Abstract 1

Title: Oral semaglutide reduces energy intake and improves eating control in type 2 diabetes

Authors: Banshi Saboo MD, PhD, John Blundell PhD , Catherine Gibbons PhD , Søren Tetens Hoff MD PhD, Susanna Lövdahl PhD, Kirsten Dahl PhD, Tine Bækdal MSc

Aim: Oral semaglutide is the first oral glucagon-like peptide-1 analogue to be approved for treatment of type 2 diabetes (T2D). We studied the effect of oral semaglutide on appetite and energy intake in subjects with T2D.

Methods: A double-blind, two-period, crossover trial (NCT02773381) randomised subjects to a 12-week period of oral semaglutide (dose-escalated to steady-state at 14 mg) followed by placebo (pbo) for 12 weeks (with 5–9 weeks’ washout in between) or vice versa. Energy intake was measured at the end of each period over 4-days, during an ad libitum lunch, evening meal and evening snack box. Appetite was rated using a visual analogue scale after overnight fast and during standardised 5-hour breakfast and 8-hour fat-rich breakfast tests. Eating and craving control were assessed using the Control of Eating Questionnaire (CoEQ). Body weight changes were assessed and reported for the first period only due to a possible rebound effect if oral semaglutide was received first.

Results: Fifteen subjects were randomised (13 males, mean age 58.2 years, HbA1c 6.9%, BMI 30.8 kg/m2); two withdrew prematurely. Energy intake was reduced with oral semaglutide vs pbo during the lunch meal, evening meal and evening snack box, leading to total daily energy intake reduction of 5096 kJ (relative difference −38.9%). Palatability ratings after the meal indicated no food aversion with oral semaglutide or pbo. There were no significant differences between oral semaglutide and pbo in overall appetite ratings pre-meal (fasting state) or during a breakfast meal. After the fat-rich breakfast mean postprandial fullness rating was significantly greater with oral semaglutide vs pbo; no other significant differences in appetite ratings were found. CoEQ indicated fewer food cravings and better eating control with oral semaglutide vs pbo. During the first period, mean ± SD body weight loss with oral semaglutide was 2.9 ± 4.3 kg vs 1.2 ± 3.2 kg with pbo, which was attributable to body fat mass loss.

Conclusion: Once-daily oral semaglutide reduced energy intake in subjects with T2D. Appetite was unchanged, control of eating improved and body weight was reduced.

Abstract 2

Title: Similar efficacy and gastrointestinal tolerability versus exposure for oral and subcutaneous semaglutide

Authors: L Srinivas Murthy MD, Christin Løth Hertz MD, PhD, Rune Viig Overgaard PhD, Andrea Navarria MD, Steen Hvass Ingwersen MSc

Aims: The lower bioavailability of oral semaglutide results in more variable plasma concentrations versus subcutaneous administration. Using populations from the SUSTAIN and PIONEER trials, we investigated if oral administration affects the efficacy and tolerability of semaglutide.

Methods: Population pharmacokinetic and exposure–response (ER) analyses were based on average semaglutide steady-state concentrations. Response data were from four trials (SUSTAIN 1, 2, 3, SUSTAIN-Japan) of once-weekly subcutaneous semaglutide 0.5 mg and 1.0 mg over 30 weeks (n=1552), and six trials (PIONEER 1, 2, 3, 5, 8, 9) of once-daily oral semaglutide 3 mg, 7 mg or 14 mg over 26 weeks (n=3003). Graphical and model-based techniques were used to investigate ER relationships for changes from baseline in glycated haemoglobin (HbA1c) and body weight, and proportions of subjects reporting nausea or vomiting.

Results: Pharmacokinetics were dose-proportional and body weight was the main covariate for exposure for both subcutaneous and oral semaglutide. ER analyses showed greater HbA1c and weight reductions, and more subjects reporting nausea/vomiting with increasing exposure. The main covariate for glycaemic effect was baseline HbA1c (larger HbA1c reductions with higher baseline HbA1c values). Exposure range was wider for oral semaglutide than subcutaneous dosing, but there was considerable overlap between oral semaglutide 7 mg and 14 mg and subcutaneous semaglutide 0.5 mg and 1.0 mg, indicating similar exposures across formulations. ER relationships were similar in SUSTAIN and PIONEER.

Conclusions: Similar ER relationships were observed for efficacy and tolerability of semaglutide, regardless of administration route, indicating that greater variability in plasma concentrations with oral semaglutide does not impact response.

Abstract 3

Title: Effects of Semaglutide and Liraglutide on Urinary Albumin To Creatinine Ratio. A Pooled study of SUSTAIN 6 and LEADER.

Authors: Jubin Jacob MD, Vlado Perkovic PhD, Stephen Bain MD, George Bakris MD, John Buse MD, PhD, Thomas Idorn MD, PhD, Kenneth Mahaffey MD, Steven Marso MD, Michael Nauck MD, Richard Pratley MD, Søren Rasmussen PhD, Peter Rossing MD, DMSc, Karen Tornøe MD, PhD, Bernard Zinman MD, Johannes Mann MD

Introduction and Aims: Urinary Albumin to Creatinine Ratio (UACR) is a marker of renal damage and renal disease progression risk; UACR reduction has been shown to correlate with renal protection. This post hoc analysis of pooled data from SUSTAIN 6 and LEADER investigated the effects of glucagon-like peptide-1 (GLP-1) analogues semaglutide and liraglutide vs placebo on UACR.

Methods: Primary outcome in both trials: major adverse cardiovascular events. Nephropathy events and change in renal biochemistry, including UACR, were secondary outcomes. In this analysis (N=11812), patients were stratified by baseline UACR class (normo-: <30 mg/g, micro-: 30–300 mg/g and macroalbuminuria: >300 mg/g). Risk of categorical changes in UACR (time to persistent: progression to micro- or macroalbuminuria, regression to micro- or normoalbuminuria, 30% reduction) and UACR change from baseline to 2 years were analysed by baseline UACR using a Cox regression model (with treatment as a fixed factor and stratified by study) and a mixed model on log-transformed values, respectively.

Results: GLP-1 analogues reduced risk of progression to micro- or macroalbuminuria and increased the likelihood of regression in albuminuria category vs placebo. The likelihood of reaching 30% UACR reduction was greater with the GLP-1 analogues vs placebo (p<0.0001 for all UACR subgroups). UACR increased relatively more from baseline to 2 years for placebo vs GLP-1 analogues for the pooled population and patients with baseline normo-, micro- and macroalbuminuria (24%, 20%, 30% and 21%, respectively).

Conclusions: This pooled analysis shows beneficial effects on UACR for GLP-1 analogues semaglutide and liraglutide vs placebo irrespective of baseline UACR.