Update on the treatment of non-small-cell lung cancer: focus on the cost-effectiveness of new agents.

BACKGROUND: The incidence of lung cancer and the cost of drug treatment have increased dramatically in the last decade. This article examines the costs of new target agents, such as tyrosine kinase inhibitors (TKIs) and anti-angiogenic drugs.
METHODS: This study uses PubMed research to focus on the topics of lung cancer, economics, and new targeted therapies.
RESULTS: The published papers only addressed TKIs and anti-angiogenic antibodies. For gefitinib, the results favored a clinical-based selection, despite the low number of studies. Erlotinib was studied in second line and as a maintenance treatment (with the studies reaching opposite conclusions in terms of cost-effectiveness). Economic analyses were not in favor of bevacizumab, but the studies on this topic were very heterogeneous.
CONCLUSION: The economic impact of a drug depends on the health care system organization. Future clinical trials must include economic analyses, particularly with TKIs in the first line.

Introduction

Significant progress in the treatment of cancer has been made since the late 1990s, notably with the development of targeted therapies in the first decade of the 21st century. These new treatments have significantly improved the prognosis of some malignancies, including lung cancer, but the cost of treatment has increased in parallel.1 In 2007, the US National Institutes of Health estimated that the direct costs of care for lung cancer patients totaled US$ 90 billion, and this figure is predicted to rise to about US$ 160 billion by 2020.2,3

Cipriano et al reported that initial management costs per lung cancer patient were US$ 6639 during the first year, with a cumulative total cost of US$ 164,768.4 Costs were slightly lower for patients over 65 years of age.

Overall costs have increased by about 22% during the last decade, while the introduction of new agents has raised drug-related costs by 11% for lung cancer patients.5 While chemotherapy (cisplatin-based doublet therapy) has become well standardized, these new drugs have modified the treatment course, leading to longer-term treatment and the need for maintenance therapy. The impact of these new drugs on the overall cost of treatment is far from negligible. The following article examines the cost of new agents used to treat lung cancer, focusing on tyrosine kinase inhibitors (TKIs) and anti-angiogenic agents.

Methods

We performed a PubMed search with the following keywords: “lung cancer,” “costs,” “targeted therapies,” “erlotinib,” bevacizumab,” or “gefitinib”. All documents featuring one of three characteristics (lung cancer, costs and targeted therapies) were collected and analyzed by two of the authors (AV and CC).

Results

Economic analyses of TKIs in lung cancer

Gefitinib

Gefitinib was the first TKI to be approved for the treatment of non-small-cell lung cancer (NSCLC). Previous data showed an average treatment cost per patient of approximately US$ 46,000 during the first two years.6 The impact of gefitinib is difficult to analyze because of the paucity of studies (Table 1).

Table 1

Economic analyses of gefitinib in NSCLC

Author	Line	Type	Main results	Reference
Chouaid et al	3rd	Modeling of a compassionate-use program. French payer’s perspective.	Total costs:€ 39,979 ± € 20,279 (10% of total costs).	7
Thongprasert et al	2nd	Model-based comparison: erlotinib, pemetrexed, and docetaxel. Thai payer’s perspective.	Gefitinib is better than erlotinib and docetaxel.	8
Horgan et al	2nd	Cost-utility analysis of INTEREST trial, gefitinib versus docetaxel.	ICER CA$ 5161; gefitinib preferred.	9
Brown et al	1st	Modeling of first-line gefitinib versus chemotherapy.	ICER £ 35,700 for gefitinib versus doublet therapy.	10
de Lima Lopes et al	1st	Modeling of cost-effectiveness of EGF TKIs versus standard care. Asian payer’s perspective.	US$ 2400; dominant strategy.	11

Abbreviation: ICER, incremental cost effectiveness ratio.

Chouaid et al performed a model-based study of compassionate-use gefitinib therapy in France (between 2002 and 2004), based on data from 106 patients.7 The total cost for each of these patients was € 40,000 ± € 20,729, with gefitinib representing about 10.7% of the overall cost (€ 4241 ± € 1424). However, this study included only highly selected patients.

A second study conducted in Thailand examined the cost-utility of second-line gefitinib for NSCLC.8 The comparators were docetaxel, erlotinib, and pemetrexed. Gefitinib proved to be the most cost-effective second-line treatment. This study adopted the perspective of the Thai health care system, and most of the costs were based on expert estimates.

Horgan et al based their study on the dataset from the INTEREST clinical trial.9 In this cost-utility study based on prospective data, the marginal cost-effectiveness of gefitinib versus docetaxel was CA$ 5161, which was considered acceptable for the North American health care system. Adverse effects and quality of life also favored the use of gefitinib rather than chemotherapy.

Brown et al conducted a study for the UK National Institute for Health and Clinical Excellence (NICE).10 Patients were not selected for EGFR-R mutations. Costs were considered too high for the British system, despite the fact that this was a selected population (IPASS trial). NICE calculated the costs as ranging from £ 25,000 to £ 65,000 per additional quality-adjusted life year (QALY).

The recently published study by de Lima Lopes et al, adopting the perspective of Asian health systems, analyzed first-line gefitinib use in patients with EGF-R mutations in comparison with chemotherapy.11 The results favored gefitinib, as confirmed by sensitivity analyses.

Erlotinib

Secondline treatment

Erlotinib was first validated in this setting (second line treatment). Among the many studies conducted, only one French study took EGF-R mutation status into account.12 The use of targeted therapies did not reduce the overall cost of treatment.4,13,14

Bradbury et al conducted an economic analysis of the BR21 registration trial.15 They showed that marginal cost-effectiveness was close to US$ 100,000 per year of life saved, which was still just acceptable for the Canadian health care system. Studies of patients’ willingness to pay for a portion of their treatment have offered similar results: patients agreed to pay, but only about 5% to 10% of the real cost of these drugs.16,17

Other studies compared erlotinib with chemotherapeutic agents, such as docetaxel and pemetrexed. Carlson et al showed that erlotinib dominated the other two products.18 Lewis et al compared erlotinib with docetaxel in a cost-utility study.19 Although the results were very similar, they tended to favor erlotinib. Furthermore, when compared to best supportive care, docetaxel was a better option than erlotinib, as confirmed in other countries, such as Brazil.20 But NICE’s recommendations were less favorable from the point of view of the UK health care system.21 Erlotinib did not result in additional costs in an Italian study.22 A recent study conducted in Canada showed that, although the effectiveness was similar with second-line docetaxel was not significant in an unselected population,23 erlotinib compared well with supportive care in terms of cost effectiveness.24 Borget et al showed that patient selection based on biological and clinical criteria led to lower costs (€ 5020 and € 5815, respectively) than in an unselected population (Table 2).12

Table 2

Results of the ERMETIC study12

ICER/QALY (euros)	Euros
No selection	43,895
Chemically guided	28,863
Biologically guided	27,209

Abbreviations: ICER, incremental cost-effectiveness ratio; QALY, quality-adjusted life years.

Maintenance treatment

While most studies have focused on pemetrexed (showing little efficacy),25 recent studies following the Saturn clinical trial have examined the possible place of erlotinib in this setting.26

In a cost-minimization study, erlotinib proved to be less costly than pemetrexed.27 When the manufacturer submitted the dossier to NICE, the UK agency redid the calculations and found that erlotinib was not cost-effective for the British health care system.28 The NICE values were approximately £ 50,000 per QALY.

Vergnenègre et al subsequently performed a cost-effectiveness study for patients with wildtype EGF-R when the disease stabilized after the end of first-line treatment.29 The study was conducted in France, Germany, and Italy. Erlotinib was found to have an acceptable cost-effectiveness ratio per QALY in France (€ 39,783), Germany (€ 46,931), and Italy (€ 27,885). At a threshold of € 50,000, erlotinib had a 50% probability of being cost-effective.

Another recently published study of patients with wildtype EGF-R compared the cost-effectiveness of erlotinib maintenance treatment versus best supportive care.24 The results were € 20,711 in the UK and € 25,124 in Germany. The authors concluded that erlotinib maintenance was medically and economically justified.

First-line treatment

There are no published data on first-line erlotinib. In 2009, Carlson et al conducted an exploratory study, showing that a pharmacogenomic test could reduce the cost per QALY.30

Many clinical trials taking EGF-R mutations into account have now been published,31 but economic analyses are still needed. It is very likely that erlotinib will prove to be cost-effective in selected populations.

Anti-angiogenic agents: bevacizumab

Anti-angiogenic therapies have recently been used in patients with lung cancer, but head-to-head comparisons with chemotherapy are rare.32

Bevacizumab is the most extensively studied anti-angiogenic drug, notably in the phase III trial by Sandler et al.33 Published articles on the costs associated with this drug have been analyzed in a general review.34

Among the five most interesting articles on the cost-effectiveness of this drug, two showed that bevacizumab had acceptable cost-effectiveness from the standpoint of German and Italian society,34–36 while the three studies suggested it was not cost-effective. It must be stressed, however, that all these publications were model-based and did not use real clinical trial data. Giuliani et al35 and Ahn et al36 postulated a dose of 7.5 mg/kg, while the others used 15 mg/kg per day. The models all adopted the payer’s viewpoint and not that of society.

Cost analyses should include the overall costs, especially as indirect costs, for patients treated with bevacizumab, could be significantly lower, through earlier return to work.37

Conclusion

New cancer therapeutics are increasingly effective but generate increasingly high costs. Societies must consequently weigh the costs and benefits, using various thresholds (for example, US$ 100,000 to 150,000 in the United States). Numerous studies have been published, but many are modelbased, and their conclusions often differ.5 General reviews are helpful but still fail to provide definitive results. Only economic analyses embedded within independently funded clinical trials can serve to inform decision makers. Many previous studies included unselected populations, but it would probably be better to select patient subgroups in which the benefits are likely to be greatest. Such economic studies are needed to ensure each patient receives the most cost-effective treatment.