New Study Strengthens the Case for EUS-RFA in NF-PanNETs

The incidence of pancreatic neuroendocrine tumors (PanNETs) is rising in the U.S.—particularly nonfunctioning PanNETs (NF-PanNETs), which represent the majority of cases. Established clinical management of NF-PanNETs is limited to either surveillance or surgical resection, yet both approaches have distinct drawbacks: surveillance carries a risk of progression or metastasis, while surgery carries risks of morbidity and mortality.

Endoscopic ultrasound-guided radiofrequency ablation (EUS-RFA) has emerged as a minimally invasive alternative, but the method’s long-term outcomes and optimal procedural strategy remain unclear.

A study led by NYU Langone Health researchers offers new insights on those questions. “Our primary goal was to establish the technical success rates, safety, and efficacy of EUS-RFA in treating NF-PanNETs across multiple tertiary and quaternary referral centers,” says lead author Tamas A. Gonda, MD. “We also wanted to identify procedural and patient-specific variables associated with treatment response.”

“In sum, our study supports EUS-RFA as a safe and effective treatment for small, asymptomatic NF-PanNETs,” Dr. Gonda says. “Not everyone responded equally, but almost everyone eventually responded.”

Building on Past Research

Prior case series, though few and small, have reported a complete response rate of 85.7 to 93 percent, with an adverse event (AE) rate of 10 to 13 percent. Supported by these findings, EUS-RFA is increasingly used to treat low-risk NF-PanNETs in select patients.

Still, Dr. Gonda notes, “To establish EUS-RFA as a standardized therapeutic option for these tumors, and to inform future clinical guidelines, we need larger, longer-term studies, with uniform treatment and follow-up protocols.”

The new study, published in Techniques and Innovations in Gastrointestinal Endoscopy, is the largest multicenter, international analysis of EUS-RFA for NF-PanNETs to date. Conducted across seven high-level care centers, the study included 46 patients who underwent the procedure between 2021 and 2024. All had tumors measuring less than 3 cm, without evidence of metastasis.

“The idea was to gain a better understanding of how EUS-RFA works in the real world, where there are differences in how patients are treated at each center, or even within a given center,” Dr. Gonda explains.

EUS-RFA was performed using a dedicated 19-gauge needle electrode and a linear array echoendoscope. Radiofrequency current was applied using a 5-, 7-, or 10-mm probe, selected at the proceduralist’s discretion. Investigators adhered to the manufacturer’s recommended energy settings, which increased with tip length. The number of applications varied based on the endoscopist’s discretion but was guided by the manufacturer’s recommendation.

Uncovering the Dynamics of Treatment Response

Consistent with earlier studies of EUS-RFA for NF-PanNETs, the researchers report a high overall response rate—91.1 percent at 12 months.

“The evolution of treatment response over time was particularly striking,” Dr. Gonda notes.

Initial response: At initial follow-up, at least 10 weeks after the index ablation, 32.6 percent of patients had a complete response (CR), 47.8 percent had a partial response (PR), and 19.6 percent had no response (NR).

Repeat RFA: Nearly 60 percent of patients with NR or PR underwent additional RFA sessions. Of these, 44.4 percent converted to CR, and 5.6 percent from NR to PR. Among non- or partial-responding patients who didn’t undergo repeat RFA, about one-third nonetheless went on to convert to CR by maximal follow-up. “It may take longer to see a response depending on how the procedure is performed, or some patients may simply take longer to respond than others,” says Dr. Gonda.

Cumulative response: By 12-month follow-up, 60 percent of patients had a CR, 31.1 percent had a PR, and 8.9 percent had NR.

The overall incidence of AEs was 15 percent after index RFA, including abdominal pain, bleeding, and acute pancreatitis. Aside from one patient who required a distal pancreatectomy, no surgical interventions were needed. No additional AEs were reported for patients who underwent repeat RFA.

Looking for Factors that Boost—or Inhibit—Efficacy

Another noteworthy finding concerned predictors of response. The team found that two technical factors were significantly associated with achieving a CR at initial follow-up: the use of a probe with a 7- or 10-mm needle tip, and a higher number of RFA applications at the index procedure. “These results suggest that if you deliver a bit more energy, you may see a quicker response,” Dr. Gonda says.

However, there were no variables predictive of response at maximal follow-up, including lesion size, type, grade, or location. Nor were any variables found to be predictive of conversion to CR among patients with NR or PR who underwent repeat RFA.

“There were no obvious tumor characteristics that made certain types of lesions better or worse candidates for treatment,” Dr. Gonda says. “Having established that the procedure is safe and effective, the next focus is standardized protocols.”