OPDP Comes out Swinging With A Warning Letter on Unsubstantiated Efficacy

August 17, 2023By Dara Katcher Levy

Well, if you, like many within industry, felt emboldened to disseminate promotional materials that may push the envelope on efficacy (so long as your risk information was tight) think again!  OPDP is back in the game, letting all the people know that they are back to run the show.

In what may be the surprise “Warning Letter of the Year” (can you tell this blogger was pretty shocked?) OPDP took issue with AstraZeneca’s professional sales aid for Breztri Aerosphere (budesonide, glycopyrrolate, and formoterol fumarate) inhalation aerosol for oral inhalation use.  Why is this Warning Letter so shocking?  Let’s take a walk down OPDP Warning Letter memory lane . . .

Whew!  Pretty egregious stuff, right?

In this most recent Warning Letter, however, OPDP did not take issue with ANY of the safety information presented about Breztri in the sales aid; rather, objections were focused solely on false or misleading claims related to efficacy.  And those efficacy presentations included disclaimers and context, yet FDA found they were insufficient to “mitigate the misleading impression” caused.  Sheesh!

Let’s go over some background information relating to Breztri and the OPDP Warning Letter.

Breztri was approved just over 3 years ago in July 2020 for the maintenance treatment of patients with chronic obstructive pulmonary disease (COPD).  The indication includes a limitation of use that Breztri is not indicated for the relief of acute bronchospasm or for the treatment of asthma.  The drug does not have a Boxed Warning, but has an extensive list of Warnings (17 in total).  The promotional piece at issue is a 12-page brochure intended for healthcare professionals, with the offending presentations appearing on pages 6 and 7.

OPDP first took issue with a headline and graph presented on page 7 of the 12-page brochure.  The headline, “Difference Observed in Time to All-Cause Mortality (Over 52 Weeks)” appears above a graph titled “Secondary Endpoint Study 1:  Time to all-cause mortality in the ITT population” and includes claims that there is an observed relative difference between Breztri v. LAMA/LABA.  The Kaplan Meier graph closely resembles one appearing in the American Journal of Respiratory and Critical Care Medicine Journal, a bi-weekly peer-reviewed medical journal published by the American Thoracic Society.  The page in the brochure calls out that the data were also included in the New England Journal of Medicine.  FDA took issue with the representation that there was a 49% observed relative difference in all cause mortality with Breztri v. LAMA/LABA because the results intended to support that claim come from a study with multiple secondary endpoints and there was a failure to show significance on endpoints ranked higher in the analysis hierarchy.  Despite AstraZeneca’s inclusion of the statement “These results are observational in nature, and any comparisons between treatment arms should be interpreted with caution,” FDA provided its typical response re disclaimer statements – that they do not “mitigate the misleading impression” of the presentation.

The second issue raised by OPDP was that the inclusion of a p value in conjunction with a claim suggested statistical significance when the results were not, in fact, statistically significant.  OPDP states that a p-value is generally understood to indicate statistical significance if it is less than 0.05.  However, in the testing strategy from the trial,

the raw p-value of each hypothesis test was compared to the corresponding critical value to determine whether the test was statistically significant.  As the p-value for the Breztri to ICS/LABA comparison (p=0.02) was greater than the critical value (0.008) for that hypothesis test, the result, per the threshold set by the testing strategy, is not statistically significant.  Therefore, the presentation of these claims (i.e., with a p-value of 0.02) creates the misleading impression that Breztri provides a statistically significant reduction in severe exacerbations compared to ICS/LABA by 20% when this has not been demonstrated.

FDA acknowledged a footnote that explained that the analysis was based on predefined Type-1 error control plan however, as above, FDA stated “this does not mitigate the misleading impression.”

What does this mean?

While one could argue that these presentations are misleading because they overstate the efficacy of Breztri, the presentations are limited to the middle of a 12-page brochure intended for healthcare professionals and the brochure otherwise describes the studies discussed and includes relevant safety information.  The claims are (presumably) not made to patients and each had context that described how the data was interpreted.  This certainly doesn’t seem like promotion that is prioritized for OPDP enforcement– while the drug has many warnings, it is not a Boxed Warning drug or an opioid, it is not subject to a REMS, nor is it used for COVID-19.  As Breztri has been approved for just over 3 years, this is not launch material for a newly approved drug, nor is it a product that has been the subject of previous compliance letters and the promotion is not a “far-reaching” campaign.

In a post-CFL world, these claims walk the line in that they could be “consistent” with the FDA-required labeling in accordance with FDA’s stated 3-factor analysis.  Clearly, OPDP thinks differently.  As part of the Warning Letter, OPDP states, “To date, no drug has been shown to improve ACM in COPD” and includes a footnote that states, “Through the issuance of this letter, FDA does not intend to convey any views on whether data that did show that Breztri improved ACM in COPD would support a change to the FDA-approved labeling for Breztri.”  This seems to suggest that FDA expects this type of outcome claim to be included in labeling.

Enforcement may also have been triggered by the manner of the presentation.  If AstraZeneca wanted to present this data in text form as opposed to colorful graphics, careful to equally present context with a description of the findings, this may not have resulted in a Warning Letter.

But a Warning Letter?  This blogger is curious to see what else OPDP has up its sleeve for 2023.  At a minimum, this letter gives industry a lot to think about as it considers Rx drug promotional materials.