This appeal is a challenge to the action of the Commissioner of Food and Drugs in terminating provisional approval of the color additive FD&C Red No. 2 ("Red No. 2") pursuant to the Transitional Provisions of the 1960 Color Additives Amendments to the Federal Food, Drug, and Cosmetic Act.1 The Certified Color Manufacturers Association and other parties (collectively referred to as "CCMA" or "appellants") are appealing from the order of the District Court granting summary judgment to the defendants-appellees and dismissing CCMA's complaint with prejudice, and thus frustrating appellants' effort to enjoin the publication of the Commissioner's order in the Federal Register. Appellants seek reversal of the District Court order, and a remand with directions to enter judgment for them and to enter an order directing the Commissioner to restore Red No. 2 to the provisional list.

Appellants urge that the District Court erred in holding that (1) the Commissioner had not acted arbitrarily or capriciously, or in excess of statutory authority, in terminating the provisional listing of Red No. 2; and (2) such action need not be preceded by notice and opportunity for comment.

We affirm the order of the District Court.

To provide a proper perspective from which to analyze the regulatory action here challenged, it is helpful first to describe the statutory framework within which the action was taken. The Color Additive Amendments of 1960 reflect a Congressional and administrative response2 to the need in contemporary society for a scientifically and administratively sound basis for determining the safety of artificial color additives, widely used for coloring food, drugs, and cosmetics. The Amendments reflect a general unwillingness to allow widespread use of such products in the absence of scientific information on the effect of these products on the human body.3 The previously used system had some glaring deficiencies, and the 1960 Amendments were designed to overcome them. This was accomplished by the establishment of a dual system of registration: a permanent listing and a provisional listing. A color additive would be permanently listed if those desirous of producing it had proven to the satisfaction of the Commissioner that it was safe for its intended use. Generally speaking, until such listing, certain color additives were provisionally listed for a reasonable interim period pending completion of the scientific investigations necessary to make a safety determination.

The statute required that a color additive be "deemed unsafe" within the meaning of various sections of the Federal Food, Drug, and Cosmetic Act for use in food, drugs or cosmetics4 unless the Commissioner5 had issued a regulation which stated that the additive had been found to be suitable and safe for general use, or for a particular use.6 The burden of establishing safety was placed on the industry. The Commissioner was required to exercise his scientific judgment7 in determining whether safety had been proven, but was expressly precluded from listing any additive for any use if it was found to induce cancer in a human or animal.8

In order to avoid a statutory presumption that all additives not permanently listed at the date of enactment were unsafe, Title II of the Amendments ("Transitional Provisions") provided for the continued use of commercially established additives such as Red No. 2 on an interim basis. This was allowed only "to the extent consistent with the public health, pending completion of the scientific investigations needed as a basis for making determinations" on the safety of the additive for permanent approval.9 This provisional listing was to expire on a "closing date" which was established as either 12 January 196310 ("initial closing date"), or such later date as the Commissioner determined. The original closing date could be postponed "if in the (Commissioner's) judgment such action (was) consistent with the objective of carrying to completion in good faith, as soon as reasonably practicable" the investigation needed for making the safety determination.11

Throughout this transitional period, while the permanent safety determinations were being made, the Commissioner was granted broad discretionary authority summarily to terminate a provisional listing "forthwith whenever in his judgment such action (was) necessary to protect the public health."12 During the period following postponement of the closing date, the Commissioner additionally was given broad authority to terminate that postponement "at any time" if he found that, inter alia, the basis for the postponement no longer existed because of a change in circumstances.13

It is thus apparent that after 12 January 1963 the exercise of either one of these discretionary powers would achieve the same result: the affected color additive would be removed from the marketplace; no regulation listing it for permanent approval would be in effect, and it therefore would be "deemed unsafe."14

In the instant case, the Commissioner exercised both of these powers. He terminated the postponement of the closing date of Red No. 2, and additionally terminated its provisional listing.

Red No. 2 is a petroleum derived color additive widely used in this country artificially to create or brighten the white, brown, purple, or red colors of various foods, drugs, and cosmetics. It is used alone or in combination with other compounds to attain the desired color. It has no apparent nutrient value and its primary function is to enhance the eye appeal of a particular product.15 It has been widely used in this country under federal regulatory control for nearly seventy years.16 Since the initiation of more sophisticated scientific investigations following enactment of the 1960 Color Additive Amendments, it has failed to receive FDA permanent approval for safety.

Acting pursuant to his authority under the Transitional Provisions, the Commissioner has postponed the closing date a total of fifteen times, either at the request of interested parties or on his own initiative.17 The most recent postponement occurred 5 January 1976 and was to expire 30 September 1976.18 A petition for permanent listing pursuant to 21 U.S.C. § 376 has been under active consideration since late 1968,19 but had not been acted on prior to the action here challenged.20 At one point the agency was on the verge of affirmatively acting on the petition;21 before such action could be taken,22 however, information became available on several studies undertaken in the Soviet Union which raised new concerns over the safety of Red No. 2. One study concluded that amaranth, the generic name for Red No. 2, was poisonous to the reproductive organs of laboratory rats; the remaining studies concluded that amaranth was carcinogenic, that is, it caused cancer.

While FDA scientists questioned the validity of the Soviet tests,23 the new studies pointed out the absence of information on the effects of Red No. 2 on reproductive physiology and raised concern over its possible carcinogenicity. Accordingly, the FDA undertook its own chronic feeding study over a two-and-one-half year period in which Osborne-Mandel rats were fed Red No. 2 in their diet at concentrations of 3%, 0.3%, 0.003%, and 0%. The duration of that study and the follow-up evaluations have been the primary reason for postponing the closing date of Red No. 2's provisional listing for the last several years.

The most recent postponement came after a meeting of the Toxicological Advisory Committee ("TAC") established by the Commissioner pursuant to the statute24 to make recommendations on the safety of Red No. 2. At its November 1975 meeting the TAC saw no reason to recommend against postponing the closing date, thus continuing the period for evaluation. It recognized that the feeding study was severely flawed because the dosage levels of Red No. 2 in the various rat groups had not been maintained and because there had been substantial tissue loss due to autolysis. It nevertheless believed that portions of the study could still be used to establish safety25 and recommended continuing review, including a biostatistical study of the data.26 Accordingly, on 14 November 1975 the Commissioner published notice of his intent to postpone the closing date of the provisional listing of Red No. 2, along with all other provisionally listed additives, stating as his reason that "(r)eview of the data submitted to establish (safety) is underway . . . ."27 After receiving no adverse comments, the closing date was formally postponed on 5 January 1976.28

In the interim, Dr. D. W. Gaylor had performed the biostatistical analysis requested by the TAC. In a memorandum to the Chairman of the TAC, dated December 1975, Dr. Gaylor presented his analysis and concluded that "it appears that feeding FD&C Red No. 2 at a high dosage results in a statistically significant increase in a variety of malignant neoplasms among aged Osborne-Mandel female rats."29 In preparation for his statistical analysis, Dr. Gaylor had isolated malignant neoplasms found in 44 rats in the low dosage group and in 44 rats in the high dosage group. His analysis resulted in findings that (1) for those animals dying before the end of the study, that is, before the passage of 131 weeks, the proportion of those dying with cancer in the high dosage group was not significantly higher statistically than those dying with cancer in the low dosage group;30 (2) for those rats sacrificed at the end of the study and found to have malignant neoplasms, it was determined that there was less than a 5%, or a 4%, probability that the higher frequency of malignant neoplasms found in the high dosage group would be due to random occurrence;31 and (3) in evaluating the experiment for all animals (interim deaths and terminal sacrifice), it was found that there was less than a 2% probability that the higher frequency of malignant neoplasms found in the high dosage group would be due to random occurrence.32

The statistical relationship thus uncovered was viewed as a matter of "great concern,"33 and a special ad hoc "Working Group"34 was convened on 14 January 1976 to review the study. It submitted a report to the Commissioner on 16 January 1976 in which it concluded: (1) given the deficiencies in the chronic feeding study, "inferences drawn by Dr. Gaylor represent the minimum possible observable differences," (2) the chronic feeding study "could never be used as a demonstration of safety," and (3) the principles of the statistical analysis were accepted, but the strength of any conclusions would depend on reanalysis "using a slight redefinition of tumor types."35

Following receipt of this report, on 19 January the Commissioner prepared, and gave notice of intent to publish, a regulation to terminate the provisional listing of Red No. 2.36 The order was signed on 23 January, to become effective on publication in the Federal Register. Prior to publication, CCMA filed its complaint in the District Court and was granted a temporary restraining order the next day. After consideration of the papers filed, the administrative record, and oral argument of the parties, the District Court vacated the temporary restraining order, denied CCMA's request for a preliminary injunction, granted summary judgment to the defendants-appellees, and dismissed the complaint with prejudice.37

Following CCMA's unsuccessful attempts to obtain a stay from this court and the Supreme Court, the regulation was published on 10 February 1976.38 The Commissioner therein reviewed the general progression of events leading to the most recent postponement of the closing date, the Gaylor analysis and the Working Group's report, and concluded that provisional listing of Red No. 2 was "no longer appropriate."39 Accordingly, he terminated the provisional listing and additionally terminated the postponement of the closing date.

He concluded that termination of the listing was "necessary to protect the public health" because of the concern engendered by the Gaylor analysis and because the available data did not overcome that concern. The Commissioner stated:

. . . The provisional listing of FD&C Red No. 2 should be terminated because such action is necessary to protect the public health, in that questions have been raised about the safety of the color additive and the available data do not permit a determination of safety.40

He terminated the postponement of the closing date on a finding that the basis for the last postponement no longer existed because of the change in circumstances brought about by the Gaylor analysis and the conclusion of the Working Group that the feeding study could not be used to demonstrate safety. He further stated:

This postponement was based on the assumption that the studies being evaluated could establish the safety of the color additive. Since it has now become clear that the studies cannot satisfactorily resolve this issue, the Commissioner finds that the basis for the postponement no longer exists and hereby terminates the postponement of the closing date in accordance with section 203(a)(2) of the transitional provisions of the Color Additive Amendments of 1960.41

"Cessat ratione, cessat" postponement.

Each action resulted in the termination of the provisional listing of Red No. 2. If we find, therefore, that the District Court was correct in determining that the Commissioner acted properly in taking either action, the effect is to sanction the general conclusion of the Commissioner.

The Transitional Provisions are silent as to the scope of review of agency action such as is involved here. Provision is specifically made for the review of a termination of a provisional listing, but only for actions taken prior to 12 January 1963. For a decision on termination made before that time, judicial review was allowed only after an aggrieved party had two opportunities to present evidence supportive of his position in post-determination agency proceedings. After thus failing to convince the Commissioner that he should reinstate a provisional listing, a party in the position of appellants could file a petition for review in an appropriate court of appeals where review would be had only on a claim that the Commissioner's determination was not based on a "fair evaluation of the entire record."42

No explanation seems readily apparent for this absence of specific review authority after 12 January 1963. No provision, however, indicates a specific intent to deprive an aggrieved party of the right to seek judicial review for any termination taken after the postponement of the original closing date. Although not urged upon us, we have independently given consideration to the possibility that Congress intended to withhold judicial review and create unreviewable discretionary authority43 in the Commissioner. We have determined that given the general presumption of reviewability44 and the failure of the agency which principally authored the statute45 to interpret it as precluding review, jurisdiction did in fact exist in the District Court to review the agency action under the Administrative Procedure Act.46

Accordingly, the appropriate standard for review of the evidence involved in the challenged agency action is that provided by section 10(e)(2)(A) of the Administrative Procedure Act,47 which requires a district court to "hold unlawful and set aside agency actions, findings and conclusions" which are found to be "arbitrary, capricious, an abuse of discretion or otherwise not in accordance with law."48 The District Court followed this standard; the parties agree that it should be applied, but predictably disagree over the result of such review.

It is well recognized that this standard of review is a narrow one which presumes agency action to be valid.49 It requires that a court not substitute its judgment for that of the agency, so long as a rational basis exists, or may be reasonably discerned, for the agency determination.50 We "must consider whether the decision was based on a consideration of the relevant factors and whether there has been a clear error of judgment."51 It is only where there is no rational nexus between the facts found and the choices made that a court is authorized to set aside the agency determination. Our task is at an end when, after conscientiously adhering to this standard, we find that such a nexus exists.

Application of this standard is made even easier where as here the discretionary authority is very broad and the statutory limitations on the exercise of that authority are relatively broad as well. The breadth of this discretion is highlighted by the following unequivocal language contained in a letter sent by the Secretary of Health, Education and Welfare to the Speaker of the House of Representatives upon the submission of certain amendments to the proposed Color Additive Amendments:

Were it not for this discretionary or judgmental aspect of the transitional provisions, it is doubtful that we could have considered submission of the bills as a departmental proposal compatible with our responsibility for the protection of the public health.52

To the extent that the instant case involves questions of statutory construction, our review is guided by the considerable deference traditionally owed the interpretation of a statute by the head of the agency charged with its administration.53 Statutory construction is, of course, more properly the assigned role of a court of law. Nevertheless, we believe that the traditional deference is heightened when the statutory interpretation involves a remedial statute designed to protect the public health and principally authored by the implementing agency.

A. Termination of the Postponement of the Closing Date

Appellants urge that the District Court erred in holding that the Commissioner had not acted in an arbitrary and capricious manner, nor in excess of statutory authority, in terminating the postponement of the closing date. In support of their position they argue that, under the transitional provisions, a postponement of the closing date taken to allow for the evaluation of data for a determination of safety may not be terminated on the basis of a change in "circumstances" when in fact the evaluation has not ceased at the time of termination.54 While this argument is initially appealing, we believe that it is based on a much too confined interpretation of the statute and of the Commissioner's actions.

It will be recalled that the statutory authority to grant a postponement is couched in broad, discretionary language and allows for the interplay of judgmental factors.55 While the statute speaks in terms of completing the necessary "scientific investigations," there is also a pervasive legislative gloss that the postponement must also be consistent with the protection of the public health.56

Here the Commissioner had postponed the closing date on the assumption that studies in progress or under evaluation could establish the safety of Red No. 2. At that time, review of the results of the feeding study had not uncovered any evidence indicating that Red No. 2 had exhibited carcinogenic effects. Dr. Gaylor's biostatistical analysis had not yet become known to the Commissioner. Once it had become known, and the Working Group had found that, in viewing it in conjunction with the generally flawed nature of the feeding study, the study could not be used to prove safety, the Commissioner determined in his judgment that a change of circumstances had occurred. The District Court held that this finding was not arbitrary and capricious, and we agree.

We cannot accept the thrust of appellant's argument that the "change in circumstances" standard is so restrictive as to preclude absolutely the Commissioner from taking discretionary action prior to the completion of the evaluation which gave rise to the postponement.57 It is reasonable to assume, as the Commissioner did,58 that the standard is broad enough to embrace situations such as present here, where evidence suggestive of carcinogenicity is uncovered following postponement and it thereafter appears that the principal study under review cannot be used to assess safety. Admittedly, all studies performed on Red No. 2 were under evaluation, but the primary study at issue was the feeding study. Once it became apparent to the Commissioner that the study could not support a finding of safety and, indeed, was suggestive of carcinogenicity, there was a rational basis for terminating the postponement of the closing date.59

Appellants nevertheless argue that the Commissioner exceeded his statutory authority by considering the factor of safety.60 They posit the theory that safety considerations are patently irrelevant under the Transitional Provisions and that they only enter into the decisionmaking process under the Permanent Provisions of the Amendments. They urge that since provisional listing for commercially established additives, such as Red No. 2, was automatic, a "presumption of safety" attached to the additive which could only be overcome by a "finding of hazard to health," or unless the postponement was otherwise terminated.61

We find this argument to be without merit since it rests on a fundamental misconstruction of the plain meaning of the statute. Initially it should be noted that the "presumption of safety" was not predicated on then existing evidence, but rather was a statutory presumption deemed necessary to avoid an automatic ban on the additive upon the date of the enactment.62 As discussed in Part I(A) of this opinion, it was always subject to broad discretionary controls. Additionally, there is no mention anywhere in the statute that a "hazard" to health must exist before discretionary action may be taken. The standard is the much less restrictive one of action "necessary to protect the public health."63

Moreover, postponement of the closing date is expressly related to safety considerations and presupposes that the studies under evaluation could support a finding of safety. It appears inevitable, then, that consideration of the "basis" for postponement in making a termination decision necessarily involves reevaluation of that factor. It is a fundamental rule of statutory construction that legislative enactments be construed in a manner designed to give effect to all parts while avoiding a result contrary to the apparent intent of the Congress.64 The argument advanced by appellants would clearly be inconsistent with that rule.

The principal issue raised by appellants is that the District Court erred in failing to find that the Commissioner had acted arbitrarily and capriciously, and in excess of statutory authority in terminating the provisional listing of Red No. 2. We shall first consider the statutory argument and then the evidentiary considerations.

The primary thrust of appellants' statutory argument is that the standard authorizing the Commissioner to act "forthwith whenever in his judgment such action is necessary to protect the public health"65 requires that there be an actual threat to the public,66 amounting for all practical purposes to an imminent health hazard.67 We believe this argument is without foundation. When Congress has intended so to restrict the discretion available to an agency to take regulatory action similar to that involved here, it has not hesitated to indicate that intent by the inclusion of appropriately restrictive language.68 Here Congress attached no such "qualifiers" and chose instead to provide for a much broader standard, designed, we think, to allow for precautionary or prophylactic responses to perceived risks. There is nothing in the statute or its legislative history to indicate that Congress intended to restrict regulatory action to situations where there was a substantial likelihood of serious harm to the public.69 The overall Congressional intent appears to be to the contrary.

Indeed, it appears that Congress clearly recognized that the need for broad authority was both necessary and desirable, given the relative value attached to color additives.70 It should be noted that we deal here with a product which has, literally and accurately, a cosmetic value. Accordingly, there is no precise statutory indication as to the level of concern necessary to justify, in the Commissioner's judgment, the taking of protective action.

Nor is the legislative history much more indicative. The only clue is contained in a letter written to the Speaker of the House of Representatives by the Secretary of Health, Education and Welfare expressing, inter alia, opposition to the suggestion of a witness at the Committee hearings that termination be allowed only on the basis of "substantial probative evidence." In successfully urging retention of the present broad standard, the Secretary stated that termination could occur when there was a "substantial question as to . . . safety."71 Counsel for appellants would have us read this statement as creating a "legislative gloss" that termination is authorized only "where the evidence of harm (is) 'substantial.' "72 Clearly no such interpretation of the Secretary's remark is justified.

Furthermore, a court must keep in mind that the Commissioner's action is in the nature of a preliminary assessment of available information and that the underlying data will be thoroughly reviewed in the context of a post-determination formal hearing. In the circumstances presented here, where the termination occurs after the initial closing date, that administrative review will take place during the formal hearings following denial of the petition for permanent listing.73

Our review of the record, in light of the broad discretion conferred on the Commissioner, convinces us that there was no abuse of discretion in terminating the provisional listing of Red No. 2. The information available to him indicated a statistically significant relationship between high dosages of Red No. 2 and the occurrence of cancer in aged female rats. That relationship concededly did not establish conclusive proof that Red No. 2 was a carcinogen, but it was at least suggestive of it, and that statistical analysis was later confirmed, albeit at lower levels of significance.74 Moreover and this was the angle from which he was to view these findings he had been advised that the principal study under evaluation could not be used to establish safety. Based on this information, he determined in the exercise of his scientific judgment that a question as to the safety of the additive existed sufficient to warrant termination. We cannot say that this was an irrational choice, nor can we say that it was beyond his statutory authority.

Courts have traditionally recognized a special judicial interest in protecting the public health,75 particularly where "the matter involved is as sensitive and frightladen as cancer."76 Where the harm envisaged is cancer, courts have recognized the need for action based upon lower standards of proof than otherwise applicable.77 While we might not have arrived at the same conclusion as the Commissioner did, we cannot say that the District Court was obligated to hold, as a matter of law, that the action was irrational under the circumstances. There was a rational nexus between the facts found and the decision made.

Therefore, in light of the broad discretionary authority available to the Commissioner, and our necessarily limited role as judges, we affirm the District Court's holding on this issue.78 We note, however, that the Commissioner has denied appellants' petition for permanent listing, and we assume that this has triggered a formal proceeding before the agency which will result in a new order being issued based on a complete reevaluation of the evidence. That is a separate proceeding from which there will arise another opportunity to seek judicial review.79

Appellants lastly urge that the District Court erred in finding that the notice and comment provisions of the Administrative Procedure Act80 did not apply to the actions challenged.81 They urge that since it is generally accepted that the Act "should be received hospitably,"82 and that exclusions from the rule are rarely found, the District Court should be reversed and the Commissioner thereby obligated to begin the termination process anew by providing notice and allowing for opportunity to comment.83

The arguments presented in support of this proposition are without merit. The Commission is authorized to terminate the postponement of the closing date "at any time"84 and may terminate a provisional listing " forthwith."85 Such discretionary authority, expressly designed to avoid any time limitations, are "so inconsistent with any notice-and-comment requirement as to compel the conclusion that a pre-determination hearing cannot be implied from the terms of the statute."86

Lastly, the fact that the Commissioner gave notice and provided opportunity for comment when he postponed the closing date and when he proposed the establishment of tolerance limitations on the use of Red No. 2 establishes no precedent or parallel for our purpose here. A review of the statutory provisions authorizing such actions indicates that they are not controlled by the terms "at any time" or "forthwith."IV. CONCLUSION

For the above-stated reasons, the judgment of the District Court is

Affirmed.