OR22-3 Closed-Loop Glucagon Pump: A Novel and Effective Strategy for Post-Bariatric Hypoglycemia

Current therapies for post-bariatric hypoglycemia (PBH) are incompletely effective. We previously reported feasibility of an open-loop glucagon system for this challenging syndrome. In this study, patients with PBH were enrolled in a double-masked, placebo-controlled, crossover trial to determine the efficacy of a closed-loop mini-dose glucagon delivery system to reduce severe hypoglycemia after a mixed meal. A novel hypoglycemia detection & mitigation algorithm was embedded in the Artificial Pancreas System connected to a continuous glucose monitor (CGM, Dexcom) driving a patch infusion pump (Insulet) filled with study drug (Xeris liquid glucagon or vehicle). After screening and enrollment, CGM were placed; participants returned after an overnight fast for the 1(st) of 2 study visits. A liquid mixed meal (Ensure Compact: 64 g CHO, 18 g protein, 236 mL) was consumed, and sensor/plasma glucose were measured serially. The system autonomously delivered up to 2 doses of study drug (300/150 mcg of glucagon or equal volume vehicle) if triggered by the hypoglycemia mitigation algorithm. If plasma glucose fell to <55 mg/dL or neuroglycopenia occurred, rescue IV dextrose was given per protocol. During a 2(nd) study visit, the protocol was repeated, with pump filled with the other study drug. Twelve participants (11F/1M, age 52+2, postoperative duration 8+1 years, mean+SEM) completed all study visits. In an additional 3 participants, the mixed meal did not trigger either alarm or hypoglycemia, so study drug was not administered, and a second visit was not conducted. For the 12 participants receiving glucagon vs. vehicle during 2 study visits, predictive hypoglycemia alerts prompted automated drug delivery at mean 94+6 vs. 89+5 (p=0.41) minutes post meal, when sensor glucose was 114+7 vs. 121+5 mg/dL (p=0.39). Four participants did not require rescue during either visit; 1 participant required rescue during both visits. Seven participants required rescue glucose after vehicle but not after glucagon (p=0.0082). Similarly, 5 participants had severe hypoglycemia (plasma glucose <55 mg/dL) after vehicle but not after glucagon (p=0.03). Nadir plasma glucose was higher in study visits with glucagon vs. vehicle delivery (67.4±2.7 vs. 58.5±1.9 mg/dL, p=0.004). Glucagon levels were not elevated at time of alert (14.6±1.4 pg/mL) but rose after glucagon delivery (1231±187 vs. vehicle 16 ±1.4 pg/mL at 30 minutes, p = 0.001). No rebound hyperglycemia occurred. Emesis occurred before study drug delivery in 2 visits. Transient pain at infusion site was reported during both glucagon (n=11 of 12) and vehicle (n=10 of 12) study visits. No other adverse advents were observed. Our data demonstrate that a CGM-guided glucagon closed-loop system can detect imminent hypoglycemia and deliver mini-dose glucagon, yielding improvements in post-meal glucose and reducing severe hypoglycemia in patients with PBH.