Study Design

This was a multicentre, randomised, double-blind study of the safety and efficacy of a combination of 30 or 45 mg of pioglitazone (Actos) and insulin when given to patients with type 2 diabetes mellitus whose glucose levels were poorly controlled by their current insulin therapy. Patients who participated in this study were at least 18 years of age, had an HbA1c value greater than or equal to 8.0%, and were on a stable, fixed dose (at least 30 units/day) of insulin for at least 30 days before the study.

The primary efficacy variable was HbA1c. The secondary efficacy variables were HbA1c responder rate defined as a percentage of patients meeting a clinically relevant target value, fasting plasma glucose (FPG), FPG responder rate, and serum lipids (ie, triglycerides, total cholesterol, high density lipoprotein (HDL) cholesterol, low density lipoprotein (LDL) cholesterol, and very low density lipoprotein (VLDL) cholesterol, and free fatty acids (FFA). A total of 345 patients were randomly assigned to each treatment arm, for a total of 690 patients in the intent-to-treat (ITT) population. The mean age for patients was 56.5 years, and mean BMI was 33.19 kg/m2. Approximately two thirds (63.3%) of the patients were Caucasian, and slightly more than half (54.6%) were male. The mean insulin dose (all forms) at Baseline for ITT patients was 69.2 units/day, and 27.1% of the patients reported use of antidiabetes therapy in addition to insulin. There were no major differences between the treatment groups with regard to any of the baseline variables.

Results

The improvements in glycaemic control observed in this study occurred in concert with overall reductions in insulin use by both treatment groups. Statistically significant difference from Baseline in insulin use were first observed at Week 4 for the 45 mg treatment group and at Week 8 for the 30 mg treatment group. The reductions were maintained throughout the remainder of the study, and there was a 4.5 and 7.3 U/d reduction with 30 mg and 45 mg of pioglitazone (Actos), respectively, at Endpoint. The reduction in insulin dose was statistically significantly greater in the 45 mg treatment group at all time points.

Introduction

Actos (pioglitazone HCl), a member of the thiazolidinediene (TZD) insulin-sensitizing class of drugs, was recently approved by the FDA for marketing in the US. Actos, developed by Takeda America Research and Development Center, Inc., is indicated for glycemic control in people with type 2 diabetes. It can be used alone or in combination with sulfonylurea, metformin, or insulin when these three agents prove to be ineffective on their own. Whether used as monotherapy or as a component of combination therapy, treatment should also include proper diet and exercise.

Clinical Study Results

Three randomized, double-blinded, placebo-controlled clinical trials were conducted in the United States to determine the safety and efficacy of Actos monotherapy. The first study lasted 26 weeks and included 408 patients with type 2 diabetes. They were randomized to receive 7.5 mg, 15, mg, 30 mg, 45 mg, or placebo once daily. Statistically significant differences were noted in the HbA1c and in the fasting blood glucose levels at endpoint in the patients who received 15, mg, 30 mg, and 45 mg compared to placebo. There was a one percentage point difference in HbA1c between both the 15 mg and 30 mg Actos patients and the control patients, and a 1.6 percentage point difference between the 45 mg patients and the control group (p < 0.05). After treatment, FBG increased by 9 mg/dL in placebo patients and decreased by 30, 32, and 56 mg/dL in patients who received 15, 30, and 45 mg of Actos, respectively (p < 0.05). A total of 260 patients participated in the second 24-week study. Patients were randomized to receive one of two forced-Actos titration regimens or placebo titration. The first Actos titration group received 7.5 mg of Actos for four weeks, with doses increasing to 15 mg and then 30 mg in four-week intervals. The second group followed a similar pattern of dosage increases, but began with 15 mg/day and went up to 45 mg/day. HbA1c levels were significantly lower in both treatment groups compared to the control group (1.5 percentage point difference for both, p < 0.05 for both).

There was a 68 mg/dL difference between patients in the higher dose Actos regimen and those in the placebo group, and a 62 mg/dL difference between those in the lower dose Actos regimen and patients in the placebo group (p < 0.05). Lastly, in a 16-week study with 197 participants, 30 mg Actos was compared to placebo. A 1.4 percentage point difference in HbA1c from placebo and a 58 mg/dL difference in FBG from placebo were observed.

Three 16-week, randomized, double-blind, placebo-controlled studies were conducted to evaluate the effects of Actos on glycemic control in type 2 diabetes patients as part of combination therapy. The addition of Actos to a treatment regimen consisting of sulfonylurea significantly reduced the mean HbA1c by 0.9% and 1.3%, with 15 mg and 30 mg of Actos, respectively. Treatment with metformin also benefited from the addition of Actos to the regimen, with a 0.8% reduction in HbA1c and a 38 mg/dL decrease in FBG. Patients with unsuccessful glycemic control with insulin alone experienced a 0.7 percentage point and 1.0 percentage point decrease in HbA1c with 15 mg Actos and 30 mg Actos, respectively. FBG levels decreased as well by 35mg/dL and 49 mg/dL with the addition of 15 mg Actos and 30 mg Actos to the insulin regimen.

What the Patient Should Know

Some of the patients in the clinical trials experienced upper respiratory tract infection, headache, sinusitis, muscle pain, and sore throat. Patients should also be aware of the possibility of moderate to mild edema and anemia. Using Actos in combination with insulin and sulfonylurea is associated with a greater risk for hypoglycemia. As a result, lowering the doses of insulin or sulfonylurea may be necessary.

Pharmacopoeias. In Europe. Also in USNF.

European Pharmacopoeia, 6th ed. (Guar). Guar is obtained by grinding the endosperms of the seeds of Cyamopsis tetragonolobus. It consists mainly of guar galactomannan. Guar is a white or almost white powder, yielding a mucilage of variable viscosity when dissolved in water. Practically insoluble in alcohol.

European Pharmacopoeia, 6th ed. (Guar Galactomannan). A yellowish-white powder. It is soluble in cold and hot water practically insoluble in organic solvents. Its main components are polysaccharides composed of D-galactose and D-mannose at molecular ratios of 1:1.4 to 1:2. The molecules consist of a linear main chain of β-(1 —>4)-glycosidically linked mannopyranoses and single α-(1 —>6)-glycosidically linked galactopyranoses.

The United States Pharmacopeia 31, 2008, and Supplements 1 and 2 (Guar Gum). A gum obtained from the ground endosperms of Cyamopsis tetragonolobus (Leguminosae). It consists chiefly of a high-molecular-weight hydrocolloidal polysaccharide, a galactomannan, composed of galactan and mannan units combined through glycosidic linkages. It is a white to yellowish-white, practically odourless, powder. Dispersible in hot or cold water forming a colloidal solution.

Adverse Effects and Precautions

Guar gum can cause gastrointestinal disturbance with flatulence, diarrhoea, or nausea, particularly at the start of treatment.

Because guar gum swells on contact with liquid it should always be washed down carefully with water and should not be taken immediately before going to bed. It should not be used in patients with dysphagia, oesophageal disease, or intestinal obstruction.

Interactions

Guar gum may retard the absorption of other drugs where this is likely to pose a problem the other drug should be taken at least an hour before guar gum.

Uses and Administration

Guar gum is used in diabetes mellitus as an adjunct to treatment with diet, insulin, or oral antidiabetics since it results in some reduction in both postprandial and fasting blood-glucose concentrations. It is given with or immediately before meals in doses of 5 g usually 3 times daily. Adverse gastrointestinal effects may be reduced by using a lower initial dose of 5 g once daily before breakfast for 1 week, then increasing to 5 g twice daily, then 3 times daily, as required. Each dose of guar gum granules should be taken stirred in about 200 mL of a cold drink. Alternatively it can be sprinkled over or mixed with food which must be taken with about 200 mL of fluid.

Guar gum is also used to slow gastric emptying in some patients with the dumping syndrome. It is also used as an adjunct in the treatment of hyperlipidaemias.

Guar gum is also used as a thickening and suspending agent, and as a tablet binder. It has been incorporated into processed foods.

Guar gum is an example of a soluble fibre. On contact with water it forms a highly viscous gel, the viscosity of which varies with such factors as its plant source or the form in which it is given.

Fibres such as guar gum reduce postprandial and fasting blood-glucose concentrations as well as plasma-insulin concentrations in healthy subjects and diabetic patients. Such reductions in blood-glucose concentrations and in glycosylated haemoglobin have been demonstrated in both type 1 and type 2 diabetes, but they have generally been small. Possible mechanisms for these effects of guar gum include a delay in gastric emptying, decreased small-bowel motility, decreased glucose absorption resulting from increased viscosity of the contents of the gastrointestinal tract, or inhibition of gastrointestinal hormones.

Guar gum also lowers serum total cholesterol and low-density -lipoprotein (LDL) cholesterol concentrations high-density-lipoprotein (HDL) cholesterol and triglyceride concentrations appear to be unaffected. The most likely mechanism is binding of bile acids, reducing their enterohepatic circulation in a similar way to bile-acid sequestrants. When used alone in patients with hype rcholesterolaemia guar gum has generally produced a modest reduction in plasma-cholesterol and LDL-cholesterol concentrations although some studies have been unable to demonstrate an effect. A few studies have suggested that the cholesterol-lowering effect is attenuated after 8 to 12 weeks of treatment but a long-term study observed a 17% decrease in total serum cholesterol that was maintained for 24 months. Some studies have shown further reductions in cholesterol and LDL-cholesterol concentrations on addition of guar gum to therapy with other li-pid regulating drugs. The usual treatment of hyper lip idaemias is discussed.

There have been suggestions that guar gum reduces appetite by promoting a feeling of fullness, but a meta-analysis has indicated that it is not effective for reducing body-weight. Products containing guar gum have, however, been promoted as slimming aids. Their use cannot be advocated because of the risk of tablets swelling before reaching the stomach and causing oesophageal obstruction.

Preparations

Proprietary Preparations

Argentina: Regudigl

Australia: Benefiber †

Brazil: Benefiber † Biofiber †

Finland: Guarem

Germany: Figur-Verlan Guar Verlan

Hong Kong: Guarem

Ireland: Guarem †

Italy: Novafibra

New Zealand: Guarcol

Spain: Fibraguar Plantaguar

Switzerland: Leiguar

UK: Resource Benefiber

USA: Benefiber

Multi-ingredient

France: Carres Parapsyllium Moxydar Mucipulgite Mulkine Seroxydar

Italy: Cruscasohn Resource Gelificata

Switzerland: Mucipulgite

Drug Nomenclature

International Nonproprietary Names (INNs) in main languages:

Adverse Effects and Precautions

As for Rosiglitazone Maleate. The effects of pioglitazone on serum lipid concentrations appear to differ from those of rosiglitazone, see below. Other adverse effects reported include upper respiratory-tract infections, haematuria, and visual disturbances. Liver function should be monitored periodically as there have been isolated reports of liver dysfunction, and the drug should be used with caution in patients with hepatic impairment (see below).

An increased incidence of bladder cancer has been seen in rats but not in mice treated with pioglitazone. Use is contra-indicated in patients with diabetic ketoacidosis. For precautions and contra-indications to the use of thiazolidinediones in heart failure see Effects on the Heart, under Rosiglitazone Maleate.

Effects on lipids. Thiazolidinediones are reported to affect serum concentrations of lipids. Compared with placebo, pioglitazone has been found to reduce triglycerides, increase high-density lipoprotein (HDL)-cholesterol, and have little or no effect on low-density lipoprotein (LDL)- and total cholesterol. In a studyof patients being transferred from troglitazone to either pioglitazone or rosiglitazone, there were decreases in concentrations of triglycerides, LDL- and total cholesterol, and an increase in HDL-cholesterol in those patients on pioglitazone, whereas the opposite occurred for rosiglitazone. Whether these effects of pioglitazone reduce cardiovascular risk in patients with type 2 diabetes is yet to be fully established, but the large prospective PROactive study did suggest that it could reduce the risk of mac-rovascular events in patients with evidence of macrovascular disease, although the risk of heart failure appears to be increased (see Diabetic Complications, below, and Effects on the Heart, under Rosiglitazone Maleate).

Effects on the liver. There have been isolated reports of hepatocellular injury in patients receiving pioglitazone.The UK and US licensed product information recommends that liver enzymes should be checked before starting therapy with pioglitazone patients with aminotransferase (ALT) concentrations more than 2.5 times the upper limit of normal should not be given pioglitazone. ALT concentrations should then be monitored periodically during treatment. If ALT concentrations rise to more than 3 times the upper limit of normal and remain so after retesting then treatment with pioglitazone should be stopped treatment should also be stopped if jaundice develops.

Interactions

When pioglitazone was given with gemfibrozil, an inhibitor of the cytochrome P450 isoenzyme CYP2C8, there was a threefold increase in the area under the concentration-time curve (AUC) of pioglitazone, and a decrease in pioglitazone dose may be needed if it is given with gemfibrozil or similar CYP2C8 inhibitors. Equally, rifampicin, a potent inducer of cytochrome P450, halves the AUC of pioglitazone when both are given, and pioglitazone dose may need to be increased.

Antibacterials. For a report of hypoglycaemia when gatifloxacin was given to a patient already receiving oral hypoglycaemics such as pioglitazone.

Pharmacokinetics

Pioglitazone is rapidly absorbed after oral doses. Peak plasma concentrations occur within 2 hours and bioavailability exceeds 80%. Pioglitazone is more than 99% bound to plasma proteins. It is extensively metabolised, primarily by the cytochrome P450 isoenzyme CYP2C8 to both active and inactive metabolites. It is excreted in urine and faeces and has a plasma half-life of up to 7 hours. The active metabolites have a half-life of up to 24 hours.

Uses and Administration

Pioglitazone is a thiazolidinedione oral antidiabetic similar to rosiglitazone. It is used in the management of type 2 diabetes mellitus. It is given as pioglitazone hydrochloride but doses are expressed in terms of the base pioglitazone hydrochloride 1.1 mg is equivalent to about 1 mg of pioglitazone. It is given orally as monotherapy, particularly in patients who are overweight and for whom metformin is contra-indicated or not tolerated. Pioglitazone may also be added to metformin or a sulfonylurea or both, or to insulin, when single-agent therapy is inadequate (but see Administration, below). The usual dose is 15 or 30 mg once daily. This may be increased to a maximum of 45 mg once daily if necessary. Pioglitazone may be taken with or without food.

Administration. Although pioglitazone is licensed for use with other antidiabetic drugs, the specifics of licensing and use may vary from country to country. In the UK, use of pioglitazone with insulin was originally considered to be contra-indicated, because of an increased risk of heart failure, although this was subsequently amended to permit dual therapy in patients who could not be given insulin plus metformin. Furthermore, although pioglitazone is licensed for use with metformin or a sulfonylurea (or both if necessary) in patients who do not respond to these drugs, NICE recommends this only in patients unsuited to combination therapy with metformin plus a sulfonylurea. However, in the USA, pioglitazone has always been licensed for use with insulin (with appropriate monitoring), metformin, or a sulfonylurea in any patient in whom single agent therapy is inadequate.

Diabetic complications. It has been suggested that, in addition to their hypoglycaemic effect, thiazolidinediones may have beneficial effects in the prevention of macrovascular diabetic complications. Studies in patients with type 2 diabetes mellitus have shown that pioglitazone may slow the progression of carotid intima-media thickness, an indicator for cardiovascular risk. A study of secondary prevention found that, compared with placebo (in addition to usual medications for glucose control), pioglitazone reduced death from any cause, myocardial infarction, and stroke. There was also a reduced need to add insulin for glucose control. However, there was no significant difference between the groups when the end-point was broader and also included acute coronary syndrome, leg amputation, and coronary or leg revascularisation. Subgroup analysis also found that pioglitazone reduced the risk of fatal and non-fatal myocardial infarction and acute coronary syndrome in patients with a history of myocardial infarction. A meta-analysis that included the results of this study with cardiovascular outcome data from studies of glycaemic control found that pioglitazone significantly reduced the risk of death, myocardial infarction, and stroke in a diverse population of patients with type 2 diabetes. However, the risk of serious heart failure was increased (see also Effects on the Heart).

It is unclear whether other thiazolidinediones might have similar effects and whether patients at lower risk might benefit. Rosiglitazone and pioglitazone are known to have different effects on lipids (above) and there is some evidence that rosiglitazone may have adverse cardiovascular effects.

Hepatitis. A small proof-of-concept study has suggested that pioglitazone 45 mg daily with a hypocaloric diet for 6 months produces greater metabolic and histological improvement in patients with nonalcoholic steatohepatitis than diet alone.

Malignant neoplasms. For references to the experimental use of pioglitazone with rofecoxib and trofosfamide as anti-ang-iogenic therapy for malignant neoplasms see Trofosfamide.

Psoriasis. It has been suggested that by binding to peroxisome proliferator-activated receptor gamma, pioglitazone may have an anti-rnflarnmatory effect in conditions such as chronic plaque psoriasis and psoriatic arthritis (see Spondyloarthropathies). In a small open-label study, oral doses of 30 mg daily were reported to improve moderate chronic plaque psoriasis in 4 of 5 patients, with definite improvement seen 1 to 3 months after starting therapy. Treatment was stopped in 1 other patient because of fluid retention. In a double-blind study, 70 patients with moderate to severe disease were treated for 10 weeks with daily doses of pioglitazone 15 mg, 30 mg, or placebo. Greater improvements were reported with pioglitazone than with placebo, and the dose of 30 mg appeared to be slightly more effective than 15 mg. There has also been a report of improvements in tender and swollen joints in a small group of patients with psoriatic arthritis who were given a high dose of pioglitazone (30 mg twice daily) for 12 weeks. Fluid retention was also reported.

Preparations

Proprietary Preparations

Argentina: Actos Cereluc Higlucem Pioglit Piotamax

Australia: Actos

Austria: Actos

Belgium: Actos

Brazil: Actos

Canada: Actos

Chile: Actosf Diabestat Tiazac

Czech Republic: Actos Glustin

Denmark: Actos

Finland: Actos

France: Actos

Germany: Actos

Greece: Actos

Hong Kong: Actos

India: Diaglit G-Tase Glita Glizone Opam P-Glitz Pepar Piomed Piosafe Piozulin

Indonesia: Actos Deculin

Italy: Actos

Japan: Actos

Malaysia: Actos

Mexico: Zactos

The Netherlands: Actos Glustin

Norway: Actos

New Zealand: Actos

Philippines: Actos Prialta Zypi

Portugal: Actos Glustin

Russia: Actos

South Africa: Actos

Spain: Actos

Sweden: Actos

Switzerland: Actos

Thailand: Actos

UK: Actos

USA: Actos

Multi-ingredient

Czech Republic: Competact Glubrava Tandemact

France: Competact Tandemact

India: Exermet P P-Glitz M Piomed M Piosafe M †

Portugal: Competact Tandemact

UK: Competact

USA: Actoplus Met Duetact

Study Design

This was a multicentre, randomised, double-blind, placebo-controlled study of the safety and efficacy of 2 doses of pioglitazone (Actos) (15 or 30 mg) in combination with insulin compared with insulin alone in patients with type 2 diabetes mellitus whose glucose levels were poorly controlled with their current insulin therapy. Patients who participated in this study were 30 to 75 years of age inclusive, had an HbA1c value greater than or equal to 8.0%, had received insulin therapy of at least 30 units/day for 4 months or longer, and were on a stable, fixed dose of insulin for at least 30 days before the study (with or without metformin, acarbose, or a sulphonylurea).

The primary efficacy variable was change from Baseline in HbA1c. The secondary efficacy variables were HbA1c responder rate defined as a percentage of patients achieving a clinically relevant target value, fasting plasma glucose (FPG), fasting C-peptide, triglycerides, and cholesterol (total, high density lipoprotein (HDL) cholesterol, and low density lipoprotein (LDL) cholesterol. Comparisons between placebo plus insulin and each of the 2 pioglitazone (Actos) plus insulin treatment groups, with respect to changes in the primary and secondary variables, were performed using the Dunnett test with estimates of means and variances obtained from a 2-way ANCOVA model. A total of 566 patients were randomly assigned to treatment and included in the intent-to-treat (ITT) population. Mean age at Baseline was 57 years, and mean body mass index (BMI) was 33.6 kg/m2. Most patients were Caucasian (73%), and slightly more than half (53%) were female. Baseline systolic blood pressure was slightly higher for the groups who received pioglitazone (Actos) plus insulin than for the group who received placebo plus insulin, but there were no differences among the treatment groups with respect to any other baseline variable.

Results

All treatment groups had statistically significant mean decreases from Baseline in HbA1c throughout the study, and both groups of patients who received pioglitazone (Actos) plus insulin had a statistically significantly greater reduction in HbA1c than the group of patients who received placebo plus insulin. At Endpoint (Week 16), those who received 15 mg of pioglitazone (Actos) plus insulin had a mean reduction in HbA1c that was 0.73% (95% CI, -100, -0.47) beyond that observed in the placebo plus insulin group, and the group who received 30 mg of pioglitazone (Actos) plus insulin had a reduction that was 1.00% (95% CI, -1.27, -0.74) beyond that observed with placebo plus insulin. The reduction in HbA1c levels was reflected in the number of HbA1c responders, as 31.6% of the patients who received placebo, 69.5% who received 15 mg of pioglitazone (Actos), and 75.1% who received 30 mg of pioglitazone (Actos) exhibited a 0.6 percentage point decrease in HbA1c, a reduction in HbA1c to a level of 6.1% or lower, or both, without requiring an increase in daily insulin dose of 25% or more.

In clinical trials, pioglitazone 45 mg given once daily significantly lowered mean fasting blood glucose levels in patients who were not previously taking antidiabetic medications by 63.7 mg/dL. Patients’ glycosylated hemoglobin levels fell by up to 1.9 percentage points compared with baseline. Also, pioglitazone significantly decreased mean triglyceride levels and increased mean HDL levels in both monotherapy and combination therapy.

Because of the problems observed with troglitazone, monitoring of hepatic enzymes is recommended with pioglitazone even though no toxicities have been observed thus far (at the start of therapy, every two months for the first year, and periodically thereafter). As with other agents that reduce insulin resistance, premenopausal anovulatory women may experience a resumption of ovulation once therapy begins, and pregnancy can result.

The most commonly reported side effects to pioglitazone in clinical trials were symptoms of upper respiratory tract infection, headache, sinusitis, muscle pain, tooth disorder, and sore throat. Weight gain was noted, as well as mild to moderate edema and anemia. Patients receiving pioglitazone in combination with insulin or sulfonylureas may be at risk for hypoglycemia, and a reduction in the dose of insulin or sulfonylureas may be necessary.

Separately, SKB strengthened its hand by announcing it would copromote rosiglitazone – likely to be approved for treatment of patients with diabetes type 2 as either monotherapy or with metformin – with Bristol-Myers Squibb, which markets metformin (Glucophage). Since rosiglitazone will probably reach the U.S. market before pioglitazone, the collaboration could be critical in quickly penetrating the insulin-resistance market. Some experts project that pioglitazone will become the market leader in the glitazone class, and Takeda’s previously announced marketing collaboration with Lilly promises to make the competition fierce.

Oral antidiabetic drugs fall into several classes (Table 5). Of particular interest are the insulin sensitizers because they specifically target insulin resistance. Other agents address different aspects of glycemic control.

Insulin secretion stimulators (secretagogues)

For more than 40 years, sulfonylureas have been the first line of therapy for individuals with type 2 diabetes. These agents directly stimulate pancreatic b-cells to produce insulin by increasing the influx of calcium. Sulfonylureas increase circulating insulin and reduce both fasting and postprandial glucose, but they are not insulin sensitizers and therefore do not address the problem of insulin resistance. Sulfonylureas lower A1C an average of 1% to 2% and offer effective glycemic control in up to 75% of patients; however, efficacy lapses over time, with about 5-10% of patients per year failing to maintain the initial glycemic control. Primary adverse effects include hypoglycemia and weight gain (typically 2-5 kg). Glimepiride (Amaryl) is emerging as the sulfonylurea of choice due to its once-a-day dosing, extrapancreatic effect, and the fact that it causes less weight gain and hypoglycemia and is priced as low as generic glyburide.

Another class of secretagogues, derivatives of meglitinide or phenylalanine, also stimulate insulin but act at a different site on pancreatic beta-cells than the sulfonylureas. Because these agents have a very short onset of action and short half-life, they must be taken immediately before every meal (compared to once-daily dosing for sulfonylureas), so treatment adherence may be an issue for some patients. They have a side effect profile similar to the sulfonylureas; however, because meglitinides are shorter-acting agents, they carry a lower risk of sustained hypoglycemia. Efficacy is similar to that of sulfonylureas. The two products currently available are nateglinide (Starlix) and repaglinide (Prandin).

Alpha-glucosidase inhibitors

Drugs in this group produce mild reductions in postprandial hyperglycemia by inhibiting the enzyme responsible for metabolizing complex carbohydrates in the small intestine. Taken right before a meal, these agents reduce glucose levels by slowing absorption of carbohydrates and delaying entry of glucose into liver and muscle tissue. Gastrointestinal side effects (ie, abdominal pain, diarrhea, and flatulence) are the most common reactions to alpha-glucosidase inhibitors (reported in up to 75% of patients), leading some patients to discontinue therapy with these drugs. Available agents include acarbose (Precose) and miglitol (Glyset). Gastrointestinal side effects can be greatly reduced if low doses are started and then gradually titrated over 10-12 weeks to the maximum and effective doses. At present, these agents are seldom used in the United States.

Thiazolidinediones

The thiazolidinediones (TZDs or glitazones) are a relatively new class of agents that reduce insulin resistance. TZDs do not stimulate the secretion of insulin but rather enhance the effects of circulating insulin by improving insulin sensitivity in muscle and adipose tissue and by inhibiting hepatic gluconeogenesis. TZDs work by stimulating certain receptors (peroxisome proliferator-activated receptor gamma, or PPAR-gamma) in the nucleus of the cells. Activation of PPAR-gamma modulates the transcription of a number of insulin-responsive genes involved in the control of glucose and lipid metabolism. In response to thiazolidinediones stimulation, the genes produce a protein called GLUT-4. Insulin works by recruiting GLUT-4 to the cell’s outer membrane. This partnership, in turn, promotes transport of glucose across the membrane and into the cell’s interior.

Examples of insulin sensitizers include pioglitazone hydrochloride (Actos) and rosiglitazone maleate (Avandia). Another drug in this class, troglitazone (Rezulin), was removed from the market because it was linked to idiosyncratic cases of hepatotoxicity. In clinical trials, there has been no evidence of drug-induced hepatotoxicity with pioglitazone or rosiglitazone, but there have been rare postmarketing case reports of liver damage in patients receiving rosiglitazone and pioglitazone (causality not established). The safe use of these agents, therefore, requires careful monitoring of liver function: ALT enzyme levels should be measured at baseline and monitored every 2 months for 1 year and periodically thereafter. Patients with hepatic impairment should not be treated with thiazolidinediones.

In large placebo-controlled trials lasting up to 26 weeks, monotherapy with pioglitazone or rosiglitazoneproduced significant improvements in A1C and fasting blood glucose concentrations (Table 6). Pioglitazone also led to significant improvements in A1C and improvements in FPG when combined with a sulfonylurea, metformin, or insulin. Rosiglitazone resulted in significant decreases in A1C and FPG levels when combined with metformin or a sulfonylurea.

In addition to reducing insulin resistance, thiazolidinediones also have effects on lipids (Table 6). In a 26-week placebo-controlled study, pioglitazone was associated with decreases in triglycerides of 9.0%, 9.6%, and 9.3% in patients treated with 15-, 30-, and 45-mg, respectively, compared with baseline. HDL (“good”) cholesterol increased by 14%, 12%, and 19% in the 15-, 30-, and 45-mg groups, respectively. No consistent differences were reported for LDL (“bad”) cholesterol and total cholesterol in patients treated with pioglitazone versus placebo.

In a similar 26-week pla-cebo-controlled study, rosiglitazone raised HDL cholesterol by 11.4% and 14.2% in doses of 4- and 8-mg per day, respectively, compared to baseline, but the drug also raised LDL cholesterol by 14.1% and 18.6%, respectively. Changes in triglycerides were variable and generally not statistically significant compared to placebo controls. A recent retrospective review of type 2 diabetes patients treated with either pioglitazone (n=525) or rosiglitazone (n=590) suggested that pioglitazone provides a greater benefit in terms of blood lipid profile than does rosiglitazone.

Because TZDs do not affect insulin secretion, they do not induce hypoglycemia. Dose-related weight gain is seen with both pioglitazone (average increase 0.5 kg to 2.8 kg) and rosiglitazone (median increase 1.0 kg to 3.1 kg). Also, a small number of patients experience mild to moderate edema and anemia. TZDs can cause fluid retention, which may lead to or exacerbate heart failure; thus, patients should be observed for signs and symptoms of congestive heart failure, and TZDs should not be used in patients with class III or IV cardiac status. Studies are currently underway to determine whether thiazolidinediones may be effective in preventing progression of insulin resistance to full-blown type 2 diabetes.

Mechanism of Action: The mechanism of action of the thiazolidinediones is still being investigated, but many of their actions seem to be mediated through binding and activation of PPARg (peroxisome proliferator-activated receptor g). PPARg is a nuclear receptor that has a regulatory role in cell differentiation, particularly fat cell differentiation. Although concentrations of PPARg tend to be highest in fat cells, the receptor is present in many other tissues, including skeletal muscle and pancreatic beta cells. The potencies of the thiazolidinediones as antihyperglycemic agents are also correlated with their binding affinities and functional agonist potencies at PPARg. Intriguingly, while some agents are virtually complete PPARg agonists, other agents, particularly pioglitazone, are partial agonists with mixed a- and g-activation. This property has been suggested as a mechanism underlying the differential effects of these agents on lipid parameters.

Available data indicate that the major effects of these agents in vivo are to increase peripheral glucose disposal, primarily by enhancing skeletal muscle uptake of glucose. Some studies have also shown an effect of the agents on endogenous glucose production. It also remains unclear whether the glucose-lowering effects of these agents are produced directly, via activation of PPARg in skeletal muscle, or indirectly, via their effects on PPARg in adipocytes and potentially mediated by reductions in free fatty acids (FFA). Reductions in FFA concentrations have been documented to improve insulin sensitivity, and all of the clinically available thiazolidinediones produce significant reductions in FFA, despite having apparently disparate effects on overall lipid profiles.