Lung cancer is the leading cause of cancer deaths worldwide ^[1] ; approximately 85% of all lung cancers occur in current or former cigarette smokers ^[2] . Recently, two important studies ^{[3, 4]} received worldwide press coverage for the discordant results reported ^[5–8] . The critical issue was whether spiral computed tomography (CT) scan screening could reduce lung cancer deaths in heavy cigarette smokers and could thus be considered the ultimate test of any effective cancer screening.

The first study ^[3] reported that low-dose CT screening could lead to a therapeutic strategy that resulted in a 10-year survival rate of 88% for patients with stage I disease. The authors, in a large collaborative study, screened 31,567 asymptomatic persons at risk for lung cancer using low-dose CT; 27,456 repeated screenings were performed 7–18 months after the previous screening. They estimated the 10-year lung cancer–specific survival rate among participants with clinical stage I lung cancer that was detected on CT screening and diagnosed by biopsy, regardless of the type of treatment received, and among those who underwent surgical resection of clinical stage I cancer within 1 month. Screening resulted in a diagnosis of lung cancer in 484 participants. Of these participants, 412 (85%) had clinical stage I lung cancer, and the estimated 10-year survival rate was 88% in this subgroup (95% confidence interval ^[CI] , 84%–91%). Among the 302 participants with clinical stage I cancer who underwent surgical resection within 1 month after diagnosis, the survival rate was 92% (95% CI, 88%–95%). The eight participants with clinical stage I cancer who did not receive treatment died within 5 years after diagnosis.

The second study reported that there is no evidence that screening with chest CT scan does anything to reduce lung cancer deaths ^[4] . Those investigators used a validated lung cancer prediction model to estimate the expected numbers of various lung cancer outcomes among the combined cohort of 3,246 participants. To assess the effectiveness of CT screening, they then compared the observed numbers of lung cancer outcomes with the numbers of expected cases. They observed a more than threefold higher number of new lung cancer cases (144 observed versus 44.5 expected) and a tenfold higher number of lung cancer resections (109 observed versus 10.9 expected). However, there were no fewer advanced lung cancer cases (42 observed versus 33.4 expected) or lung cancer deaths (38 observed versus 38.8 expected).

Several limitations of the first study ^[3] have to be highlighted: (a) the data used were the baseline examination and first CT repeat; (b) the specific survival rate was reported for 412 (1.3%) clinical stage I patients; (c) survival in the vast majority of subjects after 2 years is not valuable; (d) deaths related to surgery were reported within 4 weeks, whereas the standard is 30 days ^[9] ; and (e) biopsies were performed in 535 of the 5,646 participants (9.5%) with positive results on chest CT.

The limitation of the second study ^[4] is that the investigators used a validated lung cancer prediction model to estimate the expected numbers of various lung cancer outcomes among 3,246 participants. To assess the effectiveness of CT screening, they compared the observed numbers of lung cancer outcomes with the numbers of expected cases. In this study, the majority of individuals who died from lung cancer, despite participation in annual screening, did not have their cancer detected at an early, likely curable, stage: 34% were initially diagnosed with stage III or IV non-small cell lung cancer (NSCLC), 18% were initially diagnosed with small cell lung cancer, and 16% had no documented diagnosis of lung cancer prior to their death from this disease. In this study, in which 85% of volunteers have now completed their fifth year of CT screening ^[10] , the survival duration of stage I patients was equivalent to the survival in the other study, but the proportion of patients with stage I disease fell dramatically after the second year, and the cumulative lung cancer mortality rate at 5 years was very close to that expected without screening. A reasonable explanation for such a discrepancy is that radiological CT imaging can detect early-stage, slow-growing, or indolent disease, but is unable to prevent more aggressive and early metastatic lung cancer. Autopsy study recognized that inconsequential lung cancers are not rare ^[11] . As a possible consequence, some persons, even those enrolled in screening trials, will live and die with their lung cancer, although the degree and impact on screening remains uncertain.

The major highlight that comes into view from these studies is that the natural history of lung cancer detected by CT is unknown. The biological features of CT-detected versus symptomatic lung cancers have been tested by cDNA microarray analysis, showing a similar malignant phenotype ^[12] . Nonetheless, one has to consider that such a case–control study analyzed only a small sample of patients (37) and a limited number of genes (3,231), with nine genes differentially expressed in the two groups. According to the current vision of NSCLC carcinogenesis, accumulation during a time of multiple genetic/epigenetic alterations needs to occur in a cell prior to (malignant) transformation ^[13–16] . It cannot be assumed that the biologic behavior of lung cancer parallels anatomical size or that small lesions are equivalent to early-stage disease. In fact, there are currently no data to confirm that a primary 5-mm lung tumor (~108 cells) has a significantly better prognosis than a 10-mm tumor (~109 cells) or even a 30-mm tumor (~2.7 x 1010 cells) ^[17] . Tumors may already have demonstrated their potential to remain localized or to metastasize by the time they are visible on CT imaging. In some studies, approximately 60% of patients with radiographically detected stage I disease (< 3 cm) died from lung cancer within 5 years, despite appropriate therapy ^[18] . This suggests that a high percentage of patients have disseminated occult disease at the time of presentation.

With newer and more sensitive methods of detection, sites of isolated tumor cells and micrometastases have now become apparent ^[19] . In fact, clinical studies have confirmed that patients with small tumors can harbor malignant cells in normal-appearing lymph nodes that are detectable only by polymerase chain reaction assay ^[20–23] . Other investigations have found tumor cells or circulating endothelial progenitor cells in the peripheral blood of patients with lung cancers of all sizes and stages ^{[24, 25]} . An experimental study showed that a 1-cm tumor can shed ~3–6 million cells into the blood every 24 hours ^[26] .

These findings pose a key challenge for estimating lung cancer risk.

Susceptibility differences in the form of rare, high-penetrance genes are suggested from studies of familial aggregation of lung cancer and a linkage study. Low-penetrance genes in the tobacco smoke metabolism pathways (phase I and II enzymes) and DNA repair pathways, as well as inflammation, cell cycle–related genes, and epigenetic mechanisms (methylation) have to be considered ^[15] . Mouse models confirmed the above hypothesis and demonstrated the polygenic nature of lung cancer susceptibility ^[27] . As an example, one study showed that genetic variants underlying a previously mapped lung cancer susceptibility locus could have opposing effects on cancer risk depending on whether an oncogenic somatic mutation at the same locus subsequently arises in cis or in trans ^[27] .

Increasing evidence suggests that the relative risk for lung cancer, as the relationship between smoking and lung cancer, may be different for women and men: women have greater susceptibility to tobacco carcinogens, but a lower rate of fatal outcome than men ^{[28, 29]} . Lindell et al. ^[30] , in a retrospective study carried out in high-risk individuals who underwent annual chest CT screening for 5 years, reported more TC-positive nodules mainly in women, suggesting a higher incidence of “overdiagnosis” than in men.

Interactions between race or ethnicity and tobacco exposure have been reported. African Americans and Native Hawaiians seem more susceptible and Latinos and Japanese less susceptible ^[31] .

A recent article, which used existing data from a large lung cancer case–control study, reported the development and validation of separate risk prediction models for never, former, and current smokers ^[32] . The authors computed the absolute risks for lung cancer and presented calculations for varying risk profiles, demonstrating that a currently smoking man with an estimated relative risk close to 9 had an 8.68% 1-year absolute risk for lung cancer, compared with a 0.56% annual incidence rate for his age group. Such a degree of risk would justify more intensive surveillance and counseling. All the variables in these models have strong, biologically plausible etiologic roles in lung cancer supported by numerous studies ^[33] , including our own ^[34] . The purpose of the study by Spitz et al. ^[32] was to create a parsimonious model for assessing lung cancer risk with a minimal number of risk predictors that is realistic to use in clinical practice. In our experience, patients are agreeable to complete health questionnaires, either self-administered or by personal interview ^[35] . The next important step is to incorporate pathway-based genetic variation data in the model to acknowledge the important contribution of genetic susceptibility to lung cancer risk. Adding such data is likely to further improve the sensitivity and specificity of the models, although incorporating genetic data may not be practical for community-based settings.

As the debate over lung cancer screening continues, it appears that we must learn more on the biology of this disease and integrate this knowledge with early diagnostic strategies. Although the main goal is to reduce mortality from lung cancer, the theoretical benefits of screening must be validated by appropriate, rigorous, robust randomized trials before being introduced into routine practice.

Acknowledgments

Funding for this work came from grant “Molecular markers for early detection of lung cancer in subjects eligible for TC scan” awarded by “Fondazione Compagnia di San Paolo” Turin, and Local Government (Liguria County), Genoa, Italy. We thank Dr. Ugo Pastorino (Thoracic Surgery Unit, INT, Milan) for stimulating discussion.

REFERENCES

_{# Jemal A, Siegel R, Ward E et al. Cancer statistics, 2007. CA Cancer J Clin 2007;57:43–66.
# Peto R, Darby S, Deo H et al. Smoking, smoking cessation, and lung cancer in the UK since 1950: Combination of national statistics with two case-control studies. BMJ 2000;321:323–329.
# Henschke CI, Yankelevitz DF, Libby DM et al.; International Early Lung Cancer Action Program Investigators. Survival of patients with stage I lung cancer selected on CT screening. N Engl J Med 2006;355:1763–1771.
# Bach PB, Jett JR, Pastorino U et al. Computed tomography screening and lung cancer outcomes. JAMA 2007;297:953–961.
# CNN.com. Study: Lung Cancer Screening Doesn’t Save Lives.
# abc News. Lung Cancer Screening Does Not Boost Survival.
# NPR: Early Detection of Lung Cancer May Not Save Lives.
# Newsweek Study: Screening Fails to Reduce Lung-Cancer Deaths.
# Harpole DH Jr, DeCamp MM Jr, Daley J et al. Prognostic models of thirty-day mortality and morbidity after major pulmonary resection. J Thorac Cardiovasc Surg 1999;117:969–979.
# Pastorino U. Early detection of lung cancer. Respiration 2006;73:5–13.
# Dammas S, Patz EF Jr, Goodman PC. Identification of small lung nodules at autopsy: Implications for lung cancer screening and overdiagnosis bias. Lung Cancer 2001;33:11–16.
# Bianchi F, Hu J, Pelosi G et al. Lung cancers detected by screening with spiral computed tomography have a malignant phenotype when analyzed by cDNA microarray. Clin Cancer Res 2004;10:6023–6028.}

Laura Paleari, Pierluigi Granone, Alessia Grozio, Alfredo Cesario, Patrizia Russo
Lung Cancer Unit, National Cancer Institute, Genoa, Italy;
Surgical Pathology Unit, Catholic University, Rome, Italy;
IRCCS “San Raffaele,” Rome, Italy

Correspondence: Patrizia Russo, Ph.D., Lung Cancer Unit, Department of Advanced Diagnostic Technology, National Cancer Research Institute, Largo Rosanna Benzi 10, I-16132 Genova, Italia. Telephone: 390105737272; Fax: 390105737273; e-mail: .(JavaScript must be enabled to view this email address)