Screening for Lung Cancer: Is It Really Without Value?
Screening for Lung Cancer: Is It Really Without Value?
Abstract & Commentary
Synopsis: A number of studies have suggested that screening high-risk populations for lung cancer with chest radiographs or sputum cytology results in the early diagnosis of lung cancers at a stage of disease more likely to be cured with surgical resection; however, convincing evidence for reduced mortality rates has not emerged. Among the 33,000 male smokers between the ages of 50 and 69 years who were screened for the presence of lung cancer before entering the Finnish alpha-tocopherol beta carotene cancer prevention study, 93 were found to have lung cancer. The men whose tumors were detected by screening were twice as likely as nonscreened men to have resectable disease and the five-year survival was also twice as great (19% vs 10%).
Source: Salomaa E-R, et al. Prognosis of patients with lung cancer found in a single chest radiograph screening. Chest 1998;114:1514-1518.
Lung cancer screening has not been widely adopted because of the impression that it does not save lives. However, a number of large-scale screening studies have demonstrated the detection of asymptomatic cancers, a higher fraction of which are surgically resectable. The problem is that the mortality rate has not been conclusively demonstrated to be improved by screening.
In 1985, a Finnish research group began a large randomized cancer prevention trial to assess whether alpha-tocopherol and/or beta carotene could reduce the incidence of lung cancer in a group of high-risk male subjects, smokers between the ages of 50 and 69 years. The results of that study have been reviewed previously in these pages. However, in order to qualify for the study, potential subjects were identified and the presence of lung cancer was ruled out by subjecting the 33,743 participants to a screening chest radiograph. Of these, 93 men were found to have asymptomatic lung cancer, and thus, they were ineligible for the lung cancer prevention study. These 93 men were then compared to 239 age-matched control men who were diagnosed as having lung cancer through their own encounters with the health care system; generally, this was either through the follow-up of cancer-related symptoms (69%) or the incidental discovery of lung cancer in the course of working up another medical problem (31%).
The histology of the lung cancer in the screened cases was similar to that in the unscreened cases. However, the stage distribution was significantly different. About 54% of the screened cases had stage I disease compared to 30% of the unscreened cases. Furthermore, 37% of the screened cases underwent radical surgery with curative intent compared to 19% of the unscreened cases (P < 0.001). Early-stage patients who were not operated on with curative intent often had underlying lung disease or other severe diseases that made the radical surgery too dangerous to undertake. The survival of the screened men was also significantly better than survival of the unscreened men. Nineteen percent of screened men and 10% of unscreened men survived five years after diagnosis (P < 0.001). If men in the two groups were matched for stage, no difference in survival was detected. That is, stage I and II screened patients had the same prognosis as stage I and II unscreened patients. However, the screening appeared to exert its beneficial effects by bringing cases to medical attention at an earlier and more treatable stage.
Commentary
This study was not a randomized comparison of screened and unscreened populations. Instead, Salomaa and associates took advantage of the fact that in order to conduct an appropriate cancer prevention study, they had to assure that study participants did not have cancer at the time of study entry. Among the 33,743 individuals who agreed to participate, 93 men were found to have asymptomatic lung cancer detected by routine screening chest radiography. These men were disproportionately in early stages of disease and they had a treatment outcome compatible with the early stage of their diagnosis.
It is possible to criticize this study on a number of grounds. First, the fact that the screened men had agreed to participate in the study and the unscreened men had not may be a feature leading to selection bias. Persons who agree to participate in studies may be more "health-conscious" than counterparts who declined to participate and some lifestyle features associated with health consciousness could influence survival. Other features not balanced in the two populations such as socioeconomic status and education could also lead to disparate results.
Lead-time bias is not a valid basis for criticizing this study. Lead-time bias suggests that the better prognosis emerges from the screened cases because their survival is being measured from an earlier time point. This type of bias has been implicated in difficulties interpreting the results of prostate cancer screening programs. The major basis of this bias is the notion that the screening test is detecting tumors that do not need to be detected and would not influence patient survival. Clearly, no such benign outcome from any lung cancer has yet been documented. Furthermore, in this study, matching the screened and unscreened populations for stage and aggressiveness of surgery demonstrates that similar outcomes emerged. The main benefit of screening was, as is usually hoped for, early detection.
Of course, in these days of cost-effective and evidence-based medicine, the data presented in this paper are unlikely to lead to massive screening efforts or a change in the recommendations of the major preventive health organizations. However, for those of us trying to do the best we can for our individual patients, this paper adds some encouragement in support of the notion that chronic smokers over age 50 who are asymptomatic may benefit from annual chest radiographs. It is more likely that any cancer detected in this population will be at an earlier and more treatable stage than a cancer detected as a consequence of the symptoms it has produced.
Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.