Schering Corporation brings this appeal from the district court’s judgment sustaining the Food and Drug Administration’s view, expressed in 1990 in response to a citizen petition, that the agency is not limited to determining “bioequivalence” by applying the test set forth in 21 U.S.C. § 355(j)(7)(B). 782 F.Supp. 645. In light of the FDA’s promulgation in 1992 of regulations defining “bioequivalence,” we hold that the appeal is moot.
Schering manufactures and distributes pharmaceutical products. Two of Schering’s drugs, Lotrimin® and Proventil®, fall within the class of “non-systemically effective drugs,” or NSEDs. The effectiveness of NSEDs comes from their topical application rather than their absorption into the bloodstream. This class of drugs includes skin ointments and inhaled respiratory medicines.
As in most markets, successful pharmaceutical products invite imitation. Competitors may gain the FDA’s approval for generic copies of certain “listed” drugs under an abbreviated application process. To use the abbreviated process, the applicant must demonstrate that its product meets several criteria. See generally 21 U.S.C. § 355(j). For generic drugs with the same active ingredient, one test is whether the generic product is “bioequivalent” to the drug it copies. See 21 U.S.C. § 355(j)(3)(F). In regard to bioequivalence, the statute states:
A drug shall be considered to be bioequivalent to a listed drug if ... the rate and extent of absorption of the drug do not show a significant difference from the rate and extent of absorption of the listed drug____
21 U.S.C. § 355(j)(7)(B).
On December 4, 1989, Schering filed a citizen petition with the FDA. See generally 21 C.F.R. §§ 10.25(a), 10.30. Schering initially asserted that parties using the abbreviated application process could never establish for NSEDs a similar “rate and extent of absorption,” and thus could never meet the § 355(j)(7)(B) standard for bioequivalence. In its response to the petition, the FDA concluded § 355(j)(7)(B) created “one mandatory definition of bioequivalence, but did not purport to exclude other definitions of bioequivalence____ Congress has not removed FDA’s discretion to adopt other scientifically supportable definitions of bioequivalence that may be applicable to nonsystemieally absorbed drugs.” Letter from Carl C. Peck to Alan R. Bennett of 6/29/90, at 17 (“1990 Letter”). The FDA did not offer any additional definitions of bioequivalence in its response.
Under FDA regulations, the 1990 Letter was final agency action from which the affected parties could seek relief. See 21 C.F.R. § 10.45(d); Abbott Laboratories v. Young, 920 F.2d 984 (D.C.Cir.1990), cert. denied, — U.S. -, 112 S.Ct. 76, 116 L.Ed.2d 49 (1991). Schering filed an action in the district court challenging the FDA’s response. While the matter was pending, the FDA on November 27, 1991, approved the abbreviated application of Copley Pharmaceutical Inc. to sell a generic version of Proventil®. Schering moved for a preliminary injunction suspending the approval of the Copley application. The motion also sought to prohibit the FDA from disclosing any data from Schering’s own NSED applications that addressed the safety and effectiveness of the products. The district court denied Schering’s motion for a preliminary injunction and granted the FDA’s motion for summary judgment on the issue of statutory interpretation addressed in the 1990 Letter, finding the FDA’s position reasonable on essentially the same grounds the agency had identified.
Shortly after Schering filed notice of its appeal, the FDA issued new regulations covering abbreviated drug applications. See 57 Fed.Reg. 17,949-18,001 (1992). In a section entitled “Definitions,” the FDA provided its interpretation of the term “bioequivalence”:
Bioequivalence means the absence of a significant difference in the rate and extent to which the active ingredient ... becomes available at the site of drug action____
57 Fed.Reg. 17,997 (1992). The FDA’s discussion accompanying the 1992 regulations indicated that, at least in theory, a product with a different rate of absorption than the listed product might be bioequivalent so long as four criteria similar to those described in § 355(j)(7)(B)(ii) — unnecessary to describe *37here — were met. 57 Fed.Reg. 17,974 (1992). The FDA could give no examples of such products, however, because it was “unaware of any category of products that can differ in rate of absorption and still be considered bioequivalent.” Id.
Relying on the 1990 Letter, Schering protests that the FDA is now “free to approve generic copies of pioneer drug products using whatever different standard it might deem suitable in a given case.” Brief at 2,10, 32. Whether Schering has correctly characterized the 1990 Letter is irrelevant. The company’s argument, offered as a reason for restricting the agency to the standard set forth in § 355(j)(7)(B), entirely ignores the effect of the 1992 regulations. One of the main differences between those regulations, which Schering has not challenged, and the FDA’s 1990 Letter is that the regulations give the FDA’s definition of bioequivalence while the Letter does not. This marks the difference between this case and one in which the substance of a final rule is “essentially the same” as that of interim regulations. Union of Concerned Scientists v. NRC, 711 F.2d 370, 376 (D.C.Cir.1983); see American Maritime Ass’n v. United States, 766 F.2d 545, 554 n. 14 (D.C.Cir.1985). The regulations, not the FDA’s Letter to Schering, now govern, and will continue to govern, the abbreviated application process. See Fort Stewart Schools v. FLRA, 495 U.S. 641, 654, 110 S.Ct. 2043, 2051, 109 L.Ed.2d 659 (1990). No matter what an agency said in the past, or what it did not say, after an agency issues regulations it must abide by them. Having adopted a definition of “bioequivalence” in its regulations, the FDA is not free to adopt — in Schering’s words — whatever standard “it might deem suitable in a given case.”
An action is moot when nothing turns on its outcome. Since the regulations rather than the Letter now control, Schering must show that the Letter has some continuing consequences to Schering sufficient to keep the case alive. The only possibility relates to the FDA’s approval of the Copley abbreviated application for its competing drug. In moving for a preliminary injunction, Schering assumed that the FDA had followed the practice purportedly reflected in the Letter of using whatever substantive standard other than § 355(j)(7)(B)’s “rate and extent of absorption” test it saw fit. See Memorandum of Points and Authorities in Support of Plaintiff Schering Corporation’s Motion for Preliminary Injunction at 5-7 (Dec. 16, 1991). The materials before us do not support the assumption. An affidavit from an FDA official familiar with the Copley application states that the product satisfied the § 355(j)(7)(B) standard for bioequivalence. Declaration of Roger L. Williams, M.D. at 5 (Jan. 3, 1992). Schering has provided nothing to the contrary. See Cuomo v. NRC, 772 F.2d 972, 974 (D.C.Cir.1985); Washington Metro. Area Transit Comm’n v. Holiday Tours, Inc., 559 F.2d 841, 844 (D.C.Cir.1977). Thus, as matters now stand, all indications are that the FDA did not act on the suggestion in its 1990 Letter that it could adopt informally a standard of bioequivalence other than the one set forth in the statute. See County of Los Angeles v. Davis, 440 U.S. 625, 631, 633-34, 99 S.Ct. 1379, 1383, 1384, 59 L.Ed.2d 642 (1979); Freeport-McMoRan Oil & Gas Co. v. FERC, 962 F.2d 45, 46 (D.C.Cir.1992). Compare National Indep. Coal Operators’ Ass’n v. Kleppe, 423 U.S. 388, 393 n. 4, 96 S.Ct. 809, 812 n. 4, 46 L.Ed.2d 580 (1976).
Schering and the FDA say that the 1992 regulations are consistent with past practice. Perhaps so, in the sense that the regulations do not purport to contradict how the agency had been acting. Still, a comparison of the discussion of bioequivalence in the regulations and accompanying explanation and in the 1990 Letter reveals numerous differences, not the least of which is that the regulations contain an administrative definition of the disputed term while the Letter does not. In any event, similarities between past practice and current regulations cannot save this case from’mootness. Judicial interpretation of § 355(j)(7)(B) must occur in light of the current administrative interpretation of the provision, an interpretation now embodied in the 1992 regulations. Those regulations are not before us. So far as we can tell, nothing in the way of findings in the 1990 Letter served as a foundation for the 1992 regulations. Compare American Maritime Ass’n, 766 F.2d at 554 n. 14; Union of Concerned Scientists, 711 F.2d at 379. We cannot reach back to the 1990 Letter and issue a judgment on the meaning of the statute. As we have said, the Letter has no *38current operative effect. It cannot govern future agency actions. And even to the extent it reflected past agency practice, Schering has been unable to demonstrate any continuing consequences from the position the FDA then expressed. The case has thus “ ‘lost its character as a present, live controversy of the kind that must exist if we are to avoid advisory opinions on abstract [questions] of law.’ ” Princeton Univ. v. Schmid, 455 U.S. 100, 103, 102 S.Ct. 867, 869, 70 L.Ed.2d 855 (1982) (per curiam), quoting Hall v. Beals, 396 U.S. 45, 48, 90 S.Ct. 200, 201-02, 24 L.Ed.2d 214 (1969) (per curiam).
The judgment of the district court is vacated and the cause is remanded to the district court with instructions to dismiss the complaint as moot.
So ordered,.