Lung cancer is the primary cause of cancer-related mortality in the United States, accounting for almost 30% of all cancer-related deaths.[1] It is estimated that more than 177,000 cases of lung cancer were diagnosed in 1996.[1] Direct medical expenditures for lung cancer reached some $5 billion in 1996 and represented about 15% of all cancer-related costs.[1] In short, lung cancer presents a substantial challenge to clinicians and results in a significant human and economic burden.

Squamous cell, large cell, and adenocarcinomas of the lung comprise the class of tumors commonly referred to as non-small-cell lung carcinomas (NSCLC). Collectively, they represent more than 75% of all lung carcinomas.[2] Average 5-year survival rates for non-small-cell lung cancer are about 13%,[1] but prognosis is highly dependent on stage of disease. Five-year survival rates average about 5% for persons with stage IIIb disease and less than 2% for those with stage IV disease.[3]

Treatment Options and Dilemmas

Therapeutic options for persons with non-small-cell lung cancer are limited. Surgical excision of the tumor is the treatment of choice, and persons with stage I disease have a better than 70% survival rate at 5 years.[3] Yet, only 30% of patients are operable at presentation.[3] For individuals with unresectable locally or regionally advanced disease, radiotherapy, alone or in combination with chemotherapy, is the preferred option. For those with metastatic disease, the goals of therapy include palliation and extension of survival. Patients with metastatic disease typically receive supportive care with or without chemotherapy. However, controversy exists over the appropriate role of chemotherapy in this setting.

Recent studies comparing radiotherapy plus chemotherapy vs radiotherapy alone in patients with advanced disease but no clinically evident metastases have been inconclusive.[4-10] A meta-analysis of studies evaluating the addition of a cisplatin(Drug information on cisplatin) (Platinol)-based regimen to radiotherapy alone suggested a 13% reduction in the risk of death (hazard ratio, .87; 95% confidence interval [CI], .79 to .96).[11] This result corresponds to an absolute survival benefit of 2% at 5 years.[11]

Evidence of clinical benefit is also mixed when chemotherapy is compared to best supportive care (BSC) for stage IV patients.[12-19] A general trend favoring the use of chemotherapy has been reported, but only a few studies establish a statistically significant survival difference. In part, the equivocal results are related to the small sample sizes of the studies, [20] although meta-analyses support the finding of a modest benefit.[11,21-23]

Considering the dismal prognosis of these patients, some clinicians believe that even modest survival gains are important and should be aggressively pursued. Others disagree, maintaining that the minimal survival and palliative benefits from the use of these agents have little merit in clinical practice, especially in view of their associated toxicity and cost. These variations in belief and practice concerning the use of chemotherapy in inoperable non-small-cell lung cancer have been demonstrated by several studies.[24-28]

The purpose of this paper is to recount existing data on the economic consequences of chemotherapy in patients with inoperable non-small-cell lung cancer as they relate to the clinical risks and benefits of treatment. Data are presented that describe the cost-effectiveness of chemotherapy, and although chemotherapy’s effect on quality of life is not a primary focus, this review addresses the importance of preferences in prioritizing benefits vs costs. We conclude by commenting on the role of chemotherapy in non-small-cell lung cancer and by identifying the limitations of analyses that should be taken into account by future studies.

Methods

MEDLINE (1980-present), PDQ (September 1997), Cancerlit (1989-present), EMBASE (1989-present), and the Nursing & Allied Health (1982-present) databaseS were searched in September 1997 for studies published in English through the use of the following key terms: cost, cost-effectiveness, chemotherapy, and non-small-cell lung cancer. The reference list of relevant articles was also scrutinized, and several authors of economic analyses of therapy for non-small-cell lung cancer were queried about the existence of other published or unpublished studies. Studies that included chemotherapy but focused primarily on the role of colony-stimulating factors or antiemetics were excluded. Those evaluating patients with lung cancer, instead of non-small-cell lung cancer, were also excluded.

Various approaches to economic evaluation are possible.[29] Studies of chemotherapy in non-small-cell lung cancer have generally employed three types of analyses: costing, cost-minimization, and cost-effectiveness. Costing analysis describes the costs of providing a particular intervention and expresses the results in monetary units.

Cost-minimization analysis compares the costs of competing strategies that achieve the same clinical outcomes and selects the approach with the lowest cost. This method is limited, however, since different treatments seldom produce truly equivalent outcomes.

Cost-effectiveness analysis overcomes this limitation by simultaneously considering the comparative costs and outcomes of two or more treatment strategies. A ratio of cost to effectiveness is estimated. (The ratio has been defined as the incremental price of obtaining a unit of health effect from a given health intervention when compared to an alternative intervention. Results are reported as monetary unit per outcome gained.) Higher priority is given to interventions with lower cost-effectiveness ratios because they offer the maximum aggregate health benefit for a particular expenditure. Although any clinically measurable end point can be used, most analyses use years of life gained. We reviewed the economic evaluations for incremental cost-effectiveness ratios using life-years or quality-adjusted life-years (QALYs) gained.

Summary of Study Results

Seventeen studies were identified. Five were excluded from this review. Of the five, three were published only in abstract form,[30-32] and a fourth that considered the costs of using fazarabine for metastatic non-small-cell lung cancer was excluded because the drug had no demonstrable activity.[33] Rees reported a “notional patient benefit year cost” for two non-small-cell lung cancer chemotherapy regimens,[34] but we excluded this fifth study because the method of economic analysis used in the study is not a standard approach.

Table 1 provides a summary of the key elements of the 12 included studies.[35-46] A wide variety of economic evaluation approaches were utilized, and most analyses were reported within the last few years. Earlier analyses compared certain regimens to best supportive care or to other available products. Many of the more recent evaluations projected the clinical and economic impact of gemcitabine(Drug information on gemcitabine) (Gemzar) prior to its approval in the United States from data accumulated in phase II trials.

All of the analyses reported direct medical costs, with most taking the perspective of a third-party payor, such as the government or an insurance company. Only six reported cost-effectiveness ratios with life-years or QALYs gained as the outcome measure. In most of the identified studies, trial data were supplemented with data from other sources, such as surveys, published guidelines, or chart reviews, in order to produce the economic evaluation.

To provide a thorough discussion of the important findings and issues raised by these studies, we gave detailed consideration to four of the most comprehensive cost-effectiveness analyses. For three of these studies, researchers based their analyses on the results of individual clinical trials.