Screening for lung cancer: do risks outweigh benefits?

By Diana Zuckerman, PhD

Medicare and U.S. insurance companies will pay for lung cancer screening with low-dose CT scans (LDCT) for former heavy smokers. However, a careful review of the research indicates that this screening is unlikely to save the lives of Medicare patients. It may help save the lives of some patients ages 55-65, but may also harm some.

A major study called the National Lung Cancer Screening Trial (NLST) found fewer deaths from lung cancer in patients ages 55-74 who had smoked for at least 30 pack-years (1 pack of cigarettes per day for 30 years, or 2 packs a day for 15 years, etc) and were screened with LDCT rather than chest x-ray (CXR). Patients were 20% less likely to have died within 6.5 years of screening and 16% less likely within 7.5 years. However, the decrease in deaths from all causes (not just lung cancer) was not as impressive: 6.7% after 6.5 years. That means that these former and current smokers were dying from other diseases, some of which might also have been related to their smoking, such as emphysema, breast cancer, and COPD.

Most Medicare patients are over 65 and many are over 74, so the study above can’t predict what is likely to happen to older patients. What evidence is there that Medicare patients with a 30-pack-year habit would benefit from LDCT screening? The major issues are:

1) Would screening heavy smoker and former smokers over 65 detect cancers early enough to save lives or extend lives?
2) Would the screened patients who are diagnosed with lung cancer be healthy enough to undergo lung surgery and benefit from it?
3) Would the screened patients who are diagnosed with lung cancer be likely to die from something other than lung cancer, making lung cancer screening irrelevant to their survival?

The US Preventive Services Task Force recommended LDCT screening for people ages 55-80 who currently smoke or had quit within the last 15 years and who have a history of at least 30 pack-years of smoking. However, those experts warn that “screening cannot prevent most lung cancer deaths.” In fact, even if men and women who are screened with LDCT are diagnosed earlier and live longer with the diagnosis, they may die at the same age they would have without screening. In other words, a patient diagnosed after screening at age 65 might live to be 68 and if they were not screened they might have been diagnosed at 67 but still died at the age of 68. Even patients diagnosed with the earliest (stage 1) lung cancer will not necessarily be treated successfully.

The answer to #2 above is also difficult to know. The patients studied in the NLST were half as likely to have emphysema, diabetes, or heart disease as most smokers of the same age. Since typical Medicare patients eligible for LDCT screening would have more health problems, they would be less likely to be healthy enough to undergo surgery or other cancer treatment. And, even fewer than 6.7% would be likely to benefit by living longer.
Another major concern is that screening has health risks. LDCT screening is very inaccurate: 96% of the patients who are told their results show they may have lung cancer, do not actually have lung cancer. That is called a false positive result. Those patients will need more tests to find out whether or not they have lung cancer. Some of those additional tests, such as lung biopsies, can be dangerous or even deadly.

In addition, using LDCT every year to screen for lung cancer will expose patients to radiation that can increase their risk of lung cancer.

The Medicare Evidence Development and Coverage Advisory Committee (MEDCAC), made up of experts from across the country, recommended that Medicare not pay for LDCT screening, saying they were not confident that the benefits would outweigh the risks.
After heavy lobbying, Medicare decided to pay for LDCT screening anyway. Only time will tell how many lives will be saved – or lost – as a result.

For more information, see: D. Zuckerman, Screening for Lung Cancer: To Be or Not to Be Covered by Medicare? Thoracic Imaging, Vol 30, No. 1, 2015.