This appeal arose from a series of appeals and remands relating to the FDA’s approval of Philips Roxane, Inc.’s 1981 “new animal drug application” for a generic of bacitracin zinc. The company Alpharma, which manufactures a similar drug, filed citizen petitions asking the FDA to revoke their approval of the drug. The FDA requires applications to demonstrate a drug’s safety and efficacy for their prescribed use. A 1970s safety study sufficiently supported the conclusion that the drug was safe under the Food, Drug, and Cosmetic Act. To establish the drug’s efficacy, Philips Roxane pursued the “regulatory shortcut” of establishing “bioequivalency”: as long as they could establish the generic drug’s efficacy by demonstrating it was the “bioequivalent” to a “benchmark” drug the FDA had already found effective, then Philips Roxane would have satisfied the FDA’s efficacy requirement. Philips Roxane used a 1978 study that concluded both a “benchmark” drug and the generic were equally effective.
Alpharma appealed the FDA’s approval of the Philips Roxane generic up to the DC Circuit Court of Appeals, disputing that the 1978 study established the generic’s bioequivalence and alleging that the FDA had acted arbitrarily. At this first appeal, the DC Circuit deferred to the FDA on the question of bioequivalence, but did not accept the FDA’s response to Alpharma’s criticism of the testing methodology. The court found the FDA’s response conclusory, making “no response to cogently explain why Alpharma was mistaken.” The DC Circuit remanded the case to the FDA for a more adequate explanation.
Judge Garland wrote the majority opinion upon the case’s return to the DC Circuit in 2006. He held that the FDA’s explanation on remand adequately addressed the questions of the first appeal to the DC Circuit on how the 1978 study satisfied the bioequivalency standard in the animal drug context. However, he added that the FDA’s response also raised new problems and apparent contradictions, requiring another remand to the agency before the court could determine whether the FDA had acted reasonably. There were indications that the single-dosage approach of the 1978 study contradicted a historic practice of using the maximum approved dosage in testing, which was higher than a single dosage in this case. Furthermore, the FDA’s explanation made it unclear whether the 1978 study demonstrated the “control” or the “prevention” of the poultry disease for which it was prescribed. This mattered because the highest approved dosages were different for control and prevention. Because the FDA’s explanation also failed to provide its rationale for determining the optimally effective control dose, Judge Garland said the court was unable to determine whether the FDA’s decision was reasonable. Nevertheless, anticipating minimal harm from allowing the approval to remain in effect, the court left the approval in place pending the FDA’s renewed explanation on remand.