Journal of Thoracic Oncology:
Clinicopathological Characteristics of Screen-Detected Lung Cancers
Kawachi, Riken MD; Watanabe, Shun-ichi MD; Asamura, Hisao MD
Division of Thoracic Surgery, National Cancer Center Hospital, Tokyo, Japan.
Clinicopathological Characteristics of Screen-Detected Lung Cancers: Erratum
When this article was published in the May 2009 issue of the Journal of Thoracic Oncology, several errors appeared. These errors appeared in the Abstract as well as the text. The correct data was published in the Erratum published in the August 1009 issue.
Disclosure: The authors declare no conflicts of interest.
Address correspondence to: Dr. Hisao Asamura, Division of Thoracic Surgery, National Cancer Center Hospital, 5-1-1 Tsukiji, Chuo-ku, Tokyo 104-0045, Japan. E-mail: email@example.com
Background: The efficacy of screening for lung cancers remains controversial, and none of the guidelines for lung cancer detection recommend screening for lung cancers. The purpose of the present study was to retrospectively analyze and characterize the clinicopathological features of screen-detected (SCR) lung cancer in comparison with lung cancers detected by other means.
Patients: The records of 2281 patients who underwent lung resection for primary lung cancer between 2000 and 2006 were analyzed retrospectively. Patients were classified into three groups according to the method of detection: SCR (n = 1290), symptom-detected (SYM, n = 481), and incidental (INC, n = 568). In the SCR group, clinicopathological factors were analyzed according to the detection modality: chest x-ray (n = 1136, 82.6%), computed tomography (CT, n = 196, 13.9%), positron emission tomography (n = 22, 1.6%), and sputum cytology (n = 17, 1.3%).
Results: The percentages of smaller (≤2 cm) lung cancer (42.6%: SCR, 19.6%: SYM, 40.9%: INC), adenocarcinoma (85.8%: SCR, 58.6%: SYM, 73.1%: INC), and pathologic stage I (73.0%: SCR, 47.0%: SYM, 71.2%: INC) were higher in the SCR group than in the other two groups. The 5-year survival rates in SCR, SYM, and INC group were 79.6%, 74.6%, and 64.6%, respectively. The patients with CT-detected lung cancer had a higher incidence of smaller size (≤2 cm, 76.4%), adenocarcinoma (92.6%), and stage I (clinical: 97.2%, pathologic: 93.1%). The 5-year survival rates in the chest x-ray and CT groups were 77.8% and 91.2%, respectively.
Conclusions: SCR lung cancers were characteristically less advanced, had a smaller diameter, and were more frequently adenocarcinoma histologically. CT-screening may be able to detect early stage lung cancers, and improve the prognosis of lung cancer patients.
Lung cancer is the most common cause of cancer death worldwide not only in Japan but also in the other developed countries. In 2005, 45,189 males and 16,874 females died of lung cancer in Japan.1 Early detection and surgical resection could provide the best chance for cure of lung cancers. However, previous trials using chest x-ray (CXR) and sputum cytology (SC) in heavy smokers failed to show a reduction in mortality.2–4 Recently, several studies have shown that lung cancer can be detected in a much earlier stage.5–12 These are the most promising recent measures for early detection using computed tomography (CT).
The objective of the present study was to identify the characteristics of lung cancer detected by screening, and to clarify whether the screen-detected (SCR) group shows better survival than other groups. The objective of the study was to compare screen detected cancers to incidental (INC) or symptomatic cancers and to evaluate survival in these groups.
PATIENTS AND METHODS
From January 2000 to December 2006, 2281 patients underwent surgical resection for primary lung cancer at the National Cancer Center Hospital in Tokyo, Japan. Medical records of all patients were reviewed retrospectively. Preoperative staging routinely included CXR and chest and abdominal CT. Positron emission tomography (PET), bone scan, and brain magnetic resonance imaging were performed only when further examination was required. All patients were staged clinically and pathologically according to the International Union Against Cancer tumor node metastasis classification system.13 The histology of the tumor was described according to the World Health Organization classification.14 The present study focused on patients with non-small cell carcinoma (adenocarcinoma, squamous cell carcinoma, large cell carcinoma, and adenosquamous carcinoma).
Grouping by Method of Detection
The method of detection was categorized as SCR (n = 1279, 56.1%), symptom-detected (SYM, n = 466, 20.4%), or INC (n = 536, 23.5%). The patient characteristics are shown in Table 1. In the SCR group, clinicopathological factors were further analyzed according to the detection modality: CXR in 1047 (81.9%), CT in 176 (13.8%), PET in 20 (1.6%), and SC in 17 (1.3%). The characteristics according to the detection modality are shown in Table 2. The modality was defined as the primary method used to detect the abnormality. The type of screening was roughly divided into three groups; screening sponsored by local government, screening held by company, and screening at patients’ expense. The last screening consists of members who pay dues and are entitled to screening. Furthermore, members can choose the modality of screening depending on the price. Thus, many types of screening were enrolled in this study. The SYM group was defined as patients who complained of the kind of respiratory symptom, and the incidentally detected group was defined as patients who were detected during screening for other diseases.
Survival was calculated using the Kaplan-Meier method and differences in survival were determined by log-rank analysis. The median follow-up time for patients was 35.1 months. p values lower than 0.05 were considered statistically significant.
Characteristics of Screen-Detected Lung Cancer
Screen-detected lung cancers were smaller in diameter (<2 cm: 42.6%), less advanced (p-stage I: 70.8%), and showed a higher incidence of adenocarcinoma (85.8%). Incidentally detected lung cancers showed a similar tendency to SCR lung cancers, but SYM lung cancers were larger diameter, more advanced. Several characteristic findings were observed in CT-detected lung cancers: smaller diameter (<2 cm: 76.4%), less advanced (clinical stage I: 97.2%, pathologic stage I: 93.1%), and more frequently adenocarcinoma histologically (92.6%).
Survival According to the Method of Detection
The overall 5-year survival rate for the 2281 patients was 75.4%. The 5-year survival rates for the SCR, SYM, and INC groups were 79.6%, 74.6%, and 64.6%, respectively. The differences between the three groups were statistically significant (SCR versus SYM: p < 0.0001, SCR versus INC: p = 0.0377). The survival curves according to the method of detection are shown in Figure 1. Of the 2281 total patients, 1486 had pathologic stage I non-small cell lung cancer. In this subgroup, the 5-year survival rates overall and in the SCR, SYM, and INC groups were 89.6%, 92.9%, 84.0%, and 84.6%, respectively (Figure 2). The 30-day mortality was 3 patients in SCR, 5 patients in SYM, and no patients in INC group.
Survival of Screen-Detected Lung Cancer According to Modality
The 5-year survival rates for the CXR, CT, PET, and SC were 77.8%, 91.2%, 90.9%, and 80.9%, respectively. The difference in survival between the detection modalities was significant (p = 0.0127). Moreover, 896 patients had pathologic stage I non-small cell lung cancer, and the overall 5-year survival rates for the CXR and CT were 81.4% and 91.7%, respectively (p < 0.0001) (Figure 3).
Adenocarcinoma Equal to or Smaller than 2 cm in Diameter
Of 1762 adenocarcinomas, 733 had a maximal diameter of less than 2 cm. Of these, 477 were in the SCR group. Bronchioloalveolar carcinoma (BAC) was observed in 76 patients (6.1%) and invasive adenocarcinoma in 392 patients (31.5%). The distribution of the types is shown in Table 3. No patients with adenocarcinomas were observed in the SC group, and no patients with BACs were detected by PET. The proportion of BAC in the CT group (22.2%) was much higher than that in CXR (3.5%).
Type of Disease According to Smoking in Screen-Detected Lung Cancer
The relationship between smoking and type of disease is shown in Table 4. In the never-smoking-group, the incidence of noninvasive carcinoma such as bronchioloalveolar carcinoma was higher than that in patients with smoking history. As for advanced diseases, the incidence was more frequent in current or previous smokers.
The present study provides the latest data on screenings for lung cancers in patients who underwent surgical resection. SCR lung cancers were not only smaller (2 cm or less in diameter: 42.6%) and at a lower stage (stage I: 73.0%), but also more often adenocarcinoma (85.8%) than SYM lung cancers. In particular, such findings were more evident with CT in the SCR group (2 cm or less in diameter: 76.4%, pathologic stage I: 93.1%, and adenocarcinoma: 92.6%). The characteristics of SCR lung cancers in other reports were similar to those in the present study. Sobue et al.7 reported that 82% of patients with lung cancers had stage I lung cancer. The International Early Lung Cancer Action Project (I-ELCAP) 12 also reported that the incidence of stage I lung cancer was 85.1% (412 of 484). In another report on CT screening for patients with a smoking history by Swensen et al.,8 the incidence of clinical stage I non-small cell lung cancer was 66.7% (24 of 36) and the percentage of patients with cancer smaller than 2 cm was 91.6% (33 of 36). The high proportion of stage I in the present study is consistent with the results of several recent studies.
Previous studies have included high percentages of patients who were current or ever smokers in lung cancer screening. The present study was not limited to such patients, and the patients had a different background. The percentage of patients with a smoking history in the SCR group was only 55%, which was lower than the rates in the other groups. In the SCR group, the incidence of squamous cell carcinoma histology was 17.7% (126 of 711) in ever or current smokers, and this value was significantly higher than that in never smokers. In the report by Swensen et al.,8 the percentage of squamous cell carcinoma was 13.8% (4 of 29), which is similar to the result in the present study despite the presence of patients with a smoking history. However, the incidence of adenocarcinoma histology was 97.9% (552 of 564) in never smokers, and screening for all histologic types including adenocarcinoma must not be limited to smokers.
Another characteristic feature of the present study was a high incidence of adenocarcinoma in the SCR group (85.8%), particularly in the CT-detected subgroup (92.6%). Moreover, the incidence of BAC (≤2 cm) in CT-detected lung cancers was 22.2%, which was significantly higher than that in the CXR group (3.5%). In the I-ELCAP report,12 BAC accounted for 7.1% of adenocarcinoma, which was a much lower incidence than that in the present study, although the present study was focused on adenocarcinoma smaller than 2cm in diameter. Lindell et al.10 reported similar results, in that the incidence of BAC was 18.8% (9 of 48) by CT-screening. Screening by CT may detect very early lung cancers which could not be detected by CXR or another modality, although these results might reflect a length bias or lead-time bias.
The 5-year survival rate among 1290 patients in the SCR group was 79.6%, and this value was greater than that in patients in SYM and incidentally detected groups. In the SCR group, the 5-year survival rate of 111 patients in the CT subgroup was 91.2%, which was greater than the survival in patients in the conventional CXR and SC groups. Moreover, among these, the 5-year survival rate for patients with pathologic stage I was 91.7%. In previous reports, the 5-year survival rate of radically resected stage I non-small cell lung cancer has ranged from 60 to 80%.15–19 These reports included CXR screening and the 5-year survival rate was similar to our value of 77.8%. Sobue et al.7 reported a 5-year survival rate of 100% in their series of 29 patients who underwent resection after pathologic stage I lung cancer was detected on CT. Kaneko et al.11 showed that detection by CT permitted a 10-fold increase in lung carcinoma detection. A recent report by I-ELCAP12 demonstrated that the 10-year survival rate of resected clinical stage I lung cancer was 92%, and concluded that annual CT screening could detect lung cancer that was curable. According to these results, screening with low-dose helical CT may be able to improve the efficacy of screening in terms of reducing lung cancer mortality.
Smoking is one of the well-known risk factors for lung cancer, and smoking was related to the increasing prevalence in such natural-history model that carcinoma has been assumed to progress from a few cells to advanced-stage disease. On the contrary, inconsistent findings in other previous studies were also observed in the present study; the incidence of noninvasive cancer was lower in the smoking group, although advanced cancers were observed more frequently in smokers.
The limitations of the present study should also be addressed. First, the present study was retrospective. Detection was performed in various screenings and the screening modality was heterogeneous. Many subjects were enrolled in this study. Furthermore, the number of the people who underwent screening for lung cancer and the number of the lung cancers which were detected by lung cancer screening were not known. Moreover, the patients were limited to those who underwent surgical resection. The present study showed that screening, and particularly CT-screening, could detect lung cancers at an earlier stage and offer better survival. However, this could reflect biases of various types, such as lead-time bias and length bias, and screening did not directly reduce the mortality due to lung cancer. For a direct demonstration of the effectiveness of lung cancer screening, we must wait for the results of ongoing randomized control studies, such as the Nederlands Leuvens Screening Onderzoek trial in Europe and the National Lung Screening Trial in North America. In summary, screening for lung cancer detects early stage lung cancer. Furthermore, CT screening may detect lung cancer at an early stage that would not be detected by screening with CXR. CT screening may be effective for the detection of curable lung cancer.
1. Statistics and Information Department, Minister’s Secretariat, Ministry of Health, Labour and Welfare. Statistical Abstracts on Health and Welfare in Japan, 2006.
2. Fontana RS, Sanderson DR, Woolner LB, et al. Lung cancer screening. The Mayo Program. J Occup Med
3. Marcus P, Bergstralh E, Fagerstrom R, et al. Lung cancer mortality in Mayo lung project: impact of extended follow up. J Natl Cancer Inst
4. Bach P, Kelly M, Tate R, McCrory DC. Screening for lung cancer: a review of the literature. Chest
5. Henschke C, McCauley D, Yankelevitz D, et al. Early lung cancer detection project: overall design and findings from baseline screening. Lancet
6. Sone S, Takashima S, Li F, et al. Mass screening for lung cancer with mobile spiral computed tomography scan. Lancet
7. Sobue T, Moriyama N, Kaneko M, et al. Screening for lung cancer with low-dose helical computed tomography: anti-lung cancer association project. J Clin Oncol
8. Swensen S, Jett J, Sloan J, et al. Screening for lung cancer with low-dose spiral computed tomography. Am J Respir Crit Care Med
9. Swensen SJ, Jett JR, Hartman TE, et al. CT screening for lung cancer: five-year prospective experience. Radiology
10. Lindell RM, Hartman TE, Swensen SJ, et al. Five-year lung cancer screening experience: CT appearance, growth rate, location, and histologic features of 61 lung cancers. Radiology
11. Kaneko M, Kusumoto M, Kobayashi T, et al. Computed tomography screening for lung carcinoma in Japan. Cancer
12. International Early Lung Cancer Action Program Investigators. Survival of patients with stage I lung cancer detected on CT screening. N Engl J Med
13. International Union Against Cancer. Lung tumours. In Sobin LH, Wittekind CH, (Eds.), TNM Classification of Malignant Tumours, 6th Ed. New York: Wiley-Liss, 2002.
14. Travis WD, Colby TV, Corrin B, et al (Eds.) Histological typing of lung and pleural tumors, World Health Organization International Histological Classification of Tumors. Berlin: Springer; 1999.
15. Naruke T, Tsuchiya R, Kondo H, Asamura H, Nakayama H. Implications of stating in lung cancer. Chest
16. Williams DE, Pairolero PC, Davis CS, et al. Survival of patient surgically treated for stage I lung cancer. J Thorac Cardiovasc Surg
17. Wada H, Tanaka F, Yanagihara K, et al. Time trends and survival after operations for primary lung cancer from 1976 through 1990. J Thorac Cardiovasc Surg
18. Pairolero PC, Williams DE, Bergstralh EJ, Piehler JM, Bernatz PE, Payne WS. Postsurgical stage I bronchogenic carcinoma: morbid implications of recurrent disease. Ann Thorac Surg
19. Asamura H, Goya T, Koshiishi Y, et al. How should the TNM staging for lung cancer be revised? A simulation based on the Japanese lung cancer registry populations. J Thorac and Cardiovasc Surg
Computed tomography (CT scan); Imaging (all modalities); Lung cancer; Diagnosis and staging; Positron emission tomography (PET)
© 2009International Association for the Study of Lung Cancer
Highlight selected keywords in the article text.