A Lung Cancer Story Problem’
Abstract & Commentary
Synopsis: Screening for lung cancer with helical CT has high costs and uncertain benefits.
Source: Mahadevia PJ, et al. JAMA. 2003;298:313-322.
In 2003 the American Cancer Society estimates that there will be 171,900 new cases of and 157,200 deaths from lung and bronchial cancer in the United States.1 While this represents about 13% of all cancer diagnoses, it accounts for about 28% of all cancer deaths. There is no cancer that kills as many men and women as lung cancer. Early detection, presumably before it has metastasized, remains the Holy Grail of lung cancer treatment. Mahadevia and colleagues performed a decision and cost-effectiveness analysis of screening for lung cancer with helical computed tomography (CT), using data from the Surveillance, Epidemiology, and End Results (SEER) national cancer database.
To do this kind of analysis, they accounted for all possible outcomes and assigned probabilities to them. If there were no published data for a particular outcome, they made an educated guess. Finally, they made certain assumptions: a hypothetical study population of 100,000, aged 60 years old, 55% male, heavy smokers (> 20 pack-years), eligible for lung resection. This group was divided into 3, based on smoking status: current, quitting at the start of screening, and quitting 5 years before screening. These 3 groups were divided into screened and nonscreened. The scenarios were run with annual screening until age 80 years and follow-up to age 100 years.
The SEER database contains probabilities of what happens to smokers: alive without apparent cancer, developing and dying of lung cancer, and dying of other causes. They adjusted these probabilities based on the 3 smoking-status groups. All patients developing signs or symptoms of lung cancer underwent invasive testing. Those diagnosed with lung cancer were assigned to localized-stage non-small-cell lung cancer (NSCLC), advanced-stage NSCLC, or small-cell lung cancer (SCLC). Depending on the diagnosis, the patients could receive surgery, radiation therapy, and/or chemotherapy. Helical CT sometimes identifies indeterminate nodules that need to be followed or biopsied. Using previously published studies of lung cancer screening by helical CT, Mahadevia and colleagues obtained probabilities for these events. One interesting aspect of helical CT screening is that the cancers it picks up are usually peripheral. It tends to miss cancers hidden in endobronchial locations. These cancers are histologically different from peripheral cancers (squamous cell carcinoma vs adenocarcinoma, respectively). As might be expected, the cancers detected by helical CT are more likely to be not as advanced in clinical stage. These differences were factored in. All costs (facility and professional fees, additional monitoring, surgery, diagnostic tests, etc) were calibrated to 2001 US dollars. Quality-of-life measurements (including anxiety from testing indeterminate nodules, surgery, radiation, chemotherapy, etc) were calculated.
There were 4168 lung cancer deaths per 100,000 unscreened patients and 3615 per 100,000 screened patients (553 fewer deaths, 13% relative mortality reduction). The screened group underwent 1186 invasive tests for benign lesions. The average incremental cost for screening was $116,300 per quality-adjusted life-year (QALY). The incremental cost-effectiveness increased as the risk of cancer decreased; the cost for former smokers was $2,322,700 per QALY.
Comment by Allan J. Wilke, MD
Earlier attempts to screen for lung cancer with periodic chest x-rays and sputum cytology failed to reduce mortality.2 We will have to wait several more years before the results of the National Cancer Institute’s National Lung Screening Trial (NLST), a study comparing spiral CT and chest x-ray, will be available.3 Enrollment began in September 2000, and patients will be followed until 2009. Until then, this is the best information we are likely to get. Like any simulation (Pentagon war games or The Sims®), the results are only as good as the assumptions. The most basic assumption is that early detection will reduce mortality by "catching" cancer before it has metastasized. We can only hope that NLST will demonstrate that. If it does, then we will have to ask ourselves if we, as a society, are ready to allocate already scarce health care resources to widespread screening. It is ironic that screening was most cost-effective for the group most likely not to benefit from it: heavy smokers who haven’t quit.
Your patients are being bombarded with direct-to-consumer advertising about this technology. If you enter "lung cancer screening" into your favorite Internet search engine, you will get many links that will take you to commercial sites that offer helical CT services—for a fee or upon referral from your primary care physician. Based on this study, I won’t be ordering screening helical scans any time soon. On the other hand, I’m as big a fan of the free enterprise system as the next guy. On that basis, my advice to patients who request lung cancer screening by helical CT is "if you gotta know, and you’ve got the dough, then you pay the tow!" Since cigarette smoking causes almost all lung cancer, I will also advise them to save their money and quit their habit.
Dr. Wilke is Assistant
Professor of Family Medicine at the Medical College of Ohio in Toledo, OH.
References
1. Cancer Facts and Figures 2003. American Cancer Society. p. 4.
2. Fontana RS. Cancer. 2000;89(Suppl 11):2352-2355.
3. National Cancer Institute. http://www.cancer.gov/NLST. Accessed February 11, 2003.
Screening for lung cancer with helical CT has high costs and uncertain benefits.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content