Although not the most frequent tumor, lung cancer is by far the most common cause of death from cancer . The main reason for this dreadful prognosis for lung cancer is the fact that more than 75% of the lung cancers are diagnosed at advanced tumor stages. These facts served as the rationale for undertaking a number of trials to prove that screening for lung cancer would reduce lung cancer mortality by diagnosing the tumors at an early stage, long before they would become symptomatic and when they could still be treated curatively.
Attempts to decrease lung cancer mortality by screening were already undertaken in the early 1950s using chest radiographs or chest radiographs combined with sputum analyses [2–7]. Although screening with chest radiographs led to an increased diagnosis of lung cancer at an early stage, compared with the control group, the actual aim of the trials – a reduction of lung cancer mortality – was not reached in any of these trials.
With volumetric computed tomography (CT), an imaging modality with a very high sensitivity for the detection of pulmonary nodules had become widely available. CT enables the detection of pulmonary nodules with a diameter of as little as 3 mm with a high level of confidence. But, only after the introduction of technical options to drastically reduce acquisition dose did CT become a potential tool for lung cancer screening.
To evaluate the potential of CT as a screening tool, a number of observational screening trials using low-dose CT (LD-CT) were conducted in the 1990s. The number of participants in these trials ranged from a few hundred to more than 30 000 . The trials showed that, with LD-CT screening, lung cancer could be detected in about 1% (range 0.4–2.7%) of patients at risk (heavy smokers) even in the baseline scan . More promising, most of the detected lung cancers (56–100%) were in stage I .
An increased detection rate for lung cancer in the early stages alone, however, is a poor surrogate measure for lung cancer mortality. To prove that LD-CT screening results in a reduction of lung cancer-related mortality, randomized controlled trials (RCTs) had to be conducted. For this reason, the National Cancer Institute (NCI) of the United States sponsored a large multicentric screening trial, the National Lung Screening Trial (NLST). In addition, a number of much smaller RCTs commenced in European countries (Table 1) [9–15]. The results of the NLST, published in 2011, proved, for the first time, that screening with LD-CT is able to significantly reduce lung cancer mortality when compared with a control arm that underwent chest radiography. However, important issues, such as optimal screening frequency and interval, the problem of overdiagnosis, the management of screening-detected nodules, and the selection of a ‘high-risk’ population, require further optimization to make screening cost-effective and maximize the benefit-harm ratio.
RESULTS OF THE NATIONAL LUNG SCREENING TRIAL
The NLST enrolled 53 454 participants in 33 centers all over the United States . Active or former smokers with a smoking history of 30 pack-years or more, and an age between 55 and 74 years, were included. Former smokers were only included if they quit smoking within the previous 15 years . Upon enrollment, participants were randomly assigned to one of two screening arms in which they were screened with LD-CT or chest radiographs. In each of the two screening arms, the participants were screened three times at yearly intervals.
In the fall of 2010, after a median follow-up of 6.5 years, the trial was stopped prematurely, as the study aim was already reached at that point in time. Overall, 1060 lung cancers were diagnosed in the LD-CT arm (corresponding to 645 per 100 000 person years) and 941 in the radiography arm (572 per 100 000 person years). During the observation period, 247 deaths from lung cancer per 100 000 person years were observed in the LD-CT group and 309 deaths per 100 000 person years in the radiography group. This corresponds to a relative reduction in lung cancer mortality of 20% (95% confidence interval 6.8–26.7; P = 0.004) with LD-CT screening . All-cause mortality was 6.7% lower in the LD-CT screening arm than in the control arm (95% confidence interval 1.2–13.6; P = 0.02). In this trial, 310 persons had to be screened with LD-CT to save one life .
RESULTS FROM THE EUROPEAN SCREENING TRIALS
Eight, partially still ongoing European randomized trials were initiated. Results of three trials have been published thus far, and not all of them were able to detect a significant reduction in cancer-related mortality. Even more astonishing is the fact that all three studies showed an increase in all-cause mortality in the screening arm. Underlying reasons for these discrepancies between the US trial and the European trials are not yet fully understood. Different inclusion criteria with respect to age and the severity of smoking history likely played a role, and different types and severities of comorbidities or geographic risk factors need to be considered, as well as the method of the diagnostic and therapeutic workup of screening-detected nodules. The NELSON (Nederlands-Leuvens Longkanker Screenings Onderzoek) trial is the largest of the European trials; results are expected in 2015/2016. A pooling of the European randomized trial data is under discussion right now; it is hoped that the findings will strengthen the evidence and address open questions.
The results of the European trials published thus far can be summarized as follows.
The Detection And screening of early lung cancer by Novel imaging TEchnology and molecular assays trial
The 3-year results from the Italian DANTE trial (Detection And screening of early lung cancer by Novel imaging TEchnology and molecular assays) were published 2 years prior to the publication of the NLST . Participants eligible for this trial were active or former smokers with a smoking history of at least 20 pack-years and an age of 60–74 years.
At enrollment, all participants had a chest radiograph and underwent a 3-day cytology testing. Patients were randomly assigned to either a screening arm with 5 yearly LD-CT screening rounds, or a control arm with no screening examinations.
After a follow-up period of 33 months, lung cancer was diagnosed in 60 of 1276 (4.7%) participants of the LD-CT screening arm and in 34 of 1196 (2.8%) participants of the control arm . Although the percentage of early lung cancers (stage I) was significantly higher in the LD-CT screening arm than in the control arm (54 vs. 34%; P <0.06), there was no statistically significant difference in lung cancer mortality between the two arms (1.6 vs. 1.7%).
The Multicentric Italian Lung Detection trial
The initial results of the second Italian randomized controlled screening trial, the MILD trial (Multicentric Italian Lung Detection) were published after a follow-up period of 4.4 years . In this trial, participants were randomized in either an arm with annual LD-CT screening (1186 participants), an arm with biennial LD-CT screening (1186 participants), or a control arm without any imaging (1723 participants). After a median follow-up of 4.4 years, lung cancer was diagnosed in 29 of the annually screened patients and 20 of the biennially screened participants . Again, most of the detected lung cancers (63%) were detected at stage I. During the observation period, 216 deaths from lung cancer per 100 000 person years were observed in the annually screened arm and 109 deaths per 100 000 person years in both the biennially screened arm and the control group. This difference in lung cancer mortality between the annually screened arm and the control group, however, did not reach statistical significance.
The Danish Lung Cancer Screening Trial
The Danish DLCST (Danish Lung Cancer Screening Trial) included active or former (abstinent for <10 years) smokers with a smoking history of at least 20 pack-years . Participants were randomized in two arms, an LD-CT screening arm with annual screening over 5 years, and a control group with no screening, with 2052 participants in each arm. As in the other randomized controlled screening trials, lung cancer was more frequently detected in the LD-CT screening arm than in the control group (69 vs. 24; P >0.001). The difference in lung cancer mortality, however, did not reach statistical significance.
The NLST is the first trial to provide evidence that LD-CT screening could reduce lung cancer mortality by a magnitude of 20%. This positive effect of LD-CT on lung cancer mortality, however, could not be confirmed by the initial results of three European screening trials [19–21]. To some extent, the lack of statistical significance for the results of LD-CT screening in the above-mentioned European trials is attributable to the fact that the trials included much smaller study groups, and thus, were statistically underpowered and the follow-up period was shorter. More results from European trials are expected in the next few years and pooling of the European RCT data may potentially strengthen the evidence.
Although many US guidelines recommend lung cancer screening , there is no doubt that further refinement of screening conditions is necessary to decrease the number of false-positive findings that require follow-up scans, and to decrease costs, unwanted side-effects, and overall radiation burden. A number of publications within the last few months have addressed the issue of developing risk models to better define the population at high risk for developing cancer, and who, therefore, would most profit from screening.
Management of screen-detected nodules
In general, a volume doubling time of the nodule within 400 days was considered the threshold that was indicative of malignancy. Lesion growth was assessed by manual diameter measurements in the NLST and by automatic volumetry in the NELSON trial. Three-dimensional volumetric measurements can be considered superior to two-dimensional diameter measurements in terms of accuracy and reproducibility because the whole nodule is analyzed and asymmetric growth can be assessed more accurately. It should be noted that the assessment of the growth of subsolid lesions has different requirements (volume and density increase), and no automatic Food and Drug Administration-approved tool is yet available.
The NLST trial followed a dichotomous interpretation algorithm: any noncalcified pulmonary nodule with a maximum diameter of 4 mm detected by LD-CT or any noncalcified pulmonary nodule detected by radiography was defined as positive. Based on this definition, 24% of all baseline LD-CT examinations were rated as positive. The overwhelming majority (96.4%) of the detected nodules eventually proved to be benign in nature. Only 3.6% of the positive screens were diagnosed as malignant. The vast majority of pulmonary nodules were evaluated by follow-up LD-CT examinations. In 2.6% of the patients, a surgical procedure was performed to evaluate a positive LD-CT screening result, 25% of which were proven to be benign. The risk of major complication or death for the evaluation of a false-positive result was reported to be low (4.5 and 4.1 per 10 000, respectively) .
A rather sophisticated two-step interpretation strategy was implemented for the management of positive screening results in the NELSON trial (Fig. 1) . In this trial, pulmonary nodules were stratified according to their volume into three groups. Based on this approach, the investigators could eventually define almost 98% of all nodules as negative with a negative predictive value of 99.9% . Nodules defined as positive made up only 2.6% of all lesions, with a positive predictive value of 35.5% . A recent retrospective analysis of the data from the NELSON trial revealed that, by lowering the cut-off of the volume doubling time at the initial screening to 232 days, the false-positive rate could be further decreased by 33% .
Definition of which nodules should be followed
The rate of false positives is mainly dependent on the threshold of what is defined as ‘positive’ in the baseline screening examination. In order to reduce the number of follow-up studies, a number of investigations have focused on morphologic criteria that would be helpful to definitively differentiate benign nodules from lesions that would require follow-up.
Based on the Dutch-Belgian NELSON trial, nodules smaller than 4 mm, and spherical and smooth, had a negligible risk of developing into cancers. So-called perifissural opacities that fulfilled certain morphologic criteria (adjacent to a pleural fissure or the interlobular septum, oval or triangularly shaped, <15 mm away from the pleura) represent very likely benign lymph nodules and need not be followed. Excluding those perifissural opacities from follow-up would result in a reduction of CT scans by 30%. A recent publication based on the findings of the Canadian trial proposed a risk model that included, among other factors, the anatomic location and the size and age of the patient, to predict the risk of malignancy, and thus allow for modification of the number and interval of follow-up examinations .
The diagnosis and resection of histologically proven lung cancer in LD-CT screening does not necessarily translate into improved lung cancer mortality. Lung cancers show a broad variety of biological behaviors. Whereas some grow very rapidly and metastasize early over time, others grow so slowly that they do not affect the patients’ life expectancy, and diagnosing the latter is referred to as ‘overdiagnosis’. Data from earlier trials suggest that overdiagnosis might be a relevant problem in lung cancer screening [26,27]. In the Mayo Lung Project, after 6 years of screening, 143 lung cancers were detected in the interventional arm (chest radiographs and sputum analysis), compared with 87 in the control arm . After a follow-up of 5 more years, 10 catch-up tumors were diagnosed in the control group. The excess of 46 lung cancers in the interventional arm was considered to reflect overdiagnosis and the risk of overdiagnosis was estimated to be as high as 51% . Other investigators concluded that, based on a volume doubling time of 400 days as a cut-off to discriminate overdiagnosed (indolent) cases from genuine cases, the proportion of overdiagnosed cases would be as low as 5% .
Currently there are, however, no accepted criteria by which to discriminate indolent tumors from genuine ones. Although a volume doubling time of more than 400 days is generally considered to be a good indicator of benignancy, data from the NLST indicate that growth between two points in time is not sufficient to predict future growth . Some lung cancers appear to grow not only exponentially, but also demonstrate a fairly flat growth curve . Some tumors may even remain stable in size for a long time before showing an accelerated growth .
Important strategies to diminish the detection of indolent disease (or inconsequential disease) include reducing the frequency of screening examinations, focusing on high-risk populations, and raising the threshold for recall and biopsy .
Definition of the screening population
The efficiency of any screening program depends on the definition of the population to be screened. In the NLST, the inclusion and exclusion criteria defined a cohort of people in whom the estimated risk of developing lung cancer within 10 years ranged from less than 2% to more than 20% . By applying more stringent criteria, it is likely that the effectiveness of screening can be further enhanced. The NLST and most European trials calculated the risk of developing lung cancer mainly by the age of the individual and his/her smoking history. In the ongoing British UK Lung Screen (UKLS) trial, the individual lung cancer risk of any candidate for screening is calculated by using a risk score developed by the Liverpool Lung Project . In this model, the estimation of lung cancer risk is based on age, sex, smoking duration, family history of lung cancer, history of nonpulmonary malignant tumor, history of pneumonia, and occupational exposure to asbestos . Only candidates with more than a 5% risk of developing lung cancer in 5 years are included.
Another approach for risk prediction was based on a retrospective analysis of the NLST data [33▪▪]. This prediction model included risk factors such as age, BMI, family history of lung cancer, pack-years of smoking, years since smoking cessation, and emphysema diagnosis [33▪▪]. By limiting screening to the 20% of those participants with the highest risk, the number of people who had to be screened to prevent one lung cancer death would be as low as 161 [33▪▪]. However, screening of the 20% of the participants with the lowest risk would prevent almost no cancer death [33▪▪], indicating the importance of defining appropriate selection criteria about which population to screen to optimize the risk-benefit and also cost-benefit ratio.
Screening intervals and duration
Another issue under ongoing discussion involves the optimal screening interval and duration of the screening. If the interval is too long, the probability of diagnosing more indolent, slowly growing tumors is higher than that with shorter screening intervals, with the burden of increasing the cumulative radiation dose and overall costs. Shorter intervals, on the contrary, increase the probability of diagnosing the more aggressive cancers. Most of the randomized controlled screening trials have screened in annual intervals with three to five screening rounds [14,17,19,21]. More information regarding the optimal length of the screening interval is expected from the Italian MILD trial and the Dutch-Belgian NELSON trial. In the MILD trial, participants were randomized in either an arm with annual LD-CT screening for 10 years, an arm with biennial LD-CT screening for 10 years, or a control arm . In the NELSON trial, the screening intervals were 1, 3, and 5 years after baseline . Further empirical studies of other screening intervals are needed to assess the benefit-harm balance of less frequent screening. A twice-yearly (biennial) screening is here of particular interest; this interval schedule was found to be equally or potentially more cost-effective than annual screening, although less effective in absolute terms using calculations of 10-year outcomes from the NLST and the UKLS trial.
Cumulative radiation dose
The risk of radiation-induced cancer from the cumulative radiation dose applied by regular LD-CT screening examinations and additional LD-CT, full dose CT, or PET-CT remains to be determined. In the NLST, the average effective dose per screening examination was 1.6 mSv for men and 2.1 mSv for women . Overall, however, due to the high number of positive screening results that required further evaluation, the average effective dose after three screening rounds was estimated to be as high as 8 mSv . On the basis of these data, it was estimated that, per 2500 persons screened, there would be one cancer death caused by radiation from imaging . As the improvement in overall mortality by LD-CT is in the range of 7% , the beneficial effect of screening was greater than the risk of radiation-induced cancer.
The NLST is the first study to provide evidence that screening with LD-CT can reduce lung cancer mortality by 20%. The encouraging results of the NLST, however, are awaiting confirmation from ongoing European trials. In addition, a number of open questions regarding the definition of persons at risk, the management of screening-detected nodules, or overdiagnosis need to be addressed further. For the time being, leading American scientific societies recommend offering LD-CT screening in the population at risk as investigated in the NLST [35–37]. In contrast to the study design of the NLST, some guidelines recommend performing annual LD-CT screening examinations until the age of 74 years  or 79 years . Of paramount importance is detailed counseling on the potential harms and benefits of screening [35,37]. Screening should be performed at experienced centers by a multidisciplinary team [35–38].
Conflicts of interest
There are no conflicts of interest.
REFERENCES AND RECOMMENDED READING
Papers of particular interest, published within the annual period of review, have been highlighted as:
- ▪ of special interest
- ▪▪ of outstanding interest
1. Siegel R, Naishadham D, Jemal A. Cancer statistics, 2012. CA Cancer J Clin. 2012; 62:10–29.
2. de Hoop B, Schaefer-Prokop C, Gietema HA, et al. Screening for lung cancer with digital chest radiography: sensitivity and number of secondary work-up CT examinations. Radiology. 2010; 255:629–637.
3. Brett GZ. The value of lung cancer detection by six-monthly chest radiographs. Thorax. 1968; 23:414–420.
4. Wilde J. A 10 year follow-up of semi-annual screening for early detection of lung cancer in the Erfurt County, GDR. Eur Respir J. 1989; 2:656–662.
5. Frost JK, Ball WC Jr, Levin ML, et al. Early lung cancer detection: results of the initial (prevalence) radiologic and cytologic screening in the Johns Hopkins study. Am Rev Respir Dis. 1984; 130:549–554.
6. Fontana RS, Sanderson DR, Taylor WF, et al. Early lung cancer detection: results of the initial (prevalence) radiologic and cytologic screening in the Mayo Clinic study. Am Rev Respir Dis. 1984; 130:561–565.
7. Kubik A, Polak J. Lung cancer detection. Results of a randomized prospective study in Czechoslovakia. Cancer. 1986; 57:2427–2437.
8. Pastorino U. Lung cancer screening. Br J Cancer. 2010; 102:1681–1686.
9. Infante M, Lutman FR, Cavuto S, et al. Lung cancer screening with spiral CT: baseline results of the randomized DANTE trial. Lung Cancer. 2008; 59:355–363.
10. van den Bergh KA, Essink-Bot ML, Bunge EM, et al. Impact of computed tomography screening for lung cancer on participants in a randomized controlled trial (NELSON trial). Cancer. 2008; 113:396–404.
11. Lopes Pegna A, Picozzi G, Mascalchi M, et al. Design, recruitment and baseline results of the ITALUNG trial for lung cancer screening with low-dose CT. Lung Cancer. 2009; 64:34–40.
12. Pedersen JH, Ashraf H, Dirksen A, et al. The Danish randomized lung cancer CT screening trial: overall design and results of the prevalence round. J Thorac Oncol. 2009; 4:608–614.
13. Pastorino U, Bellomi M, Landoni C, et al. Early lung-cancer detection with spiral CT and positron emission tomography in heavy smokers: 2-year results. Lancet. 2003; 362:593–597.
14. Becker N, Motsch E, Gross ML, et al. Randomized study on early detection of lung cancer with MSCT in Germany: study design and results of the first screening round. J Cancer Res Clin Oncol. 2012; 138:1475–1486.
15. Baldwin DR, Duffy SW, Wald NJ, et al. UK Lung Screen (UKLS) nodule management protocol: modelling of a single screen randomised controlled trial of low-dose CT screening for lung cancer. Thorax. 2011; 66:308–313.
16. National Lung Screening Trial Research Team The National Lung Screening Trial: overview and study design. Radiology. 2011; 258:243–253.
17. Aberle DR, Adams AM, Berg CD, et al. Reduced lung-cancer mortality with low-dose computed tomographic screening. N Engl J Med. 2011; 365:395–409.
18. Bach P, Mirkin J, Oliver T, et al. Benefits and harms of CT screening for lung cancer: a systematic review. JAMA. 2012; 307:2418–2429.
19. Infante M, Cavuto S, Lutman FR, et al. A randomized study of lung cancer screening with spiral computed tomography: three-year results from the DANTE trial. Am J Respir Crit Care Med. 2009; 180:445–453.
20. Pastorino U, Rossi M, Rosato V, et al. Annual or biennial CT screening versus observation in heavy smokers: 5-year results of the MILD trial. Eur J Cancer Prev. 2012; 21:308–315.
21. Saghir Z, Dirksen A, Ashraf H, et al. CT screening for lung cancer brings forward early disease. The randomised Danish Lung Cancer Screening Trial: status after five annual screening rounds with low-dose CT. Thorax. 2012; 67:296–301.
22. Field JK, Oudkerk M, Pedersen JH, Duffy SW. Prospects for population screening and diagnosis of lung cancer. Lancet. 2013; 382:732–741.
23. van Klaveren RJ, Oudkerk M, Prokop M, et al. Management of lung nodules detected by volume CT scanning. N Engl J Med. 2009; 361:2221–2229.
24. Heuvelmans MA, Oudkerk M, de Bock GH, et al. Optimisation of volume-doubling time cutoff for fast-growing lung nodules in CT lung cancer screening reduces false-positive referrals. Eur Radiol. 2013; 23:1836–1845.
25. McWilliams A, Tammemagi MC, Mayo JR, et al. Probability of cancer in pulmonary nodules detected on first screening CT. N Engl J Med. 2013; 369:910–919.
26. Marcus PM, Bergstralh EJ, Zweig MH, et al. Extended lung cancer incidence follow-up in the Mayo Lung Project and overdiagnosis. J Natl Cancer Inst. 2006; 98:748–756.
27. Kubik A, Parkin DM, Khlat M, et al. Lack of benefit from semi-annual screening for cancer of the lung: follow-up report of a randomized controlled trial on a population of high-risk males in Czechoslovakia. Int J Cancer. 1990; 45:26–33.
28. Welch HG, Black WC. Overdiagnosis in cancer. J Natl Cancer Inst. 2010; 102:605–613.
29. Yankelevitz DF, Kostis WJ, Henschke CI, et al. Overdiagnosis in chest radiographic screening for lung carcinoma: frequency. Cancer. 2003; 97:1271–1275.
30. Lindell RM, Hartman TE, Swensen SJ, et al. 5-Year lung cancer screening experience. Chest. 2009; 136:1586–1595.
31. Takashima S, Sone S, Li F, et al. Indeterminate solitary pulmonary nodules revealed at population-based CT screening of the lung: using first follow-up diagnostic CT to differentiate benign and malignant lesions. Am J Roentgenol. 2003; 180:1255–1263.
32. Esserman LJ, Thompson IM Jr, Reid B. Overdiagnosis and overtreatment in cancer: an opportunity for improvement. JAMA. 2013; 310:797–798.
33▪▪. Kovalchik SA, Tammemagi M, Berg CD, et al. Targeting of low-dose CT screening according to the risk of lung-cancer death. N Engl J Med. 2013; 369:245–254.
This study investigated the benefits and harms of LD-CT screening according to lung cancer risk and proposes improved inclusion criteria to optimize the results of lung cancer screening.
34. Larke FJ, Kruger RL, Cagnon CH, et al. Estimated radiation dose associated with low-dose chest CT of average-size participants in the National Lung Screening Trial. Am J Roentgenol. 2011; 197:1165–1169.
35. Detterbeck FC, Mazzone PJ, Naidich DP, Bach PB. Screening for lung cancer: diagnosis and management of lung cancer, 3rd ed: American College of Chest Physicians evidence-based clinical practice guidelines. Chest. 2013; 143:e78S–e92S.
36. Jaklitsch MT, Jacobson FL, Austin JHM, et al. The American Association for Thoracic Surgery guidelines for lung cancer screening using low-dose computed tomography scans for lung cancer survivors and other high-risk groups. J Thorac Cardiovasc Surg. 2012; 144:33–38.
37. Wender R, Fontham ET, Barrera E Jr, et al. American Cancer Society lung cancer screening guidelines. CA Cancer J Clin. 2013; 63:107–117.
38. Field JK, Smith RA, Aberle DR, et al. International Association for the Study of Lung Cancer Computed Tomography Screening Workshop 2011 report. J Thorac Oncol. 2012; 7:10–19.