The full content of Annals is available to subscribers

Subscribe/Learn More  >
Original Research |

Estimating Overdiagnosis in Low-Dose Computed Tomography Screening for Lung Cancer: A Cohort Study

Giulia Veronesi, MD; Patrick Maisonneuve, DipEng; Massimo Bellomi, MD, PhD; Cristiano Rampinelli, MD; Iara Durli, MD; Raffaella Bertolotti, MSc; and Lorenzo Spaggiari, MD, PhD
[+] Article and Author Information

From the European Institute of Oncology and the School of Medicine, University of Milan, Milan, Italy.

Acknowledgment: The authors thank Giovanna Ciambrone for secretarial assistance and Daniela Rampazzo for secretarial assistance and COSMOS study data entry. They also thank the medical staff of the European Institute of Oncology Divisions of Thoracic Surgery and of Radiology, as well as Giuseppe Bardo and other European Institute of Oncology radiology technicians, all of whom contributed to the COSMOS study. Finally, they thank Don Ward for help with the English translation.

Grant Support: By the Italian Association for Cancer Research.

Potential Conflicts of Interest: Disclosures can be viewed at www.acponline.org/authors/icmje/ConflictOfInterestForms.do?msNum=M12-1606.

Reproducible Research Statement: Study protocol: Available at http://clinicaltrials.gov/ct2/show/NCT01248806. Statistical code and data set: Certain portions are available from Dr. Veronesi (e-mail, giulia.veronesi@ieo.it).

Requests for Single Reprints: Giulia Veronesi, MD, Division of Thoracic Surgery, European Institute of Oncology, Via Ripamonti 435, 20141 Milan, Italy; e-mail, giulia.veronesi@ieo.it.

Current Author Addresses: Dr. Veronesi, Ms. Bertoletti, and Dr. Spaggiari: Division of Thoracic Surgery, European Institute of Oncology, Via Ripamonti 435, 20141 Milan, Italy.

Mr. Maisonneuve: Division of Epidemiology; European Institute of Oncology, Via Ramusio 1, 20141 Milan, Italy.

Drs. Bellomi, Rampinelli, and Durli: Division of Radiology, European Institute of Oncology, Via Ripamonti 435, 20141 Milan, Italy.

Author Contributions: Conception and design: G. Veronesi, M. Bellomi, C. Rampinelli.

Analysis and interpretation of the data: G. Veronesi, P. Maisonneuve, M. Bellomi, C. Rampinelli, I. Durli.

Drafting of the article: G. Veronesi, C. Rampinelli.

Critical revision of the article for important intellectual content: G. Veronesi, P. Maisonneuve, M. Bellomi, C. Rampinelli.

Final approval of the article: G. Veronesi, P. Maisonneuve, M. Bellomi, C. Rampinelli, L. Spaggiari.

Provision of study materials or patients: G. Veronesi, L. Spaggiari.

Statistical expertise: P. Maisonneuve.

Obtaining of funding: L. Spaggiari.

Administrative, technical, or logistic support: L. Spaggiari.

Collection and assembly of data: G. Veronesi, I. Durli, R. Bertoletti.

Ann Intern Med. 2012;157(11):776-784. doi:10.7326/0003-4819-157-11-201212040-00005
Text Size: A A A

Background: Lung cancer screening may detect cancer that will never become symptomatic (overdiagnosis), leading to overtreatment. Changes in size on sequential low-dose computed tomography (LDCT) screening, expressed as volume-doubling time (VDT), may help to distinguish aggressive cancer from cases that are unlikely to become symptomatic.

Objective: To assess VDT for screening-detected lung cancer as an indicator of overdiagnosis.

Design: Retrospective estimation of the VDT of cancer detected in a prospective LDCT screening cohort.

Setting: Nonrandomized, single-center screening study involving persons at high risk for lung cancer enrolled between 2004 and 2005 who received LDCT annually for 5 years.

Patients: 175 study patients diagnosed with primary lung cancer.

Measurements: VDT was measured on LDCT and classified as fast-growing (<400 days), slow-growing (between 400 and 599 days), or indolent (≥600 days).

Results: Fifty-five cases of cancer were diagnosed at baseline, and 120 were diagnosed subsequently. Of the latter group, 19 cases (15.8%) were new (not visible on previous scans) and fast-growing (median VDT, 52 days); 101 (84.2%) were progressive, including 70 (58.3%) fast-growing and 31 (25.8%) slow-growing (15.0%) or indolent (10.8%) cases. Lung cancer–specific mortality was significantly higher (9.2% per year) in patients with new compared with slow-growing or indolent (0.9% per year) cancer. Sixty percent of fast-growing progressive cancer and 45% of new cancer were stage I, for which survival was good.

Limitations: This is a retrospective study. Volume-doubling time can only indicate overdiagnosis and was estimated for new cancer from 1 measurement (a diameter of 2 mm assumed the previous year).

Conclusion: Slow-growing or indolent cancer comprised approximately 25% of incident cases, many of which may have been overdiagnosed. To limit overtreatment in these cases, minimally invasive limited resection and nonsurgical treatments should be investigated.

Primary Funding Source: Italian Association for Cancer Research.


Grahic Jump Location
Figure 1.

Distribution of incident lung cancer diagnosed over the 5 years of the study, according to VDT and preoperative CT-PET.

64% of cases of cancer had a VDT <400 d. 26% were slow-growing or indolent (VDT ≥400 d). CT-PET was visually assessed as positive or negative. “Missing” indicates that CT-PET was unavailable because it was not done or was done at another hospital. CT = computed tomography; PET = positron emission tomography; VDT = volume-doubling time.

Grahic Jump Location
Grahic Jump Location
Figure 2.

Breakdown of lung cancer–specific mortality in cases of incident cancer, according to 2 lesion classifications.

CT = computed tomography; VDT= volume-doubling time. Top. Cancer classified as new (diagnosed in follow-up but not seen on previous CT), fast-growing (arising from a previously identified nodule with a VDT <400 d), or slow-growing (arising from a previously identified nodule with a VDT ≥400 d). The single patient with slow-growing cancer who died had a history of bilateral breast cancer; death was attributed to lung cancer metastasis but was not ascertained. Results of log-rank test for new vs. fast-growing vs. slow-growing nodules (P = 0.046), new vs. fast-growing nodules (P = 0.138), new vs. slow-growing nodules (P = 0.010), and fast vs. slow-growing nodules (P = 0.132). Bottom. Cancer classified according to VDT. Log-rank test for VDT subcategories (P < 0.001). Median observation times were 4.0 y for new nodules, 3.5 y for fast-growing nodules, and 3.7 y for slow-growing nodules and 4.3 y for nodules with a VDT <50 d, 3.5 y for nodules with a VDT ≥50 and <100 d, 3.4 y for nodules with a VDT ≥100 and <200 d, and 3.7 y for nodules with a VDT ≥200 d.

Grahic Jump Location
Grahic Jump Location
Appendix Figure.

Breakdowns of lung cancer–specific mortality, according to VDT, preoperative CT-PET, and pathologic stage.

CT = computed tomography; PET = positron emission tomography; VDT = volume-doubling time. Top. “Fast-growing CT-PET positive” includes new CT-PET positive nodules. “Other” comprises slow-growing CT-PET negative (n = 22), slow-growing CT-PET positive (n = 7), and fast-growing CT-PET negative (n = 28) cancer. All categories except fast-growing CT-PET positive had excellent prognoses. Log-rank test for VDT subcategories (P < 0.001). Bottom. For stage I and stages II to IV disease, a preoperative CT-PET positive nodule was a negative predictive factor. Log-rank test for preoperative CT-PET and pathologic stage (P < 0.001). Median observation times were 3.6 y for fast-growing CT-PET positive tumors vs. 3.7 y for all other tumors; 3.6 y for stage I CT-PET negative tumors, 3.6 y for stage I CT-PET positive tumors; 4.1 y for stages II to IV CT-PET negative tumors, and 4.0 y for stages II to IV CT-PET positive tumors.

Grahic Jump Location




Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).


Submit a Comment
Estimating Over diagnosis in Low-Dose Computed Tomography Screening for Lung cancer
Posted on December 21, 2012
Robert P. Young, MD, PhD and Raewyn J. Hopkins, RN, MPH
School of Biological Sciences and Faculty of Medical and Health Sciences, University of Auckland, Auckland, New Zealand
Conflict of Interest: Financial/nonfinancial disclosures: RPY, and the funding of his research, has been supported by grants from the University of Auckland, Health Research Council of New Zealand and Synergenz BioSciences Ltd.


We agree with Veronesi et al.(1), that risk models for lung cancer should help identify smokers at greatest risk of aggressive lung cancer, thereby minimizing the investigation and treatment of indolent cancers (overdiagnosis), and improving cost-effectiveness of computed tomographic (CT) screening. Using data from the PLuSS CT screening trial (2), we show that “low risk” smokers are over represented in those with indolent lung cancer.We have previously shown that lung cancer detection rates are 4-5 fold greater in current/former smokers with spirometry-defined COPD and/or CT-based emphysema compared to those with normal lungs (3). The question arises, “Is lung function (or emphysema) also related to aggressivity of lung cancer with indolent lung cancers more prevalent in smokers with no COPD?” If true, then risk models selecting smokers for CT screening based on a “disposition to COPD” would help minimize overdiagnosis. We tested this hypothesis using data from PLuSS (2) and found that in those with and without emphysema, the frequency of cancers with “slow growth” doubling time (DT) was 48% (20/42) and 48% (10/21) respectively, while in those with and without COPD (GOLD I-IV), the frequency of cancers with “slow growth” doubling time (DT) was 38% (17/45) and 72% (13/18) respectively (P<0.01). These findings suggest that airflow limitation (COPD) is a better clinical marker of aggressive lung cancers (with lower DT) than emphysema.A related observation from the PLuSS was that 87% of the slow DT cancers were adenocarcinomas (AC) or bronchoalveolar carcinomas (BAC) compared to 60% of the “rapid” or “typical” DT cancers (P<0.001). The observation that AC/BAC account for most “excess cancers” from CT screening (ie. overdiagnosed) was also reported by Saghir et al. where 82% were AC/BAC histology (4). These results concur with studies showing COPD is more closely associated with squamous and small cell cancers than adenocarcinomas, where lung function is consistently better (5). Taken together, these findings support the hypothesis that lung cancers associated with airflow limitation (COPD) are more likely to be aggressive (rather than indolent) and less likely to result in overdiagnosis. Critically, Veronesi et al. showed that early treatment of even the fastest growing cancers had good long-term survival if identified in stage 1 (1).We conclude that overdiagnosis of lung cancers detected through CT screening can be minimized by targeting CT screening to smokers with the greatest risk, in particular those with an underlying tendency or predisposition to COPD.


1. Veronesi G, Maisonneuve P, Bellomi M, Rampinelli C, Durli I, Bertolotti R, et al. Estimating overdiagnosis in low-dose computed tomography screening for lung cancer: a cohort study. Ann Int Med 2012; 157: 776- 784.

2. Wilson DO, Ryan A, Fuhrman C, Schuchert M, Shapiro S, Siegfried JM, et al. Doubling times and CT screen-detected lung cancers in the Pittsburgh Lung Screening Study. Am J Respir Crit Care Med 2012; 185: 85-89.

3. Young RP, Hopkins RJ. Diagnosing COPD and targeted lung cancer screening. Eur Respir J 2012; 40: 1063-1064.

4. Saghir Z, Dirksen A, Ashraf H, Bach KJ, Broderson J, Clementsen PK, et al. CT screening for lung cancer brings forward early disease. The randomized Danish Lung Cancer Screening Trial: status after five annual screening rounds with low-dose CT. Thorax 2012; 67: 296-301.

5. Young RP, Hopkins RJ, Christmas T, Black PN, Metcalf P, Gamble GD. COPD prevalence is increased in lung cancer independent of age, sex, and smoking history. Eur Respir J 2009; 34: 380-386.

Author's Response
Posted on January 16, 2013
Patrick Maisonneuve, DipEng, Giulia Veronesi, MD, Raffaella Bertolotti, MSc
European Institute of Oncology
Conflict of Interest: None Declared

We thank Dr Young and Dr Hopkins for their thoughtful comments and agree that current lung cancer risk prediction models and lung cancer screening protocols should be improved to minimize the investigation and treatment of indolent cancers detected through CT screening. Young and Hopkins provided some evidence that spirometry-defined chronic obstructive pulmonary disease (COPD) could help differentiating between aggressive and slow-growing lung cancers.

To corroborate this affirmation, we retrieved preoperative spirometry parameters for 107 of the 120 cases of incident lung cancer detected during the first five years of the COSMOS study (1). For 10 patients operated in other institutions or for whom pre-operative lung function parameters were not available, we used those measured at baseline screening CT (available for 61 participants). We also evaluated whether radiological evidence of emphysema, assessed visually by the radiologist at annual screening CTs was correlated with lung cancer aggressiveness.

The frequency of lung cancers with “slow growth” doubling time was respectively 20% (15/74) and 35% (16/46) in subjects with and without radiological evidence of emphysema at CT (P=0.09). It was respectively 5% (1/20) and 32% (13/41) in subjects with baseline FEV1 (% of predicted) <80% and 80% (P=0.02), 16% (5/31) and 31% (24/78) in subjects with pre-operative FEV1<80% and 80% (P=0.15), 19% (3/16) and 27% (27/101) in subjects with and without evidence of airflow obstruction (FEV1/FVC<70%) (P=0.76)(Table 1).

Our results support the inclusion of lung function tests in lung cancer risk models, to better assess individual screening interval (2), but we believe that limiting CT screening to smokers with COPD may be too restrictive as, in our study, a significant proportion of aggressive tumors were detected among subjects with preserved lung function. Currently (3), lung cancer screening is recommended to heavy smokers aged 55 years or more who had smoked at least 30-pack-years (i.e. 20 cigarettes per day for 30 years, or equivalent), a population fundamentally predisposed to COPD.

Table to appear in future print version.


1. Veronesi G, Maisonneuve P, Bellomi M, Rampinelli C, Durli I, Bertolotti R, et al. Estimating overdiagnosis in low-dose computed tomography screening for lung cancer: a cohort study. Ann Int Med 2012;157:776-784.

2. Maisonneuve P, Bagnardi V, Bellomi M, Spaggiari L, Pelosi G, Rampinelli C, Bertolotti R, Rotmensz N, Field JK, Decensi A, Veronesi G. Lung cancer risk prediction to select smokers for screening CT - A model based on the Italian COSMOS trial. Cancer Prev Res (Phila) 2011;4(11):1778-89.

3. Jaklitsch MT, Jacobson FL, Austin JH, Field JK, Jett JR, Keshavjee S, MacMahon H, Mulshine JL, Munden RF, Salgia R, Strauss GM, Swanson SJ, Travis WD, Sugarbaker DJ. The American Association for Thoracic Surgery guidelines for lung cancer screening using low-dose computed tomography scans for lung cancer survivors and other high-risk groups. J Thorac Cardiovasc Surg 2012;144(1):33-8.

Tumor volume doubling time influences lung cancer overdiagnosis
Posted on January 22, 2013
Jerome M. Reich, MD, Jong S. Kim, PhD
Earle A Chiles Research Institute, Portland State University
Conflict of Interest: None Declared

Basing their assertion on the low, annual, cause-specific (lung cancer) death rate (0.8%) in CT-trial identified, incident lung cancers whose tumor volume doubling time (TVD) was ≥ 200-days, Veronesi et al.( ) concluded that “Slow-growing or indolent cancer comprised approximately 25% of incident cases, many of which may have been overdiagnosed.” Overdiagnosis is positively influenced by TVD because it increases exposure to competing lethal morbidities and, if sufficiently lengthy, the projected time required for the cancer to increase to a lethal size may exceed an individual’s life expectancy.

However plausible, the authors’ proposed quantification of overdiagnosis is grounded on the favorable, ≤4-year cancer survival in persons with a lengthy TVD, not on the observed course of untreated cancer. (The authors employed the standard definition of overdiagnosis: “ . . . a fraction of cases of screening-detected cancer are overdiagnosed; that is, it would not have become symptomatic in the patient’s lifetime and would not have caused death.”)

Additional variables influence overdiagnosis:1. Comorbidity. Lifetime mortality is 100%. Aside from adscititious causes (Death from a vehicular collision while exiting the screening institution following identification of an aggressive lung cancer would, for example, constitute overdiagnosis.), competing cardiopulmonary morbidities, which may be asymptomatic, are the principal causes of death in smokers. Exclusion of older or symptomatic individuals and those with previous malignancies reduces the likelihood of overdiagnosis. 2. Dimension.

There is an inverse relationship between tumor size and overdiagnosis: Diameter is a function of the number of TVDs. Assuming exponential growth ( ), tumor diameter in cm, D, = cell diameter(cube root of 2)number of TVDs; D =.001(1.26)x; log form: ( ). For example, a 1-cm tumor (1A) has undergone 30 TVDs; a 5-cm tumor (1B), 37. At a constant 1A tumor VDT of 230-days (3), 1610-days (54-months) are required for a 1-cm tumor to grow to 5-cm. The growth rate ratios of diameter and volume are, respectively, D and D2: e.g., 5-fold and 25-fold for a 5-cm vs. a 1-cm tumor. The additional 4.4-years of growth required to achieve a 5-cm size increases the likelihood of overdiagnosis by increasing exposure to lethal comorbidities.

The magnitude of overdiagnosis is best assessed by the long-term deficit in diagnosed lung cancer in the control cohort of prospective screening trials. In radiographic trials, the deficit has ranged from 22 to 24% ( ). CT trials will far exceed the radiographic figure because this method disproportionately identifies bronchioloalveolar and very small, stage IA cancers.

1.Veronesi G, Maisonneuve P, Bellomi M, Rampinelli C, Durli L, Bertolotti R, et al. Estimating Overdiagnosis in Low-Dose Computed Tomography Screening for Lung Cancer. Ann Intern Med. 2012;157:776-784.

2.Geddes DM. The natural history of lung cancer: a review based on rates of tumourgrowth. Br J Dis Chest 1979;73:1–17.

3.Reich JM, Kim JS. Lung cancer growth dynamics. Eur J Radiol. 2011;80(3):e458-461. doi: 10.1016/j.ejrad.2010.08.0064.Reich JM. A critical appraisal of overdiagnosis: estimates of its magnitude and implications for lung cancer screening (Review). Thorax 2008;63:377-83

Submit a Comment

Summary for Patients

Clinical Slide Sets

Terms of Use

The In the Clinic® slide sets are owned and copyrighted by the American College of Physicians (ACP). All text, graphics, trademarks, and other intellectual property incorporated into the slide sets remain the sole and exclusive property of the ACP. The slide sets may be used only by the person who downloads or purchases them and only for the purpose of presenting them during not-for-profit educational activities. Users may incorporate the entire slide set or selected individual slides into their own teaching presentations but may not alter the content of the slides in any way or remove the ACP copyright notice. Users may make print copies for use as hand-outs for the audience the user is personally addressing but may not otherwise reproduce or distribute the slides by any means or media, including but not limited to sending them as e-mail attachments, posting them on Internet or Intranet sites, publishing them in meeting proceedings, or making them available for sale or distribution in any unauthorized form, without the express written permission of the ACP. Unauthorized use of the In the Clinic slide sets will constitute copyright infringement.


Buy Now

to gain full access to the content and tools.

Want to Subscribe?

Learn more about subscription options

Related Articles
Related Point of Care
Topic Collections
PubMed Articles

Buy Now

to gain full access to the content and tools.

Want to Subscribe?

Learn more about subscription options

Forgot your password?
Enter your username and email address. We'll send you a reminder to the email address on record.