| HOME | HELP | CONTACT US | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
EDITORIAL: LUNG CANCER |
The prominent health economist Alan Williams famously said that since death and taxes are the only certain things in life, the goal of health economics should be to postpone death and reduce taxes. It is with this civic-mindedness that Ramsey and Kessler write their provocative Commentary [1]. In it, they point out that the results from a recently released randomized trial [2] show no survival or quality-of-life (QOL) advantage for the regimen carboplatin plus paclitaxel, as compared to cisplatin plus vinorelbine in the treatment of advanced non-small cell lung cancer (NSCLC). Despite the similarities in QOL, toxicities between the two regimens varied. Cisplatin plus vinorelbine produced significantly more neutropenia, nausea, and vomiting, and carboplatin plus paclitaxel produced more sensory neuropathy. Overall toxicity caused 28% of the patients receiving cisplatin plus vinorelbine to terminate therapy compared to 15% of the patients receiving carboplatin plus paclitaxel. Because these toxicity differences did not affect QOL, Ramsey argues that the major cooperative groups should consider adopting the cheaper alternative (cisplatin plus vinorelbine) as their reference standard. To Ramsey's chagrin, the Eastern Cooperative and Southwest Oncology Groups and the majority of American oncologists have done the opposite.
Dr. Ramsey and his colleagues deserve praise for the economic analysis [3] on which their Commentary is based. Their analysis, conducted alongside a randomized Phase III trial [2], was designed to avoid many of the classic pitfalls that often make economic analyses biased, uninformative, or irrelevant. The study was cosponsored by the manufacturers of two of the competing chemotherapeutic agents (vinorelbine and paclitaxel), minimizing the prickly conflict-of-interest issue. The authors also diligently collected their cost data prospectively and monitored costs well beyond the treatment period, encompassing the entire survival history for most of the patients on trial. This approach captured downstream cost differences often ignored in similar economic analyses. The major finding of the analysis was that cancer-related health care costs averaged about $40,000 for patients treated initially with cisplatin plus vinorelbine versus almost $49,000 for those treated with carboplatin plus paclitaxel. Given the annual incidence of cases of NSCLC, most of which are treated at some time in their course with chemotherapy, this is an important finding.
In spite of the quality and relevancy of their work, it will likely have a limited impact on the prescribing patterns of U.S. oncologists in their treatment of advanced NSCLC. First of all, their QOL findings—that patients who receive cisplatin plus vinorelbine have identical QOL compared to those who receive carboplatin plus paclitaxel—have clear limitations. The QOL assessments were performed at weeks 1, 13, and 25 of therapy, just prior to a new cycle of chemotherapy. This choice of timing led to a month interval between the last cisplatin treatment and the QOL assessment. This interval may have led to recall bias and may explain why there was no significant QOL difference detected between the regimens despite the fact that only 65% and 78% of the planned doses of vinorelbine and cisplatin were infused, respectively compared to 90% of the intended carboplatin plus placlitaxel. Because chemotherapy for advanced NSCLC is by nature palliative, the study can be criticized for not paying greater attention to measuring QOL at additional timepoints.
Two other technical issues deserve noting. First, the authors' cost calculations assumed a stable price of drugs. With generic paclitaxel now entering the market and carboplatin going off-patent in less than two years, generic equivalents of these drugs will emerge and substantially drive down the cost of the carboplatin plus paclitaxel regimen. Second, the study ignored indirect costs to the patients (such as patient transportation to and from the clinics and effects of treatment on workplace productivity) and thus took only the insurer's perspectives. Because cisplatin plus vinorelbine is given on an every week schedule compared to every three weeks for carboplatin plus paclitaxel, indirect cost differences may be significant and should have been included.
In addition to these technical issues, there are ethical reasons to avoid prescribing chemotherapy based on cost. The ethical obligation of oncologists to address the needs of each individual patient is greater than their societal duty to control costs. This view is articulated well by Erich Loewy who writes that "a physician who changes his or her way of practicing medicine based on cost rather than purely medical considerations has indeed embarked on the ‘slippery slope’ of compromised ethics and waffled priorities [4]." Because of these ethical concerns, most physicians and ethicists agree that economic concerns should be addressed at a macro level and should not directly influence the practice of the individual oncologist. No patient should be denied the care that the physician deems best among available alternatives. It is society's responsibility to determine what is available.
The last consideration that will limit the impact of Ramsey's work is a structural one. Unlike most developed countries, the U.S. lacks a centralized health care system where treatment guidelines are formulated based on budgetary constraints. Nor do we have in place guiding principles for incorporating considerations of ethics and individual preferences into the discussion of health economics and the practice of oncology in an ethical manner. The development of these principles is critical because economic considerations germane to oncology may differ from those in other areas of medicine. Choosing between two nontoxic lipid-lowering agents is fundamentally less involved than making, for example, a well-informed trade-off between money and nausea and vomiting.
Death and taxes are inescapable, but they are not immutable. As such, Scott Ramsey is correct to highlight in his Commentary the issue of double-digit inflation in the cost of cancer care. As novel agents are introduced that lengthen survival but do not cure, cancer costs will rise even more steeply. The recently approved drug for non-Hodgkin lymphoma, ibritumomab tiuxetan (Zevalin), is illustrative, as it will cost approximately $20,000 for a single administration with still uncertain long-term benefit for many of the patients receiving this drug late in the course of their disease. Ramsey needs to be commended for his attempt to apply the emerging tools of health economics to this issue. It is clear that he seeks to introduce cost cutting measures in a way that will produce equivalent survival and maintain quality of life. But without accepted guidelines to direct the adoption of his work into clinical practice, in the setting of differences in toxicity and market forces that are likely to reduce cost difference between regimens, we cannot at this time endorse choosing chemotherapy for NSCLC based solely on cost. We do hope that Ramsey's Commentary encourages a national effort by the oncology community to address the rising costs of cancer care and how to apply health economics to this problem with circumspection. Indeed, the Mission and Beliefs of The Oncologist encourage our community to "resist the trend to capitulate our responsibilities in disease management to payers, gatekeepers, and hospital administrators." If we do not take on this responsibility, others will certainly try to do it for us.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | CONTACT US | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| THE ONCOLOGIST | STEM CELLS | CME | ALPHAMED PRESS JOURNALS |
