new agents for type 2 diabetesnew agents for type 2 diabetes malcolm nattrass bsc, mb, chb, phd,...

New agents for Type 2 diabetes

Malcolm Nattrass BSc, MB, ChB, PhD, FRCP, FRCPath

Consultant PhysicianDiabetes Resource Centre, Selly Oak Hospital, Birmingham, UK

Cli�ord J. Bailey BSc, PhD, FRCPE, FRCPath

Head of Diabetes Research and Senior LecturerDepartment of Pharmaceutical Sciences, Aston University, Birmingham B4 7ET, UK

Current agents for the treatment of Type 2 diabetes mellitus improve the metabolic pro®lebut do not reinstate normality. They also reduce chronic diabetic complications, but they donot eliminate them. Thus, new agents with novel actions are required to complement andextend the capabilities of existing treatments. Insulin resistance and beta-cell failure, which arecrucial components in the pathogenesis of Type 2 diabetes, remain the underlying targets fornew drugs. Recently introduced agents include a short-acting non-sulphonylurea insulin-releaser, repaglinide, which synchronizes insulin secretion with meal digestion in order toreduce post-prandial hyperglycaemia. The thiazolidinedione drugs, troglitazone, rosiglitazoneand pioglitazone represent a new class of agonists for the nuclear receptor peroxisomeproliferator-activated receptor-gamma (PPARg). PPARg increases the transcription of certaininsulin-sensitive genes, thereby improving insulin sensitivity. The intestinal lipase inhibitororlistat and the satiety-inducer sibutramine are new weight-reducing agents that may bene®tglycaemic control in obese Type 2 diabetes patients. Several further new insulin-releasingagents, and agents to retard carbohydrate digestion and modify lipid metabolism stand poisedto enter the market. The extent to which they will bene®t glycaemic control remains to beseen. However, the prospect of permanently arresting or reversing the progressivedeterioration of Type 2 diabetes continues to evade therapeutic capture.

Key words: anti-diabetic drugs; insulin resistance; sulphonylureas; metformin; acarbose;thiazolidinediones.

Type 2 diabetes mellitus is a most di�cult disease to de®ne. The phenotype di�erswidely at presentation, from a slim patient with mild symptoms and complications atdiagnosis, to the obese patient without symptoms or the obese patient with acute andintense symptoms including signi®cant weight loss. It is unclear whether thesepresentations represent di�erent genotypes, although it would seem di�cult topersevere with the view that Type 2 diabetes is a homogeneous entity.

Between genotype and phenotype, translating the former into the latter, we havethe pathogenesis of the condition, and it is fortunate that a modest degree ofhomogeneity of ®ndings and views exists in this area. Few would disagree with thegenerality of statement that the pathogenesis of Type 2 diabetes receives acontribution from de®cient insulin secretion and impaired insulin action.

1521-690X/99/020309+21 $12.00/00 *c 1999 Harcourt Publishers Ltd.

BaillieÁ re's Clinical Endocrinology and MetabolismVol. 13, No. 2, pp. 309±329, 1999

8

THE PATHOGENESIS OF TYPE 2 DIABETES

The pathogenesis of the condition is considered here only as a means of identifyinga logical approach to drug treatment. What follows is in no way a comprehensivereview of the pathophysiology of Type 2 diabetes.

Insulin secretion in Type 2 diabetes

Insulin secretion is the end result of a complex cascade of actions within the beta-cell.Nutrient metabolism by the beta-cell increases ATP concentrations, which results inclosure of the ATP-sensitive potassium channels. The decrease in potassium ion e�uxleads to depolarization of the cell membrane, allowing the voltage-dependent in¯ux ofcalcium ions. This in turn leads to the extrusion of insulin. There is no evidence tosuggest that this series of steps required for insulin secretion di�ers qualitativelybetween normal subjects and Type 2 diabetes patients. Quantitatively, however, thereis a major di�erence in the relation of stimulus to resultant secretion.

The interpretation of basal insulin levels and insulin secretion in response to stimulisuch as oral or intravenous glucose in Type 2 diabetes patients is problematical. Some®ndings of elevated insulin levels in patients with Type 2 diabetes are not consistentwith studies using highly speci®c insulin assays.1,2 Much of the apparent insulinmeasured by the non-speci®c antibodies used in radio-immunoassays can be accountedfor by pro-insulin or split pro-insulin products, as shown when speci®c immunoradio-metric assays are employed. Indeed, this reveals a particular beta-call defect in pro-insulin processing, which can be identi®ed in Type 2 diabetic patients and subjects athigh risk of diabetes.

If intravenous glucose is used as a stimulus to insulin secretion, the situation issomewhat more clear. The normal response to intravenous glucose is biphasic.3 InType 2 diabetes, ®rst-phase insulin secretion is lost, but the second phase may benormal or even exaggerated. An inadequate ®rst-phase insulin release inevitablyresults in hyperglycaemia and thus a greater stimulus to second-phase secretion. At alater stage of the disease, the second-phase response also becomes impaired.

Impaired insulin secretion could arise from either beta-cell dysfunction or reducedbeta-cell mass. In animals, more than 85% of the islet tissue must be destroyed to giveinsulinopaenic diabetes, and it is likely therefore that both lesions co-exist. Of interestis the increasing recognition that hyperglycaemia per se may impair insulin secretion.Animal studies show islet damage following prolonged exposure to hyperglycaemia,which can be prevented by treatment4, and several studies show an improved insulinresponse after the correction of hyperglycaemia.

Insulin action in Type 2 diabetes

The ®rst step in any action of insulin is its binding to a speci®c receptor.5 There followsinstigation and ampli®cation of the signal via second-messenger pathways that controlthe various membranal, cytoplasmic and genomic actions of insulin. Crucial for signaltransmission is the activation of insulin receptor tyrosine kinase activity and theautophosphorylation of tyrosine residues in the receptor. Several intracellular proteinshave been implicated as insulin receptor substrates, the phosphorylation of whichdirects the insulin signal into di�erent intracellular pathways.

310 M. Nattrass and C. J. Bailey

Decreased insulin action (insulin resistance) in diabetic patients was noted long agoby Himsworth.6 More recent techniques for studying this problem ± forearmperfusion, quadruple infusions of glucose, insulin, adrenaline and propanalol, and theeuglycaemic hyperinsulinaemic clamp technique ± produce similar results. It is clearthat in established Type 2 diabetes, there is insulin resistance, which is manifest asinadequately restrained hepatic glucose production, impaired peripheral glucoseuptake and even increased lipolysis.

Glucose-mediated glucose uptake into cells is an interesting concept that hasbeen relatively underexplored. By de®nition, glucose-mediated (insulin-independent)glucose uptake cannot be insulin resistant, yet it may be impaired in Type 2 diabetes.

The conclusion that insulin resistance is a feature of obesity, Type 2 diabetes andimpaired glucose tolerance, as well as a host of other physiological and patho-physiological conditions, is overwhelming.7 The precise lesions that interrupt theintracellular signalling cascades in the majority of Type 2 diabetic patients are as yetunclear.

Causes of insulin resistance

It is important at this stage to consider, however brie¯y, causes of insulin resistance,since they might point to logical targets for therapy. Currently available evidenceindicates that most presentations of insulin resistance in Type 2 diabetes aremultifactorial in origin. Inherited susceptibility is conferred through the geneticallydetermined levels of expression of signalling components within the pathways ofinsulin action within insulin-sensitive cells. This susceptibility to decreased insulinsensitivity is aggravated by environmental factors including diet (e.g. high fat) and asedentary lifestyle. Insulin resistance results when the collective impositions of geneticsusceptibility and environmental factors produce severe reductions in one or moresignalling pathways, seriously compromising the biological actions of insulin.

Metabolites

The concept that certain circulating metabolites cause insulin resistance has beenrecognized for many years since the pioneering work of Randle et al.8 They observedthat the addition of fatty acids or ketone bodies to the perfusion media of rat heartssigni®cantly reduced glucose oxidation by the heart. This led to the suggestion of acycle in which a fall in the blood glucose concentration, as occurs in starvation, leads tofatty acid mobilization from adipose tissue. The increase in fatty acid mobilization willlead to increased fatty acid oxidation and decreased glucose oxidation. Thus, in thiscompetitive environment of fatty acids and glucose, it would seem a logical approachto attempt to enhance glucose uptake and oxidation by the manipulation of fattyacids, either decreasing their supply to the liver or reducing their intra-mitochondrialmetabolism.

Hormones

The hormonal antagonism of insulin action induces insulin resistance. The primecounter-regulatory hormones ± catecholamines, glucagon, cortisol and growthhormone ± are clearly capable of insulin antagonism to the extent that diabetesoccurs or diabetic control is severely compromised. Other hormones, for example,thyroxine, prolactin, ACTH and vasopressin, may have actions opposing insulin when

New agents for Type 2 diabetes 311

secreted in excess or when conditions allow. The manipulation of one or morehormones for therapeutic advantage is, however, fraught with di�culty and danger,and has been little explored.

TARGETING TREATMENT IN TYPE 2 DIABETES

An ideal approach to the treatment of patients with Type 2 diabetes would be totarget the underlying defects in the pathophysiology of the condition. Later in thischapter, we discuss potential oral hypoglycaemic agents that are being developed withthis aim in mind.

Drug development in the second half of the 20th century has, however, tended notto follow this entirely logical approach but might be best described as serendipitous.Currently available treatments for Type 2 diabetes are not exempt from this. Thus,sulphonylureas, with their insulin stimulatory and extra-pancreatic e�ects, wereintroduced at a time when the pathophysiology of Type 2 diabetes was less clear and,indeed, the disease was regarded as a somewhat milder form of Type 1 diabetes.Metformin, with a glucose-lowering e�ect achieved at the same time as lowering thecirculating insulin level, was used long before there was the substantial weight ofevidence pointing to insulin resistance as part of the pathophysiology.

An alternative approach, more suitable for the pragmatist, is to identify thebiochemical picture, which is the end result of the pathogenesis, and targets therapy atany abnormalities. Of course, the biochemical abnormalities of diabetes are legion andextend far beyond glucose metabolism, but it is to that area of abnormality that wewill restrict ourselves.

It is readily apparent that when glucose pro®les are performed, two majorabnormalities are detected (Figure 1). First, the fasting plasma glucose level isabnormally high in Type 2 diabetes, and this is well recognized as part of the diagnosticcriteria of diabetes and as a target for treatment. Indeed, in the UK ProspectiveDiabetes Study (UKPDS), the fasting plasma glucose was the prime abnormalityaround which changes in treatment were made.9 Second, and somewhat under-emphasized, however, is the abnormality of an increased glucose excursion after meals.In terms of glucose control, these abnormalities represent targets for the developmentof e�ective treatments.

ESTABLISHED AGENTS

The progressive nature of Type 2 diabetes and the pursuit of good glycaemic controltypically require a succession of pharmacological interventions. When dietary modi®-cation fails to control glycaemia, the remorseless progression begins from a single oralhypoglycaemic agent, via the addition of a second or more agents, to insulin giveneither alone or in combination with an oral agent (Figure 2). Given the heterogeneityof Type 2 diabetes and the consequent multifactorial pathophysiology, there is clearly aplace for the use of di�erent agents directed at the di�erent manifestations of thedisease. The main agents currently available and their actions are summarized inTable 1.

Unfortunately, none of the currently available agents is ideal (Table 2). The bloodglucose-lowering e�ect of these agents is constrained by the natural history of thecondition. An extensive loss of beta-cell mass, function or both can severely limit the


Figure

1.Bloodglucose

concentrationthroughouttheday

in23

norm

al(non-diabetic)volunteers,12

Type2diabeticpatientsondietalone,

12Type2diabeticpatients

takingasulphonylurea,and212Type2diabeticpatients

takingmetform

inwithorwithoutasulphonylurea(Su).


Figure 2. A typical algorithm for the treatment of Type 2 diabetes mellitus.

Table 1. Actions of agents used in the treatment of Type 2 diabetes mellitus.

Agent Actions

Acarbose Slow rate of carbohydrate digestionSulphonylureas and repaglinide Stimulate insulin secretionMetformin Reduce insulin resistanceInsulin Increase peripheral glucose utilization

Decrease hepatic glucose output

Table 2. Some limitations of current treatments for Type 2 diabetes mellitus.

Agent Blood glucose-lowering e�cacy Side-e�ects and exclusions

Acarbose Mainly post-prandial, relativelymodest e�ect

Poor tolerability due togastrointestinal reactions

Sulphonylureas Generally good Hypoglycaemia, weight gainRepaglinide Mainly post-prandial Possible hypoglycaemiaMetformin Generally good, potential advantages

against syndrome XPoor renal and hepatic function, andany predisposition to hypoxic diseaseexcluded: risk of lactic acidosis

Insulin Good Hypoglycaemia, weight gain,injections required


e�ectiveness of oral agents. Most tellingly, they do not reinstate an entirely normalmetabolic state, and their use does not lead to the elimination of microvascularcomplications, even when used `intensively'.9±11 For example, in the UKPDS, morethan 50% of patients intensively treated with either an oral agent or insulin showedpoor control (a fasting plasma glucose level of over 7.8 mmol/l) within 6 years ofdiagnosis.12 Moreover, despite intensive drug treatment, 50% of patients had eitherdied or experienced a serious clinical event such as a myocardial infarction, stroke,heart failure, amputation, blindness or proliferative retinopathy within 14 years.9

While hyperglycaemia was initially reduced by the introduction of intensivetherapeutic regimens, no treatment prevented the relentless gradual deteriorationin glycaemic control, believed to re¯ect the continued decline of the beta-cell. There isthus an urgent requirement for new agents that will alter the natural history of thedisease.

RECENTLY INTRODUCED AGENTS

Drugs recently introduced for the treatment of diabetes are glimepiride, repaglinideand troglitazone (Table 3). In addition, two new anti-obesity agents (orlistat andsibutramine) have been introduced, which may be relevant to the treatment of Type 2diabetes. Further agents, namely rosiglitazone, pioglitazone and miglitol are due to beintroduced in the USA in 1999.

Glimepiride

The sulphonylurea glimepiride (Figure 3) is a preparation designed to be taken oncedaily, having its maximum plasma glucose-lowering e�ect within about 4 hours.13,14

Glimepiride shows similar e�cacy to glibenclamide, gliclazide and glipizide, reducingthe HbA1c level by 1±2% (absolute) compared with placebo. However, glimepirideprovokes a less marked stimulation of insulin secretion proportional to its e�ect uponplasma glucose, leading to the suggestion of a more potent extra-pancreatic e�ect.Thus, glimepiride enhanced insulin-stimulated glucose transport, glycogenesis andlipogenesis in isolated adipocytes associated with an increased translocation of GLUT-4glucose transporters into the plasma membrane.14 Nevertheless, the main site ofaction of glimepiride is the pancreatic beta-cell, where the drug initiates insulinsecretion by binding to the sulphonylurea receptor (SUR). This causes closure of the

Table 3. Oral agents recently introduced for the treatment of Type 2 diabetes mellitus* .

Year introduced Agent Action

1996 Glimepiride Medium-to-long-acting sulphonylurea,stimulates insulin secretion

1997 Troglitazone Long-acting PPARg agonist, enhances insulinsensitivity

1998 Repaglinide Rapid, short-acting insulin-releaser to controlpost-prandial hyperglycaemia

* Additional new agents that have a place in the treatment of Type 2 diabetes are the anti-obesity agentsorlistat and sibutramine, and the rapid, short-acting insulin analogue Lys-Pro insulin.{Available in the USA and Japan.


ATP-sensitive potassium channel, depolarization and voltage-dependent calcium ionin¯ux, with the activation of calcium-dependent regulatory proteins that directexocytosis.

Glimepiride binds predominantly to a di�erent part of the SUR from glibenclamide,that may reduce any e�ect on the potassium channels in vascular tissue.15 Tolerabilityis good, and the drug has been reported to cause fewer hypoglycaemic episodes thanglibenclamide, especially during the initiation of therapy.

Repaglinide

Repaglinide is a member of the meglitinide family of benzoic acid derivatives,developed from the non-sulphonylurea moiety of glibenclamide (see Figure 3 above). It

Figure 3. Structure of the insulin-releasing agents glimepiride, glibenclamide (glyburide), meglitinide,repaglinide, nateglinide (A4166) and KAD-1229.


is designed to be taken immediately before a meal so that insulin secretion coincideswith the time of nutrient absorption, and its insulin secretory e�ect is of rapid onsetand short duration.16 The drug binds to the SUR on pancreatic beta-cells, probably atthe same site as, and also at a di�erent site from, glibenclamide, to give a potentinsulin-releasing e�ect.17,18 The prompt, yet brief time course of action results from itspharmacokinetic properties. Absorption is quick (peak concentration being achievedin less than 1 hour, with a bio-availability of more than 80%) and is followed byfast elimination (plasma half-life being less than 1 hour), mainly through hepaticmetabolism and biliary excretion.

Preliminary reports of clinical trials in Type 2 diabetes indicate that repaglinide(0.5±4 mg with each main meal) reduces fasting plasma glucose concentration, reducesthe HbA1c level and, in addition, reduces post-prandial glucose excursions.19 This lattere�ect, which is not seen with other sulphonylureas to a comparable extent, has led tothe claim of a new class of drug, the post-prandial glucose regulator. Overall, however,the daily glucose-lowering e�ect of repaglinide is similar to that of glibenclamide.16

Like the sulphonylureas, repaglinide exerts an additive glucose-lowering e�ect withmetformin. The drug is well tolerated, and its short duration of action points to lesshypoglycaemia and particular value in those patients with irregular meal patterns.

The full potential of a rapidly acting, short-duration insulin-releaser has yet to befully realized. The obvious comparison is with the four times daily, basal-bolus regimenfor insulin in Type 1 diabetes. Against this lies a wealth of clinical experience thathighlights the di�culty of persuading patients to take drugs three times daily.

Troglitazone

The thiazolidinedione drug troglitazone (Figure 4) is the ®rst of a new class of oralagents that selectively enhance or partially mimic certain actions of insulin by

Figure 4. Structure of the thiazolidinediones: troglitazone, rosiglitazone and pioglitazone.


activating the peroxisome proliferator-activated receptor-gamma (PPARg).20,21 PPARgis a nuclear receptor expressed mainly in adipose tissue but to a small extent inskeletal muscle, liver and other tissues. It acts in a complex with the retinoid Xreceptor to increase the transcription of several insulin-sensitive genes. These includelipoprotein lipase, the fatty acid transporter protein, adipocyte fatty acid-bindingprotein, acyl CoA synthetase, malic enzyme and the glucose transporter isoformGLUT-4. Accordingly, thiazolidinediones increase the uptake of glucose and fatty acidsby adipocytes, promoting lipogenesis and adipogenesis (Figure 5). They also increaseglucose uptake, glycogenesis and glucose utilization by muscle tissue, and may reduceglucose production by the liver.22

Troglitazone exerts a slowly generated blood glucose-lowering e�ect in Type 2diabetes. A single daily dose titrated up to 600 mg achieves a maximum reduction inblood glucose by 2±3 months.23 As a monotherapy, this is typically modest, decreasingthe HbA1c level by less than 1% (absolute).24 Thus, monotherapy with this agent is notwidely recommended as a ®rst choice. E�cacy appears to be greater in patients whostill secrete signi®cant amounts of insulin, for example those with a fasting plasma C-peptide level of more than 1.5 ng/ml. Indeed, when troglitazone is given incombination with a sulphonylurea, insulin or metformin, it has an additive e�ectwith the other therapy, thus achieving greater e�cacy.22,25

A studyof Type 2 patients on insulin (average 73 units per day) found that the additionof 200 and 600 mg per day troglitazone for 6 months reduced the HbA1C level frommore than 9% to less than 8% in 30% and 57% of patients respectively (Figure 6). At thesame time, the insulin dose was decreased (by 11% in the 200 mg group and 29% in the600 mg group) in 15% and 42% of patients respectively.22 Although troglitazone is proneto causing weight gain, similar in extent to the sulphonylureas and insulin, it reducesserum concentrations of triglyceride, non-esteri®ed fatty acids (NEFAs), and insulin. A

Figure 5. Diagramatic representation of cellular mechanism of action of thiazolidinediones on an adipocyte.TZD � thiazolidinedione; PPARg � peroxisome proliferator-activated receptor-gamma; RXR � retinoid Xreceptor; GLUT-4 � glucose transporter isoform 4; FATP � fatty acid transporter protein; aP2 � adipocytefatty acid binding protein; LPL � lipoprotein lipase.


small increase in high-density lipoprotein (HDL) cholesterol and low-density lipoprotein(LDL) cholesterol may also occur, but the LDL-to-HDL ratio is generally unaltered.The increase in LDL cholesterol may be associated with an increase in particle size,which carries a lower cardiovascular risk. Preliminary reports have suggested thattroglitazone may increase the concentration of lipoprotein (a), an independent riskfactor for coronary heart disease. Conversely, the alpha-tocopherol moiety (vitamin E)of troglitazone could o�er cardioprotective anti-oxidant properties.22

Pre-clinical studies with several thiazolidinediones have observed haemodilution,cardiomegaly and anaemia, which have not surfaced as a problem in the clinical use oftroglitazone.26 In contrast, markedly raised liver enzymes and fatal episodes of hepaticfailure have emerged as the serious adverse event with troglitazone use. Raised serumalanine aminotransferase (ALT/SGPT) to more than 3 times the upper limit of normaloccurs in about 2% of patients, and greater than 1.5 times the upper limit of normal inabout 5% of patients taking troglitazone. This is most likely to occur within the ®rst6 months of therapy, and generally appears to be reversible if the problem is identi®edand the drug withdrawn.27 Unfortunately, jaundice and severe hepatic failure candevelop and, at the time of writing, 18 fatalities have been reported from 1.1 millionpatients starting troglitazone therapy during the ®rst year of its use in the USA. Fromthe case reports available, there is usually extensive hepatocellular damage withoutsigni®cant cholestasis. There is no apparent association with the drug's hepaticmetabolites or with hepatitis A, B or C.22

The hepatotoxicity represents an idiosyncratic response, and it is therefore notpossible to identify patients at risk in advance. Clearly, immediate discontinuation ofthe drug is needed by those in whom a rise in liver enzymes is observed, and strictmonitoring is mandatory. In the USA, the drug is contra-indicated in patients with abase-line ALT level of more than 1.5 times the upper limit of normal, ALT beingmonitored monthly for 8 months, bimonthly up to 1 year and periodically thereafter.Patients with an ALT level of 1.5±2.0 times the upper limit of normal should bemonitored weekly and the drug stopped if there is any deterioration. It should beborne in mind that some of the patients suited to this drug are likely to have somedegree of fatty liver and minor liver enzyme abnormalities.

Recent reports suggest that thiazolidinediones stimulate PPARg in the colonicepithelium, thus increasing the frequency of colonic tumours in genetically susceptiblemice. Conversely, the stimulation of PPARg reduced the growth of human coloncancer cells in vitro and after transplantation into nude mice, leaving this issueunresolved.28

Figure 6. E�ect of 6 months' treatment with troglitazone (200 or 600 mg once daily) in combination withinsulin on HbA1c level in a multicentre trial with 350 poorly controlled insulin-treated Type 2 diabeticpatients. Base-line indicates the value on insulin only before the addition of troglitazone. *P5 0.001compared with base-line.


Weight-reducing agents

Successful weight-management programmes for obese Type 2 diabetic patients arenotoriously di�cult to achieve and are frequently abandoned in frustration. Never-theless, there is convincing evidence that even modest weight reduction will increaseinsulin sensitivity, improve glycaemic control, improve lipid pro®les and generallyenhance prognosis.29 Two new anti-obesity agents may assist weight reduction indiabetic patients when used in conjunction with a prudent energy-restricted diet. Theycan also be used in combination with conventional anti-diabetic drugs.

Orlistat

Orlistat (Figure 7) is a gastrointestinal lipase inhibitor that binds to the active site oflipases, reducing triglyceride hydrolysis. This slows the digestion of dietary fat, causinga reduction by up to about 30% in the amount of fat absorbed.30

In a 1-year clinical trial, dietary restriction plus orlistat increased weight loss beyondthat achieved by energy restriction alone by about 4 kg in obese non-diabeticsubjects31, but by only about 2 kg in diabetic patients.32 However, the overall weightloss (over 6 kg) with diet and orlistat a�orded a useful improvement in glycaemiccontrol (a decrease in HbA1c level of more than 0.8% absolute). Moreover, in patientswith an HbA1c level of more than 8%, orlistat decreased HbA1c by 0.5% absolute.Orlistat also produced a small reduction in LDL cholesterol and triglyceride, as well asreducing waist circumference by 2.8 cm.32

An increased elimination of fat in the faeces is inevitably associated with someunpleasant gastrointestinal events, including frequent defaecation, loose oily stools,diarrhoea and sometimes even faecal incontinence. These make the drug unacceptableto some patients. Minimizing the side-e�ects requires the patient to maintain areduced fat intake. Although vitamin de®ciencies during therapy have not beenreported, it must be borne in mind that orlistat can potentially impair the absorption

Figure 7. Structure of the lipase inhibitor orlistat and the serotonin±noradrenaline re-uptake inhibitorsibutramine.


of fat-soluble vitamins. It can also alter the absorption of other medication, with aconcomitant need for dose adjustment.

Sibutramine

Sibutramine (see Figure 7) is a serotonin and noradrenaline re-uptake inhibitor thatacts centrally to enhance the post-ingestive satiety response.33 In rodents, sibutraminealso increases energy expenditure through thermogenesis, via an enhancedsympathetic stimulation of brown adipose tissue.34

One-year clinical trials in obese, non-diabetic subjects have shown that sibutramine(15 mg per day) increases weight loss by 4±5 kg beyond that obtained with an energy-restricted diet. There is a corresponding reduction in waist circumference andreduction in LDL cholesterol and triglyceride.35

Preliminary observations in obese Type 2 diabetic patients indicate an improvementin glucose tolerance with a decrease in the HbA1c level of about 0.5% (absolute) withsibutramine.36 It should be recognized that sibutramine can cause a small increase inheart rate and blood pressure, but it is not addictive and has no known e�ect upon theheart valves or pulmonary vascular function.

DESIGNING NEWAGENTS

The speci®c cellular lesions responsible for insulin resistance and beta-cell failure inmost `common' forms of Type 2 diabetes remain to be clari®ed. Multiple aetiologieswith consequent multiple pathogeneses are to be anticipated. Insulin resistanceappears to emerge through the collective in¯uence of several distinct disturbances inthe cellular signalling pathways for insulin action.37 Beta-cell failure involves defectiveglucose sensing and impaired signal coupling to insulin biosynthesis and secretion inindividual cells, as well as alterations in the turnover of the beta-cell population. Thus,a single, universally e�ective therapeutic agent is deemed unlikely. This in turnindicates that future approaches may utilize a number of therapeutic agents early inthe disease rather than follow the sequence alluded to earlier of a single agentprogressing to combination therapy late in the disease. Future developments willtherefore be aimed at the di�erent lesions of Type 2 diabetes or, where they remainunidenti®ed, at the phenotypic expression of the underlying defect.

By the time of diagnosis of Type 2 diabetes, insulin resistance is usually wellestablished, the ®rst-phase insulin response to glucose has been lost, pro-insulin±insulin processing is defective, and other features of impending beta-cell failure areevident.38,39 Many of the cellular aberrations are su�ciently advanced that they canonly be arrested or partially reversed but not normalized. Thus, the desirableproperties of a new agent to restore normal metabolic control and mitigate diabeticcomplications (Table 4) must be tempered with the realization that some cellular faultsin overt Type 2 diabetes are probably irreparable.

Table 4. Desirable properties of new agents to treat Type 2 diabetes.

Blood glucose-lowering e�cacyBene®t other components of syndrome XTreat fundamental defects of the diseaseComplementary to existing agentsSafe, well tolerated and conducive to compliance


From a practical point of view, agents are required that are capable of improvingperipheral glucose disposal and reducing (but not completely inhibiting) hepaticglucose output. These goals can be achieved in a number of di�erent ways, for exampleenhancing the insulin sensitivity of these major metabolic processes by increasinginsulin secretion su�ciently to overcome the insulin resistance present, or byindependent insulin-like e�ects.

The somewhat simple solution of raising the insulin level by enhancing its secretionhas a super®cial attraction that is lessened on closer consideration. First, the availableinsulin-releasing agents do not correct the glucose sensor defect and fail to increaseinsulin biosynthesis commensurate with secretion. Second, they fail to ameliorate therelentless decline in beta-cell function or mass that appears to be part of the naturalhistory of Type 2 diabetes.27 Third, there is the problem of increasing insulin concen-tration and the impact of this upon tissues that do not display insulin resistance.37 Also,an attack upon insulin resistance alone is not likely to be overwhelmingly successful.Lipolysis is exquisitely sensitive to inhibition by insulin, and increasing sensitivity islikely to lead to di�culty in fat mobilization and consequently obesity. Thus, enhancinginsulin sensitivity is likely to be counteracted by an increase in obesity. Anotherexample of the treatment dilemma is that insulin increases hepatic triglyceride-richvery low-density lipoprotein synthesis, an excess of which is also a risk factor formacrovascular disease. Accordingly, any new treatments will be required to balancethese issues.

NEW POTENTIAL ANTI-DIABETIC AGENTS

Many new agents have been identi®ed that could potentially reduce glucotoxicity andat least partly address the fundamental problems of beta-cell failure and insulinresistance. While some are essentially modi®cations of existing drugs, others representnovel mechanistic approaches, but, to date, none would appear to o�er an ideal orcomplete solution to the progressive nature of the disease process.40±42

Improving beta-cell function

Two short-acting insulin-releasers are well advanced in development: A4166(nateglinide) and KAD-1229 (see Figure 3). Like repaglinide, their pharmacokineticproperties enable their rapid absorption and elimination. They act directly upon theSUR of the beta-cell membrane, causing closure of the ATP-sensitive potassiumchannels (Figure 8).43,44

Another short-acting insulin-releaser is the morpholinoguanidine BTS 67 582, thatmay act directly or indirectly on the SUR (Figure 8).45

Agents that stimulate insulin secretion by actions at the beta-cell membrane that aredistal to glucose metabolism and ATP production do not appear to rectify the defect ofglucose sensing and do not enhance insulin biosynthesis. Succinate esters, whichstimulate energy production by beta-cells, have been shown to increase both thebiosynthesis and secretion of insulin, but they have widespread e�ects uponmetabolism throughout the body, which are likely to limit their application.46

Glucagon-like peptide 1 (GLP-1) 7±36 amide is an intestinal hormone that acts onthe beta-cell via a cyclic AMP-dependent mechanism to enhance glucose-inducedinsulin secretion and biosynthesis (Figure 8). While GLP-1 provides an opportunity toamplify the e�ects of glucose on the beta-cell, it requires parenteral delivery, and its


short biological half-life has limited its therapeutic potential during clinical trials.47,48

Novel routes of GLP-1 delivery, such as sublingually or intranasally, agents that restrictGLP-1 degradation (e.g. dipeptidyl-peptidase IV inhibitors), and non-peptide GLP-1receptor agonists are being explored as possible options.49 Whether GLP-1 exertsextra-pancreatic insulin-like e�ects in man is still unclear, and it remains uncertainwhether there would be any clinically signi®cant e�ect upon satiety and gastricemptying.

Other putative approaches to the ailing beta-cell include the use of a2-adrenoceptorantagonists to relieve the tonic sympathetic suppression of insulin secretion, andinhibitors of type III phosphodiesterases to enhance cyclic AMP.41,50 Achieving aselective e�ect on the beta-cell has to date thwarted their application.

Figure 8. Schematic representation of a pancreatic beta-cell showing the main pathways controlling pro-insulin biosynthesis and insulin secretion. The sites of action of insulin-releasing agents are identi®ed.PKA � protein kinase A; CAMP � cyclic adenosine monophosphate; ATP � adenosine triphosphate;IP3 � inositol trisphosphate; PLC � phospholipase C; PKC � protein kinase C; GLP-1 � glucagon-likepeptide-1; PDE � phosphodiesterase; SUR � sulphonylurea receptor; Kir � potassium inward recti®er;GLUT-2 � glucose transporter isoform 2; CCK�cholecystokinin; GIP�gastric inhibitory polypeptide.


Improving insulin action

The discovery that thiazolidinediones stimulate PPARg, which in turn activates insulin-sensitive genes22, has led to a succession of PPARg agonists designed to improve insulinsensitivity. Both rosiglitazone and pioglitazone are advanced in development (seeFigure 4)51±53, and related PPARg agonists such as the isoxazolidinedione JTT-501 areunder consideration.54 From the published information available, these agents appearto have qualitatively similar e�ects upon glucose metabolism to those of troglitazone,although there are considerable di�erences in potency. They have all been shown toreduce hyperglycaemia, hyperinsulinaemia and triglyceride concentrations in obesediabetic and glucose-intolerant animal models. However, it is not clear whether all thee�ects are consequent to the activation of PPARg or whether some e�ects dependupon the non-thiazolidinedione moiety of the molecule.

The range of PPARg agonists under development o�ers a selection of agents withdi�ering pharmacokinetic features to provide choice in this approach to the treatmentof insulin resistance. Preliminary data suggest that PPARg agonists generally provideadditive blood glucose-lowering e�cacy in combination with other classes of oralagents, and reduce the insulin requirement.22 The impact upon body weight and theadverse event pro®les await the test of widespread clinical usage.

Vanadium and trivalent chromium salts have been shown to improve glycaemiccontrol in animal models and in clinical trials. Vanadium salts are phosphataseinhibitors that may enhance the action of insulin by slowing the dephosphorylation andinactivation of insulin receptors after insulin binding. Interestingly, vanadium salts havebeen shown to reduce hepatic glucose output and increase glucose disposal in Type 2diabetes for several weeks after their administration has been discontinued.55,56 Aquestion mark over toxicity has stimulated research into peroxo-vanadium salts with ahigher potency that can be used in a lower dosage.57 Trivalent chromium has longbeen associated with improved glucose tolerance, but its mode of action is unclear.Nevertheless, its e�cacy has been clearly demonstrated in several studies of patientswith Type 2 diabetes.58

Insulin-like growth factor-1 (IGF-1) has a low a�nity for the insulin receptor and canweakly activate the post-receptor steps of the insulin signalling pathway via itsinteraction with the IGF-1 receptor.59 While this might not bene®t `common'presentations of Type 2 diabetes, it might assist patients with severe congenital insulinresistance caused by defects of the insulin receptor or in the early signalling steps.

Slowing carbohydrate absorption

Diets rich in natural vegetable ®bre play an important part in the diabetic diet. Theydelay carbohydrate digestion by impeding the accessibility of digestive enzymes andreducing sugar di�usion to the surface of the intestinal mucosa. Fibre supplementssuch as guar gum have been reported to exaggerate these e�ects, although poorpalatability and malabsorption reduce the tolerability of this approach.

Several new agents have been developed that compete with disaccharides andoligosaccharides for alpha-glucosidase enzymes in the brush border of intestinalepithelial cells.60 By inhibiting the action of the alpha-glucosidases, they slow the rateof carbohydrate digestion. The subsequent delay in glucose absorption smooths theentry of glucose into the circulation and reduces post-prandial glycaemia. New agentsin this class, namely miglitol and voglibose, are in the image of acarbose (Figure 9).61,62

Their e�cacy as monotherapy is generally held to be modest, but they may be used in


combination with any of the other classes of anti-diabetic agent. Drug titration iscritical to minimize the side-e�ect of excessive ¯atulence.

The amylin analogue pramlintide has been shown to slow the rate of gastricemptying and thereby reduce post-prandial hyperglycaemia.63 It may also inducesatiety. As it is a peptide, it needs to be delivered by subcutaneous injection, and itscoincident use with insulin is under investigation.

Modifying lipid metabolism

Adipose tissue is normally exquisitely sensitive to the anti-lipolytic e�ect of insulin.Insulin resistance allows an inappropriate release of NEFAs. These are used as anenergy source by muscle and liver, sparing glucose consumption by muscle (the Randlecycle) and promoting hepatic gluconeogenesis. Hence disturbance of lipid metabolismhas important consequences for hyperglycaemia.

Attempts to target NEFA reduction as a means of reducing hyperglycaemia haveproved attractive but so far elusive. Anti-lipolytic agents such as acipimox (Figure 10)have shown some promise, but more potent and longer-acting derivatives are requiredbefore it can be ascertained with certainty that this approach will be successful.64 Oneconfounding factor in this approach is the much-neglected hierarchy of insulin action.37

In attempting to link the suppression of lipolysis with the enhancement of glucoseuptake into muscle, it must be borne in mind that the e�ects of insulin upon lipolysisand the e�ects of insulin upon glucose oxidation by muscle occur at vastly di�erent

Figure 9. Structure of the alpha-glucosidase inhibitors acarbose, miglitol and voglibose.


circulating insulin concentrations. Thus, it may not be entirely logical to involve insulinto attempt to link the two processes when devising therapies.

Lipid-lowering ®brates also reduce plasma NEFA levels su�ciently to confer a smallimprovement in glycaemic control, but this is rarely of signi®cant clinical value.65

Agents that reduce the synthesis of fatty acids and triglycerides, for exampleben¯uorex and long-chain dicarboxylic acids, may be helpful (Figure 10), and they mayin addition have independent glucose-lowering activity.66,67

Compounds that inhibit fatty acid oxidation have been examined extensively aspotential anti-diabetic agents. The main approach has been to inhibit carnitinepalmitoyl transferase-1, the rate-limiting enzyme for the transfer of long-chain fattyacids into the mitochondrion.68 While this e�ectively reduces hyperglycaemia, it is notreadily reversed and will therefore temporarily prevent gluconeogenesis. Without thesafety net of gluconeogenesis, the patient is vulnerable to hypoglycaemia.

Figure 10. Structure of acipimox, ben¯uorex, medica 16 and CL 316 243.


The stimulation of b3-adrenoceptors in adipose tissue increases lipolysis and reducesadipose mass. In obese-diabetic animal models, this is accompanied by an improvementin glycaemic control. Achieving complete selectivity for the b3-receptor subtype hasproved elusive, and the e�cacy of putative agents similar to CL 316 243 has so farproved modest.69

Other approaches

Other approaches to the control of hyperglycaemia that have been consideredinclude glucagon receptor agonists, agents that directly stimulate glucose metabolismor inhibit gluconeogenic enzymes, and agents that inhibit glucose-6-phosphateactivity.70±73 It is too early to say whether any of these lines are likely to be fruitful.

CONCLUSION

Wemay conclude on a note of optimism that several new agents have recently becomeavailable or are soon to become available, thus extending the range of therapy forType 2 diabetes. While these will undoubtedly o�er an enhanced opportunity toimprove glycaemic control and hence diminish the number of complications, a cloudcontinues to cast its shadow over this ®eld. None of the agents currently available, andprobably none of those that we can foresee, appears to o�er the potential for asustained arrest or reversal of the pathogenic process of Type 2 diabetes once thecondition is established.

REFERENCES

1. Temple RC, Carrington CA, Luzio SD et al. Insulin de®ciency in non-insulin-dependent diabetes. Lancet1989; 1: 293±295.

2. Williams DRR, Byrne C, Clark PMS et al. Raised proinsulin concentration as an early indicator of B celldysfunction. British Medical Journal 1991; 303: 95±96.

3. Porte D & Pupo AA. Insulin responses to glucose: evidence for a two pool system in man. Journal ofClinical Investigation 1969; 48: 2309±2319.

4. Clark A, Brown E, King T et al. Islet changes induced by hyperglycemia in rats: e�ect of insulin orchlorpropamide therapy. Diabetes 1982; 31: 319±325.

5. Cheatham B & Kahn CR. Insulin action and the insulin signalling network. Endocrine Reviews 1995; 16:117±142.

6. Himsworth HP. Diabetes mellitus; its di�erentiation into insulin-sensitive and insulin-insensitive types.Lancet 1936; 1: 127±130.

7. Reaven GM. Pathophysiology of insulin resistance in human disease. Physiological Reviews 1995; 75:473±486.

8. Randle PJ, Garland PB, Hales CN & Newsholme EA. The glucose fatty-acid cycle. Its role in insulinsensitivity and the metabolic disturbances of diabetes. Lancet 1963; 1: 785±789.

9. UK Prospective Diabetes Study Group. Intensive blood glucose control with sulphonylureas or insulincompared with conventional treatment and risk of complications in patients with type 2 diabetes(UKPDS 33). Lancet 1998; 352: 837±853.

10. Panzram G. Mortality and survival in type 2 (non-insulin dependent) diabetes mellitus. Diabetologia 1987;30: 123±131.

11. Hane®eld M, Fischer S, Julius U et al. Risk factors for myocardial infarction and death in newly detectedNIDDM: the Diabetes Intervention Study, 11 year follow-up. Diabetologia 1996; 39: 1577±1583.

12. UK Prospective Diabetes Study Group. UK Prospective Diabetes Study 16. Overview of 6 years therapyof type II diabetes: a progressive disease. Diabetes 1995; 44: 1249±1258.

13. Dills DG & Schneider J. Clinical evaluation of glimepiride versus glyburide in NIDDM in a double-blindstudy. Hormone and Metabolic Research 1996; 28: 426±429.


*14. Langtry HD & Balfour JA. Glimepiride. A review of its use in the management of type 2 diabetesmellitus. Drugs 1998; 55: 563±584.

15. Bijlstra PJ, Lutterman JA, Russel FGM et al. Interaction of sulphonylurea derivatives with vascular ATP-sensitive potassium channels in humans. Diabetologia 1996; 39: 1983±1090.

*16. Graul A & Castaner J. Repaglinide. Drugs of the Future 1996; 21: 694±699.17. Malaisse WJ. Stimulation of insulin release by non-sulfonylurea hypoglycemic agents: the meglitinide

family. Hormone Metabolism Research 1995; 27: 263±266.18. Fuhlendor� J, Rorsman P, Kofod H et al. Stimulation of insulin release by repaglinide and glibenclamide

involves both common and distinct processes. Diabetes 1998; 47: 345±351.19. Wol�enbuttel BHR, Nijst L, Sels JP et al. E�ects of a new oral hypoglycaemic agent, repaglinide, on

metabolic control in sulphonylurea-treated patients with NIDDM. European Journal of ClinicalPharmacology 1993; 45: 113±116.

20. Lehmann JM, Moore LB, Smith-Oliver TA et al. An antidiabetic thiazolidinedione is a high a�nity ligandfor peroxisome proliferator-activated receptor g (PPARg). Journal of Biological Chemistry 1995; 270:12593±12596.

21. Spiegelman BM. PPAR-g: adipogenic regulator and thiazolidinedione receptor. Diabetes 1998; 47:507±514.

*22. Day C. Thiazolidinediones: a new class of antidiabetic drugs. Diabetic Medicine 1999; 16: 1±14.23. Maggs DG, Buchanan TA, Burant CF et al. Metabolic e�ects of troglitazone on monotherapy in type 2

diabetes mellitus. Annals of Internal Medicine 1998; 128: 176±185.24. Kumar S, Boulton AJM, Beck-Nielsen H et al. Troglitazone, an insulin action enhancer, improves

metabolic control in NIDDM patients. Diabetologia 1996; 39: 701±709.25. Inzucchi SE, Maggs DG, Spollett GR et al. E�cacy and metabolic e�ects of metformin and troglitazone in

type II diabetes mellitus. New England Journal of Medicine 1998; 338: 867±872.26. Ghazzi MN, Perez JE, Antonucci TK et al. Cardiac and glycemic bene®ts of troglitazone treatment in

NIDDM. Diabetes 1997; 46: 433±439.27. Watkins PB & Whitcomb RW. Hepatic dysfunction associated with troglitazone. New England Journal of

Medicine 1998; 338: 916±917.28. Seed B. PPARg and colorectal carcinoma: con¯icts in a nuclear family.Nature Medicine 1998; 4: 1004±1005.29. Wing RR, Shoemaker M, Marcus MDM et al. Variables associated with weight loss and improvements in

glycaemic control in type 2 diabetic patients. Archives of Internal Medicine 1990; 147: 1749±1753.30. McNeely W & Ben®eld P. Orlistat. Drugs 1998; 56: 241±249.31. Sjostrom L, Rissanen A, Anderson T et al. Randomised placebo-controlled trial of orlistat for weight loss

and prevention of weight regain in obese patients. Lancet 1998; 352: 167±172.*32. Hollander PA, Kaplan RA, Elbein SC et al. Role of orlistat in the treatment of obese patients with type 2

diabetes. A 1-year randomized double-blind study. Diabetes Care 1998; 21: 1288±1294.33. Jackson HC, Bearham MC, Hutchins LJ et al. Investigation of the mechanisms underlying the hypophagic

e�ects of the 5-HT and noradrenaline reuptake inhibitor, sibutramine, in the rat. British Journal ofPharmacology 1997; 121: 1613±1618.

34. Connoley IP, Heal DJ & Stock MJ. A study in rats of the e�ects of sibutramine on food intake andthermogenesis. British Journal of Pharmacology 1995; 114: 388P.

35. Weintraub M, Rubio A, Goli A et al. Sibutramine in weight control: a dose-ranging, e�cacy study.Clinical Pharmacology and Therapeutics 1991; 50: 330±337.

36. Gri�ths J, Byrnes AE, Frost G et al. Sibutramine in the treatment of overweight non-insulin dependentdiabetics. International Journal of Obesity 1995; 19 (supplement 2): 41.

37. Bailey CJ. Insulin resistance and antidiabetic drugs. Biochemical Pharmacology 1999; 58: (in press).38. DeFronzo RA. The triumvirate: b-cell, muscle, liver. A collusion responsible for NIDDM. Diabetes 1988;

37: 667±687.39. Porte D. b-cells in type II diabetes mellitus. Diabetes 1991; 40: 166±180.40. Bailey CJ, Williams G & Pickup JC. New drugs in the management of diabetes mellitus and its

complications. In Pickup JC & Williams G (eds) Textbook of Diabetes, 2nd edn, pp 84.1±84.30. Oxford:Blackwell, 1997.

41. Bailey CJ. New drugs for the treatment of diabetes mellitus. In Alberti KGMM, DeFronzo RA, ZimmetP & Keen H (eds) International Textbook of Diabetes Mellitus, 2nd edn, pp 865±881. Chichester: JohnWiley& Sons, 1997.

*42. Bailey CJ. New pharmacological approaches to glycemic control. Diabetes Reviews 1999; 7: 94±113.43. Akiyoshi M, Kakei M, Nakazaki M & Tanaka H. A new hypoglycemic agent, A-4166, inhibits ATP-

sensitive potassium channels in rat pancreatic b-cells. American Journal of Physiology 1995; 268: E185±E193.44. Kinukawa M, Ohnota H & Ajisawa Y. E�ect of a non-sulphonylurea hypoglycaemic agent, KAD-1229 on

hormone secretion in the isolated perfused pancreas of the rat. British Journal of Pharmacology 1996; 117:1702±1706.


45. Page T & Bailey CJ. Blood glucose-lowering e�ect of morpholinoguanidine derivative BTS 67 582. BritishJournal of Pharmacology 1997; 122: 1464±1468.

46. Bjorking F, Malaisse-Lagae F & Malaisse WJ. Insulinotropic action of novel succinic acid esters.Pharmacology Research 1996; 33: 273±275.

47. Nauck MA. Glucagon-like peptide 1. Current Opinion in Endocrinology and Diabetes 1997; 4: 291±299.*48. Drucker DJ. Glucagon-like peptides. Diabetes 1998; 47: 159±169.49. Pederson RA, White HA, Schlenzig D et al. Improved glucose tolerance in Zucker fatty rats by oral

administration of the dipeptidyl peptidase IV inhibitor isoleucine thiazolide.Diabetes 1998; 47: 1253±1258.50. Leibowitz MD, Biswas C, Brady EJ et al. A novel insulin secretagogue is a phosphodiesterase inhibitor.

Diabetes 1995; 44: 67±74.51. Stevenson RW, Gibbs EM, Kreutter DK et al. The thiazolidinedione drug series. Diabetes Annual 1995; 9:

175±191.52. Young PW, Cawthorne MA, Coyle P et al. Repeat treatment of obese mice with BRL 49653, a new and

potent insulin sensitizer, enhances insulin action in white adipocytes. Diabetes 1995; 44: 1087±1092.53. Hallakou S, Foufelle F, Doare L et al. Pioglitazone-induced increase of insulin sensitivity in the muscles of

the obese Zucker fa/fa rat cannot be explained by local adipocyte di�erentiation. Diabetologia 1998; 41:963±968.

54. Shibata T, Matsui K, Yonemori F et al. JTT-501, a novel oral antidiabetic agent, improves insulinresistance in genetic and non-genetic insulin resistant models. British Journal of Pharmacology 1998; 125:1744±1750.

55. Cohen N, HalberstamM, Stilimovich P et al. Oral vanadyl sulfate improves hepatic and peripheral insulinsensitivity in patients with non-insulin-dependent diabetes mellitus in vivo and in vitro studies. Journal ofClinical Investigation 1995; 95: 2501±2509.

56. Gold®ne AB, Simonson SC, Folli F et al. Metabolic e�ects of sodium metabanadate in humans withinsulin-dependent and non-insulin-dependent diabetes mellitus in vivo and in vitro studies. Journal ofClinical Endocrinology and Metabolism 1995; 80: 33111±33120.

57. Posner BI, Fraure R, Burgress JW et al. Peroxovandium compounds. A new class of potentphosphotyrosine phosphatase inhibitors which are insulin mimetics. Journal of Biological Chemistry1994; 269: 4596±4604.

58. Anderson RA, Cheng N, Bryden NA et al. Elevated intakes of supplemental chronium improve glucoseand insulin variables in individuals with type 2 diabetes. Diabetes 1997; 46: 1786±1791.

59. Moses AC, Young SCJ, Morrow LA et al. Recombinant human insulin-like growth factor 1 increasesinsulin sensitivity and improves glycemic control in type II diabetes. Diabetes 1996; 45: 91±100.

*60. Lebovitz H. a-Glucosidase inhibitors as agents in the treatment of diabetes. Diabetes Reviews 1998; 6:132±145.

61. Johnston PS, Santiago JV, Con®� RF et al. E�ects of the carbohydrase inhibitor miglitol in sulfonylurea-treated NIDDM patients. Diabetes Care 1994; 17: 20±29.

62. Goke B, Fuder H, Wieckhorst G et al. Voglibose (AO-128) is an e�cient a-glucosidase inhibitor andmobilizes the endogenous GLP-1 reserve. Digestion 1995; 56: 493±501.

63. Je�coate S, Organ K, Schoenfeld S et al. E�cacy and safety of subcutaneous pramlintide therapy in peoplewith type 2 diabetes. Diabetic Medicine 1998; 15 (supplement 1): S5.

64. Fulcher GR, Walker M, Farrer M et al. Acipimox increases glucose disposal in normal man independentof changes in plasma non-esteri®ed fatty acid concentration and whole body lipid oxidation. Metabolism1993; 42: 308±314.

65. Jones IR, Swai A, Taylor R et al. Lowering of plasma glucose concentrations with beza®brate in patientswith moderately controlled NIDDM. Diabetes Care 1990; 13: 855±863.

66. Bianchi R, Bohgers V, Bravenboer B & Erkelens DW. E�ects of ben¯uorex on insulin resistance and lipidmetabolism in obese type II diabetic patients. Diabetes Care 1993; 16: 557±559.

67. Frenkel B, Bishara-Shieban J & Bar-Tana J. The e�ect of b,b-methyl-substituted hexadecanedioic acid onplasma VLDL metabolism in rats: role of adolipoprotein C-III. Biochemical Journal 1994; 298: 409±414.

68. Foley JE. Rationale and application of fatty acid oxidation inhibitors in treatment of diabetes mellitus.Diabetes Care 1992; 15: 773±784.

69. Weyer C, Tataranni PA, Snitker S et al. Increase in insulin action and fat oxidation after treatment withCL 316, 243, a highly selective b3-adrenoceptor agonist in humans. Diabetes 1998; 47: 1555±1561.

70. Unson CG, MacDonald D, Ray K et al. Position 9 replacement analogs of glucagon uncouple biologicalactivity and receptor binding. Journal of Biological Chemistry 1991; 266: 2763±2766.

71. Stacpole PW & Greene YK. Dichloroacetate. Diabetes Care 1992; 15: 785±791.72. Vincent MF, Erion MD, Gruber HE et al. Hypoglycaemic e�ect of AIC Ariboside in mice. Diabetologia

1996; 39: 1148±1155.73. Parker JC, VanVolkenburg MA, Levy CB et al. Plasma glucose levels are reduced in rats and mice treated

with an inhibitor of glucose-6-phosphate translocase. Diabetes 1998; 47: 1630±1636.


new agents for type 2 diabetesnew agents for type 2 diabetes malcolm nattrass bsc, mb, chb, phd,...

Documents