Actos (Pioglitazone) FAERS Safety Signals

How FAERS Tracks Pioglitazone Adverse Events

The FDA Adverse Event Reporting System (FAERS) is a passive surveillance database that collects voluntary reports from clinicians, patients, and manufacturers. For pioglitazone, FAERS data have driven three major label changes since 2004.

FAERS quarterly data files show pioglitazone consistently ranking among the top 100 drugs by report volume in the diabetes therapeutic category. Between 1999 and 2024, the database accumulated reports spanning cardiac failure, neoplasms, musculoskeletal injuries, and ocular disorders 1. The system uses the Medical Dictionary for Regulatory Activities (MedDRA) preferred terms, and pioglitazone's signal profile clusters around four primary system organ classes: cardiac disorders, neoplasms (bladder), musculoskeletal disorders, and eye disorders.

Passive surveillance cannot establish causation. FAERS data triggered the formal pharmacoepidemiologic studies and randomized trial analyses that eventually confirmed or refuted each signal. The FDA's Sentinel System, launched in 2008, later provided active surveillance capability using claims data from over 100 million covered lives, and pioglitazone was among the first drugs evaluated through this infrastructure 2.

The Congestive Heart Failure Signal and Boxed Warning

Pioglitazone causes dose-dependent fluid retention that can precipitate or worsen heart failure. The FDA added a boxed warning in August 2007 after reviewing both clinical trial data and FAERS case series.

In the PROactive trial (N=5,238), pioglitazone-treated patients experienced heart failure requiring hospitalization at a rate of 11% versus 8% in the placebo group (HR 1.41 to 95% CI 1.10 to 1.80, P=0.007) 3. This 3-percentage-point absolute increase translated to a number needed to harm of approximately 33 over the 34.5-month follow-up period. The mechanism involves PPARγ-mediated sodium reabsorption in renal collecting ducts, distinct from direct myocardial toxicity.

FAERS cardiac failure reports for pioglitazone peaked between 2005 and 2008, corresponding to peak prescribing volume (exceeding 10 million US prescriptions annually). The boxed warning text states: "Thiazolidinediones, including pioglitazone, cause or exacerbate congestive heart failure in some patients" 4. NYHA Class III and IV heart failure became a contraindication.

Dr. Sanjay Kaul, a member of the FDA's Endocrinologic and Metabolic Drugs Advisory Committee, noted during the 2007 hearing: "The heart failure signal is real and clinically meaningful, but it is mechanistically distinct from ischemic cardiomyopathy. These patients retain fluid; they do not lose myocardium."

Bladder Cancer: The Most Controversial Signal

FAERS reports of bladder neoplasms in pioglitazone users first triggered FDA concern in 2005, leading to a decade of epidemiologic investigation that produced conflicting results.

The initial signal emerged from disproportionality analysis of FAERS data showing a reporting odds ratio above 2.0 for bladder cancer with pioglitazone compared to other oral diabetes drugs. The FDA required Takeda to conduct a 10-year observational study through Kaiser Permanente Northern California (KPNC). Interim 5-year results (2011) showed a statistically significant association for patients with the longest duration of exposure (HR 1.40 to 95% CI 1.03 to 1.90 for >2 years of use) 5.

The FDA issued a safety communication in June 2011 advising against pioglitazone use in patients with active bladder cancer. France suspended the drug entirely that same month. Germany followed. The EMA's Committee for Medicinal Products for Human Use (CHMP) conducted its own review and retained market authorization with a contraindication in active or prior bladder cancer 6.

The completed 10-year KPNC study (2016, N=193,099) found no statistically significant overall association (HR 1.06 to 95% CI 0.89 to 1.26), though a duration-response trend persisted for exposures exceeding 4 years 7. The FDA updated the label in December 2016, softening the language from "may increase risk" to noting the conflicting evidence while retaining the recommendation to avoid use in active bladder cancer.

A 2017 meta-analysis of 26 studies (N>1.6 million patients) calculated a pooled relative risk of 1.14 (95% CI 1.04 to 1.24) for any bladder cancer with pioglitazone exposure, though heterogeneity was substantial (I²=62%) 8. The absolute risk increase, if real, translates to approximately 1 additional case per 3,500 patient-years of treatment.

Fracture Risk in Women

FAERS musculoskeletal reports prompted the FDA to require a label update regarding fracture risk in December 2006, one of the earliest post-market safety actions for pioglitazone.

The signal was confirmed by a pooled analysis of Takeda's randomized trial database showing fracture incidence of 5.1% in women on pioglitazone versus 2.5% on comparators (relative risk 2.23, P<0.001). Fractures occurred predominantly in the distal upper and lower limbs rather than the hip or spine 9. The PROactive trial showed similar findings: fractures in 5.1% of pioglitazone-treated women versus 2.5% of placebo-treated women over a mean 34.5 months.

The mechanism involves PPARγ activation directing mesenchymal stem cells toward adipocyte rather than osteoblast differentiation, resulting in reduced bone formation. This was later confirmed by bone density studies showing 1.5% to 2% annual decreases at the hip and spine in pioglitazone-treated patients 10.

Current labeling recommends assessing fracture risk prior to initiating pioglitazone in women and considering bone density monitoring during long-term therapy. The fracture signal was not statistically significant in men in the original analyses, though subsequent observational studies have suggested a smaller but present male risk.

Macular Edema Reports

Post-market reports of new-onset or worsening macular edema in pioglitazone users led to a Warnings and Precautions addition in 2006.

FAERS cases typically described patients developing visual symptoms 2 to 12 months after starting pioglitazone, with improvement after drug discontinuation 11. Patients with pre-existing diabetic retinopathy appeared disproportionately represented. The proposed mechanism involves increased vascular permeability from fluid retention affecting the retinal vasculature.

A population-based study using British Columbia health administrative data (N=170,000 diabetes patients) found an adjusted OR of 1.3 (95% CI 0.7 to 2.3) for macular edema with thiazolidinediones, suggesting the absolute risk is small but the signal is biologically plausible 12. The label advises patients to report visual changes promptly and recommends ophthalmologic evaluation if symptoms develop.

Hepatotoxicity Monitoring: Lessons from Troglitazone

Pioglitazone's FAERS hepatic signal must be understood in context. Its predecessor troglitazone (Rezulin) was withdrawn in March 2000 after causing 90 cases of liver failure. Pioglitazone's label originally required periodic liver enzyme monitoring based on class effect concern.

Post-market data over 20+ years demonstrated that pioglitazone carries far less hepatic risk than troglitazone. FAERS hepatic failure reports for pioglitazone remained at background rates when adjusted for exposure volume. The FDA relaxed mandatory ALT monitoring requirements in 2013, changing the recommendation from "periodic monitoring" to "obtain liver tests prior to initiation and periodically thereafter based on clinical judgment" 4.

The PIVENS trial (N=247) studying pioglitazone 30 mg for non-alcoholic steatohepatitis (NASH) actually demonstrated ALT normalization in 34% of pioglitazone-treated subjects versus 19% with placebo, suggesting hepatoprotective effects in the setting of fatty liver disease 13.

Weight Gain as a Persistent FAERS Theme

Metabolic adverse events constitute a significant proportion of pioglitazone FAERS reports, with weight gain being the most frequent non-serious event reported.

In the PROactive trial, mean weight gain with pioglitazone was 3.6 kg over 34.5 months compared to 0.4 kg weight loss with placebo 3. The weight gain reflects both fluid retention (approximately 1 to 2 kg) and true adipose tissue expansion, predominantly in subcutaneous rather than visceral compartments. This redistribution from visceral to subcutaneous fat may partially explain pioglitazone's insulin-sensitizing benefit despite overall weight increase.

The clinical relevance of pioglitazone-associated weight gain has intensified in the GLP-1 receptor agonist era. Prescribers now have alternatives that produce weight loss of 10% to 15%, making the 3 to 4 kg weight gain with pioglitazone a more significant comparative disadvantage than when the drug competed primarily against sulfonylureas and insulin.

Current Regulatory Status and Label Summary

Pioglitazone remains FDA-approved and available in the United States, unlike in France where it has been suspended since 2011.

The current US prescribing information carries:

• One boxed warning (heart failure)
• Five Warnings and Precautions sections (bladder cancer, fractures, macular edema, hepatic effects, macroscopic hematuria)
• Contraindication in NYHA Class III/IV heart failure
• Contraindication in active bladder cancer (retained despite 2016 softening of evidence language)

The Endocrine Society's 2022 guidelines for type 2 diabetes management position pioglitazone as a second-line agent when insulin sensitization is the primary goal, noting particular utility in patients with NAFLD/NASH or those unable to afford newer agents 14. The American Diabetes Association's 2024 Standards of Care retain pioglitazone as an option with appropriate risk counseling.

Prescribers should document heart failure risk assessment, bladder cancer history, fracture risk (especially in postmenopausal women), and baseline hepatic function before initiating pioglitazone at the standard starting dose of 15 to 30 mg daily.