“Results of 10 years of annual LDCT in our high risk cohort show that the lung cancer detection rate remained stable over this long period and, importantly, that the proportion of early stage cancers also remained stable, implying that screening should continue for beyond ten years,” the researchers advise.
They say that, while there is ample evidence that low-dose CT screening reduces lung cancer mortality, it is not clear how long screening should continue for high-risk individuals.
To investigate, Dr. Giulia Veronesi, at the European Institute of Oncology in Milan, and colleagues assessed outcomes in a cohort of 1035 volunteers older than 50 and with a smoking history of more than 20 pack-years who were offered annual low-dose CT screening for 10 years. Compliance was good with 65% of participants completing the tenth screening round.
As reported in Lung Cancer online September 7, screening detected a total of 71 lung cancers; 23 were detected during the first two screening rounds; thereafter, the cumulative incidence was almost linear with 4-8 cancers detected at each annual screening.
The team found the proportion of stage I cancers diagnosed at each year varied, but the overall rate of 78% “remained stable over the ten years.” Average tumor size was 12.5 mm and complete resection was achieved in 58 (80.5%) cases.
Ten patients (12.1%) underwent surgery for what proved to be a benign lesion.
Overall survival among the lung cancer cases was 64% at 5 years and 57% at 10 years. Corresponding rates for those with stage I disease were 84% and 65%, Dr. Veronesi and colleagues report.
Their study, they conclude, “indicates that screening should be continued for over five years (either annually or every two-three years), and undermines the hypothesis that overdiagnosis is a major problem with lung cancer screening. If overdiagnosis were a major phenomenon, the proportion of stage I cancers would be expected to decline after the first couple of screening rounds.”
“Nevertheless,” they acknowledge, “a reliable estimate of the overdiagnosis rate can only come from randomized trials.”
The team also used their data to assess the accuracy of two lung cancer risk prediction models. “The number of lung cancers detected during the first two screening rounds was close to that predicted by the COSMOS model, while the Bach model accurately predicted frequency from the third year on,” they report.