Who should be responsible?

The government?

A basic philosophy of most health plans is to restrict coverage to those therapies that are known to provide some quantifiable health benefit. By definition, clinical trials involve therapies with an unknown benefit, particularly in Phase I or early Phase II trials. While the patient may not benefit directly from the clinical trial, the data regarding the safety and effectiveness generated during the study will certainly benefit future generations of patients, and will serve society as a whole. Those who construe these trials to serve primarily a social good advocate that funding is a governmental responsibility, and that requests for payer funding reflect cost shifting in response to declining and inadequate funding of the National Cancer Institute (NCI) and other governmental agencies.

Payers?

A variety of legislative mandates have put the onus of funding clinical research squarely on the backs of payers. Advocates of payer funding have become more vocal as government support of clinical research has dwindled. They point out that it is arbitrary to separate investigational elements of a treatment from issues of basic patient care. For example, in a Phase III clinical trial, there is theoretically adequate evidence that the "investigational" arm of the treatment performs as least as well, and hopefully better, than the standard, control therapy. Most payers would perceive Phase I trials, whose focus is primarily toxicity and not effectiveness, to clearly represent investigational therapy that would be ineligible for coverage.

The American Society of Clinical Oncology and the NCI contend that even in Phase I trials, where there may be minimal clinical evidence of effectiveness, there must be some therapeutic intent to justify toxic therapy. At the very least, there is laboratory evidence that the drug has significant antitumor activity and so some patient benefit should be expected. Furthermore, some phase I trials may involve drugs with known clinical antitumor activity, but are being used at different dosages or in unique combinations with other well-known drugs or therapy (i.e. radiation), or represent analogs of established drugs; in these situations a therapeutic effect is fully anticipated, but a Phase I trial is needed to determine toxicity.

In this analysis, the patient potentially benefits in all phases of trials. Payer funding advocates point out that although elements of the trial may be considered investigational, the majority of the care rendered in a clinical trial is basic (and high quality) patient care, such that using clinical trial participation is an arbitrary and artificial distinction. Finally, even if the patient does not benefit directly, participation in a trial should generate the outcomes data that payers rely on to develop coverage policy. Thus, funding clinical trials can be considered as research or patient care costs for the payer.

Payers argue that clinically significant outcomes, either in terms of curative potential or a significant prolongation in quality of life, should be the criteria for coverage. For example, the standard for equivalence in effectiveness is often set very low due to the ineffectiveness of many standard chemotherapy regimens for the most common malignancies. It is not difficult to be "at least as effective" as something that does not work. A minor reduction in tumor burden may have no effect on the length or quality of life; thus, therapeutic intent cannot be considered to be equivalent to patient benefit.

In many oncology trials, the standard therapy is known to have minimal to no curative potential, and yet this largely ineffective therapy is routinely funded by insurers. The advocates ask: Why shouldn't payers fund clinical trials where at least there is some assurance that the patient is being treated with a scientifically based treatment option? Subtly implicit in this line of thought is that in a quid pro quo for funding clinical trials, perhaps payers should reconsider funding those "standard" therapies known to be ineffective, such as multiple or prolonged attempts at chemotherapy in tumors that have been shown to be chemoresistant. So far, payers have been unwilling to address this emotional issue through coverage policy.

Many payers might be more receptive to funding clinical trials if the diffusion of investigational therapies could be restricted until adequate outcomes data could be collected. However, the experience with high dose therapy and bone marrow transplant, where insurance coverage was partially responsible for the rapid and uncontrolled diffusion of this technology outside of clinical trials, is a sobering testament to the influence of insurance coverage.

In this regard, the concern is not so much funding for clinical trials, but funding for the investigational therapies that are rendered outside of any clinical trial. Additionally, there is a concern regarding the quality of clinical trials. If payers were to provide funding for clinical trials, what guarantees would there be that self-serving clinical "trials" could not be rapidly concocted by institutions to support their individual programs?

Manufacturers?

The basic premise of this philosophy is that the proponents of a technology are responsible for validating its safety and effectiveness. The FDA is perhaps the world's foremost technology assessment organization, and lack of FDA approval is synonymous with remaining questions regarding a product's safety and effectiveness. Once a product receives FDA approval, the manufacturer can legally market the product for its approved indication and begin to enjoy the profits from its sale.

Payers typically consider clinical trials performed during the FDA approval process to be part of the research and development costs of the manufacturer. However, many next generation devices seeking FDA approval are similar in design to existing FDA-approved devices, so the industry contends that these devices should raise minimal issues of safety and effectiveness. Additionally, manufacturers argue that, unlike drugs whose effects are transitory, a patient may benefit from an implantable device for many years, so devices can be considered a patient care benefit.

Recently, Medicare altered its policy to provide coverage for some of these non-FDA-approved, next-generation devices. Nevertheless, many payers maintain that if a product cannot be legally marketed based on its FDA status, that its use should be part of the research and development costs of the manufacturers.

Some payers are willing to separate the research and patient care costs of certain devices, for example, funding the pre- and postoperative patient care costs associated with implanting the device, but declining funding for the device itself. Not surprisingly, manufacturers have not been receptive to this distinction between patient care and research costs.

While drug and device companies are easily recognized as economic beneficiaries of the diffusion of a new technology, the beneficiaries of surgical or medical procedures, which do not require FDA approval, are less tangible. The issue of high dose chemotherapy and autologous transplant is a good example. The rapid diffusion of this technology is a testament not only to the numbers of candidate patients and the desperate nature of their illness, but also reflects the economic incentives for institutions to offer this intense, largely inpatient treatment. While many institutions and individual physicians have reaped economic benefits from offering as yet unproven high dose therapy, their responsibility in funding the research costs to validate this procedure (particularly when performed outside of a formal clinical trial) are typically not part of the debate.

A compromise?

The recent decision by the U.S. Department of Defense to fund cancer clinical trials sponsored by the NCI reflects renewed government interest in supporting clinical research. While in the past, government support was funded through the NCI budget, support now comes from the CHAMPUS budget, and presumably reflects the view that these dollars represent individual patient care costs. Nevertheless, CHAMPUS dollars are public money and in its notice in the Federal Register, CHAMPUS points out that this support for clinical research will facilitate recruitment for NCI trials. In some respects, this funding is a blending of the strategies mentioned above. It is interesting to note, however, that the funding is restricted to Phase II and III trials. Apparently, CHAMPUS does not feel that participation in a Phase I clinical trial represents a patient care cost, but rather falls more in the category of a research cost.

In addition, CHAMPUS will restrict funding only to NCI-approved clinical trials. Therefore, those investigational therapies offered outside of an NCI-sponsored trial will not be funded; thus, patients are funneled into rigorously peer-reviewed clinical trials. This type of limited diffusion while a new technology is under investigation has been of interest to managed care companies for some time. The CHAMPUS program is a demonstration project with a cost assessment planned at the end of the year. To the extent that this program is budget neutral and is contractually and legally feasible for other managed care companies, CHAMPUS' approach may become an attractive compromise.

Elizabeth Brown, MD, is Director, Technology Assessment and Clinical Guidelines, Aetna Health Plans, Aetna Life and Casualty, Chicago, Ill. The opinions expressed in the article are those of the author and not necessarily those of Aetna Health Plans