Dahl and seventeen others, afflicted with chronic fatigue syndrome, enrolled in an experimental program to test a new medication. HEM Pharmaceuticals makes the medication, called Ampligen. The patients received the medicine as part of the testing procedure used by the Food and Drug Administration to determine whether a medicine is safe and effective. When the test was over, HEM ceased providing the medication to the patients. They sued for injunctive and other relief, claiming that HEM promised to continue providing Ampligen to them after the study ended if statistical analysis showed efficacy compared to placebo. The district court granted a preliminary injunction, requiring HEM to continue providing Ampligen for twelve months. Because HEM initially failed to comply with the injunction, the district court held it in civil contempt. HEM appeals, claiming that it should not have been so enjoined. One of the patients cross-appeals, claiming that the preliminary injunction should not have been limited to twelve months. We affirm.

I. Facts.

New medicines go through several phases of clinical evaluation before general release onto the market. HEM conducted a clinical trial with 92 patients designed to evaluate the effectiveness, side effects, and risks of Ampligen, classified by the FDA as an "investigational new drug" not yet permitted to be sold freely as a prescription medication. The study was "double blind," which means that some patients got Ampligen, some got a placebo (saline solution), and neither the doctor nor the patient knew who was getting which. The Ampligen was administered as a liquid by slow injection into a vein. All the patients signed consent forms warning of the experimental status of Ampligen and possible side effects. Although the patients were free to withdraw at any time, if they remained in the study they were required to accept the risks of treatment, forgo other drugs, not become pregnant, and submit to uncomfortable testing.

The arrangement with the experimental subjects was that they would participate in the double-blind study for a year. This was to facilitate evaluation of the safety and effectiveness of Ampligen. After the double-blind phase of testing ended, they would be entitled to receive Ampligen for a full year at no charge. The consent forms included a conditional promise of additional Ampligen after the double-blind study was completed:

If statistical analysis of the endpoints show that Ampligen TM shows efficacy compared to placebo, then following completion of all termination procedures, you understand that if you received placebo on study, you will be offered Ampligen TM and will reenter and follow the same protocol as if you had been randomized to receive Ampligen on study. If you received Ampligen TM on study, you understand that you will be offered continuation on Ampligen TM and will re-enter and follow the same protocol.

After the double-blind study, HEM applied for permission to proceed with what the FDA calls a "treatment IND." IND is an FDA acronym for "investigational new drug application." 21 C.F.R. § 312.3(b). The application, had it been granted, would have allowed the use of Ampligen "in the treatment of patients not in the clinical trials." 21 C.F.R. § 312.34(a). This procedure can be used for an experimental new drug if the disease is serious and there is no satisfactory alternative treatment. 21 C.F.R. § 312.34(b). The FDA rejected the application for safety reasons, but allowed the next phase in clinical trials, an open label study, where the doctor and patient know that the patient is receiving Ampligen.

FDA has completed its review of the treatment IND and has concluded that the data do not support expansion of ampligen treatment at this time. Therefore, the application has been placed on clinical hold until outstanding issues and concerns can be resolved.

In addition to numerous deficiencies in the application, the agency is concerned particularly about the serious and potentially life-threatening reactions that were observed during the study, and the conduct of the study in general. These serious reactions included acute hepatic (liver) toxicity, severe abdominal pain and irregular heartbeat.

In a letter to the company ... FDA Commissioner David A. Kasslar, M.D., said that HEM's study results involving 92 patients are preliminary and that data submitted thus far to FDA are incomplete and inadequate for FDA to assess safety and effectiveness in CFS. He emphasized that it was important for CFS sufferers to know that the significant side effects associated with Ampligen would have to be weighed against any claimed benefit before the drug could be approved for wide-spread use. FDA will continue to work closely with the manufacturer in an effort to resolve all issues.

. . . . .

An "open label" study of Ampligen in patients with CFS will be allowed to continue.

Thus, the FDA prohibited use of Ampligen on patients not in the clinical trials, but expressly allowed it for those who were.

II. Jurisdiction.

We have jurisdiction over this interlocutory appeal of a preliminary injunction under 28 U.S.C. § 1292 (1988).

We raised the issue of mootness sua sponte and obtained supplemental briefs. It appeared to us after argument that the one-year supply ordered by the preliminary injunction would have run its course before the case was decided. We would have no jurisdiction to review the preliminary injunction if it had been fully performed, and could have no further effect. Northwest Envt'l. Defense Ctr. v. Gordon, 849 F.2d 1241, 1244 (9th Cir.1988). But the case is not moot after all.

First, the injunction still may affect conduct by HEM for some of the patients. Graham cross appeals, claiming that the district court erred in putting a one year limit on the preliminary injunction. A second patient, Hyneman, has not yet begun treatment, so if he claims treatment, he is entitled to a year of Ampligen under the injunction. A third, Rogers-Dickie, began treatment considerably after it was ordered, and so has time left under the preliminary injunction.

Also, the validity of the injunction affects HEM, even for the patients who have received their year of Ampligen. HEM has been held in contempt and fined for violating the preliminary injunction. The civil contempt order entered against HEM depends on the validity of the preliminary injunction. In re Merrill Lynch Relocation Management, Inc., 812 F.2d 1116, 1119 (9th Cir.1987) ("If the underlying order ... is in error, the contempt order must fall."). The contempt order is not before us, but our decision on the validity of the preliminary injunction will determine whether it stands. We have been faced with this situation before, in Trans International Airlines v. International Brotherhood of Teamsters, 650 F.2d 949, 955 (9th Cir.1980). In that case we reviewed a preliminary injunction, even though the underlying dispute was settled and no further conduct was governed by the injunction, because the civil contempt order issued pursuant to that injunction would fall if we found the injunction to be in error. Id. Thus, because of the effect of this appeal on the civil contempt order, there is a live dispute.

III. HEM's appeal.

HEM makes two arguments: first, that the judicial process should have been suspended pending completion of the administrative process for reviewing the safety and efficacy of Ampligen by the FDA; second, that it had no contractual obligation to supply Ampligen at no charge for a year. The district judge was required to balance the probability of success on the merits with the possibility of irreparable injury. Associated General Contractors of Cal. v. Coalition for Economic Equity, 950 F.2d 1401, 1410 (9th Cir.1991). The grant or denial of a preliminary injunction will be reversed only where the district court abused its discretion or based its decision on an erroneous legal standard or on clearly erroneous findings of fact. Senate of Cal. v. Mossbacher, 968 F.2d 974, 975 (9th Cir.1992). Also, this injunction required affirmative conduct by Ampligen--it must provide Ampligen and inject it into the veins of the petitioners who want it. Such "mandatory preliminary relief" is subject to heightened scrutiny and should not be issued unless the facts and law clearly favor the moving party. Anderson v. United States, 612 F.2d 1112, 1114 (9th Cir.1980).

A. Primary Jurisdiction.

HEM argues that the doctrine of primary jurisdiction, which seeks to reserve to administrative agencies those questions that are peculiarly within their domain, should have prevented the district court from issuing the injunction. This doctrine applies when "protection of the integrity of a regulatory scheme dictates preliminary resort to the agency which administers the scheme." United States v. General Dynamics Corp., 828 F.2d 1356, 1362 (9th Cir.1987) (quoting United States v. Philadelphia Nat'l Bank, 374 U.S. 321, 353, 83 S. Ct. 1715, 1736, 10 L. Ed. 2d 915 (1963)). Four factors are relevant to determine whether this doctrine is applicable: (1) the need to resolve an issue that (2) has been placed by Congress within the jurisdiction of an administrative body having regulatory authority (3) pursuant to a statute that subjects an industry or activity to a comprehensive regulatory authority that (4) requires expertise or uniformity in administration. General Dynamics, 828 F.2d at 1362.

HEM claims that the District Court was faced with questions that Congress has entrusted to the FDA: effectiveness and safety of Ampligen, and whether the clinical hold was violated by the district court's order. HEM correctly argues that the agreement for a year's treatment at no charge in the consent form was conditioned on statistical analysis showing "efficacy compared to placebo." But HEM does not show that the district court took this decision away from the FDA. The FDA has not made a finding one way or the other, so far as the record shows. HEM's application for approval of the drug must imply that HEM takes the position that Ampligen does more to relieve chronic fatigue syndrome than saline solution. See 21 C.F.R. § 312.34(b)(4); 21 U.S.C. § 355(b)(1)(A). HEM has offered no evidence against the proposition that "statistical evidence of the endpoints show[s] that Ampligen shows efficacy compared to placebo," the wording of its contractual condition.

HEM argues that 21 U.S.C. § 355(d) leaves the decision of effectiveness to the FDA, and so it does, on the basis of "substantial evidence." "Substantial evidence" is defined in that statutory section to include "clinical investigations," which would be augmented by the one year of Ampligen injections on petitioners. Supervised treatment of petitioners with Ampligen under "the same protocol," as the agreement provides, will not damage the open label phase of testing for safety and efficacy--it will be part of it. HEM has offered no argument or evidence to suggest that adding a few patients to the open label phase of its study would impair the value of the study or conflict with the study protocol as approved by the FDA. It would be difficult to make practical sense of the contract if it were interpreted to mean, as HEM's position implies, that patients should get Ampligen during the double blind test, then receive none of it until the FDA approves the drug for treatment, and then go back on Ampligen for another year. The FDA determination of efficacy did not have to precede the one year of Ampligen at no charge for patients who subjected themselves to the double-blind study.

HEM also argues that the preliminary injunction erroneously takes away the decision on safety from the FDA and gives it to the district court. The language in the FDA's "Talk Paper" certainly raises safety concerns which a district court could not properly ignore. Public safety should be considered by a court when granting equitable relief. Chalk v. United States Dist. Court for the Cent. Dist. of Cal., 840 F.2d 701, 711 (9th Cir.1988). The issue here differs from the usual public safety concern--though many people would object to having an unproved and potentially life threatening drug foisted on them, the injunction in this case would provide the drug only to people who insist on it, with their eyes open to its experimental status.

Had the FDA decided that Ampligen should not be administered to human beings at all because of safety concerns, a court might well be in error if it ordered injection of Ampligen anyway because HEM had promised to do so. But that would be a distinguishable case. Here, the FDA has decided that Ampligen can be injected into humans in an open label study. This amounts to an implicit finding of sufficient safety for use in the open label study. The FDA expressly provided that the open label study could proceed, instead of applying a general "clinical hold" so that no patient could be given the drug. 21 C.F.R. § 312.42(a). The term "open label" distinguishes a study where the doctor and patient know that the medication is being administered, from a double-blind study where neither knows who is receiving a placebo. The double-blind study can establish that the drug is more effective than a placebo (since some patients with some conditions will improve without treatment, or with the psychological boost provided by a placebo). The open label study can examine the safety, efficacy, side effects, dosage, and other characteristics of the medication.

HEM's third argument is that the preliminary injunction flies in the face of the FDA's "clinical hold" on the "treatment IND." Once this jargon is translated into English, the argument loses its force. As explained above, the FDA "clinical hold" on the "treatment IND" just means that the drug is not approved for use on patients "not in the clinical trials." 21 C.F.R. §§ 312.34(a), 312.42. Because the FDA refrained from imposing a clinical hold on the open label study of Ampligen, the preliminary injunction did not contravene an FDA determination.

B. Contract obligation.

HEM argues that as a matter of contract law, petitioners' probability of success on the merits was low, because its promise was not supported by consideration. This argument is without merit. The patients submitted themselves to months of periodic injections with an experimental drug, or unbeknownst to them, mere saline solution, combined with intrusive and necessarily uncomfortable testing to determine their condition as the tests proceeded. HEM sought to have them participate in its study so that it could obtain FDA approval for its new drug.

HEM argues that because petitioners participated voluntarily and were free to withdraw, they had no binding obligation and so gave no consideration. Somehow the category of unilateral contracts appears to have escaped HEM's notice. The deal was, "if you submit to our experiment, we will give you a year's supply of Ampligen at no charge." This form of agreement resembles that in the case taught in the first year of law school, Hamer v. Sidway, 124 N.Y. 538, 27 N.E. 256 (1891). There, an uncle promised his nephew that if he would refrain from drinking, using tobacco, or playing cards and billiards until age 21, he would receive $5000. The court held that consideration had been given because the nephew had refrained from the prohibited activities during the requisite period on the faith of his uncle's promise. He had accepted the offer by completing performance. See also Restatement (Second) of Contracts § 50(2) (acceptance by performance); Mohr Park Manor, Inc. v. Mohr, 83 Nev. 107, 424 P.2d 101, 104 (1967) (recognizing unilateral contracts).2

In this case, the petitioners performed by submitting to the double-blind tests. They incurred the detriment of being tested upon for HEM's studies in exchange for the promise of a year's treatment of Ampligen. Upon completion of the double-blind tests, there was a binding contract.

Dahl seeks sanctions against HEM under Fed.R.App.P. 38 on the ground that this appeal is frivolous. While the contract argument approaches frivolousness, the primary jurisdiction argument is serious. The FDA's role in the drug approval process and the FDA's concerns about the safety of Ampligen justified HEM in pursuing the appeal. HEM's arguments were not "wholly without merit," so no sanction will be imposed. McDougal v. County of Imperial, 942 F.2d 668, 680 (9th Cir.1991).

IV. Graham's appeal.

One of the patients, Graham, cross-appeals. She argues that the district court erred in limiting the preliminary injunction to one year, because HEM had promised her free Ampligen for life if she submitted to the double-blind study. The district judge listened to testimony and studied depositions relating to this dispute, and then made careful and precise findings of fact. He found that the promise of free Ampligen for life was in fact not made, and did not believe Graham's claim to the contrary. The finding of fact is reviewed for clear error. Sega Enterprises Ltd. v. Accolade, 977 F.2d 1510, 1517 (9th Cir.1992). There was none.

AFFIRMED.