|
|
|
|
|
|
 |
 |
 |
|
 |
|
 |
|
 |
|
 |
|
 |
|
|
Screening for lung cancer with computed tomography:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Earn Category I CME credit by reading this article and the associated article and successfully completing the post-test. Successful completion is defined as a cumulative score of at least 70% correct. This material has been reviewed and is approved for 1 hour of clinical Category I (Preapproved) CME credit by the AAPA. The term of approval is for 1 year from the publication date of August 2004. |
Learning objectives
|
Some radiology clinics are now using—and many are heavily advertising—low-dose computed tomography (LDCT) as a screening tool for lung cancer. Patients are already asking about this recent application of a newer technology, and clinicians need to have an answer ready. Is this new test all that it is advertised to be? How should we answer our patients' questions about it?
Lung cancer is the No. 1 cancer killer in the United States, and it now kills more women than breast cancer. The American Cancer Society (ACS) estimates that 25% of all cancer deaths among women are due to lung cancer, while 15% are due to breast cancer. Overall, the ACS estimates that there will be 173,770 new cases of lung cancer and 160,440 deaths among men and women from it in 2004 (see Figures 1 and 2).1
Click here to view full-size graphic
Click here to view full-size graphic
A significant problem with this disease is that more than 50% of cases are found at an advanced stage, when treatment is less effective than at early stages (see Table 1).2 Lung resection currently offers the best hope for cure for most forms of lung cancer and is most effective with early-stage disease. Thus, the ability to find lung cancers early through screening may reduce the number of deaths. Note that because small cell lung cancer metastasizes early, it is not generally considered a surgical disease and so was not a focus of these studies.
TABLE 1
|
||
| Stage | TNM subset |
5-year survival, %
|
| 0 | Carcinoma in situ |
Not available
|
| 1A | T1 N0 M0 |
67
|
| 1B | T2 N0 M0 |
57
|
| IIA | T1 N1 M0 |
55
|
| IIB | T2 N1 M0 |
39
|
| IIIA | T3 N1 M0 or any T, N2 M0 |
23
|
| IIIB | T4, any N M0 or any T, N3 M0 |
5
|
| IV | Any T, any N, M1 |
1
|
| Key to staging symbols: T = Primary tumor. TX = Primary tumor cannot be assessed, or tumor cells found in sputum or bronchoscopic washings but not on radiography. T0 = No evidence of primary tumor. Tis = Carcinoma in situ. T1 = Tumor <3 cm in size, surrounded by lung or visceral pleura. T2 = Tumor with any of the following features: >3 cm; involves the main bronchus; >2 cm distal to carina; invades the visceral pleura; associated with atelectasis or obstructive pneumonitis that extends to the hilum but does not involve the entire lung. T3 = Tumor of any size that invades any of the following: chest wall, diaphragm, mediastinal pleura, pericardium; or tumor in the main bronchus <2 cm from but not involving carina; or associated atelectasis or obstructive pneumonitis of the entire lung. T4 = Tumor of any size that involves any of the following: mediastinum, heart, great vessels, trachea, esophagus, vertebral body, carina; or tumor with a malignant effusion or with satellite tumor nodules within the same lobe of lung as the primary tumor. N = Lymph node. NX = Regional lymph nodes cannot be assessed. N0 = No regional lymph node metastasis. N1 = Metastasis to ipsilateral peribronchial and/or ipsilateral hilar lymph nodes, and intrapulmonary nodes involved by direct extension of the tumor. N2 = Metastasis to ipsilateral mediastinal or subcarinal nodes. N3 = Metastasis to contralateral lymph nodes or to ipsilateral or contralateral scalene or supraclavicular nodes. M = Metastasis. MX = Presence of distant metastasis cannot be assessed. M0 = No distant metastasis. M1 = Distant metastasis present. *Tumor, Node, Metastasis. |
||
An effective screening tool is one that shows some benefit to patients undergoing the test, usually an increase in life expectancy or quality of life; causes no harm, either from the test itself or from invasive testing used to follow up a false-positive result; is cost-effective; and is widely available. Low specificity (a high rate of false-positive results) should be offset by high sensitivity (a low rate of false-negative results), which gives the ability to identify all patients who have the disease.
Both chest x-ray (CXR) and sputum cytology were thought to meet these criteria 30 to 40 years ago, until five randomized, controlled studies showed that they produced no survival benefit. Although a greater number of early-stage lung cancers were detected in the screened cohort than in the control group, no survival benefit was shown, even among patients who underwent surgery.3 Both the screened cohort and the control group had statistically the same percentage of lung cancer deaths, leading researchers to conclude that these modalities were inappropriate screening tools for this disease. Questions remain, however, regarding the studies' designs and results.
During the intervening years, little progress was made in finding a screening tool for lung cancer, until investigation into low-dose spiral CT (LDCT) was begun in the early 1990s.
Five separate studies looked at CT as a lung cancer screening tool, and all found high sensitivity but low specificity (see Table 2). The studies described here are landmark studies or well-designed large trials, representing important research and results over the past decade. More studies are currently under way.
TABLE 2
|
|||
| Study (year) | Number and characteristics | Noncalcified nodules found, total and by size of tumor (%) | Lung cancers found, total (%) and by staging* |
| Henschke et al4 (1999) |
1,000 smokers, aged >59 y (M, 540; W, 460) | 233 (23.3) >5 mm, 97 (9.7) |
27 (2.7) IA, 22 IB, 1 IIA,1 IIIA, 2 IIIB, 1 |
| Sone et al6 (2001) |
5,483, 46.1% smokers or former smokers, aged >39 y (M, 2,971; W, 2,512) | 279 (5.1) >3 mm, 170 (3.1) |
22 (0.4) IA, 20 IB, 2 |
| Sobue et al7 (2002) |
1,611, 86% smokers or former smokers, aged >39 y (M, 1,415; W, 196) | 186 (11.5) | 13 (0.81) IA, 9 IB, 1 IIIA, 2 IIIB, 1 |
| Swensen et al9 (2002) |
1,520 smokers, aged >49 y (M, 785; W, 735) | 82 (51.4) >4 mm, 475 (31.3) |
21 (1.38) IA, 13 IB, 1 IIA, 4 IIIA, 2 Limited stage small cell, 1 |
| Diederich et al10 (2002) |
817 smokers, aged >39 y (M, 588; F, 229) | 409 (50.1) >5 mm, 154 (18.8) |
11 (1.35) IA, 5 IB, 1 IIA, 1 IIB, 1 IIIA, 2 IIIB, 1 |
| Key: F, female; M, male/men; W, women. *See Table 1 for key to staging. |
|||
In the Early Lung Cancer Action Project (ELCAP), begun in 1992, Henschke and coworkers reported the results of annual LDCT scans and CXRs obtained in 1,000 smokers.4 Noncalcified nodules were found in 233 patients at the study participants' first screening, and 27 of these nodules were malignant. Although 20 of the 27 nodules were not seen on CXR, all nodules that were apparent on CXR were found on CT. Of the 27 malignancies, 23 were stage I and all but one were resectable. In the second screening, seven cancers were detected in 30 patients who had positive findings on the scans.5 Five were stage I, one was stage IIIA, and one was a limited-stage small cell lung cancer. This study showed a high rate of false-positive results with many noncalcified nodules that were not cancer. Most cancers found, however, were at an early stage.
In a 1996 Japanese study, Sone and colleagues screened 5,483 subjects aged 40 to 74 years using LDCT with repeat tests annually in each of the next 2 years.6 The initial screening detected suspicious nodules in 279 (5.1%) subjects; 22 (0.40%) patients were confirmed to have stage I lung cancer. Corresponding figures for the subsequent 2 years of the study were suspicious nodules found in 173 (3.9%) of 4,425 patients, with 25 (0.56%) lung cancers confirmed; and suspicious nodules found in 136 (3.5%) of 3,878 patients, with nine (0.23%) lung cancers confirmed. Most of the lung cancers found were stage I (55 of 60). As in the ELCAP study, researchers found a high rate of false-positive results with CT, and most of the cancers identified were at an early stage.
A third study, the Anti-Lung Cancer Association Project, subjected 1,611 patients to LDCT in addition to both CXR and sputum cytology.7 Suspicious nodules were found in 186 subjects (11.5%); 13 (0.8%) were found to be lung cancer. Eight cancers were found by CT alone, and five others were found by both CT and CXR. One additional cancer was found by sputum cytology. Interestingly, suspicious findings that warranted further investigation were associated less often with CXR (3.4%) and sputum cytology (0.8%) than with CT. The researchers reported a 71% 5-year survival rate for patients with those cancers detected by CT screening, a higher rate than is reported for patients with cancers detected by CXR (30%-40%).8
In a fourth study, three annual LDCT exams of the chest and upper abdomen plus annual sputum cytology were obtained in 1,520 subjects aged 50 or older who had smoked for 20 pack-years or more.9 Noncalcified pulmonary nodules were found in 1,049 subjects (69%); 40 patients had lung cancer. On initial screening, 782 (51.4%) patients were found to have worrisome abnormalities; 21 (1.38%) turned out to have cancer. Thirty-eight of the cancers were found with CT alone, and two were found with cytology alone. Of the 35 non–small cell lung cancers found on CT, 21 (60%) were found to be stage I. As in previous studies, a high number of false-positive results occurred, and a reasonable percentage of stage I cancers were found.
In a German study of smokers aged 40 years and older, noncalcified nodules were found in 409 of 817 subjects (50.1%); 11 of these patients (1.35%) had lung cancer found on CT.10 Six of these cancers were stage I, two each were stage II and IIIA, and one was stage IIIB. No lung cancer was found in the 298 subjects who were younger than 50 years. Although the percentage of lung cancers found was similar to previous studies, the percentage found at stage I was smaller.
The studies described in the previous section and in Table 2 all show that LDCT finds early cancers; the incidence of detected cancers ranges from 0.4% to 2.7%. All show as well, however, that LDCT had rates of false-positive results between approximately 5% and 50%. This variation may be explained by the differences in the groups studied: those studies that included patients aged 40 years and older had the lowest rates of cancers detected, and no cancers were found in patients younger than 50 years. Studies have shown an increased incidence of lung cancer with increasing age.11 Furthermore, the study that had the lowest percentage of smokers had a lower percentage of cancers found.6
Study design To date, no randomized controlled studies to evaluate CT as a screening test for lung cancer have been completed, although several are under way. Those studies that show a benefit are observational and may reflect bias. In addition, we have no long-term survival data for most of these studies, a main end point for lung cancer screening. As noted above, earlier studies of screening with CXR and sputum cytology showed that even though these modalities identified stage I cancers (although not as many as found in later studies by LDCT), there was no difference in the long-term survival of patients who were found to have lung cancer.
False-positive rate The high number of false-positive results with LDCT screening is also troubling, because false-positive results lead to expensive and invasive diagnostic testing, which leads to increased morbidity and mortality associated with radiation, biopsy, and surgical procedures. The emotional stress and turmoil that patients undergo during testing must also be considered.
Cost-effectiveness No randomized controlled studies have determined whether LDCT screening is cost effective. A scan costs at least several hundred dollars. If payment by insurance companies becomes standard and the nation's smokers and former smokers obtain the screening tests, the costs will be passed on to insurance consumers one way or another.
Whom to screen? The various studies used different inclusion criteria, with some including both smokers and nonsmokers. Screening in the United States will likely be limited to current and former smokers, but patient age and number of pack-years smoking must be defined. For example, should persons who are passively exposed to large amounts of cigarette smoke be screened?
Overdiagnosis A study of autopsy results showed that one sixth of lung cancers at autopsy were not clinically recognized and were not related to the patient's death.12 These results suggest that not all lung cancers found on a screening test will kill the patient; some are slow-growing, relatively indolent cancers that a patient dies with, but not from. One theory suggests that overdiagnosis may have been responsible for the fact that no survival benefit was associated with the increased number of early cancers detected with screening by CXR and sputum cytology. The concept of overdiagnosis suggests an increased number of cases diagnosed, early cases found, and 5-year survivals, but the same number of deaths due to the disease.
Although LDCT shows promise, the evidence is currently lacking that it increases survival after lung cancer is diagnosed. Furthermore, the high rate of false-positive results may lead to unnecessary additional procedures to determine the nature of nodules found on an LDCT scan.
Presently, professional radiology groups and the American Cancer Society do not recommend LDCT for routine lung cancer screening. The number of false-positive results must be lower before LDCT becomes a useful and widespread screening tool for lung cancer. In addition, experts must identify the ideal age and smoking history of the population who would benefit most from screening. Further studies are needed to explore the usefulness of LDCT and to address some of the attendant questions.
A patient who is at risk for lung cancer and interested in LDCT should consider undergoing the test as part of a study. For example, the National Cancer Institute is conducting the National Lung Screening Trial, comparing CXR and LDCT. In the New York area, the ELCAP group is conducting the I-ELCAP (International ELCAP), and other studies of LDCT are currently under way throughout the United States.
REFERENCES
1. American Cancer Society. Estimated new cancer cases and deaths by sex for all sites, US, 2004. Available at: http://www.cancer.org/downloads/MED/Page4.pdf. Accessed July 1, 2004.
2. Greenlee RT, Murray T, Bolden S, Wingo PA. Cancer statistics, 2000. CA Cancer J Clin. January-February 2000;50:7-33.
3. Eddy DM. Screening for lung cancer. Ann Intern Med. August 1, 1989;111:232-237.
4. Henschke CI, McCauley DI, Yankelevitz DF, et al. Early Lung Cancer Action Project: overall design and findings from baseline screening. Lancet. July 10, 1999;354:99-105.
5. Henschke CI, Naidich DP, Yankelevitz DF, et al. Early Lung Cancer Action Project: initial findings on repeat screenings. Cancer. July 1, 2001;92:153-159.
6. Sone S, Li F, Yang ZG, et al. Results of three-year mass screening programme for lung cancer using mobile low-dose spiral computed tomography scanner. Br J Cancer. January 5, 2001;84:25-32.
7. Sobue T, Moriyama N, Kaneko M, et al. Screening for lung cancer with low-dose helical computed tomography: anti-lung cancer association project. J Clin Oncol. 2002;20:911-920.
8. Ajiki W, Matsuda T, Sato Y, et al. Standard method of calculating relative survival rates in population-based cancer registries: an investigation using stomach cancer patients [in Japanese]. Jpn J Cancer Clin. 1997;43:1005-1014.
9. Swensen SJ, Jett JR, Sloan JA, et al. Screening for lung cancer with low-dose spiral computed tomography. Am J Respir Crit Care Med. 2002;165:508-513.
10. Diederich S, Wormanns D, Semik M, et al. Screening for early lung cancer with low-dose spiral CT: prevalence in 817 asymptomatic smokers. Radiology. 2002;222:773-781.
11. Bepler G, Goodridge Carney D, Djulbegovic B, et al. A systematic review and lessons learned from early lung cancer detection trials using low-dose computed tomography of the chest. Cancer Control. 2003;10:306-314.
12. Chan CK, Wells CK, McFarlane MJ, Feinstein AR. More lung cancer but better survival: implications of secular trends in "necropsy surprise" rates. Chest. August 1989;96:291-296.
Michael Weber. Screening for lung cancer with CT: An examination of the evidence. JAAPA August 2004;17:14-18.
Copyright © 2004, Advanstar Medical Economics Healthcare Communications at Montvale, NJ 07645-1742. All rights reserved.
