Has the first treatment of type 2 diabetes mellitus changed?

The aetiology of type 2 diabetes mellitus (DM2) is complex and involves numerous pathophysiological deficits affecting multiple organs. Insulin resistance (IR) in the muscle and liver, as well as β-cell dysfunction, are the main abnormalities responsible for the development and progression of hyperglycaemia.¹ In addition, IR is associated with other metabolic disorders, such as obesity, mainly visceral obesity, hypertension, dyslipidaemia, endothelial dysfunction, inflammation and a hypercoagulable state. Each of these factors, along with IR, has been associated with the development of atherosclerotic cardiovascular disease, which contributes to the increased risk of cardiovascular disease (CVD) in patients living with DM2.²

It is generally agreed that CVD is the leading cause of morbidity and mortality in DM2,³ so numerous studies have made efforts to reduce CVD through achieving better glycaemic control. However, meta-analyses that include these studies show a rather modest reduction in major cardiovascular (CV) events, with a greater benefit in non-fatal myocardial infarction, with no significant effect on stroke.⁴ On the other hand, STENO-2, which used a multifactorial approach but placed less emphasis on improving blood glucose and lowering haemoglobin A1c, showed a significant reduction in both CV events and CV mortality. Therefore, it appears that treatment that is simply focused on hyperglycaemia and not on the concomitant management of the other IR syndrome components is unlikely to produce a sustained decrease in hyperglycaemia and decrease the risk of developing CVD.⁵

The ideal treatment for DM2 is one that is able to reverse or halt at least eight of the pathophysiological deficits described in the Ominous Octet in the year 2009, in which, in addition to IR in the muscle (decreased glucose uptake), liver (increased hepatic glucose production) and progressive insufficiency of insulin secretion by the pancreatic β-cells, a dysfunction of α-cells in the pancreas (increased glucagon secretion), IR of adipose tissue (increased lipolysis), decreased gastrointestinal incretin effect, increased renal reabsorption of glucose and dysfunction of neurotransmitters in the brain (decreased satiety) are added.¹ This suggests that to prevent or delay progressive β-cell insufficiency, treatment should not simply be based on lowering A1c, but may require, along with lifestyle changes, an early combination of drugs to correct the multiple known pathophysiological deficits. In this sense, for about 5 years there has been a paradigm shift in the management of DM2, which recommends applying these modern concepts and guiding treatment according to the pathophysiology of the disease to achieve a benefit (reduction of CVD), as opposed to the classical approach of reaching a target (A1c level).⁶

Over the last two decades, the vast majority of clinical guidelines have recommended metformin as first-line therapy in DM2 due to its efficacy, low cost and well-established safety profile, so the current debate focuses on whether we should replace this treatment in favour of the SGLT2 inhibitors (sodium–glucose 2 cotransporter inhibitors) and GLP-1 analogues (glucagon-like peptide 1 receptor agonists) which have recently demonstrated both CV safety and a reduction in major CV events (including heart failure and renal protection) in randomised clinical trials as part of the regulatory requirements for approval of new antidiabetic agents. To guide treatment decisions in this patient population, it would seem logical to have CV benefit studies with metformin and comparative studies between metformin and these new agents in both primary and secondary prevention.

But what do we know about the CV benefits of metformin? Although metformin is an oral antidiabetic drug that has been widely used for many years as a first-line treatment for DM2, much of the evidence on CV outcomes comes from indirect comparative studies or sub-analyses of clinical trials designed to evaluate other antidiabetic treatments, as well as studies comparing metformin with classic antidiabetic drugs that are no longer in use, which limits the ability to extrapolate these results to routine clinical practice. The CV benefits of metformin were demonstrated in 1998 in the UK Prospective Diabetes Study (UKPDS) substudy, which included a small number of overweight patients with DM2 at low CV risk, but showed CV benefits mainly in myocardial infarction, diabetes-related death and all-cause mortality compared with conventional treatment.⁷ In addition, when patients treated with metformin were compared to those treated with sulphonylureas or insulin, patients on metformin treatment showed less diabetes-related complications, lower overall mortality and a benefit in stroke.⁷

On the other hand, it should be noted that the trials of SGLT2 inhibitors^8–10 and GLP-1 analogues^11–14 that have shown safety and CV benefits (Cardiovascular Outcome Trials [CVOT]) were conducted in patients with DM2 at high CV risk, the majority of whom had established CVD (population other than the UKPDS) and were already receiving standard treatment, which generally included metformin (70%–80%), so that the benefits observed with these drugs were mostly in combination with metformin. To understand how these drugs interact with each other and whether they may have synergistic or antagonistic effects on CV outcomes, some CVOTs with these drugs have conducted post hoc analyses to investigate whether metformin modifies the observed protective effect, with conflicting results. In the post-hoc analysis of the Liraglutide Effect and Action in Diabetes: Evaluation of Cardiovascular Outcome Results (LEADER) study, in which 76% of patients were using metformin, it was observed that patients who were not using metformin at baseline experienced more pronounced CV benefits compared to those who were using metformin.¹⁵ In contrast, the efficacy and safety of once-weekly GLP-1 receptor agonist albiglutide (HARMONY) study, with 73% of patients on metformin treatment, showed a significant CV benefit of albiglutide in patients on metformin treatment at baseline.¹⁶ Among the SGLT2 inhibitors, the hazard ratios in the Empagliflozin, Cardiovascular Outcomes, and Mortality in type 2 diabetes (EMPA-REG) and Canagliflozin and Cardiovascular and Renal Event in type 2 diabetes (CANVAS) studies suggested that the CV benefit observed was mainly due to the CV benefit obtained in patients who were not receiving metformin treatment at baseline.^17,18 However, it is important to remember the limitations of post hoc analyses and that the results should be interpreted with caution. Regarding secondary prevention, given the current evidence and the ethical principles of beneficence and non-maleficence, it will not be possible to conduct trials comparing metformin with placebo in this group of patients. In primary prevention, a recent meta-analysis of 30 CVOTs (n = 225,305) with GLP-1 analogues and SGLT2 inhibitors showed a reduction in the risk of major adverse cardiovascular events (MACE) that was consistent across CVD patients. (HR: 0.87; CI 95%: 0.83–0.92) and without CVD (HR: 0.92; CI 95%: 0.83–1.02).¹⁹ As regards metformin, there is only one ongoing active comparator trial to assess the superiority of dapagliflozin versus metformin as first-line treatment in patients with DM2 < 4 years in reducing CV events (Clinicaltrials.gov: NCT03982381), with an estimated completion date of October 2025. In summary, the evidence for metformin as a cardioprotective agent is less consistent and robust than that for SGLT2 inhibitors and GLP-1 analogues, but what impact have the CVOT results with GLP-1 analogues and SGLT2 inhibitors had on clinical guidelines?

These studies have led to a process of continuous updating of clinical guidelines for the management of DM2. The consensus published in 2018 proposed by the American Diabetes Association (ADA) and the European Association for Study of Diabetes (EASD) laid the groundwork for a significant change in the approach to DM2 by highlighting the importance of CVD prevention and personalisation of treatment to improve clinical outcomes in patients. This approach has continued to evolve in subsequent recommendations from other medical organisations and societies, such as the European Society of Cardiology (ESC) and EASD in 2019 and the ESC in 2023, further reinforcing the paradigm shift in the management of DM2 towards a holistic, patient-centred approach.

The ADA 2024 recommendations for the management of DM2 consolidate the personalised approach and include guidelines stratified by CV risk. These guidelines recommend considering GLP-1 analogues and SGLT2 inhibitors as first-line treatments, regardless of glycaemic control, in patients with established CVD, heart failure and/or chronic kidney disease. In addition, ADA suggests that initial combination therapy should be considered in people with elevated A1c levels, and metformin, SGLT2 inhibitors or GLP-1 analogues are recommended as long-term first-line treatment for patients with DM2 and prior CV events, but without heart failure. As second-line treatment, DPP-4 inhibitors, pioglitazone, acarbose and insulin should be considered. In patients with DM2 who have had previous heart failure, SGLT-2 inhibitors are recommended as long-term first-line treatment. Finally, and for the first time, the ADA Standards of Care 2024 emphasise the importance of weight control as an integral part of DM2 management. The implications of the ADA 2024 recommendations for the use of metformin as first-line therapy in DM2 include a change in the approach to initiation to one that is personalised and tailored, taking into account individual patient characteristics and comorbidities.²⁰

Recognising the CV benefits of GLP-1 analogues and SGLT2 inhibitors has had a major impact on clinical practice in the management of DM2 and CVD in general, with a shift in treatment recommendations to prioritise consideration of these drugs in patients with established CVD or multiple CV risk factors, with a broader focus on primary and secondary prevention of major CV events and on reducing the risk of hospitalisation for heart failure, with a greater emphasis on the CV risk profile of patients when selecting drugs for the treatment of DM2, and an impact on multidisciplinary care with increased collaboration between diabetes specialists, cardiologists and other healthcare professionals in the comprehensive management of patients with DM2 and CVD.

In view of the above, it does not seem justifiable to continue to consider metformin as a universal first-line treatment for all patients with DM2.

Funding

None.

Conflict of interest

Both authors declare that they have no potential conflicts of interest.