The very poor prognosis of lung cancer has barely changed in the last two decades despite all efforts. However, prognosis is better when the disease is detected earlier, so that curative surgery or radiotherapy can be applied. Lung cancer screening in the past by chest X-ray did not lead to a decrease in lung cancer mortality, because the chest X-ray has low sensitivity for early invasive stages. With the advent of the low-dose spiral CT scan it has become feasible to detect early invasive stage I lung cancer in 80-90%. Modern screening for lung cancer by spiral CT scan could possibly decrease lung cancer mortality. Despite the first favourable results of screening the question remains whether lung cancer screening will be cost-effective. These questions can only be resolved in a randomised controlled trial with lung cancer mortality as unbiased end-point. Such a study should be initiated in the Netherlands, a country with large experience in screening trials and a good health care system. Only after lung cancer screening has proven to be cost-effective can appropriate implementation be recommended to prevent uncontrolled and opportunistic diffusion of this new screening technique into clinical practice in the near future.