FEATURE ARTICLE

Apple Trees to Sodium Glucose Co-Transporter Inhibitors: A Review of SGLT2 Inhibition

John R. White, Jr., PA, PharmD

or centuries, the kidney has to SGLT2 Inhibitors proposed and utilized in the early been considered primarily J.R.R. Tolkien once wrote that “few 1900s.2 Fan organ of elimination and can foresee whither their road will During the next several decades, a regulator of salt and ion balance. lead them, till they come to its end.” phlorizin continued to be used in Although it was once erroneously Certainly, this is the case in the story trials evaluating renal physiology. thought to be the structural cause of sodium glucose co-transporter 2 By the early 1970s, research with of diabetes and in later years was (SGLT2) inhibitors. phlorizin revealed the location In 1835, French chemists isolated ignored as a regulator of glucose (proximal tubule brush border) of a substance, phlorizin, from the bark the active-transport system respon- homeostasis, it is now recognized as of trees.1 The compound was sible for glucose reabsorption and an important player in the arena of bitter in flavor and reminded them that phlorizin had a much higher glucose regulation. Today, we have a of similar extracts from the cinchona affinity for these transporters than better understanding of the physiol- and willow tree and for a time was did glucose.3 ogy of glucose transport via specific referred to as the “ from the There was a resurgence of inter- carriers such as the sodium glucose bark of apple trees.” est in phlorizin in the late 1980s co-transporters (SGLTs). Parallel to Although the best thinking at the to early 1990s concurrent with the these developments, a natural com- time concluded that phlorizin was a characterization of SGLTs and pound, phlorizin, was isolated in the reasonable candidate for the treat- the realization of a potential novel early 1800s and for decades played an ment of fevers, infectious disease, mechanism for reducing hypergly- important role in diabetes and renal and malaria, within 50 years, it was cemia.4,5 Animal studies carried out physiology research. Eventually, at discovered that high doses of phlori- in 90% pancreatectomized diabetic the nexus of these aforementioned zin caused glucosuria. Ultimately, it rats demonstrated that phlorizin- discoveries, it was recognized that was determined that chronic admin- induced glucosuria normalized the effect of a phlorizin-like com- istration of phlorizin in the canine fasting and postprandial glucose lev- pound on renal glucose transporters model produced many of the same els and reversed insulin resistance. symptoms as observed in human might offer a novel mechanism for Additionally, phlorizin administra- diabetes (glucosuria, polyuria, and the management of hyperglycemia. tion was associated with reversal weight loss). Thus, the phlorizin- This has led to the development of of first- and second-phase insulin induced diabetes animal model was secretory defects in this model.5 several potentially effective therapeu- Phlorizin has not, however, been tic modalities for the management of I N B R I E F pursued as a medication for the hyperglycemia. treatment of diabetes in humans for This article reviews the his- This article reviews the history of several reasons. The compound is phlorizin, the role of the kidney in tory of phlorizin, the role of the poorly absorbed from the gastro- glucose regulation, and the modu- kidney in glucose regulation, and intestinal tract and inhibits both lation of that regulatory system the modulation of that regulatory SGLT1 (primarily found in the via pharmacological means. It also system via pharmacological means. gastrointestinal tract) and SGLT2. offers a discussion of the results of It also offers a discussion of the Analogs of phlorizin have been clinical trials of the most salient results of clinical trials of the most developed that circumvent these two sodium glucose co-transporter salient SGLT inhibitor to date, problems. Currently, several of these inhibitor to date: . dapagliflozin. analogs, including dapagliflozin and

CLINICAL DIABETES • Volume 28, Number 1, 2010 5 FEATURE ARTICLE

, appear to be poten- glucose concentration needed to tubule. Although SGLT1 is the key tially pharmacologically viable.5,6 exceed the tubular glucose threshold transporter for glucose absorption in Additionally, at least one company is ranged between 130 and 300 mg/dl.10 the gastrointestinal tract, its impact developing anti-sense technology to Additionally, the study reported in the kidney is less significant, reduce the expression of SGLT2 and a relationship between age and an accounting for only ~ 10% of glucose increase in threshold levels. reabsorption in the nephrons. one molecule, ISIS-SGLT2Rx, is cur- rently in phase 1 clinical trials.7 This phenomenon raises an The inhibition of this transporter interesting question. Is blocking the has been of some interest pharma- The Kidney and Glucose Homeostasis reabsorption of glucose via SGLT2 cologically because blocking the As mentioned above, the kidney inhibitors going to provide a greater transporter theoretically attenuates historically has not been thought of as effect in younger patients than in gastrointestinal glucose absorption one of the major organs responsible elderly patients? and might offer a method to induce for glucose homeostasis. However, It should be noted that the weight loss or reduce postprandial we now understand the kidney plays glucose threshold, at which some hyperglycemia. In fact, at least one a major role in glucose homeostasis glucose is lost, is reached before compound is being studied for this in two ways: 1) gluconeogenesis and reaching the transport maximum potential effect. 2) glomerular filtration and reab- (Tm). The Tm is the concentration Conversely, SGLT2 is a high- sorption of glucose in the proximal at which the transport system is capacity, low-affinity transporter convoluted tubules. With today’s saturated (analogous to Vmax in found primarily in the kidney. A greater understanding of the renal metabolism). This is because some third member of this family, SGLT3, mechanisms responsible for glucose nephrons excrete glucose before is found widely throughout the body homeostasis and with modalities to others have reached their transport in skeletal muscle and the nervous manipulate that system, the kidney maximum.8 The renal transport system. SGLT3 is not thought to be a may one day play a key role in the maximum is reached at the point , but instead acts management of hyperglycemia. where all nephrons have exceeded as a glucose sensor.11 Although other their maximum capacity to reabsorb members of this family have been Filtration and Reabsorption of Glucose glucose. identified (SGLT4, -5, and -6), their The glomeruli of a normal healthy Clinically, the most common function in humans is uncertain at adult filter~ 180 g of glucose daily.8 cause of glucosuria is diabetes. this time. Under normal circumstances, virtu- Therefore, average patients will not The most prevalent and function- ally all of this glucose is reabsorbed “spill” glucose into their urine until ally important SGLT in the kidney with < 1% being excreted in the urine.9 their blood glucose concentration is SGLT2. This transporter accounts The reabsorption of glucose from the exceeds 180 mg/dl, which does not for ~ 90% of glucose reabsorption in tubules is a multistep process involv- normally occur in individuals with- the kidney and because of this has ing several transport mechanisms. out diabetes. become the focus of a great deal of The glucose, once filtered into the The first step in the reabsorption interest in the field of diabetes. This tubule, must be transported out of the of glucose from the urine involves transporter is found at a relatively tubule, through the tubular epithelial the transport of glucose from the high density on the brush-boarder cells, and then across the basolat- tubule into the tubular epithelial membrane of the S1 segment (the eral membrane into the peritubular cells. This is accomplished by SGLTs early segment) of the proximal capillary. Under optimal conditions, (Figure 1). convoluted tubule.6,12 SGLT2 binds when the tubular glucose load is SGLTs include an extensive with both sodium and glucose in the ~ 120 mg/min or less, no glucose is array of membrane proteins that tubular filtrate. These compounds lost in the urine. However, when the transport glucose, amino acids, are then translocated across the glucose load exceeds ~ 220 mg/min vitamins, ions, and osmolytes across apical cell membrane. This process (the so-called “glucose threshold”), the brush-border membrane of is called secondary active-transport glucose begins to appear in the urine. proximal renal tubules as well as the and is driven by the electrochemical The blood glucose value required intestinal epithelium.6 SGLT1 is a sodium gradient between the tubular to provide such a tubular load is not low-capacity, high-affinity SGLT. It filtrate and the cell.6 a set value in humans, but rather is a is located principally in the gastro- The average maximal reabsorp- range. One study evaluating this pro- intestinal tract but can also be found tive capacity (Tm) of the renal tubules cess reported that the required blood in the S3 segment of the proximal is variable, but typically is ≈ 375 mg/

6 Volume 28, Number 1, 2010 • CLINICAL DIABETES FEATURE ARTICLE

and glucose-6-phosphate enzyme activity sufficient to drive gluconeogenesis. The kidney is important in this regard because renal glucose production accounts for ~ 20% of all overall endogenous glucose release and is responsible for ~ 40% of glucose released secondary to gluconeogenesis.14 The role of disproportionate glu- coneogenesis as a factor worsening hyperglycemia in patients with either type 1 or type 2 diabetes is well documented.14 Significant increases in renal gluconeogenesis have been demonstrated in animal diabetes models and in human studies. Renal glucose release was reported to be proportional to hepatic glucose Figure 1. Glucose reabsorption from the glomerular filtrate through a proximal release in one study in patients with tubule epithelial cell into the blood. Reprinted with permission from Ref. 6. type 2 diabetes.15 It seems probable that abnormali- 13 min. In nondiabetic individuals, (thus reducing the expression of ties in renal glucose production and the filtered glucose load does not, SGLT2 transporters). Although both release play a role in the progres- under normal circumstances, exceed methods offer promise, pharmaco- sion of hyperglycemia in patients 375 mg/min. In this situation, the logical inhibition of SGLT2 is at a with diabetes. It is clearer, however, filtered glucose is reabsorbed and more advanced stage of development that the kidney plays a major role in returned to the systemic circulation. (phase 3 trials). glucose homeostasis and offers many Conversely, hyperglycemic patients The second step in the reabsorp- junctures at which pharmacologi- with diabetes routinely experience tion of glucose is the transport of cal intervention could be effective in a filtered glucose load of> 375 mg/ glucose through the basolateral mem- the alteration of glucose balance in min. Under these conditions, the brane and back into the peritubular patients with diabetes and perhaps reabsorptive capacity of the SGLT2 capillary. This facilitated diffusion in obese patients as well. transporters is exceeded, and the of glucose is dependent on another excess glucose is eliminated via the well-recognized family of glucose SGLT2 Inhibitors and Diabetes urine. transporter (GLUT) proteins.14 This Although the possibility of modu- Inhibition of the activity of basolateral membrane-peritubular lation of the SGLT2 system via SGLT2 in the renal tubules has capillary transfer utilizes GLUT1 anti-sense molecules is being evalu- become an area of great pharmaco- transporters in the late proximal ated, the use of SGLT2 inhibitors is logical interest in the past few years. tubule and GLUT2 transporters in much further along in development. Pharmacological and even biologi- the early proximal tubule. Several SGLT2 inhibitors have been cal interventions to modulate this developed, and some have already pathway by reducing the reabsorp- Gluconeogenesis been discarded. Clinical trials with tive capacity of the renal tubules, The production and release of glucose sergliflozin and remogliflozin (KGT- with a resultant elimination of excess into the systemic circulation can 1611) have been discontinued by are being investigated. As mentioned occur by gluconeogenesis or glycoge- GlaxoSmithKline (GSK) and Kissei above, at present, the two most nolysis. Currently, it is believed that Pharmaceuticals. Cited reasons for likely effective methods are admin- gluconeogenesis is responsible for ~ this include “result of evaluating istration of chemical inhibitors of 55% of glucose released during the circumstances including the develop- SGLT2 (thus reducing the activity of non-fed period.14 The liver and the ment status of SGLT2 inhibitors by present SGLT2 transporters) or the kidney are the only two organs in the competitors.”16 It is interesting to note, use of SGLT2 anti-sense molecules body that possess the glucoenogenic however, that Kissei/GSK are con-

CLINICAL DIABETES • Volume 28, Number 1, 2010 7 FEATURE ARTICLE

tinuing work on an SGLT1 inhibitor occurring at 1 hour (range 0.5–4 that the level of glucosuria was (KGA-3235) for the management of hours).21 One small study in healthy dose-dependent. The amount of 17 hyperglycemia in diabetes. volunteers suggested that the Tmax glucose excreted via the urine ranged Another SGLT2 inhibitor, occurs at a later time when the from 18 to 62 g of glucose daily canagliflozin, is being developed drug is administered in the fed state with doses of 2.5–100 mg daily. The by Johnson & Johnson and is cur- (median = 4 hours) but that there is average amount of glucose lost daily rently in phase 3 trials in the United very little difference in exposure to remained stable during the 14-day States. It is being studied as an the drug when fed is compared to dosing period of the study. No add-on therapy to metformin versus fasting (area under the curve [AUC] apparent change in glycemic indexes glimepiride.18 = 12,455 and 13,337 ng · hr/ml, was observed in these patients dur- The most prominent SGLT2 respectively).22 The half-life of ing the study. No treatment-related inhibitor at this time is dapa- dapgliflozin is reported to be ~ 16 serious adverse effects occurred dur- 21 ing this trial. The authors concluded gliflozin, a medication being hours. Increases in Cmax and AUC co-developed in the United States are roughly proportional to dose in that dapagliflozin demonstrated by AstraZeneca and Bristol-Myers the range of 5–100 mg (as is glucos- pharmacokinetic and pharmaco- Squibb. Currently, 20 clinical studies uria effect).21 dynamic properties amenable to have been completed or are in prog- Renal clearance of dapagliflozin once-daily dosing. ress in the United States with this is minimal (~ 3–6 ml/minute) with Another trial carried out by the compound.18 < 2.5% being excreted unchanged in same group evaluated the effects of At least three other SGLT2 inhib- the urine during a 24-hour period. dapagliflozin in 47 subjects with type itors are currently being evaluated In vitro studies have suggested that 2 diabetes.21 Researchers random- in clinical trials. Two compounds dapagliflozin may be metabolized ized patients to placebo or 5, 25, (BI 10773 and BI 44847) are being to an inactive metabolite via the or 100 mg dapagliflozin daily and developed by Boehringer Ingelheim glucuronosyltransferase enzyme assessed the effects of the drug dur- in Germany and another compound coded for by UGT1A9.20 A popula- ing a 2-week period. The treatment (YM-543) is being developed by the tion kinetic analysis of 30 healthy groups were balanced in terms of Japanese company Atstellas.19 subjects and 38 subjects with type demographics and baseline charac- 2 diabetes concluded that there teristics. The subjects were racially Dapagliflozin were no substantive differences mixed, and 18 were being treated in the dapagliflozin pharmacoki- with, and continued, metformin Dose and administration/ netic parameters between the two therapy throughout the study. pharmacokinetics groups.23 Statistically significant, dose- Dapagliflozin has been shown to have dependent reductions in fasting some anti-hyperglycemic efficacy Clinical trials serum glucose versus placebo were at doses of 2.5, 5, 10, 20, and 50 mg A thorough review of all available observed on day 14 (5 mg: 11.7%; daily in phase 2 trials.20 The majority phase 1 and phase 2 trials with 25 mg: 13.3%; 100 mg: 21.8%). of phase 3 trials currently underway dapagliflozin has been published.20 Statistically significant reductions are evaluating the effects of 2.5-, Additionally, as mentioned above, a in oral glucose tolerance test results 5-, or 10-mg tablets administered plethora of phase 3 trials are cur- were also observed at all doses of once daily.18 One study assessing rently underway. Several published active medication on days 2 and 13. the potential for QT prolongation trials offer insight into how well Urine glucose loss on day 14 was when dapagliflozin is administered dapagliflozin works and how well it is ~ 37, 70, and 70 g daily for 5, 25, and concurrently with moxifloxacin is tolerated. 100 mg, respectively. using single oral doses as high as 150 One study evaluated the effects There were no reported serious mg.18 However, it is unlikely that this of dapagliflozin in healthy sub- adverse effects or discontinuations high provocative dose would ever be jects.22 This trial assessed the safety, reported secondary to adverse used clinically. One study in Japanese tolerability, pharmacokinetics, effects in this trial. Reported adverse patients is using doses as low as 1 mg and pharmacodynamics of dapa- effects did not appear to be dose daily. gliflozin in single-ascending-dose related. The most common adverse Dapagliflozin is rapidly absorbed and multiple-ascending-dose fash- effects were gastrointestinal in after an oral dose in patients with ion, with doses ranging from 2.5 to nature and were more common in type 2 diabetes with a median Tmax 100 mg daily. The authors reported patients receiving metformin. There

8 Volume 28, Number 1, 2010 • CLINICAL DIABETES FEATURE ARTICLE

were two reports of self-resolving obvious dose relationship. There lost at least 5% of their baseline body hypoglycemia in this study in were no documented hypoglyce- weight at 24 weeks. patients taking dapagliflozin plus mic events with fingerstick glucose Urinary tract infections were metformin. concentrations ≤ 50 mg/dl. Urinary reported in 8% of placebo- and A larger trial was carried out tract infections were reported in 4.4-8.1% of dapagliflozin-taking in 389 diabetes drug–naive type 2 6% of placebo-, 9% of metformin-, patients. Also, genital infections diabetic patients.24 Researchers in and 5–12% of dapagliflozin-taking were reported in 5.1% of placebo-, this trial randomized patients at 98 patients. Also, genital infections 2% of metformin-, and 8-13.1% of clinical centers to either placebo; were reported in 0% of placebo-, 2% dapagliflozin-taking patients. Rates dapagliflozin in doses of 2.5, 5, 10, of metformin-, and 2-7% of dapa- of hypoglycemia were similar in all 20, and 50 mg daily; or extended- gliflozin-taking patients. groups in this study, and no patients release metformin (titrated to 1,500 The impact of dapagliflozin discontinued the study because of mg at week 2). The study was a hypoglycemia. on 546 patients inadequately con- prospective, 12-week, random- A very interesting trial evaluated trolled on metformin was evaluated ized, parallel-group, double-blind, the effectiveness of dapagliflozin in a recent 24-week, randomized, placebo-controlled trial and in patients with diabetes poorly double-blind, placebo-controlled, included a 2-week diet and exercise controlled with other oral anti- multicenter trial.25 The trial evalu- lead in. The primary objective of hyperglycemic agents and high-dose this trial was to evaluate mean A1C ated patients with type 2 diabetes insulin.26 Seventy-one patients in this change from baseline after 12 weeks between the ages of 18 and 77 years trial were randomized to placebo or versus placebo. Secondary measures who, despite being treated with a 10 or 20 mg of dapagliflozin. Patients included fasting plasma glucose and dose of ≥ 1,500 mg of metformin were continued on their previous percentage of patients achieving an daily continued to have A1C levels oral anti-hyperglycemic medications A1C < 7%. between 7 and 10%. After an initial but continued insulin at half of their A total of 348 patients completed 2-week lead-in phase, the patients pre-study dose. After 12 weeks, A1C the 12-week study. At 12 weeks, all were randomized to either placebo values were -0.7 and -0.7.8% for 10 dapagliflozin patient groups had or dapagliflozin in doses of 2.5, 5, and 20 mg of dapagliflozin, respec- statistically significant reductions in or 10 mg. All patients continued on tively, from baseline versus placebo. mean A1C from baseline versus pla- metformin. The primary endpoint Mean reductions in body weight cebo. Adjusted mean A1C reductions of this trial was A1C reduction at were -1.9, -4.5, and -4.3 kg for were -0.18% for placebo, -0.73% 24 weeks. Additional endpoints placebo, 10 mg, and 20 mg of dapa- for metformin, and -0.55 to -0.9% included fasting plasma glucose and gliflozin, respectively. There was a for dapagliflozin. Adjusted mean percentage change in total body low incidence of adverse events in fasting plasma glucose reductions weight at 24 weeks. this trial. However, there were more were -6 mg/dl for placebo, -18 mg/dl Adjusted mean A1C reductions genital infections reported for the for metformin, and 16 to 31 mg/dl - - at 24 weeks were -0.3% for placebo group taking 20 mg of dapagliflozin for dapagliflozin. The fraction of and -0.67, -0.7%, and -0.84% for than for placebo. patients achieving A1C < 7% at 12 2.5, 5, and 10 mg of dapagliflozin, weeks was 32% for placebo, 54% Conclusion respectively, compared to base- for metformin, and 40-59% for The kidney plays a key role in glucose line. Adjusted mean fasting plasma dapagliflozin. Also, reductions in homeostasis. The delineation of the total body weight were reported glucose reductions were -6 mg/dl for role of SGLTs throughout the body for all groups. Weight reduction at placebo and -17.8, -21.5, and -23.5 has been a major advancement in week 12 was 1.2% for placebo, 1.7% mg/dl with 2.5, 5, and 10 mg of dapa- our understanding of the kinetics for metformin, and 2.5-3.4% for gliflozin, respectively, compared to of glucose in the body. Our better dapagliflozin. baseline. The percentage change understanding of the physiology No serious adverse effects were in baseline weight was -1.02% for of renal glucose homeostasis and reported. Hypoglycemic events were placebo and -2.66, -3.66, and -3.43% recent pharmacological advances reported in 4% of placebo-treated for 2.5, 5, and 10 mg of dapa- have brought us to a point where patients, 9% of metformin-treated gliflozin, respectively, at 24 weeks. pharmacological modulation of this patients, and 6–10% of dapa- Additionally, it was noted that ~ 25% system is possible. Although diabetes gliflozin-treated patients with no of all dapagliflozin-taking patients management is an obvious candidate

CLINICAL DIABETES • Volume 28, Number 1, 2010 9 FEATURE ARTICLE

to benefit from these advances, other in type 2 diabetes mellitus. Endocr Pract development_of_remogliflozin_090703/. 14:782–790, 2008 Accessed 21 October 2009 medical conditions such as obesity 6Bakris GL, Fonseca V, Sharma K, 18National Institutes of Health: may also eventually be targeted via Wright E: Renal sodium-glucose transport: ClinicalTrials.gov [article online]. Available methods to modulate either SGLT1 or role in diabetes mellitus and potential clinical from http://www.clinicaltrials.gov. Accessed 22 October 2009 SGLT2. implications. Kidney Int 75:1272–1277, 2009 7Isis Pharmaceuticals: Pipeline [article 19Marsenic O: Glucose control by the Currently, a host of studies are online]. Available from http://www.isi- kidney: an emerging target in diabetes. Am J underway evaluating the effects of spharm.com/Pipeline/index.htm. Accessed 13 Kidney Dis 53:875–883, 2009 the SGLT2 inhibitor dapagliflozin. October 2009 20Brooks AM, Thacker SM: Dapagliflozin 8Guyton AC, Hall JE: Urine formation for the treatment of type 2 diabetes. Ann Although the published studies offer and the kidneys. In Textbook of Medical Pharmacotherapy. Published electronically on a great deal of promise, there is still Physiology. 9th ed. Philadelphia, Pa., W.B. 7 July 2009; DOI 10.134/aph.1M212 some concern regarding potential Saunders, 1996, p. 332–335 21Komoroski B, Vachharajani N, Feng Y, 9 Li L, Kornhauser D, Pfister M: Dapgliflozin, side effects, namely urinary tract Wright EM: Renal N+-glucose transporters. Am J Physiol Renal Physiol a novel, selective SGLT2 inhibitor, improved and genital infections. As data from 280:F10–F18, 2001 glycemic control over 2 weeks in patients with type 2 diabetes mellitus. Clin Pharmacol Ther phase 3 trials develop, many of 10Butterfield WJH, Keen H, Whichelow 85:513–519, 2009 MJ: Renal glucose threshold variations with these unanswered questions will be 22 age. BMJ 4:505–507, 1967 Komoroski B, Vachharajani N, Boulton resolved. D, Kornhauser D, Geraldes M, Pfister M: 11Wright EM, Hirayama BA, Loo DF: Dapgliflozin, a novel, selective SGLT2 In addition to dapagliflozin, Active sugar transport in health and disease. inhibitor, induces dose-dependent glucosuria several other SGLT2 inhibitors, J Intern Med 261:32–43, 2007 in healthy subjects. Clin Pharmacol Ther 85:520–526, 2009 an SGLT anti-sense molecule, and 12Moe OW, Berry CA, Rector FC: Renal transport of glucose, amino acids, sodium, 23Feng Y, Zhang L, Komoroski B, Pfister SGLT1 inhibitors are being devel- chloride, and water. In Brenner and Rector’s M: Population pharmacokinetic analysis of oped. Eventually, pharmacological The Kidney. 6th ed. Brenner BM, Ed. dapagliflozin in healthy subjects with type 2 Philadelphia, Pa., W.B. Saunders, 2000, p. diabetes mellitus. [Abstract]. Clin Pharmacol or biological modulation of these 378–380 Ther 84 (Suppl. 1):S93, 2008 transporters will almost certainly 13Valtin H: Tubular reabsorption. In 24List JF, Woo V, Morales E, Tang W, offer us a useful method of disease Renal Function. Boston, Mass., Little, Brown Fiedorek FT: Sodium-glucose cotransport and Company, 1983, p. 65–68 inhibition with dapagliflozin in type 2 diabe- intervention. tes. Diabetes Care 32:650–657, 2009 14Gerich JE, Woerle HJ, Meyer C, Stumvoll M: Renal gluconeogenesis. Diabetes 25Bailey CJ, Gross JL, Bastone L, List REFERENCES Care 24:382–391, 2001 JF: Dapagliflozin as an add-on to metformin 1 lowers hyperglycemia in type 2 diabetes Ehrenkranz RRL, Lewis NG, Kahn CR, 15Meyer C, Stumvoll M, Nadkarni V, patients inadequately controlled with metfor- Roth J: Phlorizin: a review. Diabetes Metab Dostou J, Mitrakou A, Gerich J: Abnormal min alone [Abstract]. Diabetologia 52 (Suppl. Res Rev 21:31–38, 2005 renal and hepatic glucose metabolism in type 1):S76, 2009 2 2 diabetes mellitus. J Clin Invest 102:619–624, Stiles PG, Lusk G: On the action of phlo- 26 1998 Wilding JPH, Norwood P, T’joen C, rizin. Am J Physiol 1903;10:61–79 Bastien A, List JF, Fiedorek FT: A study of 16 3Vick H, Diedrich DF, Baumann K: Kissei Pharmaceutical Co., Ltd.: dapagliflozin in patients with type 2 diabetes Reevaluation of renal tubular glucose trans- Discontinuation of the development of receiving high doses of insulin plus insulin port inhibition by phlorizin analogs. Am J “remogliflozin” by GlaxoSmithKline [article sensitizers. Diabetes Care 32:1656–1662, 2009 Physiol 224:552–557, 1973 online]. Available from http://www.kissei. co.fp/e_contents/press_e/2009/e20090703. 4Lee WS, Wells RG, Hediger MA: The html. Accessed 21 October 2009 John R. White, Jr., PA, PharmD, high affinity NA/glucose cotransporter: re- 17 evaluation and distribution of expression. J Pharmaceutical Business Review: is a professor in the Department of Biol Chem 269:12032–12039, 1994 GlaxoSmithKline discontinues develop- ment of remogliflozin [article online]. Pharmacotherapy at the College 5Abdul-Ghani MA, DeFronzo RA: Available from http://inwardinvestment. Inhibition of renal glucose absorption: a pharmaceutical-business-review.com/ of Pharmacy at Washington State novel strategy for achieving glucose control news/glaxosmithkline_discontinues_ University in Spokane.

10 Volume 28, Number 1, 2010 • CLINICAL DIABETES