Evaluating the early therapeutic response of radiofrequency ablation in rabbit VX2 lung tumors: a comparison of four diffusion weighted imaging models.

Background: Radiofrequency ablation (RFA) is a widely used treatment for unresectable lung tumors; however, conventional anatomical imaging has limited ability to assess early treatment response, while functional parameters derived from diffusion-weighted imaging (DWI) may offer improved sensitivity. This study aimed to evaluate and compare the diagnostic performance of four DWI models for the early identification of residual or progressive tumors after RFA in rabbit VX2 lung tumors. Methods: Fifty-four VX2 tumor-bearing rabbits underwent multi-modal DWI [mono-exponential DWI, intravoxel incoherent motion (IVIM), stretched exponential model (SEM), diffusion kurtosis imaging (DKI)] before and after RFA. Follow-up was conducted at days 1 (n=16), 3 (n=13), 7 (n=14), and 14 (n=11). Two blinded radiologists independently measured parameters, and inter-observer agreement was assessed using the coefficient of variation (CV). Statistical comparisons between groups and across time points were performed using appropriate tests ( Results: Parameters from the mono-exponential, SEM, and DKI models showed excellent inter-observer reproducibility (CVs: 2.18-8.65%), while IVIM parameters showed moderate agreement (CVs: 12.12-18.62%). Post-RFA, significant increases were observed in apparent diffusion coefficient (ADC1: b=0, 800 s/mm Conclusions: Multi-modal DWI effectively assesses RFA efficacy. Specifically, ∆Dk emerged as the most robust marker, demonstrating excellent diagnostic performance as early as day 1, while ∆ADC and ∆K provided complementary predictive value.