Evidence-Based Consensus on Positioning of SGLT2i in Type 2 Diabetes Mellitus in Indians

This article is based on previously conducted studies and does not contain any studies with human participants or animals performed by any of the authors.

SGLT2i are a promising novel class of glucose-lowering agents that interrupt the metabolic pathway without impairing β-cell function and insulin resistance [13]. Phlorizin, the first SGLT2i, was a naturally occurring substance obtained from apple tree bark. However, since it was non-selective and has poor oral bioavailability, work on its development could not continue [14]. Subsequently, selective drugs that inhibit SGLT2 receptor have now been developed.

Dapagliflozin (DAPA), canagliflozin (CANA), empagliflozin (EMPA), and recently ertugliflozin, are the four SGLT2i that have been approved by the European Medicines Agency (EMA) and the Food and Drug Administration (FDA) either as monotherapy or as an add-on to other antidiabetic agents [15]. Ipragliflozin, luseogliflozin, and tofogliflozin are approved only in Japan [16]; sotagliflozin and remogliflozin are in the clinical development stage [15]. Hence, the word ‘SGLT2i’ used in the entire consensus document will represent only DAPA, CANA, and EMPA. Details regarding approval, pharmacokinetic/pharmacodynamic profiles, contradictions, and dose adjustment required for different SGLT2i are described in Table 2 [13, 17‐22].

SGLT2i act on glucose metabolism and homeostasis through regulating the kidney function. The kidneys play a vital role in glucose homeostasis via three processes such as renal gluconeogenesis, glomerular filtration, and glucose reabsorption in the proximal convoluted tubule (PCT) [23]. In a healthy individual, the kidney can filter around 180 g of glucose daily; all of this glucose is reabsorbed in the PCT (~ 90% reabsorption occurs from the S1 segment while ~ 10% reabsorption occurs from the S2/S3 segments of PCT). This reabsorption process in the PCT is generally mediated by the SGLT receptor family (6 members, SGLT 1–6). Details of the SGLT family members are given in Table 3 [13, 15]. Among the SGLT receptor family, SGLT2 receptor is found in the S1 segment and SGLT1 receptor is found in the S2/S3 segment of the PCT. All of the reabsorbed glucose is then transported back into circulation by a passive transport mechanism through facilitative glucose transporters; as a result, the glucose concentration in the plasma increases [13, 24]. The mechanisms of action of SGLT2i and their effect on various physiological parameters are depicted in Fig. 1 [13, 25].

The SGLT2i selectively inhibit the SGLT2 receptors and reduce glucose reabsorption in the PCT, and thereby lower hyperglycaemia by increasing urinary glucose excretion (UGE). Furthermore, SGLT2i therapy results in persistent calorie loss that leads to weight loss, β-cell stress reduction and hyperinsulinaemia, increased insulin sensitivity and rate of insulin secretion. Consecutively, all these mechanisms continuously regulate blood glucose despite insulin resistance or β-cell dysfunction [13, 26‐30]. In addition, SGLT2i may also be effective in the advanced stages of T2DM (when pancreatic β-cell reserves are permanently lost) due to their insulin-independent mechanism of action [31].

SGLT2i are the most recent addition to the T2DM management armamentarium. They have some adjunct advantages such as weight and BP reduction, low risk of hypoglycaemia, and reduction in macrovascular and microvascular events [32]. Moreover, these drugs may also rectify some core defects like improvement in β-cell function and insulin sensitivity in T2DM patients. However, they also have several adverse effects, especially genital tract infections (GTIs), volume depletion, toe amputation and diabetic ketoacidosis [32]. Pertaining to their potential advantages, several guidelines have placed SGLT2i in both monotherapy and combination therapy in the management of T2DM (Table 4) [1, 33‐39].

Body weight, A1C, and systolic BP are the various parameters used to determine the responders/non-responders to SGLT2i treatment in various studies [40‐43]. In a post hoc analysis of two double-blind RCTs, patients in the SGLT2i treatment arm who had a 12-week A1C or systolic BP change lower than the median were considered as an A1C or systolic BP responder; a non-responder was defined as a 12-week A1C or systolic BP change on or above the median [40]. Similarly, in a retrospective study, SGLT2i-treated patients who had ≥ 5% body weight reduction were classified as responders and those with < 5% body weight reduction were classified as non-responders [43]. Moreover, baseline fasting C-peptide level was the important factor influencing change in body weight; it was found to be higher in the responder group than in the non-responder group (3.25 ± 1.07 vs 2.62 ± 1.02 ng/mL, P = 0.023) [43].

Evidence from numerous clinical trials, systematic reviews, and meta-analyses suggests that SGLT2i as monotherapy are effective and safe in controlling glycaemic parameters in patients with T2DM.

A systematic review and network meta-analysis (NMA) including 38 RCTs (23,997 participants with a duration of ≥ 24 weeks) investigated the efficacy of SGLT2i in patients with T2DM. Compared to placebo, all SGLT2i reduced A1C (− 0.6% to − 0.9%), fasting plasma glucose (FPG) (− 1.1 to − 1.9 mmol/l), body weight (− 1.6 to − 2.5 kg), and BP (systolic − 4.9 to − 2.8 mmHg; diastolic − 2.0 to − 1.5 mmHg) [44]. However, more hypoglycaemic episodes were reported with CANA than DAPA. Moreover, GTI episodes (odds ratios [OR] 4–6) are reported to be higher with SGLT2i compared to placebo [44]. Recently, a systematic review and NMA by Johnston et al. reported that SGLT2i as monotherapy significantly improved glycaemic parameters, induced weight loss, and reduced BP [45]. All RCTs included in this study were of ≥ 24 weeks. Compared to placebo or active comparator, SGLT2i reduced A1C (P < 0.001), body weight (P < 0.001), lipid level (P < 0.01), and BP (P < 0.001) in patients with T2DM [45]. All RCTs reported a 4–9% increase in urinary tract infections (UTIs) and GTIs with SGLT2i treatment mainly in women. In line with the evidence, another systematic review and meta-analysis including 39 RCTs (25,468 patients) reported that SGLT2i were associated with better glycaemic control (A1C reduction − 1.01% to − 0.51%), weight loss (− 2.66 to − 1.80 kg), and BP reduction (systolic − 4.77 to − 2.66 mmHg; diastolic − 1.99 to − 1.76 mmHg) compared to placebo [46]. Similarly, a meta-analysis including placebo-controlled trials reported that SGLT2i were associated with an A1C reduction of 0.5%, 0.6% and 0.6% at 12, 24 and 52 weeks, respectively [47]. Furthermore, several systematic reviews and meta-analyses of RCTs reported that SGLT2i as monotherapy were efficacious, reduced body weight and BP, and were well tolerated by T2DM patients [48‐52]. In addition, several RCTs also reported that SGLT2i were associated with clinically meaningful improvements in glycaemic parameters, and a reduction in body weight and BP in patients with baseline A1C ≥ 9% or ≥ 10% [53‐55].

Monotherapy with SGLT2i has been found to be effective and safe compared to other antidiabetic agents. A systematic review and meta-analysis including 25 RCTs reported that SGLT2i were associated with a significantly higher reduction in A1C (weighted mean difference [WMD] 0.13%, P = 0.005) and FPG (WMD 0.80 mmol/L, P < 0.00001) than DPP4 inhibitors (DPP4i) [56]. However, there was no significant difference of hypoglycaemic risk between the treatments (RR 0.99; P = 0.92). Another systematic review and meta-analysis including 39 RCTs (25,468 patients) reported that SGLT2i were similar to metformin and sulphonylureas, but superior to DPP4i (− 0.15%) regarding A1C reduction, and were better than metformin, sulphonylureas, and DPP4i in lowering body weight (− 1.04 kg, − 4.76 kg and − 2.45 kg) and systolic BP reduction (− 5.86 mmHg, − 5.44 mmHg, and − 4.43 mmHg, respectively) [46]. Details of the studies comparing SGLT2i with placebo are given in Table 5 [54, 55, 57‐60].

In addition to glycaemic benefits, SGLT2i have several extra-glycaemic benefits such as body weight reduction, BP reduction, lipid level regulation, CV risk reduction, renoprotective effect, and lowering hypoglycaemic risk. Extra-glycaemic benefits associated with SGLT2i are depicted in Fig. 2 [13, 87].

It is postulated that certain properties of SGLT2i such as osmotic diuresis, mild natriuresis, changes in nitric oxide release, and reductions in arterial stiffness may contribute to BP reduction [61]. A recent systematic review and meta-analysis of RCTs evaluated the efficacy of SGLT2i in 24-h ambulatory BP [113]. SGLT2i significantly reduced 24-h ambulatory systolic and diastolic BP by − 3.76 mmHg (95% CI − 4.23 to − 2.34; I² = 0.99) and − 1.83 mmHg (95% CI − 2.35 to − 1.31; I² = 0.76), respectively [113]. Similarly, several meta-analyses have also advocated that the addition of SGLT2i to metformin was associated with a significant reduction in BP when compared to metformin monotherapy [61‐63].

Compared to other antidiabetic drugs, SGLT2i also provide a significant reduction in BP when used as a second-line agent. In an NMA, SGLT2i significantly reduced systolic BP compared to sulfonylurea (range 4.4–5.64 mmHg), and DPP4i (range 2.26–5.79 mmHg) [63]. In another meta-analysis, the combination of metformin and SGLT2i were favoured in terms of BP reduction compared to metformin and sulfonylurea combination (pooled between-group difference 5.1 mmHg; 95% CI 4.2–6.0 mmHg), and metformin and DPP4i combination (pooled between-group difference 4.1 mmHg; 95% CI 3.6–4.6 mmHg) [61]. In addition, the durability of BP reduction with SGLT2i is similar to that of its other properties, such as glycaemic control and body weight reduction [76], which may boost its use in patients with T2DM. Moreover, evidence advocates that, unlike GLP1 RAs, SGLT2i do not trigger an increase in heart rate, despite reductions in BP [13]. Numerous RCTs also reported a significant systolic BP reduction with SGLT2i as a second-line agent (for details see Table 8).

SGLT2i have been associated with a small increase in high-density lipoprotein cholesterol (HDL-C) as well as low-density lipoprotein cholesterol (LDL-C) and a reduction in triglyceride levels [13, 87]. Meta-analyses of several RCTs reported that treatment with SGLT2i was associated with a small increase in HDL-C level in T2DM patients with no clinically significant changes in LDL-C and triglyceride level [47, 114]. In a double-blind RCT, SGLT2i treatment in T2DM patients who were uncontrolled with metformin reported an increase in HDL-C and a reduction in triglycerides compared with placebo [115]. Furthermore, SGLT2i compared to DPP4i caused a significant increase in HDL-C and LDL-C levels after 24 weeks [116].

Cardiovascular disease (CVD) remains the prominent cause of mortality and morbidity in developing countries [117]. The literature advocates that hypertension and diabetes are more important risk factors in younger Indian women than in men; women have a higher prevalence of hypertension and diabetes compared to men [118]. Moreover, abnormalities in lipid metabolism play a vital role in the development of CAD in young Indians [118]. Evidence suggests that SGLT2i provide CV benefits in patients with T2DM [119, 120]. Various possible routes that might influence the CV benefits of SGLT2i are depicted in Fig. 4 [87]. A systematic review and meta-analysis including several RCTs reported that SGLT2i reduced the risk of major adverse CV events (MACE) (RR 0.84; P = 0.006), CV death (RR 0.63; P < 0.0001), heart failure (RR 0.65; P = 0.002), and death from any cause (RR 0.71; P < 0.0001), compared to a control group. However, an adverse effect on non-fatal stroke (RR 1.30; P = 0.049) was observed. Moreover, these CV benefits of SGLT2i are a class effect rather than an individual effect [120].

Similarly, an NMA compared the effectiveness of SGLT2i against other oral antidiabetic drugs (OADs) (metformin, sulfonylureas, thiazolidinedione, and DPP4i) in preventing CV mortality and morbidity [119]. The relative risk for all-cause mortality, CV mortality, acute coronary syndrome, and myocardial infarction (MI) (only against DPP4i and placebo) were reported to be low for SGLT2i compared to all other OADs [119]. Moreover, another meta-analysis reported that SGLT2i along with their multifactorial benefits were associated with no increased risk for MACE in various subgroup populations (e.g., degree of CV risk, any other risk, and hypoglycaemia) [121]. In line with the evidence, several pooled analyses of RCTs [122, 123], CANVAS trial [124], and EMPA-REG OUTCOME trial [125] advocate that SGLT2i are associated with a lower incidence of CV mortality and morbidity compared to placebo in high-risk CV patients. Moreover, the results of the major ongoing Dapagliflozin Effect on CardiovascuLAR Events (DECLARE-TIMI 58) trial will shed light on the long-term CV effectiveness of SGLT2i in T2DM patients [126]. Thus, SGLT2i may have a better impact on women with CVD as more women are prone to heart disease in India.

In addition, several instances of real-world evidence including the CVD-REAL study [127], UK THIN database study [128], and Swedish national registries [129] have also reported that treatment with SGLT2i was associated with a lower risk of CV event, mortality and morbidity compared with other glucose-lowering agents in T2DM patients. The CVD-REAL study included 309,056 patients from six countries (US, Norway, Denmark, Sweden, Germany and the UK) who were newly initiated on either SGLT2i or other antidiabetic drugs. Compared to other antidiabetic drugs, SGLT2i were associated with lower rates of heart failure hospitalization (HHF) (hazard ratio [HR] 0.61; 95% CI 0.51–0.73; P < 0.001); death (HR 0.49; 95% CI 0.41–0.57; P < 0.001), and HHF or death (HR 0.54; 95% CI 0.48–0.60, P < 0.001) with no significant heterogeneity by country [127].

Like other benefits, SGLT2i also contribute a renoprotective effect in patients with T2DM. SGLT2i acutely reduced estimated glomerular filtration rate (eGFR), followed by progressive recovery and stabilisation of renal function [130]. In recently published RCTs (CANVAS, EMPA-REG OUTCOME), SGLT2i were associated with slower progression of kidney disease, reduced progression of albuminuria, and reduction of eGFR with lower rates of clinically relevant renal adverse events in high-risk patients [124, 131]. Moreover, a pooled analysis of 12 RCTs reported that SGLT2i were not associated with an increased risk of acute renal toxicity or deterioration of renal function in T2DM patients with normal or mildly impaired renal function [132]. Furthermore, compared to glimepiride, SGLT2i reduced the progression of renal disease over 104 weeks, and may confer renoprotective effects independently of their glycaemic effects in patients with T2DM [133].

Evidence suggests that SGLT2i improve insulin sensitivity and β-cell function in the setting of A1C and weight reduction in patients with T2DM [112, 134, 135]. SGLT2i may be beneficial in insulin-resistant patients due to their proven benefits in both improving insulin sensitivity and weight reduction. In a placebo-controlled RCT, SGLT2i significantly (P = 0.0059) improved insulin sensitivity versus placebo [134] and in another RCT [136] they significantly reduced insulin resistance (p < 0.001) in T2DM patients.

In patients with unmet needs of glycaemic control with maximum doses of OADs and insulin, the addition of SGLT2i reduces daily insulin dose requirement and therefore the cost of insulin therapy [137]. Moreover, a UK healthcare system perspective analysis confirmed that SGLT2i + metformin are more cost-effective than DPP4i + metformin [138].

Many patients with T2DM have high serum uric acid levels that might damage the kidney and cause several microvascular complications [139]. Evidence suggests that SGLT2i significantly reduced the levels of uric acid in the plasma [140, 141]. They may also inhibit reabsorption of sodium coupled-uric acid in the PCT, resulting in increased uric acid elimination [13].

A post hoc analysis of 4 RCTs reported that SGLT2i control the level of serum magnesium in hypomagnesaemia patients with T2DM, which may subsequently improve the cardio-metabolic outcome [142].

Hypoglycaemia is a major cause of concern with some antidiabetic agents; however, pertaining to their insulin-independent mechanism of action, SGLT2i produce less hypoglycaemia than other antidiabetic agents [13, 25]. Nonetheless, when SGLT2i are used concomitantly with insulin or insulin secretagogues, the risk of hypoglycaemia increases, and close monitoring of the patient is required [143]. Li et al., in a meta-analysis, found that SGLT2i compared to non-SGLT2i were associated with a significantly lower incidence of hypoglycaemic events when used as an add-on to metformin (OR 0.27; P = 0.02) [73]. SGLT2i were associated with a lower incidence of mild or moderate hypoglycaemia compared to sulfonylureas when used after metformin monotherapy (5% vs 34%) [144]; and a similar incidence of hypoglycaemia was documented when compared to DPP4i (4.2% vs 4.7%) [64].

Increased urinary glucose levels due to the use of SGLT2i results in glycosuria. This subsequently increases the risk of GTIs (balanitis and balanoposthitis in men and vulvovaginitis in women) and to a relatively lesser extent UTIs [13]. In a meta-analysis of RCTs, the incidence of suspected or confirmed GTIs was reported to be significantly (P < 0.00001) higher with SGLT2i (OR 6.41 [95% CI 3.58, 11.45] for men and 5.12 [95% CI 3.48, 7.54] for women) compared to non-SGLT2i agents, both used as an add-on to metformin [73]. However, a meta-analysis of several RCTs found inconsistent results regarding the incidence of UTI between SGLT2i and DPP4i, and between SGLT2i and sulfonylureas, when all added to metformin [13]. Similar results were also reported in a post hoc analysis conducted in Indian T2DM patients [9]. Nevertheless, the incidence of GTIs and UTIs is more common in females than in males, and can be corrected through standard treatment. The South Asian Federation of Endocrine Societies (SAFES) consensus statement advocates that perineal hygiene should be maintained to prevent GTIs and, after 3 months of GTI-free status, patients with T2DM can be prescribed with SGLT2i along with prophylactic antifungal coverage. However, SGLT2i should be avoided in the case of a previous history of upper UTIs, complicated GTIs, or refractory or resistant GTIs [13].

Glycosuria caused by SGLT2i triggers volume depletion by eliminating more fluid from the body through the process of osmotic diuresis [13, 145]. The adverse events associated with volume depletion are BP reduction, dehydration, postural dizziness, orthostatic hypotension, orthostatic intolerance, syncope, and reduction in urine output. A meta-analysis of several RCTs stated that the difference in volume depletion between SGLT2i + metformin and DPP4i + metformin was unclear. However, the study found a small difference between sulfonylureas + metformin and SGLT2i + metformin (pooled OR 1.0; 95% CI 0.6–1.7) [61]. Moreover, volume depletion-related events were reported to be higher with SGLT2i in different RCTs when compared to sulfonylureas (3% vs < 1%) [144], and DPP4i (3.0% vs 0.5%) [64]. In addition, a post hoc analysis of phase III RCTs also reported a similar incidence of volume depletion-related adverse events with SGLT2i in an Indian population vis-à-vis the overall population [84]. Hence, before prescribing SGLT2i, analysis of volume status and thorough monitoring of volume depletion-related adverse events is highly obligatory in T2DM patients.

The risk of bone fracture is increased in people with diabetes compared to their non-diabetic counterparts [146]. Patients with type 1 DM are at a higher risk of osteoporosis, and the risk of hip fracture is generally increased in patients with T2DM [1]. SGLT2i should be used with caution in patients with T2DM who have a high fracture risk [1]. Due to their osmotic diuresis effect, SGLT2i cause volume depletion which subsequently disturbs the electrolyte (serum calcium and phosphate) concentration in the body, and may show a harmful effect on bone health [147]. Moreover, SGLT2i elevate serum phosphate and subsequently enhance the secretion of a parathyroid hormone that may increase bone resorption and the risk of bone fractures [148]. In addition, hyponatraemia caused by SGLT2i might increase osteoporosis and fracture risk [147]. A pooled analysis of 9 RCTs (including the CANVAS trial; n = 4,327) reported that only patients from the CANVAS trial who were continuing CANA had shown a higher risk of fracture when compared to non-CANA patients (4% vs 2.6%) [149]. Moreover, EMPA also revealed few fracture incidents in CKD patients with T2DM in a double-bind RCT [150]. However, an international multi-centre, parallel-group, double-blind RCT reported that DAPA had no effect on markers of bone formation and resorption or bone mineral density in T2DM patients [151]. Similarly, a recent meta-analysis of 20 RCTs (n = 8,286) revealed that there was no increased risk of bone fracture with SGLT2i treatment compared to placebo in patients with T2DM (SGLT2i vs placebo; pooled risk ratio 0.67, 95% CI 0.42–1.07) [152]. Nonetheless, SGLT2i should be used cautiously in older patients, and in those with a prior history/risk of CVD, with lower baseline eGFR and taking more diuretics at baseline [149].

The literature suggests that SGLT2i are associated with improvement in CV prognosis in high CV-risk patients with T2DM [153]. However, recent evidence reported that SGLT2i (CANA) was associated with an unexpectedly increased risk of amputation in patients with T2DM [124]. Moreover, the underlying mechanisms that lead to amputation and whether the risk of amputation is particular to SGLT2i remain unclear [153]. The risk of amputation nearly doubled in patients on CANA compared to placebo in participants in the CANVAS trial (6.3 vs 3.4 participants per 1,000 patient-years, corresponding to an HR of 1.97; 95% CI 1.41–2.75) with T2DM and a high CV risk; 71% of amputations were primarily at the level of the toe or metatarsal [124]. However, a pooled analysis from 15 RCTs (phase I–III) including the EMPA-REG OUTCOME trial reported no difference in the incidence of amputation in EMPA and placebo-treated T2DM patients (both 1.1%) [154]. Data regarding DAPA on peripheral vascular disease are limited.