Continuous-time random walk and amide proton transfer-weighted imaging to predict lymphovascular space invasion in non-small cell lung cancer

Motivation: Lymphovascular space invasion (LVSI) is associated with poor outcomes in non-small cell lung cancer (NSCLC). Goal(s): To investigate whether 18F-FDG PET, a continuous-time random walk (CTRW) model, amide proton transfer-weighted imaging (APTWI)-based parameters may contribute to the non-invasive assessment of LVSI in NSCLC. Approach: A total of 78 LVSI-positive and 52 LVSI-negative NSCLC patients underwent chest multiparametric 18F-FDG PET/MRI. Results: The SUVmax and MTRasym(3.5ppm) were higher, while the ADC, α, β, and Dm were lower in the LVSI-positive group than in the LVSI-negative group. SUVmax, α, β, Dm, and MTRasym(3.5ppm) were independently associated with LVSI positivity. Impact: The analysis of tumour glucose metabolism, water molecule diffusion, temporal / spatial heterogeneities, and mobile proteins / peptides measurement at multiparametric 18F-FDG PET/MRI enables the prediction of LVSI in NSCLC.