Performance of low-dose CT screening for detecting lung cancer at the early stage and the estimated tumor growth rate according to the smoking status/age

Abstract

Objectives. To assess the utility of lowdose spiral chest CT (LDCT) screening for detecting lung cancer at the early stage and to examine the tumor volume doubling time (TVDT) according to the smoking status and age. Methods. A total of 295 patients who participated in an LDCT screening program for citizens of Nagano, Japan between 2000 and 2010 were examined with respect to the incidence, prognosis-related features (tumor size, proportion of clinical stage I tumors and histopathology) and TVDT of the lesions. Results. 1) The prevalence rate of lung cancer was similarly high in all smoking categories (504 per 100,000 subjects for the entire group), especially high in the elderly subjects (<50 years) and low in the patients in their 40's, seen only in non-smokers. The annual incidence was 84 in all subjects, being particularly low in the non-smokers and zero in the 40-year-old group. 2) The ratio of the prevalence cancer/annual repeat cancer was 6.0 (504/84), with 4.1 for smokers and 11.0 for non-smokers. 3) The prognosis-related features were significantly different between the smokers and non-smokers, with smokers having a larger tumor size and lower proportion of cstage I lesions. 4) The mean TVDT for all 69 analyzed lesions was 459 days, with 364 days for smokers and 606 days for non-smokers. The TVDT values were shorter in the elderly smokers but longer in the elderly nonsmokers, and varied widely in the current and nonsmokers, while remaining within narrow limits in the exand passive smokers. 5) The rate of possible overdiagnosis (TVDT <400 days) was 17% in smokers and 44% in non-smokers. Summary and Conclusions. 1) Compared with the findings of US studies, the rate of detection of lung cancer was lower in the Japanese smokers, while the prognosis-related features were similar in the two populations, although more favorable features were identified in the Japanese non-smokers. 2) The high rate of prevalence cancer, irrespective of the smoking status, with nearly 45% of patients having tumors <14 mm, stresses the importance of prevalence CT screening for both smokers and non-smokers. The prevalence of lung cancer was fairly high in the subjects <50 years of age, thus justifying the use of cost-effective screening. The detection of lung cancer in 40-year-old patients among the non-smoking subjects only requires further examinations. 3) For non-smokers, the lower incidence with more favorable prognosis-related features/TVDT of lung cancer stresses the importance of performing repeat scans at an inter-screening interval of <1 year. 4) Hence, in general, annual repeat screening for smokers and biennial screening for non-smokers appear to be appropriate for detecting the majority of lung cancers measuring <14 mm. However, based on the TVDT results obtained in this study, the detection of lung cancer measuring <14 mm is expected to fail in some proportion of 60-and 70-year-old smokers on annual repeat screening and 60-year-old non-smokers on biennial repeat screening, and it is necessary to identify specific risk factors rationally supporting the more frequent use of CT scans in these patients in order to avoid detecting cancer in the late stage. In contrast, no failure to detect lesions <14 mm is expected using triennial and quadrennial scans in 70-and 50-year-old non-smokers, respectively (although the number of 50-year-old patients in this study was limited), and triennial repeat screening appears to be appropriate for these subgroups. 5) The variety of TVDT values observed according to the smoking status/age should be taken into account when planning chest CT screening in the community and performing work-up studies to estimate the degree of tumor growth in the hospital.