Lung Cancer Screening
Few topics evoke as much disagreement and debate among interested parties as screening for lung cancer. National Institutes of Health-sponsored studies in the early 1970s failed to show a mortality benefit in various screening programs using chest radiography designed to detect early lung cancer. Few people on the outside of this debate are aware that none of those studies, and none of the current ongoing studies, employed a design that compared a group that was screened with a group that was not screened. Even the current National Lung Screening Trial uses, as its control group, a cohort screened with plain chest radiographs. Since the publication of National Institutes of Health screening trials in the 1970s, the mortality of lung cancer has continued to rise, and nearly every grant, review paper, or conference presentation on this topic leads with some variation of, ”Lung cancer kills more people than the next three most lethal cancers combined...breast, colon, and prostate.” With this in mind, the topic of lung cancer screening has been revisited in countless opinion pieces, editorials, review articles, and, unfortunately, original studies with various design flaws. Most of these flaws relate to the cost of screening and difficulty in subject accrual to a screening study with a truly unscreened population.
In the absence of quality data, patient advocates and some clinicians argue in favor of lung cancer screening on the basis that this is the major hope for reducing the mortality of lung cancer. These points are countered by equally vociferous arguments that this is the one end point we have repeatedly proven we cannot yet achieve with lung cancer screening. Much of the debate has centered around the use of low-dose, single breath hold, helical CT screening for lung cancer, because of some very provocative studies published by Henschke and colleagues in the 1990s on the utility of these studies in a structured program of lung cancer screening.1-3 The debate on the cost effectiveness of this approach sounds very much like the same debate that occurred almost 2 decades ago on the use of screening mammography for breast cancer and the appropriate age at which screening should begin. In order to avoid the dogma that sometimes dominates the current dialogue on lung cancer screening, perhaps the debate should not center on the CT scan itself, but on how we can effectively screen for lung cancer, even if CT screening proves to be ineffective at reducing mortality. New research on early detection methods that combine clinical history, biomarker development, and genetic risk profiling might enable advances that supplement, or even bypass, the use of CT screening. If we think of screening as a process, as opposed to a single test, then we can progress toward a solution that truly does reduce the burden of lung cancer mortality. The links below are to original articles that typify the current controversy over CT screening for lung cancer. These two articles will be referenced frequently on both sides of this argument, especially as we await publication of early results from the National Lung Screening Trial, which is linked below.
Article showing a significantly better survival in lung cancer patients whose disease was detected in a screening program.
International Early Lung Cancer Action Program investigators, Henschke CI, Yankelevitz DF, et al. Survival of patients with stage I lung cancer detected on CT screening. N Engl J Med 2006; 355:1763-1771
Article indicating that screening produced a lung cancer mortality that did not differ significantly from what would otherwise have been expected.
PB Bach, JR Jett, U Pastorino, et al. Computer tomography screening and lung cancer outcomes. JAMA 2007; 297:953-961
NLST Web site
Douglas Arenberg, MD, FCCP
Associate Professor of Medicine
Pulmonary and Critical Care
University of Michigan Medical School
