We affirm the decision of the Commissioner of the Food and Drug Administration withdrawing approval of petitioner’s new drug application.

FACTS

Brandenfels Scalp and Hair Applications and Massage is an aqueous solution of 0.25% sulfanilimide and an aqueous solution of 1.5% lanolin. The labeling reviewed by the Food and Drug Administration’s (FDA) scientific panels claimed that the treatment tends to soften the scalp, to remove dandruff scales, and to aid the scalp and hair generally. The treatment consists of topical applications of the solutions accompanied by massage along the arteries leading to the scalp. It has been marketed since the 1940’s and is sold principally by mail. As per the requirements of the 1938 Food, Drug, and Cosmetic Act, 21 U.S.C. § 301 et seq., the treatment was found to be safe by FDA in 1948.

Brandenfels was the subject of a Post Office investigation in 1946 for alleged misrepresentation of his treatment’s effectiveness. The proceeding was dismissed without prejudice in 1958.

In compliance with the 1962 amendments to the Act, petitioner in 1964 filed with the Bureau of Drugs four volumes of evidence in support of the efficacy of his treatment. Such evidence was not reviewed by the FDA’s scientific panels, apparently because of an oversight. In 1970, the Commissioner classified Brandenfels’ treatment as “possibly effective,” which meant Brandenfels had 6 months to submit “substantial evidence” of effectiveness. At the end of that time, the FDA would evaluate the data to determine whether to withdraw approval of the New Drug Application (NDA) under § 505(e) of the Act, 21 U.S.C. § 355 (e). Brandenfels requested and received an extension, and in the process of granting that extension the FDA notified Brandenfels that it would also review the data submitted by Brandenfels in 1964. In early 1972, the FDA reclassified the treatment as one “lacking substantial evidence of effectiveness.” The stated reason for the reclassification was the failure of Brandenfels to submit any evidence satisfying FDA definitions of what constituted “substantial evidence.” Brandenfels was given 30 days to notify the FDA as to whether he would seek a hearing, and, if he did want a hearing, he was required to provide reasons and analyses of the data supporting his position. Brandenfels filed his hearing request later that year.

Nearly nine years later, the Acting Director of the FDA concluded that “there was no genuine and substantial issue of fact requiring a hearing, and that the legal objections are insubstantial.” The agency’s summary judgment procedures under 21 C.F.R. § 314.200 (g) were invoked to deny Brandenfels’ request for an evidentiary hearing. The decision was based on a determination that none of the four studies submitted by Brandenfels to show efficacy were “adequate and well-controlled” within the meaning of 21 C.F.R. § 314.111 (a)(5)(h). Furthermore, the Acting Director found that statements submitted by two experts did not represent the consensus of current medical opinion and were, in any event, in the nature of unscientific testimonials not constituting substantial evidence of effectiveness as contemplated by Congress and required by the Act. Also rejected were citations to medical literature, found not to be related to the product, and testimonial letters, photographs, and the results of a questionnaire given to users, all found to lack sufficient detail to permit scientific evaluation.

Brandenfels subsequently filed a petition for reconsideration asking that the withdrawal be stayed pending reconsideration or, in the alternative, that it be vacated. The petition was denied in its entirety. The Commissioner reiterated that the studies were not adequate and well-controlled and rejected Brandenfels’ procedural arguments. This petition followed.

I. Petitioner’s Evidence of Efficacy

With respect to Brandenfels’ evidence of efficacy, we agree with the FDA that the Commissioner’s findings accurately reflected Brandenfels’ studies and data, and that the deficiencies found by the Commissioner “conclusively render the studpes] inadequate ... in light of the pertinent regulations,” as required by Weinberger v. Hynson, Westcott & Dunning, Inc., 412 U.S. 609, 622 , 93 S.Ct. 2469, 2479 , 37 L.Ed.2d 207 (1973). A study submitted by Brandenfels from the University of Oregon Medical School, for example, did not conform to any of the requirements of 21 C.F.R. § 314 .-ll(a)(5)(ii), in that it failed to explain the method of selection of subjects observed who used the treatment, or the methods of observing and recording results; it did not provide a system for quantifying symptom severity; and did not adequately establish a “control group” for the study. The other studies similarly suffered from a variety of infirmities, any one of which would be sufficient to uphold the Commissioner’s decision.

We reject petitioner’s request that we ignore FDA regulations defining “substantial evidence” of efficacy in favor of an ad hoc judicial inquiry into what constitutes such evidence. Determining reliable scientific data is not the judicial function. Congress vested that responsibility in the FDA and we will not preempt its presumed expertise. Accord, Cooper Laboratories, Inc. v. Commissioner, FDA, 501 F.2d 772, 778 (D.C.Cir.1974); Upjohn Co. v. Finch, 422 F.2d 944, 954 (6th Cir.1970).

II. Petitioner’s Allegations of Procedural Error

A. Notice

Petitioner contends that the notice of opportunity for hearing charged him with failure to furnish any evidence of efficacy, but that the withdrawal order found that his evidence was insubstantial. He contends, therefore, that the notice of hearing did not give him notice of the actual grounds for withdrawal.

Petitioner was given adequate notice of the basis for withdrawal of his new drug application. The FDA is authorized to satisfy its own notice requirements by providing holders of new drug applications with either general or specific notice of opportunity for hearing. A specific notice is acceptable under the regulations if it refers to specific requirements in the statute and regulations with which there is a lack of compliance, or if it provides a detailed description and analysis of the specific facts resulting in the notice. 21 C.F.R. § 314.200 (a)(1). The notice informing petitioner of the failure of his evidence to comply with the regulations was phrased in the specific statutory language of 21 U.S.C. § 355 (d)(5) and regulatory language of 21 C.F.R. § 314.111 (a)(5)(i), and satisfied the notice requirements.

B. Advance Copy of Order

Brandenfels also contends that the agency failed to follow 21 C.F.R. § 314 .- 200(g)(3) by not providing him with a copy of the proposed withdrawal order for comment prior to publishing the order.

Relying on 21 C.F.R. § 314.200 (g)(1), the FDA contended that it had no obligation to provide petitioner with an advance copy of the summary judgment order. That section, however, refers not to any advance notice requirement but to the information required to be contained in a specific notice of opportunity for hearing. The clear intent of 21 C.F.R. § 314.200 (g)(3) is that holders of new drug applications should receive advance notice whenever they have submitted data and/or information reasonably similar to that contemplated by the statute and regulations.

We find it unnecessary to decide whether the FDA was obliged to provide petitioner with an advance copy of its summary judgment order because petitioner showed no prejudice from the lack of an advance copy. Petitioner had ample opportunity to present evidence, if he had any, to the FDA and later to this court demonstrating that the agency had not conclusively shown that his submitted data was “inadequate or uncontrolled in light of the pertinent regulations.” Hynson, 412 U.S. at 622 , 93 S.Ct. at 2479 . He produced no such evidence and if there was any procedural error it was harmless.

C. National Academy of Science/National Research Council (NAS/NRC) Panel Review

The FDA’s failure to provide the reports and other materials submitted by petitioner in support of the efficacy of his treatment to the NAS/NRC review panels was not prejudicial. Those panels have only the function of advising and assisting the Commissioner. Hynson, 412 U.S. at 614 , 93 S.Ct. at 2475 . They do not constitute another level of review of the efficacy of the particular drug under scrutiny. In fact, all the submitted studies and data were actually considered by the Bureau of Drugs and the Commissioner before the decision was reached.

D. Separation of Functions and Ex Parte Communications

Contrary to the allegations of petitioner, the agency did not violate its restrictions on separation of functions or ex parte communications to the Commissioner. First, petitioner argues that the employees of the Bureau of Drugs who endorsed the proposed order improperly participated in the Commissioner’s decision. As the FDA notes, however, § 314.200(f) requires the Bureau to propose final dispositions to the Commissioner, and, that aside, petitioner has not alleged any improprieties.

Second, petitioner argues that Edward Basile violated the separation of functions doctrine by serving as a legal advisor to the Commissioner while he was working under the FDA Acting Chief Counsel, Jeffrey Springer, who was the FDA’s prosecutor in this case. We agree with the Commissioner, however, in finding no impropriety here: because Basile did not represent the Bureau of Drugs in this proceeding, he was properly assigned to advise the Commissioner.

Third, petitioner alleges that a medical report by Dr. Huene was furnished ex parte to the Commissioner and therefore was not properly before him. The FDA notes that the report was sent to petitioner in 1972, and that the 1977 version of the same report was provided to petitioner under the Freedom of Information Act and was placed in the administrative file where it was available to him.

Finally, petitioner claims that the FDA’s proposed rule on over-the-counter hair restorers constituted an ex parte communication because the Acting Commissioner approved the proposed rule prior to the withdrawal order at issue here. The theory seems to be that evidence considered by the Acting Commissioner in presiding over the rule-making procedure “affected his thinking” in denying Brandenfels’ request for a hearing. But petitioner points to no specific incidents of prejudice, and, furthermore, petitioner’s products were specifically excluded from the proposed rule-making. The proposed rule in and of itself did not constitute an ex parte communication.

E. Delay

Petitioner argues that the delay in this case has prejudiced him because Drs. Menne and David, two of the doctors who conducted the studies, have died, as has Brandenfels’ original attorney in the case. Although there was a long delay between petitioner’s request for a hearing and the FDA’s denial of that request, there has been no showing of prejudice, and therefore reversal under Equal Employment Opportunity Commission v. Liberty Loan Corp., 584 F.2d 853 (8th Cir.1978), is not required. In fact, petitioner has reaped a benefit since the delay has enabled him to continue marketing his product. Nor was petitioner harmed by reason of the deaths of two doctors in charge of petitioner’s studies. The issue here is not whether the views of the deceased doctors are true but whether the studies on their face complied with FDA guidelines.

III. Collateral Estoppel

We finally hold that the FDA is not estopped from denying the efficacy of the treatment because a Post Office proceeding against petitioner was dismissed without prejudice. First, the issues were not the same. The issue in the Post Office case concerned the accuracy of advertising done by Brandenfels. In contrast, the issue before the FDA is whether the data submitted satisfy the FDA’s definitions of what constitutes “adequate and well-eon-trolled” studies. Second, dismissals without prejudice do not constitute a final determination to which collateral estoppel is applied. United States v. Seckinger, 408 F.2d 146, 149 (5th Cir.1969), aff’d in pertinent part, 397 U.S. 203 , 206 n. 6, 90 S.Ct. 880 , 883 n. 6, 25 L.Ed.2d 224 (1970); IB J. Moore & T. Currier, Moore’s Federal Practice ¶ 0.443[4] (2d ed. 1982).

AFFIRMED.