Introduction: Recent randomized controlled trials (RCTs) have evaluated secondary cardiovascular (CV) safety with novel anti-diabetes drugs (ADDs) in patients with long-standing type 2 diabetes (T2DM). However, real-world evidence of the CV safety of these therapies in primary prevention is scarce.
Aims: To evaluate the risk of developing heart failure, myocardial infarction or stroke (MACE) in patients treated with dipeptidyl peptidase 4 inhibitors (DPP-4), glucagon-like peptide-1 receptor agonists (GLP-1RA) or sodium glucose co-transporter 2 inhibitors (SGLT2), compared to other oral ADDs (OOADs).
Methods: Combining a UK primary care database (THIN) and Centricity Electronic Medical Records of the USA, T2DM patients were identified from the time of first prescription (index date, ID) of DPP4 (n=108,382), GLP-1RA (n=24,831), SGLT2 (n=5,053), OOAD (n=629,803) under the conditions: ID on or after January 2005; age 18-80 yrs at ID; minimum 1 yr follow-up post ID; no insulin treatment ever; no history of CV disease, chronic kidney disease or cancer at ID; minimum 3 mths of therapy exposure; and no overlap of therapies among DPP4, GLP-1RA and SGLT2 groups. Age and follow-up time propensity score matched survival regression model with adjustments for confounders was used for risk analysis.
Results: The mean age/follow-up time in the DPP4, GLP-1RA, SGLT2 and OOAD groups were 60/3.5, 55/3.8, 55/1.7 and 57/3.9 yrs respectively. Compared to OOAD, the rates/1000 person yrs for MACE were significantly lower for SGLT2 (95% CI 8-12) and GLP-1RA (95% CI 11-13), while the rates were similar for DPP4 and OOAD (95% CI 18-19). Compared to OOAD, patients in the DPP4, GLP-1RA and SGLT2 groups had 7% (HR CI 0.90-0.95), 17% (HR CI 0.78-0.88) and 37% (HR CI 0.51-0.78) significantly reduced risk of MACE respectively during overall mean follow-up of 3.8 yrs.
Conclusions: In this primary and ambulatory care based real-world study combining representative data from the UK and USA, we observed the lowest MACE rates and risk in patients prescribed SGLT2 or GLP-1RA compared with other non-insulin ADDs. This suggests that the cardio-protective benefits reported with these drugs in RCTs may apply to the broader population of patients with T2DM and no underlying cardiovascular disease.