This case is brought under the Americans With Disabilities Act of 1990. 42 U.S.C. Secs. 12101, et seq. (the ADA). Appellant Karen E. Henderson sought a preliminary injunction against her health plan and insurance providers, maintaining that her health plan discriminated against her when it denied her physician's request for certain treatment on the ground that as to breast cancer, the treatment was experimental. The district court denied the requested relief. On expedited appeal, we entered an order (Hansen, J., dissenting) on June 16, 1995, reversing the denial of injunctive relief and remanding the case to the district court with directions to enter a preliminary injunction requiring Bodine Aluminum, Inc. and Comprehensive Health Benefits Program for Employees of Bodine Aluminum, Inc. to provide the necessary assurance to Barnes Hospital that the costs of the sought-after treatment would be paid for. This opinion, which Judge Hansen now joins, sets forth the reasons for that order.

* Henderson was diagnosed with an aggressive form of breast cancer in April 1995. Her oncologist, Dr. Gary A. Ratkin, recommended entry into a clinical trial program that randomly assigns half of its participants to a regimen of high dose chemotherapy (HDCT). This regimen is substantially more expensive than standard chemotherapy because it kills bone marrow cells and thus requires an autologous bone marrow transplant (removal of some of patient's bone marrow, and replacement after the last phases of the chemotherapy).

Because of the possibility that Henderson might participate in the more expensive marrow transplant branch of the program recommended by Dr. Ratkin, Henderson's health plan refused to pre-certify the placement. The health plan pointed to policy provisions covering HDCT only for certain types of cancer, not including breast cancer.

Henderson, who is covered as a dependent under the health plan of Bodine Aluminum, Inc., her husband's employer, brought this action under the ADA against Bodine Aluminum and its health plan (collectively, "Bodine"), claiming discrimination based on her cancer type and seeking a preliminary injunction requiring Bodine to assure payment for any possible bone marrow therapy.

Henderson filed this action on June 13, 1995. Because Henderson needed to enter the program prescribed by Dr. Ratkin no later than June 16, she sought expedited review both in the district court and here. After an evidentiary hearing on June 13, the district court filed an order denying injunctive relief on June 15. The following day we entered the order described above.

The district court, performing the preliminary injunction analysis required by Dataphase Systems, Inc. v. C.L. Systems, Inc., 640 F.2d 109 (8th Cir.1981), found that Henderson did not supply sufficient evidence to show a likelihood of prevailing on her ADA claim. The district court held that because the plan explicitly excludes most cancers from HDCT coverage, breast cancer was not improperly singled out for exclusion. (The district court acknowledged that if a federal statute like the ADA is violated, the Employee Retirement Income Security Act (ERISA) provisions allowing ERISA-covered health plans great latitude in excluding coverage for various treatments are preempted. Le v. Applied Biosystems, 886 F.Supp. 717 (N.D.Cal.1995).)

Henderson argues that HDCT for her kind of aggressive breast cancer has been proven more effective than standard chemotherapy and that it has passed beyond the experimental phase to a stage at which it can justly be called an accepted treatment. Since Bodine does cover HDCT for cancers for which it is an accepted treatment, denying HDCT treatment for breast cancer is discrimination based on disability type. Such discrimination is forbidden by the ADA, concludes Henderson.

We agree that Henderson's argument has a sufficient likelihood of success on the merits. Therefore, if the evidence shows that a given treatment is non-experimental--that is, if it is widespread, safe, and a significant improvement on traditional therapies--and the plan provides the treatment for other conditions directly comparable to the one at issue, the denial of that treatment arguably violates the ADA. See EEOC's Interim Policy Guidance on ADA and Health Insurance, (June 8, 1993), reprinted in BNA's Americans With Disabilities Act Manual, No. 18, at 70-1052 (if coverage disparities apply to dissimilar conditions, they are not disability-based distinctions) (construing ADA Sec. 501(c) (42 U.S.C. Sec. 12201(c))). Thus, the issue here is whether Henderson has shown a likelihood of eventually proving at trial that HDCT is an accepted treatment for breast cancer.

Henderson has documented a large and growing body of medical scholarship and testimony showing that HDCT is safe, increasingly widely used, and a significant improvement over standard chemotherapy. E.g., William P. Peters, High Dose Chemotherapy with Autologous Bone Marrow Transplantation for the Treatment of Breast Cancer: Yes, in IMPORTANT ADVANCES IN ONCOLOGY 1995, at 215 (Vincent T. DeVita, et al., eds. 1995). Cases dealing with the specific factual issue whether HDCT is an experimental treatment of the type usually excluded by ERISA-type health insurance plans are divided. Compare White v. Caterpillar, Inc., 765 F.Supp. 1418, 1421-23 (W.D.Mo.) (HDCT is an accepted therapy for breast cancer), aff'd, 985 F.2d 564 (8th Cir.1991), and Adams v. Blue Cross/Blue Shield of Maryland, Inc., 757 F.Supp. 661, 663 (D.Md.1991) (same), with Harris v. Mutual of Omaha Cos., 992 F.2d 706 (7th Cir.1993) (Office of Personnel Management ruling that HDCT is experimental for breast cancer not arbitrary or capricious); see Harris v. Blue Cross Blue Shield of Missouri, 995 F.2d 877, 879-80 (8th Cir.1993) (declining on mootness grounds to address district court finding that HDCT was an accepted treatment); Clark v. K-Mart Corp., 979 F.2d 965 (3rd Cir.1992) (en banc) (same). It is clear, then, that Henderson's basic contention that HDCT is an accepted treatment has a realistic chance of success, as a similar contention recently succeeded in our circuit. White, 765 F.Supp. at 1421 (finding a contrary conclusion arbitrary and capricious); see also Harris, 995 F.2d at 879-80.

With this body of medical commentary and litigation in mind, we find compelling the fact that Bodine has brought forward no evidence refuting Henderson's factual assertions. We note that Henderson was diagnosed in April and that Bodine must have known of Henderson's position before this action was filed, since Henderson's caregivers unsuccessfully applied for pre-certification. We do not believe it is unfair to expect Bodine and its sophisticated health insurance providers to promptly provide some general evidence that HDCT is not an accepted therapy for breast cancers like Henderson's. After all, such coverage issues lie at the heart of a health insurance provider's expertise, and the evidence and cases cited show that the issue of HDCT for breast cancer has been on the insurance industry's horizon for some years.

Although on the current state of the record Henderson has by any standard shown a likelihood of success on the merits, we note also that preliminary injunctions become easier to obtain as the plaintiff faces progressively graver harm. 11A CHARLES A. WRIGHT, ARTHUR R. MILLER AND MARY K. KANE, FEDERAL PRACTICE AND PROCEDURE, Sec. 2948.3 at 189-97 (1995). It is hard to imagine a greater harm than losing a chance for potentially life-saving medical treatment.

As it is undisputed that the plan covers HDCT for other types of cancer for which it is an accepted treatment, and as she has shown considerable unrebutted evidence that HDCT is an accepted breast cancer treatment with significantly higher success rates than standard chemotherapy, Henderson has shown a likelihood of success on the merits warranting injunctive relief.

The district court also held that Henderson did not show that she would suffer irreparable harm if injunctive relief were denied. The district court reasoned that Henderson had not showed that she could not secure HDCT treatment (or an equally effective alternative) from another source. We conclude, however, that requiring Henderson to make such a showing is too speculative. Although it is true there is no evidence in the record regarding different sources from which Henderson could seek HDCT treatment, or as to alternate treatments which may be as effective as HDCT, there is evidence that Henderson had to begin treatment on June 16 or be disqualified from a program that offered HDCT as a possibility; as discussed above, there is also evidence that HDCT is more effective than standard chemotherapy. Henderson faced missing perhaps her only opportunity for HDCT if injunctive relief were not granted; given the grave and progressive nature of Henderson's illness, this constitutes irreparable harm.

The district court also held that Henderson's bid for injunctive relief was improper because she asked only for monetary relief. It is usually impossible to get injunctive monetary relief because cash is the classic remedy that can await trial. Classic Components Supply, Inc. v. Mitsubishi Elec. America, Inc., 841 F.2d 163, 164-65 (7th Cir.1988) (if money will fully compensate claimant, harm to claimant is not irreparable) (citing Sampson v. Murray, 415 U.S. 61, 88-91, 94 S.Ct. 937, 951-53, 39 L.Ed.2d 166 (1974)). But this principle does not apply here. Henderson seeks no money for herself, but is rather seeking specific performance under her health insurance policy; Henderson in effect asks for a declaration that the plan requires Bodine to guarantee payment to her caregivers in the event they decide to administer HDCT. At bottom, Henderson seeks urgent medical treatment, not money. Urgent medical treatment is the kind of equitable relief that cannot abide trial.

The district court also held that Henderson violated a contractual duty under the plan to wait 60 days after a precertification denial before suing. In cases of rapid, life-threatening illnesses, patients seeking treatment are not required to exhaust contractual or administrative procedures before coming to federal court.

Appellees Healthy Alliance Life Insurance Company and Alliance Blue Cross and Blue Shield argue that because they do not employ Henderson's husband but merely provide insurance services to the Bodine Aluminum, Inc. and its health benefits program, they are not responsible for giving the assurance of payment that Henderson seek. We express no opinion on this issue, leaving it instead to be decided in the first instance by the district court.

It is for these reasons, then, that we reversed the order denying a preliminary injunction and remanded for entry of preliminary injunctive relief and for trial.

Finally, we note the general rule that "the findings of fact and conclusions of law made by a court granting a preliminary injunction are not binding at trial on the merits." University of Texas v. Camenisch, 451 U.S. 390, 395, 101 S.Ct. 1830, 1834, 68 L.Ed.2d 175 (1981). This general rule applies in this case so that none of the substantive issues discussed in this opinion are foreclosed from being litigated on remand.