Lung Cancer Screening at Any Price?

Most biomedical research focuses on the quest to discover new methods for reducing morbidity and mortality from serious illnesses. Once these interventions are created and developed, the next steps in science are directed toward confirmation of the efficacy and effectiveness of these interventions and then estimating their potential impact. On some level, for most investigators, this represents the be-all and end-all of research.

The last decade of medical practice and the changes that have occurred have led to a dramatic alteration in how clinicians perceive the utility of these interventions. No longer is it sufficient that a medication or procedure prolong life or even cure the patient. Today, medical costs also play a role in assessments and judgments of what interventions to use and under what circumstances to use them. In an era of limited resources, these judgments enable physicians to choose among interventions based on these considerations. Expensive drugs or treatments that cure a few individuals may not stand up against less effective but cheaper programs that can cure many. Because of this focus on the societal benefits of therapy, cost-effectiveness has become an important, even crucial, part of health policy decision-making with regard to the use of expensive, or even inexpensive, treatment modalities.^{1 - 4} Hence, from a health policy perspective, a cost-effectiveness policy regarding testing and preventive treatment options may save millions of dollars each year on the costs of introducing highly expensive, but marginally effective, technology.

Cancer is certainly one of the most important health concerns in the United States, and lung cancer leads all other types of cancer in mortality.⁵ But because of the high frequency of lung cancer mortality and the relatively short survival time, the actual costs of lung cancer to the health care economy are among the lowest of the major cancers in the United States.⁶ Furthermore, because so much of lung cancer incidence can be attributed to tobacco use, smoking cessation must remain the first and foremost priority in reducing the burden of lung cancer in the population.

Despite such efforts, and despite some significant success in reducing the rates of cigarette smoking in the United States,^{7 - 8} further strategies are needed to cope with the lung cancer epidemic. A new development in this area that has generated much interest and controversy has been the use of spiral helical computed tomography (CT) scanning for the detection of very early lung cancers,^{9 - 11} and the apparent improvements in lung cancer treatment and survival.^{9 ,12 - 13} Many disagree about the true benefits of spiral CT screening for lung cancer.^{14 - 15} However, the National Cancer Institute (NCI) has recently initiated a large randomized trial to test its efficacy more adequately.^{16 - 17} As with most large, expensive, and long-term screening trials, both the public and the medical community are left to wonder what to do in the interim, until the results of the trial are available. At the same time, though, commercially available cancer screening programs have been rapidly disseminated nationwide, fueled by direct promotion to consumers and funded by direct payment on the part of consumers.¹⁸

In this issue of THE JOURNAL, the decision and cost-effectiveness analysis by Mahadevia and colleagues¹⁹ evaluates whether lung cancer screening using helical CT might be considered an appropriate strategy for adult smokers and those who have recently quit smoking. The authors find that such an approach is very expensive from both health policy and societal perspectives. The investigators have assembled the best data available on the outcomes associated with lung cancer screening, as well as the costs associated with most of the elements involved in the early diagnosis, treatment, and long-term outcomes of lung cancer. Furthermore, they have conducted painstaking analyses, with appropriate attention to sensitivity analyses and other accepted methods, to explore and compare the relative costs of this type of screening in various subgroups and under different sets of circumstances. Based on their results, even with the assumption that helical CT screening will truly prove to be efficacious, it currently appears to be too expensive in most reasonable and foreseeable situations.

Some caveats must be borne in mind. This advance in imaging technology is a brand-new modality and approach to lung cancer, and is still in its infancy. Cost-effectiveness analyses are highly dependent on large numbers of assumptions, and are only as good as those assumptions. With improvements in the available technology, and with more extensive evidence regarding its efficacy (including its sensitivity, specificity, and anticipated reductions in unnecessary workups of false-positive findings), this type of cost-effectiveness analysis will need to be updated and reconsidered on a frequent basis. Certainly, 10 or more years from now, when the results of the NCI randomized trial are known, these analyses will need to be revamped.

Another consideration is the actual assessment of quality of life and quality-adjusted life-years that are the primary outcomes of this analysis. The authors appropriately use the average quality-adjusted life-years for each health state²⁰ ; however, there is almost certainly a wide distribution of individual preferences. Thus, there may be many individuals for whom their level of preferences may justify this type of screening, even under the parameters postulated in this article.¹⁹ Likewise, the opposite may also hold true.

Furthermore, the other major area in which Mahadevia et al made assumptions is with regard to the actual cost of each of the factors within the model. Future treatments may change and become less invasive (surgery is still needed to treat positive findings) and less expensive. Also, the technology may improve and the costs of the scanning may decline, and indeed almost surely will. Improvements in the technology may significantly enhance its actual efficacy from the estimates used by the investigators in their analyses.

In addition, costs are sensitive to the present value of money.¹ A study that requires annual screening examinations is highly dependent on the discount rate; the lower this rate, the more favorable the cost-effectiveness of spiral helical CT scanning would be.¹ Therefore, the cost-effectiveness analyses, as embodied in this study, represent a dynamic and fluid model.

Cost-effectiveness estimates, along with the estimates of efficacy, remain critically important components of health policy decision-making and assessment of the use of this or any other health intervention. The estimates reported by Mahadevia et al represent the reality of current data and should lead to some hesitation and thoughtful consideration of whether to use this screening modality at the present time in a given individual. Until more data are available and the NCI randomized trial is completed, physicians, patients, and policy makers should be conservative about accepting this new, as-yet not fully tested, and relatively very expensive strategy of using helical CT scanning in screening for lung cancer.