Sanofi Seeks Summary Judgment in Generic LOVENOX Dispute; Administrative Record Highlights Disagreements at FDA Over Generic Approval

December 14, 2010

By Kurt R. Karst – 

When we last left you in the battle over FDA’s approval of a generic version of sanofi-aventis U.S. L.L.C.’s (“sanofi’s”) blockbuster anti-coagulant drug LOVENOX (enoxaparin sodium injection), Judge Emmet G. Sullivan of the U.S. District Court for the District of Columbia denied sanofi’s Motion for Preliminary Injunction.  Sanofi did not appeal the decision, but the case has moved forward for a decision on the merits.  In a Motion for Summary Judgment sanofi filed earlier this week, the company continues to allege that that FDA’s approval of generic LOVENOX is unlawful and should be set aside.  Sanofi’s brief also highlights some interesting disagreements that allegedly took placed within FDA on generic LOVENOX.

Sanofi sued FDA on July 26th requesting that the court issue a declaratory judgment that FDA acted unlawfully in approving Sandoz Inc.’s ANDA No. 77-857, as well as a temporary restraining order and preliminary injunction directing FDA to immediately suspend and withdraw approval of the ANDA, and a permanent injunction under the same terms.  (Sanofi later agreed to consolidate its temporary restraining order request with its preliminary injunction request.)  In addition to the approval of ANDA No. 77-857, Sanofi also challenged FDA’s July 23rd response to a February 2003 citizen petition, in which the Agency outlined five criteria (i.e., standards for identity) that an ANDA applicant needs to demonstrate sameness of its active ingredient as compared to LOVENOX.  Those five criteria are: (1) equivalence of physicochemical properties; (2) equivalence of heparin source material and mode of depolymerization; (3) equivalence in disaccharide building blocks, fragment mapping, and sequence of oligosaccharide species; (4) equivalence in biological and biochemical assays; and (5) equivalence of in vivo pharacodynamic profile. 

In sanofi’s initial court filings the company set forth three merits arguments as to why a preliminary injunction is necessary: (1) FDA exceeded its authority under the FDC Act (specifically FDC Act § 505(j)(2)(A)) by requiring Sandoz to submit studies beyond what is permitted for ANDAs (i.e., immunogenicity studies that, according to Sanofi, are studies intended “to demonstrate safety and effectiveness,” rather than, as FDA argued, chemistry, manufacturing, and control information); (2) FDA departed from Agency precedent by approving ANDA No. 77-857 when the product has not yet been fully characterized; and (3) FDA approved ANDA No. 77-857 without sufficient evidence that the drug product has the “same” active ingredient as LOVENOX (as required by FDC Act § 505(j)(2)(A)). 

Judge Sullivan noted in his denial of sanofi’s Motion for Preliminary Injunction that the court’s decision was not a final decision on the merits and that “Sanofi may be able to establish that there are grounds for overturning the grant of Sandoz’s ANDA.”  Sanofi’s Motion for Summary Judgment tries to do just that, building on the company’s previous merits arguments.  According to sanofi’s latest motion:

First, the plain language of the FDCA precludes FDA from requiring immunogenicity testing, as it did here, as part of the ANDA approval process. Section 505(j)(2)(A) of the Act sets forth eight categories of information FDA is permitted to require of an ANDA applicant and expressly forbids FDA from mandating submission of additional information.  Immunogenicity testing is not encompassed by any of the categories.  FDA’s argument to the contrary – that a provision of the Act authorizing it to require “a full description of the methods used in, and the facilities and controls used for, the manufacture, processing, and packing of [a generic] drug,” encompasses immunogenicity testing – does not withstand scrutiny.  FDA’s interpretation of the FDCA would upset the three-tiered framework Congress crafted for approval of new drugs and render Section 505(b)(2) of the Act wholly superfluous.

Second, FDA’s decision must be set aside as arbitrary and capricious under the [Administrative Procedure Act] because of the Agency’s utter failure to acknowledge and provide a persuasive justification for its departure from established precedent. . . .  Prior to its unlawful approval of Sandoz’s ANDA, FDA had consistently recognized that where a drug is not fully characterized, it is impossible to conclude that it is the same as a previously approved drug. . . .

Third, and finally, FDA’s approval decision is flatly inconsistent with the FDCA’s mandate that an ANDA may not be approved unless its active ingredient is the “same as” that of the RLD.  In contravention of this statutory directive, FDA approved Sandoz’s ANDA based on application of a five-factor test that relies on unsubstantiated conjecture lacking any basis in either science or the text of the FDCA.  Moreover, the evidence before the Agency makes clear that Sandoz’s drug did not even satisfy the Agency’s own ad hoc standard for determining active ingredient sameness. [(Internal citations omitted)]

Sanofi’s Motion for Summary Judgment also brings to light some interesting tensions within FDA concerning the five criteria that an ANDA applicant needs to demonstrate sameness of its active ingredient as compared to LOVENOX.  According to sanofi, information in the administrative record shows that scientists within FDA’s Office of New Drug Quality Assessment (“ONDQA”) “expressed disagreement” with the Office of Generic Drugs’ (“OGD’s”) conclusions about the five criteria for assessing active ingredient sameness in generic LOVENOX.  According to sanofi:

ONDQA’s scientists, including FDA’s own expert on heparin and heparin-based products, Dr. Ali Al Hakim, explained that a showing of active ingredient sameness for enoxaparin is impossible without complete characterization of all of enoxaparin’s oligosaccharide chains.  ONDQA argued that, because enoxaparin consists of a mixture of oligosaccharides, it cannot be determined which components of the mixture contributed to enoxaparin’s activity.  ONDQA further explained that OGD’s five criteria are inadequate to demonstrate sameness and criticized OGD for relying on “inference” to do so.   ONDQA asserted that OGD’s proposed methodology was contrary to both the law and FDA policies.  [(Internal citations omitted)]

This disagreement was resolved in a July 20, 2010, intra-agency memorandum, according to sanofi, in which Dr. Keith Webber, Deputy Director of the Office of Pharmaceutical Science (and acting OGD Director) determined that the five criteria are a “valid approach . . . for purposes of ANDA approval.”  According to sanofi, Dr. Webber “[rejected] ONDQA’s argument that OGD’s approach was inconsistent with FDA policy involving drugs that had not been fully characterized [and] asserted that ‘[d]epending on the kind of drug at issue, there may be different ways to show active ingredient sameness’ and that OGD’s approach was acceptable as applied to enoxaparin.”

FDA’s five criteria for assessing active ingredient sameness with respect to generic LOVENOX were also raised in a December 10th citizen petition submitted on behalf of Teva Neuroscience, Inc. (“Teva”) concerning FDA’s ability to approve generic versions of the company’s COPAXONE (glatiramer acetate injection), a drug product approved for the reduction of frequency of relapses in relapsing-remitting multiple sclerosis.  In May 2010, FDA indicated in a response to a different Teva citizen petition concerning generic COPAXONE that the Agency might be considering applying similar criteria to that drug.  Specifically, FDA stated that “given the complexity of Copaxone, we may require that any ANDA sponsor demonstrate active ingredient sameness through a multi-criteria test or series of tests, each criterion of which captures different aspects of the active ingredient’s ‘sameness,’ and which together would provide overlapping evidence by which an ANDA applicant could demonstrate active ingredient sameness within the meaning of the Act and FDA regulations.” 

Teva argues in its latest petition, among other things, that “none of the five criteria that provided reasonable assurance of sameness between generic enoxaparin and Lovenox® are viable in the Copaxone® context” (emphasis in original), and that generic COPAXONE should not be approved unless and until: (1) the glatiramer acetate in COPAXONE has been fully characterized and a generic applicant proves that its proposed drug product contains exactly the same polypeptide sequences (in the same amounts and with the same structures) as COPAXONE; or (2) all polypeptide sequences that contribute to the therapeutic effects of COPAXONE’s glatiramer acetate have been identified and a generic applicant proves that its proposed drug product contains exactly the same clinically relevant polypeptide sequences (in the same amounts and with the same structures) as COPAXONE (and that any differences in the non-clinically active polypeptides in the proposed generic drug do not “undermine the clinically active polypeptides’ safety, efficacy, toxicology, and immunology profiles”).