The thiazolidinediones (TZDs) troglitazone, rosiglitazone and pioglitazone have had a troubled history. These agents all work upon the nuclear peroxisome proliferator activated receptor (PPAR) Gamma-receptor, and modulate the activity of multiple genes. They were heralded as the first drugs to truly address insulin resistance, the central problem in type 2 diabetes. Animal and human studies indeed showed the TZDs to be efficacious in lowering insulin resistance, glucose levels and HbA1c. Unfortunately, these agents are highly pleiotropic, and affect many other regulatory pathways unrelated to glucose metabolism. These encompass redistribution of adipose tissue and weight gain, fluid retention, osteopenia and possibly altered cancer risk. Troglitazone was banned because it caused hepatic necrosis, and rosiglitazone has been banned in Europe and restricted in the USA because of increased cardiovascular morbidity. Pioglitazone has been banned in France and black-box warnings have been applied in the US because of bladder cancer. This class is likely to remain out of favour unless some way can be found of dissociating its useful and unwanted effects.
Mechanism of action of TZDs
Thiazolidinediones modulate the action of Peroxisome-Proliferator–Activated Receptor (PPAR) Gamma. The PPARs are a family of nuclear receptors that regulate gene expression in response to the binding of certain fatty acids. The downstream effects on gene regulation can either be activating or repressing, depending on which ligand binds to the receptor. The PPARγ subtype is primarily expressed in adipose tissue and is thought to play a crucial role in adipocyte proliferation and differentiation, fatty acid uptake and fatty acid storage. By promoting the storage of fat in adipose tissue rather than burdening the liver or muscle with fatty acid overload, TZDs are thought to improve insulin sensitivity.
Clinical efficacy of TZDs
Figure 1. Chemistry of the thiazolidinediones pioglitazone and rosiglitazone The first thiazolidinedione, troglitazone, was developed around 1994 and marketed in 1997. The beneficial effects of troglitazone on insulin sensitivity, glucose levels and free fatty acid concentrations was clearly demonstrable but soon reports about serious hepatotoxicity emerged and troglitazone was withdrawn from the market in 2000. Hepatotoxicity did not emerge as a class effect, and pioglitazone and rosiglitazone (figure 1) were developed for clinical use.
Several trials established the efficacy of TZDs as glucose-lowering agents, with HbA1c decreasing by about 1-1.5%. In line with their presumed mechanism of action, the TZDs also seem efficacious in disease states associated with abnormal distribution of fatty acids over the tissues, such as Non Alcoholic Fatty Liver Disease (NAFLD) or lipodystrophies.
However, there is no consistent and undisputed evidence for a beneficial effect of the TZDs on diabetes associated morbidity (e.g. microvascular complications) or mortality.
Much was made of a small (2.1% absolute) risk reduction for a secondary combined macrovascular endpoint in the pioglitazone-treated arm compared to the control arm of the PROACTIVE study. However, the primary endpoint in the PROACTIVE study did not show any difference; in the subgroup adequately treated with statins pioglitazone conferred no benefit; and despite a study protocol aiming at similar glycaemic control the control arm had a higher HbA1c at endpoint. In line with this, in the RECORD study no difference in any relevant hard endpoint could be found for those assigned to rosiglitazone compared to controls. Finally, an independent meta-analysis of rosiglitazone study data revealed that rosiglitazone is associated with an increased risk of fatal myocardial infarction (but not an increased risk of all-cause mortality).
Adverse effects of TZDs
Apart from the increased risk for myocardial infarction with rosiglitazone, there are various other side-effects which caution against the use of TZDs. Fluid retention and an increased risk of hospitalisation for heart failure were noted as a side-effect of TZDs from the start. This fluid retention also partially explains the weight gain that is associated with TZD use. The fluid retention tends to be more severe when combining TZDs and insulin, which is therefore contra-indicated.
An intriguing finding is the increased risk for fractures associated with their use. It has been put forward that PPARγ modulation affects the preferential differentiation of mesenchymal stem cells towards adipocytes rather than osteocytes, ultimately leading to more brittle bones.
Finally, the TZDs, particularly pioglitazone, are associated with an increased risk for bladder cancer although overall cancer risk seems not increased and risk for colon and liver cancer may be somewhat reduced.
^ Grey A et al. The peroxisome proliferator-activated receptor-gamma agonist rosiglitazone decreases bone formation and bone mineral density in healthy postmenopausal women: a randomized, controlled trial. J Clin Endocrinol Metab. 2007;92:1305-10.