Lung Cancer Screening Debate Continues to Draw Breath


Lung Cancer Screening Debate Continues--the Latest Update  from the Medical Imaging Talk BlogThe last few months have seen substantial drama related to the potential introduction of coverage for CT-based lung cancer screening for high-risk Medicare patients. This blog post is a quick review of the history and current state of the debate.

It should be noted that “high-risk patients” in this instance means “asymptomatic adults aged 55 to 80 years who have a 30 pack per year smoking history and currently smoke or have quit smoking within the past 15 years.” This is a small percentage of the approximately 50 million Medicare beneficiaries. But it is a group that has, for obvious reasons, a high mortality rate with respect to lung cancer.

First Among Cancers

The aim, of course, is to use medical imaging to detect lung cancer in the early stages when it is most treatable and thus save lives. By body count, lung cancer is the most deadly cancer out there right now. Currently, almost 160,000 people in the United States die each year from it. It causes more deaths each year than from the next three types of cancer put together.

So lung cancer screening seems like a no-brainer. But, like most things in life, it is not so simple.

Enter Low-Dose Computed Tomography (LDCT)

The debate, as with all screening questions, is primarily about whether the evidence supports the idea that the particular screening methodology and technology in combination reliably detects cancers (i.e., has useful true positive and true negative rates). Furthermore, it must do so better than alternative screening approaches (in this case sputum tests or projection chest X-rays).

First Do No Harm

Also of concern is whether a particular screening program does substantially more good than harm. Because we are talking about a condition that currently kills 16 percent of those diagnosed with it, “harm” seems like an odd thing to be worried about. But as experience has shown us in other areas, false positives can lead to medical interventions that have their own consequences and risks. While lung surgery is a completely reasonable risk-benefit choice if you actually have lung cancer, it is not something you want to inflict on a patient who is actually healthy.

And, while great strides have been made in reducing radiation exposure from CT exams (it is called “low dose CT” for a reason), any amount of radiation exposure to the lungs runs a very small, but non-zero risk of causing the same cancer that we are trying to detect.


To settle the clinical questions, the National Lung Screening Trial (NLST) was conducted in 2011. It concluded that the program would reduce mortality by at least 20 percent (actual success rates may be much higher). Further, they concluded as many as 20,000 Americans could be saved each year if a CT lung cancer screening program were fully implemented.

The United States Preventive Services Task Force’s (USPSTF) reviewed the results of the trial and agreed, giving CT lung cancer screening a relatively high “B” rating. Amongst other things, this triggered the process for deciding Medicare coverage.

Surprise Upset

At the end of May, the Medicare Evidence Development and Coverage Advisory Committee (MEDCAC) convened a meeting to review whether Medicare should cover CT lung cancer screening based on the USPSTF’s recommendation and other evidence. To the shock and surprise of many, given the seemingly strong evidence, MEDCAC voted against recommending Medicare coverage. While this does not decide the matter, it will no doubt have a significant influence on the final decision by CMS.

It is Not About the Money (Really)

Implementing CT lung cancer screening is potentially a very expensive undertaking for Medicare – some estimates are $9.3 billion if you include the cost of treatment of the cancers found. Therefore, some people in the imaging community have raised concerns about whether the rejection by MEDCAC was based even partially on cost concerns rather than the science. CMS maintains that this is not the case.

Talking about potential program cost gets into the uncomfortable question of how much is the right amount of money to spend to save a life. I prefer to interpret this as, “How can we save the most lives with a given amount of money?” Since we all die eventually and effective medicine just helps us to forestall this, the comparative metric is usually phrased as cost-per-life-year. Even this calculation has to be done carefully to avoid lead-time-bias.

An actuarial analysis published in 2012 made a strong case in favor of the costs-effectiveness of a focused CT lung screening program. It found that the cost-per-life-year was significantly better than screening programs for other forms of cancer already in place. Notably, some private insurers have chosen to cover CT lung cancer screening, having already accepted the evidence of the studies and analysis.

The Community Hospital Question

One of the MEDCAC panel members questioned whether community hospitals had sufficient expertise to perform the screening and delivery of follow-up interventions. Limiting screening to “centers of excellence” was floated as an alternative, despite the access difficulties this would present. This comment angered a number of community hospital radiologists who either participated in the NLST or already provide the service outside of Medicare. Others have pointed out that this could be dealt with through national accreditation measures similar to those that exist for mammography.

Meanwhile in Washington, DC…

In early June, the debate reached the floors of the Senate and Capitol. In a rare bipartisan alignment, 45 Senators and 134 Representatives signed letters requesting that CMS reach a decision on the matter quickly. The House letter, in particular, set a goal deadline of February 8, 2015 for a final decision on coverage.

Coverage Outside of Medicare

While CMS considers the question relative to Medicare, low dose CT lung cancer screening is gaining ground elsewhere. Because of the USPSTF’s “B” rating, in the US all private insurers will be required to cover it in 2015 under the rules of the Affordable Care Act. As mentioned above, some private insurers have already started coverage.

Canadian provinces and health care systems elsewhere in the world have produced their own positive recommendations based on the NLST research and that of their own scientific bodies.

The Future?

For now we are all waiting for the final CMS ruling. Of course this debate could all be moot if a non-imaging approach is discovered in the near future to give comparable or better sensitivity to CT screening. Some researchers are betting on detection of volatile organic compounds (VOCs) – basically smelling the tell-tale signs of cancer on a patient’s breath. Devices for this already exist commercially for laboratory use but need to be proven in the real world.

Do you have an opinion on this debate? Please feel free to leave me a comment.

Learn about McKesson’s enterprise medical imaging solutions, including QICS for Exam Protocols, which helps standardize the management of standardized diagnostic imaging exam protocols.

Leave a Reply