was 40 months, consistent with a high rate of disease control. Side effects of Lu-DOTA-Tyr3-octreotate are primarily hematological and nephrotoxicity rates are lower than observed with Y labeled compounds. An international phase III trial of Lu-DOTATATE versus high-dose octreotide is now open for patients with midgut NETs whose tumors have progressed on standard-dose octreotide.
Abstract and Introduction
Abstract
Objective: To describe recent advances in the treatment of gastroenteropancreatic neuroendocrine tumors (GEP-NETs).
Methods: A review of the published English language literature on GEP-NET therapy with a focus on practice-changing clinical trials.
Results: Somatostatin analog (SSA) treatment remains a cornerstone of GEP-NET therapy, primarily for patients with hormonally functional tumors and midgut carcinoids. The biologic agents everolimus and sunitinib have similar tumor-stabilizing effects in pancreatic NETs and are both approved to treat progressive low-intermediate-grade tumors. Their role in nonpancreatic NETs remains controversial. Cytotoxic chemotherapy is effective against pancreatic NETs, but modern prospective data is lacking. Radiolabeled SSAs will likely become more widely available once phase III randomized studies are completed.
Conclusions: New treatment options for GEP-NETs have become available and highlight the necessity of developing predictive biomarkers that will allow for appropriate and individualized therapy selection.
Introduction
Neuroendocrine tumors (NETs) arise in the diffuse neuroendocrine system and are characterized by their ability to synthesize, store, and secrete a variety of neuroamines and peptides. Gastroenteropancreatic neuroendocrine tumors (GEP-NETs) are subdivided into 2 categories: carcinoid tumors of the luminal gastrointestinal (GI) tract and pancreatic NETs. Carcinoid tumors originate in the enterochromaffin cells of the gut and have distinctive clinicopathologic features based on the site of origin. Tumors originating in the embryonic midgut (ileocecum) often secrete serotonin and other vasoactive substances, giving rise to the malignant carcinoid syndrome. In contrast, hindgut tumors (distal colon and rectum) rarely produce hormones. Pancreatic NETs are thought to arise in the pancreatic islets of Langerhans and can release a variety of peptide hormones, including gastrin, insulin, and glucagon; however, the majority of pancreatic NETs are hormonally silent.
Recent years have seen a significant expansion in treatment options available for patients with advanced GEP-NETs. This review focuses on major therapeutic categories including somatostatin analogs (SSAs), radiolabeled SSAs, cytotoxic drugs, angiogenesis inhibitors, and mammalian target of rapamycin (mTOR) inhibitors. Recent genetic sequencing studies have already identified new potential targets that will undoubtedly alter the future treatment landscape.
SSAs
SSAs have had a significant impact on patients with hormonal symptoms associated with functioning GEP-NETs. In recent years, studies have established that SSAs can inhibit neuroendocrine tumor growth, primarily in patients with small-bowel carcinoid tumors (see "Antiproliferative Effects of SSAs" below). The original rationale for SSA development was the discovery that native human somatostatin acts as an endocrine inhibitor. Among other functions, somatostatin reduces GI secretion, inhibits peristalsis, and impedes the secretion of other GI hormones, such as serotonin. Somatostatin exerts its effects by binding with 5 somatostatin receptor subtypes (ssts1–5), which belong to a family of G protein-coupled receptors. The short half-life of native human somatostatin (approximately 2 minutes) limits its clinical utility. As a result, SSAs with an improved pharmacokinetic profile have been synthesized by eliminating enzymatic cleavage sites within the peptide chain.
The SSAs currently available in clinical practice are octreotide and lanreotide. Both compounds are cyclic octapeptides with very similar somatostatin receptor affinities; they bind avidly to sst2 and moderately to sst5. The first clinical trial of octreotide evaluated the drug in 25 patients with malignant carcinoid syndrome. Flushing and diarrhea were promptly palliated in 22 patients (88%), and major reductions in urine 5-hydroxyindoleacetic acid (5-HIAA) were observed in 18 cases (72%), leading to the approval of octreotide by the U.S. Food and Drug Administration (FDA) for the management of the carcinoid syndrome. A subsequent crossover trial comparing octreotide versus lanreotide in 33 carcinoid syndrome patients demonstrated no significant differences in symptom control or reduction of tumor markers. Other small trials and clinical series have demonstrated that both SSAs are effective at palliation of hormonal syndromes associated with pancreatic NETs. Rapid reversal of the watery diarrhea, hypokalemia, and achlorhydria (WDHA) syndrome associated with vasoactive intestinal peptide (VIP)omas has been observed with SSA use. Glucagonoma syndrome, particularly the necrolytic migratory erythema (NME) rash, is likewise treatable with SSAs. The insulinoma syndrome is more refractory to SSAs, likely due to the fact that most insulinomas express relatively low levels of the sst2 receptor. Caution must be exercised when treating advanced insulinomas with SSAs due to the potential for hypoglycemia exacerbation by inhibition of the counterregulatory hormone, glucagon. The gastrinoma syndrome can also be palliated with SSAs; however, high-dose proton-pump inhibitors (PPIs) exert even tighter control of acid secretion.
Octreotide was originally developed as an immediate-release subcutaneous (SC) formulation and was studied at doses of 100 to 500 mcg 2 to 3 times daily. More recently, long-acting depot formulations of octreotide and lanreotide have been developed, enabling monthly dosing. The initial clinical trial of depot-octreotide long-acting repeatable (LAR) for the treatment of the carcinoid syndrome investigated doses of 10, 20, and 30 mg given intramuscularly (IM) every 4 weeks. While there was no correlation between octreotide LAR dose and symptomatic improvement, suppression of urine 5-HIAA appeared to be inferior in the 10-mg arm compared to the 20- and 30-mg arms. As a result, the FDA label recommends a starting dose of 20 mg with titration to 30 mg in patients with suboptimal symptom control. Depot lanreotide is administered as a deep SC injection at doses ranging from 60 to 120 mg every 4 weeks. The formulation is available as a prefilled syringe and allows for the achievement of therapeutic plasma levels within 24 hours. Although approved within the European Union for the management of the carcinoid syndrome, the sole FDA-approved indication in the U.S. is for the management of acromegaly.
Both octreotide and lanreotide are exceptionally well-tolerated agents. Side effects are generally mild and may include nausea, steatorrhea, and bloating. Supplemental use of pancreatic digestive enzymes may be considered in patients complaining of gas or steatorrhea associated with SSA use. Long-term administration of SSAs can result in increased rates of biliary stone and sludge formation due to inhibitory effects of SSAs on gallbladder contractility. Maximally tolerated doses (MTD) of SSAs have not been identified in clinical trials. In routine practice, above-label doses are commonly administered to patients who experience suboptimal control of their hormonal syndromes. Patients who experience symptom exacerbation towards the final week of each treatment cycle may benefit from an increased frequency of administration (i.e., every 3 weeks). In other cases, supplemental dosing of SC short-acting octreotide may serve to control breakthrough symptoms, such as flushing or diarrhea. Evidence suggests that somatostatin receptors may be saturated at an octreotide LAR dose of 60 mg; thus, higher doses may be associated with a minimal marginal benefit.
Treatment and Prophylaxis of the Carcinoid Crisis
The term "carcinoid crisis" refers to acute flushing and hypotension caused by the massive release of serotonin and other vasoactive substances into the circulation. Triggers of carcinoid crisis include general anesthesia, epinephrine, and physical manipulation of tumors; therefore, carcinoid crises most commonly occur in an intraoperative setting. Patients who have carcinoid syndrome or are prone to serotonin release should receive a prophylactic dose of octreotide 250 to 500 mcg SC or intravenously (IV) prior to a surgical or interventional procedure. Patients who develop intraoperative hypotension should receive bolus IV doses of 500 to 1000 mcg until symptom control is achieved. Alternatively, an IV infusion of octreotide may be delivered after a bolus dose.
The Antiproliferative Effect of SSAs
While SSAs were initially developed to control hormonal syndromes, accumulating data suggest that they can also inhibit tumor growth. The putative mechanisms of growth inhibition fall into 2 categories: "direct" and "indirect." The direct antiproliferative effect involves an interaction between SSAs and somatostatin receptors on tumor cells, leading to inhibition of cellular proliferation and apoptosis. Although the precise signaling transduction pathways are not fully understood, the initial steps appear to involve activation of phosphotyrosine phosphatases (PTPs), as well as modulation of the mitogen-activated protein kinase (MAPK) pathway. The indirect antiproliferative effect occurs through suppression of circulating growth factors, such as vascular endothelial growth factor (VEGF), growth hormone (GH), and insulin-like growth factor (IGF), as well as the interaction of SSAs with somatostatin receptors on endothelial cells.
Until recently, evidence of the antiproliferative effects of SSAs derived from single-arm phase II trials documented disease stabilization in roughly 50% of patients with progressive GEP-NETs who were treated with octreotide or lanreotide Higher level evidence establishing the antiproliferative effect of octreotide emerged with the publication of the PROMID study. This randomized phase III trial compared octreotide LAR 30 mg versus placebo in 85 patients with advanced midgut NETs. The primary endpoint, time to tumor progression (TTP), increased from 6 months in the placebo arm to 14.3 months in the octreotide LAR arm (P = .000072). Serious adverse events were evenly balanced. Multivariate analysis suggested that patients with resected primary tumors and low tumor burden benefitted most significantly from treatment with octreotide versus placebo. Based on the results of the PROMID trial, octreotide LAR therapy is considered an appropriate first-line systemic therapy for patients with metastatic, unresectable midgut carcinoid tumors and is endorsed for this indication by the National Comprehensive Cancer Network (NCCN), as well as other guidelines.
As a general rule, tumors that are slow-growing, mitotically inactive, and express somatostatin receptors are most likely to be effectively controlled with SSAs. The majority of midgut carcinoid tumors meet these criteria. It is still unclear whether the growth of non-midgut GEP-NETs can be controlled with SSAs. The CLARINET study was designed to address this question by randomizing patients with hormonally nonfunctioning GEP-NETs to treatment with depot-lanreotide versus placebo. The target accrual has been achieved, and the results are pending.
Alpha-interferon
Interferons (IFNs) can inhibit tumor growth through a variety of mechanisms including immune stimulation, inhibition of angiogenesis, and induction of cell cycle arrest. IFNs can also upregulate somatostatin receptor expression, thus potentiating the effect of SSAs. Alpha-IFN (also known as leukocyte IFN) has been studied extensively in NETs. Single-arm studies of alpha-IFN reported objective response rates in the 5 to 10% range, with higher rates of disease stabilization and palliation of hormonal syndromes. In vitro studies showing synergy between alpha-IFN and SSAs led to several randomized studies investigating SSA/IFN combinations. One trial of patients with carcinoid syndrome who had become refractory to octreotide reported symptomatic improvement in 49% of patients after the addition of alpha-IFN. Another study reported disease stabilization in 14 of 21 patients for whom alpha-IFN was added to pre-existing octreotide therapy.
Several randomized clinical trials have investigated SSAs alone versus in combination with alpha-IFN. One study of 68 patients with metastatic midgut carcinoid tumors who had undergone prior hepatic embolization evaluated octreotide alone versus octreotide combined with alpha-IFN (3 million units administered 5 times per week). The 5-year survival in the combination arm was 57% versus 37% in the octreotide monotherapy arm. Due to the small number of patients, the results were not statistically significant (P = .13). The hazard ratio (HR) for tumor progression was 0.28 (P = .008). Another randomized study of 109 GEP-NET patients compared octreotide alone versus octreotide combined with alpha-IFN (4.5 million units administered 3 times per week). The median survival was prolonged in the combination arm (51 vs. 35 months), but the results did not achieve statistical significance (P = .38). The response rates in both arms were <6%. A 3-arm trial of 80 therapy-naïve patients with advanced GEP-NETs compared subcutaneous lanreotide alone to alpha-IFN (5 million units 3 times weekly) or the combination of the 2 drugs. In this trial, TTP was similar in all 3arms. Among 11 patients who were treated with lanreotide monotherapy and crossed over to combination therapy, 1 patient showed measurable inhibition of tumor growth.
Given the underpowered nature of these randomized trials, it is difficult to draw any definitive conclusions regarding the effects of alpha-IFN on overall survival (OS) or PFS. Midgut carcinoid tumors appear to be most sensitive to the antisecretory and antiproliferative effects of alpha-IFN. Enthusiasm for this drug is somewhat tempered by potential side effects, which include flu-like symptoms, depression, and myelosuppression. However, it is important to note most GEP-NET trials have studied low doses of alpha-IFN, which are fairly tolerable.
mTOR Inhibitors
The mTOR enzyme is a serine/threonine kinase that lies downstream of the phosphatidylinositide 3-kinase (PI3K)/protein kinase B (AKT) pathway and regulates cell growth, proliferation, and metabolism in response to environmental factors. The importance of mTOR in neuroendocrine cancers is highlighted by the fact that patients with germline mutations of the TSC2(tuberous sclerosis 2) gene, an endogenous inhibitor of mTOR, are prone to develop pancreatic NETs. Somatic mutations in mTOR pathway genes, including TSC2, PTEN(phosphatase and tensin homolog) and PIK3CA(phosphatidylinositol-4,5-bisphosphate 3-kinase) occur in roughly 15% of pancreatic NETs. Other alterations in PI3K/AKT/mTOR pathway genes, including amplifications of AKT1/2, are observed in nearly one-third of small bowel carcinoid tumors.
The oral mTOR inhibitor everolimus has been extensively studied in GEP-NETs. In the RADIANT 1 trial, 160 patients with advanced, progressive pancreatic NETs were evaluated in 2 strata: everolimus monotherapy (n = 115) or everolimus plus octreotide (n = 45). Response rates and median PFS were 9% and 9.7 months in the monotherapy arm versus 4% and 16.7 months in the combined therapy arm, respectively. The RADIANT 2 trial randomly assigned 429 patients with hormonally functional carcinoid tumors to treatment with everolimus (10 mg) plus octreotide versus placebo plus octreotide. On central radiographic review, median PFS increased from 11.3 months on the control arm to 16.4 months on the experimental arm (HR 0.77, P = .026). While clinically significant, the primary endpoint fell short of its prespecified statistical significance threshold. One potential factor contributing to the lack of statistical significance was loss of progression events due to discrepancies in central versus local radiographic review. Objective responses were observed in only 5 patients in the everolimus arm (2%) versus 4 patients in the placebo arm (2%). There was no significant trend toward improvement in OS, a fact that may be attributable to the crossover design of the study. Everolimus is not currently approved by the FDA for the treatment of carcinoid tumors.
The RADIANT 3 trial randomized assigned 410 patients with low- and intermediate-grade pancreatic NETs to treatment with everolimus (10 mg) versus placebo. Despite an objective response rate of only 5% in the everolimus arm, the study demonstrated a clinically and statistically significant improvement in PFS, the primary endpoint. Median PFS increased from 4.6 months on the placebo arm to 11 months on the everolimus arm (HR 0.35, P<.001). OS differences were not observed, possibly due to the crossover design. As a result of the significant improvement in PFS, everolimus was approved by the FDA for treatment of patients with advanced pancreatic NETs.
Side effects of everolimus include aphthous oral ulcers, skin rash, diarrhea, hyperglycemia and cytopenias. Pneumonitis is a relatively rare but potentially serious toxicity. As an analog of rapamycin, everolimus has immunosuppressive properties and can cause atypical infections. While most toxicities are mild, cumulative and chronic side effects may impact patient quality of life.
Angiogenesis Inhibitors
Neuroendocrine tumors are among the most highly vascular of solid tumors and frequently express the VEGF ligand and receptor (VEGFR), which are key drivers of angiogenesis. Increased levels of circulating VEGF have been correlated with tumor progression. Consequently, inhibition of the VEGF pathway has been a promising therapeutic target. Most VEGF-targeting agents can be divided into 2 categories: tyrosine kinase VEGFR inhibitors and circulating VEGF inhibitors.
The tyrosine kinase receptor inhibitor sunitinib targets VEGFR-1, -2, and -3, as well as stem-cell factor receptor (cKIT) and platelet-derived growth factor receptor (PDGFR). In a phase II trial of 109 patients, treatment with sunitinib yielded objective response rates of 2.4% and 16.7% in carcinoid and pancreatic NETs, respectively. Based on the relatively high response rates in the latter cohort, a multi-national phase III study was launched that randomly assigned patients with low- and intermediate-grade pancreatic NETs to sunitinib (37.5 mg daily) versus placebo. The study was discontinued on interim analysis after enrollment of 171 patients and demonstrated a statistically significant improvement in median PFS from 5.5 months on the placebo arm to 11.1 months on the sunitinib arm. There was a trend toward improvement in OS, which was not statistically significant in the mature analysis. The objective response rate associated with sunitinib was 9.3%. Side effects of sunitinib included diarrhea, nausea, fatigue, cytopenias, hypertension, and palmar-plantar erythrodysesthesia. Despite these toxicities, no overall differences were noted between study groups in global health-related quality of life. As a result of this study, sunitinib is FDA approved to treat pancreatic NETs.
Bevacizumab is a monoclonal antibody to circulating VEGF-A. In a randomized phase II trial, 44 patients with metastatic carcinoid tumors were randomly assigned to treatment with bevacizumab or pegylated alpha-IFN (PEG-IFN) for 18 weeks, followed by administration of both agents in combination. At the week 18 time point, the rate of PFS was 95% on the bevacizumab arm versus 68% on the PEG-IFN arm. Moreover, the objective radiographic response rate in the bevacizumab arm was 18% versus 0% with PEG-IFN. On functional computed tomography (CT) scans performed at baseline and on day 2 of therapy, bevacizumab treatment yielded an average reduction in tumor blood flow of 49%. An ongoing phase III study led by the South West Oncology Group (SWOG) is comparing bevacizumab to alpha-IFN in patients with metastatic carcinoid tumors with a primary endpoint of PFS. The study has completed accrual, and the results are pending.
Cytotoxic Chemotherapy
Several factors predict the sensitivity of GEP-NETs to cytotoxic chemotherapy, including tumor grade, differentiation, and primary site. Poorly differentiated neuroendocrine cancers, which can be divided into small cell and nonsmall cell categories, are generally characterized by a high mitotic rate, a Ki-67 proliferation index exceeding 50%, and tumor necrosis. The treatment of these aggressive malignancies is generally similar to the treatment of small cell lung cancer. A study evaluating cisplatin and etoposide in GEP-NETs reported a response rate of 67% in poorly differentiated tumors versus 7% in well-differentiated tumors. Another study of cisplatin and etoposide in poorly differentiated GEP-NETs demonstrated a response rate of 42%. The durations of response in both studies were short (8–9 months) and with median survivals of only 15–19 months.
The chemosensitivity of low- to intermediate-grade tumors correlates with primary tumor site. Among GEP-NETs, pancreatic primary tumors appear to be most highly sensitive to cytotoxic agents, including streptozocin (STZ), dacarbazine, temozolomide, and fluoropyrimidines. One of the first cytotoxic drugs tested in pancreatic NETs was the nitrosourea STZ, an agent that is specifically toxic to pancreatic islet beta-cells. Based on promising phase II data, 2 randomized trials were conducted by the Eastern Cooperative Oncology Group (ECOG) in the 1970s and 1980s. The first of these studies reported response rates of 63% with STZ plus 5-fluorouracil (5-FU) versus 36% with STZ monotherapy. The second trial compared STZ plus doxorubicin versus STZ plus 5-FU and reported response rates and TTPs of 69% and 20 months versus 45% and 6.9 months, respectively. STZ plus doxorubicin also had a significant advantage in terms of survival (median OS 2.2 versus 1.4 years).
The high response rates associated with STZ-based regimens were subsequently questioned due to the partial reliance on nonradiographic response criteria, such as improvement in tumor markers or reduction in hepatomegaly. For example, in a small retrospective series of 16 patients, only 1 patient achieved an objective response by CT scan criteria. A more recent retrospective study investigating the combination of STZ, 5-FU, and doxorubicin in 84 pancreatic NETs reported a response rate of 39% using objective radiographic criteria and a median response duration of 9.3 months. Enthusiasm for STZ-based chemotherapy is somewhat tempered by toxicity concerns, including myelosuppression, nausea, and renal insufficiency.
In recent years, the oral alkylating agent temozolomide has emerged as an active agent in pancreatic NETs. Temozolomide shares a metabolite with dacarbazine and is converted to the active alkylator MTIC (3-methyl-[triazen-1-yl]-imidazole-4-carboxamide), which induces DNA methylation at the O-6 position of guanine. A phase II study investigating the combination of temozolomide and thalidomide reported an objective response rate of 45% in the pancreatic NET subset of patients (5 of 11 patients). A recent retrospective study of temozolomide combined with capecitabine in 30 chemo-naïve pancreatic NET patients reported an objective radiographic response rate of 70% and median progression-free survival (PFS) of 18 months. A phase II trial of temozolomide combined with bevacizumab reported a response rate of 33% and median PFS of 14.3 months in the pancreatic NET cohort, which consisted of 15 patients. Side effects of temozolomide are relatively tolerable when patients receive appropriate nausea prophylaxis and include thrombocytopenia and lymphopenia. The data supporting temozolomide-based regimens in pancreatic NETs consists, thus far, of retrospective series and small cohorts of prospective trials. An ECOG-sponsored prospective randomized trial of temozolomide monotherapy versus capecitabine plus temozolomide will provide much needed prospective data in a large cohort of patients.
Well-differentiated carcinoid tumors appear to be significantly more resistant to cytotoxic chemotherapy. A randomized clinical trial of doxorubicin plus 5-FU compared with STZ plus 5-FU reported response rates of 16% and median PFS of 5 months in both arms. In the phase II study of temozolomide plus thalidomide, the response rate in patients with carcinoid tumors was only 7% (compared to 45% in the pancreatic NET cohort). In the phase II study of temozolomide plus bevacizumab, the response rate in the carcinoid tumor cohort was 0% (compared to 33% in the pancreatic NET cohort). There are several potential explanations for the differences in chemosensitivity. In 1 study, carcinoid tumors expressed a much higher level of methyl-guanine-methyl-transferase (MGMT), a DNA repair enzyme, compared to pancreatic NETs. Additionally, the low mitotic activity of certain carcinoid tumors, particularly those originating in the midgut, may confer higher resistance to cytotoxic agents than pancreatic NETs, which tend to have a higher proliferative rate.
Radiolabeled SSAs
An emerging systemic treatment modality consists of radiolabeled SSAs. This targeted form of systemic radiation, also known as peptide receptor radiotherapy (PRRT), enables the delivery of radioactive isotopes to somatostatin receptor-expressing tumors. Nearly 80% of GEP-NETs express somatostatin receptors as evidenced by radiotracer uptake on somatostatin-receptor scintigraphy (SRS, OctreoScan®). Radiolabeled SSAs consist of a radionuclide isotope, an SSA (peptide), and a chelator used to bind them. Commonly used SSAs include octreotide and octreotate, an analog with enhanced binding to sst2. Chelators include DPTA (diethylenetriamine-penta-acetic acid) and DOTA (tetraazacyclododecane-tetra-acetic acid). The main selection criterion for PRRT is evidence of strong radiotracer expression (ideally higher than normal liver tissue) on OctreoScan or another somatostatin receptor imaging modality.
Early clinical trials of PRRT used octreotide labeled with high, cytotoxic doses of In-pentetreotide, the isotope used in OctreoScans. Although symptomatic responses were observed in some cases, objective radiographic responses were rare, possibly due to the short tissue penetration of Auger electrons emitted by the In isotope. The next generation of radiolabeled SSAs used yttrium (Y), a high-energy β-particle emitter with a maximum tissue penetration range of approximately 12 mm. In 1 study, objective responses associated with Y-DOTA-Tyr3-octreotide (Y-DOTATOC) were observed in 28% of 87 patients with advanced GEP-NETs. However, in a subsequent multicenter trial of GEP-NETs using doses of Y-DOTATOC up to 886 mCi in 4 cycles, a response rate of only 9% was reported. Another phase II study enrolled 90 patients with metastatic hormonally functioning carcinoid tumors who experienced symptom progression while on SSAs. Treatment consisted of Y-DOTATOC at a dose of 4.4 GBq times 3 cycles. While the majority of patients (74%) appeared to obtain clinical benefit, only 4 patients (4%) achieved a partial radiographic response. The median PFS and OS were 16.3 and 26.9 months, respectively. Toxicities associated with Y-DOTATOC include renal insufficiency, which is partially ameliorated by concurrent amino acid infusions. In 1 large institutional series of 1,109 patients, 102 patients (9%) had severe permanent renal toxicity. Acute GI toxicities, including nausea and vomiting, are primarily attributable to the amino acid infusions. Myelosuppression is another side effect of PRRT, with grade 3 and 4 hematological toxicities occurring in approximately 13% of cases.
A newer generation of radiolabeled SSAs employs lutetium-177 (Lu), a medium energy β emitter with a maximal tissue penetration of 2 mm and a half-life of 6.7 days. Lu also emits γ rays allowing for scintigraphy and dosimetry using the same compound. The most commonly used peptide-isotope compound is Lu-DOTA-Tyr3-octreotate (Lu-DOTATATE). In 1 series of 310 patients with GEP-NETs, an objective response rate of 30% was reported. Response rates were particularly high in pancreatic NETs ranging from 36% in nonfunctioning pancreatic NETs to approximately 40 to 60% in functional gastrinomas, insulinomas, and VIPomas. Median TTP was 40 months, consistent with a high rate of disease control. Side effects of Lu-DOTA-Tyr3-octreotate are primarily hematological and nephrotoxicity rates are lower than observed with Y labeled compounds. An international phase III trial of Lu-DOTATATE versus high-dose octreotide is now open for patients with midgut NETs whose tumors have progressed on standard-dose octreotide.