We're starting to go in circles. Let me make sure I get your description of SNY's argument: because the FDA got additional data it should not have, it approved a drug it would not otherwise have approved. OK, not the intent of FDCA, but OK.
However, that subsection allows the FDA broad latitude to determine bioequivalence, as mPower points out in his citation of it. At the USP's 3rd Workshop on the Characterization of Heparin Products, it was proposed that a LMWH's immunogenic profile can be used to determine bioequivalence.
As such, the FDA can reasonably hold that their request was not for safety and efficacy data per se, but for bioequivalence data. Of course, the FDA requested the data way before Dr. Walenga presented her data . . . I can't find much reference to the concept prior to a couple of years ago, but that doesn't mean the concept was unknown to the FDA; the different immunogenicity profiles of LMWH and their generics (obviously these would be foreign ones approved by looser regulatory agencies) was well known. Certainly it was central in the contaminated heparin issue of '08, not TOO much after the FDA made the request . . . unless there is a smoking gun email or letter that says it was for safety & effectiveness. We only saw the PR, which made no mention of safety or effectiveness; the letter received by Momenta/Sandoz is not a public document at this juncture.
As for the argument that this data somehow effectively allowed the generic applicant to rely on SNY's pioneering confidential info, the FDA's response to the CP addresses that. I just don't see it unless the "chinese wall" broke down. And neither of us think that happened.
Obviously, I don't think SNY's argument will fly, either, but I believe it more strongly than you do because of the above. I don't think this is ample ammunition for a judge to rule their way.