Low-Dose CT Lung Cancer Screening Still Debated, Despite Evidence

Ingrid Hein

June 18, 2020

Despite mounting evidence that low-dose CT screening reduces lung cancer mortality in people at high risk, the uptake of screening in the United States has been slow, and some researchers caution that the risks involved need to be better understood.

It has been almost 10 years since the landmark National Lung Screening Trial (NLST) provided the scientific evidence used by the United States Preventive Services Task Force (USPSTF) to recommend annual screening for adults 55 to 80 years of age who have a 30 pack-year smoking history and currently smoke or have quit in the previous 15 years.

But just 4.2% of Americans who qualified for screening in 2018 were tested, according to an American Lung Association (ALA) report. If everyone at high risk had been tested, 48,000 American lives could have been saved.

Now that we have consistent data from the NELSON confirmatory trial, can we finally believe NLST?

Final results from the NELSON trial, published earlier this year, support those from NLST.

Mortality was 24% lower with low-dose CT screening than with no screening in the NELSON cohort, which consisted of 13,195 men and 2594 women at high risk for lung cancer because they were current or former smokers.

"With the NELSON results, the efficacy of low-dose CT screening for lung cancer is confirmed," wrote the authors of an editorial accompanying the NELSON results. "Our job is no longer to assess whether low-dose CT screening for lung cancer works: it does. Our job is to identify the target population in which it will be acceptable and cost-effective."

That sentiment is echoed by Michael Gould, MD, from Kaiser Permanente Southern California.

"Lo and behold, we have confirmation of NLST results from NELSON," Gould told Medscape Medical News. "Now that we have consistent data from the NELSON confirmatory trial, can we finally believe NLST?"

Even though NELSON confirms the benefits of screening in clinical trials, many questions remain about how lung cancer screening translates into everyday practice, said Gould, who had been scheduled to discuss the trials and the state of lung screening at the American Thoracic Society 2020 International Conference, which will now run virtually in August.

For starters, the target population needs more scrutiny. Research has shown that outside of clinical trials, the harms of screening can sometimes outweigh the benefits.

In 2018, the rate of overdiagnosis was shown to be 67.2% in the Danish Lung Cancer Screening Trial (DLCST).

And 56% of people screened with low-dose CT had false-positive results that required follow-up testing and procedures, according to a 2017 study of current and former heavy smokers. That rate is more than double the 18.5% false-positive rate in NLST.

"Only 20% of NLST participants were over age 65," Gould said. "The NELSON cohort was younger."

And although the USPSTF recommends lung screening in high-risk people, "there were some in the NLST cohort whose risk was not particularly high." Others in the trial, he said, had a high risk, but some of those had one or more comorbid conditions, "so the risk was unbalanced."

"Risk is more complicated than simply saying that anyone who meets the NLST criteria should get scanned," he added.

Weighing risks and benefits needs to be done on a patient-by-patient basis, Gould said. "Do they have the ability to tolerate surgery? What's important to them? We can't just say, 'you have a 30-pack-year history, go get a test'."

Often, he said, it's the people who have the most to gain from screening who are also at highest risk from biopsies and surgical and nonsurgical treatments because of comorbidities.

The NLST population might also have cast a wider net for those eligible for screening; NELSON had a lower threshold for amount smoked (30 vs 15 pack-years). "NLST points to scanning a bigger population and lighter smokers," Gould said.

Psychological Risks of Screening

Neither the NLST nor NELSON reported relevant psychological aspects of harm from CT screening for lung cancer, two researchers reported in a letter responding to the NELSON findings.

The trial-participation request letters, which were sent to 606,409 people in the general population, "in order to identify 15,792 persons (2.6%) who were eligible to participate, may have caused fear," wrote Jes Lindholt, MD, DMSc, and Rikke Søgaard, PhD, from Odense University Hospital in Denmark.

"That raises the question: Do people want to be screened? I can't understand why the US and Britain consider it so definitive to start a screening program," Lindholt told Medscape Medical News.

In addition to a psychological cost, he questioned the financial cost–benefit ratio of a screening program. "What strikes me is that they haven't done any cost analysis on any of these randomized trials."

"Of the 203 men who got the diagnosis of lung cancer, 160 (78.8%) died from lung cancer. Whether screening actually improved or prolonged their remaining lifetime should be considered," Lindholt and Søgaard wrote.

Challenges of Implementation

Despite the extensive trials, there are still questions about how to implement screening in the real world. "Did NLST select patients who were, on average, healthier and less likely to have complications?" Gould asked.

Everyday practice might not find the same favorable outcomes as NLST, he suggested. "Can the results of the NTLST be replicated in real-world settings?"

"Not yet," he said. Hospitals and health systems are struggling to implement screening.

Follow-up and tracking are not where they should be. General practitioners don't have the same resources as the NLST researchers had, he explained. They were able to remind patients to come back for another test and call them with the results, all under the umbrella of implementation, "and they're still not on target."

Getting people scanned is key, said Michael Barry, MD, from Massachusetts General Hospital in Boston, who is a current member of the USPSTF and is working on new lung cancer screening recommendations to be published this summer.

"We have an implementation problem," he said. "The heavier smokers are being way underscreened."

People need to have more information to review the pros and cons of screening, Barry said. "We've got large trials that show that benefits outweigh the harms, but we could benefit from implementation research. This is an issue for many screening tasks," he said.

Eight million Americans meet the eligibility requirements for lung cancer screening with low-dose CT, according to a 2019 report from the American College of Radiology.

Screening tests are covered by Medicare but getting people to the clinic has not been easy. In 2018, Saved by the Scan, a big-budget national advertising campaign launched by the ALA, featured ex-smokers who survived lung cancer because of early detection with a low-dose CT scan, as reported by Medscape Medical News.

And many people being scanned are not part of the USPSTF target group. In 2017, lung cancer screening was reported "by 12.5% of smokers who met USPSTF criteria and 7.9% of smokers aged 55–80 years who did not meet USPSTF criteria," according to a recent analysis of data from the Behavioral Risk Factor Surveillance System published by the Centers for Disease Control and Prevention (CDC).

The CDC report concludes that some people are being screened without needing screening, and that "avoidance of screening inconsistent with USPSTF criteria could reduce the potential for harms such as over diagnosis and overtreatment."

Gould said he agrees that this factor needs to be looked at. "There is underutilization in those who need screening, and maybe overscreening in those who aren't at risk."

There are also epidemiologic data that show that black Americans are at higher risk at a younger age for the same level of smoking. "So should there be a lower threshold for smoking and lower age, particularly in the African American population?" Gould asked.

The NELSON trial had significant results in a population younger than that in NLST, he pointed out. "That needs to be considered."

Smokers Dismiss Medical Advice

People in the high-risk group need to better understand the benefits of screening, said Christine D. Berg, MD, an NLST researcher from the National Cancer Institute.

"We know the uptake of lung cancer screening has been slow," she said.

She described encouraging her neighbor, a heavy smoker, to get screened. "But she said she didn't want to know if she had lung cancer, so she didn't go."

"Now she's dead," Berg continued. Unfortunately, "what we see is that those who continue to smoke, and smoke heavily, are not likely to heed medical advice."

The fear of finding out you have lung cancer needs to be overcome, she said. Smokers need to understand that they can add a decade to their lives if lung cancer is detected early.

Some places in the United States have better screening rates than others. "We see a lot of variation from state to state," she said. For instance, in Massachusetts, 12.3% of high-risk people have been screened; in Nevada, the rate is just 0.5%.

There are many reasons for that. First, there are logistics. Screening covered by Medicare must be done in a certified center "with good equipment and that can track results," Berg said. That might be one hurdle. But the greater hurdle is the patients themselves.

I think the probability of false-positives and problems from biopsy have changed dramatically over the last 10 years.

There are studies that point to risks associated with invasive procedures, such as biopsy after screening, which can lead to complications, even when no cancer is found. "My answer to that is, if you need a biopsy, check the data. The Society of Thoracic Surgeons has a database of all the complications, and it's publicly accessible. You can find hospitals in your region that report data," she explained, and "that have highest volume and lowest complication rates."

Second, imaging has improved since the NLST trial. "We have a better ability to estimate cancer in the nodules we find," Berg explained. Nodules that previously needed a biopsy to confirm malignancy can now be assessed with AI and machine learning.

"I think the probability of false-positives and problems from biopsy have changed dramatically over last 10 years," she said.

And we are catching more lung cancer earlier and saving lives. Overall, early detection is increasing and late-stage detection is decreasing. "We're bending the curve, making progress," she said.

In 2019, the 5-year survival rate for lung cancer was 21.7%, up from 17.2% a decade earlier, according to the ALA. Much of that is because of early diagnosis, when the disease is still curable, which could be related to increased screening.

"NELSON showed benefit to CT screening, and is useful in helping convince some of the skeptics," Berg said.

Diagnosis is also improving with new technologies. Electronic health records can be scanned to identify patients at increased risk, and patient portals can send reminders, notifications, and other educational information to encourage patients to discuss options with their doctor, which could improve the national lung cancer prognosis, Gould said.

At the end of the day, it still comes down to the patient and doctor having a conversation about the risks and benefits.

"But we have to get to that point," Gould said. "We need to continue to develop tools to facilitate that conversation. It's complicated, and there's a lot of information to weigh."

"We're still working out how to do that," he added.

Barry, Gould, and Berg have disclosed no relevant financial relationships.

American Thoracic Society (ATS) 2020 International Conference.

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.