Clinical Care/Education/Nutrition ORIGINAL ARTICLE

Nateglinide Improves Early Insulin Secretion and Controls Postprandial Glucose Excursions in a Prediabetic Population

CAROLA SALORANTA, MD1 KAJ LAHTI, MD4 Diabetes Trial (5), and the TRIPOD Study CHRISTIANE GUITARD, MD2 PATRICK BRUNEL, MD2 (6) have suggested that pharmacotherapy ECKHARD PECHER, MSC2 LEIF GROOP, MD5 (with metformin, acarbose, and thiazo- 3 PEDRO DE PABLOS-VELASCO, MD lidinediones, respectively) is a valid ap- proach to diabetes prevention in patients with IGT at high risk of developing dia- betes, such as those with a family history OBJECTIVE — The purpose of this study was to evaluate the metabolic effectiveness, safety, and tolerability of nateglinide in subjects with impaired glucose tolerance (IGT) and to identify of diabetes (5) or women with recent ges- a dose appropriate for use in a diabetes prevention study. tational diabetes (6). These data raise the question of whether another agent, acting RESEARCH DESIGN AND METHODS — This multicenter, double-blind, random- by a different mechanism, may be equally ized, parallel-group, fixed-dose study of 8 weeks’ duration was performed in a total of 288 or more effective in the prevention of type subjects with IGT using a 2:2:2:1 randomization. Subjects received nateglinide (30, 60, and 120 2 diabetes. mg) or placebo before each main meal. Metabolic effectiveness was assessed during a standard- The D-phenylalanine derivative, ized meal challenge performed before and after the 8-week treatment. All adverse events (AEs) nateglinide, is a new insulinotropic agent were recorded, and confirmed hypoglycemia was defined as symptoms accompanied by a self- monitoring of blood glucose measurement Յ3.3 mmol/l (plasma glucose Յ3.7 mmol/l). that specifically targets mealtime glucose excursions by restoring or replacing early RESULTS — Nateglinide elicited a dose-related increase of insulin and a decrease of glucose prandial insulin release that is markedly during standardized meal challenges, with the predominant effect on early insulin release, impaired early in the progression of type 2 leading to a substantial reduction in peak plasma glucose levels. Nateglinide was well tolerated, diabetes (7) and, indeed, in IGT (8). Un- and symptoms of hypoglycemia were the only treatment-emergent AEs. Confirmed hypoglyce- like most sulfonylureas (SUs), nateglinide mia occurred in 28 subjects receiving nateglinide (30 mg, 0 [0%]; 60 mg, 5 [6.6%]; 120 mg, 23 is found to have a low risk of hypoglyce- [26.7%]) and in 1 (2.3%) subject receiving placebo. mia in patients with diabetes (9,10) be- CONCLUSIONS — Nateglinide was safe and effective in reducing postprandial hyperglyce- cause its stimulation of insulin release is mia in subjects with IGT. Preprandial doses of 30 or 60 mg nateglinide would be appropriate to rapidly reversible and glucose dependent use for longer-term studies to determine whether a rapid-onset, rapidly reversible, insulinotropic (11,12). Since nateglinide addresses the agent can delay or prevent the development of type 2 diabetes. key factor that determines whether an in- dividual with IGT progresses to overt di- Diabetes Care 25:2141–2146, 2002 abetes (7), it may be a particularly attractive option for use in diabetes pre- vention. everal studies have shown that (1–3); however, such changes in lifestyle However, an insulin secretion agent weight loss and exercise can delay or are notoriously difficult to maintain. Very used in subjects with minimal fasting hy- S prevent progression from impaired recently, results from the Diabetes Pre- perglycemia must have an acceptably low glucose tolerance (IGT) to type 2 diabetes vention Program (4), the STOP-NIDDM hypoglycemic potential. Further, an agent used for an extended period of time ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● must have an excellent overall safety and From the 1Helsinki University Hospital, Department of Medicine, Helsinki, Finland; 2Novartis Pharma, AG, tolerability profile. Thus, before nateglin- Basel, Switzerland; the 3University of Las Palmas de Gran Canaria, Faculty of Medicine, Las Palmas, Canary Islands, Spain; the 4Primary Health Care Center, Vaasa, Finland; and the 5Department of Endocrinology, ide could be used in a diabetes prevention University of Lund, Malmo, Sweden. study, it would be essential to demon- Address correspondence and reprint requests to Christiane Guitard, Clinical Research, Novartis Pharma, strate, in the target prediabetic popula- AG, CH-4002, Basel, Switzerland. E-mail: [email protected]. tion, that it has these properties. The Received for publication 4 April 2002 and accepted in revised form 11 September 2002. C.G. and E.P. hold stock in, C.S. has received funds from, and P.d.P.-V. has received honoraria from purpose of the present study was to assess Novartis. the safety, tolerability, and metabolic ef- Abbreviations: AE, adverse event; AUC, area under the curve; DCCT, Diabetes Control and Complica- fectiveness of nateglinide in subjects with tions Trial; FPG, fasting plasma glucose; IGT, impaired glucose tolerance; IRI, immunoreative insulin; IGT and to identify an appropriate dose OGTT, oral glucose tolerance test; SAE, serious AE; SMBG, self-monitoring of blood glucose; SU, sulfonyl- urea; UKPDS, U.K. Prospective Diabetes Study; ULN, upper limit of normal. for use in long-term studies to assess its A table elsewhere in this issue shows conventional and Syste`me International (SI) units and conversion potential to delay or prevent the develop- factors for many substances. ment of type 2 diabetes.

DIABETES CARE, VOLUME 25, NUMBER 12, DECEMBER 2002 2141 Nateglinide effects in IGT

RESEARCH DESIGN AND returned for assessment of FPG and fruc- compared between groups by Fisher’s ex- METHODS tosamine at weeks 2 and 4 and for stan- act test (two-sided, not adjusted for mul- Study protocol. This multicenter, dou- dard laboratory tests, FPG, fructosamine, tiplicity). Efficacy analyses were ble-blind, randomized, parallel-group and HbA1c at week 8. The final dose was performed to assess the effects of nateg- fixed-dose study of 8 weeks’ duration was followed by a standardized meal linide on prandial glucose and insulin performed to define the maximum toler- challenge. (3-h area under the curves [AUCs] calcu- ated dose of nateglinide in a prediabetic During the meal challenge, a fasting lated by the trapezoidal method), FPG, Ϫ population. The effects of nateglinide on plasma sample ( 15 min) was obtained, HbA1c, and fructosamine. The change prandial glucose and insulin, fruc- then 150 ml of water (week 0) or study from baseline at week 8 was analyzed by tosamine, fasting plasma glucose (FPG), medication with water (week 8) was taken ANCOVA with treatment and center as and HbA1c levels were also assessed. A and a 500 kcal breakfast meal was initi- factors and baseline as covariate. Sum- total of 288 nondiabetic subjects (FPG ated within 10 min and consumed within mary statistics of absolute values and Ͻ7.0 mmol/l) with IGT were randomized the next 15 min. The breakfast was com- changes from baseline by treatment group to receive 30 mg (n ϭ 83), 60 mg (n ϭ posed of orange juice (180 ml), toasted and time point were calculated for the in- 76), or 120 mg (n ϭ 86) nateglinide or bread (60 g), jam or preserves (20 g), but- tent-to-treat population. No formal placebo (n ϭ 43) before (within 10 min ter or margarine (10 g), whole milk (120 hypothesis testing was performed. P val- of) each main meal using a 2:2:2:1 ran- ml), and coffee or tea with nonnutritive ues derived from the ANCOVA were ex- domization. To determine the effects of sweetener if desired. Plasma samples for plorative and were not adjusted for nateglinide on prandial glucose and insu- measurement of glucose and insulin were multiplicity. lin levels, a standardized meal challenge obtained at 10, 20, 30, 40, 60, 120, and was performed before the first dose (week 180 min after initiation of the meal. RESULTS 0) and after the final dose (week 8). All subjects were provided with a Characteristics of the study popula- The study was carried out in 32 cen- home glucose monitoring device and tion. Table 1 reports the baseline charac- ters in six countries and enrolled male and were instructed to perform a fingerstick teristics of the study population. There female subjects with an FPG Ͻ7.0 mmol/l glucose test for all episodes of suspected were no differences among treatment and with IGT confirmed by two 2-h post– hypoglycemia. Confirmed hypoglycemia groups in demographics or in metabolic Ն glucose challenge values 7.8 mmol/l was defined as an event for which SMBG indexes (HbA1c, FPG, and fructosamine). but Ͻ11.1 mmol/l measured during oral value was Յ3.3 mmol/l, corresponding to Nearly all subjects were Caucasian, glucose tolerance tests (OGTTs) per- a plasma glucose Յ3.7 mmol/l. women outnumbered men by a ratio of formed within 1 year and repeated within Glucose was measured by an enzy- ϳ60:40, mean age was ϳ57 years (with 2 weeks before study entry. Subjects matic method (Boehringer Mannheim, ϳ25% of each treatment group Ͼ65 agreed to maintain their prior diet and to Mannheim, Germany) and fructosamine years), and the mean BMI in each group provide written informed consent to par- was measured by a spectrophotometric was 30 kg/m2.Reflecting the selection cri- ticipate. Main exclusion criteria were method (Roche Diagnostics, Basel, Swit- teria, mean FPG ranged from 5.9 to 6.2 prior or concomitant antidiabetic agents, zerland) using a Cobas Bio analyzer mmol/l in the four groups; HbA1c and investigational drugs within 4 weeks, car- (Roche). Blood levels of HbA1c were mea- fructosamine levels were near their re- diovascular events within 6 months, liver sured by high-performance liquid chro- spective ULNs. disease, or persistent elevations of liver matography standardized to the Diamat Safety and tolerability. Nateglinide was enzymes. The study protocol received In- ion-exchange method (upper limit of nor- very well tolerated. The overall incidence stitutional Review Board approval at all mal [ULN] ϭ 6.0%). Insulin was mea- of AEs, other than symptoms of hypogly- sites and was conducted in accordance sured by radioimmunoassay (Pharmacia, cemia, was low and similar among all with the U.S. Code of Federal Regula- Uppsala, Sweden). All laboratory samples treatment groups. No deaths occurred tions, the rules governing medical proce- were analyzed at a central laboratory during the study, and the incidence of se- dures in the European Community and (CRL, Brussels, Belgium). rious AEs (SAEs) was 1 (1.2%), 0, 1 the Declaration of Helsinki. Data analysis. Baseline and demo- (1.2%), and 2 (4.7%) in the nateglinide Visits and evaluations. During the base- graphic variables were compared among 30, 60, and 120 mg and placebo groups, line (week 0) visit, subjects underwent groups with an F test for numeric param- respectively. Only one SAE was suspected physical exams, electrocardiograms, and eters and by a Mantel-Haenszel test for to be related to study medication, but standard laboratory tests (hematology, catergorical variables. The primary safety upon unblinding, this subject was found chemistry, and urinalysis) after an over- evaluation was based on the incidence of to be in the placebo group. Three (3.6%), night fast. Samples were obtained for confirmed hypoglycemia. Acceptable two (2.6%), and none of the subjects in assessment of FPG, HbA1c, and fruc- safety was defined prospectively as an in- the 30, 60, and 120 mg nateglinide tosamine, then the subjects underwent cidence of severe (grade B—requiring groups and one (2.3%) in the the week 0 standardized breakfast meal assistance from an outside party) hypo- placebo-treated group, respectively, challenge. Dosing was initiated before glycemia Ͻ0.1% in the combined nateg- withdrew due to any AE other than hypo- lunch, and subjects were instructed to linide groups and a discontinuation rate glycemia. The incidence of clinically no- take one tablet before each main meal for of Ͻ5% within a dose group. Binomial table laboratory parameters (hematology a minimum of two and a maximum of tests were performed on these two pri- and biochemistry) was low in all subjects four meals per day and to not take study mary safety criteria. In post hoc analyses, and none were dose-related. medication if a meal was missed. Subjects the frequency of adverse events (AEs) was Because the purpose of this study was

2142 DIABETES CARE, VOLUME 25, NUMBER 12, DECEMBER 2002 Saloranta and Associates

Table 1—Baseline demographic and background characteristics (randomized population)

Nateglinide 30 mg 60 mg 120 mg Placebo n 83 76 86 43 Male sex 34 (41.0) 32 (42.1) 36 (41.9) 19 (44.2) Caucasian race 82 (98.8) 76 (100.0) 86 (100.0) 42 (97.7) Age (years) 56.4 Ϯ 11.0 57.6 Ϯ 10.9 57.2 Ϯ 10.4 58.4 Ϯ 10.8 Ͻ65 years 60 (72.3) 57 (75.0) 65 (75.6) 30 (69.8) Ն65 years 23 (27.7) 19 (25.0) 21 (24.4) 13 (30.2) BMI (kg/m2)* 30.2 Ϯ 5.3 30.4 Ϯ 5.2 30.0 Ϯ 6.6 30.2 Ϯ 5.6 FPG (mmol/l)† 6.09 Ϯ 0.08 5.88 Ϯ 0.08 6.20 Ϯ 0.08 6.20 Ϯ 0.15 Ϯ Ϯ Ϯ Ϯ HbA1c (%) 5.76 0.06 5.70 0.06 5.87 0.07 5.78 0.10 Fructosamine (␮mol/l) 249 Ϯ 3 251 Ϯ 4 258 Ϯ 3 254 Ϯ 5 Data are n (%), *means Ϯ SD, or †means Ϯ SE. There were no statistically significant differences among groups in any baseline or demographic characteristic based on Cochran and Mantel-Haenszel tests or F tests. to determine the maximum tolerated dose number of events and sample size. All jects in the 120-mg group occurred dur- and thus to define an appropriate dose for plasma glucose levels during any con- ing the morning. a long-term diabetes prevention study, firmed hypoglycemic episode in the 60- Effects on glucoregulation. Figure 1 de- particular attention was paid to hypogly- mg nateglinide group and in the group picts the immunoreative insulin (IRI) and cemia. There were no incidents of severe who took placebo were Ͼ3.1 mmol/l. In glucose profiles during standardized hypoglycemia (requiring assistance from the 120-mg group, 11 subjects (12.8%) breakfast challenges performed at weeks an outside party) during this study. Given experienced hypoglycemia with a plasma 0 and 8 of treatment with nateglinide (60 the sample size of 245 subjects in the glucose equivalent Յ3.1 mmol/l; how- mg, before meals). This dose of nateglin- combined nateglinide groups, the bino- ever, the other 17 confirmed episodes had ide modestly and selectively increased the mial CI for the true underlying incidence plasma glucose values Ͼ3.1 mmol/l. In all early insulin response to the meal and ranged from 0.0 to 1.2%. A total of three groups, a missed or delayed meal, stren- greatly reduced the prandial glucose ex- subjects on nateglinide (1.2%) discontin- cursion. A lower dose produced a smaller uous exercise, or stress were the most fre- ued because of hypoglycemic symptoms increment of insulin and decrement of quently assigned precipitating factors, (one on 60 mg and two on 120 mg). As glucose, and a higher dose produced a reported in Table 2, the incidence of con- but no recognizable precipitating factor larger and more prolonged augmentation firmed hypoglycemia in the nateglinide- was identified by the subjects in most ep- of meal-stimulated insulin release and a treated groups was 0 in the 30 mg, 5 isodes observed with the 120-mg dose. marginally greater reduction of postpran- Ͻ (6.6%) in the 60 mg, 23 (26.7%) in the Most hypoglycemic events occurred 4h dial glucose levels. Ͻ 120 mg, and 1 (2.3%) in the placebo from the last meal and took 30 min to This is illustrated in Fig. 2, which group. There was clear statistical evidence resolve following oral carbohydrate in- shows the change from week 0 to week 8 for a higher incidence in the 120-mg take. All events occurred during the day, in the 3-h prandial AUCs of insulin and group, but other analyses should be inter- mainly during the afternoon, but many of glucose in the four treatment groups. The preted with caution in view of the low the symptoms experienced by the sub- effects of nateglinide to increase insulin

Table 2—Subjects with any AE or confirmed hypoglycemia (plasma glucose <3.7 mmol/l)

Nateglinide Treatment 30 mg 60 mg 120 mg Placebo n 83 76 86 43 Subjects with at least one AE 31 (37.3) 25 (32.9) 44 (51.2)* 13 (30.2) Signs or symptoms consistent with 10 (12.0) 10 (13.2) 26 (30.2)† 6 (14.0) hypoglycemia‡‡ Confirmed hypoglycemia 0 5 (6.6)§ 23 (26.7) 1 (2.3) Plasma glucose Յ3.1 mmol/l 0 1¶ 11 (12.8)# 0 Plasma glucose Ͼ3.1, Յ3.7 mmol/l‡‡ 0 5 (6.6)** 17 (19.8)† 1 (2.3) P values are post hoc calculations based on Fisher’s exact text (two-sided, not adjusted for multiplicity). *P Ͻ 0.05 vs. placebo and 60 mg; †P Ͻ 0.005 vs. 30 mg, P Ͻ 0.05 vs. 60 mg; ‡included in this category was increased sweating, weakness, headache, tremor, dizziness (excluding vertigo), fatigue, blurred vision, hunger, feeling hot, and nervousness; §P Ͻ 0.05 vs. 30 mg; ʈP Ͻ 0.001 vs. placebo, 30 mg, and 60 mg; ¶this value Ͻ3.1 mmol/l was not an accurate determination and resulted from improper handling of the sample according to the investigator; #P Ͻ 0.05 vs. placebo, P Ͻ 0.01 vs. 60 mg, P Ͻ 0.001 vs. 30 mg; **P Ͻ 0.05 vs. 30 mg; ††P Ͻ 0.05 vs. 60 mg, P Ͻ 0.01 vs. placebo, P Ͻ 0.001 vs. 30 mg; ‡‡whole-blood glucose Ͻ2.8, Յ3.3 mmol/l.

DIABETES CARE, VOLUME 25, NUMBER 12, DECEMBER 2002 2143 Nateglinide effects in IGT

lowering effect. Thus, by restoring a phys- iologic insulin profile, nateglinide essentially normalized glucose toler- ance—precisely what would be desired in an agent to be tested in a diabetes preven- tion study. It would also be necessary to have an exceptionally “clean” safety and tolerabil- ity profile, since the agent would be taken for a prolonged period of time in subjects with minimal hyperglycemia. In the present study, as described previously (14), nateglinide was very well tolerated. The incidence of AEs other than hypogly- cemia in nateglinide-treated subjects was indistinguishable from that in subjects who took a placebo. The non–hypoglyce- mia-related AEs that occurred more fre- quently in nateglinide than in the subjects who took a placebo were in no way dose related or suspected to be related to study medication. With regard to hypoglycemia, it is noteworthy that no severe hypoglycemia (requiring assistance from an outside party) occurred with any dose of nateglin- ide, likely due to its rapidly reversible (15) and glucose-dependent (12) insuli- notropic action. No hypoglycemia oc- curred in any patient receiving the 30 mg dose of nateglinide, and all hypoglycemia Figure 1—Plasma insulin (IRI) and glucose during standardized breakfast challenges performed that occurred in subjects given 60 mg was prior to the first dose at week 0 and after the final dose at week 8 with 60 mg nateglinide in 72 accompanied by a plasma glucose Ͼ3.1 nondiabetic subjects with IGT. Mean Ϯ SEM. mmol/l. In contrast, of the patients in the 120-mg group that did have confirmed hypoglycemia (26.7%), nearly half had and to curb prandial glucose excursions for a long-term study to determine one or more episodes during which were dose related, but only the highest whether an insulinotropic agent could de- plasma glucose levels fell below 3.1 dose (120 mg) significantly increased the lay or prevent the occurrence of type 2 mmol/l. In light of the increased inci- total insulin exposure. diabetes, as has recently been established dence and severity of hypoglycemia in As would be predicted from the char- for concerted lifestyle modification and subjects receiving 120 mg nateglinide, acteristics of the study population, the for metformin (4), acarbose (5), and tro- and given the minimal additional meta- duration of study, and the mechanism of glitazone (6) monotherapy. It was found bolic benefit it provides, this highest dose nateglinide action, this agent had no sig- that nateglinide elicited a dose-dependent does not appear to be a good option in nificant effect on FPG or on fasting insulin increase of early prandial insulin levels subjects with IGT. levels. The change from baseline HbA1c and reduction of prandial glucose excur- The 120 mg dose of nateglinide is was not statistically significant except in sions—both the amplitude of the basal to used safely and effectively in the vast ma- the 120-mg group (⌬ϭϪ0.12 Ϯ 0.04, peak spikes (13) and the 3-h AUCs. This jority of patients with type 2 diabetes P Ͻ 0.05). Nateglinide did, however, has been observed previously in patients treated with nateglinide, and it was re- modestly but significantly decrease fruc- with type 2 diabetes (9) and in healthy cently reported that in a 24-week study in tosamine levels (⌬ϭ30 mg, Ϫ8.3 Ϯ 0.4; volunteers (11). Since the present study patients with only moderately elevated 60 mg Ϫ8.6 Ϯ 2.1; and 120 mg, Ϫ8.1 Ϯ population represents a midpoint be- FPG (Յ7.8 mmol/l), 120 mg nateglinide ␮ Ͻ 2.0 mol/l; P 0.05 vs. placebo), al- tween the populations studied previ- preprandially normalized HbA1c levels though the effect was not dose related. ously, this finding is not in itself (Յ6.0%) in 40% of subjects, with an ac- unexpected. However, it does highlight ceptable incidence of hypoglycemia CONCLUSIONS — The goal of this the power of early insulin release. Thus, (5.3%, plasma glucose Յ3.3 mmol/l) study was to assess the safety, tolerability, the lowest dose tested (30 mg) had no (16). However, it is not surprising that a and metabolic effects of nateglinide in a effect on overall insulin exposure, as as- lower dose would be appropriate in non- prediabetic population and to thereby es- sessed by the 3-h AUC, but produced diabetic subjects with IGT. From the tablish the feasibility of using nateglinide more than half the maximal glucose- present findings, it appears that either 30

2144 DIABETES CARE, VOLUME 25, NUMBER 12, DECEMBER 2002 Saloranta and Associates

maintained diet therapy without tolbut- amide developed diabetes (25). In summary, augmenting early insu- lin secretion is a remarkably effective mechanism to control postprandial hy- perglycemia. Low-dose nateglinide could be used to determine the potential of this mechanism to prevent or slow the pro- gression of type 2 diabetes and may well become a therapeutic option for predia- betic subjects with only moderately dis- turbed glucose homeostasis.

Acknowledgments— The authors gratefully acknowledge the participating investigators (listed below) and their patients: Spain: Prof. Ramon Gomis (Barcelona), Dr. Pedro De Pablos (Las Palmas), Dr. Domingo Acosta (Sevilla), Dr. Concha Sanabrina (Cividad Real), Dr. Bernardo Costa (Mora d’Ebre), Dra Josefina Olivan (Sevilla). Ger- many: Prof Markolf Hanefeld (Dresden), Dr. Scholz (Leipzig), Dr. Ernest Schell (Nurn- berg), Dr. Michael Panijel (Frankfurt), Prof Ul- rich Foelsch (Kiel), Dr. Alexander Mann (Hamburg). Italy: Prof. Mario Maioli (Sassari), Prof. Georgia Pagani (Bergamo), Prof. Antonio Tiengo (Padova), Dr. Giuseppe Seghieri (Pis- toia), Dr. Felice Paleari (Monza), Dr. Lamberto De Giorgio (La Spezia). The Netherlands: Dr. MGAAM Nijpels (Zwaag). Sweden: Prof. Leif Groop (Malmo¨), Dr. Jan Eriksson (Umea), Dr. Figure 2—Change from week 0 to week 8 in the insulin (IRI) and glucose AUCs during stan- Ϯ Ͻ Ͻ Oerjan Johansson (Valdermarsvik). Finland: dardized breakfast challenges. Least squares mean SE. *P 0.025; **P 0.001 vs. placebo. Prof Sinikka Aija¨nseppa¨ (Helsinki), Dr. Carola Saloranta (Helsinki), Dr. Carol Forsblom (Hel- or 60 mg would be acceptable for a long- factor for cardiovascular disease (23). sinki), Dr. Bo Isomaa (Jakobstad), Dr. Kaj term study to determine whether this Lahti (Vaasa), Dr. Bjorn Forsen (Narpes), Dr. Since early insulin release and the result- Pasi Vanhala (Pieksamaeki). rapid-onset/short-duration insulino- ant suppression of hepatic glucose pro- tropic agent can delay or prevent manifest duction is a major determinant of diabetes in a prediabetic population and postprandial glucose levels (8), an agent References reduce the increased mortality and mor- such as nateglinide, which selectively 1. Tuomilehto J, Lindstrom J, Eriksson J, bidity associated with IGT (17,18). augments early insulin release, may be a Valle TT, Hamalainen H, Ilanne-Parikka P, Keinanen-Kiukaanniemi S, Laakso M, Although the Diabetes Control and particularly effective approach to reduc- Complications Trial (DCCT) in patients Louheranta A, Rastas M, Salminen V, ing cardiovascular morbidity and Uusitupa M: Prevention of type 2 diabetes with type 1 diabetes and the U.K. Pro- mortality. spective Diabetes Study (UKPDS) in pa- mellitus by changes in lifestyle among Accordingly, a diabetes prevention subjects with impaired glucose tolerance. tients with type 2 diabetes have study using nateglinide, which was N Engl J Med 344:1343–1350, 2001 convincingly established the benefits of shown here to substantially reduce post- 2. Viswanathan M, Snehalatha C, Vis- tight glycemic control to reduce micro- prandial hyperglycemia in subjects with wanathan V, Vidyavathi P, Indu J, Ram- vascular complications (19,20), macro- achandran A: Reduction in body weight vascular complications were not IGT, may allow prospective assessment of helps to delay the onset of diabetes even in significantly reduced by insulin or oral the role of glucose spikes, not only in the non-obese with strong family history of secretagogues. Many correlative studies development of overt diabetes, but also in the disease. Diabetes Res Clin Pract have suggested that postprandial hyper- the risk of macrovascular disease (24). 35:107–112, 1997 glycemia, or glucose spikes rather than The promise of this approach was dem- 3. Pan X-R, Li G-W, Hu Y-H, Wang J-X, total glycemic exposure, may play a key onstrated many years ago by a small study Yang W-Y, An Z-X, Hu Z-X, Lin J, Xiao in which none of the 23 men with IGT J-Z, Cao H-B, Liu P-A, Jiang X-G, Jiang role in the cardiovascular morbidity and Y-Y, Wang J-P, Zheng H, Zhang H, Ben- mortality associated with IGT and diabe- who continued treatment with the short- nett PH, Howard BV: Effects of diet and tes (13,17,21,22). Indeed, it has been re- acting SU, tolbutamide, and diet devel- exercise in preventing NIDDM in people ported that postprandial hyperglycemia, oped diabetes during the 10-year follow- with impaired glucose tolerance. Diabetes but not impaired fasting glucose, is a risk up, whereas 13% of the 98 subjects who Care 20:537–544, 1997

DIABETES CARE, VOLUME 25, NUMBER 12, DECEMBER 2002 2145 Nateglinide effects in IGT

4. Knowler WC, Barrett-Connor E, Fowler Diabetes Care 23:1660–1665, 2000 intensive treatment of diabetes on the de- SE, Hamman RF, Lachin JM, Walker EA, 11. Kalbag JB, Walter YH, Nedelman JR, velopment and progression of long-term Nathan DM: Reduction in the incidence of McLeod JF: Mealtime glucose regulation complications in insulin-dependent dia- type 2 diabetes with lifestyle intervention with nateglinide in healthy volunteers: betes mellitus. N Engl J Med 329:977–986, or metformin. N Engl J Med 346:393–403, comparison with repaglinide and pla- 1993 2002 cebo. Diabetes Care 24:73–77, 2001 20. UK Prospective Diabetes Study Group: 5. Chiasson JL, Josse RG, Gomis R, Hanefeld 12. Hu S, Wang S, Dunning BE: Glucose-de- Intensive blood-glucose control with sul- M, Karasik A, Laakso M: Acarbose for pre- pendent and glucose-sensitizing insulino- phonylureas or insulin compared with vention of type 2 diabetes mellitus: the tropic effects of nateglinide: comparison conventional treatment and risk of com- STOP-NIDDM randomised trial. Lancet to sulphonylureas and repaglinide. Int J plications in patients with type 2 diabetes 359:2072–2077, 2002 Exp Diabetes Res 2:63–72, 2001 (UKPDS 33). Lancet 352:837–853, 1998 6. Buchanan T, Xiang A, Peters R, Kjos S, 13. Temelkova-Kurktschiev TS, Koehler C, Tan S, Marroquin A, Ochoa C, Goico J, Henkel E, Leonhardt W, Fuecker K, 21. Donahue RP, Abbott RD, Reed DM, Yano Azen S: Prevention of type 2 diabetes by Hanefeld M: Postchallenge plasma glu- K: Postchallenge glucose concentration treatment of insulin resistance: compari- cose and glycemic spikes are more and coronary heart disease in men of Jap- son of early vs. late intervention in the strongly associated with atherosclerosis anese ancestry: Honolulu Heart Program. Diabetes 36:689–692, 1987 TRIPOD study (Abstract). Diabetes 51 than fasting glucose or HbA1c level. Dia- (Suppl. 2):A35, 2002 betes Care 23:1830–1834, 2000 22. Barrett-Connor E, Ferrara A: Isolated 7. Weyer C, Bogardus C, Mott DM, Pratley 14. Starlix [Package Insert]. Novartis Pharma- postchallenge hyperglycemia and the risk RE: The natural history of insulin secre- ceuticals Corporation, East Hanover NJ, of fatal cardiovascular disease in older tory dysfunction and insulin resistance in 2001 women and men: The Rancho Bernardo the pathogenesis of type 2 diabetes melli- 15. Leclercq-Meyer V, Ladriere L, Fuhlen- Study. Diabetes Care 21:1236–1239, tus. J Clin Invest 104:787–794, 1999 dorff J, Malaisse WJ: Stimulation of insu- 1998 8. Mitrakou A, Kelley D, Mokan M, Vene- lin and somatostatin release by two 23. Tominaga M, Eguchi H, Manaka H, Iga- man T, Pangburn T, Reilly J, Gerich J: meglitinide analogs. Endocrine 7: rashi K, Kato T, Sekikawa A: Impaired Role of reduced suppression of glucose 311–317, 1997 glucose tolerance is a risk factor for car- production and diminished early insulin 16. Saloranta C, Hershon K, Ball M, Dickin- diovascular disease, but not impaired fast- release in impaired glucose tolerance. son S, Holmes D: Efficacy and safety of ing glucose: the Funagata Diabetes Study. N Engl J Med 326:22–29, 1992 nateglinide in type 2 diabetic patients Diabetes Care 22:920–924, 1999 9. Hollander PA, Schwartz SL, Gatlin MR, with modest fasting hyperglycemia. J Clin 24. The NAVIGATOR Trial Steering Commit- Haas SJ, Zheng H, Foley JE, Dunning BE: Endocrinol Metab 87:4171–4176, 2002 tee: Nateglinide and valsartan in impaired Importance of early insulin secretion: 17. Glucose tolerance and cardiovasclar mor- glucose tolerance outcomes research: ra- comparison of nateglinide and glyburide tality: comparison of fasting and 2-hour in previously diet-treated patients with diagnostic criteria. Arch Intern Med 161: tionale and design of the NAVIGATOR type 2 diabetes. Diabetes Care 24:983– 397–405, 2001 trial (Abstract). Diabetes 51:A116, 2002 988, 2001 18. Saydah SH, Miret M, Sung J, Varas C, 25. Sartor G, Schersten B, Carlstrom S, Me- 10. Horton ES, Clinkinbeard C, Gatlin M, Gause D, Brancati FL: Postchallenge hy- lander A, Norden A, Persson G: Ten-year Foley J, Mallows S, Shen S: Nateglinide perglycemia and mortality in a national follow-up of subjects with impaired glu- alone and in combination with metformin sample of U.S. adults. Diabetes Care 24: cose tolerance: prevention of diabetes by improves glycemic control by reducing 1397–1402, 2001 tolbutamide and diet regulation. Diabetes mealtime glucose levels in type 2 diabetes. 19. The DCCT Research Group: The effect of 29:41–49, 1980

2146 DIABETES CARE, VOLUME 25, NUMBER 12, DECEMBER 2002