Once-daily oral HRS-7535, an investigational nonpeptide glucagon-like peptide-1 receptor agonist that does not require fasting administration or injection, reduced hemoglobin A1c levels compared with placebo among patients with type 2 diabetes inadequately controlled with metformin in a 16-week phase 2 randomized controlled trial.
The double-blind trial, published in JAMA Network Open, included 194 patients aged 18 to 75 years at 44 centers in China. Patients were randomly assigned to receive once-daily oral HRS-7535 at doses of 15 mg, 30 mg, 60 mg, or 90 mg, or placebo, all added to stable metformin therapy. Eligible patients had hemoglobin A1c levels of 7.5% to 11.0%, body mass index of 19.0 to 40.0, body weight of at least 50 kg, and stable body weight for at least 3 months prior to screening.
HRS-7535 is a glucagon-like peptide-1 receptor agonist (GLP-1 RA). The study researchers noted that oral semaglutide is the only approved oral GLP-1 RA but requires strict fasting administration to optimize absorption. HRS-7535 is a small-molecule nonpeptide agent that does not require fasting administration, which could have practical implications for adherence if efficacy and safety are confirmed in larger trials.
The 60-mg and 90-mg groups used prespecified dose-escalation regimens. Both groups started at 30 mg. The 60-mg group increased the dose by 15 mg weekly for 2 weeks, and the 90-mg group increased the dose by 30 mg every 4 weeks until reaching the target dose. As a result, patients in the 60-mg and 90-mg groups were exposed to target doses for only 8 to 12 weeks, a limitation for interpreting both efficacy and safety.
The primary end point was change in hemoglobin A1c level from baseline to week 16. Patients had a mean baseline hemoglobin A1c level of 8.5%, mean age of 52 years, and mean body mass index of 26.7. Overall, 177 of 194 patients completed the treatment period.
At week 16, least-squares mean hemoglobin A1c reductions were 1.19 percentage points with HRS-7535 15 mg, 1.59 percentage points with 30 mg, 1.82 percentage points with 60 mg, and 1.64 percentage points with 90 mg compared with 0.25 percentage points with placebo. Placebo-adjusted reductions ranged from 0.94 percentage points with the 15-mg dose to 1.57 percentage points with the 60-mg dose.
The proportion of patients who achieved hemoglobin A1c levels below 7.0% ranged from 49% to 63% across HRS-7535 dose groups compared with 15% among patients who received placebo. Reductions in fasting plasma glucose, 2-hour postprandial plasma glucose, and 7-point self-monitored blood glucose levels were also greater with HRS-7535 than with placebo.
The researchers noted that hemoglobin A1c reduction with the 60-mg dose was similar to that with the 90-mg dose, suggesting the lower dose may provide robust glycemic control. That finding may be relevant for phase 3 dose selection, although this phase 2 trial was not designed to establish an optimal benefit-risk dose.
Rescue antihyperglycemic therapy was required in 3% of patients receiving 15 mg, 8% receiving 30 mg, no patients receiving 60 mg, and 3% receiving 90 mg compared with 31% of patients who received placebo.
The primary analysis used a mixed-effects model for repeated measures and a hypothetical estimand, meaning data collected after premature treatment discontinuation or initiation of rescue antihyperglycemic therapy were excluded. The researchers also reported a treatment-policy sensitivity analysis using copy-reference multiple imputation, in which missing data were imputed using placebo reference patterns, as well as an ad hoc analysis excluding patients without measurable postbaseline pharmacokinetic exposure. Findings were consistent across these analyses. No formal adjustment for multiplicity was applied, so secondary end point and multiple dose comparisons should be interpreted cautiously.
Weight reduction was modest in this population. Patients receiving HRS-7535 90 mg had a 2.63% reduction in body weight at week 16 compared with 1.30% among patients who received placebo, although the between-group difference was not statistically significant. Lower HRS-7535 doses produced smaller changes that were generally similar to placebo. Changes in waist circumference were small across groups, with numerically greater reductions at higher HRS-7535 doses.
In an accompanying invited commentary, Robert M. Cohen, MD, of the Division of Endocrinology, Diabetes & Metabolism, Department of Medicine at the University of Cincinnati College of Medicine and Cincinnati Veterans Affairs Medical Center in Ohio, cautioned against assuming that results from this study would directly predict outcomes in other populations or trial designs.
"When reading any diabetes clinical trial, but in particular those from countries you may not be familiar with, it is important to review the anthropometrics and demographics with a question of how this might alter expected outcomes or introduce a bias from what may be seen with a different participant selection, participant source, or study design," Dr. Cohen wrote.
Dr. Cohen noted that patients with type 2 diabetes in major East Asian countries generally have lower body mass index than patients in Western countries and may develop type 2 diabetes at lower body mass index. The mean body mass index in this trial was about 27, below what many US physicians commonly see among patients with type 2 diabetes. As a result, the modest weight effects observed in the study may not predict the magnitude of weight change in higher-body-mass-index Western populations, and comparisons with other GLP-1 RA or dual agonist trials should be made cautiously. He also noted that lower baseline hemoglobin A1c levels leave less room for improvement and that 16 weeks may be too short for hemoglobin A1c to fully reflect glucose lowering that takes time to reach steady state — meaning the observed reductions may not fully capture the longer-term glycemic response. Separately, whether the effect is durable over time remains unknown.
"Suffice it to say, the study by Guo et al demonstrates statistically significant effects of HRS-7535 in this population with this study design," Dr. Cohen wrote.
Adverse events occurred in 72% to 85% of patients receiving HRS-7535 and 72% of patients receiving placebo. Gastrointestinal adverse events were the most common treatment-related events and were predominantly mild or moderate. Across HRS-7535 dose groups, nausea occurred in about 8% to 34% of patients, vomiting in about 3% to 16%, and diarrhea in about 5% to 15%. Most gastrointestinal events occurred during treatment initiation or dose escalation. The researchers noted that daily gastrointestinal adverse event rates were higher in the 60-mg group than in the 90-mg group during days 14 to 56, suggesting that titration every 2 weeks may not have allowed sufficient time for gastrointestinal tolerability to develop — a finding that could be relevant for future dose-escalation strategies.
Documented hypoglycemia occurred in 9 patients receiving HRS-7535 and no patients receiving placebo. All events were level 1; no level 2 or level 3 hypoglycemia was reported.
Prespecified adverse events of special interest occurred in 5 patients receiving HRS-7535 and no patients receiving placebo; all were lipase elevations greater than 3 times the upper limit of normal. Dose-related increases in both amylase and lipase levels were observed with HRS-7535. No pancreatitis, abnormal blood calcitonin findings meeting criteria for adverse events of special interest, liver function–related adverse events of special interest, severe hypoglycemia, or alanine aminotransferase or aspartate aminotransferase elevations greater than 3 times the upper limit of normal were reported.
Heart rate and pulse findings were consistent with known GLP-1 RA class effects. At week 16, mean heart rate increased by 3.0 to 5.4 beats per minute in the 30-mg, 60-mg, and 90-mg groups compared with 0.9 beats per minute with placebo. Sinus tachycardia occurred in 10% of patients receiving 15 mg, 3% receiving 30 mg, 8% receiving 60 mg, and 10% receiving 90 mg compared with 3% of patients receiving placebo. One patient in the 30-mg group died of cerebral infarction, which investigators considered unrelated to study treatment.
The researchers noted several limitations. The double-blind treatment period lasted 16 weeks, with limited exposure to the target 60-mg and 90-mg doses. Longer-term studies are needed to assess durability of glycemic control, weight effects, cardiovascular and kidney outcomes, and safety. The study also enrolled adults in China with relatively low baseline body weight and body mass index compared with populations in many global type 2 diabetes trials, which may limit generalizability.
The phase 3 OUTSTAND program includes two 52-week trials evaluating HRS-7535 among patients with inadequate glycemic control despite lifestyle intervention or other glucose-lowering therapies.
The study was supported by Jiangsu Hengrui Pharmaceuticals. The funder was involved in the design and conduct of the study, study management, data collection, analysis, interpretation, and report writing, but had no role in manuscript preparation, review, approval, or the decision to submit the manuscript for publication. Medical writing support was provided by Lin Dong, PhD, of Jiangsu Hengrui Pharmaceuticals. Zi Ye, PhD, Yimei Xu, MD, and Pan Liu, MS, reported employment by Jiangsu Hengrui Pharmaceuticals during the study. No other study disclosures were reported.
Invited commentary author Dr. Cohen reported receiving grant support from Dexcom Inc paid to his institution and holding equity in Abbott Laboratories outside the submitted work. No other disclosures were reported.
Source: JAMA Network Open
