Lung Cancer Screening, towards a Multidimensional Approach: Why and How?

• Main Authors: Jonathan Benzaquen, Jacques Boutros, Charles Marquette, Hervé Delingette, Paul Hofman
• Format: Article
• Language: English
• Published: MDPI AG 2019-02-01
• Series: Cancers
• Subjects: lung cancer; artificial intelligence; screening
• Online Access: https://www.mdpi.com/2072-6694/11/2/212

Early-stage treatment improves prognosis of lung cancer and two large randomized controlled trials have shown that early detection with low-dose computed tomography (LDCT) reduces mortality. Despite this, lung cancer screening (LCS) remains challenging. In the context of a global shortage of radiologists, the high rate of false-positive LDCT results in overloading of existing lung cancer clinics and multidisciplinary teams. Thus, to provide patients with earlier access to life-saving surgical interventions, there is an urgent need to improve LDCT-based LCS and especially to reduce the false-positive rate that plagues the current detection technology. In this context, LCS can be improved in three ways: (1) by refining selection criteria (risk factor assessment), (2) by using Computer Aided Diagnosis (CAD) to make it easier to interpret chest CTs, and (3) by using biological blood signatures for early cancer detection, to both spot the optimal target population and help classify lung nodules. These three main ways of improving LCS are discussed in this review.