Objectives <p>Mindfulness meditation (MM) has been introduced as a promising alternative in Internet gaming disorder (IGD). However, previous intervention studies have primarily relied on node-centric approaches, neglecting critical edge-level interactions that may capture essential features of intervention-induced dynamic brain changes.</p> Method <p>We conducted a pre-registered, single-blind randomized controlled trial. Ultimately, 66 participants completed the 1-month training (MM group, <i>n</i> = 31; progressive muscle relaxation [PMR] group, <i>n</i> = 35), comprising baseline assessment, eight training sessions, and post-intervention assessment. We quantified nodal entropy from edge-centric functional connectivity as a metric of dynamic complexity and used Allen Human Brain Atlas (AHBA)–based imaging transcriptomics to explore transcriptomic correlates of entropy changes.</p> Results <p>MM demonstrated superior efficacy over PMR in reducing addiction severity and craving. At the level of network dynamics, MM specifically increased normalized entropy within the limbic network, signifying a restoration of dynamic flexibility. This enhancement in limbic entropy significantly moderated the therapeutic effect on craving attenuation. Furthermore, these macroscale dynamic changes were spatially correlated with transcriptomic profiles enriched in pathways related to RNA metabolism and sensory transport, providing exploratory transcriptomic context for the observed entropy changes.</p> Conclusions <p>MM alleviates IGD by enhancing the dynamic complexity of the limbic network, effectively countering neural rigidity. By bridging behavioral outcomes, edge-centric dynamics, and gene expression, this study demonstrates that nodal entropy is a promising dynamical biomarker for monitoring therapeutic response in behavioral addiction.</p> Preregistration <p>ChiCTR2300075869, registered on September 18, 2023.</p>