<p>Though lung cancers due to tobacco consumption or smoking corresponds to 25% of all cases of lung cancer throughout the world, there are several causes which lead to lung cancer in non-smokers (LCNS) as well. Women are more susceptible to LCNS with a particularly high incidence in south-east Asia. Adenocarcinoma is more common in non-smokers. Exposure to environmental tobacco smoke, cooking fumes, hormone therapy, dietary factors, asbestos, occupational exposure to smoke are some of the leading causes of LCNS. Polymorphism of genes encoding pro and anti-inflammatory molecules induces lung cancer. The mutation frequency of epidermal growth factor receptor (EGFR), KRAS and p53 varies in cases of lung cancer among smokers and non-smokers. Epigenetic alterations including differential methylation patterns, and variations in immunogenic responses further distinguish LCNS from tobacco associated lung cancer. Recent advances in targeted therapies including crizotinib, osimertinib, gefitinib as well as erlotinib exert therapeutic effects in LCNS and have improved the clinical outcomes. In parallel, emerging computational approaches including mathematical modelling and artificial intelligence and machine learning are being increasingly used to predict disease progression. This review summarizes available reports on the epidemiology, risk factors, molecular genetics, epigenetics, immunological aspects along with probable therapeutic approaches in LCNS, while also highlighting the growing role of computational and data driven frameworks in improving disease understanding.</p>