On February 7, United States District Judge Amy Berman Jackson granted defendant FDA's motion for summary judgment and denied plaintiff Sanofi-Aventis's ("Sanofi's") cross-motions in Sanofi's challenge of FDA's decision of a citizen petition filed by Sanofi regarding generic versions of Lovenox® (enoxaparin sodium injection), a blood thinning drug. As we reported here in our coverage of FDA's response to Sanofi's citizen petition, Sanofi had requested that FDA withhold generic approvals until enoxaparin had been fully characterized, for any abbreviated new drug application ("ANDA") referencing Lovenox® that did not use an equivalent manufacturing process or did not show proof of equivalent safety and efficacy through clinical trials.
In that decision, FDA had concluded that enoxaparin had been adequately characterized to approve generic applications derived from natural sources and that five criteria would be sufficient to demonstrate "sameness" of enoxaparin in generic products compared to Lovenox®. In FDA's opinion, these elements would demonstrate sameness of enoxaparin, including its 1,6-anhydro ring structure, without (a) the need for a complete characterization of all of the different polysaccharides of exoxaparin, (b) using the same manufacturing processes by Sanofi, or (c) clinical trial data to demonstrate safety and efficacy. FDA's administrative record showed later, however, that while the Office of Generic Drugs ("OGD") supported this five-part test, the Office of New Drug Quality Assessment ("ONDQA") thought that the test was insufficient and the only way to demonstrate enoxaparin sameness was to fully characterize all of the different polysaccharides of enoxaparin.
Following FDA's decision on the citizen petition, FDA approved Novartis's/Sandoz's version and Sanofi filed a complaint that included a motion for a temporary retraining order ("TRO") and preliminary injunction ("PI") to prevent the sale of Sandoz's version and have FDA withdraw approval of Sandoz's version. Since then, the District Court consolidated and denied Sanofi's TRO/PI motion, based on the Court's finding that Sanofi was unlikely to success in its three claims. When these same claims were reviewed by the Court as part of FDA's motion for summary judgment, the Court concluded that all of Sanofi's contentions could be answered in terms of statutory construction:
- the FDA acted within its statutory authority when it called for Sandoz to file immunogenicity data as part of its ANDA;
- it did not unlawfully depart from agency precedent by approving a generic before the listed drug had been fully characterized; and
- it reasonably found that the active in the generic drug was the same as the active ingredient in Lovenox.
For the first conclusion, the Court granted Chevron II deference to FDA, with principal cited authority to Serono Labs., Inc. v. Shalala, 158 F.3d 1313 (D.C. Cir. 1998), finding that neither the Food, Drug, and Cosmetic Act ("FD&C Act") nor FDA's regulations or guidances specifically prevented FDA from requesting immunogenicity data as part of an ANDA. Instead, the Court disagreed with Sanofi that FDA was bound to consider only the enumerated ANDA requirements and agreed with FDA that Congress left the ANDA requirements ambiguous to permit FDA "to utilize its expertise to determine what information it needs to make the assessment it is required to make. . .," including incorporating elements from requirements for new drug approval applications ("NDAs") as necessary. The Court found Serono supported the Court's conclusion, because there the Federal Circuit permitted FDA to use animal studies to evaluate approvability of an ANDA for a generic version of Pergonal® (menotropins, an equal mixture of the naturally occurring follicle-stimulating hormone and luteinizing hormone, which was also regulated as a drug though naturally -occurring and also involving a complex manufacturing process). Further, contrary to Sanofi's assertion, FDA's request for immunogenicity data did not render the product an NDA product approved under Section 505(b)(2) of the FD&C Act, because the proposed product was a generic product, not a modification of the listed drug.
For the second conclusion, the Court referred to its TRO/PI analyses, finding that FDA did not depart, as Sanofi suggested from president for three other products that have not been approved as ANDAs because they were not fully characterized: Premarin® (conjugated estrogens derived from pregnant mares' urine). Hyaluronidase (a recombinant human enzyme to help with the absorption of other drugs), and Omnitrope® (somatropin, a type of recombinant growth hormone). For this, the Court noted that its review is "highly deferential" to FDA because the matter related to "complex scientific information within the agency's technical expertise."
For the third conclusion, the Court found that FDA's own records and internal dispute between OGD and ONDQA that was resolved in a well-reasoned opinion did not demonstrate an arbitrary or capricious decision, an abuse of discretion, or was otherwise not in accordance with the law. For this, once again the Court noted that FDA was entitled to a heightened level of deference given the highly technical nature of the issue. Furthermore, the Court found FDA's five-pronged approach for sameness reasonable and that the immunogenicity data was not an additional element for "sameness," but spoke more to potential impurities.
In view of this decision and our previous experience evaluating decisions where FDA's bioequivalence determinations have been challenged in court, the Court's outcome does not appear surprising. Few petitioners have successfully challenged FDA's bioequivalence determinations in Court, even when FDA ultimately reversed its own bioequivalence recommendations. This is what we have observed, in part because Congress drafted these provisions broadly to permit FDA to have a high level of discretion as long as its decisions were not plainly erroneous (e.g., using an incorrect conversion factor) or inconsistent with its own regulations. To the extent petitioners have succeeded in a remand for explanation, FDA typically has worked out a solution with the petitioner or new drug applicant outside of court rather than creating a negative precedent.
While Lovenox® was approved as an NDA rather than as a "biological product," the District Court's decision suggests that FDA's bioequivalence determinations for biosimilars, once articulated, most likely on a case-by-case basis, will be difficult to challenge, unless FDA does not adequately explain its reasons or makes erroneous conclusions. Furthermore, the Court's decision suggests that FDA will continue to be granted extensive leeway deciding when a product is still generic or requires a 505(b)(2) NDA based on additional requested studies.
