Pioglitazone: The forgotten,cost-effective cardioprotective drug for type 2 diabetes

Background

Type 2 diabetes mellitus (T2DM) is a cardiometabolic disease^1,2 that affects both the microvasculature (retinopathy, nephropathy, neuropathy) and macrovasculature [myocardial infarction (MI), stroke]. The microvascular complications primarily are related to the level of glycaemic control,^3,4 whereas hyperglycaemia is a relatively weak risk factor for the macrovascular complications^3,5 which represent the major cause of mortality in T2DM patients.^6,7 Long-term cardiovascular (CV) outcome trials have generally demonstrated no or only slight reduction in CV events with intensive glycaemic control.^3,8–10 In contrast, treatment of more traditional CV risk factors (blood pressure, dyslipidaemia) consistently has been associated with major CV benefits in T2DM patients. ¹

The results of recent CV outcome trials have documented that glucose-lowering agents in two different classes significantly reduce the MACE (major adverse cardiovascular events) endpoint (composite of CV mortality, non-fatal MI, non-fatal stroke). In both the EMPA REG OUTCOME trial ¹¹ and in the CANVAS program, ¹² the sodium glucose transporter-2 (SGLT2) inhibitors, empagliflozin and canagliflozin, reduced MACE by 14% and 13%, respectively, although the relative contributions of the three individual components of the composite outcome differed. In LEADER ¹³ and SUSTAIN-6, ¹⁴ therapy with the glucagon-like-peptide receptor agonists (GLP-1 RAs), liraglutide and semaglutide, resulted in reductions in MACE of 13% and 26%, respectively, and also with differential contributions from the composite elements. Importantly, empagliflozin and liraglutide were each associated with significant reductions in CV mortality as well. With the robust results of these large, long-term, CV outcome trials, we are entering a new era of T2DM treatment where glucose-lowering drugs that address both glycaemia, as well as CV risk, are now preferred in patients with cardiovascular disease (CVD) over those therapies that simply lower HbA1c. ¹⁵

In the midst of the newfound interest in the SGLT2 inhibitors and GLP-1 RAs, the established anti-atherogenic benefits of the thiazolidinedione (TZD), pioglitazone, have been overlooked. ¹⁶ The recent results of the IRIS (Insulin Resistance Intervention after Stroke) trial ¹⁷ should rekindle interest in pioglitazone as a cardioprotective drug, an effect which actually was established more than a decade ago. Because pioglitazone is now generically available, it represents a more affordable option than either an SGLT2 inhibitor or a GLP-1 RA. ¹⁸ Furthermore, it can be combined with these and other glucose-lowering agents, including the SGLT2 inhibitors or GLP-1 RAs, to minimize side effects.^15,18–20 Pioglitazone also has a number of other demonstrated benefits, including amelioration of insulin resistance, preservation of beta-cell function, durable glycaemic control, improvement of multiple factors of the metabolic syndrome and reversal of hepatic steatosis [nonalcoholic fatty liver disease (NAFLD)]/nonalcoholic steatohepatitis (NASH) making it an attractive option for the treatment of many patients with T2DM, particularly those at risk for CV events. In this review, we examine the CV, glycaemic and other metabolic benefits of pioglitazone and provide strategies to maximize the drug’s benefit: risk ratio.

CV benefit

A substantial body of evidence, including large randomized prospective CV outcome trials,^{16,17,21–23} real-world observational studies^24–26 and smaller studies of regression of coronary atherosclerosis ²⁷ and carotid intima thickness, ²⁸ has demonstrated that pioglitazone reduces both atheroma progression and related CV events. These investigations were initiated because of a substantial literature dating back many decades that has linked insulin resistance to premature coronary heart disease (CHD). ²⁹ These data begged the question as to whether reducing insulin resistance with an insulin-sensitizing drug would provide a CV benefit. The question is also of historical importance, since the notion that a diabetes drug could reduce CV events had eluded investigators for years. In PROactive, 5238 T2DM patients with a prior CV event were randomized to pioglitazone or placebo and followed for a mean of 2.9 years. ¹⁶ Although the primary endpoint, a broad composite that included leg revascularization procedures, fell short of statistical significance [hazard ratio (HR) = 0.90, p = 0.09], the ‘main secondary endpoint’, MACE, was significantly reduced (HR = 0.84, p = 0.027) (Figure 1), on par with the effect size in the aforementioned recent positive trials of newer glucose-lowering agents.^11–15 In PROactive participants with a prior MI (n = 2445) or prior stroke (n = 948) pioglitazone therapy were associated with robust 28% and 47% reductions in recurrent MI ³⁰ and recurrent stroke, ³¹ respectively. The primary endpoint in PROactive ¹⁶ should be interpreted in the context that leg revascularization historically has not been included as an endpoint in CV outcome trials since it is refractory to antihypertensive, lipid-lowering and glucose-lowering therapy.^32,33 Consistent with PROactive, a meta-analysis of published pioglitazone studies and reported to the Food and Drug Administration (FDA) demonstrated a 25% reduction in CV events.^21,22

OPEN IN VIEWER

Figure 1.

(a) Kaplan–Meier plot of time to MACE endpoint (cardiovascular mortality, non-fatal MI, non-fatal stroke) in T2DM patients treated with pioglitazone (PIO) or placebo (Plc) in PROactive. Redrawn with permission from Dormandy et al. ¹⁶ (b) Pioglitazone reduces recurrent MI in diabetic patients with a previous MI in PROactive. Redrawn with permission from Erdmann et al. ³⁰ (c) Pioglitazone reduces recurrent stroke in diabetic patients with a previous stroke in PROactive. Redrawn with permission from Wilcox et al. ³¹ (d) Meta-analysis of all published studies (excluding PROactive) in which the effect of pioglitazone versus placebo or active comparator on cardiovascular events is examined. Redrawn with permission from Lincoff et al. ²¹

Based upon (1) evidence that insulin resistance was a strong risk factor for stroke as well CHD, ² (2) the consistently positive results observed in these CV outcome trials,^{16,17,21,22,24–28} and (3) the reduction in recurrent stroke (by 47%) and MI (by 28%) in T2DM individuals in PROactive, ¹⁶ the National Institutes of Health initiated the IRIS study. ¹⁷ In 3876 non-diabetic, insulin-resistant individuals with a recent transient ischaemic attack (TIA) or stroke, pioglitazone reduced fatal/non-fatal stroke or MI by 24% (p = 0.007) over a mean of 4.8 years (Figure 2). ¹⁷ In a follow-up report from this study, ³⁴ pioglitazone reduced the risk of any stroke by 25% (p = 0.01) and decreased the risk of acute coronary syndrome by 29% (p = 0.02), with most of the drug’s effects on type 1 MIs (HR = 0.62, p = 0.03), particularly large infarcts (HR = 0.44, p = 0.02). ³⁵ These results compare favourably with results obtained with aspirin and anti-platelet drugs,^36–38 as well as with statins, ³⁹ which are now widely used for stroke prevention.^36–38 Notably, the positive beneficial CV effects of pioglitazone in all of these studies occurred on the background of widespread use of evidence-based CV therapies including anti-platelet agents suggesting that pioglitazone can effectively address ‘residual CV risk’.

OPEN IN VIEWER

Figure 2.

Effect of pioglitazone versus placebo on recurrent stroke and myocardial infarction in the Insulin Resistance Intervention after Stroke (IRIS) study. Drawn from the data in Kernan et al. ¹⁷

Observational ‘real-world’ data also support the CV benefits of pioglitazone. For example, a retrospective analysis of 91,511 patients in the UK Research General Practice Database (GPRD) who were followed for 7.1 years demonstrated that pioglitazone decreased all-cause mortality by 39% compared with metformin. ⁴⁰ In a separate analysis of 27,457 GPRD patients who had a second agent added to metformin monotherapy, pioglitazone therapy was associated with a significantly decreased HR for all-cause mortality (HR = 0.71) and the combined endpoint of all-cause mortality/major adverse CV events (HR = 0.75). ²⁴ In a more recent observational study, ²⁵ pioglitazone significantly reduced both CV (HR = 0.58) and non-CV (HR = 0.63) mortality in a large (n = 62,266) European cohort of diabetic patients. In a study which compared 56,536 patients with T2DM who were first-time users of pioglitazone or insulin, propensity scores showed a 67% reduction in all-cause mortality in favour of pioglitazone. ²⁶ In a meta-analysis of nine randomized controlled trials, pioglitazone significantly reduced the risk of major CV events in patients with diabetes [HR = 0.83, 95% confidence interval (CI) = 0.72–0.97] and prediabetes or insulin resistance (HR = 0.77, 95% CI = 0.64–0.93). ⁴¹ The results of this meta-analysis are consistent with a previous one by Lincoff et al. ²¹ Finally, consistent with the IRIS study, another retrospective study from the UK using Clinical Practice Research Datalink (CPRD) found a HR of 0.63 for incident stroke in T2DM patients who were users of pioglitazone versus other glucose-lowering drugs. ⁴²

Smaller mechanistic studies are consistent with the findings from these large prospective and observational studies and meta-analyses. In the PERISCOPE study, pioglitazone, compared with glimepiride, retarded the progression of coronary atherosclerosis as measured by intravascular ultrasound (IVUS), ²⁷ while in the CHICAGO study, pioglitazone slowed the rate of increase in carotid intimal thickness, a surrogate measure of atherosclerosis. ²⁸ Pioglitazone has been shown to reduce intracoronary plaque volume in non-diabetic ⁴³ and type 2 diabetic ⁴⁴ subjects and to prevent restenosis after stent placement. ⁴⁵

One negative pioglitazone study to consider is the recent CV outcome trial from Italy, TOSCA-IT. ⁴⁶ A total of 3041 T2DM patients with suboptimal glycaemic control on metformin monotherapy were randomized to either pioglitazone or a sulphonylurea and followed for a mean of 4.8 years. Because only 11% had a prior history of CVD, this was essentially a primary prevention population. The primary outcome (all-cause death, non-fatal MI, non-fatal stroke and urgent coronary revascularization) occurred at a similar frequency between the two groups: pioglitazone 6.8% versus sulphonylurea 7.2% (HR = 0.96, p = 0.40). Unfortunately, the study had some methodological limitations, including its unblinded design and the fact that many patients in the pioglitazone arm had either terminated their participation early (10%) or had stopped the study drug (28%), likely stemming from controversy about the drug’s safety that had arisen during the trial. Furthermore, the CV event rate, 1.5 per 100 person years, was very low, rendering the study greatly underpowered to detect any effective CV events. This issue was underscored by an a posteriori per-protocol analysis focusing on just those patients taking their study drug. Here, a secondary outcome that included peripheral vascular events was significantly reduced by study drug (HR = 0.67, p = 0.03). Of course, this outcome must be interpreted cautiously.

In summary, a large body of evidence from clinical trials to observational studies to mechanistic investigations document that pioglitazone effectively prevents recurrent CV events in both diabetic and non-diabetic individuals, most likely through a beneficial effect on atherosclerosis. The only negative study to our knowledge, TOSCA-IT, involved patients generally without CVD and had multiple interpretative challenges.

Appropriate concern has been raised about ‘heart failure’ (HF) in PROactive. Due to increased renal sodium retention, all TZDs are associated with oedema, ⁴⁷ which is a nonspecific sign of HF by clinicians. Since the HF cases in PROactive were not adjudicated, it remains possible that at least some of the excess events may have reflected cases of oedema without cardiac decompensation. HF is typically an ominous diagnosis in patients with diabetes, with a 5-year mortality in excess of 50%.^1,45 Given that mortality in this cohort of individuals with HF in PROactive was decreased (not increased), albeit not significantly, it is possible even likely that not all (probably many) patients diagnosed with HF actually had HF. In fact, in smaller trials, pioglitazone has been demonstrated to have no deleterious effect on left ventricular (LV) function,^48,49 to actually improve diastolic dysfunction,^48–50 to reduce blood pressure,^48,49 and to increase myocardial insulin sensitivity.^48,49 Interestingly, prior to the concern about HF, observational data suggested that this drug class might actually decrease mortality after HF admissions. ¹⁶ Thus, participants in PROactive who were diagnosed with HF did not experience any increase in CV events compared to placebo-treated individuals. ¹⁶ In IRIS, HF was not increased, although this cohort of cerebrovascular patients had less CHD than did participants in PROactive.^17,51 Also, the IRIS protocol allowed for dose reductions in the setting of significant weight gain or oedema. These observations are consistent with previous findings that pioglitazone improves diastolic function in diabetic rats ⁵² and humans^48,53 by positively influencing matrix remodelling.^52,53 A recent meta-analysis suggests that pioglitazone also reduces both new onset and recurrent atrial fibrillation by 27%. ⁵⁴

Metabolic effects of pioglitazone

The insulin resistance syndrome (IRS), originally referred to as the metabolic syndrome, comprises a cluster of cardiometabolic disorders, each representing an independent CV risk factor. ² Pioglitazone improves each component of the IRS (Table 1) (reviewed in previous studies^2,55–58). It enhances insulin sensitivity and effectively reduces plasma glucose levels and HbA1c while also lowering blood pressure and having a favourable effect on the plasma lipid profile. The latter includes a reduction in triglycerides and free fatty acids (FFAs), increase in high-density lipoprotein (HDL) cholesterol and conversion of small dense low-density lipoprotein (LDL) particles to larger, more buoyant, less atherogenic ones. The drug also shifts fat from visceral abdominal depots, from liver and from skeletal muscle to subcutaneous abdominal depots,^59–62 thereby ameliorating lipotoxicity.^2,55,63–66 It normalizes adipocytokine secretion, especially adiponectin, improves endothelial dysfunction and reduces circulating concentrations of the procoagulant plasminogen activator inhibitor-1 and the pro-inflammatory mediator C-reactive protein (CRP).^66,67 Although pioglitazone improves multiple CV risk factors, both preclinical^68–70 and clinical^{16,17,21–23,27,28,71}, data suggest that pioglitazone exerts direct anti-atherogenic effects on the arterial wall.

OPEN IN VIEWER

Table 1.

Effect of pioglitazone on established CV risk factors.

CV risk factors	Effect of pioglitazone
Obesity (visceral)	Improves – redistributes fat a
Hypertension	Decreases BP
Hypertriglyceridaemia	Decreases TG
Low HDL cholesterol	Increases HDL
Small dense LDL particles	Converts to larger more buoyant LDL
Endothelial dysfunction	Improves
Hyperglycaemia	Durable decrease in HA1c
Inflammation (hsCRP)	Reduces
Lipotoxicity	Reverses
NASH/NAFLD	Improves
PAI-1	Reduces
Hyperinsulinaemia	Decreases
Insulin resistance	Improves

CV: cardiovascular; BP: blood pressure; TG: triglyceride; HDL: high-density lipoprotein; LDL: low-density lipoprotein; hsCRP: high-sensitive C-reactive protein; NASH/NAFLD: nonalcoholic steatohepatitis/nonalcoholic fatty liver disease.

a

Although subjects treated with pioglitazone may gain weight, visceral, hepatic and muscle fat content are decreased.

Pioglitazone transacts its effects through activation of the nuclear hormone receptor peroxisome proliferator–activated receptor-gamma (PPARγ). ⁷² PPARγ receptors are expressed in endothelial cells, arterial smooth muscle cells and monocytes/macrophages, providing a pathway for direct anti-inflammatory, antioxidant and other protective actions of pioglitazone.^73–77 Pioglitazone is the only true insulin-sensitizing antidiabetic agent ⁷⁸ and insulin resistance has been independently associated with atherosclerotic CVD in many cross-sectional and prospective studies.^{2,29,58,79–84}

Pioglitazone is a potent insulin sensitizer

The core pathophysiologic defects in T2DM are insulin resistance in muscle and liver and beta-cell failure.^2,55,85 Collectively, these three pathophysiologic disturbances have been termed the TRIUMVIRATE.^2,55 Insulin resistance in liver results in excess glucose production during the sleeping hours and is the primary factor responsible for fasting hyperglycaemia, while insulin resistance in muscle is the primary factor responsible for postprandial hyperglycaemia. Impaired suppression of hepatic glucose production and reduced liver glucose uptake following a meal also contribute to the postprandial hyperglycaemia.^55,85 Progressive beta-cell failure^55,85–88 accentuates the insulin resistance in liver and muscle. In addition, the adipocyte is resistant to insulin,^2,55,85 resulting in accelerated lipolysis and increased circulating plasma FFA concentrations.^89,90 Elevated plasma FFA in turn exacerbate the muscle insulin resistance, ⁹¹ stimulate hepatic gluconeogenesis and inhibit hepatic glucose uptake ⁹² and impair beta-cell function. ⁹³ Pioglitazone improves insulin sensitivity in skeletal^{2,55,72,94–96} and cardiac^48,49 muscle, in liver ⁹⁷ and in adipose tissue ⁹⁸ via multiple mechanisms: PPARγ activation, stimulation of the insulin signal transduction system, improved glucose transport/glycogen synthesis/glucose oxidation, increased mitochondrial function, reduced plasma FFA levels and reversal of lipotoxicity.^2,55,63–66

Pioglitazone improves beta-cell function

Insulin resistance is the earliest detectable disturbance in the natural history of T2DM.^2,55,85 However, overt diabetes does not develop in the absence of beta-cell failure and progressive decline in insulin secretion.^2,55,85–88 Although not well appreciated, TZDs, including pioglitazone, in addition to their insulin-sensitizing action, exert a potent effect to preserve beta-cell function^99,100 and durability of glycaemic control has been demonstrated in eight long-term, double-blind, placebo-controlled or active comparator studies for up to 5 years (reviewed by DeFronzo ⁵⁵ ). Multiple studies performed in subjects with impaired glucose tolerance (IGT) also have demonstrated a potent action of TZDs to augment beta-cell function (reviewed in previous studies^86,100,101). For example, in the ACT NOW study,^100,102 conversion of IGT to T2DM was reduced by 72% and improvement in the insulin secretion/insulin resistance (disposition) index (gold standard measure of beta-cell function) was the strongest predictor of diabetes prevention. ¹⁰⁰ The improvement in beta-cell function is related to stimulation of PPARγ receptors on the beta cell, enhanced beta-cell sensitivity to glucose and reversal of lipotoxicity.^55,103

Pioglitazone improves NASH/NAFLD

NAFLD has reached epidemic proportions in the United States and worldwide ¹⁰⁴ and is the precursor for NASH. ¹⁰⁵ Diabetic patients with NASH are at high risk for cirrhosis and hepatocellular carcinoma. ¹⁰⁶ Patients with NAFLD/NASH are markedly resistant to insulin, often have the metabolic syndrome and are also at increased risk for CVD.^107–110 Because pioglitazone improves insulin sensitivity, corrects multiple components of the IRS, ameliorates lipotoxicity and protects against atherosclerotic CVD, it would be an excellent agent for the treatment of NAFLD and NASH. Indeed, multiple studies have demonstrated that pioglitazone consistently reduces hepatic fat content and reverses hepatic fibrosis.^59–62 No other antidiabetic agent other than rosiglitazone, a TZD, has shown benefit in the treatment of NAFLD/NASH.^111,112

Safety concerns

Summary

As we transition to a new evidence-based era of T2DM management in patients with CVD, ¹⁵ it is imperative that we choose therapies that not only improve glycaemic control but also improve CV outcomes–the latter representing the greatest cause of mortality in this population. Pioglitazone has been shown to reduce MACE (MI, stroke and CV mortality) in multiple studies including PROactive, ¹⁶ IRIS, ¹⁷ meta-analyses of multiple prospective studies;^21,22 to reduce CV events and mortality in several large observational studies;^24–26,40 to retard the anatomical progression of coronary and carotid atherosclerosis in PERISCOPE, ²⁷ Chicago ²⁸ and ACT NOW (Table 2). ¹⁰² Pioglitazone is the only available insulin-sensitizing agent and has a potent beneficial effect to improve and preserve beta-cell function, leading to a durable reduction in HbA1c (Table 2). Pioglitazone also corrects multiple components of the metabolic syndrome and is an effective treatment for NASH/NAFLD. Side effects remain a concern but can be mitigated by optimizing dosing strategies and combining therapy with other medications (metformin, SGLT2i, GLP-1 RA) that promote weight loss and sodium excretion. The benefit to risk ratio of pioglitazone is very favourable when caution is employed to avoid the known side effects of the drug (Table 2). Moreover, pioglitazone is now generically available and some 50 times less expensive than many branded glucose-lowering drugs with recent CV benefits. It, therefore, represents a highly affordable option for the treatment of patients with T2DM, especially those with prevalent CVD.

OPEN IN VIEWER

Table 2.

Benefits and risk associated with pioglitazone therapy.

Benefit	Risk
Potent, durable HbA1c reduction	Fat weight gain – but decreases visceral, hepatic and muscle fat content
Low risk of hypoglycaemia	Fluid retention/heart failure
Improves insulin resistance	Bone fractures (distal long bones, trauma related)
Improves beta-cell function	Bladder cancer (not established)
Prevents IGT progression to T2DM
Improves cardiovascular risk factors (HDL, triglyceride, blood pressure, inflammation)
Reduces microalbuminuria
Decreases cardiovascular events in high-risk diabetic patients (PROactive, IRIS, meta-analysis)
Reduces cardiovascular events in diabetic patients with chronic kidney disease
Improves endothelial dysfunction
Improves NASH/NAFLD

IGT: impaired glucose tolerance; T2DM: type 2 diabetes mellitus; HDL: high-density lipoprotein; NASH/NAFLD: nonalcoholic steatohepatitis/nonalcoholic fatty liver disease.