GSK Avandia CV Risk Concealment (1999–2010)

Introduction

Avandia (rosiglitazone) is a thiazolidinedione-class diabetes drug approved by the FDA in May 1999. Marketed by GlaxoSmithKline (GSK), it was prescribed to millions of patients with Type 2 diabetes to improve insulin sensitivity. By the mid-2000s it had annual sales exceeding $3 billion. In 2007, a landmark meta-analysis by Cleveland Clinic cardiologist Steven Nissen shook the global diabetes treatment community and triggered one of the largest pharmaceutical fraud investigations in US history.

The Nissen Meta-Analysis (2007)

In May 2007, the New England Journal of Medicine published a meta-analysis by Dr. Steven Nissen and Kathy Wolski that pooled data from 42 randomised clinical trials involving Avandia. The analysis found that rosiglitazone was associated with a statistically significant 43% increase in the risk of myocardial infarction (heart attack) and a 64% increase in cardiovascular death risk (though the latter did not reach conventional statistical significance). The publication triggered immediate regulatory review in both the US and Europe.

The FDA convened an advisory committee in July 2007, which voted that Avandia posed a cardiovascular risk and that stronger warnings were needed, though it did not recommend withdrawal. The European Medicines Agency suspended Avandia in September 2010, effectively ending its sale in Europe.

Senate Finance Committee Investigation

The Senate Finance Committee investigation, led by Senators Max Baucus and Charles Grassley and concluded in February 2010, found that GSK had possessed internal data suggesting Avandia''s cardiovascular risk as early as 1999 — the year of approval. An internal GSK meta-analysis completed in 2006, a year before Nissen''s publication, had reached similar conclusions about elevated myocardial infarction risk. GSK had not submitted this internal analysis to the FDA.

The Senate report characterised GSK''s conduct as a failure to disclose material safety information to regulators. Former GSK CEO Jean-Pierre Garnier and other executives faced questioning in the investigation.

FDA Restrictions and Reversal

In September 2010, the FDA restricted Avandia to patients who had already been taking it and who could not be adequately controlled with other medications — effectively limiting new prescriptions to near-zero. The agency also required a Risk Evaluation and Mitigation Strategy (REMS). In 2013, following a re-analysis of the RECORD trial data, the FDA lifted the restrictions after concluding the cardiovascular risk was lower than previously estimated — a conclusion that itself generated scientific controversy.

The $3 Billion DOJ Settlement

In July 2012, GSK agreed to a $3 billion settlement with the US Department of Justice — at the time the largest criminal and civil healthcare fraud settlement in US history. The settlement covered multiple allegations including: illegal promotion of Avandia for uses not approved by the FDA, failure to disclose safety data on Avandia to the FDA, and kickbacks to physicians. GSK pleaded guilty to criminal misdemeanour charges.

The Avandia-specific component addressed the failure to disclose cardiovascular safety data and the promotion of the drug to physicians in ways that minimised the risk signal.

Verdict

Confirmed. Senate Finance Committee findings, DOJ settlement ($3 billion, 2012), and GSK''s own internal documents confirm that the company possessed cardiovascular risk data that was not disclosed to regulators in a timely manner. The Nissen meta-analysis confirmed the signal that GSK''s internal 2006 analysis had already identified. Millions of diabetes patients were exposed to elevated cardiovascular risk without adequate warning.

How GSK Became Aware of Cardiovascular Risk

Rosiglitazone's mechanism of action involves improving insulin sensitivity through activation of peroxisome proliferator-activated receptors (PPARγ). However, thiazolidinediones carry a known liability: they cause weight gain, fluid retention, and changes in lipid profiles—all established cardiovascular risk factors. From the drug's approval in 1999 onward, GSK collected adverse event reports through its postmarketing surveillance systems. These reports documented heart attacks, strokes, and congestive heart failure in patients taking Avandia. Concurrently, clinical trials examining Avandia's efficacy also recorded cardiovascular events. Rather than aggregating and analyzing this data systematically and disclosing novel findings to the FDA, GSK allowed the safety signal to accumulate internally for years. This pattern was not unique to GSK—postmarketing surveillance often lags in analysis—but what distinguished this case was the company's subsequent decision to suppress the analysis once it was completed. By the early 2000s, enough adverse event reports existed that internal analyses became standard procedure. Yet these analyses, when completed, contradicted the drug's approved safety profile and remained within company walls.

The 2006 Internal Meta-Analysis: The Smoking Gun

In 2006, GSK commissioned an internal meta-analysis that pooled available trial data on rosiglitazone and cardiovascular outcomes. This analysis reached conclusions similar to what Steven Nissen would independently publish one year later: rosiglitazone carried an elevated risk of myocardial infarction. The existence of this internal 2006 analysis became the central finding of the Senate Finance Committee investigation. The Committee noted that the analysis represented material new safety information that GSK was obligated under FDA regulations to report to the agency. Instead of submitting the analysis through postmarketing safety update channels (required by law for significant new findings), GSK kept it internal. The company possessed financial incentives to suppress disclosure: Avandia was among its top-selling medications, generating over $3 billion in annual revenue at its peak. An internal finding of elevated MI risk would have threatened sales and triggered mandatory label changes and potentially restrictions. The analysis was discoverable only through subsequent litigation and congressional investigation. The Senate Finance Committee's report explicitly stated that GSK executives were aware of this analysis and its implications yet chose not to escalate the safety signal to regulators. This constituted both a violation of postmarketing reporting obligations and, in the Committee's view, a conscious choice to prioritize commercial interests over patient safety.

The Senate Finance Committee Investigation (2010)

When the Senate Finance Committee, led by Senators Max Baucus (D-MT) and Charles Grassley (R-IA), investigated GSK's conduct in 2009–2010, committee investigators used subpoena power to compel internal company documents. These documents revealed the full scope of GSK's knowledge. The Committee concluded that GSK had possessed information about Avandia's cardiovascular risks as early as 1999—the year of FDA approval. The company's internal 2006 meta-analysis, completed before Nissen's peer-reviewed publication, demonstrated that GSK knew of the risk signal independently. The Committee found that GSK had failed to meet its obligation under FDA regulations to report new safety information to the agency as it became available. Additionally, the Committee documented that GSK had promoted Avandia to physicians through marketing materials and sales presentations that downplayed or omitted cardiovascular risks, even as internal data suggested otherwise. The Committee's February 2010 report became public evidence of what it characterized as "systematic" suppression of safety information. The investigation's conclusions provided the factual foundation for the Department of Justice's decision to pursue criminal charges against GSK in addition to civil fraud penalties.

The Regulation-Evidence Trap: When 2013 Re-Analysis Did Not Vindicate Non-Disclosure

A crucial counter-argument to the Avandia case centers on the FDA's 2013 decision to lift its 2010 restrictions. In 2013, following re-analysis of the RECORD trial (a large prospective study of over 4,000 patients designed to measure cardiovascular outcomes), the FDA concluded that rosiglitazone's true cardiovascular risk was lower than Nissen's meta-analysis suggested. This finding has led some observers to argue that the initial restrictions were overreaching and that Avandia was safer than the 2007 outcry indicated. However, this retrospective re-assessment does not address the core violation. GSK's obligation was not to predict accurately what future evidence would show, but to disclose to regulators material safety information available to the company as it emerged. Even if rosiglitazone's true cardiovascular risk was substantially lower than the Nissen analysis indicated—a possibility the 2013 re-analysis suggests—this does not retroactively justify GSK's decision to withhold its internal 2006 meta-analysis from the FDA. Non-disclosure of material adverse information is a regulatory breach independent of whether the information ultimately proves overestimated. The 2013 re-analysis addressed the scientific question of rosiglitazone's actual risk; it did not address the legal and ethical question of GSK's conduct. Moreover, the Nissen meta-analysis, though later revised downward, was based on legitimate trial data and represented a genuine scientific concern in 2007. Patients who remained on Avandia during the years of concealment faced uncertainty about risks that the company possessed but did not share—a harm that continues regardless of what later evidence revealed.

Aftermath, Regulatory Precedent, and Patient Impact

The July 2012 DOJ settlement—$3 billion in criminal and civil penalties—became a benchmark for healthcare fraud enforcement for years. Under the settlement, GSK pleaded guilty to criminal misdemeanor charges and agreed to significant compliance obligations, including enhanced pharmacovigilance procedures and mandatory disclosure of safety data to the FDA. The settlement also included $700 million in civil settlements with individual patients who sued GSK for Avandia-related injuries. Beyond financial penalties, the case reshaped pharmaceutical postmarketing surveillance expectations. The FDA strengthened requirements for timely disclosure of internal safety analyses and increased scrutiny of companies' postmarketing data reporting. Diabetes treatment guidelines, which had emphasized thiazolidinediones as first-line agents, were revised to de-emphasize these drugs in favor of safer alternatives such as metformin, GLP-1 agonists, and SGLT2 inhibitors. For the estimated 28 million Americans who took Avandia between 1999 and 2010, the case highlighted a structural vulnerability in pharmaceutical oversight: companies possess access to internal safety data before regulators do, and financial incentives can motivate suppression. Tens of millions of patients worldwide took Avandia during the years GSK concealed its internal risk findings. Many experienced serious cardiovascular events—myocardial infarctions, strokes, congestive heart failure—without knowing that their pharmaceutical company possessed data suggesting such risks. The case remains a textbook example of why independent regulatory oversight and transparency requirements are essential to pharmaceutical safety.