Metformin not as first-line therapy anymore in T2DM patients at high/very high CV risk.
The most disturbing proposal for endocrinologists/diabetologists concerns the
the downgrading of metformin use in T2DM patients at CV risk.
The, 2020 ESC guidelines recommended that in drug-naïve T2DM patients with a high/very high CV risk profile (a vast majority of patients with T2DM, as already discussed), a pharmacological monotherapy using an SGLT2i or a GLP-1 RA should be initially referred to a metformin monotherapy while
metformin should be added in the second-line only if HbA1c is not a target.
This recommendation is based upon the recent robust findings of CVOTs with both SGLT2 and GLP-1 RAs in T2DM patients with established CVD. These data support the use of these new agents according to evidence-based medicine criteria, which contrasts with the limited available evidence concerning metformin as a cardioprotective agent.
This innovative view, which most endocrinologists might consider somewhat iconoclastic, is not officially endorsed by the ADA-EASD and the diabetes community yet. It may be challenged for several reasons. First of all, it should be emphasized that a vast majority (about three quarters) of T2DM patients at high risk of CVD enrolled in recent CVOTs were treated with metformin as background therapy (as monotherapy but more often as combination therapy) at inclusion. Thus, the new agent (either an SGLT2i or a GLP-1 RA) was used as add-on therapy.
Even if some exploratory analyses did not show any significant interaction when comparing patients
receiving metformin or not, these subgroup analyses are only exploratory. Of interest, a metaanalysis of CVOTs with dipeptidyl peptidase-4 (DPP-4) inhibitors showed that metformin use might moderate the effect of DPP-4 inhibitors on CV outcomes.
A trend for a reduction in MACEs was observed when a gliptin was added in metformin-treated patients, while a pattern for an increase in MACEs was noticed in T2DM patients not treated with metformin; this difference in overall DPP-4 inhibitor effect between metformin user and nonuser subgroups was statistically significant. To our knowledge, such a metaanalysis of CVOTs separating T2DM patients treated or not with metformin as background therapy has been carried out neither for SGLT2is nor for GLP-1 RAs.
Everybody would agree that the CV protection with metformin was obtained in a limited subgroup of obese
patients with newly-diagnosed T2DM, thus in individuals with a low CVD risk. This population is entirely different from the patients recruited in recent CVOTs, who generally had a long-duration of T2DM (> 10 years) and a high/very high risk of CVD, most of them
having established ASCVD.
Interestingly, however, the CV protection in the group previously treated with metformin persisted after an observational follow-up of 10 years after the end of the UKPDS. Because of these results, the 2019 ESC guidelines consider that metformin should be considered in overweight patients with T2DM without CVD and at moderate CV risk (Class IIa, Level C).
In the absence of a dedicated CVOT with metformin, meta-analyses of available data have been performed to evaluate the potential impact on CVD of this old oral glucose-lowering drug. When RCTs were considered, no reduction in the incidence of CV events could be detected. None of the trials was designed to test this hypothesis, and all of them recruited a vast majority of patients without established ASCVD. A more recent meta-analysis that combined results from some RCTs and mainly retrospective cohort studies
in patients with coronary artery disease showed that metformin reduces both all-cause and CV mortality significantly.
Multiple potential mechanisms have been shown in human studies that support the concept of CV protection with metformin beyond those provided by reduced blood glucose, including improvements in hemostatic function, reduced low-grade inflammation and oxidative stress, and inhibition of crucial steps in the process of atherosclerosis. Observational studies have shown that metformin exerts protective effects even in patients considered at higher risk, such as patients with renal impairment, stable coronary heart disease, and stable HF.
However, also if some evidence suggests that metformin has a protective effect on coronary artery beyond its glucose-lowering effects, future prospective cohort-based studies and RCTs are needed to identify CV 'at-risk' population who may potentially benefit from metformin. A population-based, longitudinal-cohort study using a nationwide US commercial insurance claims database is ongoing. It has the objective to compare the effectiveness of SGLT2is relative to metformin for reducing subsequent CV events in patients with T2DM who are new users of
A specific CVOT comparing the effect of metformin versus placebo is also underway. Because it is not possible anymore to perform such a trial in patients with T2DM for ethical reasons, this trial will recruit patients with impaired glucose tolerance.