For the past few years, safety questions have hovered over Advair, a long-acting beta agonist that includes a steroid and a $6 billion seller. An FDA alert was issued in 2005; last November, an FDA panel recommended more warnings for kids both Advair and Glaxo’s Serevent, which doesn’t include a steroid; and the agency asked drugmakers for more data as a prelude to an advisory committee meeting later this year.

In an attempt to get out in front of the controversy, Glaxo sponsored a review in the Annals of Internal Medicine shows that Advair decreases the risk of severe side effects, doesn’t appear to alter the risk for hospitalization, and may not change the risk for asthma-related deaths when compared with the use of steroids alone. In reaching this conclusion, the researchers conducted a meta-analysis of 66 trials involving a total of 20,966 patients.

However, in an accompanying editorial by Kevin Weiss says the review is well done, but the source of the data - randomized efficacy trial - limits its ability to answer a key question: Does the use of long-acting beta-agonists combined with inhaled steroids enhance effeciveness and reduce risks?

Weiss - a Northwestern University professor, ceo of the non-profit American Board of Medical Specialties, and a Merck asthma consultant - says the researchers tripped up by mixing Glaxo’s SMART study, a large, prospective, real-world clinical trial, with several smaller trials for closely supervised patients. He also writes that major adverse event rates were higher in SMART, suggesting samples were different; different attention levels were given to measuring adversing events, or both. And the meta-analysis didn’t address higher mortality rates in African-Americans.

And so, Weiss concludes, the study may have “ultimately helped answer the question about safety when long-acting beta-agonists are used with inhaled steroids in an idealized clinical environment, but they did not resolve the controversy about the safety of long-acting beta-agonists (with or without steroids) in an environment that more closely reflects actual clinical practice.”

By the way, GlaxomithKline supported data collection, analyses, manuscript preparation, and data interpretation, while the authors say approved the manuscript for submission. Four of seven authors are Glaxo employees. The lead author, Eric Bateman, is a Glaxo consultant, has done speaking for Glaxo, received Glaxo grants and provided expert testimony for Glaxo in trials. Here is the dislcosure info.