Editorial For reprint orders, please contact [email protected]
Caution: screening for lung cancer Expert Rev. Resp. Med. 3(3), 203–205 (2009)
David E Midthun, MD Associate Professor of Medicine, Mayo Clinic, Rochester, MN 55905, USA Tel.: +1 507 284 0561 Fax: +1 507 284 3154 [email protected]
“More people die from lung cancer than from breast, colon and prostate cancer combined, yet we are not screening for lung cancer. Why not? ” When dealing with a disease that is common and frequently lethal, early detection makes sense – find the disease in early stage when treatment is more likely curative, and lives will be saved. The 5‑year survival rate of 16% for lung cancer pales by comparison with 88% for breast cancer, 65% for colon cancer and 100% for prostate cancer  . More people die from lung cancer than from breast, colon and prostate cancer combined, yet we are not screening for lung cancer. Why not? The answer is complicated, but it is essentially because no screening test (chest radiography, sputum cytology or computer tomography [CT]) has proven to save lives from lung cancer. The diagnosis of lung cancer is now primarily made in the pursuit of symptoms such as cough, hemoptysis or shortness of breath, and detection is too often at a stage that curative intervention, namely surgery, is no longer possible. An effective screening tool is desired. Prevention through smoking avoidance or cessation remains key, but improving outcomes for patients with lung cancer will result from early detection and effective treatment. We know that screening with chest radio graphy or CT leads to the detection of more cancers, more early-stage cancers and a marked improvement in survival. Each of those measures is a necessary outcome, but, even taken together, they are insufficient to show that screening is efficacious. The use of a screening test introduces biases that are inherent in screening and may result in the observation of improved survival without affecting mortality. The most significant of these include lead time, length time and overdiagnosis bias. Lead-time bias occurs
when a cancer is detected earlier than it would have been in the absence of screening; intervention occurs earlier, but the natural history of disease is not changed. As the measure of time between detection and death, apparent survival is lengthened, suggesting benefit, but true survival (time from disease onset to eventual death, whether it be due to lung cancer or not) is something we cannot measure. As a result of applying a screening test, although apparent survival is improved, mortality may remain unchanged.
“As the measure of time between detection and death, apparent survival is lengthened, suggesting benefit, but true survival … is something we cannot measure.” Length-time bias shows an apparent improvement in survival when that improvement is actually due to selective detection of cancers with a less progressive course and missing cancers that have the most rapidly progressive course. Application of a screening tool at specified intervals would have a higher likelihood of detecting a cancer with a more indolent course than one with a more rapid course that may present with symptoms between screenings. The result is to show an apparent improvement in survival through the better outcome of the more indolent cancers found with screening. Overdiagnosis bias is an extreme form of length-time bias in which indolent lung cancers are detected. These are cancers
© 2009 Expert Reviews Ltd
that would not have altered the expected survival when compared with the normal population. Patients with overdiagnosed lung cancers would have died of a cause other than lung cancer and the lung cancer may have gone undetected. People die with rather than die from these cancers. The natural history of these cancers does not need to be altered with detection and treatment. Advances in CT technology allow for scanning of the entire chest in less than 15 s and in a single breath hold. Prospective single-arm studies of low-dose spiral CT (LDSCT) screening have created considerable excitement about the potential to effectively screen for lung cancer. Usual clinical practice identifies only approximately 20% of lung cancers in stage I. CT screening series report that approximately 50–90% of cancers identified at baseline or with subsequent screening were stage I [2–9] . These findings suggest an ability of CT to shift the diagnosis from late-stage detection to early-stage detection when compared with historical clinical practice. The largest single-arm observational report of CT screening was the International Early Lung Cancer Action Program (IELCAP)  . Over 30,000 participants received a baseline scan and a total of 484 cancers were found; 412 (85%) of these were clinical stage I non-small cell carcinomas and 375 (77%) were surgical stage I. Actuarial 10‑year lung cancer survival was 80% overall, and the actuarial survival for those with clinical stage I was 88%. The median follow-up in this study was only 40 months and only one patient with cancer had reached the 10‑year mark  . The high percentage of cancer found in stage I suggests that LDSCT may have the potential to shift identification of a lung cancer to stage I from a more advanced stage. The calculated survival results are impressive. However, in the absence of a control arm, one cannot be certain that screening is not simply identifying more cancers in early stage while not changing the number of advanced cancers or the number of deaths from lung cancer. The IELCAP study conclusions also need to be interpreted in the light of significant conflicts of interest [10,11] .
“If CT screening is shown to achieve a 20% reduction in mortality from lung cancer, this would be the equivalent, in lives saved, of eliminating mortality from prostate cancer.” A subsequent report highlighted the concern that, despite earlier diagnosis, CT may not result in lives saved with screening. Investigators applied a validated lung cancer prediction model to three prospective, single-arm observational studies of CT screening combining 3246 participants  . CT screening found three times the number of expected cancers, resulted in ten times the expected number of resections (109 out of 144 cancers were resectable), and identified more than the expected number of advanced-stage cancers. Despite having a 94% actual 4‑year survival for participants with clinical stage I cancers who underwent surgery, CT screening did not result in a reduction in the expected 204
number of deaths from lung cancer. These results emphasize the need to evaluate the effectiveness of screening by mortality rather than survival. Reports of the single-arm prospective studies have resulted in a great deal of interest among physicians and the public. Results of the nonrandomized trials appear to be insufficient evidence to move CT screening from the realm of personal choice for those of means to a matter of public policy. A randomized, controlled trial is the only available method to demonstrate mortality reduction from screening. The NCI has sponsored the National Lung Screening Trial, which has enrolled over 50,000 participants aged 55–74 years, and current or former smokers of at least 30 packyears  . Participants were randomized between low-dose fast spiral CT and chest x-ray at baseline, 1 and 2 years with followup. There is great anticipation that this study will show mortality reduction with low-dose fast spiral CT scanning and that this effectiveness will be at a cost that is not prohibitive. Several randomized trials are underway in Europe [14–16] .
“The natural history of these cancers does not need to be altered with detection and treatment.” If CT screening is shown to achieve a 20% reduction in mortality from lung cancer, this would be the equivalent, in lives saved, of eliminating mortality from prostate cancer. A 25% reduction in mortality would have the equivalence of eliminating deaths from breast cancer. Given the number of lives lost to lung cancer each year, the stakes are high. Results from these randomized trials will be available in the next few years. Until a positive effect by mortality reduction from either of these studies is demonstrated, it is not likely that screening for lung cancer will be covered by health insurance plans. What is a clinician to do in the mean time? The answer is not clear. Screening with a chest radiograph or LDSCT has been shown to find more lung cancers, more cancers at an early stage, and improve survival, but neither has demon strated benefits in terms of lives saved. Potential harms include concerns, follow-up and unnecessary interventions resulting from false-positive scans, cost and the fact that radiation from screening CT may actually induce cancers. In the absence of clearly demonstrated benefit, the decision to case-find lung cancer with screening requires a discussion of the risks and potential benefits between the physician and the patient. Given the potential harm in unnecessary biopsies and surgery, increasing survival in the absence of mortality reduction suggests that screening benefit may not outweigh the risk. Financial & competing interests disclosure
The author has no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties. No writing assistance was utilized in the production of this manuscript. Expert Rev. Resp. Med. 3(3), (2009)
Caution: screening for lung cancer
Jemal A, Siegel R, Ward E et al. Cancer statistics. CA Cancer J. Clin. 58(2), 71–96 (2008).
Kaneko M, Eguchi K, Ohmatsu H et al. Peripheral lung cancer: screening and detection with low-dose spiral CT versus radiography. Radiology 201(3), 798–802 (1996).
Sone S, Takashima S, Li F et al. Mass screening for lung cancer with mobile spiral computed tomography scanner. Lancet 351(9111), 1242–1245 (1998).
Henschke CI, McCauley DI, Yankelevitz DF et al. Early lung cancer action project: overall design and findings from baseline screening. Lancet 354(9173), 99–105 (1999).
Diederich S, Wormanns D, Semik M et al. Screening for early lung cancer with low-dose spiral CT: prevalence in 817 asymptomatic smokers. Radiology 222(3), 773–781 (2002).
Hillman BJ. Economic, legal, and ethical rationales for the ACRIN National Lung Screening Trial of CT screening for lung cancer. Acad. Radiol. 10(3), 349–350, (2003).
McWilliams A, Mayo J, MacDonald S et al. Lung cancer screening: a different paradigm. Am. J. Respir. Crit. Care Med. 168(10), 1167–1173 (2003).
The International Early Lung Cancer Action Program Investigators. Survival of patients with stage I lung cancer detected on CT screening. N. Engl. J. Med. 355(17), 1763–1771 (2006).
Blanchon T, Brechot JM, Grenier PA et al.; for the ‘Depiscan’ Group. Baseline results of the Depiscan study: a French randomized pilot trial of lung cancer screening comparing low dose CT scan (LDCT) and chest X-ray (CXR). Lung Cancer 58(1), 50–58 (2007).
Infante M, Lutman FR, Cavuto S et al. DANTE Study Group. Lung cancer screening with spiral CT: baseline results of the randomized DANTE trial. Lung Cancer 59(3), 355–363 (2008).
Xu DM, Gietema H, de Koning H et al. Nodule management protocol of the NELSON randomized lung cancer screening trial. Lung Cancer 54(2), 177–184 (2006).
Chabner BA. Smoke, then fire: lung cancer screening studies under further scrutiny. Oncologist 13(4), 348–349 (2008).
Swensen SJ, Jett JR, Hartman TE et al. CT screening for lung cancer: five-year prospective experience. Radiology 235(1), 259–265 (2005).
Schwartz RS, Curfman GD, Morrisey S, Drazen JM. Full disclosure and the funding of biomedical research. N. Engl. J. Med. 358(17), 1850–1851 (2008).
Henschke CI, Naidich DP, Yankelevitz DF et al. Early lung cancer action project: initial findings on repeat scanning. Cancer 92(1), 153–159 (2001).
Bach PB, Jett JR, Pastorino U, Tockman MS, Swensen SJ, Begg CB. Computed tomography screening and lung cancer outcomes. JAMA 297(9), 953–961 (2007).