Nature communicationsMar 25, 2026

Combined GLP-1 and GLP-2 treatment in advanced type 2 diabetes: lab tests and early human trial

GLP-1 Therapies Weekly Brief ↗PubMed ↗DOI ↗OA ↗

Updated Jun 28, 2026

Abstract

Essence

The bispecific GLP-1/GLP-2 agonist PG-102 improved glycemic control without worsening catabolic weight loss in advanced diabetic mice and had an acceptable early safety profile in people.

Evidence

Preclinical db/db mouse comparisons against semaglutide and tirzepatide suggested beta-cell preservation and enhanced glucose uptake, and a single-center randomized double-blind placebo-controlled phase 1 multiple-ascending-dose trial in 24 overweight adults found mostly mild to moderate gastrointestinal adverse events with no serious adverse events or discontinuations.

Caveat

The human trial was small, short, and designed for safety in overweight adults rather than efficacy in type 2 diabetes, so the clinical glucose-lowering benefit remains unproven.

Simplified

PG-102 is a potency-optimized bispecific Fc fusion protein targeting GLP-1 and GLP-2 receptors. In db/db mouse models of advanced diabetes characterized by uncontrolled hyperglycemia and catabolic weight loss, PG-102 achieved superior and sustained glycemic control compared with semaglutide or tirzepatide while preserving body weight, uncoupling glycemic control from catabolic weight loss. Mechanistic studies indicated that these effects were not driven by acute insulinotropic activity, but by β-cell preservation and enhanced glucose uptake. These benefits required dual GLP-1R/GLP-2R engagement, as PG-102 outperformed monospecific Fc fusion agonists or their combination and promoted coordinated receptor trafficking with delayed internalization. We conducted a randomized, double-blind, placebo-controlled multiple ascending dose phase 1 study at a single center in the Republic of Korea in adults with overweight (BMI 25-30 kg/m²). Twenty-four participants were randomized within three weekly dose cohorts (15 mg, 30 mg, and 30/60 mg; n = 6 per cohort) in a 6:2 ratio to receive PG-102 (n = 18) or placebo (n = 6). All randomized participants (PG-102 n = 18; placebo n = 6) received at least one dose and were included in the safety analysis. Safety and tolerability were predefined as the primary endpoint and assessed by treatment-emergent adverse events (TEAEs). TEAEs occurred in 5/6 (83.3%) participants in each PG-102 cohort and 4/6 (66.7%) in placebo; treatment-related AEs occurred in 5/6 (83.3%) and 3/6 (50.0%), respectively. No serious adverse events or discontinuations due to adverse events occurred. Gastrointestinal events were mild to moderate. These findings support bispecific GLP-1/GLP-2 agonism as a mechanistically distinct incretin strategy in advanced T2D. ClinicalTrials.gov identifier: NCT06309667.