United States v. Articles of Drug, 624 F. Supp. 776 (N.D. Ill. 1985)

*777 Michael S. O'Connel, Asst. U.S. Atty., Chicago, Ill., Ann H. Wion, FDA Associate Chief Counsel for Enforcement, Rockville, Md., for plaintiff.

Joan M. Hall, Daniel J. King, Jenner & Block, Chicago, Ill., Alan H. Kaplan, Thomas O. Henteleff, David E. Kaplan, Kleinfeld, Kaplan & Becker, Washington, D.C., for defendants.

GETZENDANNER, District Judge:

In this drug enforcement action, the United States seeks seizure and condemnation of certain over-the-counter ("OTC") drug products, Promise toothpaste with Fluoride and Sensodyne-F toothpaste. The government alleges that these products are "new drugs" within the meaning of 21 U.S.C. § 321(p) (1), and were introduced into interstate commerce without an approved New Drug Application ("NDA"), as required by 21 U.S.C. § 355(a). The government also seeks injunctive relief against the manufacturer Block Drug Company, Inc. ("Block"), and two of its officers, Leonard N. Block and James A. Block, to restrain further marketing of these products. Block has filed a claim in the in rem proceeding.

On July 30, 1984, this court denied the defendants' motion to dismiss and ruled *778 that the Compliance Policy Guide ("CPG") of the Food and Drug Administration ("FDA") did not bar the present action pending completion of further administration review; the court also ruled that a remand of proceedings to the FDA would be inappropriate. United States v. Articles of Drug ... Promise Toothpaste for Sensitive Teeth, 594 F. Supp. 211 (N.D.Ill. 1984). The opinion denying that motion contains an extensive overview of new drug regulations and the FDA's review of OTC dentifrices, which will not be repeated here. The matter is currently before the court on the government's motion for summary judgment, or alternatively, for a preliminary injunction. For the reasons set forth herein, the motion for summary judgment is granted.

The sole issue presented by the motion for summary judgment is whether, as a matter of law, Promise with Fluoride and Sensodyne-F are "new drugs" within the meaning of § 321(p) (1). Under the Food Drug and Cosmetic Act, a drug product must be classified as a new drug unless its manufacturer can show that it is "generally recognized among experts" to be "safe and effective" for its recommended uses. 21 U.S.C. § 321(p) (1). It is established by regulation that the "newness of a drug may arise by reason ... of" combining two or more drugs. 21 C.F.R. § 310.3(h). See also 21 C.F.R. § 330.10(a) (4) (iv). Thus, even though the component parts of a new drug may be generally recognized as safe and effective, the combination itself may not be. See United States v. Articles of Food and Drug ... Coli-Trol 80, 518 F.2d 743, 746 (5th Cir. 1975); United States v. An Article of Drug ... Entrol-C Medicated, 513 F.2d 1127, 1129 (9th Cir.1975); United States v. An Article of Drug ... Neo-Terramycin, 540 F. Supp. 363, 376 (N.D.Tex.1982), aff'd 725 F.2d 976 (5th Cir. 1984); United States v. X-Otag Plus Tablets, 441 F. Supp. 105, 111 (D.Colo.1977), aff'd in part, remanded in part, 602 F.2d 1387 (10th Cir.1979); United States v. An Article of Drug ... Mykocert, 345 F. Supp. 571, 575 (N.D.Ill.1972). The rationale for this rigorous treatment of combination drugs is that a single component of a drug may react with other components, thereby reducing the total effect of the drug or producing unexpected side effects. Coli-Trol 80, 518 F.2d at 746; X-Otag Plus Tablets, 441 F. Supp. at 111.

The Food Drug and Cosmetic Act nowhere defines the term "generally recognized among experts ... [as] safe and effective." Weinberger v. Hynson, Westcott & Dunning, Inc., 412 U.S. 609, 629, 93 S. Ct. 2469, 2483, 37 L. Ed. 2d 207 (1973); United States v. Articles of Drug ... 5,906 Boxes, 745 F.2d 105, 119 n. 22 (1st Cir.1984), cert. denied, sub nom. Alcon Laboratories, Inc. v. United States, ___ U.S. ___, 105 S. Ct. 1358, 84 L. Ed. 2d 379 (1985). At a minimum, however, courts have construed "general recognition" to require a consensus of expert opinion based on the same "substantial evidence" of effectiveness based on "adequate and well-controlled investigations" which would be required for approval of a new drug application under 21 U.S.C. § 355(d). See, e.g., Weinberger v. Bentex Pharmaceuticals, Inc., 412 U.S. 645, 653, 93 S. Ct. 2488, 2494, 37 L. Ed. 2d 235 (1973) (reach of scientific inquiry for effectiveness under § 355(d) and § 321(p) is "precisely the same"); Hynson, 412 U.S. at 628-32, 93 S. Ct. at 2482-84 (procedural requirements applicable for approval of new drugs must be applied to the definitional requirements for "new drugs"); 5,906 Boxes, 745 F.2d at 116-19 (same); United States v. An Article of Drug ... 4,680 Pails, 725 F.2d 976, 985-87 (5th Cir.1984) (general recognition requires a showing of both expert consensus in fact and substantial evidence (i.e., adequate and well controlled investigations) to support that consensus); United States v. Western Serum Co., 666 F.2d 335, 338 (9th Cir.1982) (consensus of informed opinion must be "founded on well-controlled clinical tests").

Many courts have gone further and held that, in order for a drug to be "generally recognized" as effective, the data relied on for that opinion must not only constitute "substantial evidence" within the meaning *779 of § 355(d) but must also be published and available generally to experts in drug evaluation. See, e.g., United States v. Undetermined Quantities ... Equidantin, 675 F.2d 994, 1001 (8th Cir.1982), cert. denied sub nom. Performance Products, Inc. v. United States, 460 U.S. 1051, 103 S. Ct. 1497, 75 L. Ed. 2d 929 (1983); United States v. 41 Cases ... Naremco, 420 F.2d 1126, 1130 (5th Cir.1970); United States v. Articles of Drug ... Hormonin, 498 F. Supp. 424, 432 (D.N.J.1980), aff'd, 672 F.2d 904 (3d Cir.1981); United States v. Articles of Drug ... Colchicine, 442 F. Supp. 1236, 1243 (S.D.N.Y.1978). As the Supreme Court said in Bentex Pharmaceuticals, "[w]hether a particular drug is a `new drug' depends in part on the expert knowledge and experience of scientists based on controlled clinical experimentation and backed by substantial support in scientific literataure." 412 U.S. at 652, 93 S. Ct. at 2493. See Premo Pharmaceutical Laboratories, Inc. v. United States, 629 F.2d 795, 803 (2d Cir.1980) ("[n]ormally the general expert consensus is supported not only by clinical experience but by publication in the scientific literature"); 21 C.F.R. § 330.10(4) (ii) ("[g]eneral recognition of effectiveness shall ordinarily be based upon published studies which may be corroborated by unpublished studies and other data"). The reason behind the publication requirement is inherent in the statutory concept of a "new" drug. As explained in Premo Pharmaceutical, "only those pre-existing drug products found through careful study by scientific experts to have been generally recognized on the basis of usage and documentation to be safe and effective [are] entitled to the exemption...." 629 F.2d at 802-03. Unawareness of the drug by experts generally precludes its qualifying as a pre-existing drug. Id.

Finally, any drug which has become generally recognized as safe and effective due to clinical investigations, but which has not otherwise "been used to a material extent or for a material time" under the conditions prescribed, recommended, or suggested in the labeling of the drug must be considered a "new drug" requiring approval of an NDA before being marketed. 21 U.S.C. § 321(p) (2). See Hynson, 412 U.S. at 631, 93 S. Ct. at 2484 (Act is designed to exempt only those drugs on the market which "have mustered the requisite scientifically reliable evidence of effectiveness long before they are in a position to drop out of active regulation by ceasing to be a `new drug.'") The government in this case has not argued that Promise with Fluoride and Sensodyne-F fail to meet this particular requirement, and the court will not address the issue further.

Pursuant to the above cases, the government can prove lack of "general recognition" by proving an absence of material fact as to any of the following three issues: (1) general recognition in fact among the nation's experts that the seized drugs are safe and effective for their intended use; (2) the existence of adequate and well-controlled studies which constitute the "substantial evidence" of safety and effectiveness § 355(d) requires for approval of an NDA; and (3) generally available scientific literature substantiating an expert consensus of safety and effectiveness. The government has not alleged that the seized drugs are unsafe in any way, and the parties accordingly have confined their discussion to the question of "general recognition." The actual effectiveness of the drug is not in issue. Undetermined Quantities ... Equidantin, 675 F.2d at 1001; Premo Pharmaceutical Laboratories, 629 F.2d at 803-04; Tyler Pharmacal Distributors, Inc. v. H.E.W., 408 F.2d 95, 99 (7th Cir.1969). With these overall standards in mind (to be elaborated further), the court now addresses the evidence presented by the parties.

The following facts are undisputed. On January 1, 1983, defendant and claimant Block Drug Company began to market the products Promise with Fluoride and Sensodyne-F. (Def. Response to Pltf's First Set of Interrogatories Nos. 19 and 20). Each of these products contains a combination of .76% sodium monofluorophosphate (commonly *780 known as "MFP") to prevent cavities, and 5% potassium nitrate to desensitize dentin in people with hypersensitive teeth. The toothpaste thus serves to maintain anti-cavity fluoride protection for people who suffer from dentinal hypersensitivity and who require special toothpastes for that reason.

In 1980, sodium mfp was classified by the FDA's Advisory Review Panel on OTC Dentifrices and Dental Care Drug Products to be generally recognized as safe and effective for preventing cavities. 45 Fed. Reg. 20682-90. Given the numerous clinical trials for fluoride compounds, the panel and the American Dental Association (ADA) both recommended that satisfaction of certain laboratory testing profiles could replace further clinical investigations in determining whether a particular toothpaste containing an approved fluoride combination is safe and effective for preventing cavities. Id. at 20677-78. The panel also recommended, however, that products combining an anticaries agent such as sodium mfp with another agent be considered new drugs and be required to submit NDAs. Id. at 20674, 20690.

In May 1982, the panel issued a proposed monograph for oral discomfort products. 47 Fed.Reg. 22712 (1982). The panel concluded that potassium nitrate, the other active ingredient in defendant's toothpaste, was safe at specified dosages, but concluded that there was insufficient data to establish its effectiveness as a tooth desensitizer. Id. at 22754-55. The Panel recommended that certain combinations containing potassium nitrate be permitted; none of these combinations included mfp or any other anticaries agent. Id. at 22719-23. The panel also concluded that sodium monofluorophosphate was safe for use as a desensitizing agent, but that the data were insufficient to establish its effectiveness for the purpose. Id. at 22752-53.

The proposed monograph did not consider any data which had become available after 1978, and the panel invited the submission of comments and new data. 47 Fed.Reg. 22759. In response, both Block and Richardson-Vicks, Inc., manufacturer of Denquel, a toothpaste containing potassium nitrate, submitted post-1978 data on potassium nitrate, including at least six studies concerning the effectiveness of potassium nitrate, as well as copies of several studies published in scientific or medical journals. The ADA and the Proprietary Association, a health organization, also submitted comments on potassium nitrate; both organizations recommended it for its effectiveness as a tooth desensitizer. (Def.Ex. 8 at 2; Ex. 9 at 9-10). While there appears to be significant scholarly dispute on this point, see Mandel Dep. Exs. 8 & 9, at least one of the government's experts, admitted his position to be a minority view. (Ash Dep. at 14). For purposes of the present motion, the court will therefore assume that 5% potassium nitrate is recognized by a majority of qualified experts as safe and effective for desensitizing teeth, and will confine its discussion to defendant's combination of 5% potassium nitrate and .76% sodium mfp in a single product.

On October 18, 1982, Block also filed data on the safety and effectiveness of its MFP/potassium nitrate combination. This data contained no clinical testing of the combination, but discussed the rationale for the combination, toxicology data, fluoride bioavailability to demonstrate anticaries effectiveness, and nitrate bioavailability to demonstrate desensitizing effectiveness. On April 25, 1983, after it had begun marketing the drug, Block submitted additional fluoride bioavailability tests and animal anticaries studies, 4- and 8-week human safety studies, and the results of several clinical studies on potassium nitrate previously submitted by Richardson-Vicks.

On September 7, 1983, Block submitted the results of a recently completed 12-week clinical investigation (hereafter "Silverman study"), which tested a 5% potassium nitrate toothpaste (Vick's "Denquel"), a Block sodium MFP/potassium nitrate toothpaste, and a placebo toothpaste, containing only the toothpaste vehicle used in the seized products (i.e., no active ingredients) *781 (Def. Ex. 22). The study did not include any testing of a .76% sodium mfp toothpaste, and did not test for cavity prevention. Sixty-eight patients were enrolled in the study, all of whom had complained of hypersensitive teeth and showed physical symptoms associated with hypersensitivity. A dental history taken before admission indicated that approximately 80% of these patients used fluoride toothpastes regularly before starting the study. The results showed that both Promise and Denquel reduced tooth hypersensitivity to a statistically significant greater level than the placebo toothpaste. There was no statistically significant difference between Promise and Denquel. The Silverman study is not published and defendant has not identified any other clinical study testing the combination product. (Def. Response to Pltf's First Set of Interrogatories, Nos. 7 and 30).

From this point on the material presented by the parties consists of opinion evidence from various experts in the field of dentistry or periodontics and differs sharply. The plaintiff has produced affidavits from three experts who have testified that they only became aware of products containing the combination sodium mfp and potassium nitrate when contacted by FDA employees in connection with this litigation. See Ash Declaration ¶ 6; Carlos Declaration ¶ 10; Trummel Declaration ¶ 5. Each of these experts has declared that he does not recognize defendant's products as safe and effective for their labeled indications, Ash at ¶ 13, Carlos at ¶ 9, Trummel at ¶ 9, and that he does not believe the drugs to be generally recognized by qualified experts as safe and effective. Ash at ¶ 14, Carlos at ¶ 10, Trummel at ¶ 10.

These experts have also declared that they are unaware of any published article concerning a combination of sodium mfp and potassium nitrate for either anti-cavity protection or tooth desensitizing, and believe it is scientifically unsound to extrapolate solely from data on sodium mfp or potassium nitrate to make a determination about the effects of the combination. Ash at ¶¶ 9-10, Carlos at ¶¶ 5-7, Trummel at ¶¶ 6-7. Finally, these experts have declared that they do not consider the Silverman study relied on by Block to be adequate and well controlled, because of inadequate selection of subjects, inadequate duration of the study, and inadequate methodology. Ash at ¶ 8, Trummel at ¶ 7(b), Carlos at ¶ 7. In particular, these experts criticize the failure to test the combination product against a fluoride toothpaste without potassium nitrate, so as to eliminate the possibility that the mfp and not the potassium nitrate was responsible for the decrease in dentinal hypersensitivity. Id.

To rebut the above submissions, the defendants have submitted three affidavits from experts who claim to be more qualified than plaintiff's experts in the area of caries prevention and/or dentinal sensitivity. These experts uniformly agree that Block's mfp/potassium nitrate combination is generally recognized as safe and effective for both anticaries and sensitivity treatment, and refer to at least ten other experts who share this opinion. (Mandel Declaration ¶¶ 6, 16; Ciacio Declaration ¶¶ 8, 10; Featherstone Declaration ¶ 21). According to these experts, there is no chemical, physical, or physiological basis, as substantiated through in vitro clinical studies, for concluding that sodium mfp and potassium nitrate will interact to affect the safety or efficacy of either active ingredient. Thus, the generally available scientific literature on the two different ingredients can rationally be extrapolated to form a basis for concluding that the combination found in Promise is also safe and effective. Mandel at ¶¶ 17(d), (e), 19; Ciancio at ¶¶ 13(B), 18; Featherstone at ¶¶ 14, 15, & 19(c). It is significant that the government experts, while disputing this assertion, have conceded that they have no theoretical basis for assuming that potassium nitrate is likely to interact with sodium mfp. See, e.g., Ash Dep. at 75.

Defendants' experts have also disagreed with the criticisms of the government experts regarding the Silverman study. Mandel at ¶ 17; Ciancio at ¶ 19; Featherstone at ¶¶ 18-20. In particular, these experts maintain that it is questionable whether *782 sodium mfp has a desensitizing effect, and therefore that Silverman's failure to test Promise against a fluoride toothpaste does not invalidate the research results, particularly since 80% of the test participants had used fluoride toothpaste before entering the study. Id. These experts further note that the duration of the experiment 12 weeks is in excess of the 8-week period recommended by the FDA expert advisory panel and considered adequate by the ADA in testing for the safety and effectiveness of potassium nitrate alone. Id. For purposes of the present motion, the court has assumed that the opinions expressed by defendants' declarants represent those of the majority of qualified experts. The court will assess whether relevant statutory standards nonetheless compel a conclusion that Promise with Fluoride and Sensodyne-F are "new drugs."

In its memoranda for summary judgment, the FDA argues, as it apparently has in other cases, see, 5,906 Boxes, 745 F.2d at 119 n. 22, that any "conflict" of opinion among medical experts is sufficient to mandate a legal conclusion that general recognition of Promise's effectiveness does not exist. Many cases contain a statement to this effect. See, e.g., Premo Pharmaceutical Laboratories, 629 F.2d at 803 ("either the unawareness of the drug product by experts generally or a genuine dispute among qualified experts regarding a drug product's safety and effectiveness preclude its qualifying for exclusion as `generally recognized.'"). Accord, Equidantin, 675 F.2d at 1000; Western Serum Co., 666 F.2d at 338; United States v. An Article of Drug ... "Bentex Ulcerine," 469 F.2d 875, 880 (5th Cir.1972), cert. denied, 412 U.S. 938, 93 S. Ct. 2772, 37 L. Ed. 2d 397 (1973); United States v. Articles of Drug ... Hormonin, 498 F. Supp. 424, 431 (D.N. J.1980), aff'd, 672 F.2d 904 (3d Cir.1981).

Despite this impressive array of authority, the court must disagree. The above line of cases can be traced back to dicta in United States v. 354 Bulk Cartons ... Trim Reducing-Aid Cigarettes, 178 F. Supp. 847 (D.N.J.1959) and Merritt Corp. v. Folsom, 165 F. Supp. 418 (D.D.C.1958), in which it was stated that a mere conflict between qualified experts establishes a lack of general recognition as a matter of law. In United States v. 7 Cartons ... Ferro-Lac, 293 F. Supp. 660 (S.D.Ill.1968), aff'd in part, vacated in part, 424 F.2d 1364 (7th Cir.1970), the district court thoughtfully pointed out the inadequacies of this statement as a rule for general application:

[T]he government's contention cannot be accepted as a sound principle of procedural law for application to this statute. The statute provides a test of general recognition. The rule stated in Merritt and subsequent cases would equate the statutory language, "generally recognized" to "unanimously recognized." There is no justification for such equation. There is nothing in the statute to indicate that Congress intended "generally recognized" in other than its commonly understood meaning. The adverb, "generally," is defined, inter alia, to mean "... extensively, though not universally" ....

Where, as here, each party has submitted affidavits in support of its position, the court must examine those affidavits to determine whether a genuine difference of expert opinion is presented. If such difference of opinion appears, decision must abide the results of a trial of the issues.

293 F. Supp. at 662-63 (citations omitted). See also 5,906 Boxes, 745 F.2d at 119 n. 22. ("[U]nanimity among experts is not required to demonstrate `general' recognition.") United States v. Articles of Food and Drug (Coli-Trol 80), 518 F.2d 743, 746 (5th Cir.1975) ("It is not enough to show merely a conflict in the evidence of general recognition, for even properly conducted studies may produce disagreement. What is required is not unanimous recognition but general recognition.")

The implications of this principle for the present case are obvious. The government *783 should not be given license to seize non-new drugs simply because it can find a small contingent within the scientific community dissenting from the view that the drugs are safe and effective. By the same token, however, the manufacturer cannot avoid summary judgment simply by recruiting experts to testify that a particular drug is generally recognized as effective: "[t]he congressional purpose of removing the complex medical and scientific aspects of drug disputes from the courts to the expert province of the FDA would be frustrated if any and every divergence in expert opinion could serve to send a new drug case to the jury." 5,906 Boxes, 745 F.2d at 119 n. 22. Hence, a district court must scrutinize opposing affidavits carefully to determine whether there really is a triable issue of disputed fact as to general recognition. Id.

As explained earlier, the relevant issue on the present motion is not whether potassium nitrate is generally considered by experts to be "safe and effective" for desensitizing teeth, but whether the combination of potassium nitrate and sodium mfp is safe and effective both for combating cavities and dentinal sensitivity. The government argues that the conflict among experts in this case is sufficient to preclude a finding of general recognition at trial. In particular, the government stresses defendants' failure, after deposition, to impeach the credentials of those experts supporting the government's position. While the question is close, the court finds on the present record that it cannot determine the existence of a "genuine" difference of opinion among experts without improperly assessing the probative weight of the evidence presented. See United States v. An Article of Food ... 345/50-Pound Bags, 622 F.2d 768, 773 (5th Cir.1980) (court should not "weigh" conflicting evidence on motion for summary judgment in drug seizure case). The court has reviewed the depositions of the various experts and finds no basis on which one set of experts can undisputedly be preferred over the other. On the government's side, the court finds significant that defendants' experts all appear to rely on unpublished data provided to them by defendants. Nonetheless, defendants have presented evidence that many if not most experts recognize 5% potassium nitrate as safe and effective (government expert Timothy Ash admitted upon cross-examination that his view in this regard was a minority position). Defendants have also presented evidence that there is no reasonable scientific basis for distinguishing the effects of sodium mfp and potassium nitrate when combined from their effects when used separately, and that most scientists therefore recognize defendants' products as safe and effective. Finally, defendants have submitted a survey which shows that 75% of periodontists recommend a desensitizing toothpaste containing fluoride, and that over fifty percent recommend Promise. (Def.Ex. 17). While there are apparent grounds raised by the government for criticizing the survey, and while defendants' inability to impeach the government witnesses thoroughly suggests a bona fide dispute among qualified experts, the court cannot conclude on the present record that the government has established a lack of general recognition in fact which would warrant a grant of summary judgment. The court therefore turns to the other arguments raised by the government.

Given the rigors of the FDA's combination policy, many courts have held that a combination of known drugs be considered a "new drug" absent controlled clinical experimentation and substantial support in the scientific literature regarding the combination in its exact form or dosage. See Entrol-C Medicated, 513 F.2d at 1129; Neo-Terramycin, 540 F. Supp. at 376; Mykocert, 345 F. Supp. at 575; United States v. Asper-Sleep [1970-1973 Transfer Binder] Food Drug Cosm.L.Rep. (CCH) ¶ 40,836 (N.D.Ill. April 30, 1971). Cf. International Nutrition, Inc. v. H.H.S., 676 F.2d 338, 341 (8th Cir.1982) (upholding FDA rejection of NDA for animal drug combination product where manufacturer combined two approved *784 drugs without showing that the combination itself was safe and effective).

Although FDA policy regarding OTC drugs does not always require increased effectiveness, see 37 Fed.Reg. 9470 (1972) (increased effectiveness not necessary absent a decrease in safety), each ingredient described as active must nonetheless make a contribution to the effect claimed or suggested in the labeling. 21 C.F.R. § 330.10(a) (4) (iv). Proof that the active ingredients are separately recognized as effective does not satisfy this requirement. Accordingly, courts have ruled that studies supporting the safety and effectiveness of a combination drug will not be considered "adequate and well-controlled" within the meaning of 21 U.S.C. § 355(d) unless those studies compare the effectiveness of the combination with the effectiveness of each active ingredient as well as with a control group. Masti-Kure Product Co. v. Califano, 587 F.2d 1099, 1101-03 (D.C.Cir.1978); Neo-Terramycin, 540 F. Supp. at 377.

In the present case, it is undisputed that the only clinical (i.e., human) study conducted of Promise with Fluoride and Sensodyne-F failed to test the effects of the sodium mfp/potassium nitrate combination against the desensitizing effects of sodium mfp toothpaste alone. Defendants' expert Sebastian Ciancio argued that such a procedure is scientifically acceptable since it is "questionable" whether sodium mfp has a desensitizing effect (Ciancio Decl. at ¶ 19). None of the defendants' experts testified affirmatively that sodium mfp does not aid in treating dentin sensitivity, however, and only one testified that there is no scientific evidence confirming that possibility. (Ciancio Decl. at ¶ 19). This vagueness is notable since defendants' experts elsewhere categorically and affirmatively rejected the government's purported theory that sodium mfp and potassium nitrate might chemically react with one another to weaken their respective effectiveness in dentinal hygiene and treatment. (Mandel Decl. at ¶ 17(d)-(e); Ciancio at ¶ 13(B); Featherstone at ¶¶ 14-15).

Defendants' experts also point out that 80% of the test participants, chosen because they were suffering from dentinal hypersensitivity, had reported using a fluoride toothpaste before entering the study, and that the sodium mfp/potassium nitrate combination, although clinically and statistically more effective than the placebo in relieving hypersensitivity, was neither clinically nor statistically different in effect from the use of potassium nitrate alone. From these facts, defendants' experts contend that most scientists would interpret the results as indicating that the potassium nitrate component of the combination was primarily responsible for the combination's effectiveness in treating dentinal hypersensitivity.

The court does not dispute the logic of this argument. It is clear, however, that "substantial evidence" of effectiveness as defined in 21 U.S.C. § 355 is as much a term of law as it is one of scientific consensus, and that the evidence of effectiveness required for approval of a new drug application is, if anything, lesser than the evidence of effectiveness required for determining that a drug escapes "new drug" status altogether. Given defendants' failure to rebut the government's evidence that sodium mfp is potentially effective for hypersensitivity treatment and the Silverman study's failure to test the mfp/nitrate combination against mfp alone, there is no adequate and well-controlled study that potassium nitrate contributes to the desensitizing effects of fluoride toothpaste. Summary judgment for the government is therefore warranted, and the defendants must await FDA approval before engaging in further marketing of the seized drug products.

Defendants have vigorously argued that the Silverman study, when considered in light of the nitrate bioavailability studies and the clinical literature on potassium nitrate alone, constitutes substantial evidence which would support a conclusion that the mfp/nitrate combination is generally recognized as safe and effective for desensitizing teeth. While certain conditions *785 may make it scientifically appropriate to extrapolate solely from data on potassium nitrate and sodium mfp separately to make a determination about the effects of the combination, the statute and applicable FDA regulations require more. None of Block's experts was willing to say that the Silverman study on the combination was unnecessary to their determination that the combination product was safe and effective. (Mandel Dep. at 18-19; Ciancio Dep. at 21; Featherstone Dep. at 8.) Ciancio specifically stated that the Silverman study was necessary to such determination. (Ciancio Dep. at 21.) Because the Silverman study did not demonstrate that each ingredient contributes to the claimed effect, the necessary scientific basis for a judicial conclusion of general recognition simply does not exist.

In further opposition to the government's motion, defendants argue that the government is not entitled to rely on "new drug" case law concerning prescription products in determining whether nontoxic over-the-counter items such as Sensodyne-F require premarket clearance. The Act clearly requires premarket approval of "new" OTC drugs as well as new prescription drugs, however, 21 U.S.C. § 331, and the regulations bear out that the OTC combination policy is essentially the same as that used for prescription drugs. 37 Fed. Reg. 9464, 9470 (1972). While the FDA's OTC policy has been applied less strictly than its prescription policy, see Def. Brief at 20-21, that difference does not license manufacturers to ignore the strictures of the Act as to premarket clearance for all "new drugs."

Defendants' clinical studies are secondly inadequate to support a finding of general recognition since the mfp/nitrate combination has never been tested for its effectiveness in preventing cavities. The Silverman study never purports to study the anticaries effect of the drug, and defendants concede that no other clinical study exists. Defendants instead argue that Promise with Fluoride satisfies the laboratory testing profiles propounded by the FDA in 1980 for various fluoride dentifrice formulations, and that clinical testing is therefore not required. See 45 Fed.Reg. 20666, 20677. That same FDA monograph, however, expressly restricts approved OTC anticaries drug products to those containing "single active ingredients." Id. at 10674. The language about laboratory testing profiles in lieu of clinical tests is followed by a list of approved fluoride/abrasive combinations, none of which contains potassium nitrate. Accordingly, any general scientific opinion that mfp/potassium nitrate in combination is safe and effective for cavity prevention is based on the prediction that the potassium nitrate will not affect the fluoride's effectiveness, and not on controlled clinical studies testing the product itself. This predictive evidence, whatever its wide acceptance among qualified experts, is statutorily insufficient for side-stepping the administrative review procedures of the Food Drug and Cosmetic Act.

Defendants correctly point out that the Supreme Court, in dicta, once stated that certain circumstances might allow for a finding of "general recognition that a drug is efficacious ... without the kind of scientific support necessary to obtain approval of an NDA." Bentex Pharmaceuticals, 412 U.S. at 652, 93 S. Ct. at 2493. Defendants in this case suggest that the known chemical inability of potassium nitrate and mfp to interact, as has been scientifically demonstrated, avoids the need for controlled clinical studies demonstrating their effect when combined. Defendants have been unable to point to any cases which follow up on this dictum, however, and the overall tenor of the decision in Bentex Pharmaceuticals is to incorporate the standards of § 355 into the definition of "new drug" under § 321(p). The circumstance most likely envisioned by the Supreme Court is one of longstanding market acceptance so that the body of experience alone, without controlled research, could support a finding of general recognition. The relatively recent marketing of defendants' products precludes such an exception *786 here, and the court will not require a jury trial because of an undefined "possibility" suggested in passing by the Supreme Court.

The final ground on which the government asserts the propriety of summary judgment is that of publication. Because the statute requires that expert recognition of a drug's effectiveness be "general," several courts have held that the data on which that consensus is based be generally available to the community of experts before a determination of "general recognition" can be made. See, e.g., Equidantin, 675 F.2d at 1001; Naremco, 420 F.2d at 1130; Hormonin, 498 F. Supp. at 432; Colchicine, 442 F. Supp. at 1243; Mykocert, 345 F. Supp. at 574. Since it is undisputed that the only clinical in vivo study concerning defendants' products is unpublished, the government argues that absence of such publication establishes newness as a matter of law.^[1]

Defendants correctly point out that some of the authorities cited by the government do not require publication per se but instead merely note that the presence or absence of published literature should "normally" or "generally" be considered in determining whether a drug is generally recognized as effective. Bentex Pharmaceuticals, 412 U.S. at 652, 93 S. Ct. at 2493; Premo Pharmaceutical Laboratories, 629 F.2d at 803; 21 C.F.R. § 330.10(4) (iii). There is no statutory or regulatory requirement that the data supporting effectiveness be published, only that it be "substantial." Defendants therefore argue that the various studies submitted to the FDA concerning the mfp/potassium nitrate combination, including the Silverman study, constitute publicly available evidence on which a finding of general recognition could be based.

The court disagrees. The "publication" requirement imposed by many courts is based on a common sense understanding that a "lack of documented sources of information perforce curtails the widespread knowledge of the drug's effectiveness or safety" and is therefore often determinative of "newness." Mykocert, 345 F. Supp. at 574. No judicial opinion has allowed a new drug case to go to the jury on the basis of unpublished data alone, and the court agrees with the government that there is no principled basis for doing so here. A manufacturer should not be allowed to create general recognition of safety and effectiveness by sending unpublished data to the FDA for administrative review and then arguing that the submission itself creates the public availability which would obviate the need for FDA review altogether. In the present case, the lack of published clinical studies conclusively shows that Promise with Fluoride and Sensodyne-F are "new drugs" within the meaning of the Act. The fact that most scientists would extrapolate from other largely available data that the drugs are safe and effective does not change this conclusion.

The court has not addressed all of the arguments raised by the government, in part because it finds the above discussion dispositive, and also because many of the more refined criticisms made by the government go to credibility determinations which are inappropriate on a motion for summary judgment. The court does determine, however, that the government has met its burden of demonstrating as a matter of law that Promise with Fluoride and Sensodyne-F are "new drugs" within *787 the meaning of the Food Drug and Cosmetic Act. The government is therefore entitled to destroy the seized drugs and to enjoin defendants from further marketing of Promise and Sensodyne-F. The court agrees with defendants, however, that the scope of the injunction as requested goes beyond what is necessary or appropriate: the injunction should be limited to the new drugs listed in the complaint, with appropriate language to cover the possibility that defendants might attempt to market similarly labeled and/or formulated drugs under a slightly different form or name. The court will not enjoin defendants from marketing any other new drug for which no approved NDA is on file, as the government has not demonstrated the likelihood of such an event.

Accordingly, the government's motion for summary judgment is granted. The parties should confer for the purposes of preparing a draft final order. The parties should report for status on December 5, 1985. If the parties cannot agree on the language of a final order, at that time each side should submit a proposed order to the court. The defendants should also inform the court at that time whether they intend to appeal, and, if so, whether they intend to seek a stay pending appeal.

It is so ordered.

[1] After this opinion was issued, counsel for the government alerted the court to the fact that the Silverman study was published last February, after the defendants filed their opposition memorandum but well before the court ruled. Although this development was mentioned in a footnote to the government's reply brief, that footnote escaped the court's attention. The court does not believe that this subsequent publication, approximately 1½ years after the case was filed, affects its conclusion that the reliance of Block's experts on largely unpublished data precludes general recognition as a matter of law.