Letters Section Editor: Robert M. Golub, MD, Senior Editor.
To the Editor: Dr Bach and colleagues1 statistically modeled whether earlier detection of lung cancer with low-dose helical CT and subsequent longer survival after diagnosis necessarily translates into a lower lung cancer mortality rate. According to the authors' predictive model,2 the number of expected lung cancer deaths among 3 single-group cohort studies was no lower than would have been expected without screening. The authors correctly caution prudence about widespread dissemination of CT screening until the results of large-scale randomized controlled trials—including the National Lung Screening Trial (NLST) and several European trials—are completed; the earliest results may be available in 2009. However, the authors raise concerns that the NLST may not be adequately designed to observe a reduction in lung cancer mortality, should one exist, given the number of annual screens in the NLST and the planned duration of follow-up. We would like to address these concerns.
Berg CD, Aberle DR. Computed Tomography Screening for Lung Cancer. JAMA. 2007;298(5):513-516. doi:10.1001/jama.298.5.513-b