RETNET: Exploring Therapies for Liver Metastases in NETs

For patients with neuroendocrine tumors (NETs), the development of liver metastases is a pivotal event, occurring in 9 of 10 cases. The control of these metastases is the primary determinant of both patient survival and symptomatic relief. Consequently, liver-directed therapy, specifically transarterial embolization (TAE), has been a cornerstone of treatment for nearly half a century.

“Embolization of symptomatic neuroendocrine liver metastases started in the 1970s in Europe, and then in the 1980s [physicians] started doing chemoembolization, where they mix some concentrated chemotherapy in with the embolics that we were injecting in the arteries to the tumors in the liver,” explained Michael Soulen, MD, in an interview with Targeted Oncology®.

“If you look at the guidelines for neuroendocrine tumors, it's pretty clear that for progressive or symptomatic liver metastases, liver active therapy is indicated, but there are a number of techniques for doing it, and there is no recommendation or preference among them because there was no data. So even though we have done this for a very long time, there were no randomized trials among techniques of liver embolization,” said Soulen, director of interventional oncology at the Abramson Cancer Center at the University of Pennsylvania.

The RETNET study (NCT02724540)^1,2 was designed to fill this critical evidence gap by directly comparing the primary techniques in a multicenter, global, randomized setting.

What was the design and methodology of RETNET?

The RETNET trial was an international, pragmatic study designed to evaluate the efficacy, toxicity, and health-related quality of life associated with TAE (or bland embolization) and conventional transarterial chemoembolization (cTACE).¹ Soulen explained that bland embolization involves “plugging up the arteries” to the tumors without the use of drugs, while chemoembolization, as the name suggests, comprises “emulsify[ing] drugs with an oily contrast agent and instill[ing] that to fill the tumor vasculature.”

Soulen noted that RETNET initially also included an arm assessing drug-eluting embolics, where drugs are loaded onto particles that are then instilled into tumor vasculature. However, the drug-eluting embolics were too toxic, and the arm was closed at the first safety review. This finding was so definitive that the National Comprehensive Cancer Network guidelines were promptly updated to advise against using drug-eluting embolics for NETs.

Between 2017 and 2022, 151 patients were randomly assigned, with 78 patients undergoing TAE and 73 undergoing cTACE. The primary end point was hepatic progression-free survival (HPFS).

Patients enrolled had NET liver metastases of any grade or origin that were progressive, symptomatic, or represented a high tumor burden (> 25%). Baseline demographics were well matched between the TAE and cTACE arms. Primary tumor sites included the midgut (54%) and pancreas (36%), with the majority of patients having grade 2 tumors (56%).

What were the key findings?

The analysis revealed no significant difference in effectiveness of TAE and cTACE at controlling disease. This outcome aligns with historical observations, as a significant superiority of 1 technique likely would have become apparent over the 40 to 50 years of their concurrent use, according to Soulen. The median HPFS was 18 months for TAE vs 13 months for cTACE (HR, 1.40; P =.234), a difference that was not statistically significant.

The median overall PFS was 15.9 months for TAE vs 11.1 months for cTACE (HR, 1.43; P =.133), also not statistically significant. Further, changes in the carcinoid symptom severity score were similar between the groups, indicating comparable effectiveness in managing hormonal symptoms.

However, and surprisingly, the trial demonstrated that bland embolization carried a higher risk of serious adverse events than conventional chemoembolization.

“Many thought there’s chemotherapy in the chemoembolization arm, so that’s going to be worse,” Soulen said. “The adverse events were about doubled in the bland embolization arm,” with serious adverse events about 50% higher with bland embolization.

Major complications occurred significantly more frequently in the TAE arm at 44% (n = 34) compared with 29% (n = 21) in the cTACE arm. In contrast, grade 3 to 4 toxicities were more frequent in the cTACE arm at 54% (n = 42) compared with 36% (n = 26) in the TAE arm.

This distinction in safety metrics highlights a complex risk profile. While cTACE was associated with more high-grade toxicities (likely laboratory abnormalities or expected post-embolization symptoms), TAE led to a greater number of serious complications requiring an elevated level of care.

What are the future directions?

The RETNET trial provides definitive, high-level evidence that fundamentally clarifies the role of embolization therapy for NET liver metastases. The trial data set is still undergoing analysis. Two key areas remain to be explored. First, a full analysis of [patient-reported outcomes] has not yet been completed, but will provide crucial insight into the patient experience and quality of life with each therapy. Secondly, the adverse event data, currently based on investigator assessment, requires a more detailed independent review to fully understand the nature and cause of the observed differences in safety between the TAE and cTACE arms.

REFERENCES:

1. Soulen M. Randomized embolization trial for neuroendocrine tumors (RETNET). Presented at: 2025 North American Neuroendocrine Tumor Society Symposium; October 23-25, 2025; Austin, TX.

2. Randomized embolization trial for neuroendocrine tumor metastases to the liver. ClinicalTrials.gov. Updated January 6, 2025. Accessed October 24, 2025. https://www.clinicaltrials.gov/study/NCT02724540