Benefits and Risks of Lung Cancer Screening

Lung cancer is the third most common cancer in women and the second most common cancer in men in Germany [1]. It has the highest mortality among the solid tumors. Although there has been progress in the therapy of lung cancer in the last decade, the overall prognosis across all stages has not been substantially improved. Only 15% of patients are alive after 5 years. In contrast, if one looks at other types of cancer such as prostate cancer, colorectal cancer, or breast cancer, substantial improvements in long-term survival have been achieved in the last few years. A large proportion of this positive trend is due to established screening programs for these tumor types. For colorectal cancer, 5-year survival increased from 50% in the 1970s to 62% in the 1990s. For breast cancer, this proportion increased from 75 to 86%. Currently, 97% of patients with prostate cancer are alive after 5 years [2].There are a number of reasons that make lung cancer screening especially attractive. For one, lung cancers have the highest mortality among the solid tumors worldwide. Also, the prognosis of lung cancer is strictly stage-associated. Treatment is only curative in the early stages. However, only 15-20% of patients are diagnosed in stage I, as the tumors are frequently asymptomatic.From the 1950s until the 1970s, a number of studies on lung cancer screening were conducted. Usually, these were studies that used chest X-rays for imaging alongside sputum cytology. During this time, a total of four non-randomized non-controlled studies, two non-randomized but controlled trials, and four randomized controlled trials were published on lung cancer screening [3,4,5,6,7,8,9,10,11,12]. These studies showed an improved stage distribution in favor of earlier stages, better resectability of the tumors, and also improved survival. However, an effect on overall mortality could not be demonstrated. Recently, data were published from the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial (PLCO) [13,14]. In this large screening trial (n = 154,942), the active-arm participants (n = 77,477), who were aged 55-74 years and had no further specific risk factors, had chest X-rays performed which were compared at baseline and annually repeated for 3 years. Of the 67,038 people screened with chest X-ray, 8.9% had positive findings in their baseline examination, and of these, a diagnosis of lung cancer could be made in 2.1% (n = 126). However, there was no difference in the incidence of lung cancer found between the control group and the screened group. In a mortality analysis, there was no difference in mortality between the screening group and the control group. Thus, the results are comparable to the chest X-ray studies mentioned earlier. When evaluating lung cancer screening studies, a number of methodological problems should be considered. Among these are statistical issues, with a special focus on disease-specific and/or all-cause mortality. Another important factor is the stage shift. As screening yields more frequent diagnoses of early-stage lung cancer, this should result in a reduction in later stages. Last but not least, bias should be considered [15]: i) Lead time bias: screening only leads to earlier diagnosis, but early treatment may not result in increased overall survival; ii) Length time bias: rapidly growing tumors become symptomatic quickly and thus evade the screening, therefore, screening results in the selection of slowly growing tumors; iii) Overdiagnosis bias: an extreme form of length time bias in which very slowly growing tumors are selected that do not affect the prognosis.Technical advances in imaging techniques for the lung, from chest X-rays to cross-sectional imaging using computed tomography (CT), have led to renewed interest in lung cancer screening. The technique of low-dose CT (LDCT) is based on the fact that even with examination protocols that use less radiation, soft tissue-dense pulmonary nodules of a few millimeters can be detected. Compared to conventional CT from the 1990s, radiation exposure with LDCT is less by a factor of 10. Compared to chest X-ray, radiation exposure with LDCT is 2- to 3-fold higher. In the 1990s, 12 different single-arm non-controlled studies were initiated worldwide (table 1). These feasibility studies were supposed to demonstrate the higher sensitivity of LDCT as compared to conventional chest X-ray [16]. The first two Japanese studies showed a 2- to 3-fold higher rate of lung cancer diagnosis with LDCT as compared to conventional chest X-ray. The American Early Lung Cancer Action Program (ELCAP) enrolled a total of 1,000 participants who were older than 60, had a smoking history of more than 10 pack-years, were functionally operable, and had no cancer history [17,18]. The screening consisted of an annual CT scan of the thorax and a conventional chest X-ray. This study also showed that LDCT detected 3 times as many non-calcified pulmonary nodules, 4 times as many malignant lesions, and 6 times as many cases of stage I disease. The ELCAP was then extended to the International-ELCAP (I-ELCAP). The I-ELCAP enrolled 31,567 asymptomatic participants for the baseline screening and 27,456 for the follow-up screening. A total of 484 participants were diagnosed with lung cancer. Of these, 412 (85%) had stage I disease, and the 10-year survival calculated for this group was 88% [19]. The other single arm studies were in keeping with these findings. One problem associated with these studies was the high proportion of detected pulmonary nodules and the subsequent surgical treatment of benign nodules. In order to demonstrate the efficacy of lung cancer screening by LDCT, ultimately randomized controlled studies with the endpoints of disease-specific and overall mortality were required. The single-arm non-randomized non-controlled studies provided the rationale for such an investigation.On the basis of the feasibility studies, randomized controlled studies were initiated worldwide. The largest trial was the National Lung Screening Trial (NLST). The endpoint was the reduction in lung cancer-specific mortality by at least 20% [20]. Patients enrolled had the following profile: i) age 55-74 years; ii) cigarette smoking history of at least 30 pack-years; iii) former smokers who quit within the past 15 years; iv) asymptomatic and functionally operable; v) no prior malignant disease; and vi) no CT scan within 18 months prior to enrollment in the study. Enrolled patients were randomized to either an LDCT arm with annual screening for 3 years or to a chest X-ray arm also with annual screening for 3 years. Enrollment occurred over 20 months, from August 2002 to April 2004. A total of 53,454 participants were enrolled, and the median follow-up was 6.5 years. Screening with LDCT reduced lung cancer-specific mortality by 20%. A subgroup of the PLCO trial was cross-analyzed with a matched group of the NLST. There was no evidence that screening with chest X-ray alone was significantly different from the control group in the PLCO trial with respect to lung cancer-specific mortality. Furthermore, screening with LDCT was demonstrated to reduce overall mortality by 6.7%. In patients with non-small cell lung cancer, LDCT screening led to a stage shift to earlier stages. This could not be demonstrated for small cell lung cancer. In the screening arm, 129 more patients were diagnosed with lung cancer as compared to the control arm (1,060 vs. 941). The sensitivity and specificity of LDCT screening were 94 and 73%, respectively [20]. For comparison, the sensitivity in the chest X-ray arm was 60-70% across all 3 screening rounds. In the NLST, non-calcified nodules ≥4 mm or other radiological signs of lung cancer were considered positive. The rate of positive findings was high at 24.2% for LDCT as compared to 6.9% for chest X-ray. Guidelines were developed for the management of positive findings; however, these were not obligatory. Most findings could be worked up with non-invasive measures. The number of invasive procedures per diagnosed lung cancer was relatively low. Complications were rare at 1.4% in the LDCT arm and 1.6% in the chest X-ray arm. Worse complications occurred primarily in patients in whom lung cancer was diagnosed. The rate of complications in patients who underwent an invasive workup and had lung cancer was 11.2%. In comparison, serious complications in patients who underwent an invasive workup but did not have lung cancer was 0.06%.Outside the United States (US), three European randomized controlled screening studies have so far yielded results based on a screening and a control group [21,22,23,24]. These trials were generally smaller by a factor of 10 as compared to the NLST. The risk profiles and ages of the enrolled participants also differed. Hence, a direct comparison with the NLST does not make sense. In the Detection and Screening of Early Lung Cancer by Novel Imaging Technology and Molecular Assays Study (DANTE Study), an annual screening with LDCT was compared with observation. Patients aged 60-74 years and with a smoking history of more than 20 pack-years were enrolled. A total of 1,276 participants were randomized. In this study, screening was also shown to be advantageous in terms of lung cancer-specific mortality and overall mortality [21,22]. The Danish Lung Cancer Screening Trial (DLCST) was a single center study that also compared LDCT screening with observation. Persons 50-70 years of age with more than a 20 pack-year history of smoking were enrolled, with a total of 2,052 participants randomized. In this trial, no benefit was found in the screening arm with respect to disease-specific and overall mortality [23]. Similarly, the Multicentric Italian Lung Detection (MILD) Trial was a single-center trial that enrolled younger subjects, aged 49 years and older, who had more than a 20 pack-year history of smoking. The subjects were randomized to the LDCT screening arm or to an observation arm. This study also failed to show a difference in mortality (table 2) [24].In summary, the NLST currently stands as the largest worldwide randomized controlled lung cancer screening study. More than 50,000 people with a defined risk profile were enrolled in this trial. LDCT screening compared to screening by chest X-ray lowered lung cancer-specific mortality by 20% and overall mortality by 6.7%. The number needed to screen (NNS) to prevent 1 lung cancer death was 320; the NNS to prevent a death overall was 219 over 6.5 years. One can compare these results with the NNS in other cancers. The NNS to prevent breast cancer death in mammography studies is 1,329 after 11-20 years of follow-up. To prevent a colorectal cancer death, the NNS by colonoscopy is 817 Thus, there are and data that make screening by LDCT in a defined risk group a cancer is the of all cancer worldwide. More than men and women in be diagnosed with lung cancer this more than of the the fact that lung cancer is by far the number one cancer, there has been a screening for at least not the results of the NLST were published Screening for lung cancer in the NLST showed a 20% reduction in and participants were needed to screen to prevent 1 lung cancer These findings were with but a screening is and associated with screening for lung cancer also to be considered of an screening for lung cancer has been an for many years. Early randomized screening trials that of chest X-ray and sputum did not show a mortality benefit from screening CT screening for lung cancer has been in CT which of of the lung within one using only a radiation Early CT screening studies initiated by the ELCAP in the and in Germany by a group led by of the of These studies provided data on and risk of lung cancer However, due to the single-arm of the studies was to the of LDCT screening on lung cancer mortality in an the the NLST enrolled 53,454 participants aged 55-74 years who were or former smokers of a per for 30 at were randomized to either LDCT (n = or chest X-ray (n = annually for 3 screening [20]. LDCT screening were considered positive and to lung cancer if at least 1 non-calcified 4 mm in largest other for lung and chest X-ray were positive if non-calcified or 24.2% of CT and 6.9% of chest X-ray were positive. who all 3 had a rate of screen and of these were or did not show lung cancer Lung cancer-specific mortality were per and per in the LDCT and chest X-ray This in a 20% reduction in lung cancer mortality in the LDCT arm and an risk reduction for lung cancer death by 4 per 1,000 The rate of death from was reduced in the LDCT arm as compared with the arm by = To and patients the results from the NLST in an the National Cancer a and for lung cancer screening, which was in with the NLST (table The and risk for for were used are less than to small randomized controlled trials of low-dose CT screening for lung cancer are has been in all trials, for the Lung Screening Trial European trials have substantially than NLST, and only one the lung cancer screening which is the largest with is at to show a reduction in lung cancer mortality of at least at 10 years after The has enrolled participants the trial has participants the trial has enrolled participants and the study has randomized participants to LDCT no screening, respectively the study has randomized participants to to and to annual LDCT The trial has the only European trials have in common that LDCT screening is compared with a arm, and for all trials only and former smokers have been reduction in with CT screening has been shown in the and studies more lung cancers detected in the LDCT and more early-stage there were of cancers compared to and there was no mortality reduction with LDCT screening. However, these trials were to a mortality reduction of 20% as with the NLST data from the and trials become in of the in mortality data from this trial can be in evidence for the and of LDCT screening for lung one by on of a the American Cancer the American of the American of and the National Cancer and the other by to the randomized controlled trials have the of LDCT screening on lung cancer-specific mortality among with smoking of which the NLST was considered the most This large trial that screening with LDCT was associated with in lung cancer-specific and all-cause mortality. small European trials showed no benefit of screening studies have the of LDCT screening affect smoking It has been that LDCT screening may result in of or may an for smoking the smoking of have not found evidence that or are by in screening malignant and benign non-calcified nodules, the Although most LDCT screening studies have on the nodules the and of are studies, the rate per of screening was but from 3 to in randomized controlled trials and from 5 to in Most studies that of nodules detected were benign Most a detected further but the management protocols were in the The of further CT imaging among screened from 1 to The of further tomography imaging among screened less from to The of invasive of detected nodules was generally but In the NLST, of patients who were not found to have lung cancer underwent an invasive such as or of patients who were not found to have lung cancer had a or In the study, these were and respectively is no for findings on screening LDCT, and sensitivity is by of lung cancer within 1 of a screening study as In the 6 studies this sensitivity of LDCT for lung cancer from to a rate of only study on complications from LDCT screening is the NLST. the of death within months of a of a detected was per screened by LDCT, and 5 per screened by chest X-ray. of the after a were to follow-up as and per occurred within months the only an imaging study. the of a a of a detected was per screened by LDCT, and 10 per screened by chest X-ray. The of complications occurred after surgical and in patients with lung cancer. The rate of complications in patients with lung cancer who underwent was only on patients who had nodules detected by LDCT that to be death occurred within among and complications occurred among to lung cancers screening that not the if During lung cancers were in the LDCT arm and in the chest X-ray arm of the NLST. The is that a lung cancer detected by screening with LDCT was an that a non-small cell lung cancer detected by LDCT was an and that a lung cancer detected by LDCT was an The number of cases of found among the participants needed to be screened in the NLST to prevent 1 death from lung cancer was of radiation with LDCT is to be per however, chest CT or to further detected rapidly the exposure and for most of the radiation exposure in screening It is that NLST participants per over the 3 screening and over the screened of from such radiation from and studies, based on from and also many studies of the NLST these 1 cancer death by radiation from imaging per subjects the benefit in lung cancer in NLST is than the radiation risk studies among LDCT screening. In the and trials, screening did not affect overall of or long-term In the the studies increased or compared with baseline among participants with positive or results and of cancer compared with baseline among with results NLST has the of lung cancer screening, evidence that LDCT screening can reduce lung cancer-specific and all-cause mortality In of these the in Lung Cancer has of a for lung cancer screening in to who have a history of 30 pack-years of smoking or of up to 15 years as an which that in the be to the of CT screening in this over the all lung cancer screening with LDCT The National Cancer Lung Cancer Screening provided the first after the evidence and has from the American Lung the American of the American for and the American Cancer and the of These should be screening in the only for which there is a demonstrated benefit is that defined by the smokers with at least 30 pack-years of who are aged to and who quit less than 15 years screening to other than in the NLST is not at this time is in the of a study The of LDCT screening is primarily by the risk of lung cancer of death A of the NLST that participants to their risk for lung cancer showed that the number of lung cancer increased with an in the risk of death from lung cancer. the number of participants who to be screened to prevent 1 death from among the 20% of participants at risk to among the 20% at highest Also, the number of participants with results on screening per lung cancer death from among the 20% of participants at risk to among the 20% at highest risk the use of an that risk to for screening is to be and reduce to people with the least risk of lung cancer risk has been developed in The PLCO cancer screening trial lung cancer risk was developed from the largest to A of this has been to the NLST and selected people for screening who a diagnosis of lung cancer in which have in 12 In one should be lung cancer screening results from the NLST to other with different and When the NLST selection to an study in the found that only of lung cancer cases that occurred in their the NLST of cases not have been in a screening trial based on NLST A reduction in mortality from lung cancer of as shown in the NLST, with a reduction in mortality in the study associated with screening be with the have been from lung cancer screening, in terms of complications from procedures to to be nodules In the NLST, 4 in 1,000 from lung cancer. However, in the LDCT arm, in 1,000 had a to an invasive such as or and 3 in 1,000 had a from such invasive for can the number of further imaging studies or invasive to is In the and trials, are used to nodules, than all nodules of a specific as as in the NLST from these trials important that could to better screening programs is of specific in lung cancer screening. The data from a of the NLST that of in the LDCT arm with lung cancer and of in the LDCT arm with non-small cell lung cancer may be cases of In other if these had not the NLST, not have a lung cancer diagnosis or at least not for the 5 years This is most in patients with a diagnosis of invasive an and long-term These data the as to the and the of therapy if a diagnosis of invasive is and the to a that could all lung tumors In a of all participants were not to a screening that results in more than 1 per 1 from death due to cancer The number of cases of found among the participants who to be screened in the NLST to prevent 1 death from lung cancer was shown with the NLST, LDCT screening in a with controlled selection can to a in the rate of death from lung cancer. To which these results can be achieved if screening is on a is from surgical especially for benign nodules, are a In the NLST, the mortality from such was which to that from For comparison, in mortality of pulmonary is in the of and in this in the have of an lung cancer screening One frequent for the and of to the screening are the of and the trial and the of the European randomized controlled CT screening trials further mortality data on CT screening and also within the European These trials also a risk of the and a radiological which reduce the number of and also with the management nodules from these trials become in to European randomized lung cancer CT screening trial published a with to the of CT screening and the of CT screening trials in The that CT screening is not for in at this time and that should the of the screening trials in order to the following i) of for lung cancer CT screening ii) of radiological for use in screening iii) of for the workup of from CT screening iv) Guidelines for of nodules from lung cancer CT screening v) for surgical and of nodules lung cancer CT screening and vi) of smoking lung cancer CT screening These to be an lung cancer screening can be established within this of the are to a and lung cancer screening in and to the evidence on this important and their to the and their of that European be can that that following the NLST results time is to a lung cancer screening by annual LDCT in a defined risk [20]. is more in of the and the other to for the results of the randomized controlled trial in and the of the other European randomized controlled CT screening trials for prior to CT screening in in order to prior to the with a screening the following this there not by there a from with for the management of lung cancer screening in is a third to the The first and of the NLST results the 20% reduction in lung cancer mortality and reduction in overall mortality as far as at the in first the results in Germany until the European have been and and for this not the of of to the of such important findings for the number of lung cancer patients from this and even the early and NLST may the benefit of a further if reduced to LDCT screening in this the 3 performed in NLST and the on the of such an Although and thus far the risk of further screening may relatively than the benefit from of further diagnosed lung cancers in other participants in the years to This alone may the of a and screening in of the two the that is and be in Germany based on the data published up to this even by the smokers performed based on soft of in history and as by by their an to may even be than no at all but is This be the for a and lung cancer screening as as in with a for LDCT as as a for findings for further imaging and the of do have in Germany a rapidly growing number of Cancer by Cancer as as Lung Cancer by the Cancer who are more and more control and for diagnosis and therapy radiation with tumor and follow-up. As this is not the in many other European the be for such a In no other the be less than in such an this lung cancer screening could be as an with or following findings from European randomized could only be if the and their worked in the of this the the the for the for and the of and the of with the of the Cancer such as smoking programs and but also could be to this important if lung cancer screening was to be for as as be for the if was performed in the of such a than on an basis in a of and and is and a should be than so time which and could be achieved and the of this technique in order to this with so many cancer in the The did not a of interest of The did not a of interest