Physician, patient education needed to increase lung cancer screening rate

When the U.S. Preventive Services Task Force endorsed lung cancer screening for Americans who meet age and smoking history criteria, proponents hailed the decision as a major step toward reducing mortality from the disease.

The National Lung Screening Trial showed screening with low-dose CT reduced lung cancer deaths by 20% among those at high risk for the malignancy compared with chest x-ray screening.

Those results served as the foundation for USPSTF’s 2013 recommendation that individuals aged 55 to 80 years with a 30 pack-year smoking history who either still smoke or quit within the prior 15 years undergo annual screening with low-dose CT. CMS subsequently supported Medicare coverage of CT screening for this population from ages 55-77.

“The primary benefit of low-dose CT screening alone is the opportunity to detect early lung cancer when it is most likely to be curable with surgical resection or other local therapy,” Denise Aberle, MD, professor of radiology and bioengineering at UCLA David Geffen School of Medicine and a researcher on the NLST, told HemOnc Today.

Five years after the USPSTF recommendation, however, screening rates remain staggeringly low.

Only 1.9% of the 7.6 million eligible current and former heavy smokers nationwide underwent screening in 2016, according to data that Pham and colleagues presented at this year’s ASCO Annual Meeting.

Low uptake may be attributable to many factors, including lack of familiarity among providers about CT screening and eligibility criteria.

“Although we are beginning to see the use of electronic medical records and system alerts to remind physicians of patient eligibility, the vast majority of patients are screened because their health care provider specifically referred them after meeting with them in clinic,” Aberle said. “For many, screening simply hasn’t become incorporated into their regular practice.”

Further, successful lung cancer screening programs require a dedicated team of qualified health care providers, which might not be available at every institution.

“The database systems and human resources are expensive, and the time involved in doing this is substantial,” Aberle said.

HemOnc Today spoke with lung cancer specialists about the low uptake of low-dose CT screening; the benefits and challenges of establishing screening programs; the potential need to broaden screening criteria and coverage; and how members of the clinical community can encourage more patients to undergo screening.

Basis for screening

Lung cancer remains the leading cause of cancer death among men and women. Early detection improves prognosis for most subtypes.

In the NLST, adults underwent three annual screenings with low-dose CT (n = 26,722) or single-view posteroanterior chest radiography (n = 26,732).

Researchers observed 247 deaths from lung cancer per 100,000 person-years in the low-dose CT group and 309 deaths per 100,000 person-years in the radiography group, equating to a 20% (95% CI, 6.8-26.7) relative reduction in mortality from lung cancer with low-dose CT screening.

“The knowledge we have learned from this tremendous study is still the guiding principle of what we do today,” Tawee Tanvetyanon, MD, MPH, medical oncologist and manager of the comprehensive lung cancer screening program at Moffitt Cancer Center, said in an interview. “There has not been any study of this magnitude ever attempted.”

The goal of screening is to detect lung cancer among at-risk individuals before they present with signs or symptoms.

“The benefit is we find lung cancer early and prevent people from dying by treating it before it advanced,” Peter J. Mazzone, MD, MPH, FCCP, director of the lung cancer program and director of education for the Respiratory Institute at Cleveland Clinic, told HemOnc Today. “The only study that has shown we can do that by CT scan is the NLST. With the system used in that trial, there was a 16% to 20% reduction in lung cancer deaths with three annual scans. Nobody is quite sure what it would mean if someone is scanned throughout their adult life.”

Screening also may identify other lung diseases, such as those associated with tobacco, or significant calcific coronary artery disease.

“In some cases, unexpected disease of other organs, and even significant osteoporosis, may be found,” Aberle said.

Still, there are risks associated with screening. This requires all screening programs to be of high quality.

“The specificity of low-dose CT is very low,” Paul A. Bunn Jr., MD, distinguished professor in the division of medical oncology and James Dudley chair in lung cancer research at University of Colorado, said in an interview. “About a quarter of the people will have one or more nodules, and 96% of the nodules are not cancer. It takes a lot of visits and workup to determine if the nodule is malignant or not.”

Many people who are scanned are found to have benign nodules.

“The screening program has to be able to manage those lung nodules well, and we have to make sure we aren’t doing too many additional tests on nodules that aren’t cancer,” Mazzone said. “There is a chance we will hurt as many patients as we help if we don’t manage the nodules well.”

Effectively distinguishing small malignant nodules from those that are benign is key to reducing lung cancer mortality.

“An advantage of screening is that you find many small nodules that are malignant that can be cured with surgery or with radiation if a patient isn’t a surgical candidate,” Bunn said. “Finding those small nodules increases the cure rate.”

Proper management of nodules that pose an intermediate risk may be especially challenging.

“Risks include the observation of indeterminate lung nodules that are not cancer, but that require continued imaging surveillance or biopsy, which can result in unnecessary medical, financial and emotional costs,” Aberle said.

Cost is a concern, but “it costs more to treat somebody with immunotherapy and targeted therapy once they have advanced-stage lung cancer than it does to treat it at an early stage,” Mazzone said. “So, you could argue, honestly, cost-effectiveness is in favor of screening.”

False-positive results from screening can lead to additional imaging, which can result in additional radiation exposure, greater anxiety, and other tests or surgeries.

“The key disadvantage is that the test may show findings that are not cancer,” Tanvetyanon said. “This can cause a false alarm and inconveniences.

“In my opinion, though, the advantages far outweigh the disadvantages,” he added. “The key advantage is that the test can help detect lung cancer early, and this can save many lives.”

Consistently low uptake

Despite the advantages of screening, multiple studies have reported low uptake.

In a research letter published in JAMA Oncology, Jemal and colleagues reported screening rates with low-dose CT scan among high-risk smokers only increased from 3.3% in 2010 — prior to the USPSTF recommendation — to 3.9% in 2015, after the recommendation.

In their ASCO study, Pham and colleagues identified 7.6 million current and former heavy smokers who were eligible nationwide for screening. However, only 1.9% of eligible people (n = 141,260) underwent screening at accredited sites.

“This report likely underestimates the current rate of lung cancer screening, but there is no disagreement that uptake of screening has been and remains low — very low,” Robert A. Smith, PhD, cancer epidemiologist and vice-president, cancer screening at the American Cancer Society, told HemOnc Today.

Source: Miriam Falco.

Cost remains a barrier to implementation of successful screening programs.

Institutions must purchase or develop the necessary database systems to record findings, track individual nodules, ensure communication with patients and providers, and follow up with patients, Smith said.

Tanvetyanon called lung cancer screening programs “a work in progress.”

“Yes, the number 1.9% may seem low, but I think it is a meaningful number given the large group of eligible participants,” he said.

The rate should improve with time, Mazzone said.

“It’s hard to know whether the 1.9% is totally accurate,” he said. “It is true uptake is relatively slow but, given how new lung cancer screening is, that is not necessarily a bad thing. As providers and patients are educated, uptake will continue to grow.”

The low screening rates in the United States likely are multifactorial.

For one, CT screening is a newer technology.

“If you look at the evolution and adoption of mammography, the first dedicated X-ray unit for breast imaging was developed in 1969. It was over 20 years later that Congress passed the Mammography Quality Standards Act, which ensured all women have access to mammography for breast cancer detection,” Aberle said. “Unfortunately, evidence-based medical procedures are very slow to become mainstream.”

Also, there is no tracking system in the U.S. for lung cancer screening, which could help identify targeted populations and manage the screening process.

Levine Cancer Institute at Atrium Health launched a mobile lung cancer screening unit to address another potential challenge: patient access.

The 35-foot unit — which houses a portable, full-body, 32-slice CT scanner — offers uninsured individuals who meet age and smoking history criteria free lung cancer screenings and access to lung cancer information and, if necessary, treatment interventions.

“We hope that, by increasing the screening availability and, thus, getting patients in earlier for care, we will catch lung cancer at a much earlier, much more treatable stage,” Mellisa Wheeler, director of disparities and outreach at Levine Cancer Institute, told HemOnc Today when the mobile unit was launched.

Smith said primary care physicians are the “gatekeepers” to lung cancer screening, and a lack of referrals contribute to low uptake.

“Many PCPs and subspecialists are not very familiar with CT screening and the current eligibility criteria, which poses a problem since they are responsible for various preventive measures in patients,” Smith said.

Shared decision-making also is an important — but potentially resource-demanding — aspect of lung cancer screening.

“It is important to incorporate shared decision-making into the screening program to ensure national coverage for patients,” Aberle said. “This requires human resources.”

Mazzone and colleagues found that shared decision-making visits helped patients better understand eligibility requirements for low-dose CT screening as well as its benefits and harms.

Their survey of 423 patients showed that, prior to their initial visit, participants had poor understanding of age eligibility criteria (8.8%), smoking eligibility criteria (13.6%), the benefits of screening (55.2%) and the harms of screening (38.4%).

Immediately following the visit, participants’ knowledge improved significantly in all areas (P = .03 to P < .0001).

Successful screening also requires that physicians understand eligibility requirements so they can refer appropriate patients.

“The selection criteria must also be understandable to primary care physicians and easy to implement in the context of very busy practices,” Smith said.

Impact of stigma

Stigma surrounding a lung cancer diagnosis remains prevalent and can contribute to poor screening uptake.

“Stigma, nihilism and fatalism may be factors [for low uptake], both among patients and providers,” Smith said.

Smokers have been stigmatized for the past 50 years to believe they deserve tobacco-associated diseases, Aberle said.

“They suffer guilt, shame, embarrassment, the sense of the inevitability of death if diagnosed with lung cancer, and denial,” she said.

Patients often will assume blame for their diagnosis, Bunn added.

“Some of the [stigma] is going away as there are more advocacy groups, treatments are improving and smoking rates declining, but there are societal factors — besides costs, time and expense of screening — to consider as barriers,” he said.

Eligible adults from underserved populations may forgo screening due to the implications of a positive result.

“Underprivileged groups may just not want to know they may have lung cancer because they are struggling to feed families and meet the basic requirements of life,” Aberle said. “How would they pay for the costs associated with treating a lung cancer?”

Although stigma may contribute to patient resistance to screening in some cases, data suggest this might not be a uniform trend.

“Some patients are likely to be resistant to screening, but we observed high rates of acceptance to be in the NLST and high rates of participation in all of the screening rounds,” Smith said. “What little evidence we have has shown that most patients presented with the opportunity for lung cancer screening wish to have the exam.”

Resistance may stem from a fear of the results or adults assuming they are not at high enough risk to justify screening.

“I don’t think that is a major reason of low uptake,” Mazzone said. “It is just a matter of education and developing high-quality programs, and then I think patients will come.”

Defining eligibility

When addressing low uptake, it also is important to consider whether the population examined in the NLST truly represents a real-world population that stands to benefit from screening.

Several studies have evaluated other approaches to screening.

In one, Cheung and colleagues compared USPSTF screening criteria with individualized, risk-based eligibility using data from the National Health Interview Survey in 2005 (n = 5,460), 2010 (n = 5,155) and 2015 (n = 6,971).

Using USPSTF criteria, the number of screening-eligible smokers in the U.S. declined from 9.5 million in 2010 to 8 million in 2015. However, using a risk-based approach, that number only declined by approximately 800,000.

Further, researchers calculated 2,617 more deaths from lung cancer could have been prevented using individual risk-based criteria rather than USPSTF criteria in 2005. The number of additional potentially prevented deaths increased to 5,115 in 2015.

A decline in smoking rates also could affect the number of individuals who qualify for screening.

According to CDC, the percentage of American adults who smoke declined from 20.9% in 2005 to 17.8% in 2013.

A retrospective study by Wang and colleagues suggested this decreased rate could limit the number of adults eligible for screening and, thus, increase lung cancer mortality.

Researchers identified 1,351 individuals form Olmstead County, Minnesota, who developed incident primary lung cancer between 1984 and 2011.

From 1984-1990 to 2005-2011, the proportion of patients diagnosed with lung cancer who had at least a 30 pack-year smoking history decreased (86.6% vs. 77.2%), whereas the proportion of those who had quit smoking more than 15 years prior increased (30.6% vs. 45.8%; P < .001 for both).

More than half (58.6%) of those diagnosed with lung cancer from 1984-1990 would have been eligible for screening using USPSTF criteria, compared with 43.3% of those diagnosed from 2005-2011 (P < .001). The eligibility rate declined during those time periods among both men (60% vs. 49.7%; P = .03) and women (52.3% vs. 36.6%; P = .005).

“The existing screening program will become less effective at reducing lung cancer mortality in the general population if they continue to use the same criteria,” study author Ping Yang, MD, PhD, epidemiologist at Mayo Clinic in Rochester, Minnesota, said in a press release. “We don’t want to penalize people who succeeded in smoking cessation.”

However, when it comes to eligibility, adhering to the current recommendations may be more of an immediate concern than redefining USPSTF criteria.

“The screening strategy is correct, but the proportion of people adhering to the strategy is low,” Bunn said. “Probably less than 10% of the people who meet the guidelines are actually undergoing screening, so it’s a huge problem. It’s not the guidelines at fault, it’s why people aren’t adhering to the guidelines.

“Lung cancer treatments have gotten much better. That is all the more reason to do screening,” he added.

Efforts may be better directed to improve existing lung cancer screening programs.

“As long as screening in the eligible group is performed in high-quality settings — where management of the CT findings is done well — then we still feel this is an effective strategy to screen for lung cancer,” Mazzone said.

Call to quit

Beyond early detection of lung cancer, a major power of screening programs may be their ability to engage with at-risk adults and encourage them to quit smoking.

“I believe that the single greatest added benefit of screening should be more effective smoking cessation,” Aberle said. “A good program should provide or ensure that there is a strong and immediately available program for smoking cessation and follow-up of patients who are considering quitting or who need pharmaceutical assistance to stop.”

A screening program can raise awareness about the dangers of lung cancer.

“This alone can serve as a reminder for smokers to quit,” Tanvetyanon said.

CMS coverage stipulates that individuals be offered counseling to quit smoking, but not that a smoking cessation program be incorporated into the screening program.

“To my mind, it just makes obvious sense to do so and it enhances the benefits of the screening program significantly,” Aberle said.

An individual who decides to undergo screening may be more likely to quit smoking.

“A current smoker who has chosen to undergo lung cancer screening appears to be more likely to be successful in quitting smoking than an adult not attending screening who is only in a smoking cessation program,” Smith said.

Still, if screening itself won’t encourage an individual to quit, it may represent a teachable moment for physicians.

“This can only be effective when screening is combined with a proactive, readily available program for smoking cessation,” Aberle said. “Targeted counseling, close follow-up of patients, and the ability to prescribe nicotine-replacement therapies and other medications for smoking cessation should be central to a good screening program.”

A randomized pilot trial conducted by Kathryn L. Taylor, PhD, behavioral scientist and professor of oncology at Georgetown Lombardi Comprehensive Cancer Center, and colleagues showed that a telephone-based intervention after lung cancer screening quadrupled smoking cessation rates.

The analysis included 92 current smokers with at least a 20 pack-year smoking history scheduled to undergo lung screening.

Researchers randomly assigned study participants to telephone counseling or usual care. Intent-to-treat analyses showed a 17.4% smoking quit rate among those assigned telephone counseling and a 4.3% quit rate among those assigned usual care (P < .05).

“The goal is to get people to think about quitting and think about strategies and methods to quit that they may not have used in the past,” Taylor told HemOnc Today when the study was published. “Even if they choose to not use these strategies right away, they at least have an understanding of what is out there in terms of evidence-based methods for when they are ready to quit. It is a teachable moment.”

Still, evidence from studies on whether screening programs can influence smoking behaviors has been weak, Mazzone said.

“There were concerns that someone finding out that their scan was clear might lead them to smoke more and be less likely to quit, whereas others felt it was a great opportunity to encourage someone to quit smoking and impact smoking rates,” he said. “There hasn’t been a strong signal in either direction.”

More research is necessary to determine how best to properly connect smoking cessation to screening.

“Hopefully, over time, we will get a better idea of what we should be doing and will become more effective at helping active smokers quit when they visit us for screening,” Mazzone said.

Need for improvements

Various measures can be taken to increase uptake among screening-eligible individuals.

One way is to raise awareness about the benefits of screening.

“Lung cancer screening can be improved over time with continued emphasis on how many lives that have been saved by the low-dose CT scan,” Tanvetyanon said. “One way to do this is by the help of media ... to send powerful messages about lung cancer screening when occasions arise.”

Most importantly, more education is needed for PCPs.

“One of the problems is that oncologists see the patients last and the pulmonologists second to last,” Bunn said. “We need new educational efforts to be directed to PCPs.”

Most PCPs know about lung cancer screening, Tanvetyanon said, but discussing patient eligibility is another to-do item on their very busy schedules.

“There are probably too many things on their list, and we respect that,” Tanvetyanon said. “That is why our program [at Moffitt] accepts self-referrals. I would encourage everyone to do what they can do.”

Members of the primary care community are becoming increasingly aware of who should be screened, Mazzone said.

“They have a lot on their plate and they are doing a good job of identifying patients to screen,” he said. “As this becomes a more commonly noted screening, like colonoscopy and Pap smear, I’m quite confident the physician community will know more about it.”

Bunn said electronic quality measures also could make a difference, but there are barriers to achieving good quality pathways.

“The electronic medical records aren’t good at capturing smoking history, so it can be difficult to figure if a patient meets the guidelines,” Bunn said. “Further, it is hard to develop a quality measure, but education and quality measures might help a lot to improve screening.”

EMRs have been beneficial for patient-reported outcomes and other quality measures, but their benefit for screening has been limited.

“[We need] electronic medical records systems that can aggregate data elements of risk and alert physicians and patients about screening and better automate systems for ensuring follow-up; computer-aided systems that automatically detect and measure nodules, provide coronary artery calcium scores, measure bone density, provide measures of extent of emphysema and fibrosis, and map these to appropriate recommendations and management strategies; and molecular testing of readily accessible biospecimens like blood or buccal/nasal epithelial cells that can increase the performance and benefits of screening by increasing the positive predictive value of a positive screen,” Aberle said.

Erkmen and colleagues presented a study at last year’s International Association for the Study of Lung Cancer Multidisciplinary Symposium in Thoracic Oncology that showed an EMR-centered database could be useful for obtaining important quality and outcomes data from patients undergoing lung cancer screening.

Researchers created a mechanism to measure metrics and outcomes using existing EMR functions for 373 patients to provide the documentation and data entry required by CMS for billing.

Researchers used a flowsheet function of the Epic EMR system to create custom fields that asked about demographics, smoking status, history of smoking, lung cancer screening eligibility, shared decision-making, smoking cessation counseling and results of low-dose CT screening. Patients also completed follow-up questionnaires.

Ninety-eight percent of patients documented a shared decision-making visit and 90% reported using a shared decision-making tool within the flowsheet. A total of 221 active smokers reported at least a discussion of smoking cessation counseling, and 290 patients reported low-dose CT screening results.

“We were using our existing Epic platform to collect diverse data,” Erkmen said during her presentation. “We can tailor these data based on what we need. We need more information about race, education and socioeconomic barriers.”

Other research is evaluating means of more personalized screening options and biomarkers to better define eligibility criteria.

Tanner J. Caverly, MD, MPH, from the VA Center for Clinical Management and University of Michigan Medical School, and colleagues developed a risk prediction tool that tailors which patients should be screened based on clinical benefit and personal preferences.

Researchers developed a state-transition microsimulation model using data from two large randomized trials to determine the effect of personal preferences on the benefit of screening among patients who underwent 3 years of annual low-dose CT screening vs. no screening.

Data showed that moderate differences in patient preferences about the downsides of screening influenced whether screening was appropriate among eligible individuals who had a less than 0.3% annual cancer risk or less than 10.5 years life expectancy. However, highly negative views about screening and its drawbacks did not impinge on the benefits of screening among individuals with a higher risk and longer life expectancy.

“Our results support the importance of personalizing the harm-benefit assessment of low-dose lung cancer screening for informing screening decisions rather than uniformly recommending or withholding a recommendation for eligible patients,” Caverly and colleagues wrote.

In a study published in July in JAMA Oncology, researchers reported a new blood test may serve as a biomarker-based risk assessment for improving eligibility criteria for lung cancer screening.

The study included 63 ever-smokers with lung cancer and 90 matched controls. Results showed an integrated risk prediction model that combined smoking exposure with a biomarker score yielded an area under the curve of 0.83 (95% CI, 0.76-0.9) compared with 0.73 (95% CI, 0.64-0.82) for a model based on smoking exposure alone (P = .003 for difference).

“The question boils down to, are there other individuals who don’t meet current eligibility criteria who are at risk?” Mazzone said. “Can we expand the pool by finding other high-risk patients?

“But, the big concern now is we have 7 million to 9 million adults who would be eligible but who we are not all reaching right now,” he added. “Let’s get it right, and as we get it right and grow the programs, we can learn more about how we can expand the risk pool through biomarkers that are being developed or otherwise.” – by Melinda Stevens

Click here to read the , “Should a model-based approach be used to identify which individuals should undergo lung cancer screening?”

References:

Carlson AS, et al. Abstract OA01.03. Presented at International Association for the Study of Lung Cancer Multidisciplinary Symposium in Thoracic Oncology; Sept. 14-16, 2017; Chicago.

Caverly TJ, et al. Ann Intern Med. 2018;doi:10.7326/M17-2561.

Cheung LC, et al. Ann Intern Med. 2017;doi:10.7326/M17-2067.

Erkmen CP, et al. Abstract OA01.02. Presented at International Association for the Study of Lung Cancer Multidisciplinary Symposium in Thoracic Oncology; Sept. 14-16, 2017; Chicago.

Huo J, et al. JAMA Intern Med. 2016;doi:10.1001/jamainternmed.2016.9016.

Infante M, et al. Am J Respir Crit Care Med. 2015;doi:10.1164/rccm.201408-1475OC.

Integrative Analysis of Lung Cancer Etiology and Risk (INTEGRAL) Consortium for Early Detection of Lung Cancer. JAMA Oncol. 2018;doi:10.1001/jamaoncol.2018.2078.

Jemal A, et al. JAMA Oncol. 2017;doi:10.1001/jamaoncol.2016.6416.

Kinsinger LS, et al. JAMA Intern Med. 2016;doi:10.1001/jamainternmed.2016.9022.

Mazzone PJ, et al. Chest. 2016;doi:10.1016/j.chest.2016.10.027.

Pham D, et al. Abstract 6504. Presented at: ASCO Annual Meeting; June 1-5, 2018; Chicago.

Taylor K, et al. Lung Cancer. 2017;doi:10.1016/j.lungcan.2017.01.020.

The National Lung Screening Trial Research Team. N Engl J Med. 2011;doi:10.1056/NEJMoa1102873.

Wang Y, et al. JAMA. 2015;doi:10.1001/jama.2015.413.

For more information:

Denise Aberle, MD, can be reached at daberle@mednet.ucla.edu.

Paul A. Bunn Jr., MD, can be reached at paul.bunn@ucdenver.edu.

Peter J. Mazzone, MD, MPH, FCCP, can be reached at mazzonp@ccf.org.

Robert A. Smith, PhD, can be reached at robert.smith@cancer.org.

Tawee Tanvetyanon, MD, MPH, can be reached at tanvett@moffitt.org.

Disclosures: Smith reports American Cancer Society received an unrestricted educational grant from AstraZeneca to support the National Lung Cancer Roundtable. Aberle, Bunn, Mazzone, Smith and Tanvetyanon report no relevant financial disclosures.