David F. Yankelevitz, MD

In 2010, the long-awaited findings from the National Lung Screening Trial (NLST) revealed that participants who received low-dose helical computed tomography (CT) scans had a 15% to 20% lower risk of dying of lung cancer than participants who received standard chest x-rays. In response, the U.S. Preventive Services Task Force (USPSTF) recommended low-dose CT screening for lung cancer in high-risk populations. Despite proof that screening saves lives, lung cancer screening programs still face an uphill battle. To discuss this important issue The ASCO Post spoke to lung cancer expert, David F. Yankelevitz, MD, a practicing diagnostic radiologist in New York, New York.

Low Rates of Screening Continue

Since the USPSTF’s recommendation on low-dose computed tomography screening for lung cancer, do we have a general sense of how it has been rolled out across the country?

The general sense is that it’s not rolling out well in terms of the number of people who are being screened. An abstract presented at the 2018 ASCO Annual Meeting calculated that only about 2% of the eligible populations are being screened for lung cancer.¹ It’s important to note that the eligible population is a pretty high-risk group. The rough estimates conclude that about 20% of lung cancers are occurring in the group eligible for screening, so in essence, we’re screening just 2% of those 20%, a small fraction of those who develop lung cancer.

Adoption of Screening: ‘A Slow Process’

As a follow-up, do we have a good picture of barriers to adoption of low-dose CT screening?

There are several thoughts about barriers to screening that have been reviewed in papers and presentations. The gentlest of the explanations is that it is simply a slow process, not unlike other early-detection programs, such as routine mammography for certain populations of women. Others conclude that the process is being slowed by a lack of awareness on the part of

I personally think the main barrier is that lung cancer screening has not been widely embraced by the medical community.

— David F. Yankelevitz, MD

Tweet this quote

clinicians and patients. Those factors are certainly plausible, as is the lack of available resources.

I personally think the main barrier is that lung cancer screening has not been widely embraced by the medical community. Moreover, there has been a constant stream of negative press on the screening issue, focusing on the potential harms and downplaying the benefits. In other words, skeptics consider screening a close call—one that doesn’t justify the benefits over the risks and requires a shared decision-making discussion between provider and patient. I just think that physicians are generally not overly enthusiastic about lung cancer screening. Many major societies do not recommend such screening, such as the American Academy of Family Physicians.

Misleading Negative Messages

One of the chief concerns among skeptics of low-dose CT screening is the problem of overdiagnosis. Has that issue been looked at in rigorous studies?

To me, the chief concern is a lack of understanding about the benefits of screening, which are substantial. Critics have distorted that issue, contending that the 20% mortality reduction shown in the NLST only translates into 20% of the screen detected lung cancers being cured. That is a fundamental misrepresentation of the NLST findings, and I personally cannot think of a more serious error in medicine today.

Negative messages are being promulgated: Four out of five people whose lung cancer is detected while in a screening program will die anyway. The NLST did three rounds of screening with 6 years of follow-up. So much of the time, the researchers weren’t even screening, and there were a lot of cancers diagnosed in the screening arm after screening had stopped. You can’t expect screening to have led to cures for those patients, and that is only part of the explanation. So it’s wrong to conflate that conclusion with the NLST findings.

I suggest that perhaps 80% of screened lung cancers can be cured. And if that statement alone were being put out in the medical and lay communities, people would embrace screening in high-risk populations. In my view, patients would want to know their chances of developing lung cancer over the next several years and their chances of being cured if they are screened vs not being screened. If they hear that they have a 20% chance of developing lung cancer and they’ll be cured 4 out of 5 times if screened, vs dying 9 of 10 times if not screened, that will compel many of them to request screening.

However, that message is not getting out. Even the shared decision-making literature is misleading. The charts used often wrongly indicate that 3 out of 1,000 individuals screened will live, plus incorrect information on false-positive results. So, even when screening information is disseminated, it discourages patients about lung cancer screening.

Overdiagnosis: ‘A Malformed Concept’

Could you briefly discuss the issue of overdiagnosis?

The whole concept of overdiagnosis is being poorly defined—not just for lung cancer but for all cancers. Suffice it to say, although people tend to think of overdiagnosis as finding slow-growing indolent cancers that would not become deadly, that is not what the epidemiologists who have defined this concept actually mean. For that epidemiologic definition, which is distinct from being a clinically defined entity, cancer is not a real cancer unless you die of it. So, the person with the aggressive cancer who happens to die of a heart attack or in a car accident, he or she also has an overdiagnosed cancer. In reality, it is a malformed concept.

The benefit of lung cancer screening has been terribly underestimated, and as a result, we are missing a huge opportunity to save thousands of lives.

— David F. Yankelevitz, MD

Tweet this quote

I’m not saying there are not slow-growing cancers that would not need to be treated. However, one clinical aspect that makes lung cancer unique is that we can determine which are the indolent cancers by their radiologic appearance. Moreover, not only can we detect slow-growing cancers, but we also know how to monitor them. And monitoring these cancers, rather than treating them, is now standard care in the field. In addition, this approach is endorsed by many major medical organizations and is part of best practices guidelines. So, malignant nodules can now be monitored and managed annually to see whether they grow. In short, the issue of “overdiagnosis” is far more manageable in lung cancer than other cancers.

Radiation Exposure

Are there other possible harms associated with low-dose CT that raise concerns, such as radiation exposure?

With the doses we use, there is virtually no sound evidence that radiation in screening poses a medical threat. Overtesting has gotten a lot of press, but again, there are no sound data to show any possible harm that would outweigh the benefits of screening for lung cancer in high-risk populations.

Perfect Time for Smoking Cessation Measures

Does low-dose CT screening offer potential ways to address smoking cessation and other preventive measures?

The screening process is a perfect time for clinicians to address smoking cessation measures with their patients. Every time there is a doctor-patient interaction, there should be a reminder about the need to quit smoking. Moreover, there are a lot of other valuable findings on the scans. For instance, a percentage of these people have previously unknown emphysema and coronary artery disease. That information can help their doctors develop proper interventions.

Closing Thoughts

Would you like to share a few closing thoughts on this important issue?

We really need to make sure we present information from randomized trials such as the NLST in a meaningful manner that is clinically relevant to the patient population. The benefit of lung cancer screening has been terribly underestimated, and as a result, we are missing a huge opportunity to save thousands of lives. ■

DISCLOSURE: Dr. Yankelevitz owns stock or has other ownership interests in Accumentra, is a consultant/advisor with Grail, and has patents/royalties/other intellectual properties with Cornell University and General Electric for management of various lung abnormalities including lung nodules.

REFERENCE

1. Pham D, Bhandari S, Oechsli M, et al: Lung cancer screening rates: Data from the lung cancer screening registry. 2018 ASCO Annual Meeting. Abstract 6504. Presented June 1, 2018.