BACKGROUND: Dendritic cell (DC) vaccine efficacy is limited by suboptimal migration to vaccine site-draining lymph nodes (VDLNs). In mice, vaccine site conditioning with inflammatory cytokines or mature DCs increases DC trafficking and the induction of antigen-specific T cells. We assessed the impact of DC migration to VDLNs on clinical outcomes in patients with newly-diagnosed glioblastoma (GBM) by randomizing patients to one of two vaccine site conditioning strategies. METHODS: We compared effects of vaccine site conditioning with either unpulsed DCs or tetanus-diphtheria (Td) toxoid prior to a vaccine of 111Indium-labeled DCs and monitored subsequent migration to VDLNs using SPECT/CT imaging. Clinical responses following vaccine site conditioning before DC vaccination were evaluated with progression-free survival (PFS) and overall survival (OS). A preclinical model using DCs derived from fluorescent transgenic mice was employed to explore DC migration to VDLNs in the context of vaccine site conditioning. The capacity for increased DC migration to induce tumor-specific immune responses in mice was evaluated. RESULTS: In patients with GBM, migration of injected DCs to VDLNs via Td conditioning was significantly increased compared to controls (P = 0.04), and the extent of migration was strongly associated with clinical outcomes (HR = 0.845 P = 0.027 for PFS; HR = 0.820 P = 0.023 for OS; Cox proportional hazards model). Furthermore, patients randomized to Td demonstrated improved OS compared to controls (median OS 18.5 months vs >32.5 months, P = 0.013). In accordance with our clinical observations, our preclinical model showed that Td conditioning significantly increased DC migration systemically to VDLNs (P = 0.0001). Td conditioning resulted in a significant elevation in serum levels of the chemokine macrophage inflammatory protein-1 alpha (MIP-1α) (P = 0.028). In vitro chemotaxis assays revealed a selective increase in human and murine DCs migrating towards serum from Td-conditioned patients and mice compared to controls. Furthermore, this preferential migration was severely abrogated with MIP-1α blockade (P = 0.02). In mice, Td conditioning in MIP-1α knockout mice only induced baseline migration efficiency compared to controls (P = 0.003). Td conditioning in mice dramatically suppressed tumor growth and improved DC vaccines in a tumor antigen-specific manner (P = 0.0001). Moreover, Td-conditioned MIP-1α knockout mice demonstrated significantly increased tumor burden compared to that of normal mice receiving Td prior to DC vaccination (P < 0.05). CONCLUSIONS: Td conditioning increased DC migration and tumor responses in a preclinical murine model corroborating observations from our phase I/II clinical trial. Our studies demonstrate a relationship between dendritic cell trafficking and antitumor efficacy and highlight MIP-1α as an important chemokine for improving dendritic cell immunotherapy. SECONDARY CATEGORY: Clinical Neuro-Oncology.