Abstract
Abstract INTRODUCTION: The ultimate success of dendritic cell (DC) vaccination for the active immunotherapy of neoplasia involves a wide variety of variables including the method of DC generation, loading, and maturation; the DC dose and route of administration. A maturation cocktail comprising IL-1b, TNF-a, IL-6, and PGE2 (ITIP) is commonly employed clinically, as is the use of the single inflammatory cytokine TNF-a. Here we compared these two methods with respect to DC maturation, phenotype, and function. METHODS: CD14+ monocytes were isolated from normal donor G-CSF-mobilized peripheral blood progenitor cells using the mini-MACS system (Miltenyi Biotec), and cultured for 6 days in AIM-V medium containing GM-CSF and IL-4 (Cellgenix). Immature DCs were loaded with both acute myelogenous leukemia (AML) tumor lysate and mRNA as previously described1 and then matured with the ITIP cocktail, TNF-α alone, or a combination of TNF-α and CD40 agonism (Becton-Dickinson). Mature DCs were analyzed phenotypically and then cultured with autologous T-cells. RESULTS: Data generated from three different normal donors indicated that DCs matured with TNF-α were deficient in important costimulatory surface molecules CD80, CD83, and CD40 as well as in IL-12 production (p=0.001 in representative experiment). Analysis of T-cell functionality by IFN-γ ELISpot also demonstrated greatly decreased Th-1 functionality among lymphocytes stimulated by TNFα-matured DCs in comparison to ITIP-matured DCs (p=0.004 in representative experiment). Additionally, lymphocytes stimulated by TNFα-matured DCs were not able to proliferate significantly whereas ITIP-matured DCs mediated robust lymphocyte expansions consisting of up to 70% CD8+ cells. The addition of a CD40 agonist antibody to TNF-α during the DC maturation process could partially, but not completely overcome these observed functional deficiencies. CONCLUSIONS: The data suggest that DCs loaded with primary tumor antigens and then matured with TNF-a might be deficient in the induction of type 1 T-helper responses. The absence of critical costimulatory markers and IL-12 secretion suggest that DCs matured with TNF-α might be suboptimal for the priming of naïve CD8 responses. These findings might help to improve the clinical efficacy of future DC vaccination studies.