**6. Therapeutic implications**

#### **6.1. SGLT2 inhibitors**

explain the quantity of glycogen stored in diabetic kidneys. A major part of the high renal glucose release found in subjects with diabetes may be determined by increased renal glyco‐

Diabetes represents the most common single cause of end-stage renal disease (ESRD) in the United States and Europe. There can be several factors responsible for this, including an increased prevalence of T2DM, longer life spans among patients with diabetes, and better recognition of kidney disease [40]. Comparing with subjects with type 1 diabetes mellitus, only a smaller fraction of those with T2DM develop ESRD, but due to the increased prevalence of T2DM, these individuals represent more than a half of those with diabetes on dialysis. There are numerous variables in progressing to nephropathy, including racial/ethnic variability because Native Americans, Hispanics and African Americans are at much greater risk of

The first clinical signs of nephropathy are represented by low, but abnormal, levels (≥30 mg/day or 20µg/min) of albumin in the urine (previously referred to as microalbuminuria). The detection of albumin in the urine increases the risk of progression to persistent albumi‐ nuria, progressive decline in glomerular filtration rate (GFR), increased blood pressure and cardiovascular morbid-mortality. But because T2DM may be present for many years before diagnosis, a higher proportion of individuals with T2DM have microalbuminuria and overt nephropathy shortly after diagnosis. It is known that without treatment, 20-40% of patients with T2DM and microalbuminuria progress to overt nephropathy. Nevertheless, after 20 years from the onset of nephropathy, only 20% will have progressed to ESRD [40]. The explanation comes from the greater risk among subjects with diabetes and chronic kidney disease of dying

Several clinical trials indicate that the onset and development of diabetic nephropathy may be significantly influenced by numerous interventions including tight glucose control and also use of angiotensin-converting enzyme inhibitors or angiotensin receptor blockers. This is the reason why annual screening for microalbuminuria is critical since it can lead to early diagnosis of nephropathy. Numerous studies, including the well-known Diabetes Control and Compli‐ cations Trial and United Kingdom Prospective Diabetes Study indicated that intensive glycemic control represent a very important step in reducing the risk of developing microal‐

New data from the Action in Diabetes and Vascular Disease: Preterax and Diamicron MR Controlled Evaluation (ADVANCE) trial offers hope regarding the benefic effects of tight glucose control on decreasing the risk of nephropathy [41]. In the ADVANCE trial, after almost 5 years, subjects that were on intensive glycemic control had a 10% relative reduction in the combined outcome of major macrovascular and microvascular events. This was happening mainly because of a 21% relative reduction in the risk of developing nephropathy. The intensive glucose control is also important because it is associated with a 9% reduction in new

developing ESRD than non-Hispanic white subjects with T2DM [40].

from cardiovascular disease than progressing to ESRD.

buminuria and overt nephropathy [40].

genolysis [6].

12 Treatment of Type 2 Diabetes

**5. Diabetic nephropathy**

SGLT2 is highly specific for (several authors consider that it is found only in) the proximal tubules of the kidney, as compared to SGLT1 or GLUT2, therefore it is a preferred target for more specific renal pharmacologic interventions. Thus, the idea of interfering with the activity of the SGLT2 has gained much attention [2].

Inhibition of SGLT2 transporter 'resets' the reabsorption system by lowering the threshold for glycosuria, resulting the correction of the hyperglycemia [1]. Reduction of the blood glucose level can improve insulin resistance in muscle by increasing insulin signaling, GLUT4 and glycogen synthase activity [1].

The history of SGLT2 inhibitors starts in 1835 when phlorizin was found in the root bark of apple tree [42]. Many years after, it was found to be a non-specific SGLT1 and SGLT2 and it could increase glucosuria and reduce blood glucose levels and normalize insulin sensitivity in a pancreatectomized animal model of T2DM [43]. Nevertheless, it could not become a treatment for diabetes due to numerous side effects. Being non-selective and inhibiting SLGT1 at the intestinal brush border, it can cause serious problems regarding the absorption of dietary glucose. Inhibition of SGLT1 can result in glucose–galactose malabsorption and cause diarrhea, events that occur naturally in SGLT1 deficiency [44]. Moreover, in the intestine, phlorizin is poorly absorbed and is rapidly hydrolyzed to phloretin, a substance that blocks GLUT1, leading to disturbance in glucose uptake in several tissues [45]. Highly-specific inhibitors of SGLT2 have subsequently been developed in order to overcome some of these shortcomings.

Ellsworth et al [46] discovered a group of *C*-aryl glycosides that includes dapagliflozin [47] and canagliflozin [48]. They are resistant to degradation produced by β-glucosidase enzymes in the gastrointestinal tract. Moreover, dapagliflozin has a very high sensitivity for SGLT2 compared to SGLT1, blocking renal glucose reabsorption by almost 40–50%. Using this treatment, they can be excreted up to 80–85 g of glucose per day [47]. Clinical trials evaluating the treatment with dapagliflozin, either as monotherapy or in association with metformin or with insulin in subjects with T2DM have demonstrated its efficacy in reducing glucose and HbA1c levels [3]. Pharmacokinetics and bioavailability of dapagliflozin are not influenced by a high-fat meal and there are no reports regarding any interactions with several other drugs used in the treatment of T2DM [3].

Human trials analyzing canagliflozin are more limited than for dapagliflozin. It has been indicated that both drugs have similar therapeutic characteristics [3]. Canagliflozin could induce an important, dose-dependent decrease in the mean renal glucose threshold to approximately 60 mg/dl (3.33 mmol/l) [49].

There are numerous other SGLT2 inhibitors including sergliflozin, remogliflozin, ipraglifozin and empagliflozin. Some of them, such as ipraglifozin and empagliflozin, are being tested in phase III trials and are promising very good results while other compounds have disappointed in clinical trials due to possible side effects (sergliflozin) or to susceptibility to hydrolysis by β-glucosidase enzymes (sergliflozin and remogliflozin) [3].

As already mentioned, patients diagnosed with FRG often gave higher urinary glucose excretion of almost 120 g per day. It remains unclear why treatment with SGLT2 inhibi‐ tors cannot achieve the same levels of glycosuria even when the maximal doses are used. Moreover, SGLT2-null mice can only reabsorb up to a third of the filtered glucose [29], but subjects taking dapagliflozin reabsorb ~50% at the highest doses. Moreover, the nonselec‐ tive inhibitor phlorizin completely blocks reabsorption. One possible explanation may be that SGLT1 has a greater role in the kidney than it was previously imagined [50]. There are some theories that include antisense nucleotide technology to knock out SGLT2 in order to achieve a higher degree of blockade of glucose reabsorption than SGLT2 inhibition. Preliminary data in human subjects with T2DM with moderate or severe renal impair‐ ment indicate that SGLT2 inhibition determines proportionally less glycosuria than in subjects with preserved renal function [51]. These findings confirm that a low GFR in subjects with T2DM is accompanied by a comparable loss of tubular absorptive capacity that represents the anticipated consequence of nephron loss [3].

The approach of lowering hyperglycaemia in T2DM by blocking glucose reabsorption has many attractions. One of them is represented by the activity of SGLT2 inhibitors that is not dependent on pancreatic β-cell function, which deteriorates over time. This is the only class of drugs that present this mechanism of action. Other drugs such as the insulin secretagogues [glinides, sulphonylureas, dipeptidyl peptidase-4 (DPP-4) inhibitors and glucagon-like peptide 1 (GLP-1) agonists] and insulin sensitizers (thiazolidinediones and metformin) depend on insulin secretion. The insulin independence of their action indicates that the risk of hypoglycaemia is very low [6].

As a consequence, the liver can react to the induced glycosuria by increasing glucose release. The mechanism for increased liver glucose excretion is not well understood. The relative small decrease in plasma glucose but also insulin concentrations after massive glycosuria may stimulate endogenous glucose release. Other additional mechanisms are not excluded. Moreover, glucose output is usually not decreased enough to attain and maintain normal glucose values in patients with T2DM treated with SGLT2 inhibitors [52]. Adaptation of glucose metabolism to massive glycosuria needs further investigation.

Osmotic diuresis accompanies glycosuria. It is usually detected an increase in urine output with acute SGLT2 inhibition; while chronic administration of SGLT2 inhibitors is accompanied by an excess urine volume of 200–600 ml per day. As a consequence, haematocrit increases are noted but they are moderate and clinical signs of volume depletion, such as tachycardia and orthostatic hypotension, are rarely met [52].

SGLT2 inhibitors determine glucose and sodium reabsorption blocking and natriuresis also occurs. Changes in serum sodium concentration are not frequent with chronic SGLT2 inhibi‐ tion because at the nephron level, reduced sodium reabsorption in the proximal segment determines the increase of sodium delivery to the juxtaglomerular apparatus, and the inhibi‐ tion of the renin-angiotensin-aldosterone system (RAAS) occurs. In experimental diabetic rats fed a high-salt diet [53], SGLT2 inhibition could prevent blood pressure increase. This effect may be countered by an activation of the RAAS if volume depletion appears as a consequence of excessive diuresis. SGLT2 inhibition in patients with T2DM also determines the reduction of blood pressure levels (by 2–5 mmHg) [52]. The possible explanations may be the enhanced natriuresis and RAAS deactivation [3]. Because most of the individuals with T2DM also present high blood pressure, this effect is of great importance in clinical practice.

induce an important, dose-dependent decrease in the mean renal glucose threshold to

There are numerous other SGLT2 inhibitors including sergliflozin, remogliflozin, ipraglifozin and empagliflozin. Some of them, such as ipraglifozin and empagliflozin, are being tested in phase III trials and are promising very good results while other compounds have disappointed in clinical trials due to possible side effects (sergliflozin) or to susceptibility to hydrolysis by

As already mentioned, patients diagnosed with FRG often gave higher urinary glucose excretion of almost 120 g per day. It remains unclear why treatment with SGLT2 inhibi‐ tors cannot achieve the same levels of glycosuria even when the maximal doses are used. Moreover, SGLT2-null mice can only reabsorb up to a third of the filtered glucose [29], but subjects taking dapagliflozin reabsorb ~50% at the highest doses. Moreover, the nonselec‐ tive inhibitor phlorizin completely blocks reabsorption. One possible explanation may be that SGLT1 has a greater role in the kidney than it was previously imagined [50]. There are some theories that include antisense nucleotide technology to knock out SGLT2 in order to achieve a higher degree of blockade of glucose reabsorption than SGLT2 inhibition. Preliminary data in human subjects with T2DM with moderate or severe renal impair‐ ment indicate that SGLT2 inhibition determines proportionally less glycosuria than in subjects with preserved renal function [51]. These findings confirm that a low GFR in subjects with T2DM is accompanied by a comparable loss of tubular absorptive capacity

The approach of lowering hyperglycaemia in T2DM by blocking glucose reabsorption has many attractions. One of them is represented by the activity of SGLT2 inhibitors that is not dependent on pancreatic β-cell function, which deteriorates over time. This is the only class of drugs that present this mechanism of action. Other drugs such as the insulin secretagogues [glinides, sulphonylureas, dipeptidyl peptidase-4 (DPP-4) inhibitors and glucagon-like peptide 1 (GLP-1) agonists] and insulin sensitizers (thiazolidinediones and metformin) depend on insulin secretion. The insulin independence of their action indicates that the risk of

As a consequence, the liver can react to the induced glycosuria by increasing glucose release. The mechanism for increased liver glucose excretion is not well understood. The relative small decrease in plasma glucose but also insulin concentrations after massive glycosuria may stimulate endogenous glucose release. Other additional mechanisms are not excluded. Moreover, glucose output is usually not decreased enough to attain and maintain normal glucose values in patients with T2DM treated with SGLT2 inhibitors [52]. Adaptation of

Osmotic diuresis accompanies glycosuria. It is usually detected an increase in urine output with acute SGLT2 inhibition; while chronic administration of SGLT2 inhibitors is accompanied by an excess urine volume of 200–600 ml per day. As a consequence, haematocrit increases are noted but they are moderate and clinical signs of volume depletion, such as tachycardia and

approximately 60 mg/dl (3.33 mmol/l) [49].

14 Treatment of Type 2 Diabetes

β-glucosidase enzymes (sergliflozin and remogliflozin) [3].

that represents the anticipated consequence of nephron loss [3].

glucose metabolism to massive glycosuria needs further investigation.

hypoglycaemia is very low [6].

orthostatic hypotension, are rarely met [52].

Several phase III clinical trials of dapagliflozin reported the decrease of serum uric acid concentrations [54]. Sodium and urate are handled together in several physiological circum‐ stances, and also in response to several drugs such as diuretics and antihypertensives. Several sodium-dependent phosphate transporters may also excrete urate into the urine. Therefore, the excretion of urate determined by SGLT2 inhibitors is explained by this mechanism. GLUT9 might represent an alternate explanation. GLUT9 represents an antiporter that exchanges glucose for uric acid; his two isoforms act together to reabsorb glucose from the tubule lumen in exchange for uric acid [55].

Another important effect of SGLT2 inhibition is weight loss. Clinical trials in patients with T2DM have reported a decrease of 2.5–4.0% of body weight [52]. At first, this weight loss is predominantly due to fluid depletion, but soon after that appears the loss of subcutaneous and visceral depots of adipose tissue. This effect is caused by an important caloric loss through the urine. Nevertheless, body weight loss remains constant after several months of treatment [3].

Clinically, the most frequent and undesired effect of SGLT2 inhibitors is represented by high incidence of genitourinary infections. These infections were observed more frequent in women than in men taking SGLT2 inhibitors and tend to occur in susceptible subjects; these include postmenopausal women, history of urinary tract infections or poor hygiene. Interestingly, studies with dapagliflozin in addition to metformin reported a not significant difference in incidence of genitourinary infections between individuals in the placebo and treatment groups [56], while in subjects receiving dapagliflozin in addition to insulin, the difference was significant [57]. This might explain a possibly increased risk of this adverse effect in patients with advanced T2DM (when immune function may be defective) [3].

The incidence of genitourinary infections tends to decrease in time, with long-term treatment, when the adaptation to the treatment is installed or exclusion of susceptible individuals over time appears. More important, infections of the upper urinary tract, that tend to be more severe than those of the lower urinary tract, are not frequent, although the reported patient exposure is presently too limited to rule out this adverse event [56].

Another reported event was a very small, but consistent, increase in PTH levels (<2.0 ng/l) together with increased plasma phosphate concentration. The increased PTH might indicate a mild form of secondary hyperparathyroidism but the available studies so far offer very few data regarding the long-term effects of SGLT2 inhibitors on bone metabolism, making room for other clinical studies on this important issue.

There have been reports regarding several cases of bladder cancer and breast cancer, in subjects with T2DM receiving treatment with dapagliflozin [58]. Trials with large numbers of patients with different SGLT2 inhibitors are required to assess any associated increased risks of breast or bladder cancer [3].

Theoretical safety and tolerability concerns also include impairment in renal function [54]. Although, until now, there are no data indicating that the SLGT2 inhibitors would determine or be responsible for deterioration of renal function, the few clinical studies investigating these drugs have relatively short duration (6-12 months). Moreover, several authors are speculating that SGLT2 inhibitors may play an important role in preventing diabetic nephropathy. First, improved glycaemic control decreases the risk of diabetic nephropathy and other diabetic complications [40]. Second, by increasing the quantity of sodium in the juxtaglomerular apparatus, the use of SGLT2 inhibitors may determine a protective effect on the kidney, independently of glucose decreased.

In T2DM, the high quantity of glucose and sodium absorbed in the proximal tubule reduces the quantity of sodium to be delivered to the juxtaglomerular apparatus. Thus, the glomerulotubular feedback reflex is activated; this leads to high renal plasma flow, increased intraglomerular pressure and elevated GFR. All these processes can induce normal salt delivery to the juxtaglomerular apparatus; however this can result in increased intra-glomerular pressure. All these alterations in renal hemodynamic lead to renal hypertrophy and eventually the result is represented by diabetic nephropathy [59]. SGLT2 inhibitors may prevent diabetic nephrop‐ athy by inhibiting the glomerulo-tubular feedback reflex and, therefore increasing sodium delivery to the distal nephron [1]. Nevertheless, this therapy is contraindicated in patients with estimated GFR (eGFR) <45 mL/min/1.73 m2 and must be used at lower doses at eGFRs of 45-60 mL/min/1.73 m2 [60]. New clinical trials are expected to evaluate the efficacy and safety of SGLT2 inhibitors.

The pathogenesis of type 2 diabetes combines numerous defects in many tissues. Therefore, there is no single antidiabetic drug that can compensate all the metabolic disturbances, and a good treatment for diabetes will require the use of multiple drugs in combination. Having a unique pharmacokinetic and a special mechanism of action, the SGLT2 inhibitors can be used not only as monotherapy [61] but also in combination with currently available antidiabetic agents [62,63].
