Researchers Call for Transparency When Black Box Warnings Are Removed

Posted 24 March 2016 By Michael Mezher


A team of Harvard researchers are calling for the US Food and Drug Administration (FDA) to develop guidelines detailing its decision-making process for removing black box warnings from drug labels.

In a recent article appearing in Drug Safety, James Yeh, Ameet Sarpatwari and Aaron Kesselheim of Bringham and Women's Hospital and Harvard Medical School argue that by making the process to remove such warnings from drugs "uniform and transparent," FDA could increase public confidence in its ability to safeguard public health.

When FDA learns of new safety issues for an approved product, it can add warnings to the product's label, or oblige the manufacturer fulfill certain postmarketing requirements such as conducting additional studies or implementing a Risk Evaluation and Mitigation Strategy (REMS).

Black boxed warnings are the strongest type of warning FDA can place on a product's label, and are reserved for serious or potentially life threatening adverse events.

The researchers say these warnings are a "powerful tool" for FDA to communicate risk information, but worry that a lack of insight into the agency's decision-making process for removing boxed warnings could create confusion among drugmakers, and create skepticism among patients and healthcare providers.

For example, the researchers point to two recent cases where drugmakers petitioned FDA to lift a black box warning from one of their products.

In one case, Pfizer unsuccessfully petitioned FDA to lift a boxed warning for severe psychiatric events, including suicidality, from its smoking cessation drug Chantix. Five years after FDA recommended a boxed warning on the label for Chantix, Pfizer asked the agency to remove the warning based on data from five short-term observational studies.

However, after convening a joint meeting of its Psychopharmacology Drugs and Risk Management Advisory Committees, FDA opted to maintain the black box warning for the drug.

In another case, GlaxoSmithKline (GSK) managed to persuade FDA to remove the black box warning for its blockbuster diabetes drug, Avandia, using an independent analysis of data from a large randomized open-label controlled trial.

Establishing Guidelines

In both cases, the researchers say FDA's decision to impose a black box warning was based on less evidence than was necessary to support the removal of a warning. This is necessary, they argue, as FDA must act quickly to protect patients when safety issues emerge.

However, when it comes to removing a warning, the researchers argue that FDA should require more substantial evidence than it used to impose the warning in the first place.

"To avoid frequent flip-flopping, we believe it is ethically justified for the FDA to require a greater burden of proof, or a greater level of certainty provided by the evidence at hand, to justify removal of a boxed warning than it did to impose one," they write.

The type of study necessary to support the lifting of a black box warning, the researchers say, should be dependent on factors related to the safety issues the warning addresses. For example, a randomized controlled trial would be more appropriate when trying to study an adverse event with a small effect size, whereas, observational studies would be better suited to detecting rare adverse events.

The researchers also call on FDA "to provide greater clarity about the source of data used to impose a boxed warning."

Before removing a boxed warning, the researchers suggest that FDA should hold a public hearing to get input from experts. When a boxed warning is removed, the researchers argue that FDA should share its rationale publicly. By taking these steps, the researchers say FDA could increase public confidence in, and reduce skepticism about, its decision-making process.

Ethical and Practical Considerations in Removing Black Box Warnings from Drug Labels

Share this article:

Categories: Drugs, Postmarket surveillance, News, US, FDA

Tags: Boxed warning, Black box warning, Chantix, Avandia

Regulatory Exchange: Latest Updates From the Community