Pulmonary Fibrosis (PF) is a chronic lung disease caused by an abnormal wound healing response to an injury that may lead to the formation of excess fibrous connective tissue, also known as scarring. This reduces lung function resulting in breathing difficulties and eventual death. PF is a common side effect of radiation therapy (RT) in the chest area; however, its staging and prognosis remains still elusive, mainly because of the highly het- erogeneous course of the disease in different patients [1]. Thus, there is a clinical need to characterise the development of radiation-induced PF and to determine appropriate radiation dosage that minimises the risk of fibrosis development after RT. Existing prog- nostic models rely on demographic characteristics and/or clinical indices and have had limited success. In this work, we present a patient-specific in silico 3D model of lung cancer regression and radiation-induced fibrosis in the organ, which incorporates individ- ualised information and enables the accurate description of PF development. The model is a system of partial differential equations that govern the evolution of cancerous cells, fibroblasts and lung parenchymal tissue. Our in silico model is applied on physiologically accurate lung geometries, generated using Computed Tomography (CT) scans. These correspond to 10 patients diagnosed with non- small cell lung cancer, and were treated at the Bank of Cyprus Oncology Centre, Cyprus. The Hounsfield data, the radiation dosage distribution and the position of the cancerous lesion are used as initial conditions for each simulation. The numerical solution exhibits the development of fibrous tissue in the lungs which follows the clinically observed patterns of fibrosis development and resembles the irregular “honeycombing” appearance seen in widespread pulmonary fibrosis. Our simulation results demonstrate that the model can realistically reproduce fibrosis progression exhibited in lung cancer patients who received specifically low-dose stereotactic RT as part of lung cancer treatment. The predicted fibrotic regions, in terms of volume and location, are in very good agreement with the fibrotic tissue observed in the CT scans of patients, which highlights the potential of our in silico approach to improve radiotherapy planning and minimise treatment impact.