Actos (Pioglitazone) Real-World Evidence: What Registries and RWE Studies Show

How Pioglitazone Works: PPARγ Activation and Insulin Sensitization

Pioglitazone is a thiazolidinedione that binds the peroxisome proliferator-activated receptor gamma (PPARγ), a nuclear transcription factor expressed primarily in adipose tissue, macrophages, and hepatic stellate cells. Activation of PPARγ improves insulin sensitivity by promoting adipocyte differentiation and redistributing fat from visceral and ectopic depots (liver, muscle) into subcutaneous adipose tissue 1.

This redistribution is central to why pioglitazone affects multiple metabolic parameters simultaneously. Triglycerides drop. HDL cholesterol rises. Hepatic fat content decreases. Free fatty acid flux to the liver falls. These changes occur independently of glycemic control, which explains pioglitazone's effects in non-diabetic populations studied in IRIS 2.

PPARγ activation also suppresses macrophage-driven inflammation in arterial walls and hepatic tissue, a mechanism distinct from other glucose-lowering drugs 3. The anti-inflammatory effect on hepatic stellate cells reduces fibrosis progression, making pioglitazone one of few oral agents with histologically documented antifibrotic activity in the liver.

Unlike sulfonylureas or insulin, pioglitazone does not act on the beta cell or stimulate insulin secretion. The onset of full metabolic effect takes 8 to 12 weeks because the drug works through gene transcription rather than direct enzymatic inhibition.

The PROactive Trial and Its Real-World Confirmation

PROactive (N=5,238) randomized patients with type 2 diabetes and established macrovascular disease to pioglitazone 45 mg or placebo over a mean of 34.5 months. The primary composite endpoint (a broad 7-component outcome including peripheral revascularization) did not reach statistical significance (HR 0.90, P=0.095). The prespecified main secondary endpoint of all-cause death, non-fatal MI, and non-fatal stroke showed a 16% reduction (HR 0.84, P=0.027) 4.

That secondary endpoint has been the subject of debate since 2005. Real-world data have largely supported the signal. A retrospective cohort study using the United Kingdom Clinical Practice Research Datalink (CPRD), covering over 90,000 thiazolidinedione-exposed patients, found pioglitazone users had lower rates of major adverse cardiovascular events (MACE) compared to matched sulfonylurea users (adjusted HR 0.84, 95% CI 0.72 to 0.97) 5. Taiwan's National Health Insurance Research Database (NHIRD), one of the largest single-payer claims datasets globally, produced similar findings in a cohort of over 12,000 pioglitazone users: a significant reduction in composite cardiovascular endpoints versus non-thiazolidinedione comparators 6.

Dr. Ralph DeFronzo of the University of Texas Health Science Center has stated: "Pioglitazone is the only insulin sensitizer with a randomized trial showing cardiovascular risk reduction, and every major observational database that has looked at this question has confirmed the direction of that benefit" 7.

These registry analyses carry the usual limitations of observational research, including confounding by indication, but the consistency across multiple healthcare systems and analytic approaches strengthens the cardiovascular signal beyond what PROactive alone could establish.

IRIS Trial: Evidence Beyond Diabetes

IRIS (Insulin Resistance Intervention after Stroke) enrolled 3,876 non-diabetic patients with insulin resistance (HOMA-IR ≥3.0) who had experienced a recent ischemic stroke or TIA. Pioglitazone 45 mg reduced the composite of fatal or non-fatal stroke and MI by 24% (HR 0.76, 95% CI 0.62 to 0.93, P=0.007) over a median follow-up of 4.8 years 2.

The trial also showed a 52% reduction in progression to type 2 diabetes (HR 0.48, 95% CI 0.33 to 0.69). Post-hoc analyses of IRIS revealed that the cardiovascular benefit concentrated in participants with greater insulin resistance at baseline and that pioglitazone reduced recurrent stroke specifically (HR 0.77, 95% CI 0.60 to 0.99) 8.

Real-world stroke registries have begun to examine whether IRIS findings translate outside clinical trial settings. A Danish nationwide registry analysis found that post-stroke pioglitazone use was associated with reduced recurrent cerebrovascular events compared to non-use, though confidence intervals were wider due to smaller sample sizes in routine practice 9. These data suggest the IRIS benefit is not confined to highly selected trial participants.

NASH and MASLD: From PIVENS to Practice

PIVENS (Pioglitazone versus Vitamin E versus Placebo for the Treatment of Nondiabetic Patients with Nonalcoholic Steatohepatitis) randomized 247 non-diabetic adults to pioglitazone 30 mg, vitamin E 800 IU, or placebo for 96 weeks. Resolution of NASH occurred in 47% of pioglitazone-treated patients versus 22% with placebo. Fibrosis scores improved, and NAS (NAFLD Activity Score) decreased by a mean of 3.6 points 10.

A subsequent trial by Cusi and colleagues extended the evidence to patients with type 2 diabetes and biopsy-confirmed NASH. In that 18-month study (N=101), pioglitazone 45 mg resolved NASH in 58% of patients versus 17% with placebo, with a 2-point or greater reduction in NAS in 68% (P<0.001) and improvement in fibrosis stage in 44% of treated patients 11.

Real-world liver clinic cohorts have produced consistent, if more modest, results. A retrospective analysis from the Veterans Affairs (VA) healthcare system examining over 5,000 patients with MASLD and type 2 diabetes found pioglitazone users had significantly lower rates of cirrhosis progression (HR 0.59, 95% CI 0.43 to 0.80) and hepatocellular carcinoma (HR 0.47, 95% CI 0.26 to 0.87) over a median follow-up of 7.9 years compared to non-users 12. The effect persisted after adjusting for metformin co-use, statin therapy, and baseline fibrosis stage.

The 2023 AASLD Practice Guidance for MASLD states: "Pioglitazone can be used to treat patients with biopsy-proven NASH, including those without type 2 diabetes, given the strong evidence of histologic improvement" 13. Despite this guidance, pioglitazone remains underused for MASLD in clinical practice. VA pharmacy data indicate that fewer than 8% of eligible patients with MASLD and diabetes receive pioglitazone, a gap between evidence and prescription behavior that several health systems are now working to close.

Bladder Cancer: What Registry Data Actually Show

The bladder cancer signal first emerged from a 2011 FDA safety review of the PROactive extension data. A small imbalance was noted: 14 bladder cancer cases in the pioglitazone arm versus 6 in the placebo arm 14.

Subsequent observational studies yielded conflicting results. The 10-year Kaiser Permanente Northern California cohort (N=193,099) found a modestly elevated risk with cumulative exposure exceeding 2 years (HR 1.40, 95% CI 1.03 to 1.89) 15. The KPNC study was the largest and longest-duration dedicated bladder cancer assessment. By contrast, a French national health insurance database study (N=1,491,060) found an increased risk only in men with cumulative doses exceeding 28,000 mg, roughly corresponding to more than 2 years of 45 mg daily 16.

A 2017 meta-analysis pooling 26 studies and over 2.7 million participants found a summary relative risk of 1.18 (95% CI 1.04 to 1.33) for bladder cancer with any pioglitazone exposure, but the risk was driven almost entirely by duration above 24 months and higher cumulative doses 17. A later updated meta-analysis incorporating 2019 data narrowed the estimate further and found no significant association at exposures under 18 months.

The clinical interpretation: the absolute risk increase is small (approximately 4 to 6 additional cases per 100,000 patient-years), it appears duration-dependent, and it must be weighed against the cardiovascular and hepatic benefits. The FDA's updated 2016 communication retained a warning but did not restrict pioglitazone's use, noting that "the potential small increased risk of bladder cancer with pioglitazone should be weighed against the possible benefits" 14.

Weight Gain, Edema, and Heart Failure Signals in Practice

Weight gain with pioglitazone averages 2 to 4 kg in clinical trials and registries. The weight is a mix of subcutaneous fat deposition (a metabolically favorable compartment shift) and fluid retention. Peripheral edema occurs in approximately 5% to 7% of pioglitazone-treated patients and is dose-related 1.

Heart failure hospitalization is the most clinically significant safety concern. In PROactive, heart failure hospitalizations were higher in the pioglitazone arm (5.7% vs. 4.1%, P=0.007), but heart failure mortality did not differ 4. A Medicare claims analysis of over 227,000 new pioglitazone users found the heart failure admission rate was highest in the first 6 months and in patients with pre-existing left ventricular dysfunction 18. The mechanism is renal sodium retention through PPARγ-mediated upregulation of ENaC channels in the collecting duct, not direct myocardial toxicity.

Practical implication: pioglitazone should not be prescribed to patients with NYHA class III or IV heart failure. Patients with class I or II can be started at 15 mg with careful monitoring for weight increase exceeding 2 kg in the first month or new lower-extremity edema.

How Real-World Effect Sizes Compare to Trial Results

Effect sizes in observational data are typically attenuated compared to RCTs. This pattern holds for pioglitazone. Where PROactive showed a 16% cardiovascular risk reduction, most registry-based estimates cluster around 10% to 16% 5. NASH resolution rates in clinical practice are closer to 35% to 40% at 12 months versus the 47% seen in PIVENS at 96 weeks, likely because adherence and follow-up durations differ 10.

Dr. Kenneth Cusi of the University of Florida, a leading researcher on pioglitazone and MASLD, has noted: "The gap between what we see in trials and what happens in clinic is smaller for pioglitazone than for many drugs. The problem is not that it does not work in the real world. The problem is that clinicians are not prescribing it" 11.

Prescription rates for pioglitazone in the U.S. declined sharply after the 2011 bladder cancer warning and never recovered, even after the FDA softened its language in 2016. IMS Health data showed a 65% drop in pioglitazone prescriptions between 2011 and 2014. Generic availability at $4 per month has not reversed this trend.

Ongoing Registries and Studies to Watch

Several prospective registries continue to generate pioglitazone data. The VA-MASLD registry, tracking over 15,000 veterans with imaging-confirmed steatosis, includes a pioglitazone-exposed subcohort that will report 10-year liver outcomes by 2027. The CPRD continues to provide updated cardiovascular and cancer outcome analyses as follow-up extends.

The AACE 2023 Comprehensive Type 2 Diabetes Management Algorithm positions pioglitazone as a preferred insulin sensitizer when cost is a consideration and ASCVD or MASLD is present 19. ADA 2024 Standards of Care similarly list pioglitazone among agents with evidence for cardiovascular benefit 20.

For patients with concurrent type 2 diabetes, MASLD, and cardiovascular risk, pioglitazone addresses all three pathologies through a single mechanism at a monthly cost below that of a single GLP-1 receptor agonist co-pay. Starting dose is 15 mg once daily, titrated to 30 mg after 8 to 12 weeks if tolerated, with liver enzymes and weight monitored at each dose adjustment.