Hepatic Embolotherapies for Metastatic Neuroendocrine Tumors: An Interview With Jonathan Strosberg, MD

Dr Strosberg is a Senior Member at Moffitt Cancer Center in Tampa, Florida. He spoke with IO Learning at CIO 2019 about his experience with various embolization techniques and the latest data on these treatment modalities.

What modalities are available for treatment of neuroendocrine tumors, and what role did the RETNET trial play?

Embolization has been a widely used therapy for metastatic liver dominant neuroendocrine tumors for decades, starting with bland embolization and chemoembolization. More recently, drug-eluting beads came into the foreground, and then radioembolization using yttrium 90 resin or glass microspheres. There’s been a long-standing debate as to which is the optimal embolization modality. Nearly all of the data are based on retrospective series. There are few prospective data and virtually none of it is randomized.

The only randomized trial that I’m aware of is a study comparing bland embolization with chemoembolization, but it didn’t meet accrual and was halted midway. At that point, the trial hadn’t shown much of a difference, but it was highly under-powered, so the debate continues. RETNET is an attempt to answer at least part of this question by comparing bland embolization, chemoembolization, and drug-eluting beads in a 3-arm study (not radioembolization). The drug-eluting bead arm was terminated for safety reasons. There seemed to be a higher toxicity with drug-eluting beads in this particular population, especially with respect to biliary injury and necrosis. There have been previous reports of this toxicity, suggesting that this particular technology may be inferior or more toxic, but we’re still randomizing patients to the bland and chemoembolization arms. I’m not sure there’s a huge difference between those two modalities.

Of course, neuroendocrine tumors are treated as one entity, when in fact it is an extremely heterogeneous disease and our systemic therapies are not identical. Pancreatic neuroendocrine tumors are much more chemosensitive, for example, than the average small bowel neuroendocrine tumor. Similarly, the notion that there is a “one size fits all” for embolization is probably mistaken. However, at this point, prospective trials are just beginning, and we need to group together different questions.

What type of differentiation do you hope to see in future studies?

The RETNET trial is a start. I don’t know if it’s going to be possible in the future, but it would be nice to see the different types of tumors, such as small bowel and the pancreas, at least segregated by primary side.

Is long-term toxicity a problem with any modalities?

We did bland embolization for 7 years. After several years, we switched primarily to radioembolization and then subsequently, we started to see indications of long-term toxicity.

When did this toxicity become apparent in your patients?

It became apparent at more than 6 months in most patients, although there were a few patients whose livers became fibrotic and small within a short period of time. We started to see a cirrhotic fibrotic appearance to the liver and a decrease in liver size, and then jaundice. This occurs in 10%-20% of patients, and has not been seen with bland embolization.

It’s possible that the affected patients received a relatively high radiation dose. The doses have not been standardized, so it’s difficult to determine if dose was a factor. Nonetheless, we stopped performing radioembolization routinely for typical slow-growing, bilobar liver disease. At this point, several publications have reported similar findings, and we have presented some of our data at meetings and will hopefully publish soon.

Can you tell us more about bilobar liver disease? 

I think there is a role for radioembolization if one is performing a single-lobe embolization and can preserve liver function with the contralateral lobe. I tend to use radioembolization in more aggressive, localized tumors, as it is probably a more effective way to eradicate individual tumors. One can direct a large amount of radiation to a small area, which may be one area where radioembolization is superior to other modalities. In fact, we’ve seen cases where patients with aggressive non-resectable, non-ablatable tumors have not responded to bland embolization and then respond well to radioembolization.

Are there any other issues regarding radioembolization-induced liver disease?

One issue is that many of the early radioembolization studies in neuroendocrine tumors did not have very good long-term follow-up. They were initial reports. It’s only recently that we’ve started to see more long-term issues arise.

Can you tell me more about situations in which there is a role for radioembolization?

This is more based on anecdotal experience than data, but I find that bland embolization doesn’t work that well in patients with relatively rapidly growing tumors. As I said earlier, neuroendocrine tumors are extremely heterogeneous and liver-directed embolotherapy is not a “one size fits all” treatment. There are almost certainly cases where radioembolization is better than bland embolization, and other cases where bland embolization is better than radioembolization. I’m not sure trials can determine which method is better in any given situation. For example, if a patient has a highly vascular tumor that’s going to pick up a lot of the radiation, and the operator can spare the surrounding liver, radioembolization works quite well and there is little collateral toxicity.

Can you tell us more about differences between chemoembolization and bland embolization?

The differences are relatively minor, and some institutions prefer bland embolization. We generally use bland embolization, while others use chemoembolization. However, it depends on the situation. For example, based on their general chemo-resistance, small-bowel neuroendocrine tumors do not generally have superior outcomes with chemoembolization.

Your presentation mentioned a combination of peptide receptor radionuclide therapy (PRRT) and radioembolization. Can you elaborate? 

There’s been some conflicting data regarding the combination, but in theory, it’s obvious that if the liver disease is extensive and the patient has already received a large radiation dose in the past, further radiation in the form of PRRT can lead to trouble. We have already seen this in 1 patient. These cases will be sporadic because they really depend on how much radiation was received in the past and how much additional radiation is delivered by PRRT.

Have clinical trials for embolization been able to individualize treatment strategy?

When it comes to studies for systemic therapies, oncology research almost always separates pancreatic neuroendrocrine tumors from gastrointestinal neuroendocrine tumors, although within the gastrointestinal neuroendrocrine tumor category, each primary site is actually quite unique. Embolization studies, however, have not separated these two groups. Ideally, this will be done in future studies, because these tumors are quite different.

Can you give an example of how you tried to individualize your treatment strategies?

We don’t individualize too much. When we do an embolization, we almost always use bland embolization. When I see a tumor grow relatively aggressively, particularly if it’s a pancreatic neuroendocrine tumor, I’ll sometimes use chemoembolization in the hope that the tumor is chemosensitive, even though most patients with pancreatic neuroendocrine tumors have already received chemotherapy. More often, we’ll do radioembolization if the tumors are growing rapidly in a limited part of the liver.

Any final thoughts?

First, embolization continues to be a really important treatment modality for liver dominant disease in neuroendocrine tumors, despite the growth of systemic treatments. We see higher response rates with embolization than with almost any other treatment, and we use it for symptomatic control in patients with hormonal syndromes. Second, in my opinion, radioembolization is probably not the best strategy for a typical slow-growing tumor, such as a small-bowel neuroendocrine tumor with diffuse liver metastases.