Adams v. Blue Cross/Blue Shield of Maryland, Inc., 757 F. Supp. 661 (D. Md. 1991)

*662 Richard Carter, Douglas M. Coleman, Hudgins, Carter & Coleman, Alexandria, Va., for plaintiffs.

Peter H. Gunst, Robert B. Levin, Frank, Bernstein, Conaway & Goldman, Baltimore, Md., for defendant.

GARBIS, District Judge.

In 1990, Plaintiffs Alexandra Adams and Kelly Whittington were diagnosed as having advanced breast cancer, a potentially fatal disease. Both women were advised by their treating physicians that High Dose Chemotherapy with Autologous Bone Marrow Transplant (hereinafter "HDCTABMT") would be the best available care for them. Because the treatment is very expensive, costing approximately $100,000.00, Plaintiffs requested their insurance carrier Blue Cross-Blue Shield of Maryland (hereinafter "Blue Cross") to confirm in advance that it would pay for the treatment. However, Blue Cross denied coverage for both, relying upon a policy provision which excluded coverage for "experimental and investigative" treatments. While acknowledging that the policy would cover the use of HDCT-ABMT for several other diseases,^[1] Blue Cross took the position that the treatment was "experimental" when used to treat breast cancer, and that Plaintiffs were not eligible for coverage. This Court is asked to decide whether under the benefit plan Blue Cross must pay for Plaintiffs' treatment.

Procedurally, Plaintiffs seek to obtain a declaratory judgment compelling Blue Cross-Blue Shield of Maryland to pay for their treatment. Because the health insurance plans under which Plaintiffs are covered *663 are "employee welfare benefit plans," this case is governed by the Employee Retirement Income Security Act of 1974 (ERISA), 29 U.S.C. § 1001 et seq. Accordingly, federal jurisdiction for this case is conferred by 28 U.S.C. § 1332.

In view of the high degree of public interest in this case, the Court believes it appropriate, at the price of some repetition, to provide a summary of its decision, followed by a detailed discussion of the issues.

Essentially, Blue Cross contends that it can deny coverage for the Plaintiffs' HDCT-ABMT treatment because a policy provision in the Blue Cross benefit plan states that "the plan will not pay for services ... that are experimental or investigative in nature." The plan provision specifically defines the term "experimental or investigative" to mean "any treatment ... not generally acknowledged as accepted medical practice by the suitable medical specialty practicing in Maryland, as decided by us." Blue Cross argues that HDCT-ABMT is experimental under that definition and, accordingly, is not covered by the plan.

As discussed at length below, this Court has concluded that High Dose Chemotherapy with Autologous Bone Marrow Transplant is not an "experimental" treatment for Plaintiffs' breast cancer under the definition of "experimental" prescribed in the Blue Cross plan. This Court finds that at the time Blue Cross decided to deny coverage to Plaintiffs, the suitable medical specialty practicing in Maryland, i.e., Maryland oncologists, generally acknowledged HDCT-ABMT as accepted medical practice. The Court heard convincing trial testimony that in April and July of 1990, the dates on which the Blue Cross decisions were made for Alexandra Adams and Kelly Whittington, respectively, Maryland oncologists considered HDCT-ABMT to be accepted medical practice. The Court also heard testimony that Maryland oncologists regularly refer breast cancer patients to institutions which administer the treatment, such as Duke University Medical Center, Johns Hopkins Medical Center, and Georgetown University Medical Center. Experts testified that the treatment was at the relevant time being offered at many major medical centers across the country, and the Court finds this, too, to be persuasive evidence that the treatment was generally acknowledged as accepted medical practice.

Disregarding the specific plan language, Blue Cross decided to deny Plaintiffs coverage based on its own independent evaluation of published scientific research results, completely ignoring the consensus of opinion of members of the Maryland oncological community. As is discussed at length below, Blue Cross's decision was both incorrect and unreasonable. The benefit plan does not allow Blue Cross to form its own independent evaluation of the treatment from scientific data, selectively dismissing the opinion of the very decision makers to which the plan requires Blue Cross to defer, i.e., medical oncologists practicing in Maryland.

Blue Cross also argued that the treatments were experimental based on the fact that the treatments at issue were to be given on research protocol at teaching hospitals. But the fact that treatment is provided at teaching centers "on protocol" does not alter the fact that at the relevant times Maryland oncologists generally acknowledged the treatment to be accepted medical practice. Of course, researchers maintain an interest in collecting further information about HDCT-ABMT. However, physicians refer their patients for HDCT-ABMT for the primary purpose of medical treatment, and under the plan definition it is of little consequence that the treatment also provides information to research investigators.

Indeed, physicians deliver many of today's accepted medical treatments at major teaching hospitals, whose practice it is to collect data on the patients they treat. The author Rose Kushner, a breast cancer victim who was for many years a source of comfort, information and inspiration to women with breast cancer, observed prior to her own death from the disease in 1990 that it is of no great moment that treatment *664 given at teaching hospitals also provides the opportunity for data collection.

It is a personal decision whether or not to go to a research hospital. [Such institutions] do make certain demands on patients that non-research hospitals do not make. Physicians, nurses, psychologists, and social workers often interview and question patients. Members of the medical staff may want to examine their wounds ... As for being part of a "random sample," I may have been, for all I know. My operation may have been done using a new type of scalpel, while a sample of other women were cut by an older instrument. I honestly don't know. But what difference does it make?

Kushner, "Why Me?" at 25 (1977).

As discussed below, whether HDCT-ABMT treatment is experimental as defined by the Blue Cross plan does not turn on a determination by Blue Cross based upon its own evaluation of the scientific data. Nor does the question turn on whether the treatment is provided on an experimental protocol. Rather, it turns only upon whether at the relevant times medical oncologists practicing in Maryland generally acknowledge the treatment as accepted medical practice. This Court finds that they did at the pertinent time and that Blue Cross improperly denied insurance coverage to Mrs. Adams and Mrs. Whittington.

Breast cancer is frequently treated with chemotherapy drugs which are designed to kill cancer cells. Administered in low doses, chemotherapy can be given to patients on an outpatient basis and is usually administered in several cycles of continuing treatment. However, the effectiveness of low-dose chemotherapy for more advanced forms of breast cancer is limited because the dosage frequently is not toxic enough to kill the more aggressive cancer cells.

Currently, doctors use a variety of dosages and drug mixtures, or "cocktails", in administering low-dose therapy (sometimes called "conventional chemotherapy"). To date, no standard dosages or preparative regimens exist for low-dose chemotherapy. Indeed, ongoing research continues to evaluate the effectiveness of different dosages and preparative regimens.

High Dose Chemotherapy with Autologous Bone Marrow Transplant is a procedure in which physicians administer relatively high doses of chemotherapy drugs in conjunction with a bone marrow "transplant." In addition to destroying cancer cells, high doses of chemotherapy also destroy the patient's bone marrow and blood cells, both of which play an essential role in immunity to disease and infection. Stripped of her bone marrow and certain blood cells, the patient is open to opportunistic infection and disease following treatment with high-dose chemotherapy. To counteract this potentially lethal side effect, a portion of the patient's bone marrow is harvested from the patient prior to her treatment with toxic chemotherapy drugs. After chemotherapy treatment is completed, the healthy marrow is then reinfused to "rescue" the patient. The marrow quickly multiplies to replace the marrow destroyed during high-dose chemotherapy.

Both Plaintiffs participated in a Blue Cross group health plan issued by their respective spouse's employer. The plan contains an exclusion provision which provides, in pertinent part, that "the plan will not pay for services and supplies ...

(3) that are experimental or investigative in nature or for the Covered Services related to them. The terms Experimental and Investigative mean the use of any treatment, procedure, facility, equipment, drug, device, or supply not generally acknowledged as accepted medical practice by the suitable medical specialty practicing in Maryland, as decided by us."

Plaintiff Alexandra Adams is a thirty-four year old, premenopausal mother of two children who was diagnosed in 1990 as *665 having Stage II^[2] or Stage III breast cancer. At the time Mrs. Adams' cancer was detected, it had spread to 18 of the 27 lymph nodes located in the region surrounding the breast. Her cancer was also described as "estrogen receptor negative," which means that the cancer cells do not have special receptors on their surface which would enable physicians to use hormone therapy in an attempt to destroy the cells. Mrs. Adams underwent surgery to remove the breast but, because of her age, the high number of lymph nodes involved, and the fact that her cancer was estrogen receptor negative, Mrs. Adams was thought to be at high risk for recurrence.

Both Mrs. Adams' personal physician, Dr. Enser Cole, and an outside consulting physician, Dr. Nancy Davidson of Johns Hopkins, recommended that she consider HDCT-ABMT. Dr. Cole suggested that she obtain treatment at Duke University Medical Center.

Prior to undergoing treatment at Duke, Mrs. Adams requested pre-authorization of insurance coverage for the procedure from Blue Cross. On April 12, 1990, Dr. William Peters, Mrs. Adams' treating physician at Duke University, wrote to Blue Cross on Mrs. Adams' behalf to request coverage. Dr. Peters' letter included articles, abstracts of studies conducted on HDCT-ABMT, and survival curves demonstrating the results of those studies.

Prior to receiving Mrs. Adams' request for preauthorization of insurance coverage, Blue Cross' corporate medical director, Dr. Arthur Keefe, had already formulated plan guidelines concluding that HDCT-ABMT was "experimental." Dr. Keefe based his initial determination almost exclusively on a 1988 technology evaluation of the procedure conducted by the Blue Cross National Association. In reviewing the technology evaluation, Dr. Keefe made the policy determination that the Blue Cross plan did not cover HDCT-ABMT when used to treat breast cancer, and subsequently formulated policy guidelines for the Blue Cross-Blue Shield of Maryland benefit plan to that effect.

Pursuant to those guidelines, the Blue Cross claims department issued a rejection letter to Mrs. Adams dated April 30, 1990. Soon after, on June 14, 1990, Dr. Peters sent an appeal letter on Mrs. Adams' behalf to Blue Cross, including another survival curve and the names of additional experts for Dr. Keefe to consult for further opinion. Despite a cursory review of the material in both of Dr. Peters' letters, Dr. Keefe never contacted any of the experts named in the letters. In addition, Dr. Cole and Dr. Davidson, both of whom had recommended the treatment, wrote in support of Mrs. Adams' appeal, citing opinions by members of the Maryland oncological community that the procedure was accepted medical practice. Neither Dr. Keefe nor anyone else from Blue Cross ever formally responded to Mrs. Adams' appeal, either orally or in writing.

Plaintiff Kelly Whittington is a twenty-nine year old, premenopausal mother of two, who was diagnosed in 1990 as having Stage IV breast cancer. As with Mrs. Adams, her cancer is also described as "estrogen receptor negative." Because Mrs. Whittington's disease had progressed to such an advanced stage, and because of her age and estrogen receptor status, she was at very high risk for recurrence despite surgery to remove the cancerous breast.

Mrs. Whittington's oncologist, Dr. Frederic Kass of Maryland, referred Mrs. Whittington to the Vincent Lombardi Cancer Center at Georgetown University Medical Center in Washington, D.C. Physicians at Lombardi recommended that Mrs. Whittington undergo HDCT-ABMT.

In July of 1990, Dr. Thomas Spitzer, Mrs. Whittington's treating physician at Georgetown, *666 contacted Blue Cross on Mrs. Whittington's behalf, and was told that she would not be covered for the procedure. On July 26, Dr. Spitzer wrote a letter of appeal to Dr. Keefe and Blue Cross, to which neither Dr. Keefe nor anyone from Blue Cross ever responded. On August 22, 1990, Mrs. Whittington's attorney wrote to Dr. Keefe, enclosing the phone numbers of nine Maryland oncologists with the suggestion that Dr. Keefe consult them. He did not. Mrs. Whittington never received any formal notification from Blue Cross that her appeal had been denied.

To evaluate the validity of Plaintiffs' claim, the Court must interpret and apply the relevant terms of the Blue Cross insurance contract. Before undertaking an analysis of the contract, it is important to set forth the appropriate standard of review under which to evaluate Blue Cross' interpretation and application of its contractual agreement.

In Firestone Tire and Rubber Co. v. Bruch, 489 U.S. 101, 109 S. Ct. 948, 956, 103 L. Ed. 2d 80 (1989), the Supreme Court held that "a denial of benefits challenged under [ERISA] is to be reviewed under a de novo standard unless the benefit plan gives the administrator or fiduciary discretionary authority to determine eligibility for benefits or to construe the terms of the plan." DeNobel v. Vitro Corp., 885 F.2d 1180, 1186 (4th Cir. 1989) (citing Bruch, 109 S.Ct. at 956-957). The Court further held that if the benefit plan in fact grants power to the trustee to construe disputed or doubtful terms, "the trustee's interpretation will not be disturbed if reasonable." Id. (citing Bruch, 109 S.Ct. at 954). Later courts have interpreted the Bruch deferential standard of "abuse of discretion" to be the equivalent of the pre-Bruch "arbitrary and capricious" standard. See DeNobel v. Vitro, 885 F.2d 1180.

The Court must first determine whether the benefit plan at issue in fact grants the plan administrator [Blue Cross], as trustee, discretionary authority to determine eligibility for benefits or to construe disputed or doubtful terms. The Bruch Court did not identify any specific language which must appear in a benefit plan for the plan to grant such discretionary authority. However, subsequent decisions have held that the benefit plan language must demonstrate on its face a clear and unequivocal intent to vest the plan administrator with discretionary authority. See e.g., DeNobel, 885 F.2d at 1187; Graham v. Federal Express Corp., 725 F. Supp. 429, 434 (W.D. Ark.1989).

In cases in which courts have found specific plan language to have granted discretionary power, the plan language has clearly and unequivocally vested a very broad discretionary power in the plan administrator. For example, in DeNobel v. Vitro, 885 F.2d 1180 (4th Cir.1989), the benefit plan vested an Administrative Committee with the power "to determine all benefits and resolve all questions pertaining to the administration, interpretation and application of Plan provisions, either by rules of general applicability or by particular decisions ..." Id. at 1186. Other courts have looked to similarly precise language conferring broad authority on plan administrators. See Guy v. Southeastern Iron Workers' Welfare Fund, 877 F.2d 37, 39 (11th Cir.1989) (arbitrary and capricious standard where plan conferred "upon the trustees `full and exclusive authority to determine all questions of coverage and eligibility' and `full power to construe the provisions of [the] Trust'"); Lowry v. Bankers Life & Casualty Retirement Plan, 871 F.2d 522, 524-25 (5th Cir.1989) (instrument granted "permissive authority to the [administrators] to `interpret and construe' the [plan] and the power `to determine all questions of eligibility and status'"); Retirement and Sec. Program for Employees of Nat'l Rural Elec. Coop. Ass'n v. Oglethorpe Power Corp. Retirement Income Plan, 712 F. Supp. 223, 226 (D.D.C.1989) (trustees vested with "authority to determine all questions arising in connection with the [plan], including its interpretation").

The relevant plan language in the Blue Cross contract excludes as "experimental" those procedures or treatments which are *667 "not generally acknowledged as accepted medical practice by the suitable practicing specialty in Maryland, as decided by us." (emphasis added). The Court finds this language vague and ambiguous and, in any event, certainly not clear and unequivocal. The Court cannot find from such language that those who drafted the Blue Cross plan intended to secure for Blue Cross the power to determine eligibility benefits or to construe ambiguous or disputed terms, much less the power to change definitions set forth specifically in the plan. The language "as decided by us" is far more vague and uncertain in its effect than the broad and clear language, cited above, which courts have held to confer discretionary authority to plan administrators. Such language does not manifest the requisite clear and unequivocal intent to confer discretionary power to the fiduciary beyond an initial ability to grant or deny benefits based on the terms of the plan. See Overcash v. Blue Cross-Blue Shield, 94 N.C. App. 602, 381 S.E.2d 330, 334 (1989) (plan language which provided benefits only for medically necessary, reasonable and customary charges "as determined by [defendant]" did not grant Blue Cross discretionary authority to determine eligibility for benefits or interpret contract terms). To embroider such intent onto the language would undermine the purposes of ERISA.

"As the Court pointed out in Bruch, ERISA abounds with trust language and terminology. The purposes of ERISA are to promote and protect the rights of employees and their beneficiaries in employer-employee benefit plans. Bruch, [109 S.Ct.] at 955. Absent a clear grant of discretion to the administrator, application of the highly deferential arbitrary and capricious standard of review does not promote these goals." Brown v. AMPCO-Pittsburgh Corp., 876 F.2d 546, 550 (6th Cir.1989).

Thus, this Court finds that the contractual language does not entitle Blue Cross to judicial deference with respect to either resolving disputes over eligibility determinations or the interpretation or definition of ambiguous terms.

Even if the Court were somehow persuaded that the phrase "as decided by us" granted Blue Cross some discretionary authority, at best such imprecise language could only be said to confer authority to resolve disputes over eligibility, but could in no way be interpreted to grant the authority to "construe doubtful or disputed terms." "Language requiring trustees to make a final determination of an employee's eligibility under the plan does not necessarily confer discretionary authority to render decisions with regard to ambiguous provisions of the plan." Baxter v. Lynn, 886 F.2d 182, 188 (8th Cir.1989). Accord, Dzinglski v. Weirton Steel Corp., 875 F.2d 1075, 10769 (4th Cir.1989) (final authority to determine all matters of eligibility merely describes trustees mandatory role in accepting or rejecting claims submitted to the Fund.)

Certainly the "as decided by us" language does not grant, as Blue Cross argues, the authority to define the meaning of the phrase "experimental" or "investigative." Indeed, the exclusion provision itself sets forth the definition of those terms. It cannot be within Blue Cross' power to deviate from that contractual definition. Nor does the plan language grant discretionary authority to Blue Cross in defining any of the disputed terms within the contractual definition. Accordingly, the Court undertakes to review the Blue Cross decision and plan interpretation under a de novo standard of review.

The pertinent contract plainly sets forth the definition for "experimental and investigative"an experimental procedure is a procedure "not generally acknowledged as accepted medical practice by the suitable practicing medical specialty in Maryland."

The Court finds that the term "generally acknowledged" is not ambiguous. The term refers to a consensus of acknowledgement by practicing Maryland oncologists.

The parties debate the meaning of the term "accepted medical practice." Blue Cross argues that "accepted medical *668 practice" is a "standard practice" which has (1) proven itself through a rigorous process of clinical testing and amassing of scientific evidence, (2) has known risks and benefits, and (3) is a practice not in the process of being tested to gather generalizable knowledge.^[3]

In an attempt to provide legal support for its proposed definition, Blue Cross wanders far afield from cases dealing with the acceptability of medical treatment and relies instead upon the venerable, but irrelevant, decision of the United States Court of Appeals for the District of Columbia Circuit in Frye v. United States, 293 F. 1013 (D.C.Cir.1923). In that case, the court established the criteria for determining whether to admit expert testimony based on some scientific technique or treatment. Under the Frye standard, experts can only testify when their testimony relates to a technique that is "generally accepted by the relevant scientific community." In evaluating whether a technique is generally accepted by the scientific community, Frye looks to whether a consensus exists within a community of objective scientific observers, that is, experts or scientists whose livelihood is not intimately connected with the technique. See People v. Tait, 99 Mich.App. 19, 297 N.W.2d 853, 856 (1980) (interpreting Frye).

This Court rejects the use of the Frye standard in the context of this case to define "accepted medical practice". The Frye standard is focused upon the evidentiary use of a putatively reliable technique, e.g., the use of polygraphs, see U.S. v. Alexander, 526 F.2d 161 (8th Cir.1975), the use of hypnosis in memory restoration, see U.S. v. Valdez, 722 F.2d 1196 (5th Cir. 1984), or DNA profiling, see U.S. v. Two Bulls, 918 F.2d 56 (8th Cir.1990). The Frye standard is used to screen out testimony based on a controversial scientific techniques because such testimony threatens to redirect the focus of a trial away from its central issues, and toward the controversy of the technique.

"All scientific aids and devices go through an experimental and testing stage, and during these stages there may be considerable scientific controversy. During this period of controversy the danger is that a trial may actually result in a trial of the technique rather than a trial of the issues involved in the case."

Reed v. State, 283 Md. 374, 389, 391 A.2d 364 (1978) (citing State v. Cary, 56 N.J. 16, 264 A.2d 209 (1970)). However, the objective scientific evaluation of a technique for purposes of determining the admissibility of expert testimony in a trial is far different from the practical evaluation of a medical treatment to decide whether it is accepted medical practice. In the first case, scientists are asked to quantify a technique's reliability, predictability, and precision so that a jury can rely upon it to determine facts. In the second, physicians practicing the scientific art of medicine must strike a practical balance between the risks of a particular treatment, the effectiveness of the treatment, and the availability of other options. The Frye standard, valid though it may be for ruling upon the admissibility of evidence, is totally irrelevant to the issue before this Court.

Furthermore, the Court finds the Blue Cross definition of the term based on the Frye standard to be arbitrary and capricious because it is inconsistent with the language of the plan documents, see DeNobel v. Vitro, 885 F.2d at 1188 (citing Reiherzer v. Shannon, 581 F.2d 1266, 1273-73 (7th Cir.1978)) (interpretations which are inconsistent with plan language are unreasonable, and are therefore "arbitrary and capricious"). The exclusion provision in the Blue Cross plan looks specifically to a community of practitioners who use practical criteria in evaluating the acceptability of a treatment. In particular, the contract looks to the standard of practice adopted by local practicing medical oncologists and not to a community of objective scientific observers as is true of the Blue Cross definition for "accepted medical practice" based on the Frye standard. Thus, the *669 Blue Cross Frye definition^[4] renders meaningless the provision language "accepted by the suitable practicing medical specialty in Maryland." The Blue Cross interpretation is not only incorrect, but to the extent it directly contradicts explicit plan language, it is also arbitrary and capricious.

This Court interprets the term "accepted medical practice" consistently with the focus of the contractual language on the Maryland medical community standards. Plaintiffs' experts, as well as relevant Maryland case law, have defined "standard care" or "accepted medical practice" to be a practice utilizing the degree of skill and diligence employed by the ordinary prudent practitioner in his or her field and community. See Clark v. U.S., 402 F.2d 950, 951 (4th Cir.1968). In determining what is "accepted medical practice," a practice which conforms to "accepted medical standards," courts usually look to expert testimony regarding what practice is accepted in the community of practitioners. See, e.g., Thimatariga v. Chambers, 46 Md.App. 260, 262, 416 A.2d 1326, 1329 (1980); Hitch v. Hall, 42 Md.App. 260, 262, 399 A.2d 953, 955 (1979); Alimchandani v. Goings, 39 Md.App. 353, 360, 386 A.2d 789, 793 (1978).

Significantly, the concept of "acceptable medical practice" is located not only in the degree of approval among practitioners, but also in the customary practices among reasonable and prudent physiciansnot merely what practitioners judge to be accepted but what they use in treating their own patients. See King, In Search of the Standard of Care: "Accepted Practice Formula," 28 Vand.L.Rev. 1213, 1236 (1975). If the consensus of practitioners were to find it reasonable to employ a particular procedure or treatment, a court would be hard pressed to find that the treatment was not "accepted medical practice."

In summary, for purposes of interpreting the contract at issue, the Court adopts the standard legal definition for "accepted medical practice" used in evaluating the care rendered by a treating physician. Thus, the question before the Court becomes whether, at the relevant times, a consensus of Maryland oncologists considered HDCT-ABMT to be an appropriate treatment option offered by the ordinary prudent and reasonable medical oncologist exercising due care for his or her patient. As previously noted, this definition requires the Court to look to expert testimony regarding the judgment and customary practices of oncologists practicing in Maryland. Thus, the definition is internally consistent with the contractual language, in stark contrast to the Frye standard offered by Blue Cross.

Having defined the pertinent contract terms and determined the appropriate standard for judicial review, the Court must now determine de novo whether Blue Cross' denial of coverage for Plaintiffs' treatment with HDCT-ABMT was correct under the contract language. In this regard, Blue Cross' decision to deny benefits to Mrs. Adams and Mrs. Whittington is incorrect if the Court finds that the treatment for those particular patients was, as of April or July of 1990, "generally acknowledged as accepted medical practice by the suitable medical specialty practicing in Maryland."

Blue Cross acknowledges from the outset and throughout trial that Dr. Keefe did not consult, or consulted minimally, Maryland medical oncologists in arriving at his decision to deny coverage for Plaintiffs' claims. At trial, Dr. Arthur Keefe testified that as corporate Medical Director in charge of medical policy, he alone was responsible for the decision to deny coverage to Mrs. Adams and Mrs. Whittington. Dr. Keefe testified that in deciding to deny coverage, he relied heavily on a 1988 technical evaluation prepared by the Technical *670 Evaluation Committee (TEC) in association with the National Blue Cross-Blue Shield Association. After reviewing the relevant scientific literature, the TEC committee reported in its Uniform Medical Policy Memorandum, issued in December of 1988, that HDCT-ABMT used as treatment for breast cancer was experimental. Dr. Keefe relied on the TEC evaluation to formulate company guidelines which excluded coverage for HDCT-ABMT as treatment for breast cancer. Dr. Keefe made the initial coverage determination pursuant to those guidelines, and after receiving appeal letters from both Plaintiffs, he reviewed the material briefly, looking for any new scientific data which might cause him to rethink his position. Rather than relying on local expert medical opinion, Dr. Keefe relied on Blue Cross' evaluation of scientific data, as well as his own independent review.

Despite the fact that Dr. Keefe did not consult Maryland medical oncologists, Blue Cross nevertheless maintains that Dr. Keefe in fact made the correct decision. In particular, Blue Cross seems to argue that the Blue Cross TEC evaluation reflected a national consensus, which coincided with the consensus in Maryland, that HDCT-ABMT was not yet accepted medical practice. At trial, Blue Cross presented five experts in support of its contention that HDCT-ABMT was, as of April and July of 1990, still experimental and investigational.

Blue Cross relied most heavily on Dr. David Eddy, a biostatistician hired by Blue Cross to prepare the 1990 National Association Report, which he did after reviewing the newest scientific literature on the subject. At trial, Dr. Eddy defined an "experimental" technique as a technique about which there exists a number of unanswered fundamental questions, including questions about benefits and about potential harm from the treatment. Dr. Eddy gave his opinion that based on his review of the data, HDCT-ABMT was still experimental because a number of questions remained unanswered with regard to potential benefits, in particular overall survival rates, as well as potential harm, namely, toxicity rates. In his testimony regarding the nature of the Blue Cross decision to classify the treatment as experimental, Dr. Eddy testified, as did Dr. Keefe, that Blue Cross considered it improper to rely on practitioner opinions to determine whether a procedure was experimental; instead, independent Blue Cross analysis of the scientific data was necessary. As a result, most of Dr. Eddy's testimony focused on a review of the published scientific research results.

Like several other Blue Cross experts, Dr. Eddy also found it highly significant that Phase III randomized clinical trials^[5] had not yet been completed for HDCT-ABMT, but were in the process of being conducted. Dr. Eddy acknowledged that Phase III studies were not necessary in all cases, and pointed to the use of HDCT-ABMT for non-Hodgkins lymphoma as an example of a "home-run" treatment, i.e., a treatment which had produced such dramatic results in Phase II studies that Phase III results were not required in order to classify the treatment as "accepted medical practice." However, Dr. Eddy testified that Phase I and II studies for HDCT-ABMT in the treatment of breast cancer had not achieved such "home-run" results. In particular, Dr. Eddy noted several times that significant long-term survival rates in breast cancer patients treated with HDCT-ABMT had not yet been demonstrated, at least not in studies which he considered to be reliable. Dr. Eddy found fault with studies which used historical controls, in part because he believed the historical controls *671 to be poorly matched with experimental subjects.

Finally, Dr. Eddy testified that no consensus existed in the medical community with regard to the use of HDCT-ABMT in the treatment of breast cancer. In support of his opinion, Dr. Eddy pointed to the scientific literature in which authors had concluded their articles with a statement that more research was needed to confirm their results. Dr. Eddy also noted that when administering the procedure, physicians told patients that their treatment was part of a research project, as is required by informed consent law.

In addition to Dr. Eddy, Blue Cross presented several other experts: two practicing Maryland oncologists, Dr. Serpick and Dr. Feldman, as well as an AIDS researcher, and an oncologist practicing in Los Angeles. All of these experts opined that HDCT-ABMT was experimental in April and July of 1990, based on their review of the scientific literature, which they believed left many questions unanswered with respect to the procedure. In particular, the Blue Cross expert witnesses pointed to questions regarding (1) whether the procedure resulted in long-term survival; (2) toxicity rates; (3) the appropriate "cocktails", or combinations of chemotherapy drugs, to use; and (4) the effect of using growth factors to stimulate bone marrow replacement after untainted bone marrow has been reinfused.

Most if not all of the Blue Cross experts testified only peripherally about the opinion of members of the Maryland oncological community. Dr. Serpick, founder of Mustard, Onkovin, Procarbazine and Prednisone (MOPP) therapy for Hodgkin's disease and a practicing Maryland oncologist, testified minimally about the views or practices of Maryland oncologists. What little relevant testimony he did offer was based on his contact with Maryland oncologists in the Baltimore Oncology Club, which has 15 to 20 members, and other informal contact. Dr. Serpick reported that Maryland oncologists did not generally acknowledge the treatment as "accepted medical practice" (as he defined the term a procedure for which investigators are not working on any hypothesis, for which all benefits and risks are known) as of either April or July of 1990.

Dr. Feldman, who is Chief of Oncology at Mercy Hospital in Baltimore, drew similar conclusions based on his own review of the literature and informal conversations with other doctors. However, like Dr. Serpick, Dr. Feldman premised his discussion on a definition of "accepted medical practice" which turned on whether the practice was well-described in the literature, had been tested by many persons, produced acceptable toxicity ratios and well-described results. Importantly, despite their conclusions, both Dr. Serpick and Dr. Feldman admitted that given the appropriate patient, they too would have referred their patients for treatment with the procedure. Blue Cross' other experts defined the term similar to Dr. Feldman and Dr. Serpick, but failed to discuss the views of Maryland oncologists with any degree of specificity.

The Court does not accept the opinions of the Blue Cross experts because they based their opinion on a definition of the term "accepted medical practice" which is inconsistent with the contract language. Instead of focusing testimony on the opinion of members of the Maryland oncological community, the Blue Cross experts concentrated on their own independent evaluations of the scientific data. Indeed, Blue Cross' primary expert, Dr. Eddy, is not a practicing physician but is an expert in biostatistics. However, scientific data only provides statistical results from research. After reviewing the relevant scientific data, the practicing medical community must make an overall value judgment about whether a treatment is accepted, that is, has a sufficiently acceptable risk-benefit ratio to justify offering, or indeed recommending, the treatment as an option. According to the contract, Blue Cross must defer to that practical medical judgment if a consensus among Maryland oncologists agrees that the treatment is accepted medical practice. Thus, the Court accords little weight to the opinion of the Blue Cross experts.

*672 In contrast to the Blue Cross experts, Plaintiffs' expert witnesses utilized a practical definition of what constitutes "accepted medical treatment," one which is both consistent with the standard legal definition discussed earlier and, most importantly, consistent with the contract language. Appropriately, Plaintiffs' experts focused on whether a majority of Maryland oncologists viewed the treatment as accepted medical practice, or were in fact referring their patients for the procedure. Based on Plaintiffs' experts' testimony, this Court finds that a consensus existed as of April and July of 1990 among Maryland oncologists that HDCT-ABMT for breast cancer was an "accepted medical practice."

Drs. Roy Beveridge, Enser Cole, Martin Abeloff, Thomas Spitzer, Marc Lippman and Nancy Davidson, all of whom are, or regularly practice with, practicing Maryland oncologists, testified convincingly and with specificity that a consensus of Maryland oncologists consider the treatment to be accepted medical practice. These doctors, by virtue of their leadership positions in the Maryland oncological community, particularly with respect to HDCT-ABMT in the breast cancer field, were at the relevant time in contact with a large number of Maryland oncologists and were in a position to become acquainted with their views. For example, Dr. Roy Beveridge, who is Director of the Bone Marrow Transplant Program at Fairfax Hospital in Virginia, testified to having had hundreds of conversations with Maryland oncologists dating from 1987 to the present, in the process of setting up and operating the Bone Marrow Transplant Program. Dr. Nancy Davidson, who is a practicing medical oncologist in Maryland, testified as well that she could identify fifty or sixty Maryland oncologists who have referred or discussed patients with her with regard to using HDCT-ABMT to treat breast cancer. These doctors, as well as the other Plaintiffs' experts, testified that as of April and July of 1990, HDCT-ABMT for breast cancer was "generally acknowledged as accepted medical practice" by Maryland oncologists.^[6] By contrast, the majority of the Blue Cross experts were far less knowledgeable about the views of Maryland oncologists, and testified minimally with regard to those opinions.

Moreover, as Plaintiffs' experts point out, and as the district court noted in Pirozzi v. Blue Cross-Blue Shield of Virginia, 741 F. Supp. 586, 591 (E.D.Va.1990), HDCT-ABMT treatment is in use at many major medical centers, including the most highly respected medical centers around the country. "This is convincing evidence that the treatment has `scientifically proven value' and is `in accordance with generally accepted standards of medical practice.'" Id.

In summary, the Court is persuaded by Plaintiffs' expert testimony with regard to the opinion of a consensus of Maryland oncologists, the referral practices of those oncologists, and the number of major medical centers around the country which provide the treatment. Weighing that evidence against the opinion of Blue Cross experts, the Court finds by a clear preponderance of the evidence that as of April and July of 1990, HDCT-ABMT was generally acknowledged as accepted medical practice by Maryland oncologists.

Although it is not necessary to support the preceding conclusion, the Court determines *673 that even if it were to accept the Blue Cross definition of the term "accepted medical practice," which turns on scientific criteria, HDCT-ABMT would satisfy Blue Cross' purely scientific criteria.

First, the Court finds from expert testimony that at the relevant time, HDCT-ABMT satisfied the five scientific criteria developed by Blue Cross' own Technology Evaluation Committee.^[7] In July of 1990, the procedure was evaluated according to the five TEC criteria by a panel of highly qualified physicians, including Dr. Gianni Bonadonna, a noted expert in breast cancer who is the Director of Division of Medical Oncology at the National Cancer Institute in Milan, Italy, and Dr. Vincent DeVita, Chief Physician at Memorial Sloan Kettering Cancer Center in New York. The evaluation team issued a report entitled "High Dose Chemotherapy and Autologous Bone Marrow Support for Breast Cancer: A Technology Assessment," in which they concluded that HDCT-ABMT as treatment for breast cancer satisfied all five of the TEC criteria.^[8] Similarly, Dr. Lippman and Dr. Abeloff both testified that as of April, 1990, HDCT-ABMT satisfied all five of the TEC criteria.

Second, with regard to the Blue Cross experts' concern about long-term survival rates, this Court is persuaded by Plaintiffs' experts' testimony that at the time Blue Cross decided to deny benefits to Plaintiffs, HDCT-ABMT had already demonstrated dramatic increases in complete and overall response rates, that is, the rate of tumor shrinkage, as well as significant improvement in disease-free survival.

In supporting Mrs. Adams' plea for Blue Cross payment, Dr. Peters submitted a package of material in April of 1990 for Dr. Keefe to review, which included graphs of results from the Duke clinical trials illustrating survival rates for women with Stage IV and Stage II/III breast cancer, which are, respectively, the prognoses for Mrs. Whittington and Mrs. Adams. The data illustrated that for women with Stage IV metastatic breast cancer who are treated early in the disease with induction therapy (standard low-dose therapy) followed by high-dose therapy with bone marrow support, studies had demonstrated that 70% had achieved total remission as compared to a 15% to 20% "complete response" rate under standard low-dose therapy alone. Of those women treated with prior chemotherapy *674 and high-dose therapy with bone marrow support, 25% remained disease free at a 3 and ½ year follow-up, versus a 20% rate for standard therapy. Dr. Peters testified that his conclusions were supported by data from the Dana Farber Cancer Institute at Harvard, and the M.D. Anderson Institute in Houston.

While overall survival rates may be similar between low-dose therapy and HDCT-ABMT for Stage IV patients, the Court nevertheless notes a significant benefit in disease-free survival to be gained from using HDCT-ABMT. That is, even if a substantial number of those women were to die tomorrow and the overall survival rate were to drop to the level of low-dose therapy, at any one point in time more women will have lived longer free of disease than if they had been treated with low-dose therapy. The Court considers it highly unreasonable for Blue Cross to ignore the benefits of disease-free survival in favor of concentrating only on long-term survival rates.

For women with Stage II/III primary breast cancer involving 10 or more lymph nodes, the data published by Duke, made available to Blue Cross in Dr. Peters' package, was even more promising. Dr. Peters testified that as of the time Blue Cross denied coverage to Mrs. Adams, Duke trials had demonstrated that after 30 months, 70% of the women treated with HDCT-ABMT following adjuvant therapy, i.e., therapy following surgery to remove the breast, achieved disease-free survival, compared to a mere 20% of women treated with low-dose therapy. In addition, those women treated with HDCT-ABMT achieved an overall survival rate of 80% at a followup of 40 months, compared to 60% for women treated with low-dose therapy.

Beyond dramatically increased complete response rates and increased disease-free survival rates, the Court notes that women undergoing HDCT-ABMT spend far less time "on therapy", that is, receiving the treatment, than do their counterparts on low-dose therapy. Dr. Karen Antman, an expert in oncology at the Dana Farber Institute affiliated with Harvard University Medical Center, testified that a woman with Stage IV disease may receive low-dose treatments every 3 to 4 weeks until she dies, compared to the single treatment administered for HDCT-ABMT. Dr. Peters also testified that "off therapy" time translates into less time in which the patient experiences the toxic side effects associated with chemotherapy, such as nausea and hair loss, and a corresponding better quality of life for the patient. The Court is persuaded by expert testimony that the increase in off-therapy time, when coupled with the dramatic increase in complete response and disease-free survival rates, may serve to offset whatever increase in toxicity may result from using HDCT-ABMT.

In addition to the statistics outlined above, Dr. Peters and Dr. Antman testified that Dr. Eddy's review of the scientific literature was seriously flawed. In particular, in assessing the performance of patients receiving HDCT-ABMT, Dr. Eddy included data from women who had refractory disease, that is, women who were not responsive to prior chemotherapy, and who were as a result less likely to respond to higher doses of chemotherapy. Dr. Eddy committed a methodological error by including members of this group, which obviously would be the poorest candidates for success with high-dose chemotherapy, in comparison to a select group receiving low-dose chemotherapy who had received no prior chemotherapy, and had thus not demonstrated that they were "nonresponders." In reviewing studies, Dr. Eddy also mischaracterized the Eder study, a study with lower success rates than others, as a study involving an "up-front" patient group that is, women receiving HDCT-ABMT without prior chemotherapy, when in fact the study involved patients who had received prior chemotherapy and had demonstrated that they did not respond well to chemotherapy. In fact, Dr. Antman testified that a more correct comparison between refractory, or non-responsive, patients receiving low-dose treatment and refractory patients receiving high-dose therapy demonstrates support for HDCT-ABMT's superiority. Although all women in the refractory group receiving high-dose *675 therapy with bone-marrow support eventually relapsed, 30% of those women achieved complete remission (complete response), as compared to only 10-20% of those receiving low-dose therapy. The Court also noted Dr. Peters' testimony that if non-responsive Stage IV patients are treated early in the disease, 15% of those patients can remain disease free for as long as seven years.

With respect to the Blue Cross expert witnesses' concern over the incomplete Phase III studies, the Court finds such concern to be of no moment. Both sides agree that completed Phase III research is not always necessary in order to judge a treatment to be accepted medical practice. In particular, for some treatments, earlier Phase I and Phase II studies may have produced such dramatic positive results, the so-called "homerun" results, that Phase III studies are not required. For example, both Blue Cross and Plaintiffs agree that HDCT-ABMT as a treatment for lymphoma was just such an instance. The parties disagree on whether the research on HDCT-ABMT as used to treat breast cancer produced "home-run" results.

However, Dr. Antman testified for Plaintiffs, and the Court was convinced, that the data produced in Phase I and II research studying the use of HDCT-ABMT to treat breast cancer is as dramatic and convincing as was the data for lymphomas. Specifically, data on toxicity rates, complete response rates, and most importantly, rates for two-year disease-free survival, are strikingly similar for HDCT-ABMT as treatment for breast cancer and as treatment for lymphomas.

Of additional significant import is the testimony from Dr. Peters and Dr. Antman that conducting a Phase III random clinical trial would be ethically acceptable, as evaluated by institutional review boards, only if the treatment was potentially as good as, if not better than, low-dose therapy. Investigators could not ethically conduct randomized clinical trials if HDCT-ABMT was known with any certainty to be less effective than low-dose chemotherapy. Contrary to Dr. Eddy's assertions, then, the fact that Phase III studies are being conducted to determine if HDCT-ABMT is better than low-dose therapy indicates that the treatment is anticipated by an institutional review board to be at least as good as low-dose therapy, as evidenced by the results from Phase I and Phase II studies. The Court heard testimony from Dr. Peters that, contrary to the Blue Cross argument, the very fact that Phase III trials are being conducted indicates a general consensus that the treatment appears potentially to be as effective as low-dose chemotherapy, may be better, and in any event, is probably an acceptable treatment option. In fact, Dr. Peters testified that to make Phase III studies ethically acceptable, researchers have had to offer HDCT-ABMT to those patients who relapse on low-dose therapy, precisely because data demonstrates HDCT-ABMT's superiority.

The Court recognizes that questions remain with respect to overall survival, but notes that research need not prove a treatment completely curative in order for it to have sufficient merit to be judged an "accepted medical practice." Questions also may remain with respect to both the use of colony stimulating growth factors, designed to accelerate marrow replacement in the "rescued" patient, and the design of an optimal preparative regimen, i.e., combination of drugs. However, as Dr. Antman noted, those questions, which are the focus of current study, are secondary questions at best, and do not preclude a judgment that the treatment is currently "accepted medical practice." In fact, Dr. Antman points out that with respect to low-dose therapy, research is currently being conducted to determine optimal preparative regimens.

Furthermore, many of today's accepted medical treatments are offered under a research protocol. Given the definition of "experimental" under the plan, the fact that the treatment is administered as part of an experimental protocol designed to facilitate the collection of data does not necessarily mean that the treatment is by definition experimental. See Dozsa v. Crum & Forster Insurance Co., 716 F. Supp. 131, 138-39 (D.N.J.1989).

*676 Thus, even if one accepts the Blue Cross definition of "accepted medical practice" as dependent upon scientific criteria, the Court finds that Blue Cross's determination that HDCT-ABMT was experimental and investigative runs counter to the evidence presented at trial in this case. Questions with regard to the net health benefits of the procedure, in particular, response rates and disease-free survival, have been sufficiently answered so as to allow Maryland oncologists to choose HDCT-ABMT as an appropriate treatment alternative, and to justify the conclusion that the treatment is an accepted medical practice. The Court finds it necessary to defer, as Blue Cross should have done under the terms of its own plan, to the opinion of the consensus of Maryland oncologists.

The foregoing reasoning is premised on the Court's decision to review the Blue Cross decision de novo. As discussed above, the phrase "as decided by us" is of insufficient scope and precision to entitle Blue Cross to judicial deference, either with respect to determining eligibility for benefits, or construing disputed or ambiguous terms. However, even if the Court were to find the "as decided by us" language to have granted discretionary authority to Blue Cross, thus warranting review under an "arbitrary and capricious" standard, the Court finds that Blue Cross' decision to deny coverage was indeed "arbitrary and capricious."

A decision to deny benefits can be held arbitrary and capricious if the plan administrator's interpretation is unreasonable, see Bruch, 109 S. Ct. at 954, or if the administrator "entirely failed to consider an important aspect of the problem [or] offered an explanation for its decision that runs counter to the evidence." Motor Vehicle Mfrs. Ass'n v. State Farm Mutual Auto, In. Co., 463 U.S. 29, 43, 103 S. Ct. 2856, 2866, 77 L. Ed. 2d 443 (1983). The plan administrator must "examine the relevant data and articulate a satisfactory explanation for [his] action." Id.

The Court finds the Blue Cross decision to be arbitrary and capricious for two reasons. First, as discussed above, the Court finds the Blue Cross interpretation of "accepted medical practice" to be unreasonable because it is inconsistent with the language in the exclusion provision. Thus, the Blue Cross decision deriving from an unreasonable definition of terms is itself unreasonable. Alternatively, accepting arguendo Blue Cross' focus on scientific criteria, Blue Cross' decision was unreasonable because it ran counter to the evidence before Dr. Keefe.

Secondly, Blue Cross' decision to deny benefits was arbitrary and capricious because it failed to consider the most relevant aspect of all the opinion of the Maryland oncological community. In this case, Blue Cross had absolutely no evidence before it with regard to whether Maryland oncologists generally acknowledged HDCT-ABMT to be accepted medical practice. In fact, both Dr. Keefe and Dr. Eddy stated at trial that professional opinions of practicing oncologists were unreliable. Instead, Dr. Keefe unilaterally drew his own conclusions, based upon the National Association Report and his own review of the data, that Maryland oncologists could not possibly have considered the treatment to be accepted because, in Dr. Keefe's opinion, no proof of the treatment's efficacy yet existed.

This Court finds it completely unreasonable for Dr. Keefe to have relied on outside scientific opinion, and his own independent review of the literature, while selectively ignoring the professional opinions of Maryland oncologists and Plaintiffs' treating physicians. If Blue Cross had wished the plan administrator to unilaterally review the scientific data and make the final determination with regard to whether a procedure was accepted, it could have drafted its contractual exclusion to reflect that intent. For example, in Reilly v. Blue Cross & Blue Shield United of Wisconsin, 846 F.2d 416 (7th Cir.1988), the court reviewed a Blue Cross benefit plan which excluded treatments "not yet recognized as accepted medical practice by Blue Cross & Blue Shield United." (emphasis added). The Blue Cross plan at issue in this case did not *677 contain such a provision. Yet Blue Cross arbitrarily and capriciously acted as if it did.

It is no small wonder that with respect to whether Maryland oncologists generally acknowledged HDCT-ABMT as accepted medical practice as of April or July of 1990, Dr. Keefe reached the wrong conclusion. As discussed above, this Court finds that had Dr. Keefe chosen to consult with the doctors listed in the material submitted to him by Plaintiffs, he would have discovered that HDCT-ABMT was generally acknowledged as accepted medical practice by Maryland oncologists. His failure to consult the very medical experts to which the plan defers is nothing if not unreasonable. Under the terms of the plan, Dr. Keefe should have deferred to the opinion of the Maryland oncological community, as does this Court in its decision today.

For the foregoing reasons, this Court finds in favor of the Plaintiffs, and rules that Blue Cross-Blue Shield of Maryland must pay for High Dose Chemotherapy with Autologous Bone Marrow Transplant treatments for Mrs. Alexandra Adams and Mrs. Kelly Whittington.

[1] Blue Cross covers HDCT-ABMT when used to treat Hodgkins Lymphoma, Non-Hodgkins Lymphoma, acute leukemia, testicular cancer, and neuroblastoma.

[2] Breast cancer is classified into stages for purposes of prognosis and treatment. In Stage II, the tumor in the breast measures from 2 to 5 centimeters and/or has spread to the lymph nodes under the arm, the "axillary" nodes. In Stage III, the tumor measures more than 5 centimeters and involves a great number of axillary nodes or other lymph nodes or tissue near the breast. In Stage IV, the most advanced stage of breast cancer, the cancer has spread or "metastasized" to other organs of the body.

[3] In light of the plan provision's definition of "experimental," whether the treatment at issue is being tested to gather research information is not relevant, as discussed in detail below.

[4] In addition to the Frye standard, Blue Cross also relies heavily on the Belmont Report, a report issued by a presidential commission on biomedical research, in support of its narrow definition for "accepted medical practice." The Belmont Report standard is in all respects identical with the Frye standard and suffers from the same deficiencies. Namely, the standard is inconsistent with a contractual exclusion which looks to the judgment of practicing clinical oncologists, rather than to evaluation by independent scientific observers.

[5] Clinical trials are usually classified into one of three phases, and research usually progresses in order from Phase I to Phase III. Phase I involves initial testing for the feasibility of conducting an experiment, to include determining the appropriate dosage of drugs administered within a treatment, and to evaluate the toxic effects produced by the drugs. Phase II studies are conducted to determine whether the treatment has any observable effect in the patients receiving it. Phase III, the randomized clinical trial, is the last stage of research, and involves using a control group of subjects who do not receive the treatment in order to compare the results from the group receiving the treatment. These studies are conducted to confirm and quantify the effectiveness of the treatment.

[6] Dr. Martin D. Abeloff, who is the Clinical Director at the Cancer Center at Johns Hopkins, testified that based on his contact with Maryland oncologists, he estimated that of one hundred practicing Maryland oncologists, in excess of fifty have actually referred patients for the procedure. Dr. Enser Cole is a practicing oncologist, whose partner in practice is in charge of the Community Physician's Program, sponsored by Johns Hopkins, which brings in physicians from many parts of Maryland. Dr. Marc Lippman, who is Director of the Vincent T. Lombardi Cancer Center affiliated with Georgetown University Medical Center, sent out a letter in 1989 to approximately 60 or 70 board certified oncologists, and 90% responded that HDCT-ABMT was the best available therapy for certain subsets of patients with breast cancer. All of the above Plaintiffs' experts concluded based on their contact with Maryland oncologists that at the relevant times, HDCT-ABMT was "generally acknowledged as accepted medical practice" by Maryland oncologists.

[7] The five criteria developed by the Technical Evaluation Committee are as follows:

(1) The technology must have final approval from the appropriate government regulatory bodies.

(2) The scientific evidence must permit conclusions concerning the effect of the technology on health outcomes.

(3) The technology must improve the net health outcome.

(4) The technology must be as beneficial as any established alternatives.

(5) The improvement must be attainable outside the research setting.

[8] The team's evaluation produced the following conclusions: HDCT-ABMT ...

1. Produces a higher frequency of objective response in all settings in which it has been tested;

2. Results in even some poor-prognosis patients obtaining long-term disease-free survival;

3. Produced results equivalent to, and with certain regimens, superior to, long term results for conventional therapies for early metastatic breast cancer.

4. Treatment programs combining induction chemotherapy and ABMT produce a higher complete response rate, higher overall response rate, and usually an increased median duration of progression-free survival when compared to programs not involving ABMT. Further, it appears that a limited number of patients with an otherwise very poor prognosis can achieve long-term disease-free survival using this approach.

5. The evidence strongly favors the conclusion that high-dose chemotherapy and autologous bone marrow support when used as consolidation in Stage II or III breast cancer involving large numbers of axillary lymph nodes is superior to alternative currently available therapeutic approaches. While randomized trials would strengthen the scientific position, the magnitude of the therapeutic benefit evidenced already may raise in some patient's minds concern that such trials would be unethical if they denied access to the high-dose therapy.

6. Conduct of appropriately designed randomized and non-randomized clinical trials should be encouraged wherever feasible and ethical to enhance the understanding of the role of various components of differing treatment programs on patient's well-being.