KENNEDY, Circuit Judge.

Defendants-appellants Otis R. Bowen, Secretary of Health and Human Services, et al. ("the Government") appeal from a judgment for plaintiff-appellee Norwich Eaton Pharmaceuticals, Inc. ("Norwich") in an action seeking a declaration of rights under the Federal Food, Drug, and Cosmetic Act, 21 U.S.C. § 301-392 ("FFDCA"). The Government argues that the District Court exceeded the appropriate scope of review and that the District Court erroneously found that the Food and Drug Administration's ("FDA") denial of Norwich's citizen petition was arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law. We find that the District Court reviewed the agency action under the appropriate standard, but we reverse its determination that the agency's action was invalid.

I.

Norwich is licensed by Reckitt and Colman, Ltd., an English pharmaceutical company, to distribute buprenorphine hydrochloride in the United States under the trade name Buprenex Injectible ("Buprenex"). Buprenex is a new analgesic drug used for treatment of moderate to severe pain. Because it is derived from the opiate thebaine, buprenorphine hydrochloride is a narcotic drug subject to control by the Drug Enforcement Administration ("DEA") under the Controlled Substances Act, 21 U.S.C. § 801 et seq. ("CSA"). As a derivative of thebaine, buprenorphine hydrochloride originally was automatically classified as a Schedule II substance under the CSA, which meant that it was subject to the most stringent controls applicable to a drug with legitimate medical use. Because of its unique molecular structure, however, buprenorphine hydrochloride does not have the physical dependence properties of the other opiates. This led to its reclassification in 1985 as a Schedule V drug, which is the lowest level of control short of decontrol.

On October 31, 1979, Norwich filed a New Drug Application ("NDA") with the FDA for Buprenex. At that time, buprenorphine hydrochloride had not been approved for marketing in the United States. On October 14, 1981, Dr. Marion Finkel of the FDA sent Norwich a letter stating that the NDA would be approved if Norwich made certain changes in the labeling. Dr. Finkel also stated that the drug was not marketable until the DEA had completed its final rulemaking procedures. At that time, efforts were underway before the DEA to reclassify the drug from Schedule II to Schedule V of the CSA.

On December 29, 1981, Dr. Finkel sent a second letter to Norwich, in which she stated:

We have completed the review of this application and have concluded that the drug is safe and effective for use as recommended in the labeling. As discussed in a telephone conversation between Dr. Alexander Neill and Dr. Marion J. Finkel on December 22, 1981, you have agreed to the conditional Phase IV bioavailability study and the labeling revisions outlined in our October 14, 1981 letter. Accordingly, the application is approved.

As you know, the Division is preparing documents to affect lesser controls for buprenorphine. This, however, in no way impairs your approval to market buprenorphine in its current controlled substances schedule.

Joint Appendix at 59 (emphasis added).

On January 27, 1982, Norwich submitted revised draft labeling to the FDA. On February 3, 1982, John M. Kolbas, the president of Norwich, sent a letter to the FDA acknowledging that the FDA had "approved" Buprenex and noting that the drug was currently subject to Schedule II controls. The letter stated further: "Although the letter approving the Buprenex NDA indicated that we may market the product under Schedule II controls, there are many factors which adversely affect the feasibility of this option." Id. at 383. Kolbas requested a meeting regarding the scheduling issue. On May 24, 1982, the FDA sent a letter to Norwich acknowledging receipt of Norwich's "supplemental application" providing for labeling changes and requesting additional changes in the label. Norwich submitted further labeling changes in September, 1984 and February and March, 1985. On February 28, 1985, the DEA ruled that buprenorphine hydrochloride was a Schedule V Narcotic Controlled Substance. On June 20, 1985, the FDA made its last request for changes in the labeling. On June 24, 1985, the FDA received final printed labeling from Norwich. On June 28, 1985, the labeling was approved. Norwich began marketing Buprenex on June 29, 1985.

Meanwhile, the Drug Price Competition and Patent Term Restoration Act, Pub.L. No. 98-417, 98 Stat. 1585 (amending 21 U.S.C. § 355) ("the Act"), became law on September 24, 1984. Title I of the Act is intended "to make available more low cost generic drugs by establishing a generic drug approval procedure for pioneer drugs first approved after 1962." H.R. Rep. No. 857, Part I, 98th Cong., 2d Sess. 14, reprinted in 1984 U.S. Code Cong. Ad.News 2647, 2647. In addition, Title I provides a limited period of protection from competition to certain previously approved pioneer drugs by granting non-patent marketing exclusivity of ten years to new chemical entities approved on or after January 1, 1982 and on or before September 24, 1984, see 21 U.S.C. § 355(j)(4)(D)(i), (c)(3)(D)(i), and non-patent marketing exclusivity of five years to drugs approved after September 24, 1984, see 21 U.S.C. § 355(j)(4)(D)(ii), (c)(3)(D)(ii). Title II provides an incentive for pioneer drug research by "restor[ing] ... some of the time lost on patent life while the product is awaiting pre-market approval." H.R. Rep. No. 857, Part I, supra, at 15, reprinted in 1984 U.S. Code Cong. Ad.News at 2648.

On August 19, 1985, Norwich filed a petition with the FDA seeking a declaration that Buprenex was entitled to exclusivity in marketing. Norwich argued that at the time of the 1981 "approval" there was no approved label under which the drug could have been marketed; thus, there was no valid approval. Norwich contended that approval actually came when the final printed label was submitted to the FDA in June, 1985. Norwich argued in the alternative that the approval became effective on January 8, 1982, which was the date on which Norwich received the FDA's letter of December 29, 1981. The FDA provided a detailed response denying this petition and stating that Buprenex had been approved on December 29, 1981, before the statutory date allowing exclusivity.

Norwich brought suit in the United States District Court for the Southern District of Ohio. The District Court, 645 F. Supp. 321, held that Buprenex was not approved on December 29, 1981, but rather on June 28, 1985, the date on which the FDA approved an application for labeling that contained a Schedule V designation under the CSA. The Government appeals.

II.

The Government first raises a procedural issue; i.e., that the District Court went beyond the administrative record in reviewing the agency's action. The District Court was required to review the agency's decision that Buprenex was approved in 1981 to determine whether that decision was "arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law," as specified in 5 U.S.C. § 706(2)(A). The scope of review under this section is narrow: "In applying that standard, the focal point for judicial review should be the administrative record already in existence, not some new record made initially in the reviewing court." Camp v. Pitts, 411 U.S. 138, 142, 93 S.Ct. 1241, 1244, 36 L.Ed.2d 106 (1973). The Government asserts that the District Court erroneously conducted a trial de novo when it reviewed the FDA's action.

As this Court recently noted in Upjohn Mfg. Co. v. Schweiker, 681 F.2d 480 (6th Cir. 1982), de novo review of agency action is the exception rather than the rule, unless required by statute. "`[ D] e novo review is appropriate only where there are inadequate factfinding procedures in an adjudicatory proceeding, or where judicial proceedings are brought to enforce certain administrative actions.'" Id. at 483 (quoting Camp v. Pitts, 411 U.S. 138, 142, 93 S.Ct. 1241, 1244, 36 L.Ed.2d 106 (1973)).

The District Court did not conduct a trial de novo here. The District Court did admit evidence not in the administrative record. The Government objected, and the District Court responded that it could not determine whether the administrative record was adequate without seeing the documents that Norwich was offering. As the Ninth Circuit has recognized:

It will often be impossible, especially when highly technical matters are involved, for the court to determine whether the agency took into consideration all relevant factors unless it looks outside the record to determine what matters the agency should have considered but did not.

Asarco, Inc. v. EPA, 616 F.2d 1153, 1160 (9th Cir. 1980). Thus, consideration of evidence outside the administrative record is proper under some circumstances, e.g., "for background information . . . or for the limited purposes of ascertaining whether the agency considered all the relevant factors or fully explicated its course of conduct or grounds of decision." Id. (citations omitted).

Moreover, the District Court stated in its Order that it based its decision upon its "review of the administrative record, the memoranda and the arguments by counsel." Joint Appendix at 263. The Order does not reveal that the District Court used any evidence outside the administrative record in reaching its decision. Thus, we find that the District Court did not conduct a trial de novo and that the Government's contention that the District Court based its decision on information outside the administrative record is without merit.

III.

On the merits, the Government appeals from the District Court's determination that Buprenex was not approved until June 28, 1985. We begin by noting that the proper issue for this Court to address, as it was for the District Court, is whether the FDA's determination on December 6, 1985 that Buprenex had been approved on December 29, 1981 was arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.

The District Court found that the Secretary's decision that the December 29, 1981 letter constituted final approval of the Buprenex NDA was contrary to law because the FDA could not approve a drug before it received final printed labeling. Therefore, the Court concluded that the approval date was June 28, 1985, which is the date that the FDA approved the final printed labeling permitting sale of Buprenex as a Schedule V drug. Adoption of this date would give Norwich a 5-year exclusivity period. See 21 U.S.C. § 355(j)(4)(D)(ii), (c)(3)(D)(ii).

We cannot agree with the District Court's characterization of the law. We see nothing in either the statute or the regulations which would prohibit the FDA from approving a drug based upon proposed labeling as it purported to do in December, 1981.

The District Court began by noting that 21 U.S.C. § 355(b) requires the sponsor of a new drug to submit "specimens of the labeling proposed to be used for such drug" in its NDA before the application will be considered. The Court also quoted 21 C.F.R. § 201.57(h)(1), which requires that the schedule of a drug controlled by the DEA be stated in the labeling. The Court found that the DEA did not issue its final rule on the scheduling of buprenorphine hydrochloride until February 28, 1985, and that Norwich's label submissions before that date were thus insufficient to allow approval. The District Court also relied on 21 C.F.R. § 314.1(c)(4)(e) in finding that the FDA had a stated policy not to approve an NDA until the labeling was in its final form. This regulation, in its 1981 version, stated: "An application will not ordinarily be approved prior to the submission of the final printed label and labeling of the drug." 21 C.F.R. § 314.1(c)(4)(e) (1981). The District Court thus concluded that the FDA's letter of December 29, 1981 merely indicated Buprenex's "approvable" status under 21 C.F.R. § 314.100(d) and that "the FDA's `approval' of the Buprenex NDA was contingent upon and not effective until approval of labeling in its final form." Joint Appendix at 264.

We disagree. The FDA's interpretation that it was authorized in 1981 to approve a drug before final labeling was submitted does not conflict with the statute then in effect. Both the version of section 355(b) in effect in 1981, see 21 U.S.C. § 355(b)(6) (1976), and the amended version, see 21 U.S.C. § 355(b)(1)(F) (Supp. III 1985), require submission of "specimens of the labeling proposed to be used for such drug," but do not specifically require submission of the final labeling. Thus, the statute cannot be read to require submission of final printed labeling before approval. Nor do the regulations mandate this conclusion. The use of the word "ordinarily" in 21 C.F.R. § 314.1(c)(4)(e) (1981) logically implies that on some occasions the FDA will approve an application based on draft labeling. Moreover, the District Court's finding that the FDA's actions violated 21 C.F.R. § 201.57(h)(1) seems to ignore that buprenorphine hydrochloride was scheduled in December, 1981, although not under the classification desired by Norwich. Thus, the FDA approved Buprenex as a Schedule II drug in December, 1981, subject to Norwich's compliance with the agreed-upon changes in the draft label.

Because of its conclusion that approval did not come until June 28, 1985, the District Court did not address Norwich's alternative contention that the approval contained in the letter dated December 29, 1981 became effective on January 8, 1982, which is the date on which Norwich received the letter. If Norwich is correct, Buprenex is entitled to a 10-year exclusivity period. See 21 U.S.C. § 355(j)(4)(D)(i), (c)(3)(D)(i). We find that this interpretation is not mandated by the statute or the regulations.

The FDA regulation in effect at the time the letter was sent stated that "the application shall be approved on the date of the notification." 21 C.F.R. § 314.105 (1981). (The FDA has since amended the regulation to specifically provide that "[t]he date of the agency's approval letter is the date of approval of the application." 21 C.F.R. § 314.105(a) (1986).) The FDA interprets the 1981 regulation as meaning that approval was effective on the date on which it issued the approval letter, not the date on which the drug company received the letter. The FDA's interpretation is reasonable and enables it to fulfill its statutory mandate. Under 21 U.S.C. § 355(c), the FDA is required within 180 days of receiving an application for approval to either approve it or provide the applicant with notice of an opportunity for hearing on the question of whether the application is "approvable." Unless the FDA knows the date of notification, it cannot determine whether it fulfilled its statutory mandate.

"An agency's construction of a statute it is charged with enforcing is entitled to deference if it is reasonable and not in conflict with the expressed intent of Congress." United States v. Riverside Bayview Homes, Inc., 474 U.S. 121, ___, 106 S.Ct. 455, 461, 88 L.Ed.2d 419 (1985) (citations omitted); see also Aluminum Co. of America v. Central Lincoln Util. Dist., 467 U.S. 380, 389, 104 S.Ct. 2472, 2479, 81 L.Ed.2d 301 (1984). "[I]t is not necessary that the agency's construction of the statute be the only permissible one. Rather, its construction `must be upheld unless that view is plainly unreasonable.'" Ohio v. Ruckelshaus, 776 F.2d 1333, 1339 (6th Cir. 1985) (quoting National Steel Corp. v. Gorsuch, 700 F.2d 314, 321 (6th Cir. 1983)), cert. denied, ___ U.S. ___, 106 S.Ct. 2889, 90 L.Ed.2d 977 (1986).

We find that the FDA's determination that it had issued a valid approval for Buprenex in 1981 was a reasonable interpretation under the FFDCA and the regulations promulgated thereunder and was not in conflict with expressed congressional intent. Thus, we see no grounds for holding that the FDA's determination was not in accordance with law.

IV.

The Government further contends that the FDA's determination that Buprenex's approval date was December 29, 1981 did not violate the prohibitions in 5 U.S.C. § 706(2)(A) against arbitrariness or capriciousness. We agree.

In its citizen petition, Norwich asserted that the regulatory history of Buprenex was "unique and potentially confusing in that, although an `approval' letter was sent to Norwich Eaton on December 29, 1981, there did not exist at that time FDA approved labeling under which the product could have been marketed." Joint Appendix at 26. The company asked the FDA to declare that Buprenex was entitled to non-patent marketing exclusivity, asserting that the effective approval date was June 28, 1985, or, in the alternative, January 8, 1982. The FDA addressed each of the arguments raised by Norwich in its petition and concluded that the approval was effective on December 29, 1981.

The FDA began by rejecting Norwich's definition of "date of approval," as contained in the Act, as meaning either the date on which a "meeting of the minds" occurred between the agency and the applicant as to the terms of the approval, or the date on which the applicant received the approval letter. The FDA also rejected Norwich's suggestion that an NDA could not be approved until the drug involved was cleared for marketing under the CSA. Rather, the FDA interpreted "date of approval" to mean "the date of [sic] which the agency exercises its authority under the Federal Food, Drug, and Cosmetic Act to approve a new drug application." Joint Appendix at 98.

The FDA pointed out that the language of its letter of December 29, 1981 clearly indicated that Buprenex was approved for marketing subject to Norwich making the agreed-upon changes in the label, and that FDA approval of the revised labeling was not a prerequisite to marketing. Under the provisions of the Act, the date of approval, not the date of marketing or distribution, determines whether exclusivity is available.

Kolbas' letter of February 3, 1982 to the FDA acknowledged that Buprenex was approved. Although the FDA admitted that it recommended additional changes to the labeling submitted by Norwich on January 27, 1982, it found these recommendations to be "irrelevant" to the date of approval. Norwich could have prepared final printed labeling in accordance with the December 29, 1981 letter and could have marketed Buprenex without further agency action. The 1981 letter requested submission only of final printed labeling and a market package of the drug when available. Moreover, Dr. Finkel, of the FDA, had told Norwich on December 22, 1981 that draft labeling need not be submitted. Instead of preparing final labeling and marketing the drug, however, Norwich choose to submit a supplement containing draft labeling. The FDA assigned these submissions an "S" file number, indicating that it regarded them as supplements, not amendments, to the original NDA. ("Amendments" are used to modify pending applications. See 21 C.F.R. § 314.6 (1981 through 1984 eds.). "Supplements" effect changes in applications that have already been approved. See 21 C.F.R. § 314.8 (1981 through 1984 eds.).) Although the FDA commented on and suggested changes to this draft labeling, these submissions in no way prevented Norwich from going ahead and marketing Buprenex in accordance with the provisions of the 1981 letter.

The FDA asserted that its June 28, 1985 letter merely approved an amendment to the 1982 labeling supplement to the previously approved NDA. It pointed out that the 1985 letter was signed by a division director, who was authorized solely to approve supplemental, not new, drug applications. Thus, if the FDA were to accept Norwich's assertion that the 1981 approval was not effective, the 1985 "approval" still would not be effective since it would not have been approved by someone with proper authority.

The FDA then stated that Norwich was incorrect in thinking that ability to market a drug under the CSA was a condition precedent to approval under the FFDCA. The agency admitted that its letter of October 14, 1981 to Norwich erroneously indicated that final rulemaking under the CSA would be necessary before legal marketing could occur. As the December 29, 1981 letter stated, however, buprenorphine hydrochloride was scheduled at the time of the 1981 approval. Thus, Norwich could have marketed the drug at the time of the 1981 approval as a Schedule II drug. Its decision not to do so was a marketing decision, not a result compelled by law.

The FDA acknowledged that a FDA employee had refused to certify Buprenex as being approved for marketing in the United States in December, 1984, pending the agency's receipt of the address of the manufacturing facility and copies of the final printed labeling. However, the agency asserted that this refusal reflected agency policy not to provide certification without this information. It did not indicate that the FDA did not consider the drug approved.

The FDA also noted that a copy of a May 12, 1982 recommendation of the Department of Health and Human Services to the DEA that buprenorphine hydrochloride be rescheduled was provided to Norwich's counsel. This recommendation clearly stated that the drug was approved, thus providing Norwich's counsel with notice of the agency's interpretation of the 1981 letter.

Finally, the agency admitted that Buprenex was initially included in the fifth edition of the agency's Approved Prescription Drug Products publication, but was omitted in later supplements. The FDA explained that Buprenex was dropped from the list because agency policy at the time was to list only those approved drugs being marketed. The agency stated that since it now appeared that Buprenex would be marketed, the drug would again appear in the publication.

In reviewing for a violation of 5 U.S.C. § 706(2)(A), "the court must consider whether the decision was based on a consideration of the relevant factors and whether there has been a clear error of judgment." Citizens to Preserve Overton Park v. Volpe, 401 U.S. 402, 416, 91 S.Ct. 814, 823, 28 L.Ed.2d 136 (1971) (citations omitted). The reviewing "court may not concern itself with the wisdom of agency action." Federal Property Mgt. Corp. v. Harris, 603 F.2d 1226, 1230 (6th Cir. 1979) (citing United States v. Allegheny-Ludlum Steel Corp., 406 U.S. 742, 749, 92 S.Ct. 1941, 1946, 32 L.Ed.2d 453 (1972)). "The court is not empowered to substitute its judgment for that of the agency." Overton Park, 401 U.S. at 416, 91 S.Ct. at 824. Rather, "[i]f the reasoning behind the agency's action is logical, . . . that action must be allowed to stand. That is true even if an alternative course of action would also be logical . . . ." Harris, 603 F.2d at 1231.

The FDA's denial of Norwich's citizen petition fully considered the arguments raised in the petition and rested upon a rational basis supported by the administrative record. Therefore, we find that the agency's determination in 1985 that Buprenex was approved in 1981 does not violate 5 U.S.C. § 706(2)(A).

V.

Accordingly, we REVERSE the District Court. The case is REMANDED to the District Court with directions to enter judgment for the Government.