Neuroendocrine Tumors: Is There a Standard Treatment?

Neuroendocrine tumors have long presented physicians and patients with a paradox: while they are generally indolent in nature, they can also be quite difficult to treat. The review by Strosberg et al in this issue of GCR illustrates that, while the inherent nature of these tumors may not be changing, the therapeutic options available to treat neuroendocrine malignancies are undergoing a rapid evolution.¹ In particular, novel “targeted” therapies hold promise for patients suffering from advanced disease. Despite these advances, however, there remains little consensus on a standard treatment approach for these patients.

In some respects, the unique clinical characteristics of neuroendocrine tumors may also be their Achilles heel. Neuroendocrine tumors have long been known to be highly vascular, and more recent studies have demonstrated elevated expression levels of vascular endothelial growth factor (VEGF), together with its receptor. ² The development of specific therapies targeting the VEGF pathway created an opportunity to advance beyond traditional cytotoxic therapies in this disease.

Bevacizumab, a monocolonal antibody directing against the VEGF ligand was one of the first such treatments evaluated in neuroendocrine tumors.³ A randomized phase II study in advanced carcinoid tumor patients suggested that treatment with bevacizumab was associated with objective tumor responses and improvement in progression-free survival when compared to treatment with alpha interferon. A second phase II study, demonstrated that treatment with sunitinib, a tyrosine kinase inhibitor targeting not only the VEGF receptor but also PDGFR, RET, and c-Kit, resulted in tumor responses in 16% of pancreatic neuroendocrine tumor patients.⁴ Most recently, similar antitumor activity was reported with the tyrosine kinase inhibitor sorafenib.⁵

VEGF pathway inhibitors are not the only promising new treatment approach for neuroendocrine tumors. The mammalian target of rapamycin (mTOR) is a serine threonine kinase that functions downstream of a number of receptor tyrosine kinases, including the VEGF receptor, and serves a central role in regulating cell growth. Inhibitors of mTOR represent a second class of “targeted” agents that have shown preliminary evidence of activity in neuroendocrine tumors. While the results of a phase II study evaluating temsirolimus were disappointing, responses were observed in 18% of carcinoid patients and 20% of pancreatic neuroendocrine tumor patients in a phase II study of the oral mTOR inhibitor everolimus.^6,7

These initial results have led to the development of large, multi-institutional randomized studies to more definitively evaluate the efficacy of VEGF pathway inhibitors and mTOR inhibitors in neuroendocrine tumors. One such study, led by the Southwest Oncology Group (S0518), is comparing treatment with bevacizumab alone to treatment with alpha interferon in patients with advanced carcinoid tumors. In a parallel effort, an international industrysponsored study is comparing treatment with sunitinib to placebo in pancreatic neuroendocrine tumor patients. Everolimus is currently being evaluated in two definitive phase III studies, one enrolling pancreatic neuroendocrine tumors and the other carcinoid tumors.

Evidence that both VEGF pathway inhibitors and mTOR inhibitors are active in neuroendocrine tumors suggests that additional agents targeting the broadly defined receptor tyrosine kinase/PI3-Kinase/AKT/mTOR cell signaling pathway may also be active in this disease type. Newer agents in development, including inhibitors of IGF1-R and PI3-Kinase, target different aspects of this same pathway. Preliminary evidence of activity in neuroendocrine tumors has already been reported with an IGF1-R inhibitor; it will be interesting to observe if activity with IGF1-R inhibitors and related compounds is confirmed.⁸

Amid the excitement associated with these novel agents, it is perhaps easy to forget that older treatment strategies may still be effective. Somatostatin analogs are clearly established as an effective treatment for patients with symptoms of hormonal hypersecretion. A number of recently developed somatostatin analogs may yield incremental improvements in symptom control. An as yet unanswered question is whether somatostatin analogs may also help control tumor growth—and if so, whether they should also be used in patients with nonsecretory tumors.

Given the high prevalence of somatostatin receptor expression on neuroendocrine tumors, as well as the efficacy of “cold” somatostatin in controlling symptoms, the use of radiolabeled somatostatin analogs is inherently appealing. A number of different radiopeptides incorporating Indium-111, Yttrium-90, or Lutecium-177 have been associated with both biochemical and radiologic responses in phase I and phase II studies.^9–16 Despite the promise of this approach, there is still little consensus on the optimal isotope and dose schedule for these agents, nor is there a broad understanding of their tolerability. Rigorous studies of these agents, incorporating consistent entry criteria and established dosing procedures, are warranted to better define their role in the treatment of neuroendocrine tumor patients.

The use of cytotoxic chemotherapy may also still play a role in the treatment of selected patients with neuroendocrine tumors. In most carcinoid tumor patients, there appears to be little if any benefit associated with traditional cytotoxic regimens. In contrast, streptozocin-based regimens are active in patients with pancreatic neuroendocrine tumors; in fact, streptozocin is currently the only FDA-approved drug for this indication.¹⁷ Because patients with advanced neuroendocrine tumors may have an excellent performance status for years, the routine use of these regimens, in light of their potential toxicity, is often questioned. Newer chemotherapy regimens incorporating temozolomide may provide a more attractive systemic alternative and are currently being evaluated in clinical studies.^18,19

Translating these many treatment choices into a rational treatment program, remains a daunting task. The current lack of a standardized treatment approach for patients with advanced disease is related to several factors. Among the most important of these factors has been the limited availability of randomized studies directly comparing different treatment modalities. The indolent nature of neuroendocrine tumors, while beneficial from the standpoint of the patient, has made it difficult to design practical end points for such clinical trials. Most studies of neuroendocrine tumors using traditional response criteria have reported high rates of stable disease, and the variable clinical course of these tumors has made it difficult or impossible to determine whether stable disease represents drug effect or the natural history of the disease. For similar reasons, the prolonged survival times of patients with advanced neuroendocrine tumors make the evaluation of survival data challenging. Current randomized trials are appropriately evaluating time to disease progression rather than overall survival as a primary end point, and therefore provide one means to obtain efficacy data for new drugs in a timelier manner. Alternatively, the development of surrogate end points of response, including the validation of biochemical markers as response end points, may expedite the development of novel agents for this disease.

The ongoing randomized studies of bevacizumab, sunitinib, and everolimus represent a first step in better defining treatment options for neuroendocrine tumor patients. If successful, these studies should also lead to a clearer path to regulatory approval in neuroendocrine tumors, facilitating the development of additional therapeutic agents in this disease. We can, perhaps, look forward to a time in the near future when neuroendocrine tumors will be not only indolent but also susceptible to a broad range of therapeutic agents.