Once-weekly TG103, a novel long-acting GLP-1*/Fc fusion protein, demonstrates noninferiority to dulaglutide for HbA1c reduction in Chinese patients with type 2 diabetes (T2D) inadequately controlled with metformin, according to a phase III trial presented at EASD 2026.
This multicentre, open-label, active-controlled, noninferiority trial enrolled 632 Chinese patients (mean age 51.3 years, 42.6 percent female) with T2D (mean baseline HbA1c 8.4 percent) inadequately controlled with metformin.
Participants were randomized 1:1:1 ratio to receive once-weekly subcutaneous injections of TG103 at 7.5 mg (n=210) or 15 mg (n=211) or dulaglutide 1.5 mg (n=211), as add-on therapy to metformin for a 28-week treatment period. At week 24, patients initially receiving TG103 7.5 mg were switched to TG103 15 mg. [EASD 2026, abstract 836]
At week 28, patients treated with either TG103 15 mg or dulaglutide achieved similar reductions in HbA1c, with a 1.38-percent decrease from baseline.
According to Dr Xiaoling Cai from Peking University People's Hospital in Beijing, China, the estimated treatment difference met the prespecified noninferiority margin of 0.4 percent (one-sided p<0.0001), demonstrating that TG103 is noninferior to dulaglutide.
The TG103 15-mg cohort also achieved substantial mean HbA1c reductions as early as week 8 (1.18 percent) and at week 12 (1.39 percent), which were sustained through week 52.
At week 52, the TG103 15-mg group demonstrated a greater reduction in HbA1c from baseline compared with the dulaglutide group (1.31 percent vs 1.25 percent).
Furthermore, a higher proportion of patients in the TG103 15-mg group achieved an HbA1c target of <7 percent than those in the dulaglutide group at weeks 28 (53.9 percent vs 52.2 percent) and 52 (52.2 percent vs 46.8 percent).
Among patients who began treatment with TG103 7.5 mg and were uptitrated to 15 mg, the mean HbA1c decreased by 1.14 percent, and 44.2 percent reached an HbA1c target of <7 percent at week 52. These outcomes were numerically comparable to those observed with dulaglutide.
“Overall, once-weekly TG103 (7.5 or 15 mg) demonstrated sustained glycaemic improvement, with marked efficacy observed at week 8, stabilized at week 12, and maintained at week 52,” Cai said. “These findings suggest that TG103 is a promising add-on treatment option for patients with T2D inadequately controlled with metformin [monotherapy].”
Safety
The most common treatment-emergent adverse events (TEAEs) were gastrointestinal disorders (eg, diarrhoea, nausea, and vomiting), the incidences of which were low and comparable across treatment groups (36.7 percent in the TG103 7.5-mg group, 36.5 percent in the TG103 15-mg group, and 29.4 percent in the dulaglutide group).
Rates of hypoglycaemia and TEAEs leading to treatment discontinuation were also low among all three groups. Most TEAEs were mild or moderate in severity.
“Overall, TG103 showed a favourable safety profile when administered with a simple titration regimen, and no new safety signals were identified,” Cai said.