New lung cancer screening guideline comes with many caveats

The US Preventive Services Task Force, an independent expert panel that makes evidence-based recommendations about clinical preventive services, recently updated their lung cancer screening (LCS) recommendation for the first time since 2013.

The new guideline recommends annual CT screening for lung cancer for adults aged 50 to 80 years who have a 20 pack-year smoking history (smoking the equivalent of one pack a day for 20 years) and currently smoke or have quit within the past 15 years. Previously the USPSTF had recommended starting screening at age 55 for adults with a 30 pack-year smoking history. This change makes 6.4 million more adults in the US eligible for screening, an 81% increase from the previous guideline.

The new recommendation was hailed by some experts as a policy that will save lives, improve access to screening, and reduce disparities in cancer deaths. Others warn that expanding screening without targeted outreach, quality control, and shared decision making will lead to overdiagnosis and overtreatment without improving health equity. This is a complex issue and one worth exploring, so let’s take a look.

Screening benefits and harms

The USPSTF gave this recommendation a B grade, meaning they “conclude moderate certainty that annual screening for lung cancer with CT has a moderate net benefit.” Not exactly a rave review. How did the USPSTF reach this conclusion? They looked at six randomized controlled trials and used modeling to compare the benefits and harms for screening at different ages. They found that the number cancer deaths prevented with screening was “substantial” while the harms of screening were “moderate,” hence the moderate net benefit.

How much is that moderate benefit? The USPSTF asserts that expanding screening to age 50 and 20 pack-years would prevent about one more death per 1000 people screened (122 per 100,000) compared to its previous recommendation in 2013. They calculated this using both trial data and model simulation.

There are two studies of LCS that were large enough to detect significant cancer deaths: the National Lung Screening Trial (NLST, n = 53 454) and Nederlands-Leuvens Longkanker Screenings Onderzoek (NELSON, n = 15 792). The NLST trial included patients age 55-80 with a 30-pack year smoking history. In the trial out of 1000 people screened annually for three years, 3 lung cancer deaths were avoided over 6.5 years and 4 deaths were avoided overall. To put this in context, among the 1000 people screened, 18 cancer deaths were not avoided.

The NELSON trial included a broader group of patients, age 50-80 and with 20-pack year smoking hospital. In this trial, out of 1000 people screened, 7-8 lung cancer deaths were prevented after ten years, but there was no significant overall mortality benefit.

A person at high risk who undergoes lung cancer screening is five times more likely to have an unnecessary invasive procedure and 83 times more likely to have a false positive, than they are to avoid dying of lung cancer.

The USPSTF cautions that there are harms to screening as well: For every 1000 people screened, 250 will have a false positive result and 17 will have an invasive procedure unnecessarily, according to the NLST results. Other studies showed that as many as 400 in 1000 screened could have follow-up tests and procedures from an incidental finding — an abnormality on a different body part picked up by the CT scan. The USPSTF writes that as many as two-thirds of lung cancers found and treated in the trials may have been overdiagnosed–that is, the cancer that was found never would have harmed the patient.

And that’s not counting the financial cost for patients. While preventive screening scans are covered under the Affordable Care Act, follow-up procedures such as lung biopsies, PET scans, and lung surgeries can cost thousands of dollars. And that’s not including complications; according to a 2019 study in JAMA Internal Medicine, the financial cost of complications from invasive lung procedures range from $6,320 for minor complications to $56,845 for major complications on average. Patients with high-deductible plans may find themselves with bills amounting to tens of thousands of dollars because of a screening they believed was covered by insurance.

The USPSTF uses the best evidence and models available, but even these tools can’t give clinicians and patients a definitive right answer. In an accompanying editorial in JAMA Oncology, Dr. Mayuko Ito Fukunaga, professor pulmonology at the University of Massachusetts Medical School, and colleagues explain why defining the “right” population for screening is so difficult:

We recognize the necessity for the USPSTF and payers to select hard cut points for determining eligibility. However, as the CISNET results dramatically illustrate, the benefits and harms of LCS exist along a spectrum. The models can help guide decisions away from inefficient scenarios, but they cannot by themselves determine the “right” criteria because each choice involves trade-offs between harms and benefits. For example, who is to say that an estimated number needed to screen to prevent 1 lung cancer death of 37 (NLST-like estimate) is really better than 45 (2021 USPSTF recommendation), both with wide confidence intervals? Or that increasing the estimated lung cancer mortality reduction from 9.8% to 13.0% is worth an additional 192 000 CT scans per 100 000 persons?… We encourage stakeholders from across the LCS continuum to vigorously debate the inherent trade-offs in selecting eligibility criteria.

Trials vs the real world

What makes the USPSTF’s new guideline even more complicated is that the guideline is based on trial data, but clinical trials differ from real-world clinical settings in many ways. First, the population of trial patients is usually very different. In an accompanying JAMA editorial, professor of radiology at UNC Dr. Louise Henderson and colleagues point out that “Neither the NLST nor the NELSON trial populations were representative of the general US population eligible for lung cancer screening. The NLST underrepresented non-White participants (<10% non-White) and included participants who were healthier, thereby more likely to tolerate curative-intent surgery.”

This means that the moderate benefit from these trials may not transfer over as well to patients that have more chronic conditions, for whom getting more tests and treatments is much more risky. In fact, a 2019 analysis in JAMA Internal Medicine found that among patients in the community, the rate of complications from invasive procedures to assess lung abnormalities was about 22-24%, more than twice the rate found in the NLST.

Another difference is that trials are usually done in hospitals with high standards for screening accuracy and quality. When hospitals with less training and fewer resources undertake broad screening programs, this can mean more false positives and complications. “You have to have a good program,” said Dr. Otis Brawley, oncologist at Johns Hopkins University, in The Washington Post. “A number of centers that are offering it should not be offering it. So those centers are perpetuating disparities, not reducing them.”

“I worry about people (getting screened) in nonacademic settings. I worry about people in rural settings. Is the benefit-to-harm ratio for lung cancer screening going to be the same in the nonacademic setting?”

Dr. Otis Brawley

While the Centers for Medicare and Medicaid require a shared decision making visit before patients get screened for lung cancer, these conversations rarely include a discussion of potential harms or use decision aids. In fact, one doctor quoted in USA Today said, “To be able to order a lung cancer screening test, you have a quick five-minute discussion with the patient about the risk and benefits, they have to be asymptomatic and then they can go off for their scan.” Some doctors may see SDM as just another box to check before ordering the test.

Lastly, the USPSTF model assumes that everyone eligible will have access to screening, and that everyone who starts screening will adhere to the program. In the real world, uptake of screening is only 5% and adherence has been as low as 12% in some populations. Making sure that every screening program matches those of the trials in screening accuracy and interpretation, patient risk, adherence, and avoidance of unnecessary biopsies is a tall order, if not impossible.

The equity question

Some experts are hopeful that the USPSTF recommendation will reduce racial and gender disparities in lung cancer deaths. Black men and white women are both at higher risk than white men for lung cancer at lower ages and smoking histories, so theoretically expanding the eligibility will make this screening more available to these populations.

But there is no guarantee that simply making more people eligible will mean that more people who have been historically underserved will now be helped. “Unfortunately, lowering the age and pack-year requirements alone does not guarantee increased equity in lung cancer screening,” wrote Dr. Yolonda Colson and colleagues at Massachusetts General Hospital, in a JAMA editorial. Colson et al recommend targeting screening based on other “risk factors such as Black race, chronic obstructive pulmonary disease, family history, and occupational exposures.” The USPSTF declined to do this because there isn’t trial data showing that screening specifically helps these populations (demonstrating why including more patients of different races and chronic conditions in clinical trials really matters!)

Ironically, it’s possible that expanding eligibility could actually increase health disparities, because more people are now eligible but there is little targeted outreach to underserved populations. People who already have regular access to care will likely be the first to get screened when they are eligible. In nine states Medicaid doesn’t even cover this screening, so without changing this policy, it will be hard to reach high-risk low-income people.

“When we do things in medicine that don’t work, we hurt people. We hurt not just the people who are getting the test or the treatment that should not be sued. We are also hurting other people who need the hospital, who are crowded out of the hospital and can’t get the services they need.”

Dr. Otis Brawley

Expanding screening eligibility may also shift hospital priorities away from cancer prevention and treatment, which could have negative health impacts. “Unfortunately there are some hospitals that are shutting down navigator programs that would keep women in breast cancer treatment that would ultimately save more lives in order to establish lung cancer screening programs, which are sexy and might bring more prestige to the hospital, and maybe more money,” said Dr. Brawley, in an interview with Managed Care Executive in 2020.

When the USPSTF expanded colon cancer screening to a younger age group we asked, Is lowering the screening age for everyone the best way to increase CRC awareness and access to care for Black people? Now, with this change in lung cancer screening guidelines, we’re still skeptical that expanding screening guidelines across the board will have a net benefit for equity.