Lung Cancer Screening | 4 |
INTRODUCTION
Lung cancer is the leading cause of cancer-related death in the United States (1). The association between tobacco use and lung cancer has been reported since the 1950s, and smoking cessation is the key intervention for preventing lung cancer (2). However, tobacco cessation is difficult to achieve, with only 20% to 30% of smokers who attempt to quit remaining tobacco free for at least 2 years (3). The wide use of tobacco is the main reason that lung cancer continues to burden society. At the time of diagnosis, the majority of patients with lung cancer have stage III or IV disease, and over half have distant metastases (4). Since most early stage lung tumors are asymptomatic, only 15% of lung cancers are localized at the time of detection (1). Due to the high frequency of late stage at diagnosis, the 5-year survival rate for lung cancer is only 16%, with little recent improvement (5). The high prevalence and mortality of lung cancer highlights the great impact that successful screening could have on this disease.
Cancer Screening Bias
Lead-time bias and overdiagnosis present potential problems for lung cancer screening. Lead-time bias occurs when screening detects a cancer earlier, but does not alter time of death. In other words, screening improves survival time from diagnosis to death, but does not change the mortality (the number of people dying of the disease). Overdiagnosis refers to the diagnosis of a slow-growing cancer that would be unlikely to progress in a clinically relevant way and would not have impacted on the patient’s longevity. Thus, these cancers would not have presented as disease if it were not for the screening test. Overdiagnosis is a particular concern when a screening test leads to invasive diagnostic testing or therapy, since potential complications from these procedures may cause more harm than good.
Conflict of interest: Dr. Jett has a research grant paid to National Jewish Health from Oncimmune, Inc. (Biomarker).
CHEST RADIOGRAPHY FOR LUNG CANCER SCREENING
Early Studies
In the early 1980s, several clinical trials evaluating chest radiography (chest x-ray) and sputum cytology for lung cancer screening reported no reduction in lung cancer mortality (6). The Mayo Lung Project was one of the largest early trials to study lung cancer screening. From 1971 to 1983, this study enrolled 9,211 men 45 years of age or older who were smoking more than a pack of cigarettes per day. Participants were randomized into two groups: (a) those undergoing chest radiography and sputum cytology every 4 months for 6 years, or (b) those getting standard of care, which included annual chest radiography. Despite a higher rate of detection of lung cancer, lower stage at diagnosis, and greater candidacy for surgical resection in the intervention group, there was no difference in all-cause or lung cancer-specific mortality between the two groups (7). The detection of lower stage cancers and an improvement in survival time from diagnosis to death without impacting mortality demonstrated lead-time bias in this study. Long-term follow-up reported that the higher prevalence of lung cancer in the intervention group (585 vs. 500 lung cancers) persisted long after completion of the screening period, suggesting overdiagnosis (8).
The Prostate, Lung, Colorectal, and Ovarian Trial
The prostate, lung, colorectal, and ovarian (PLCO) trial, published in 2011, gave further support to prior studies demonstrating a lack of utility of chest radiography as a lung cancer screening tool (9). The PLCO study enrolled 154,901 subjects between the ages of 55 and 74 years with the goal of determining the impact of screening for prostate, lung, colorectal, and ovarian cancer on mortality. The selection criteria did not include a history of smoking. Subjects in the control arm underwent imaging only if the treating physician felt it was indicated, while those in the intervention arm underwent screening with a chest radiograph annually for 4 years. All subjects remained in follow-up for 13 years. At the end of the study period, there were no differences between the arms in lung cancer detection rate, clinical stage at time of diagnosis, or mortality. The lack of mortality benefit was also noted in high-risk subjects with a smoking history of at least 30 pack-years, either current smokers or those who had quit within the past 15 years. Thus, this study confirms that chest radiography is not an effective method for lung cancer screening in the general population.
COMPUTED TOMOGRAPHY (CT) FOR LUNG CANCER SCREENING
Early Studies
The negative results of studies of chest radiography led to the evaluation of CT for lung cancer screening. Helical, multidetector row CT scanners have much greater sensitivity than standard chest radiographs for lung abnormalities, but repeated imaging results in exposure to larger amounts of radiation. The advent of low-dose CT (LDCT) lessened this concern due to the use of much lower doses of radiation than standard chest CT scans (<2 mSv vs. 7 mSv).
In 2005, the Mayo Clinic published a 5-year prospective study using annual LDCT as a screening tool in high-risk patients (10). Each of the 1,520 participants had at least a 20 pack-year smoking history and had been smoking within the past 10 years. Data from the earlier Mayo Lung Project with chest radiography was used as the control arm for comparison to LDCT. Sixty-eight lung cancers were diagnosed, 31 on the initial LDCT screen, 34 on subsequent screens, and three in the screening interval. Compared to historical controls using chest radiography, more of the lung cancers found on incidence LDCT were stage IA (47% vs. 21%) and fewer were stage III/IV (33% vs. 45%). In addition, among the lung cancers detected on subsequent screens, 61% were stage I tumors (10). Overall, the rate of early-stage cancer detected by LDCT screening was significantly higher than the 15% early-stage detection rate noted in current practice (1), suggesting a favorable stage shift with LDCT screening.
The Early Lung Cancer Action Program (ELCAP) was a larger analysis of 31,567 subjects who underwent annual LDCT screening (11). Once again, screen-detected lung cancers were mostly stage I (85%), but the lack of a control arm undermined the ability to detect any improvement in mortality (11). One early effort at randomization was the Detection and Screening of Early Lung Cancer by Novel Imaging Technology and Molecular Assays (DANTE) trial. The DANTE study enrolled 2,450 men from 60 to 74 years old with at least a 20 pack-year smoking history who were randomized to either annual CT screening or no screening. Long-term follow-up data from this study showed that lung cancer-specific mortality was similar in both arms (LDCT, 543 per 100,000 person-years vs. control, 544 per 100,000 person-years) (12). Although the DANTE study was limited by its small sample size and unisex enrollment, LDCT did not appear to improve lung cancer-specific or all-cause mortality.
The National Lung Screening Trial (NLST)
The randomized National Lung Screening Trial (NLST) enrolled 53,454 men and women between 55 and 75 years of age with at least a 30 pack-year smoking history and use of tobacco within the last 15 years at 33 centers in the United States. Participants were randomized to undergo either LDCT or standard chest radiography annually for 3 years. A positive finding was defined as any noncalcified nodule ≥4 mm in longest dimension on axial CT images or any noncalcified density on chest radiography. More subjects had positive findings by LDCT than chest radiography (24.2% vs. 6.9%), but the majority of nodules in both groups were benign: 96.4% with LDCT vs. 94.5% with chest radiography (13). Despite the large number of benign nodules, the malignancy detection rate was higher with LDCT than with chest radiography. During the 3-year screening period, LDCT detected 649 lung cancers as compared to 279 with chest radiography. This trend continued through the follow-up period (median follow-up of 6.5 years) with 1,060 lung cancers diagnosed in the LDCT arm and 941 in the chest radiography arm, suggesting potential overdiagnosis during the LDCT screening period. There was also a notable stage shift between the two arms. In the LDCT arm, 40% of the cancers detected were stage IA and 22% were stage IV as compared to 21% stage IA and 36% stage IV in the chest radiography arm (13).
For the first time, the NLST revealed a significantly lower lung cancer-specific mortality rate with LDCT screening with a relative risk reduction of 20.3% compared to chest radiography (247 deaths per 100,000 person-years vs. 309 deaths per 100,000 person-years) (13). In other words, 320 individuals need to be screened to prevent one death from lung cancer. All-cause mortality was also significantly lower in the LDCT arm (6.7% risk reduction; 95% CI, 1.2 to 13.6; P = .02) (13). It should be noted that the NLST did not dictate a standardized diagnostic algorithm for subjects with positive screening results. Recommendations on further diagnostic evaluation or timing of follow-up scans were provided to subjects and their referring health care providers based on the specific findings of screening studies. Despite this lack of uniformity in the subsequent evaluation of abnormal findings, both the use of invasive procedures in patients with ultimately benign lesions and the complication rate from such interventions were very low.
Despite the clear mortality benefit demonstrated in the NLST, the use of LDCT as a screening tool continues to be debated due to the high false-positive rate and unclear cost-effectiveness. There is also concern about the potential for overdiagnosis with LDCT screening. Adenocarcinoma was the most commonly detected histologic subtype of lung cancer diagnosed, with bronchioloalveolar carcinoma (BAC; adenocarcinoma in situ, [AIS]) being found much more commonly with LDCT than with chest radiography (14.7% vs. 4.7%) (13). Since BAC/AIS is a slow growing, noninvasive tumor, there is concern that its diagnosis would not impact on mortality, but could lead to potential harm due to unnecessary interventions (14).
On-Going Studies
The Dutch-Belgian Randomized Lung Cancer Screening Trial (NELSON) is a randomized, controlled trial comparing LDCT screening to no screening in high-risk participants between 50 and 75 years of age (15). It differs from the NLST trial in two major ways: (a) high-risk tobacco exposure is defined as a smoking history of at least 15 cigarettes per day for 25 years or at least 10 cigarettes per day for 30 years, and former smokers must have quit less than 10 years ago; and (b) radiologic criteria for a positive scan include volumetric nodule changes over time with the goal being to reduce subsequent imaging and invasive diagnostic testing. The NELSON trial is powered to detect a 25% decrease in lung cancer-specific mortality after 10 years. Since 2003, 15,822 subjects have been randomized. Preliminary results from 7,155 subjects screened with LDCT found 196 screen-detected cancers in 187 subjects, for a sensitivity of 84.6%, and specificity of 98.6% (15). Of lung cancers detected with LDCT, 64% were stage I, a finding that is similar to that reported in the NLST (16). Several other randomized trials of LDCT screening for lung cancer are on-going, including the United Kingdom Lung Cancer Screening Trial (UKLS), the Multicentric Italian Lung Detection Trial (MILD), the Italian Lung Cancer CT Screening Trial (ITALUNG), and the German Lung Cancer Screening Intervention Study (LUSI).
CT Screening Recommendations
The U.S. Preventive Services Task Force (USPSTF) published a statement on the utility of LDCT lung cancer screening in 2013 that was based on the results of the NLST and on mathematical modeling. The USPSTF recommended “annual screening for lung cancer with low-dose CT in adults aged 55 to 80 years who have a 30 pack-year smoking history and currently smoke or have quit within the past 15 years” (17). The overall strength of this recommendation was “Grade B,” meaning that the data created a high certainty of at least a moderate net benefit or a moderate degree of certainty of a moderate to substantial net benefit.
In 2015, the U.S. Centers for Medicare and Medicaid Services (CMS) determined that Medicare would cover the cost of lung cancer screening with annual LDCT in adults from 55 to 77 years of age who have a 30 pack-year smoking history and currently smoke or have quit within the past 15 years. Medicare coverage also includes a visit with a physician or qualified nonphysician practitioner for counseling and shared decision making on the benefits and risks of lung cancer screening. Importantly, the CMS statement requires counseling on smoking cessation as part of the screening process.
CT SCREENING INTERPRATATION GUIDELINES
Since the invention of CT in 1972, advances in technology have resulted in the development of helical and multidetector row CT scanners that decrease scan time and low-dose scanners that still provide high-resolution images. Both of these developments were essential for the progress of lung cancer screening since they allow faster, more convenient imaging with high resolution and less radiation exposure (18). Another key advance in CT-based lung cancer screening is the development of a standardized system of cancer risk assessment that could guide recommendations for follow-up imaging.
The international, multidisciplinary Fleischner Society has published guidelines to evaluate CT-detected lung nodules in both high- and low-risk patients (19). The Fleischner guidelines offer a detailed, systematic approach for the evaluation of incidentally discovered, not screen-detected, pulmonary nodules. These recommendations center on several factors affecting the risk of malignancy, including size, density of attenuation, border characteristics, calcification, and growth over time (20). It should be noted, that this system was intended for data obtained from standard-dose CT, not LDCT, in a population that is not undergoing screening. For subjects undergoing annual screening with LDCT more selective guidelines are needed to avoid excessive follow-up imaging of lower-risk findings.
Lung Imaging Reporting and Data System (Lung-RADS™)
The American College of Radiology developed the Lung Imaging Reporting and Data System (Lung-RADS) to standardize pulmonary nodule reporting and management within the context of LDCT screening (www.acr.org/Quality-Safety/Resources/LungRADS). The Lung-RADS differs from Fleischner guidelines in that it assumes future screening LDCT scans will be completed on a yearly basis (Table 4.1). In the Lung-RADS system, a positive nodule is defined as ≥6 mm, as compared to ≥4 mm in the Fleischner guidelines. McKee et al. evaluated the performance of Lung-RADS in 2,180 patients by comparing institutional standard LDCT interpretation to a retrospective reevaluation of the scans using Lung-RADS (21). Overall, the positive predictive value (PPV) for malignancy was 17.3% with Lung-RADS vs. 6.9% with standard NLST guidelines. This improvement in PPV was largely due to the higher cutoff of 6 mm for defining a positive finding in the Lung-RADS leading to a decrease in false-positive scans. Pinsky et al. retrospectively applied Lung-RADS criteria to the NLST’s baseline scan and found not only a reduction in the false-positive rate (12.8% for Lung-RADS vs. 26.6% for NLST interpretations), but also a decrease in sensitivity (84.9% for Lung-RADS vs. 93.5% for NLST) (22). In addition, this study reported a marked reduction in the number of subsequent CT scans required when Lung-RADS was applied, with the avoidance of an estimated 5,707 follow-up scans. Currently, Lung-RADS is recommended for use in lung cancer screening programs, but prospective studies of the performance of Lung-RADS in real-world situations are still needed.
PRESCREENING LUNG CANCER RISK ASSESSMENT
It is extremely important to select the appropriate population to consider for LDCT lung cancer screening, since the vast majority of CT-detected pulmonary nodules will be benign even in a high-risk cohort (10–13). The Pan-Canadian Early Detection of Lung Cancer Study validated this point by reporting that only 144 out of 12,029 pulmonary nodules detected on CT screening scans were malignant (23). The goal of this study was to develop an accurate model that could predict the probability that a nodule detected on LDCT screening would be malignant. The authors concluded that the ideal screening subject should have a higher probability of developing a malignant pulmonary nodule than the general population. Using such a risk prediction model to identify candidates for screening would increase the PPV of the screening test.
A good risk prediction model should have appropriate discrimination, calibration, and accuracy. With regard to lung cancer, discrimination is the ability of the model to distinguish those who died from lung cancer from those who did not. Calibration is the probability that an individual will die from cancer in a defined period of time (e.g., a 30% chance of death over the next 10 years). The closer this risk assessment is to the observed death rate, the more reliable the model. Accuracy refers to how well the risk prediction model functions, as measured by internal and external validation. Internal validation is determined in the patient cohort from which the model was derived, while external validation comes from applying the model to other cohorts.
Validated lung cancer risk assessment tools include the Tammemagi Model, the Bach Model, the Spitz Model, and the Liverpool Lung Project Model. Tammemagi et al. used the prospective data from the PLCO trial to compare a cohort of 70,962 nonsmokers to a cohort of 38,254 subjects who had a smoking history (24). The parameters included in the model were age, socioeconomic status, education, body mass index, recent chest radiography, smoking status, pack-years, and years of being tobacco free. In the validation cohort, the model demonstrated high reliability (area under the curve, 0.784) (www.brocku.ca/lung-cancer-risk-calculator).
All models evaluating lung cancer risk include length, quantity, and quit time of tobacco use. The large sample size of the NLST afforded the development of an absolute-risk prediction model for lung cancer mortality by retrospectively applying these characteristics (25). The prediction model for the LDCT arm stratifies the 5-year risk of death from lung cancer into five groups spanning low to high risk. When this risk stratification model was applied to the screened cohort, the number of false-positive results declined from 1,648 in the lowest risk group to 65 in the group having the highest risk of lung cancer-related death. Since screening of the lowest risk group prevented only 1% of lung cancer deaths, this risk-assessment model suggests that lung cancer screening may be indicated only in higher risk subjects.
Tobacco use is the strongest risk predictor for the development of lung cancer. Tanner et al. demonstrated an association between smoking and increased mortality in a subgroup analysis of the NLST cohort (26). This study found that current smokers had a higher risk of lung cancer-related death than former smokers (HR 2.14–2.29). Based on these risk-assessment models, multiple organizations have released statements on the minimum tobacco exposure required to be an optimal candidate for LDCT lung cancer screening (Table 4.2).
Table 4.2 Organizational recommendations on lung cancer screening | ||
Organization | Recommendation | Publication date |
American Cancer Society (27) | Annual LDCT screening in adults aged 55–74, who have at least a 30 pack-year smoking history and currently smoke or have quit within the past 15 years. Thorough informed consent and shared decision making should be offered prior to screening, along with smoking cessation counseling. | March 2013 |
American College of Chest Physicians/American Society for Clinical Oncology (28) | Annual LDCT screening in adults aged 55–74, who have at least a 30 pack-year smoking history and currently smoke or have quit within the past 15 years. Screening should only be performed in settings capable of offering comprehensive care similar to that provided in the NLST trial. | May 2013 |
U.S. Preventative Services Task Force (29) | Annual LDCT screening in adults aged 55–80, who have at least a 30 pack-year smoking history and currently smoke or have quit within the past 15 years. Screening should be discontinued once a patient has not smoked for more than 15 years, develops a life expectancy-limiting illness, or refuses curative thoracic surgery, if indicated. | December 2013 |
European Respiratory Society/European Society of Radiology (30) | Annual LDCT screening in adults aged 55–80, who have at least a 30 pack-year smoking history and currently smoke or have quit within the past 15 years. Multidisciplinary medical resources, shared decision making, and smoking cessation interventions should be offered. | April 2015 |
RISKS OF LUNG CANCER SCREENING
Lung cancer screening with LDCT carries multiple risks, including cumulative radiation exposure, complications from diagnostic procedures, financial cost, and anxiety from incidental findings.
Radiation Exposure
Although one LDCT scan results in only ≤2 mSv of radiation exposure as compared to 7 mSv with standard chest CT (31), there remains concern regarding the cumulative effect of serial scans. The risk of radiation-related malignancy versus the risk of death from lung cancer was evaluated in a systematic review, which concluded that the benefit of screening in higher-risk subjects outweighs the risk of cumulative radiation exposure (32).
Unnecessary Interventions
About 27% of the NLST-screened population underwent additional imaging or invasive procedures, including thoracotomy, thoracoscopy, mediastinoscopy, bronchoscopy, and needle biopsy. Within the LDCT arm, 16 participants (0.06%) died within 60 days of their procedure, and 10 of these may have died from lung cancer rather than from complications of the procedure (13). Although the rate of serious complications was already low in the NLST, defining a higher cutoff point for a positive nodule, as is being done with Lung-RADS, should lead to even fewer interventions and thus, fewer complications.
Cost
The cost of LDCT screening is a major concern given the number of additional tests that will result from its implementation. A cost-effectiveness analysis of the NLST found that LDCT screening costs an additional $1,631 per person as compared to no screening (33). The estimated life years per person gained through LDCT screening was calculated to be 0.0316, which led to a predicted value of $52,000 per life year gained. This cost is similar to that incurred by screening for breast or colorectal cancer. Another study using simulation modeling showed that the cost in a commercially insured population for LDCT lung cancer screening was actually lower than that of screening for other cancers (34). This study reported that the cost of LDCT screening was $0.76 per member per month (PMPM) as compared to $0.95, $1.10, and $2.50 for colorectal, cervical, and breast cancer screening, respectively (34).
If the screening program incorporates smoking cessation counseling, cost-effectiveness improves by 20% to 45% (34). The improved cost-effectiveness stems from an increase in the number of quality-adjusted life years (QALY) saved. The NLST data for current smokers (n = 15,489) was analyzed using longitudinal regression models to predict annual smoking cessation rates in relation to abnormalities detected on LDCT scans. There was a statistically significant association between smoking cessation and the type of abnormality detected. Specifically, a subject with a LDCT revealing a new or growing nodule concerning for cancer was at lower risk of remaining a continuing smoker (OR 0.66, 95% CI 0.61–0.72; P < .001) (35). Several other studies, including the NELSON trial, evaluated the effect of LDCT screening on smoking cessation. To date, there has been no consistently detected difference in cessation rates between those undergoing screening versus those in the control arm of a prospective study (36). Smoking cessation counseling is a requirement for CMS reimbursement, but clearly, more effective measures are needed to achieve higher smoking cessation rates.
Incidental Findings
Another area of concern regarding LDCT screening is the identification of abnormalities unrelated to lung cancer. The goal of the NLST was to decrease lung cancer mortality by detecting cancer at an early, curable stage. However, LDCT also identified other abnormalities, such as coronary calcifications, thoracic aortic aneurysms, and thyroid masses, with greater frequency than chest radiography (7.5% vs. 2.1%) (13). These data suggest that LDCT screening would result in more testing and referrals to address these incidental findings, increasing the total cost of care and the potential for harmful complications from added interventions.
LUNG CANCER BIOMARKERS
A biomarker is any biological substance indicating the risk, presence, or activity of a particular disease. Sources of biomarkers include breath, sputum, urine, and blood (37). A classic example of a biomarker is cholesterol for the risk assessment of coronary artery disease. The identification and validation of a biomarker for the risk assessment of lung cancer would be invaluable. Additionally, with the growing implementation of lung cancer screening, there is great interest in defining biomarkers that could distinguish lung cancer from benign pulmonary nodules.
Pepe et al. described the five phases of biomarker development (38). Very few biomarkers for the detection of lung cancer have progressed to prospective screening (phase 4), and none have demonstrated a reduction of the burden of disease (phase 5). While the list of putative lung cancer biomarkers is extensive, the most promising blood biomarkers evaluated for clinical use include micro-RNA, proteins, and autoantibodies.
Micro-RNA
Micro-RNA (miRNAs) are short noncoding RNAs that are released by tumor cells or the tumor microenvironment into the circulation and are highly stable and quantifiable in plasma or serum (39). Sozzi et al. developed a plasma-based miRNA signature classifier (MSC) consisting of 24 miRNAs, which they tested in a validation set of 69 patients with screen-detected lung cancer and 870 control subjects enrolled in the Multicenter Italian Lung Detection Trial (40). The MSC had a sensitivity and specificity for lung cancer detection of 87% and 81%, respectively, with a PPV of 27% and negative predictive value (NPV) of 99%. The LDCT arm in this study had a false-positive rate of 19.4% with LDCT alone, while the combination of LDCT and MSC reduced the false-positive rate to 3.7% (40). Prospective screening trials using a variety of miRNA assays are currently underway.
Protein Signatures
Blood protein classifiers measure the presence of proteins associated with various lung carcinomas. A panel of 11 proteins was tested in a retrospective validation study evaluating indeterminate pulmonary nodules 8 to 30 mm in diameter. Using a cancer prevalence estimate of 23%, this multicenter, case-control study of 141 patients with pulmonary nodules, including 78 cancers, identified likely benign nodules with a 90% NPV, and 26% PPV (41). This protein-based classifier is now being evaluated in a prospective trial in over 700 patients with noncalcified nodules. Another panel of seven proteins was developed in a cohort of 94 patients with non-small cell lung cancer and 269 long-term smokers with benign pulmonary nodules (42). The test performed better for squamous cell cancer and the utility of this test to improve the PPV of LDCT screening is currently being explored.
Autoantibodies
Autoantibodies are identifiable in patients with a wide variety of cancers and may be present months to years before clinical diagnosis. Classic examples of autoantibodies in lung cancer are the anticalcium channel antibodies in Lambert–Eaton myasthenic syndrome and the antineuronal nuclear antibody (anti-Hu) in neurologic paraneoplastic syndromes associated with small cell lung cancer. Thus, detection of autoantibodies could be a potential avenue for earlier malignancy detection. A panel of autoantibodies to lung cancer-associated antigens (Early CDT Lung) was validated in case-control studies that demonstrated a sensitivity of 36% to 39% and specificity of 89% to 91% (43). A clinical audit of the Early CDT Lung panel was performed in 1,613 patients at high risk for lung cancer, some of whom had indeterminate pulmonary nodules (44). Six-month follow-up was performed for all participants. Sixty-one (4%) were diagnosed with lung cancer and 25 of these were blood test-positive (true positive), resulting in a sensitivity of 41%. Thirty-six lung cancers had a negative biomarker test (false negative), resulting in a specificity of 87%. A positive Early CDT Lung test was associated with a five-fold increase in the chance of having lung cancer. Among lung cancers with a positive test, 57% were Stage I or II (43). Currently, two large prospective trials evaluating the clinical utility of the Early CDT Lung biomarker assay are underway (45,46).
Biomarker Summary
While LDCT screening results in a 20% reduction in lung cancer-specific mortality, it is estimated that only 25% to 30% of patients with lung cancer in the United States would meet the current criteria for LDCT screening (47). Thus, a validated biomarker assay could have a substantial impact by identifying other lung cancers when they are asymptomatic or in an early stage. Biomarkers could also be used to identify high-risk individuals who should undergo LDCT screening outside of the current guidelines. At this time, there are no standard biomarkers that have been fully validated for use in clinical practice for determining the risk of lung cancer or for delineating benign from malignant pulmonary nodules.
CONCLUSION
Lung cancer screening has developed beyond a topic of research interest to one with practical utility. Nevertheless, it is important to continue to refine the screening system in order to yield more precise data. The implementation of LDCT screening must include good communication with patients regarding the risks and benefits of this intervention. A screening program should also be committed to smoking cessation counseling, as this will have the greatest impact on the incidence of lung cancer in any population and will improve the cost-effectiveness of the screening program. A multidisciplinary team is required to guide the choice of diagnostic interventions for screen-detected pulmonary nodules. This team should include radiologists, radiation oncologists, thoracic surgeons, medical oncologists, pathologists, and pulmonologists. Hopefully, improvements in lung cancer risk assessment with modeling, derived not only from patient demographics and characteristics but also from validated biomarker assays, will afford more precise determinations of who will benefit from screening.
In summary, the positive results of the NLST have led to a cascade of approvals from a variety of national and international organizations (Table 4.2) and an agreement for financial compensation through the CMS in the United States. The current recommendation is to offer annual screening for lung cancer with LDCT in adults age 55 to 77 years who have at least a 30 pack-year smoking history and currently smoke or have quit within the past 15 years. Of course, the potential risks of screening need to be balanced against the potential benefits, such as harm that may result from procedures done on those who ultimately do not have cancer. However, as demonstrated in the NLST, such complications are relatively rare. Overdiagnosis is also a concern since LDCT will discover clinically insignificant malignancies that create anxiety and potential morbidity for the patient, and added costs for the health system.
Counseling patients before screening on the potential outcomes will help with coping and allow rational decision making; thus, counseling in a formalized visit with a provider is mandated by CMS. Financial cost to the health system can be decreased through smoking cessation counseling and by limiting screening to those at the highest risk for lung cancer. In addition, annual LDCT screening should stop once a patient has not smoked for 15 years or develops a health issue that substantially limits his or her life expectancy or reaches the upper age limit of 77 years. The CMS guidelines for lung cancer screening are presented in Table 4.3. Adherence to these mandates is required by any program in the United States that wishes to be reimbursed for lung cancer screening.
Table 4.3 CMS lung cancer screening guidelines | ||
Patient criteria | Provider criteria | Facility criteria |
• Age 55–77 years • No signs or symptoms of lung cancer • Tobacco smoking history of at least 30 pack-years • Current smoker or one who has quit smoking within the last 15 years • Willingness to undergo diagnostic or curative procedures/surgery, if indicated • Willingness to adhere to annual LDCT screening as long as indicated | • Written order supplied for LDCT scan • Lung cancer screening counseling and shared decision-making visit that is furnished by a physician or qualified nonphysician practitioner • Counseling that includes benefits and harms of screening, follow-up diagnostic testing, overdiagnosis, false-positive rate, total radiation exposure, and impact of comorbidities on ability to undergo diagnostic and treatment procedures • Counseling on importance of adherence to annual lung cancer LDCT screening • Counseling on the importance of maintaining cigarette smoking abstinence if former smoker; or the importance of smoking cessation if current smoker; furnishing of information about tobacco cessation interventions | • Performs LDCT with volumetric CT dose index of ≤3.0 mGy for standard size patients with appropriate reductions for smaller patients and appropriate increases for larger patients • Utilizes a standardized lung nodule identification, classification, and reporting system • Collects and submits data to a CMS-approved registry for each LDCT lung cancer screening performed • All CMS-approved registries must have the capacity and capability to collect data from any Medicare-eligible imaging facility that furnishes lung cancer screening with LDCT |
Source: Adapted from www.cms.gov/Medicare/Medicare-General-Information/MedicareApprovedFacilitie/Lung-Cancer-Screening-Registries.html. | ||
