The high prevalence of type 2 diabetes mellitus (T2DM) is global in nature and continues to increase dramatically, particularly in low- and middle-income countries [1]. Present estimates indicate that approximately 463 million people worldwide have diabetes and that number is expected to increase to a staggering digit of 700 million by 2045 [1]. Prevalence, incidence and heterogeneity of the disease itself among those diagnosed, and progression of the underlying complications vary greatly by geographical location and ethnic groups due to differences in genetic predisposition, cultural factors, socioeconomic differences, and lifestyle factors such as dietary habits and physical activity [2-5]. Most individuals with diabetes are classified as having T2DM [1] but present with their own unique combination of insulin resistance and impaired insulin secretion, i.e., β-cell failure [6], while the relative contribution of these two components varies greatly from patient to patient and over time [5,6]. Additionally, the rate of disease progression varies between individuals and cannot be predicted accurately at the time of diagnosis. Even if multiple clinical, laboratory, and genetic markers have been reported to be associated with faster progression after diagnosis [7], it is not possible to account for the effect of additional behavioural patterns affecting compliance with, and thereby impact of the dietary and exercise regimens; important predictors for disease progression [7].

The aim of this review is to discuss the advancement in the recent scientific and clinical evidence regarding (1) the established knowledge base regarding the unmet clinical needs and the factors contributing to lack of optimised outcomes; (2) the key learnings from the Vildagliptin Efficacy in combination with metfoRmin For earlY treatment of T2DM (VERIFY) study exploring the long-term benefits of a novel strategy of early combination; and (3) yet to be explored future aspects of more optimised clinical management of individuals newly diagnosed with T2DM (Fig. 1).

Pathogenesis of T2DM is complex, and stems from multiple metabolic imbalances, abnormalities and defects affecting multiple organ systems [8,9]. Therefore, addressing these drivers of progressive disease by initiating pharmacological monotherapy, as is suggested by most treatment algorithms, may not be sufficient. Therefore, early initiation of combination therapy has been proposed by many as a more holistic approach to delay the deterioration of glycaemic control with possible favourable effect on preservation of β-cell function early on in the disease [10]. Traditionally, recommendations and treatment guidelines have focused on achieving and maintaining glycosylated hemoglobin (HbA1c) targets, while there has been a growing interest in assessing other glycaemic parameters as indicators of response, namely glycaemic fluctuation over 24hours, including fasting and postprandial glucose values, as well as time spent in extreme hypoglycaemia and hyperglycaemia [11]. In addition to the potential contribution of glycaemic variability as a predictor of outcomes and progression of diabetic complications [12], minimising glucose variability through appropriate selection of components of drug therapy could lead to superior, individualised disease management, and provide better quality of life for patients with diabetes [11,13]. However, regardless of the parameters used to determine whether adequate glycaemic control is being achieved and maintained, we should aim to introduce a favourable, disease-modifying effect and/or critically assess the ability of the chosen regimen to halt the progression of the disease and/or its underlying complications.

Failure of the pancreatic β-cell to compensate for the progressively increasing insulin resistance has been recognized as the hallmark of T2DM, and the rate of deterioration of glycaemic control parallels that of declining β-cell function and/or mass [14,15]. While prevention of T2DM seems mechanistically feasible, albeit challenging [16,17], attempts to preserve the remaining β-cell functionality are best undertaken as early along the disease continuum as possible, even if there is limited evidence for identification of the optimal strategy for β-cell preservation. Some of the best evidence available relates to targeting normoglycemia early, by intensive insulin therapy at the time of diagnosis (Fig. 2) [18]. A shorter time-interval between diagnosis and intensive insulin therapy seem to predict a higher probability for achieving remission up to 2 years [19]. However, clinical implementation of such a regimen might be shadowed by the complexity of a multiple-injection regimen at disease onset, albeit for a short interval. The conceptual success has only been shown in individuals with sustained hyperglycaemia due to delayed diagnosis while dosing of low-dose basal insulin in individuals with newly diagnosed diabetes and low baseline, or pre-diabetes with complications, did not demonstrate impact on β-cell functionality or disease progression [20]. Additionally, there is no clinical guidance on the most optimal second-line therapy for maintenance of achieved remission after intensive insulin therapy.

Several orally administered drug classes have demonstrated a possible effect on β-cell preservation. The benefits of thiazolidinediones (TZD) were established when rosiglitazone demonstrated glycaemic durability superior to metformin and glyburide [21] while improvement in β-cell function was demonstrated to be the predominant underlying pathophysiological mechanism in inducing glycaemic remission from impaired glucose tolerance (IGT) to normal glucose tolerance with pioglitazone, and thus potentially the entire class [22].

Incretin-based therapies have demonstrated the ability to preserve β-cell function in animal models, although the clinical implication of these preliminary data in human studies is still not fully understood [23-25]. Long-term administration of glucagon-like peptide-1 receptor agonists (GLP-1 RAs) compared with insulin glargine or placebo after 4 weeks of intensive insulin therapy resulted in improved or superior β-cell function; this effect was either partially maintained [26,27], or lost after cessation of the incretin therapy [28]. Among dipeptidyl peptidase-4 inhibitors (DPP-4i) a positive effect on β-cell function has been shown in numerous clinical studies: treatment with saxagliptin versus placebo resulted in halted reduction in homeostasis model assessment of β-cell function (HOMA-β) in patients with long-term diabetes and concomitant cardiovascular (CV) disease [29]. An effect that was most pronounced in treatment-naïve patients or those receiving metformin monotherapy. Similar results have been reported in an analysis of all phase III studies of linagliptin on HOMA-β versus comparators [30] and with vildagliptin versus placebo, even though this improved effect on β-cell function was not sustained after a 4-week washout period [31]. A large meta-analysis of incretinbased therapies on β-cell and insulin resistance confirmed that these therapies not only demonstrate an increase in HOMA-β and fasting C-peptide levels, but also achieve a reduction in insulin resistance (measured as HOMA-IR) and fasting plasma glucose (FPG) in comparison with placebo [25].

Numerous studies have also shown that glucosuric agents such as sodium-glucose cotransporter-2 inhibitors (SGLT-2i), which exert their effect by inhibiting renal glucose reabsorption and producing glucosuria [32], improve insulin secretion by ameliorating the glucotoxic effect of chronic, sustained hyperglycaemia in individuals with T2DM [33-35]. In patients treated with dapagliflozin, insulin secretion significantly increased in a euglycaemic insulin clamp during a standardised oral glucose tolerance test, while there was no amelioration of β-cell function among patients receiving placebo [36]. The administration of empagliflozin was also shown to improve pancreatic β-cell function, as measured by the insulin secretion/ insulin resistance index during a hyperglycaemic clamp [33] and the improvement in insulin secretion as induced by another SGLT-2i, ipragliflozin, in Japanese patients was demonstrated to have a more persistent effect even after the cessation of drug administration [34]. In the absence of the SGLT-2 receptors on the islet cells, the exact mechanism beyond removal of the glucotoxicity underlying this beneficial effect of SGLT2is on insulin secretion remains unknown.

All the evidence indicates that alleviation of sustained hyperglycaemia, regardless of the chosen approach, is the most important contributor to β-cell preservation. Possibly, targeting normoglycaemia early at disease onset, or even at the prediabetes stage, and maintaining normoglycaemia, safely, by implementing a more aggressive, early therapeutic strategy may change the natural history of disease [9]. The Restoring Insulin Secretion (RISE) Adult Medication Study assessed the benefit of metformin, GLP-1 RAs, or insulin glargine on β-cell function in early diabetes. It included a cohort of 10 to 19-year-old children and adolescents [35], and measured β-cell function in individuals with IGT or newly diagnosed T2DM. Again, the results demonstrated that β-cell function improved through active treatment, but the favourable effect was lost after the treatment was stopped. The study results were also indicative of an unmet need for alternative approaches in the youth cohort due to a more aggressive form of the disease in youth [37]. In summary, all of the studies to date have reported on differences in outcomes based on specific individual used, not the timing or combination of therapies.

In T2DM, the treatment goals are linked to prevention or delay of chronic complications and maintenance of the patients’ quality of life [38]. Therefore, treating to target HbA1c levels is fundamental to achieving disease control, i.e., reducing time spent in hyperglycaemia and preventing chronic, micro- and macrovascular complications and premature mortality [39-41]. Previously, due to lack of sufficient evidence on the use of early combination therapy, stepwise treatment intensification has been the standard approach to achieve glycaemic control as recommended by the American Diabetes Association (ADA)/ European Association for the Study of Diabetes (EASD) consensus on treatment algorithm [42]. The standard of care for first-line drug therapy has been metformin monotherapy with lifestyle measures followed by step-wise treatment intensification to achieve stringent HbA1c targets (mostly; HbA1c <7.0%). However, the recent inclusion of CV status of the patient has added new nuances to the recommendations. The European Society of Cardiology (ESC) guideline developed in collaboration with EASD suggested that instead of metformin, SGLT-2i, and GLP-1 RAs should be administered as first-line monotherapy in patients with atherosclerotic CV disease or with high CV risk [43]. It is recommended that this monotherapy approach, independent of choice of therapy, should be intensified with an additional drug if monotherapy does not achieve or maintain the individualised HbA1c target after 3 months [44]. The preferred third-line treatment approach includes insulin initiation or a triple combination of oral blood glucose lowering drugs [42,44].

Despite these updated recommendations, metformin is still the most common initial monotherapy and it will celebrate its 100th birthday in 2022 without a detailed mode of action ever being described. It does, however, have many desirable pharmacodynamic effects: it lowers both basal and postprandial plasma glucose, suppresses excessive hepatic glucose production through a reduction in gluconeogenesis increases glucose uptake and insulin signalling, decreases fatty acid and triglyceride synthesis and increases fatty acid β-oxidation [45-47]. It may also increase glucose utilisation in peripheral tissues [45,46], and reduce food intake and intestinal glucose absorption, and has been reported to increase the concentration of both incretin hormones (enhance DPP-4 activity and intact GLP-1 concentrations) and even stimulate GLP-1 synthesis or inhibiting DPP-4 activity (through improved metabolic control) [48]. In other words, metformin, despite being a widely used drug with an impressively complex, dose- and duration dependent effect on glycaemia [49], it is still ‘only’ a single component and usually a starting point for a series of failing therapy options. We should, instead, consider the favourable effects of a strategic approach, as there are challenges with the metformin, or indeed any, monotherapy approach: non-response on an individual and population level [50], on top of often delayed diagnosis [51].

Our understanding of the genetic basis of T2DM is expanding with nearly 80 susceptibility loci identified, and attempts have been made to correlate phenotype of glucose disposition with particular genotypes [52,53]. Accordingly, the individual response to medications differs from one patient to another. Clinical variables such as disease duration, age, baseline HbA1c, and body mass index have been found to predict clinical responses to metformin, and sulfonylureas (SUs), basal insulin, or incretin-based therapies, whereby the clinical responses to each medication was differently affected by these variables [54-56]. In addition, the field of pharmacogenomics in diabetes is expanding with identification of susceptibility loci to individual drug classes [57]. This means that when most patients present with an HbA1c value more than 1.5% to 2% units of HbA1c above the diagnostic criteria and treatment target level for newly diagnosed patients at diagnosis, the current recommendations for monotherapy limit realistic opportunities to sufficiently lower glycaemia to achieve target levels at onset of diagnosis [10,58]. Furthermore, once diagnosed, many patients with T2DM will likely experience one treatment failure after another, often due to delayed intervention. In a substantial number of newly diagnosed patients, their treating physicians fail to intensify therapy within 6 or even 12 months of failure of metformin monotherapy [59] and less than half of the individuals with self-reported diabetes are likely to achieve the goal of HbA1c below 7.0% [60] while more than a third will have an HbA1c value above 7.0% [61].

The inability to achieve target glycaemic control is multifactorial and includes procrastination by the patient, physician and society [62]. These factors include the difficulty for the person living with diabetes to understand importance of glycaemic control, leading to unrealistic expectations that the condition will resolve spontaneously, rather than progressively worsening over decades [63,64]. The ‘wait and see’ approach of the sequential treatment intensification also creates uncertainties for the undecided physician. Clinical inertia—a failure to initiate or intensify treatment in a timely manner, but also, failure to establish appropriate targets and to escalate treatment to achieve treatment goals [65]—is common and may expose patients to long-term elevation of glycaemia that can negatively impact their prognosis [64]. In reality, median time to treatment intensification, unfortunately, exceeds 12 months even after persistent HbA1c above 7.0% for 6 months on metformin monotherapy [66]. Insufficient adherence to “goal setting” and lack of adequate patient-physician communication were proposed to be contributors to clinical inertia according to results of an online patient-physician survey [64]. Though physicians well appreciated the risks associated with poor diabetes control, only 25% of people living with T2DM reported they were worried about developing diabetes complications, whereas the rest were unconcerned or believed the risk was remote [64]. Clinical inertia seems to accelerate rather than diminish [67,68] despite the recent focus on it on guidelines [42].

Initiation of combination therapy, or earlier intensification of the failing monotherapy might also be linked to an unfounded fear of potentially increased risk of (potential synergistic) adverse events of multiple regimens [69]. As a result, patients may be exposed to prolonged, avoidable hyperglycaemia prior to treatment intensification, which can lead to increased risk of microvascular and macrovascular complications [62,70]. With the onset of complications, therapy intensification can become complex with increased risk of adverse drug reactions [71]. In these circumstances, providers should consider first whether aggressive treatment with the goal of attaining lower HbA1c values is warranted. Nonetheless, the vast majority of newly diagnosed patients with T2DM do not fall into this category and will require, as well as better tolerate, a more aggressive treatment approach. Individuals who are diagnosed with T2DM today might live longer with their disease than they lived without it. Due to earlier disease onset, elevated number of people with young onset diabetes, we have an increasing number of people living with sub-optimally managed T2DM for several decades [72].

A proactive approach to treatment intensification was proposed more than 10 years ago as an attempt to minimise time spent in a constant state of hyperglycaemia [73]. During this time, the ADA/EASD/American Association of Clinical Endocrinologists (AACE) treatment guidelines started to support initiation of metformin therapy at diagnosis parallel to lifestyle modification [74,75]. Initial combination therapy was to be considered in individuals with an HbA1c above 7.5% [74] or 9.0% [75] and later the associations added a text indicating that ‘Initial combination therapy may be considered in patients presenting with HbA1c levels more than 17 mmol/mol (1.5%) above their target’ [38]. It follows that an initial combination regimen was already recommended for use in patients with diagnostic HbA1c value above 7.5% (Fig. 2).

Thus, the debate started to centre on the need for combination therapy at higher initial levels of HbA1c, and whether at least for those levels indicative of delayed diagnosis or more aggressively progressing early disease, initiating combination therapy based on pathophysiology is preferred over adding stepwise therapy based on demanding glycaemic goals. However, it was less clear is whether initial combination therapy would carry greater benefits than a sequential approach in timely diagnosed, mostly asymptomatic individuals presenting with HbA1c levels close to or at the diagnostic criteria (HbA1c 6.5%). Given this dilemma, a study was needed to verify if an early combination strategy provided more favourable outcomes and if the achieved effect on glycaemia could be more durable versus the recommended sequential therapy [76]. A theoretical presentation of the available evidence, with the diagnostic level of glycaemia contrasted against the known or assumed duration of T2DM, both from double-blinded, randomised clinical studies but also, real-world evidence and conceptual attempts to rejuvenate the β-cell functionality in newly diagnosed patients, is displayed in Fig. 2.

The multinational VERIFY study provides the most compelling evidence regarding opportunities for treatment optimisation in early T2DM [77]. VERIFY enrolled patients who were diagnosed with T2DM within 24 months; HbA1c levels between 6.5% and 7.5%; no prior diabetes treatment or maximum of 4 weeks of metformin. The study was reflective of the increasingly heterogeneous worldwide T2DM population and enrolled a cohort with ethnic, gender, regional, and socioeconomic diversity—40% of the recruited patient population were non-Caucasian and over half were female [78]. The choice of exploring the early combination of a DPP-4i with metformin was supported by the complementary mode of action of its two components; glucose-dependent β-cell stimulation by vildagliptin [79] and concomitant insulin sensitisation by metformin [80]; as well as the established favourable safety profile of both drugs [81]. Both components have also been reported to reduce hepatic glucose production [10,80,82]; a key component and driver of early, progressive T2DM.

In the VERIFY study, half of the individuals received the early combination immediately after a short, individualised metformin up-titration phase while those patients who initially received metformin monotherapy received the same combination therapy with metformin and vildagliptin following their first treatment failure (HbA1c twice, consecutively, 13 weeks apart above 7.0%). Apart from halving the risk of glycaemic progression in 5 years following treatment initiation and delaying the time to primary treatment failure with more than 2 years versus the sequential approach, early combination therapy also reduced the risk of time to secondary treatment failure (when all patients were receiving the combination) by 26% [77]. These results provide an additional, confirmatory piece of evidence to the array of studies previously hinting that only the early normalisation of glycaemia can help to attenuate diabetes progression (Fig. 1) [83,84].

As the clinical response of one individual to a particular medication can differ from the response of another individual, one of the most frequently asked questions is whether all patients will benefit from the early combination strategy. The median time to initial treatment failure by strategy in the VERIFY study demonstrated the overall lack of benefit with initial metformin monotherapy, in all systematically studied global populations and predefined sub-groups [85]. Additionally, the time to second failure analysis confirmed that better late than never does not apply in diabetes: later introduction of the combination strategy did not provide long-term durability after sequential introduction of a second-line agent. In this way, the full efficacy of the added-on medication, vildagliptin, did not fully materialise versus how efficaciously it induced a rapid reduction in hyperglycaemia but also, provided a remission-like, long-term durability over 5 years when dosed as early combination instead. It also seems impossible to predict on the individual level who will respond to metformin monotherapy; the HbA1c levels were significantly lower throughout 5 years in those receiving the combination strategy versus those who could be considered as ‘metformin responders.’ Therefore there were very few indicators of superior benefits with initial monotherapy strategy in the VERIFY study. All the clinical benefits, including the intriguing, statistically non-significant trend for improved CV outcomes, were observed among those with early combination despite some individuals having demonstrated relatively good glycaemic control on initial metformin monotherapy [77].

Two synergistic drugs dosed earlier is better [86] and provides a more pronounced and rapid decline in glycaemia than one [77] but the clinically compelling outcomes of the VERIFY study extend beyond the scientific findings. That is, that early combination treatment in newly diagnosed patients with earlystage T2DM provides increased glycaemic durability of target HbA1c levels. VERIFY also demonstrated that achieving these targets required less frequent interventions, and thereby investment of clinical resources, for treatment intensification over time.

Similar results to those of VERIFY have been observed in a meta-analysis of 15 randomised clinical trials evaluating initial, metformin-based, dual combination therapies versus metformin monotherapy in treatment-naïve patients with T2DM (Fig. 2) [87]. Compared with metformin, the combination regimens with TZDs, insulin secretagogues, DPP-4is or SGLT-2is provided statistically significant, albeit absolutely modest reductions in HbA1c (−0.43%; 95% confidence interval [CI], −0.56 to −0.30), increases in attainment of glycaemic goal, HbA1c below 7.0% (relative risk, 1.40; 95% CI, 1.33 to 1.48) and reductions in FPG [87]. The population included had a mean age range of 48.4 to 62.7years; HbA1c between 7.2% to 9.9% and T2DM duration of 1.6 to 4.1years, with median follow‐up of 6months. Similarly, Cai et al. [88] demonstrated in a meta-analysis of 36 studies with most widely used therapeutic regimens, that compared with metformin monotherapy, initial metformin-based combination therapies exhibited significant reductions in HbA1c (Fig. 2). In this analysis, most of the combination therapies induced a similar risk of hypoglycaemia, with the exception of combinations of SU/glinide and metformin or combinations of TZD and metformin. Compared with DPP-4i monotherapy, initial DPP-4i-based dual combination therapies showed a significant decrease in HbA1c and a similar risk of hypoglycaemia [88].

Addressing complex pathophysiological mechanisms may also require acceleration from dual to triple combination: AbdulGhani et al. [89] evaluated initial combination therapy with metformin, TZD (pioglitazone) and a GLP-1 RA (exenatide) compared with add-on therapy to metformin with an SU and then insulin (Fig. 2). In this study, as opposed to the initial triple therapy, the conventional approach began with metformin up-titration, followed by addition of the SU and ultimately, insulin glargine. Interestingly, the time-related change in glycaemia was equally effectively managed with a rather aggressive but efficaciously administered sequential therapy during the first 6 months [89]. However, over time the HbA1c control with the initial triple therapy was superior, due to the components used and the complementary mechanisms of action while the role of the timing of addition of the components was unclear. In other words, this study demonstrated that appropriate, timely intensification of sequential titration of therapy may also lead to achievement of glycaemic targets. Therefore, starting with initial combination therapy may not necessarily fully resolve the issue of clinical inertia still prevalent outside clinical trials [90], but, rather, just postpone it to the next stage of required treatment intensification.

Nevertheless, the benefits of early, timely, therapy intensification are additionally supported by a number of real-world studies (Fig. 2). In patients newly diagnosed with T2DM, early intensification of diabetes therapy (timely addition of anti-diabetes medicine, changes in metformin regimen, or nutritionist/dietician referral) due to metformin failure resulted in a more rapid attainment of HbA1c goals compared with late or no treatment intensification [59]. Real‐world observational studies have shown the benefit of intensifying therapy with basal insulin earlier in the treatment paradigm: in an analysis of a US managed‐care claims database, earlier insulin initiation in T2DM patients on one or two oral anti-diabetic (OAD) medications at baseline resulted in the greatest reductions in HbA1c, compared with patients on three or more OADs [91] In another retrospective database analysis, significantly more patients on one OAD at baseline (early insulin initiation group) achieved their glycaemic target, compared with patients receiving two, or three or more OADs before insulin initiation; in addition, decrease in HbA1c was significantly greater in patients on one OAD at baseline [92].

Mechanistically, the effects of agents other than vildagliptin in combination with metformin, or other oral regimens, based on rational consideration of mode of action will provide further insight into pathophysiological processes in early T2DM. Nevertheless, generalisability of the VERIFY results remains to be demonstrated for all the other combinations because numerous studies addressing the same scientific question have contributed to pronounced differences observed in both the efficacy, and safety and tolerability profile of components within the same drug class [93].

The particular drugs and combinations assessed need to be studied comparatively evaluating their short-term effectiveness in HbA1c reduction as well as their long-term benefits on β-cell function, durability, and reduction of diabetic complications [94]. Overall, the risk-to-benefit ratio of each drug must be carefully weighed, because patients with newly diagnosed diabetes make up a vulnerable population. Initial combination therapy entails prescription of multiple drug classes with their accordant known and unknown risks in previous exposed or less studied populations. Prescribing incompatible medications may result in unduly exposing many individuals to harm, as the long-term safety of anti-diabetes medications in newlydiagnosed patients has yet to be determined for all (due to extensive focus on CV outcomes trials and later stage patients, mostly in the Western world).

One of the paradoxes of diabetes management is that early drug treatment is assumed to be considerably less complex than late intervention. Yet the foundation of care is set at the time of diagnosis and more explicit clinical guidance is needed when introducing a new treatment paradigm, a strategy. Using two drugs with complementary mechanisms of action targeting different physiological abnormalities will be required in most patients to achieve adequate glycaemic targets, prevent β-cell failure and achieve durability versus what can be achieved with the current standard-of-care.

Implementing the current, updated treatment recommendations in practice will require a change, from identification and diagnosis to patient follow-up. Recommendations and tangible protocols will need to facilitate early T2DM detection and diagnosis: one of the past and future challenges of successful replication of the VERIFY study results in practice will be related to underdiagnosis and delayed diagnosis of T2DM worldwide [1]. In VERIFY, only half of the screened individuals met the glycaemic inclusion criteria reflective of optimal diagnostic range and glycaemic levels for disease-modifying effect for preservation of the β-cells. The lowest individual HbA1c values at screening (after a locally confirmed diagnosis of T2DM) were within the normoglycaemic range (<5.5%) while the highest value was beyond the upper limit of the central laboratory (>18.5%) [123]. There were significant geographical variations, with the absolute screening failure rate indicating over 80% being outside the desired range [123]. This may be due to regional differences in aptitude of healthcare systems to identify patients early; less structured and limited access to public healthcare infrastructure, socioeconomic inequality in combination with less accurate early identification and late or missed diagnoses, confirm the existence of clinical diagnosis inertia and the unmet need for early diagnosis and intervention to curb the epidemic of diabetes. Despite the VERIFY results indicating early benefits (and other evidence indicative of opposite prospects for treatment success for those with delayed diagnosis), the guidelines will need to provide more explicit guidance for early disease detection.

Under recent clinical guidance, and not least based on the results from the VERIFY study, the role of metformin as the cornerstone of early diabetes care requires careful reassessment, among many new questions of importance (Fig. 1). The latest revisions of the guidelines already indicated that eventually UKPDS is the only study which has demonstrated signs of CV risk reduction with metformin while all the other evidence indicates presence of a summarised, neutral effect [124]. It is indeed prime time to explore outcomes of other early combinations without the current widely used component of metformin.

Nevertheless, even if outcomes have gained in importance, one should not conceptually mix outcomes with the clinical determinants of safety and their role in success of any chosen regimen(s). The VERIFY results include an excellent tolerability profile and lack of discontinuations due to adverse events, despite achievement of low, pre-diabetic glycaemic levels [77]. We also learned that every agent added will not necessarily result in additional drug-induced adverse effects in all patients even if there are historical data indicating that adverse drug reactions with older drugs were more prevalent with two-drug combinations as opposed to monotherapy [125]. Conservatively, sequential use and titration of therapy, if applicable, will allow physicians to judge the adverse effect profile of each agent and address it appropriately, whereas use of well-tolerated modern anti-diabetic agents should remove this clinical challenge and, often unconscious fear without foundation. Use of sequential therapy will add to the complexity of the regimen and potentially negatively influence compliance without a favourable effect on safety and tolerability. No evidence exists that would suggest that earlier combination of therapy as opposed to sequential titration would result in greater adherence or compliance of patients. The two different treatment arms in the VERIFY study had similar adherence to the therapeutic regimens in a double-blind setting, and randomised clinical trials can seldom address adherence in an optimal way. Claims of initial combination therapy masking an excellent response to one element of the combination, or a poor response to another, has been rejected as a hypothesis, at least in the VERIFY study [77]. Additionally, there is no evidence to date that a change of timing of the same agents as early combination as opposed to sequential titration of therapy will have any favourable benefits on body weight. Adding an SU earlier in the course may increase weight gain earlier, but if aggressive titration is achieved over a few months, the impact of weight gain and its timing will be minimal. With use of SUs or e.g., insulins, hypoglycaemia would be an issue regardless of timing. Instead, unselective use of early combination therapy may ignore the heterogeneous presentation of the disease at onset and result in pathophysiologically incorrect treatment of individuals from the start—identification of individual drivers of the disease and the best combination components will induce durability and be the first step towards precision medicine. It is hardly justifiable to believe that newly diagnosed patients will maintain adequate glycaemic control with monotherapy or even just by lifestyle modification for a prolonged period of time. One would be ignoring the progressive nature of the disease and the current evidence demonstrating durability for the first time, with potential to reduce the clinical inertia associated with starting and optimising glycaemic control.

It could be argued that the narrow baseline glycaemia in the VERIFY study was not reflective of clinical reality but in the future, physicians may also face challenges in justifying continued treatment when impressive glycaemic control on or close to normoglycaemic levels have been achieved and diabetes ‘remission’ has been coined as a term. As T2DM is a progressive disease with underlying mechanisms for complications and early loss of β-cell function and insulin sensitivity, cessation of successful treatment and ‘drug-holidays’ justified by financial claims will jeopardise the unique response to early combination treatment. Therefore, we must shift from a concept of sequential treatment failures to new concepts of long-term diabetes remission, disease modifying effect and conversion of glycaemia back to pre-diabetic levels.

There are several studies estimating the cost-effectiveness of the new anti-diabetes agents, not strategic approaches, and it is difficult to quantify the cost-effectiveness of softer outcomes such as fewer hypoglycaemic events or improved quality of life [126]. Conceptually, initial combination therapy may induce wider clinical use of newer and more costly anti-diabetes agents. However, its cost-effectiveness must be individually assessed against the known, currently progressive disease management strategies, and generated costs, which are often covered out of pocket by the individual instead of the society potentially reimbursing for the drug. The cost of these novel agents remains a significant hurdle to their clinical use in many regions of the world. The emerging evidence for long-term efficacy and safety out of the VERIFY study now provides preliminary data on the impact of excess time spent in avoidable hyperglycaemia using validated pharmaco-economic models. Increasing awareness of the benefits of sustained and continuous disease control is essential to achieve alignment between physicians and policy makers. The incoming Glycemia Reduction Approaches in Diabetes: a Comparative Effectiveness (GRADE) study will additionally respond to many of these questions [94].

The recently reported VERIFY study results have provided a wealth of new evidence confirming long-term glycaemic durability and tolerability of the studied early combination in the management of newly diagnosed, treatment-naïve patients worldwide. These results have also contributed to changes in clinical treatment guidelines and standards of care. Clinical implementation and individualised treatment decisions based on VERIFY results might face barriers beyond the existing scientific evidence. Appropriate, timely intensification of sequential titration of therapy may also lead to achievement of glycaemic targets while there is a wealth of data indicating this approach is challenging and less successful in clinical practise due to extensive presence of clinical inertia. While there is no single answer to the question if the early combination genuinely and always is better, in all individuals, these unanswered questions, as highlighted in the current diabetes guidelines, should not become obstacles when changing the practice of medicine, one patient at a time, when carefully considering each individual case.