Endoscopic small intestinal ablation is an investigational therapy for patients with type 2 diabetes (T2D) designed to replicate the metabolic benefits of gastric bypass surgery. To date, ablation has been limited to 10-15 cm of the post-ampullary duodenum due to limitations with currently available devices. In our first-in-human study (NCT05887635), we demonstrated that radiofrequency vapor ablation (RFVA) with a novel through-the-scope catheter was feasible for duodenal mucosal ablation (DMA), though glycaemic responses were heterogeneous. Major unanswered questions in the field are whether endoscopic small intestinal ablation is durable and whether the procedure can be safely repeated to maximise metabolic benefit over time. To address these limitations, we developed proximal intestinal mucosal ablation (PIMA), which is an extended ablation approach targeting up to 60 cm of proximal small bowel. Through PIMA, we hypothesised that (1) small intestinal ablation with RFVA could be safely repeated, and (2) extending the ablation length would elicit improved glycaemic control even in patients with an inadequate response to prior DMA.

This prospective, single-centre study (NCT06724822) evaluated PIMA, a newly developed small intestinal ablation technique using the second-generation circumferential RFVA catheter. This catheter was specifically designed to safely enable a longer ablation (up to 60 cm) with its mesh-tip and ability to pass through-the-scope. Patients with an inadequate glycaemic response to initial DMA (defined as <0.5% glycated haemoglobin [HbA1c] reduction or HbA1c >7.5% at the end of 6-months follow-up from our initial DMA cohort) were enrolled. All suitable participants underwent PIMA between Sep-2024 and May-2025. Continuous glucose monitoring (CGM) was initiated at baseline, and PIMA was performed via an adult colonoscope (Olympus HQ190L) under general anaesthesia. RFVA was delivered at a dose of 275J with double application to the post-ampullary intestinal mucosa. Post-procedure, patients followed a 14-day modified diet, continued oral glucose-lowering agents (GLA), and received standard lifestyle advice for T2D. Primary innovation endpoints included (1) Procedural feasibility, (2) Safety and tolerability of repeat RFVA, and (3) Early metabolic efficacy captured by HbA1c and CGM parameters over six months.

Nine patients were screened and six (mean age 52.5 years [IQR 48–58], 83.3% female) underwent PIMA at an average of 13.9 months after DMA. During initial DMA, mean HbA1c reductions from screening to 3-, 6-, and 12-months were 0.8% (SD 0.9), 0.4% (SD 0.7), and –0.3% (SD 0.9). At re-treatment, mean baseline HbA1c was 9.0% (SD 0.9), body mass index (BMI) 30.3 kg/m² (SD 4.0), and patients were on a median of one GLA. PIMA was successful in all patients with an average length of treated bowel of 57.5 cm (IQR 51-63) and 51.0 (SD 11.8) ablations. Average procedure time was 74.5 minutes (SD 20.1) and catheter time 62.3 minutes (SD 19.8).  PIMA was well tolerated, with a maximum median visual analogue score for pain of 1 (IQR 0–2) on day 3. Follow-up enteroscopy with biopsies was normal in all patients. Five patients have completed 6-month follow-up with no serious adverse events and one related adverse event (abdominal pain). Median HbA1c reduction was 1.9% (IQR 1.2–2.5; p=0.002) at 3-months and 1.8% (IQR 1.7–2.3; p=0.0006) at 6-months. BMI decreased by 1.3 kg/m² (SD 1.0; p=0.03) and 1.0 kg/m² (SD 1.8; p=0.35), respectively, while time-in-range (3.9-10.0 mmol/L) at these time points was 87% (IQR 75–91) and 74% (IQR 73–81).

PIMA with RFVA appears safe and feasible as a repeat endoscopic therapy for T2D. The greater HbA1c reduction observed with PIMA suggests that ablation length may be a key determinant of metabolic efficacy.