Le dépistage du futur du cancer broncho-pulmonaire : perspectives et mythes

Lung cancer is a bona fide candidate for screening, since the definition of a targeted highrisk population is easy (over 50-55 years age smokers), and since prognosis of resected early stage lung cancer patients remains good (88 % overall survival at 10 years). The NLST randomized trial on more than 50,000 high-risk subjects has proved the major interest of low-dose CT scan for such a screening, since it is actually the only screening tool, all cancer types comprised, to have demonstrated a 20 % decrease in specific cancer mortality and a 6 % decrease in overall mortality. However, low-dose CT scan induced a high rate of false-positive nodules, which often needed invasive diagnostic procedures to prove their benignity. Various biological assays have been proposed as alternative tools or complement to low-dose CT-scan screening. Studies based on morphometric cytology analysis on exfoliated cells, or on nucleic acids (DNA, micro-RNA) extracted from those cells, are limited by the difficulty to obtain a representative expectoration from asymptomatic subjects. Studies of exhaled condensates showed such materials were not correlated with biological abnormalities of bronchial epithelium. Circulating blood biomarkers are easier to obtain and study. Proteomic studies helped to define new diagnostic circulating peptides, which could be measured using a more simple and costeffective ELISA technology in future larger prospective trials. Micro-RNA signatures have proved spectacular predictive values, especially in Italian screening trials, but will also need new prospective trials, in which they could be used as a complement to low-dose CT-scan, within a decisional algorithm. Such algorithm, in front of a suspect lung nodule, would help with accuracy to lower the false-positive rate and the percentage of unneeded invasive and potentially hazardous biopsies for benign incidental nodules.