Abstract: To address the inadequate decision reliability of large language models in oncology multidisciplinary team (MDT) consultations caused by the lack of collaboration mechanisms and evidence-based quality control, a multi-agent framework named OncMDT with dynamic orchestration and reflection coordination is proposed. The framework dynamically instantiates a subset of specialist agents based on patient clinical features and constructs a multi-stage collaborative reasoning workflow mapping the actual clinical agenda. An iterative retrieval strategy with planning and verification feedback is introduced to achieve role-aware guideline evidence addressing. Before consensus generation, a reflection and coordination mechanism is designed to structurally review and correct inter-disciplinary conflicts, evidence deviations, and missing intervention dimensions, combined with a case memory bank utilizing dual-layer manifold retrieval for empirical reference. Evaluated on a benchmark dataset comprising 120 cases with varying complexity across four cancer types, OncMDT achieves a decision consistency rate of 85.1%, a completeness rate of 87.5%, and a citation accuracy of 82.1%, consistently outperforming baseline methods with statistically significant margins (paired bootstrap test, p＜0.05,where p denotes the p-value). The above results preliminarily indicate that this framework can effectively simulate multidisciplinary collaborative logic on the synthetic virtual case evaluation set, offering a promising methodological approach for enhancing the evidence-based rigor of complex oncological decision-making.