Elsevier

Lung Cancer

Volume 67, Issue 2, February 2010, Pages 177-183
Lung Cancer

Lung cancer screening using low-dose computed tomography in at-risk individuals: The Toronto experience

https://doi.org/10.1016/j.lungcan.2009.03.030Get rights and content

Abstract

Objective

The Department of Medical Imaging at the University Health Network in Toronto is performing a lung cancer screening study, utilizing low-dose computed tomography (LDCT) as the modality. Baseline and annual repeat results are reported on the first 3352 participants, enrolled between June 2003 and May 2007.

Methods

Enrollment was limited to those aged 50 years or older, with a smoking history of at least 10 pack-years, no previous cancer and general good health. A helical low-dose CT (LDCT) of the chest was performed using 120 kVp, 40–60 mA, images were reconstructed with 1–1.25 mm overlapping slices. The primary objectives were the detection of parenchymal nodules and diagnosis of early stage lung cancer. Baseline LDCTs were termed positive if at least one indeterminate non-calcified nodule 5 mm or larger in size, or non-solid nodule 8 mm or larger in size was identified. Follow up periods for individuals with a positive baseline LDCT were determined by nodule characteristics.

Results

The median age at baseline was 60 years (range 50–83), with a median of 30 pack-years of cigarette smoking (range 10–189). Baseline CT evaluations were positive in 600 (18%) participants. To date, 2686 (80%) of the participants have returned for at least one annual repeat screening LDCT. Biopsies have been recommended for 82 participants since the study began, and 64 have been diagnosed with screen-detected cancer (62 lung, two plasmacytoma of the rib). A total of 65 lung cancers have been diagnosed (62 screen-detected, 3 interim), 57 are NSCLC (82% with known stage are stage I or II) and the rate of surgical resection was 80%. Sensitivity and specificity of the protocol in successfully diagnosing early stage lung cancers were 87.7% and 99.3%, respectively.

Conclusions

Data indicate that LDCT can identify small lung cancers in an at-risk population. The diagnostic algorithm results in few false-positive invasive procedures. Most cancers are detected at an early stage, where the cancer is resectable with a greater potential for cure. Long-term follow up of lung cancer cases will be carried out to determine survival.

Introduction

In Canada, lung cancer is expected to be diagnosed in 23,900 people in 2008 and cause 20,200 deaths, ranking it first in cancer mortality for both men and women [1]. Similar figures are reported in other western countries [2], [3]. A key problem is that most tumors are diagnosed at advanced stages, when potential intervention has little effect on survival and mortality [3]. There is considerable interest in shifting detection to stages at which intervention can be curative, and lung cancer screening using low-dose computed tomography (LDCT) is under evaluation in research centers worldwide [4], [5], [6], [7], [8], [9], [10].

Early LDCT data originated in Japan and was largely based on mass screening [8], [9]. The Early Lung Cancer Action Program (ELCAP) was established in the United States in 1992, and focused on subjects considered at-risk for developing lung cancer, based on age and smoking history [6]. ELCAP was subsequently expanded to additional centers and countries to become the International Early Lung Cancer Action Program (I-ELCAP). Several randomized trials, in the United States (National Lung Screening Trial, NLST) [5] and Europe (NELSON [10] and ITALUNG [7]), were established in recent years; first reports are expected within the next few years.

The Department of Medical Imaging at the University Health Network (UHN) in Toronto has been conducting a lung cancer screening study since 2003, using a single-arm study design to screen at-risk individuals. It is currently the only Canadian center in the I-ELCAP [6]. The protocol identifies LDCT scans as positive based on the characteristics of detected parenchymal nodules. Accordingly, participants with negative results are asked to return for annual repeat scans, while those with positive results are designated for more immediate follow up.

The Toronto study is currently one of the larger prospective screening studies. Following the publication of the baseline results involving the first 1000 Toronto participants [11], this paper reports on the baseline and annual repeat screening evaluations of 3352 patients, who had baseline scans performed between September 2003 and May 2007.

Section snippets

Screened population

Since September 2003, the UHN Department of Medical Imaging in Toronto has been conducting a lung cancer screening study. While independently funded, the study is the only Canadian site sharing data with the multi-institutional I-ELCAP. Local eligibility criteria were: aged 50 years or older, at least 10 pack-years history of smoking, no prior history of cancer (exception of non-melanotic skin cancer) and general good health. Participation in the study was voluntary, with sources including

Study population

By May 2007, 3352 participants were enrolled in the study (Table 1). The study population had a median age at baseline of 60 years (range 50–83), and a median of 30 pack-years of smoking (range 10–189). More than half (54%) of the participants were female.

Baseline CT evaluations

Using I-ELCAP definitions described above, baseline CT evaluations were positive in 600 (18%) participants (Fig. 1). Follow up LDCT for these individuals were recommended to take place 1 month (n = 44, 7%), 3 months (n = 521, 87%), or 6 months (n = 

Discussion

The results demonstrate that baseline and annual repeat lung cancer screening, combined with a defined algorithm for nodule surveillance, results in a high overall detection rate of lung cancers (1.9%). The lung cancer detection rate is in agreement with other studies, ranging between 0.4% [8] and 2.7% [6] and depends on the enrollment criteria. Similar reports in the literature [14], have noted that most tumors are prevalence cancers arising from nodules detected at baseline (87%), less (7%)

Conclusion

In conclusion, the lung cancer screening program at the University Health Network in Toronto has successfully recruited a large number of participants. The program involves a diagnostic regimen that guides radiologists and surgeons in determining the follow up procedures to be used for each screening participant. This results in a minimization of invasive procedures and unnecessary follow up. Preliminary results are positive: a high proportion of early stage diagnoses, high rate of resection in

Conflict of interest

None declared.

Acknowledgements

The establishment of the lung cancer screening program at the University Health Network in Toronto was enabled by a generous donation to the Princess Margaret Foundation from the family of Lusi Wong. We thank the staff, fellows and CT technologists in the chest section of the Department of Medical Imaging.

References (23)

  • G. Picozzi et al.

    Screening of lung cancer with low dose spiral CT: results of a three year pilot study and design of the randomised controlled trial “Italung-CT”

    Radiol Med (Torino)

    (2005)
  • Cited by (0)

    View full text