Date: 2012-06-09
Type of information:
phase: 3
Announcement: results of Harmony 6 and Harmony 7 studies
Company: GSK (UK)
Product: albiglutide
Action
mechanism: Albiglutide is an investigational biological, injectable form of human GLP-1 and is not currently approved anywhere in the world. GLP-1 is a peptide that acts throughout the body to help maintain normal blood-sugar levels and to control appetite. Normally, GLP-1 levels rise during a meal to help the body utilise and control the elevation in blood sugar levels. However, GLP-1 is rapidly degraded, resulting in its short duration of action. In people with type 2 diabetes, GLP-1 secretion in response to a meal is reduced. Albiglutide is an investigational medicine which fuses human GLP-1 to human albumin. It is designed to have the potential to extended duration of action and allow for weekly or less-frequent injections.
GSK is developing albiglutide as a once-weekly subcutaneous injection. All of the medication is contained within a proprietary injector pen for simple reconstitution and subcutaneous administration using a fine gauge needle by the patient.
Disease: type 2 diabetes
Therapeutic area: Metabolic diseases
Country:
Trial
details: Harmony 6 was a 52-week randomised open label multicentre study among patients with Type 2 diabetes who were inadequately controlled on intermediate or long-acting insulin. Albiglutide (30mg with optional titration to 50mg) was compared to Lispro, each administered in combination with long-acting insulin glargine.
Harmony 7 was a 32-week head-to-head study comparing albiglutide (50mg) to once-daily liraglutide (1.8mg).
Latest
news: GSK has announced that detailed findings from a Phase III study comparing the investigational glucagon-like peptide-1 (GLP-1) receptor agonist albiglutide to prandial insulin (Lispro), were presented at the American Diabetes Association Meeting (ADA) in Philadelphia, USA, and show that the effect is maintained out to 52 weeks.
Harmony 6 results:
As previously disclosed, at 26 weeks, albiglutide showed clinically and statistically significant reductions in HbA1c from baseline and non-inferiority versus Lispro (reduction of 0.82% vs 0.66%; p<0.0001 for non-inferiority) which met the primary endpoint of non-inferiority. The effect on HbA1c was maintained on continued treatment at 52 weeks (reduction from baseline of 1.01% vs 0.85%; p=NS [i]). In addition, t he proportion of patients achieving a clinically meaningful HbA1c target level (ADA goal of <7.0%) by week 52 was 45% in the albiglutide arm compared to 30% (p=NS for treatment difference) in the Lispro arm.
Data presented at the ADA also show that weight changes from baseline observed at 26 weeks for patients in the albiglutide arm (-0.73kg loss) vs the Lispro arm (+0.81kg gain) were sustained at 52 weeks (-0.96kg vs +1.66kg; p<0.0001 for treatment difference). Fasting Harmony 6, a 52-week randomised open label multicentre study among patients with Type 2 diabetes who were inadequately controlled on intermediate or long-acting insulin, albiglutide (30mg with optional titration to 50mg) was compared to Lispro, each administered in combination with long-acting insulin glargine. As previously disclosed, at 26 weeks, albiglutide showed clinically and statistically significant reductions in HbA1c from baseline and non-inferiority versus Lispro (reduction of 0.82% vs 0.66%; p<0.0001 for non-inferiority) which met the primary endpoint of non-inferiority.
The effect on HbA1c was maintained on continued treatment at 52 weeks (reduction from baseline of 1.01% vs 0.85%; p=NS [i]). In addition, t he proportion of patients achieving a clinically meaningful HbA1c target level (ADA goal of <7.0%) by week 52 was 45% in the albiglutide arm compared to 30% (p=NS for treatment difference) in the Lispro arm.
Data presented also show that weight changes from baseline observed at 26 weeks for patients in the albiglutide arm (-0.73kg loss) vs the Lispro arm (+0.81kg gain) were sustained at 52 weeks (-0.96kg vs +1.66kg; p<0.0001 for treatment difference). Fasting plasma glucose (FPG) decreased from baseline in both study arms throughout the 52 week period (-27mg/dL in the albiglutide arm compared to -16mg/dL in the Lispro arm; p =0.0281 for treatment difference).
Adverse events were higher in the albiglutide arm than the Lispro arm over the 52-week treatment period, with the most common being nausea (13% vs 2%), diarrhoea (14% vs 6%), and injection site reactions (10% vs 5%). No severe hypoglycaemic events (low blood glucose levels) were reported for albiglutide (1% for Lispro) and minor hypoglycaemia (?70mg/dL) was higher in the Lispro arm compared to the albiglutide arm (39% vs 23%).
Harmony 7 results:
Results of Harmony 7, a 32-week head-to-head study comparing albiglutide (50mg) to once-daily liraglutide (1.8mg), were also presented as a poster at the ADA. As previously disclosed, albiglutide demonstrated a statistically significant reduction in HbA1c from baseline (-0.78%; p<0.0001) but did not meet the pre-specified primary endpoint of non-inferiority to liraglutide (-0.99%; 95% CI: 0.08-0.34%; p=NS). The proportion of patients achieving a clinically meaningful HbA1c target level of <7.0 at week 32 was higher in the liraglutide arm (52%) compared to the albiglutide arm (42%; p=0.0023 for treatment difference).
Weight decreased from baseline with both albiglutide and liraglutide, however a greater difference was seen in the liraglutide arm (-2.2kg) vs the albiglutide arm (-0.64kg). FPG change from baseline at week 32 was -22 mg/dL for albiglutide and -30 mg/dL for liraglutide (p=0.0050 for treatment difference).
Overall, adverse events occurred at similar rates in both arms of the study but gastro-intestinal (GI) events occurred more frequently in the liraglutide arm than the albiglutide arm; nausea (29% vs 10%) and diarrhoea (14% vs 15%) were the most common GI adverse events. In the albiglutide treatment group, the incidence of GI events was generally stable from week 1 throughout the study period, with no apparent effect of uptitration at week 6.
Injection site reactions, generally of mild intensity, occurred with greater frequency in the albiglutide arm (13%) vs the liraglutide arm (5%). No severe hypoglycaemic events were reported and minor hypoglycaemia was higher in the liraglutide group compared to the albiglutide (17% vs 13%).
