We are talking about all these exciting new treatments in immunotherapy, but the reality is surgery remains the only way to cure a neuroendocrine tumor. The only way to get rid of it right now is to cut it out. Surgery is always the first question for someone with localized disease. That is clearly the first option. Particularly for the low-grade tumors, it is often curative, and patients can often do quite well. Radiation therapy can have a role in patients who are locally symptomatic, so those with bone metastases, for example. External beam radiation sometimes can be helpful in that context. One of the more exciting developments in recent years has been the development of peptide receptor radiotherapy. That is a targeted radiation, a way of administering radiation systemically in a targeted fashion that turns out to be highly effective as a treatment option.

In recent years, there has almost been an explosion in treatment options for neuroendocrine tumors that sometimes actually makes it difficult to decide which treatment to choose first, second, and third. Really, the first set of treatment options for most patients, I would say, are somatostatin analogues. These are agents that target the somatostatin receptor, which is expressed on 90% of neuroendocrine tumors. There are 2 commonly used ones that are used, lanreotide and octreotide. They have very few side effects. They are given as monthly injections. They have a couple benefits. One is that they can decrease hormone secretion for tumors that secrete hormones. Another important action that has been identified more recently is the fact that these agents can also slow tumor growth and improve progression-free survival (PFS).

They have very few adverse effects. They are highly effective in many patients, and that is usually our first treatment option for patients with advanced neuroendocrine tumors. Another therapeutic option that has been around for a long time is, of course, standard cytotoxic chemotherapy. The story here differs depending on what type of neuroendocrine tumor a patient has. For extra-pancreatic neuroendocrine tumors, standard cytotoxic chemotherapy tends not to be particularly effective. There are obviously some exceptions. Where cytotoxic chemotherapy can have a role is in the treatment of advanced pancreatic neuroendocrine tumors. There, what we have learned more recently is that the oral chemotherapy drug temozolomide can be quite active. In fact, a combination of temozolomide and capecitabine was compared to temozolomide alone in a randomized phase 2 trial. That trial showed that the combination of temozolomide and capecitabine was associated with encouraging response rates in the order of 30% to 40%, improved progression-free survival, and improved overall survival, too.

Other agents that are effective in neuroendocrine tumors include so-called targeted therapies, molecularly targeted therapies. A couple classes are active here. One of those classes are M4 [muscarinic acetylcholine receptor] inhibitors and everolimus is used in both pancreatic neuroendocrine tumors and extra-pancreatic neuroendocrine tumors. VEGF) inhibitors form an important class of therapeutic agents for neuroendocrine tumors. We will be talking more about that, but several tyrosine kinase inhibitors (TKIs) have shown activity in this setting. Several are now being approved so that they are more widely available. Finally, an exciting area is immunotherapy, which at least, at this point, seems to be primarily active in the aggressive neuroendocrine tumors, the high-grade neuroendocrine tumors. The big question, as we have been discussing here, is how might it be possible to make low-grade tumors responsive to immunotherapy? One additional important class of therapeutic agents for neuroendocrine tumors is one that I also mentioned previously, peptide receptor radiotherapy. This is an interesting approach that has shown very good result and is now FDA approved. The idea here is to take a somatostatin analogue, attach a radioisotope to it, and then use the somatostatin analogue to deliver very targeted radiation therapy to the tumor wherever it might be in the body. This is a highly effective approach for both pancreatic and extra-pancreatic neuroendocrine tumors.

One of the big challenges with advanced neuroendocrine tumors is deciding which treatment to give. Usually, the first choice of treatment is a relatively easy one. It is usually a somatostatin analogue based on the low side effect profile and ease for the patient and efficacy. Where it becomes much more challenging is for patients who progress on somatostatin analogues. Here, we are really faced with a choice. We can use cytotoxic chemotherapy, at least for pancreatic neuroendocrine tumors. We have got molecularly targeted agents, like the MTOR (mechanistic target of rapamycin) inhibitors, the VEGF inhibitors, tyrosine kinase inhibitors, and now peptide receptor radiotherapy. All of these are active approaches. In many ways, it depends on the patient. It depends on what the side effect profile is, what the comorbidities might be of the patients, and whether the goal is to just stop tumor growth or actually to try to shrink the tumors in situations where there is high tumor bulk. It ends up being a very patient-specific, and sometimes subjective, decision to decide which treatment to use first, second, and third. We really do not have any hard data at this point on tumor sequencing and outcomes.

Videos

Matthew H. Kulke, MD, provides a broad overview of initial therapy options available for patients with neuroendocrine tumors.

Neuroendocrine Tumors: Options for Initial Therapy

Neuroendocrine tumors are actually a very broad category of malignancies. We tend to classify them broadly in 2 big groups. One group would be pancreatic neuroendocrine tumors and the other group would be extra-pancreatic neuroendocrine tumors. Beyond that, pancreatic neuroendocrine tumors can be classified based upon the type of hormone they secrete, and they can secrete a whole range of hormones. People have heard of insulinomas, gastrinomas, etc. Then, the extra-pancreatic neuroendocrine tumors are classified, or subclassified, based on their site of origin. Common sites of origin include the lung, with bronchial neuroendocrine tumors, or, for example, small intestine neuroendocrine tumors. That is 1 big way to classify them. Then, the other important way to classify them, and something that really drives treatment, is the histology. Neuroendocrine tumors can range from very low-grade, well-differentiated tumors that tend to grow slowly to very aggressive, high-grade neuroendocrine tumors. Based on the histology, we sometimes select treatments differently.

One of the challenges does relate to the histology. There is this range of behaviors from aggressive to very indolent. Sometimes, it is a little bit tricky to decide just how you want to treat them. Another challenge can be in the treatment of low-grade well-differentiated neuroendocrine tumors. They tend to grow very slowly, which is a good thing, but at least historically it has been really challenging to identify treatments that work, especially treatments that can help shrink the tumors. That is something that has changed in recent years, and that has made the field really exciting.

Molecular testing is an interesting topic in neuroendocrine tumors. Across the board in most cancers, we tend to—almost reflexively now—get molecular testing to identify potentially-targetable mutations. In most neuroendocrine tumors, and particularly the low-grade neuroendocrine tumors, it turns out there are very few mutations. There are very few particularly in extra-pancreatic neuroendocrine tumors. In pancreatic neuroendocrine tumors there are 3 common ones: 

. Those are all tumor-suppressor genes and therefore hard to target therapeutically. I think the answer to that question is that while we do get genomic testing and molecular testing, it does not always lead to a potential targetable mutation or potential decision that can help us with treatment. One exception to that is, in some cases, neuroendocrine tumors are associated with rare inherited genetic syndromes like 

. Von Hippel-Lindau disease is another one. For patients like that, you certainly do want to get germline testing, family counseling, genetic counseling, et cetera.

Not a lot is known about the microenvironment. That is an active area of research and a very important, I would say, area of research in neuroendocrine tumors right now. The aggressive neuroendocrine tumors do appear to be responsive to checkpoint inhibitors, suggesting that the environment is not particularly immunosuppressive. The big challenges right now are the low-grade neuroendocrine tumors where they seem to be very quiet genomically and potentially have an immunosuppressive microenvironment. A big question right now is can you change that, and can you potentially make these low-grade neuroendocrine tumors responsive to immunotherapy?

Expert insight on the subtypes of neuroendocrine tumors and current understanding of the disease microenvironment.

Matthew H. Kulke, MD, Dana-Farber Cancer Institute

Articles

Neuroendocrine Tumors: Options for Initial Therapy

The Nature of Neuroendocrine Tumors