OBJECTIVES

To evaluate the efficacy of the oral GLP-1 receptor agonist semaglutide compared to liraglutide (an injectable GLP-1 receptor agonist), empagliflozin (an SGLT-2 inhibitor), sitagliptin (a DPP-4 inhibitor), and no treatment beyond ongoing background antihyperglycemic treatment in adults with T2DM.

METHODS

We performed a systematic literature search in MEDLINE, Embase, and Cochrane Central Register of Controlled Trials through October 2019 to identify trials of oral semaglutide and our active comparators measuring intermediate outcomes (e.g., HbA1c) or key measures of benefit (e.g., cardiovascular outcomes). We also included trials measuring key benefits of injectable semaglutide to inform the effects of semaglutide as a molecule. Data informing the comparison of semaglutide and our comparators on key benefits were synthesized in fixed effects Bayesian NMAs.

RESULTS

We included 12 randomized controlled trials (RCTs)—eight from the Phase III PIONEER program and four cardiovascular outcomes trials (CVOTs) of our active comparators and injectable semaglutide. Results from the PIONEER trials showed oral semaglutide reduced HbA1c and body weight at follow-up (26-78 weeks) more than placebo, sitagliptin, and liraglutide and reduced HbA1c but had similar effects on body weight compared to empagliflozin. Results from our NMA on findings from the CVOTs showed semaglutide (oral and injectable) reduced the risk of major cardiovascular events more than placebo (hazard ratio [HR]: 0.76; 95% credible interval [95% CrI]: 0.63-0.93) and sitagliptin (HR: 0.77; 95% CrI: 0.61-0.96); no significant differences were found between semagltuide and liraglutide or empagliflozin. Additionally, NMA results showed empagliflozin reduced the risk of hospitalization for heart failure more than semaglutide (HR: 0.63; 95% CrI: 0.42-0.95).

CONCLUSIONS

Oral semaglutide produced significant reductions in HbA1c and body weight compared to placebo, sitagliptin, and liraglutide and significant reductions in HbA1c compared to empagliflozin. The effect of oral semaglutide on key benefits could only be assessed by indirect comparisons that are potentially susceptible to effect modification.