Lung cancer screening and prevention

CLINICAL PEARLS

Lung cancer mortality has been declining since the 1990s, yet it remains the leading cause of cancer mortality and is still expected to cause more than 120,000 deaths in 2023 in the United States.
National and local efforts to curtail smoking initiation and encourage cessation are estimated to have been associated with an avoidance of 8 million premature deaths and an estimated extended mean life span of 19 to 20 years.
Several trials evaluating lung cancer screening using low-dose computed tomography (LDCT) have been completed in the past 2 decades, with some currently ongoing. They have differed considerably from each other in gender ratios, age cutoffs, and total number of pack-years of smoking burden for participants. The National Lung Screening Trial was the first prospective randomized trial to demonstrate that screening individuals at high risk for lung cancer with LDCT could decrease mortality.
In 2013, the US Preventative Services Task Force provided a “grade B” recommendation for annual screening for lung cancer using LDCT in adults aged 55 to 80 years with a 30 pack-year smoking history, who currently smoke or quit within the past 15 years. A revision in July 2020 lowered the age threshold for initiation of screening to 50 years, and the minimum smoking history was lowered to 20 pack-years.
National Comprehensive Cancer Network guidelines and the Lung-RADS tool risk-stratify and standardize findings on LDCT and delineate next steps for additional evaluation and follow-up.

Epidemiology, risk factors, and prevention of lung cancer

Lung cancer is the most common cause of cancer death in both men and women worldwide, accounting for 11.4% of all new cancers in 2020 and 18% of all cancer deaths. ^, In the United States, lung cancer has been closely linked to tobacco use and specifically cigarette smoking. Since the publication of the first Surgeon General’s report on smoking and health in 1964, the production, sale, and consumption of tobacco products has been subject to regulation at the state and federal levels.

The morbidity and mortality in the general population attributable to tobacco use, and specifically attributable to lung cancer caused by tobacco use, is an area of extensive research. Geographically, lung cancer mortality tracks closely with areas of high smoking prevalence. Moreover, differences are observed between males and females that also track with smoking prevalence, as regions with highest male smoking prevalence (Eastern and Southeastern Asia and Eastern Europe) have markedly higher mortality due to lung cancer among men compared with regions with the highest female smoking rates (European countries, followed by Oceania and North/South America). In the United States in particular, the magnitude of the association between smoking and lung cancer mortality has increased over time as smoking has become less common among women, with an estimated relative risk of 2.7 in the 1960s to 25.7 in the 2000s). National and local efforts to curtail smoking initiation are estimated to have been associated with an avoidance of 8 million premature deaths and an estimated extended mean life span of 19 to 20 years.

Multiple related analyses have investigated the combined effects of smoking and other behavioral and environmental risk factors. In this vein, careful separation of the effects of tobacco smoke from the incident effects of environmental smoke is difficult, in part because of the confounding nature of the exposures; smoking is more prevalent among members of certain occupations, in certain socioeconomic classes, and in certain geographic regions. This nonrandom variation, therefore, should inform public health efforts to define the appropriate intervals for screening among different populations.

Second-hand smoke

The establishment of a causal link between tobacco smoke and lung cancer subsequently generated interest for the ripple effects of antismoking policy on the long-term cancer risks of never-smokers who are affected by passive smoke exposure. In a pooled analysis of two large European case-control studies of lung cancer patients, odds of lung cancer incidence were 18%–23% higher with ever-smoking or long-term smoking, respectively, by a spouse. This was similar for exposure in the workplace, with 16%–27% higher odds. A dose-response relationship for increasing duration of exposure to secondhand smoke was also demonstrated.

Air pollution, environmental, and occupational exposures

Similar to tobacco smoke, fine particulate matter and ambient air pollution contribute to lung cancer risk. The relative contributions of occupational exposures have also been differentiated from classic “smoke exposure.” In analysis of data from the National Lung Screening Trial (NLST), exposure to welding fumes and metallic particulate matter from foundry gases was associated with increased risk of lung cancer in ever-smokers. ^, Specifically, increasing years of employment in welding and foundry work were related to increased lung cancer risk among heavy smokers, and squamous cell subtype in particular. A joint effect of those who were employed in both professions, compared with those who were never metalworkers, was also found.

Traffic-related pollutants, specifically nitrogen oxides, have been examined in multiple observational studies, with a meta-estimate for change in lung cancer associated with a 10-μg/m ³ increase in exposure to NO ₂ being 4% (95% confidence interval [CI], 1%−8%). This ultimately led to the classification of outdoor air pollution including fine particulate matter and diesel exhaust as a group 1 carcinogen by the International Agency for Research on Cancer. Recently, data from the Adventist Health and Smog Study-2 (AHSMOG-2), which enrolled nearly 100,000 nonsmokers from all 50 US states and five Canadian provinces, showed that lung cancer incidence increased 43% (95% CI adjusted hazard ratio [aHR], 1.11–1.84) for each 10-μg/m ³ increment in fine particulate matter. Additionally, in the subset of patients who spent over an hour per day outdoors, this amounted to a 68% increase in lung cancer incidence (95% CI aHR, 1.28–2.22), and for those who had lived 5 or more years at their enrollment address, this was associated with a 54% increase in lung cancer incidence (95% CI aHR, 1.17–2.04). These estimates shed light on the variability of this association in nonsmokers versus smokers. In a Chinese cohort that included smokers, the contribution of fine particulate matter to lung cancer incidence was estimated to be 5.5% for a 10 mg/m ³ increase in 2-year average (95% CI, 3.8%–7.2%) for men and a 14.9% increase for women (95% CI, 12.0%–17.8%).

Radon

Radon, an inert radioactive noble gas, has been linked to the increased incidence in lung cancer in miners and laboratory animals. However, empiric evidence showing increased lung cancer risk from residential exposure only emerged in the past 20 years. Because radon becomes trapped in enclosed spaces, public health interest in the magnitude of this association soon developed. Because the concentration of radon in residences is much lower than in mines, the strength of the association, and its dependence on other coincident risk factors, has also been investigated extensively. A meta-analysis of eight epidemiological studies found that for nonsmokers, the risk of developing lung cancer for an exposure of 150 Bq/m ³ was increased by 18% (95% CI, 0.8−1.6) and there was a significant dose-response relationship. When examined in conjunction with smoking status in a pooled analysis of European epidemiological studies, the absolute risks of lung cancer by age 75 years at usual radon concentrations at 100 and 400 Bq/m ³ were estimated to be approximately 0.5% and 0.7%, respectively, for lifelong nonsmokers; however, for smokers, this risk was approximately more than an order of magnitude greater (12% and 16%, respectively).

Chronic inflammation and diet

Smaller but not insignificant contributions to increased risk of lung cancer have been explored from an epidemiological standpoint. Prospective screening studies have further characterized the added risk incurred by the inflammatory potential of diet in heavy smokers. While a higher dietary inflammatory index has not been shown to increase lung cancer risk (despite association with higher reported dyspnea and radiological evidence of emphysema), adherence to a traditional Mediterranean diet has been associated with lower lung cancer risk. Furthermore, in a meta-analysis of studies examining fruit and vegetable intake, significant inverse dose-response relationships were observed for 100 g/day increase of fruits and vegetables across 18 studies.

History and evolution of lung cancer screening

History

Lung cancer remains the leading cause of cancer mortality in part due to the fact that nearly half of patients are diagnosed at advanced stage. Like many malignancies, the prognosis for patients with early-stage lung cancer is significantly better than when lung cancer is diagnosed at later stages. An ideal screening technique would therefore enable detection of disease that is localized and small, before development of symptoms or advanced or metastatic disease, particularly at a point when cure or successful treatment is no longer possible. Effective screening tests have been available for the three other cancers with highest mortality in the United States, specifically breast, colorectal, and prostate cancers, which have contributed to improving mortality in each of these cancers. Lung cancer mortality has also been declining since the 1990s, largely due to decreases in smoking, yet lung cancer remains the leading cause of cancer mortality and is still expected to cause approximately 130,000 deaths annually. There has been significant interest in developing and improving screening methods for lung cancer to diagnose patients at earlier stages, to increase the opportunity for treatment at a point to improve prognosis.

Early attempts at developing a screening test focused largely on chest radiography, often in conjunction with sputum cytology, though those modalities have since been shown to not be effective screening methods. The NLST, initially published in 2011, was the first prospective randomized trial to demonstrate that screening individuals at high risk for lung cancer with low-dose computed tomography (LDCT) could decrease mortality. Other studies have since demonstrated similar findings with LDCT, and ongoing and future studies will continue to improve on screening modalities and guidelines, define the optimal eligibility criteria for screening, and improve implementation of guidelines.

Screening with chest radiography and sputum cytology

Interest in lung cancer screening goes back to at least the 1950s and 1960s, when lung cancer incidence and mortality rates were increasing dramatically and smoking had been established as a risk factor for developing lung cancer. Initial studies that evaluated screening with chest radiography, with or without concurrent sputum cytology analysis, have conclusively shown no associated mortality benefit, and screening with chest x-ray (CXR) and sputum cytology is not recommended. One of the earliest studies by the Mass Radiography Service in London randomized more than 55,000 men to either chest radiography every 6 months or chest radiography only at the beginning and end of the 3-year study period. While more resectable lung cancers were detected in the semiannual screening group (43.6% vs. 29.0%), there was no difference in mortality between the two groups. In another pilot study published by the American Cancer Society and Veterans Administration, 14,607 veterans older than the age of 45 were evaluated with chest radiographs at 6-month intervals combined with evaluation for sputum cytology and questionnaires on smoking habits and respiratory symptoms; this screening intervention also did not show a mortality benefit.

The National Cancer Institute (NCI) Cooperative Early Lung Cancer Detection Program subsequently sponsored three parallel randomized studies in the 1970s at the Mayo Clinic, Johns Hopkins Medical Institutions (JHMI), and Memorial Sloan-Kettering Cancer Center (MSK) to determine whether detection of lung cancer could be improved by adding sputum cytology techniques (namely bronchoscopy) to regular examination with chest radiography to reduce lung cancer mortality. The Mayo study evaluated 10,933 male smokers older than age 45 with baseline CXR and sputum cytology and then randomized patients with negative screens to CXR and sputum cytology every 4 months or usual care, which included recommending an annual CXR. While more lung cancers were diagnosed and at earlier stages in the screening arm, the study found no mortality benefit of continued screening every 4 months compared with usual care. A similar study of 6364 patients in Czechoslovakia also found an increased incidence of early-stage, resectable lung cancer in the screening group but also failed to demonstrate a difference in mortality.

Conversely, the NCI Johns Hopkins and Memorial Sloan-Kettering studies were designed specifically to assess the role of sputum cytology in screening. In the two studies, a combined 20,427 male smokers older than the age of 45 had annual dual screening with annual CXR plus sputum cytological testing every 4 months or annual CXR alone. ^, Neither study found that the addition of sputum cytology to an annual radiographic screening provided any benefit, with similar lung cancer mortality between the groups. Notably, the long-term survival for patients in these studies was better than what would have been expected in similar patients at the time, though due to the design of these studies to specifically evaluate sputum cytology, they were unable to determine whether screening with CXR alone provided a survival benefit.

The three NCI and Czechoslovakian studies ultimately did not show a benefit in lung cancer mortality in patients undergoing screening. While sputum cytology was not found to be a useful adjunct to screening with chest radiography, these studies were unable to determine whether screening with chest radiography alone provides a mortality benefit over no screening at all, largely due to selection and lead time bias, lack of proper control groups in the studies, and insufficient statistical power. To address these concerns and determine the value of annual CXR in lung cancer screening, lung cancer screening with CXR was included in the Prostate, Lung, Colorectal and Ovarian (PLCO) trial, which was a randomized controlled trial evaluating screening techniques for prostate, lung, colorectal, and ovarian cancers conducted between 1993 and 2001.

For the lung cancer portion of the PLCO study, 154,901 male and female patients, including smokers and nonsmokers, between ages of 55 and 74 were randomized to usual medical care or a baseline CXR and then annual CXR for 2 (nonsmokers) or 3 years (smokers). ^, At 13 years of follow-up, the incidence rates of lung cancer, as well as stage and histological findings, were similar between the usual care and screening groups (19.2 vs. 20.1 cases per 10,000 person-years). Furthermore, the groups had a similar total number of lung cancer deaths (1230 vs. 1213 deaths) and mortality (mortality rate ratio [RR] 0.99; 95% CI 0.87−1.22). This similarity in mortality persisted in a subset analysis of high-risk patients who also met criteria for the NLST, specifically those patients with at least a 30 pack-year smoking history who were either current smokers or had quit in the prior 15 years. With the results of the PLCO study, it was conclusively demonstrated that screening for lung cancer with annual chest radiography provides no mortality benefit ( Table 1.1 ).

TABLE 1.1 ■

Table From the National Lung Screening Trial: Overview and Study Design Summarizing the Randomized Controlled Trials of Lung Cancer Screening With Chest Radiography With or Without Sputum Cytology

Study	Intervention	No. of Participants	No. of Lung Cancers Detected at First Screening (Prevalence)	No. of Lung Cancers Detected After First Screening	No. of Stage III and IV Cancers ^a	Lung Cancer Mortality ^b ^, ^c	5-Year Survival (%) ^b
Memorial Sloan-Kettering					173	NA	35
Experimental arm	Annual chest radiography, sputum cytology every 4 months	4968	30	146	...	...	...
Control arm	Annual chest radiography	5072	23	155	...	...	...
Johns Hopkins					NA		NA
Experimental arm	Annual chest radiography, sputum cytology every 4 months	5226	39	194	...	3.4/1000 PY	...
Control arm	Annual chest radiography	5161	40	202	...	3.8/1000 PY	...
Mayo Lung Project			91 in all ^d
Experimental arm	Chest radiography, sputum cytology every 4 months	4618	...	206	123 ^b	4.4/1000 PY	35
Control arm	Recommended annual chest radiography, sputum cytology	4593	...	160	119 ^b	3.9/1000 PY	19
Czechoslovakian randomized control trial			19 in all ^d				NA
Experimental arm	Chest radiography and sputum cytology every 6 months × 3 years, annually after year 3	3171	...	108	53	7.8%	...
Control arm	Chest radiography and sputum cytology annually after year 3	3174	...	82	46	6.8%	...

a Numbers of stage III and IV lung cancers include all cancers (prevalence and later) for the Memorial Sloan-Kettering RCT. Numbers of stage III and IV lung cancers do not include prevalence cancers for the Mayo Lung Project and Czechoslovakian RCT.

b Lung cancer mortality and 5-year survival was at 5–8 years of follow-up for the Memorial Sloan-Kettering RCT, at 8 years of follow-up for the Johns Hopkins Study, at 20 years of follow-up for the Mayo Lung Project, and at 15 years of follow-up for the Czechoslovakian RCT.

c Mortality data for the Johns Hopkins RCT include prevalence cases. Mortality data for the Mayo Lung Project and the Czechoslovakian RCT do not include prevalence cases. PY = person-years.

d Not included in remainder of Table.

Screening with low-dose computed tomography

As studies in screening with chest radiography and sputum cytology continued to show no mortality benefit in lung cancer, interest in screening eventually shifted toward the use of computed tomography (CT) for screening, which has been shown to have higher sensitivity for detection of early-stage disease compared with radiographs. The development of low-dose helical chest CT (LDCT) allowed for fast, high-resolution cross-sectional imaging during a single breath hold at lower radiation doses than conventional standard-dose chest CT. ^,

Early lung cancer action program

The Early Lung Cancer Action Project (ELCAP), initiated in 1992, was the first to study early diagnosis of lung cancer in cigarette smokers using LDCT. The project enrolled asymptomatic smokers (with at least a 10 pack-year history) age 60 or older and with no history of previous cancer and provided both chest radiographs and LDCT for each patient. The early findings of the ELCAP study demonstrated that many patients with lung cancer using LDCT were diagnosed with low-stage, resectable disease. Since the publication of their initial screening results in 1999, the program expanded to multiple sites in New York state (NY-ELCAP) and now to include other national and international screening sites (I-ELCAP). The initial studies provided quantitative data suggesting a benefit of screening for lung cancer with LDCT.

National lung screening trial

While several nonrandomized, observational trials suggested that screening with LDCT could lead to detection of early-stage lung cancer, ^, the NLST was the first prospective, randomized trial to demonstrate a lung cancer−specific mortality benefit for lung cancer screening with LDCT. The NLST was a large prospective, multicenter, randomized controlled trial funded by the NCI that compared LDCT with chest radiography in patients at high risk for lung cancer. Specifically, eligible patients were older (55−74 years old) current or former smokers with at least a 30-pack year history of cigarette smoking; former smokers were to have quit within the previous 15 years. A total of 53,454 patients were enrolled between 2002 and 2004 and were randomized to three annual screening examinations with LDCT ( n = 26,722) versus chest radiography ( n = 26,732) at 33 US medical centers. Positive screening results were defined as noncalcified nodules greater than or equal to 4 mm in diameter on LDCT or any noncalcified nodule or mass for radiography. The trial was stopped early after a preplanned interim analysis in 2010 found that the primary endpoint of the trial had been reached; specifically, a 20.0% (95% CI, 6.8−26.7; P = 0.004) relative reduction in the rate of death from lung cancer was found in the LDCT group compared with the chest radiography group, and the number of patients needed to screen with LDCT to prevent one death from lung cancer was 320. The study also showed a significant reduction in all-cause mortality of 6.7% (95% CI, 1.2−13.6; P = 0.02) in the LDCT group, though when lung cancer deaths were excluded from this comparison the reduction in overall mortality was no longer significant. Importantly, since the NLST only evaluated LDCT versus CXR and therefore did not have a “usual care” arm, the subgroup analysis of the previously mentioned PLCO trial provided a way to evaluate the benefits of LDCT compared with usual care, as well as the value of CXR in high-risk patients.

In the NLST trial, at a median follow-up duration of 6.5 years, adherence to the screening protocols were high in both arms: 95% in the LDCT group and 93% in the radiography group. In all three rounds of screening, there were higher rates of positive screening results in the LDCT group (overall screening period: 24.4% vs. 6.9%), as well as rates of identification of findings suspicious for abnormalities other than lung cancer (7.5% vs. 2.1%). The majority of further diagnostic evaluations that resulted from these identified abnormalities were additional imaging (chest radiography, chest CT, positron emission tomography [PET] or PET/CT), and invasive procedures were done infrequently in both groups. Among those patients who underwent invasive diagnostic procedures, the adverse event rate was relatively low (at least one complication: LDCT, 1.4% vs. radiography, 1.6%). Sixteen patients (10 of whom had lung cancer) in the LDCT group and 10 in the radiography group (all of whom had lung cancer) died within 60 days of an invasive diagnostic procedure. Across the three rounds of screening, 96.4% of the findings in the LDCT group and 95.5% of the findings in the radiography group were found to be false positives. Overall, a total of 1060 lung cancers (645 cases per 100,000 person-years) were diagnosed in the LDCT group compared with 941 (572 cases per 100,000 person-years) in the radiography group, with more cancers diagnosed at lower stages in the LDCT group. Among those lung cancers diagnosed during the study, more patients in the LDCT group were diagnosed by positive screening tests, compared with being diagnosed after a negative screening test or during follow-up, suggesting that screening with radiography alone misses some cancers at lower stages.

Updated results of the NLST were published in 2019, with median follow-up time of 11.3 years for incidence and 12.3 years for mortality. The updated analysis found that LDCT diagnosed 1701 lung cancers compared with 1681 by CXR (RR, 1.01; 95% CI, 0.95−1.09) and lung cancer deaths in each group of 1147 versus 1236 in the LDCT and CXR arms, respectively (RR, 0.92; 95% CI, 0.85−1.00). The long-term results suggest number of patients needed to screen (NNS) to prevent one death from lung cancer of 303, which is similar to the original NNS of 320. The long-term results did not find an increase in lung cancer incidence in the LDCT group compared with the CXR group.

The NLST was the largest trial to support the large-scale feasibility of LCS with LDCT and the first randomized trial to show a definitive mortality benefit to LDCT screening. The results also suggested that risks of screening include high false-positive rates and requirement of further diagnostic testing or procedures. Limitations of the study included lack of a control arm of usual care. However, results from the PLCO trial suggest that a similar reduction in mortality would have been seen had the control arm been usual care rather than radiography. Additional limitations include limited generalizability to the overall population given the high rates of adherence to screening in the study, selection bias, and performance of screening at relatively specialized centers in the study. Further questions remain including the impact of overdiagnosis and potential for radiation-induced cancers from screening, management of incidental and nonpulmonary findings, the cost-effectiveness of screening with LDCT, and appropriate implementation and eligibility criteria.

NELSON trial

The Dutch-Belgian Nederlands–Leuvens Longkanker Screenings Onderzoek (NELSON) trial also demonstrated a benefit to screening with LDCT. In this study, 15,822 patients aged 50 to 74 years and with significant smoking histories were randomized into a screening group that would receive LDCT at years 1, 2, 4, and 6.5 compared with usual care. Notably, the definition of a positive nodule in the NELSON trial differed from the NLST; the NELSON trial used nodule volume to categorize nodules and volume doubling time, which has been shown to be more accurate and have fewer false positives than the NLST diameter criteria, as well as the initial Lung Imaging Reporting and Data System (Lung-RADS) criteria that was developed by the American College of Radiology (ACR) to standardize lung cancer screening CT reporting and management recommendations.

Unlike other smaller European trials, the NELSON trial was powered to detect a reduction in lung cancer mortality of 25% or more at 10 years. The results of the NELSON trial, presented at the 2018 World Conference on Lung Cancer, showed that the majority of lung cancers diagnosed in the LDCT group were diagnosed at earlier stages, compared with the majority of cancers in the control arm being diagnosed at stage III/IV. The trial found a statistically significant 26% lung cancer mortality benefit for men in the LDCT group. While the majority of patients in the trial were men (84%), the trial also looked specifically at gender differences in mortality benefit, as the results of a subgroup analysis of NLST and the German LUSI (LUng cancer Screening Intervention) trial have suggested that there may be an greater benefit to screening in women. ^, Indeed, the NELSON trial found a significant 39% reduction in lung cancer mortality among women. The mortality benefit of screening with LDCT confirms the findings of a survival benefit among high-risk patients from the NLST and, in fact, are more favorable while also suggesting a significant gender difference in the benefit of screening favoring women.

Other randomized trials of screening with low-dose computed tomography

A number of other randomized studies have evaluated LDCT as a screening modality for lung cancer, employing a variety of different eligibility criteria, recruitment strategies, size, and screening designs. All have included high-risk patients, or those with significant smoking histories, but differed in eligibility criteria of pack-year smoking histories and age. Unlike the NLST, most included a control arm of standard of care that did not include a screening modality. In addition, as conducted, most of these studies were not sufficiently powered to detect a lung cancer−related mortality difference.

The DANTE (Detection And screening of early lung cancer with Novel imaging TEchnology) trial was an Italian study comparing screening with LDCT to usual care in 2450 men aged 60 to 74 with 20+ pack-year smoking histories. ^, Patients in both arms received a baseline examination, CXR, and sputum cytology, and the intervention arm also received five annual LDCTs. In their long-term follow-up results, at a median follow-up of 8.35 years, they found a similar lung cancer mortality between the two groups, though were statistically underpowered to make a statement on the efficacy of LCS with LDCT.

The DLCST (Danish Lung Cancer Screening Trial) was a single-center trial from Denmark that also compared LDCT to usual care in 4104 patients aged 50 to 70 with a minimum 20 pack-year smoking history. Similar to the DANTE trial, they found no difference in lung cancer mortality, as well as all-cause mortality, though more cancers were detected in the screening group and at lower stages. Both trials had shorter smoking exposures than the NLST trial and were underpowered to demonstrate a mortality benefit of screening.

The MILD (Multicentric Italian Lung Detection) trial randomized 4099 patients to annual or biennial screening with LDCT versus usual care with no screening. Notably, the screening and control groups differed significantly with the control group being younger, with a higher proportion of females, more current (compared with former) smokers and shorter pack-year smoking history. The trial found a 39% decrease in lung cancer−specific mortality in the LDCT arms compared with usual care at 10-year follow-up (hazard ratio [HR], 0.61; 95% CI, 0.39−0.95; P = 0.017), as well as a nonsignificant 20% decrease in all-cause mortality.

The Depiscan trial was a French randomized trial of LDCT versus CXR that suggested noncalcified nodules are more often detected with LDCT, but it also interestingly evaluated the feasibility of enrollment in screening by general practitioners. They found that enrollment to trial was difficult, with less than half of invited general practitioners actually enrolling patients for the study, and almost 20% of patients withdrawing consent for the study after enrolling.

Other randomized trials include the German LUSI trial that examined 4052 long-term smokers aged 50 to 69 randomized to screening with 5 annual LDCT versus usual care without screening; the UKLS (UK Lung Cancer Screening) trial, which randomized 4055 patients aged 50 to 75 at high risk for lung cancer based on a population-based survey and risk prediction model to a screening arm with LDCT or a control arm with no screening; and the Italian ITALUNG trial, which randomized 1613 patients aged 55 to 69 with at least a 20 pack-year history in the prior 10 years to screening with LDCT for 4 years or usual care. ^, ^, Similar to other smaller European trials, these trials were not powered to demonstrate overall differences in lung cancer mortality, but the LUSI trial did show a mortality benefit specifically among women (HR, 0.31; 95% CI, 0.1−0.96; P = 0.04).

International lung screen trial

The NLST trial and NELSON trials, as well as others, used eligibility criteria to identify patients at higher risk for developing lung cancer on the basis of age and risk factors, namely significant smoking history. Smaller trials have evaluated risk scores as potential alternatives for selection of appropriate patients for screening. The International Lung Screen Trial (ILST), the initial interim results of which were recently published in the Journal of Thoracic Oncology, aims to answer the question of whether risk prediction models or patient selection criteria such as those used in the NLST are superior for identifying individuals appropriate for LCS. The ILST is a multicentered trial that is enrolling 4000 participants, randomized to receive two annual screenings if they fit the US Preventative Services Task Force (USPSTF) criteria versus a 6-year risk of development of lung cancer of ≥1.5% based on the PLCO _m2012 risk assessment model. The PLCO _m2012 risk prediction model, developed by Tammemagi and colleagues, is based on data from 80,375 participants in the PLCO trial and incorporates a number of patient factors, including age, race, personal medical history, family cancer history, and smoking history. ILST participants will receive two annual screens with LDCT and be followed for 6 years. Those patients who do not fit USPSTF criteria or do not have a ≥1.5% risk based on PLCO _m2012 criteria are not offered screening, but some will be followed to assess for lung cancer outcomes.

At the time of the interim analysis reported, 3673 patients had been enrolled and scanned across international centers. In the patients meeting PLCO _m2012 criteria only (i.e., not USPSTF criteria), 16 cancers had been diagnosed in 795 patients. In the patients meeting USPSTF criteria alone (i.e., not PLCO _m2012 criteria), one cancer had been diagnosed in 478 patients. Of those patients meeting both criteria, 46 cancers had been diagnosed among 2400 patients. This translates to a higher positive predictive value of lung cancer risk in patients meeting PLCO _m2012 compared with USPSTF criteria (1.94% vs. 1.63%, P = 0.34) and odds ratio [OR] favoring PLCO criteria (McNemars OR, 16; 95% CI, 2.49−670.96; P = 0.0003). More lung cancers were diagnosed in the PLCO _m2012 -positive group. These results suggest that using PLCO _m2012 criteria may be superior for classification accuracy of lung cancer screening outcomes, and that patients who fit USPSTF criteria but not PLCO _m2012 criteria are at low risk (0.2% baseline risk) and are unlikely to benefit from screening. The final results of this study, expected to be published in the next few years, as well as ongoing and future studies, will further refine appropriate eligibility criteria for screening ( Table 1.2 ).

TABLE 1.2 ■

Table From Screening for Lung Cancer CHEST Guideline and Expert Panel Report [Mazzone] Summarizing Design of (A) Randomized Controlled Trials for Lung Cancer Screening With Low-Dose Computed Tomography and Trial Results (B)

Adapted from: Mazzone PJ et al: Screening for lung cancer: CHEST guideline and expert panel report. Chest. ePub, 2021

A
Study	Sample Size	Age (Years)	Smoking History	Smoking Cessation (Years Since Quit)	Screening Interval and Duration	Follow-Up (Years)	Definition of Positive Result ^a
LDCT vs. CXR
NLST	53,454	55–74	≥30 pack-years	≤15	3 Annual screens	6.5 (median)	≥4 mm
Depiscan	765	50–75	≥15 cigarettes/day for ≥20 years	<15	3 Annual screens	NR	>5 mm
LDCT vs. usual care (no screening)
DANTE	2472 males	60–74	≥20 pack-years	<10	5 Annual screens; baseline CXR for both study arms	8	>5 mm
DLCST	4104	50–70	≥20 pack-years	<10	5 Annual screens	10	>15 mm or rapid growing 5- to 15-mm nodules (>25% increase in volume on 3-month repeat CT)
NELSON	15,822	50–75	≥15 cigarettes/day for ≥25 years or ≥10 cigarettes/day for ≥30 years	<10	4 Screening rounds; interval after baseline; 1 year, 2 years, and 2.5 years	7	Volume >500 mm ³ or volume 50–500 mm ³ with VDT <400 day on 3-month repeat CT
ITALUNG	3206	55–69	≥20 pack-years	≤10	4 Annual screens	6	≥5 mm solid nodule, a ground-glass nodule ≥10 mm, or any part-solid nodule
MILD	4099	≥49	≥20 pack-years	<10	2 Study arms: 5 annual screens; or 3 biennial screens	5	Volume >250 mm ³ or rapid rowing 60–250 mm ³ (>25% increase in volume on 3-month repeat CT)
LUSI	4052	50–69	≥15 cigarettes/day for ≥25 years or ≥10 cigarettes/day for ≥ 30 years	<10	4 Annual screens	3	≥5 mm
UKLS	4055	50–75	LLPv2 risk ≥5%		1 Screening	10	Volume >500 mm ³ or volume 50–500 mm ³ with VDT <400 days on 3-month repeat CT
LSS	3318	55–74	≥30 pack-years	< 10	One screening	1	≥ 4mm

B
Study	No. Randomized	Age (Years): Mean ± SD or Median (IQR)	Male (%)	Pack-Years Median (IQR)	Active Smokers (%)	Positive Results ^a at T ₀	Positive Results ^a by End of Screening Period	Lung Cancer Mortality RR (95% CI)
NLST	53,454	61 ± 5	59	48 (27)	48.1	7191 (27.3%)	10,287 (39.1%)	0.85 (0.75–0.96)
Depiscan	765	56 (NR)	71	30 (NR)	64	24%	NR	NR
DANTE	2472	64.6 ± 3.5	100	45 (28.5)	56	199 (15.6%)	471 (37%)	1.01 (0.70–1.44)
DLCST	4104	58 ± 5	55	36 (13)	75.3	155 (7.6%)	241 (11.8%)	1.03 (CI 0.66–1.60)
NELSON	15,822	59 (IQR: 6)	84	42 (19)	55	120 (1.6%)	2.0% (overall) 6.0% (at least 1 positive scan)	NR
ITALUNG	3206	61 ± 4	64	40 (NR)	66	426 (30.3%)	1044 (46.1%) ^b	0.70 (0.48–1.04)
MILD	4099	Annual: 57 (NR) Biennial: 58 (NR)	Annual: 68 Biennial: 69	Annual: 39 (NR) Biennial: 39 (NR)	Annual: 69 Biennial: 68	Annual: 177 (14%) Biennial: 158 (15%)	NR	Annual: 2.48 (0.98–6.29) Biennial: 1.24 (0.42–3.70)
LUSI	4052	58 (IQR: 5)	66	36 (18)	61	451 (22.2%)	805 (39.7%)	NR
UKLS	4055	67 ± 4	75	NR	39	536 (26.9%) ^c	NR, single screen	NR
LSS	3318	NR	58	54 (NR)	57.9	340 (20.5%)	573 (34.5%)	NR

CXR, Chest radiograph; DANTE, Detection and Screening of Early Lung Cancer by Novel Imaging Technology and Molecular Essays Trial; DLCST, Danish Lung Cancer Screening Trial; ITALUNG, Italian Lung Cancer Screening Trial; LLPv2, Liverpool Lung Project version 2; LSS, Lung Screening Study; LUSI, German Lung Cancer Screening Intervention Trial; MILD, Multicentric Italian Lung Detection Trial; NELSON, Nederlands-Leuvens Longkanker Screenings Onderzoek Study; NR, not reported; UKLS, United Kingdom Lung Screening Study; VDT, volume doubling time. See Table 1 legend for expansion of other abbreviations.

CI , Confidence interval; DANTE , Detection and Screening of Early Lung Cancer by Novel Imaging Technology and Molecular Essays Trial; DLCST , Danish Lung Cancer Screening Trial; IQR , interquartile range; ITALUNG , Italian Lung Cancer Screening Trial; LSS , Lung Screening Study; LUSI , German Lung Cancer Screening Intervention Trial; MILD , Multicentric Italian Lung Detection; NELSON , Nederlands-Leuvens Longkanker Screenings Onderzoek Study; NLST , National Lung Screening Trial; NR , not reported; RR , risk ratio; UKLS , United Kingdom Lung Screening Study. See Table 1 and 3 legends for expansion of other abbreviations.

a If benign features were present, the nodule was considered negative.

a Number of patients with positive results, not number of nodules; see previous table for definition of positive result in each study.

b The total number of positives from T ₀ to T ₄ is 1,044; unable to determine if this excludes positive results from the baseline (T0) screen.

c If including follow-up imaging at 1 year (since a single screen trial), the number would be 1015 (50.9).

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here