Tumor > Malignant > Lungcancer

Screening population and at what interval?:

In general, it is desirable to screen a patient population that would be expected to have a high prevalence of disease- ie: patients that are at increased risk for developing lung cancer. The average prevalence for lung cancer in most screening studies is between 0.3-2.3% [74]. Cigarette smoking is the overwhelming dominant risk factor for developing lung cancer [54]. Cigarette smoking causes more than 80% of lung cancers [77]. The relative risk for developing lung cancer in a male smokers is about 10 times that for non-smokers and the incidence of lung cancer directly correlates with the pack-years of cigarettes smoked (heavy smokers risk is 15-35 times greater) [45,54]. Generally, screening studies have focused on patients with smoking histories of 10-20 pack years or greater and for patients that were at least 50 years of age. COPD increases the risk for lung cancer by a factor of 2-3 [28]. Another risk factor is occupational exposure to carcinogens (such as asbestos, uranium, radon, arsenic, chromium, nickel, or mustard gases) [54]. A history of lung cancer in a first degree relative is also a risk factor (and may indicate a specific lung cancer susceptibility gene) [54].

The National Comprehensive Cancer Network recommends annual CT screening for two groups of high risk individuals [99,104]:

Group 1:
- Age 55 to 74-80 years
- 30 pack year or more smoking history
- Current of former smokers who quit within the past 15 years

Group 2:
- Over 50 years of age
- Over 20 pack year smoking history
- Current or former smokers who quit for any length of time
- At least one additional risk factor (excluding second hand smoke exposure): Personal history of smoking related cancer; family history of lung cancer in a first degree relative; chronic lung disease including emphysema and pulmonary fibrosis; or known exposure to pulmonary carcinogens

An appropriate screening interval has not yet been established. In a clinical setting at this time, yearly screening is probably reasonable, however, it should be understood that some cancers will become evident between the screening studies [7]. One suggestion is to perform a minimum of two annual screening exams followed by repeat screening every 2-3 years thereafter [56].

Screening Protocol and Radiation Dose:

Radiation dose is directly proportional to tube current (mA) at a constant peak tube voltage (kVp), slice width, and pitch [26]. Helical CT screening employs the use of a low dose exam (click to view protocols) that adjusts mA, kVp, and pitch. The potential benefit of low dose CT for the detection of early lung cancer must be weighed against any possible risk of inducing malignancy by using ionizing radiation [21]. All screening exams utilize a decreased kVp and mA in order to decrease the radiation dose to the patient. The radiation exposure to the lungs is approximately 3 to 10 times higher than that of a two view plain film CXR (single view 3.2 mrem [25]), but is only one-sixth that of a conventional CT [10,33]. The dose to the breasts is equivalent to that of one mammographic film of each breast [33]. The average effective dose has been reported to be 0.6 mSv (60 mrem) in men and 1.1 mSv (110 mrem) in women [21]. These doses do not include any high resolution or follow-up CT studies. The dose for follow-up thin slice exams can be decreased by using lose dose parameters and limiting the exam to the area of the nodule [21]. To put the dose in perspective, natural background radiation exposes the average American to 360 mrem per year (470 mrem if living in Denver) and the radiation dose from a lumbar spine series (5 views) is approximately 200 mrem [25]. One must also keep in mind that human cells can repair low dose radiation cell damage rapidly [25]. Another way to look at radiation risk is in terms of things associated with a one in a million chance of dying: Receiving 10 mrem of radiation; smoking 1.4 cigarettes; eating 40 tablespoons (600 mL) of peanut butter; spending 2 days in New York City; driving 40 miles in a car; and flying 2500 miles in a jet [25].

The Committee on the Biological Effects of Ionizing Radiation (BEIR V) estimated that the risk of cancer death is 0.04% per rem of effective body dose [25]. BEIR VII estimated the overall chance of a radiation induced solid tumor fatality from a 10 mGy (1 rad) CT exam to be about 0.00041- which is a very small risk [64]. This data was based upon information from therapeutic high-dose radiation and from survivors of the atomic bombs in Japan [25]. In determining radiation risks from screening CT exams one assumes that there is no threshold needed for cancer induction (stochastic effect) and that high-dose data can be applied to low dose studies [25]. Estimates by the International Commission on Radiological Protection predict that examination of 100,000 individuals with one low dose CT exam will induce three (men) to six (women) additional cancers within the next 15 to 20 years [21]. The likelihood for inducing malignancies in theory increases with every additional thin-section follow-up CT or annual repeat exam [21]. A 50 year old female smoker who underwent annual CT lung screening until age 75 would incur an estimated radiation related lung cancer risk of 0.85% in addition to her otherwise expected lung cancer risk of approximately 17% (men would incur a 0.23% increased risk) [37]. It has been estimated that a mortality benefit of greater than 5% may be necessary to outweigh the potential radiation risks associated with screening [37]. This would seem very possible given that by using screening CT, the likelihood for detecting early stage lung cancer is also increased [6,7]. In the end, the risk for inducing malignancy must be balanced against the detection rate of potentially curable cancer (between 600 to 800 cases of Stage I cancer per 100,000 screened individuals assuming a prevalence of only 1%) [6,7,21]. Identification of the proper patient cohort for screening is essential for minimizing radiation risks while maximizing the gain from early lung cancer detection. Some ways to reduce overall radiation risk would be to begin screening at age 60 years, or screen every two years, rather than annually [37]. Also- follow-up studies can be performed with adjustment to the scanning parameters based upon the patients weight in order to decrease patient radiation exposure (i.e.: 70 mAs for a 70 kg patient) [60].

The use of breast shields in screening CT has been evaluated and although there is an associated dose reduction, the authors note that because a breast shield increases noise in the anterior lung, faint and small nodules with ground-glass opacity may be missed [94].

Radiologic guidelines:

Nodules identified on screening CT should be defined based upon their characteristics of size, shape, margins, location, and if benign calcifications are present. Lung-RADS, from the ACR, is an established set of guidelines for reporting and monitoring nodules detected on lung screen CT [104]. Unfortunately, there is variability in application of these guidelines with disagreement in some specific scenarios [104]. It has also been reported that subsolid nodules that are classified as category 2 or 3 have a higher risk of malignancy than that which should ordinarily be associated with those categories (3% vs 1% for category 2, and 13% vs 1-2% for category 3) [106]. Ground glass nodules measuring 10-19 mm have a higher rate of malignancy compared to GGNs smaller than 10 mm (6% vs 1.3%), despite both being classified as category 2 [106]. Lung-RADS version 1.1 defines benign perifissural nodules (PFNs) as a category 2 classification with a less than 1% probability of malignancy- benign features as a solid PFN include smooth margins; an oval, lentiform, or triangular shape; and a maximum diameter less than 10 mm (or a volume less than 524 mm²) [107]. Additionally, some specific scenarios are not addressed by Lung-RADS including isolated hilar or mediastinal adenopathy or pleural effusion in the absence of a lung nodule [104].

In one study, only 2.7% of patients had nodules that were recommended for biopsy [91]. Of those patients that underwent biopsy, 84% had malignant nodules (16% false positive- but this represented only 0.42% of the total study population) [91]. In that study, all spiculated nodules were malignant [91].

1- Clearly benign: Patients with negative scans or obviously benign lesions (calcified granulomas) can be followed with yearly screening.

2- Clearly malignant: Patients with nodules that have CT characteristics suggestive of malignancy (spiculated or lobulated margins (lobulated margins indicates asymmetric growth), air-brochogram sign), should have the lesion resected. A poorly defined nodule border confers a 6.6 fold increased risk of lung cancer [103].

3- Indeterminate: An indeterminate nodule is solid, smooth-edged, and does not show characteristically benign calcifications. Small ground-glass nodules and mixed nodules (with solid and ground-glass components) can also be considered in this category. The risk of cancer in a non-solid nodule increases with lesion size, particularly the solid component [45]. According to ELCAP data, 18% of nodules with pure ground glass opacity were malignant, and 63% of nodules with mixed ground-glass opacity were malignant [44]. Subsequently, data from screening studies revealed part-solid nodules on 5% of baseline screening exams (as opposed to solid nodules which are found on 30% of baseline screening exams)- about 2.7% of these proved to be malignant and 20% resolved or decreased in size at followup [101]. A new part solid nodule can be identified at 0.8% of annual repeat screening and about 1.5% prove to be malignant, whereas almost 70% resolve or decrease in size [101]. Patients with indeterminate nodules will require serial CT evaluation for stability. The follow-up interval for indeterminate nodules is often dictated by the individual subject and their physician. Sites experienced in lung cancer screening have adopted the following strategy based on the diameter of the nodule.

A. Nodule less than 5 mm: The likelihood for malignancy in nodules less than 5 mm in size is low [45,51]. Available data indicate that less than 1% of nodules under 5 mm are malignant [54]. Nodules less than 3 mm in diameter have only a 0.2% chance of being malignant, while the risk of malignancy is 0.9% for nodules 4-7 mm in diameter [78].

B. Nodules or mixed nodules between 5 to10 mm: Thin slice helical CT at 3, 6, 12 and 24 months. Short term follow up scans at 3 and 6 months may be confined to the lesion only to decrease patient radiation exposure. Partially solid nodules have the highest probability of malignancy (40-50% in some studies) [78]. Biopsy/removal of nodules that increase in size is recommended (25-30% malignant). Nodules that do not grow in volume in 6 months have a smaller risk of malignancy (less than 10%) [45].

C. Ground glass nodule: Non-solid nodules can be found in up to 4% of baseline screening exams and develop on approximately 0.7% of followup screening studies [96]. Pure ground glass nodules have a higher likelihood for representing a malignancy (about 34%) [78]. Partially solid nodules are even more likely to be malignant [97]. In one study, 63% of semi-solid nodules (with a ground-glass component) were found to be malignant [98]. Pure ground glass nodules tend to grow slowly with volume doubling times of approximately 800 days [78]. Some authors recommend followup of pure ground glass nodules (even new GG nodules that develop on followup screening exams) with one year intervals [96]. However, the recommendations vary [96]. Some authors have suggested that the term indolent lesion of epithelial origin be used to describe GG nodules, rather than the term lung cancer [96]. Focal ground glass nodules can also represent inflammation, atypical adenomatous hyperplasia, and focal interstitial fibrosis [84].

D. Nodules larger than10 mm: Consider biopsy of all of these nodules (30-80% malignant)- the risk of malignancy increases to 18% for nodules 8-20mm, and 50% for those larger than 20 mm [78]. If of adequate size, these nodules may be studied with PET scanning or with CT contrast enhancement. Unfortunately, because bronchoalveolar cell carcinoma is often PET-negative, PET scans should be discouraged if the nodule is predominantly ground-glass [49].

For patchy or non-solid nodules, a short term 6-8 week follow-up CT may sometimes be performed following a course of antibiotics to assess for nodule resolution as a large percentage of these nodules will resolve [45,76].

4- Cancer associated with cystic airspaces: The cysts associated with lung cancer demonstrate wall thickening, ground glass attenuation, and mural nodularity which may slowly evolve over time which necessitates long term followup [108]. Cancer associated with cystic spaces is most commonly adenocarcinoma (88% of cases) and it can be seen in both smokers and non-smokers (up to one-tthird of patients) [108].

Limitations of Screening CT:

A. Incidental nodules: The identification of small non-calcified nodules is becoming so common that the suspicion raised by such a finding is becoming negligible [36]. One or more non-calcified nodules can be detected in 51% to 92% of screened patients [27,34,78], but only approximately 1% of these nodules prove to be malignant (i.e.: 99% of nodules identified are false-positive findings) [18,27]. This high false positive rate leads to increased patient anxiety, and additional medical costs related to CT scans, biopsy, and surgical intervention [18]. Incidental findings which require further evaluation or medical action can be detected in up to 14% of patients- again, increasing costs. Investigators have reported that up to 12% of people who are screened for lung cancer with CT undergo invasive biopsies that ultimately reveal one or more benign processes [66].

B. Missed cancer: Lung cancers can be missed on screening CT exams due to detection or interpretation errors [29]. Between 32% to 56% of cancers can be missed on initial screening exams [7,12,29,95]. Up to 23% of cancers are identified, but ot classified as malignant [95]. These lesions are often very small (less than 1 cm), not well defined (ground glass attenuation), or are overlapped with, obscured by, or similar in appearance to normal structures [29,55]. Cancers are also more commonly missed due to interpretation errors in patients with underlying lung disease (with scarring or fibrosis) [29,55,71]. Generally, when detected on follow-up exams, the lesions are still Stage I, however, up to 17-18% can be beyond stage I [95]. Another limitation of screening is that unaided intra- and interobserver agreement in detecting pulmonary nodules on low dose CT of the lungs is relatively low [52,72] and there is wide variation as to what is classified as a pulmonary nodule [72]. Thinner collimation for screening exams (5 to 8mm) and automatic computer aided nodule detection programs can improve identification of nodules that would otherwise go undetected [29,30,31,55,59]. These programs work best with thin slices images and a small reconstruction interval [50]. The one drawback of automated detection systems are that some nodules/cancers will still not be detected (up to 27% [59]) and there are a large number of false-positive findings (from 3 to 13 per CT study) [59]. Most missed nodules on computer aid detection are small (under 4 mm), have low attenuation, and are not completely surrounded by lung parenchyma [59]. Double reading has been shown to result in the detection of 2.7% more cancers and 19% more nodules, but this was not statistically significant [90]. Computer aided detection (CAD) has been shown to be useful for the detection of small nodules, but the drawback is a very large number of false positives [100]. In one study, CAD systems detected up to 56-70% of small cancers that were not detected by the radiologist [100]. However, in this same study, CAD failed to detect 20% of cancers that were identified by the radiologist [100]. There is also wide performance variability between different CAD systems [100].

C. Measuring lesion growth: Another limitation of the exam is the difficulty in accurately measuring the growth of small nodules. Two-dimensional CT measurements have been shown to be unreliable in assessing for small changes in lesion size [39,41]. This is because a doubling in volume amounts to only a 26% increase in nodule diameter and a significant change in size may be difficult to appreciate for a small nodule [57]. For example, a 5 mm nodule can double in volume, but its diameter will increase by only 1.25 mm (well within the error of a transaxial measurement [39]). Volumetric nodule reconstructions provide better detection of lesion growth (particularly in the z-axis) and remove the guess work of image registration between separate examinations [19,23,35,41]. Generally, volumetric measurements show good interobserver measurements [58]. However, even automated volumetric measurements can vary considerably- particularly for small nodules (under 5 mm in size), those of ground-glass attenuation [38,58,62,79], and those with spiculated margins [65]. In particular,  cardiac pulsations, patient motion, and variations in breath hold can affect the accuracy of volumetric measurements [38,43,79]. Nodule location can also affect the volume measurement- juxtapleural and juxtavascular nodules have been shown to exhibit greater volume measurement variability compared to completely intra-parenchymal nodules [65]. Image compression of greater than 20:1 can also affect the accuracy of volume measurements- particularly for ground-glass nodules [53]. Section thickness and varying the window center can also affect the volume measurement [46,48,65]. Nodules located near the heart show as much as 34% volume change during the cardiac cycle [79]. ECG-gated CT imaging may aid in determination of nodule volume by decreasing pulsation artifacts [43]. There is also significant variability in lesion size and volume observed on serial scanning of the same nodule [58,79]. For this reason, even when applying volumetric analysis, a threshold change in volume must be established for true nodule growth [40,79].

D. Stability over time: Although many people quote that lesion stability over a two year period indicates that the nodule is benign, this statement was taken out of content from the original article in which it appeared. Caution must be used when applying this criteria to pulmonary nodules [13]. In fact, no growth over a two year period has a positive predictive value for benignity of only 65% (sensitivity is 40%, and specificity is 72%) [13]. All two year stability implies is that the lesion has a doubling time of at least 730 days (most malignant lesions have doubling times between 20 and 400 days) [13]. Bronchoalveolar cell carcinoma is particularly problematic as the lesion may appear stable for a period of 2 to 7 years before the diagnosis of BAC is made [14]. In one lung cancer screening study, 13 of 48 cancers (27%) had volume doubling times of longer than 400 days [69]. These data support extended follow-up for small nodules- even non-solid nodules [54].

E. Cancers that develop in the interval between screening exams: The benefit of screening for lung cancer is to detect stage I lesions. Because lung cancer can have a wide variation in doubling times (from 32 days to over 1000 days in one study [23]), new cancers can arise between screening exams.

In the International Early Lung Cancer Action Program, new nodules were detected in about 5% of patients and of these, 5% were found to be malignant [102]. In the national lung screening trial (NLST), new nodules developed in about 3% of patients [102]. Of these nodules, about 6% were malignant and the risk for malignancy increased with nodule size (from 1.1% for nodules < 4 mm, 13% for nodules ≥ 8 mm, and 24% for nodules ≥ 20 mm) [102]. Because survival is related to stage, it is also important to know what percentage of interval cancers are stage I. Some authors note that up to 23% of cancers are not visible on a previous screening exam and of these, up to 29% can be beyond stage I when detected [95]. Other authors indicate that during follow-up screening, pooled US data indicates that about 60% of cancers detected after the initial prevalence screening exam are stage I, meaning up to 40% are higher stage [74]. Additionally, in the NLST, cancers associated with new nodules had a significantly poorer survival compared to cancers detected at baseline screening likely due to more rapid growth and aggressive characteristics [102,103]. In one study, based upon doubling times- annual screening would result in detection of 94% of T1 tumors, while biennial screening would have detected only 78% of tumors at this size [23]. This suggests that caution and further research are needed prior to changing the screening interval to more than 12 months [95]. 

In a retrospective study, it has also been shown that high-risk patients have a very high incidence of lung cancer developing even after a series of three prior negative lung screen CT exams (median of 7 years since completing prior screening and a lung cancer prevalence of up to almost 21%) ][105]. In that study, when the previous screening CT exams were reviewed, 43% of the malignancies had nodules (5-11mm) at the tumor site that had remained stable during a minimum 2 year interval [105].

F. Screening Bias- survival versus mortality: Screening trials suffer from several forms of bias which can affect survival data, but not generally impact on patient mortality. These include over-diagnosis bias, length bias, and lead-time bias [68]. In over diagnosis bias cancers are detected that might never have become symptomatic during a patient's lifetime because of competing causes of mortality- in other words coexisting heart disease might have resulted in the patient's death prior to the development of symptomatic lung cancer [68]. In length bias (related to tumor aggressiveness), expected survival appears increased by detecting more patients with less aggressive disease (who have longer survival) and fewer patients with more aggressive disease- hence, there appears to be a survival benefit from the screening test [68,74]. Slowly progressing lesions are disproportionately detected at screening due to a longer window for detection [80]. A lead-time bias occurs when patients are accorded extended survival times solely because cancer was detected earlier due to screening, though death occurred at the same time as would have happened without screening (i.e.: the intervention yields no benefit) [68]. For example, if a patient is diagnosed with lung cancer in 2010 and dies from lung cancer in 2013 their survival is three years. If that same patient had undergone screening in 2007 and the cancer had been detected and removed, but they still died from lung cancer in 2013, their survival would be increased to 6 years, but their mortality did not change.

Although detecting lung cancer at an earlier stage will obviously result in improve patient survival, studies have not yet demonstrated a decrease in mortality with lung cancer screening [66,68]. The study by Bach et al [66] revealed that screened patients were diagnosed with lung cancer in far greater numbers than would have occurred in the absence of screening and the majority (67%) were stage I or stage II. However, there was no decrease in overall mortality based upon predicted models [66]. Granted, the study has limitations as the mortality estimates depend on the validity of prior risk factor analyses [66]. Contradicting the Bach model, the Mayo clinic predicted that screening may reduce lung cancer specific mortality (8% reduction in lifetime lung cancer mortality after 5 screening exams) [77]. To date, no study has included a "no screening" comparison group which would permit determination of the true effect of screening on mortality without screening biases [68,77]. Additionally, because of the small number of patients in Bach study, the 95% confidence interval for their data might allow for a lung cancer mortality reduction as large as 30% [66].

G. Surgical complications: Surgical intervention even in early stage lung cancer carries definite risks. In the I-ELCAP study mortality following lung cancer resection was only 0.5% [67]. However, the average mortality in the US following lung cancer resection is 5% (1% when done by an experienced thoracic surgeon [74]), and the frequency of serious complications ranges from 20-40% [66].

H. Incidental findings: Other incidental findings that merit further evaluation occur in 14-49% of patients including other neoplasms, adrenal, renal and liver masses, vascular aneurysms, and coronary artery calcifications [68]. Studies have shown that a coronary calcium score derived from the lung screen CT exam can carry prognostic significance regarding all-cause mortality (with a strong association between an increasing calcium score and cardiovascular events and all-cause mortality) [93]. These findings lead to additional health care costs. However, when viewed from a different perspective, the detection of clinically significant additional findings can result in early intervention. Extra-pulmonary malignancies are detected in 0.5 to 1.1% of patients- most commonly renal cancers and lymphoma [86]. The detection of significant coronary artery calcifications on screening CT has been shown to be associated with an increased risk for cardiovascular death and all-cause mortality [83,92].

I. Additional obligations on the radiologist: Patients must be understand that a negative screen does not preclude the subsequent development of lung cancer, even between scans and that some lung cancers may not be detected by CT screening. A mechanism must be in place that ensures the patient is contacted with results of the CT exam. Appropriate physicians need to be available to council and treat the patient with a positive result. Patients need to understand that a large number of small lung nodules are benign and they must be willing to following screening guidelines should a nodule be detected. Finally, patients must be aware that screening is not a "one time" exam, but rather a process that requires them to submit to yearly evaluation.

J. Patient compliance: Patients with negative baseline scans and patients that continue to smoke have been shown to be less compliant with followup examinations (adherence to followup below 50%) [109].

Conclusion:

CT is clearly superior to chest radiographs for the detection of lung cancer- particularly early stage (Stage IA) lesions [19]. Unfortunately, all lung cancer screening studies lack a control group, so they suffer from screening biases. Regardless of these limitations, survival for Stage IA tumors can be as high as 80% at 5 years. Therefore early detection of lung cancers should lead to overall improved lung cancer survival. The exam may be cost effective with suitable specifications for screening (the cost per life-year saved can be as low as $10,000) [7]. This is below the cost for existing screening programs for and cervical carincoma [7]. However, the cost benefit is directly related to the prevalence of the disease in the screened patient population and higher estimates have also been determined [74]. One should also bear in mind that not all other screening examinations which are currently employed have been shown to result in a survival benefit. Although prostate cancer screening is in widespread use, it is not yet validated for providing a clear benefit in terms of reducing mortality from prostate cancer [35]. Unfortunately, if you screen for lung cancer, you will miss lung cancer in up to 56% of cases [7].

Patients at risk for lung cancer should make their own decisions regarding their need for CT lung cancer screening after consultation with a physician [35]. Patients need to understand that screening is not a "one-time exam", but rather a yearly requirement if it is to be implemented properly and that cancers can be missed even on CT imaging. Annual screening for at risk patients has been recommended for 25 years [102].

REFERENCES

1. Chest 1996; Arita T, Matsumoto T, Kuramitsu T, et al. Is it possible to differentiate malignant mediastinal nodes from benign nodes by size? Reevaluation by CT, transesophageal echocardiography, and nodal specimen. 110: 1004-08 

2. J Clin Onc 1998; Eberhardt W, Wilke H, Stamatis G, et al. Preoperative chemotherapy followed by concurrent chemoradiation therapy based on hyperfractionated accelerated radiotherapy and definitive surgery in locally advanced non-small cell lung cancer: Mature results of a phase II trial. 16: 622-34

3. J Thorac Imag 1999; Bepler G. Lung cancer epidemiology and genetics. 14: 228-234 

4. Lancet 1998; Sone S, et al. Mass screening for lung cancer with mobile spiral computed tomography scanner. 351: 1242-1245

5. Br J Cancer 2001; Sone S, et al. Results of three-year mass screening programme for lung cancer using mobile low-dose spiral computed tomography scanner. 84: 25-32

6. Lancet 1999; Henschke CI, et al. Early lung cancer action project: overall design and findings from baseline screening. 354: 99-105

7. Cancer 2001; Henschke CI, et al. Early lung cancer action project. Initial findings on repeat screening. 92: 153-159

8. Radiol Clin North Am 2000; Henschke CI, Yankelevitz DF. CT screening for lung cancer. 38: 487-495

9. Cancer 2000; Okamoto N. Cost-effectiveness of lung cancer screening in Japan. 89: 2489-2493

10. Cancer 2000; Kanedo M, et al. Computed tomography screening for lung carcinoma in Japan. 89(11 Suppl):2485-8

11. Radiology 1996; Kaneko M, et al. Peripheral lung cancer: screening and detection with low-dose spiral CT versus radiography. 201:798-802

12. Radiology 1999; Kakinuma R, et al. Detection failures in spiral CT screening for lung cancer: analysis of CT findings. 212: 61-66

13. AJR 1997; Yankelevitz DF, Henschke CI. Does 2-year stability imply that pulmonary nodules are benign. Feb 168 (2): 325-328 (No abstract available)

14. Chest 1998; Dumont P, et al. Bronchoalveolar carcinoma: histopathologic study of evolution in a series of 105 surgically treated patients. Feb;113(2):391-395

15. J Thorac Cardiovasc Surg 1998; Inoune, et al. Diagnositc assessment of 1310 patients with non-small cell lung cancer who underwent complete resection from 1980-1993; 116: 407-411 

16. Chest 1999; Adebonjo SA, et al. Impact of revised stage classification of lung cancer on survival. A military experience. 115: 1507-1513

17. Chest 1997; Mountain CF. Revisions in the international system for staging lung cancer. 111: 1710-17

18. Society of Thoracic Radiology Course Syllabus 2000; Swenson SJ. Lung cancer screening: Low dose CT. 32-36

19. AJR 2000; Boiselle PM, et al. Lung cancer detection in the 21st century: Potential contributions and challenges of emerging technologies. 175: 1215-1221 (No abstract available)

20. Radiology 2001; Patz EF, et al. CT screening for lung cancer: not ready for routine practice. 221: 587-591

21. AJR 2002; Diederich S, et al. Screening for early lung cancer with low dose spiral CT: prevalence in 817 asymptomatic smokers. 222: 773-781

22. AJR 2002; Henschke CI, et al. CT screening for lung cancer: frequency and significance of part-solid and nonsolid nodules. 178: 1053-1057

23. Radiology 2002; Winer-Muram HT, et al. Volumetric growth rate of stage I lung cancer prior to treatment: serial CT scanning. 223: 798-805

24. Chest 2002; Nawa T, et al. Lung cancer screening using low-dose spiral CT. Results of baseline and 1-year follow-up studies. 122: 15-20

25. AJR 2002; CT screening: why I do it. Brant-Zawadzki M. 179: 319-326

26. AJR 2002; Prasad SR, et al. Standard-dose and 50%-reduced-dose chest CT: comparing the effect on image quality. 179: 461-465

27. AJR 2002; Swensen SJ. CT screening for lung cancer. 179: 833-836

28. Radiology 2002; Garg K, et al. Randomized controlled trial with low-dose spiral CT for lung cancer screening: feasibility study and preliminary results. 225: 506-510

29. Radiology 2002; Li F, et al. Lung cancers missed at low-dose helical CT screening in a general population: comparison of clinical, histopathologic, and imaging findings. 225: 673-683

30. Radiology 2002; Armato SG, et al. Lung cancer: performance of automated lung nodule detection applied to cancers missed in a CT screening program. 225: 685-692

31. Radiology 2003; Brown MS, et al. Lung micronodules: automatic method for detection at thin-section CT- initial experience. 226: 256-262

32. Radiology 2003; Shah PK, et al. Missed non-small cell lung cancer: radiographic findings of potentially resectable lesions evident only in retrospect. 226: 235-241

33. Radiology 2003; Earnest F, et al. Respecting patient autonomy: screening at CT informed consent. 226: 633-634

34. Radiology 2003; Swensen SJ, et al. Lung cancer screening with CT: Mayo clinic experience. 226: 756-761

35. Thorax 2002; Mulshine JL, Smith RA. Lung cancer 2: screening and early diagnosis of lung cancer. 57: 1071-1078

36. Radiology 2004; Henschke CI, et al. C screening for lung cancer: suspiciousness of nodules according to size on baseline scans. 231: 164-168

37. Radiology 2004; Brenner DJ. Radiation risks potentially associated with low-dose CT screening of adult smokers for lung cancer. 231: 440-445

38. Radiology 2004; Kostis WJ, et al. Small pulmonary nodules: reproducibility of three-dimensional volumetric measurement and estimation of time to follow-up CT. 231: 446-452

39. Radiology 2004; Revel MP, et al. Are two-dimensional CT measurements of small noncalcified pulmonary nodules reliable? 231: 453-458

40. Radiology 2004; Revel MP, et al. Pulmonary nodules: preliminary experience with three-dimensional evaluation. 231: 459-466

41. Radiology 2004; Jennings SG, et al. Lung tumor growth: assessment with CT- comparison of diameter and cross-sectional area with volume measurements. 231: 866-871

42. Radiology 2004; Tiggers S, et al. Routine chest radiography in a primary care setting. 233: 575-578

43. Radiology 2004; Boll DT, et al. Volumetric assessment of pulmonary nodules with ECG-gated MDCT. 233: 1217-1223

44. Radiology 2004; Li, F, et al. Malignant versus benign nodules at CT screening for lung cancer: comparison of thin-section CT findings. 233: 793-798

45. Chest 2004; Libby DM, et al. Managing the small pulmonary nodule discovered by CT. 125: 1522-1529

46. Radiology 2005; Zhao B, et al. Pulmonary metastases: effect of CT section thickness on measurement- initial experience. 234: 934-939

47. Radiology 2005; Swensen SJ, et al. CT screening for lung cancer: five-year prospective experience. 235: 259-265

48. Radiology 2005; Goo JM, et al. Volumetric measurement of synthetic lung nodules with multi-detector row C: effect of various image reconstruction parameters and segmentation threshold on measurement accuracy. 235: 850-856

49. AJR 2005; Lindell RM, et al. Lung cancer screening experience: a retrospective review of PET in 22 non-small cell lung carcinomas detected on screening CT in a high-risk population. 185: 126-131

50. Radiology 2005; Kim JS, et al. Automated detection of pulmonary nodules on CT images: effect of section thickness and reconstruction interval- initial results. 236: 295-299

51. Radiol Clin N Am 2005; Hartman TE. Radiologic evaluation of the solitary pulmonary nodule. 43: 459-465

52. AJR 2005; Leader JK, et al. Pulmonary nodule detection with low-dose C of the lung: agreement among radiologists. 185: 973-978

53. Radiololgy 2005; Ko JP, et al. Effect of CT image compression on computer-assisted lung nodule volume measurement. 237: 83-88

54. Radiology 2005; MacMahon H, et al. Guidelines for management of small pulmonary nodules detected on CT scans: a statement from the Fleischner society. 237: 395-400

55. Radiology 2005; Li F, et al. Computer-aided detection of peripheral lung cancers missed at CT: ROC analysis without and with localization. 237: 684-490

56. Radiology 2006; Gur D. Lung cancer screening: radiology's opportunity here and now. 238: 395-397

57. J Nucl Med 2006; Bunyaviroch T, Coleman RE. PET evaluation of lung cancer. 47: 451-469

58. AJR 2006; Goodman LR, et al. Inherent variability of CT lung nodule measurements in vivo using semiautomated volumetric measurements. 186: 989-994

59. AJR 2006; Yuan R, et al. Computer-aided detection in screening CT for pulmonary nodules. 186: 1280-1287

60. AJR 2006; Revel MP, et al. Software volumetric evaluation of doubling times for differentiating benign versus malignant pulmonary nodules. 187: 135-142

61. AJR 2006; Mascalchi M, et al. Risk-benefit analysis of x-ray exposure associated with lung cancer screening in the Itallung-CT trial. 187: 421-429

62. Radiology 2006; Gietema HA, et al. Pulmonary nodules detected at lung cancer screening: interobserver variability of semiautomated volume measurements. 241: 251-257

63. AJR 2006; Kakeda S, et al. Effect of temporal subtraction technique on interpretation time and diagnostic accuracy of chest radiography. 187: 1253-1259

64. Radiology 2006; Boone JM. Multidetector CT: opportunities, challenges, and concerns associated with 64 or more detector rows. 241: 334-337

65. AJR 2007; Petrou M, et al. Pulmonary nodule volumetric measurement variability as a function of CT slice thickness and nodule morphology. 188: 306-312

66. JAMA 2007; Bach PB, et al. Computed tomography screening and lung cancer outcomes. 297: 953-961

67. JAMA 2007; Black WC, Baron JA. CT screening for lung cancer. Spiraling into confusion? 297: 995-997

68. Thorax 2007; Black C, et al. Population screening for lung cancer using computed tomography, is there evidence of clinical effectiveness? A systemic review of the literature. 62: 131-138

69. Radiology 2007; Lindell RM, et al. Five-year lung cancer screening experience: CT appearance, growth rate, location, and histologic features of 61 lung cancers. 242: 555-562

70. Radiology 2007; NY Early Lung Cancer Action Project. CT screening for lung cancer: diagnoses resulting from the New York early lung cancer action project. 243: 239-249

71. Radiographics 2007; Park CM, et al. Nodular ground-glass opacity at thin-section CT: histologic correlation and evaluation of change at follow-up. 27: 391-408

72. Radiology 2008; Gierada DS, et al. Lung cancer: interobserver agreement on interpretation of pulmonary findings at low-dose CT screening. 246: 265-272

73. Radiology 2008; Li F, et al. Lung cancers missed on chest radiographs: results obtained with a commercial computer-aided detection program. 246: 273-280

74. AJR 2008. Ravenel JG, et al. Screening for lung cancer. 190: 755-761

75. AJR 2008. Li F, et al. Improved detection of small lung cancers with dual-energy subtraction chest radiography. 190: 886-891

76. Chest 2006; Libby DM, et al. CT screening for lung cancer. The value of short-term CT follow-up. 129: 1039-1042

77. Radiology 2008; McMahon PM, et al. Estimating long-term effectiveness of lung cancer screening in the Mayo CT screening study. 248: 278-287

78. Radiology 2009; Xu DM, et al. Smooth or attached solid indeterminate nodules detected at baseline CT screening in the NELSON study: cancer risk during 1 year of followup. 250: 264-272

79. AJR 2009; Rampinelli C, et al. In vivo repeatability of automated volume calculations of small pulmonary nodules with CT. 192: 1657-1661

80.  Radiology 2010; Pickhardt PJ, et al. Colorectal and extracolonic cancers detected at screening CT colonography in 10286 asymptomatic adults. 255: 83-88

81. Radiology 2010; de Hoop B, et al. Screening for lung cancer with digital chest radiography: sensitivity and number of secondary work-up examinations. 255: 629-637

82. Radiology 2010; de Hoop B, et al. Computer-aided detection of lung cancer on chest radiographs: effect on observer performance. 257: 532-540

83. Radiology 2010; Shemesh J, et al. Ordinal scoring of coronary artery calcifications on low-dose CT scans of the chest is predictive of death from cardiovascular disease. 257: 541-548

84. AJR 2011; Goo JM, et al. Ground-glass nodules on chest CT as imaging biomarkers in the management of lung adenocarcinoma. 196: 533-543

85. Radiology 2011; Freedman MT, et al. Lung nodules: improved detection with software that supprresses the rib and clavicle on chest radiographs. 260: 265-273

86. Radiology 2011; Rampinelli C, et al. Extrapulmonary malignancies detected at lung cancer screening. 261: 293-299

87. JAMA 2011; Oken MM, et al. Screening by chest radiograph and lung cancer mortality. The prostate, lung colorectal, and ovarian (PLCO) randomized trial. 306: 1865-1873

88. N Engl J Med 2011; National lung screening trial research team, Aberle DR, et al. Reduced lung-cancer mortality with low-dose computed tomographic screening. 365: 395-409

89. AJR 2011; Larke FJ, et al. Estimated radiaiton dose associated with low-dose chest CT of average-sized participants in the national lung screening trial. 197: 1165-1169

90. Radiology 2012; Wang Y, et al. No benefit for consensus double reading at baseline screening for lung cancer with the use of semiautomated volumetry software. 262: 320-326

91. AJR 2012; Wagnetz U, et al. CT screening for lung cancer: implication of lung biopsy recommendations. 198: 351-358

92. Radiology 2012; Sverzellati N, et al. Relationship and prognostic value of modified coronary artery calcium score, FEV1, and emphysema in lung cancer screening population: the MILD trial. 262: 460-467

93. AJR 2012; Jacobs PC, et al. Coronary artery calcium can predict all-cause mortality and cardiovascular events on low-dose CT screening for lung cancer. 198: 505-511

(94) AJR 2013; Kim YK, et al. Reduced radiation exposure of the female breast during low-dose chest CT using organ-based tube current modulation and a bismuth breast shield: comparison of image quality and radiation dose. 200: 537-544

(95) AJR 2014; Xu DM, et al. Retrospective review of lung cancers diagnosed in annual rounds of screening. 203: 965-972

(96) Radiology 2015; Yankelevitz DF, et al. CT screening for lung cancer: nonsolid nodules in baseline and annual repeat rounds. 277: 555-564

(97) Radiology 2006; Winer-Muram HT. The solitary pulmonary nodule. 239: 34-49

(98) Radiol Clin N Am 2005; Hartman TE. Radiologic evaluation of the solitary pulmonary nodule. 43: 459-465

(99) J Am Coll Radiol 2015; McKee BJ, et al. Experience with a CT screening program for individuals at high risk for developing lung cancer. 12: 192-197

(100) Radiology 2016; Liang M, et al. Low-dose CT screening for lung cancer: computer-aided detection of missed lung cancers. 281: 279-288

(101) AJR 2016; Henschke CI, et al. CT screening for lung cancer: part-solid nodules in baseline and annual repeat rounds. 207: 1176-1184

(102) AJR 2017; Pinsky PF, et al. Lung cancer risks associated with new solid nodules in the national lung screening trial. 209: 1009-1014

(103) Radiology 2018; Liu Y, et al. Radiologic features of small pulmonary nodules and lung cancer risk in the national lung screening trial: a nested case-control study. 286: 298-306

(104) AJR 2018; Carter BW, et al. Screening for lung cancer: lexicon for communicating with health care providers. 210: 473-479

(105) Radiology 2018; Kavanagh J, et al. Importance of long-term low-dose follow-up after negative findings at previous lung cancer screening. 289: 218-224

(106) Radiology 2019; Hammer MH, et al. Cancer risk in subsolid nodules in the National Lung Screening Trial. 293: 441-448

(107) Radiology 2020; Godoy MCB. Conservative management of juxtapleural nodules at low-dose CT lung cancer screening: is this proudent? 297: 719-720

(108) AJR 2021; Mendoza, D, et al. Clinicopathologic and longitudinal imaging features of lung cancer associated with cystic airspaces: a systemic review and meta-analysis. 216: 318-329

(109) AJR 2021; Barbosa EJM, et al. Real-world lung cancer screening performance, smoking behavior, and adherence to recommendations: Lung-RADS category and smoking status predict adherence. 216: 919-926