Role of dendritic cell phenotype, determinant spreading, and negative costimulatory blockade in dendritic cell-based melanoma immunotherapy.

MART-1 27-35 -peptide-pulsed immature dendritic cells (DCs) resulted in immunologic and clinical activity in a prior phase I trial. A phase 2 cohort expansion was initiated to further characterize the phenotype and cytokine milieu of the DC vaccines and their immunologic activity in vitro and to further examine a possible link between clinical activity and determinant spreading. In an open-label phase 2 trial, 10 7 autologous ex vivo generated DCs pulsed with the HLA-A*0201 immunodominant peptide MART-1 27-35 were administered to 10 subjects with stage II-IV melanoma. The experimental vaccines were administered intradermally in a biweekly schedule for a total of three injections, and blood for immunologic assays was obtained before each administration and at three time points after. DC vaccine preparations had wide intra- and interpatient variability in terms of cell surface markers and preferential cytokine milieu, but they did not correlate with the levels of antigen-specific T cells after vaccination. Of four patients with measurable disease, one had stable disease for 6 months and another has a continued complete response for over 2 years, which is confounded by receiving a closely sequenced CTLA4 blocking antibody. The DC vaccines induced determinant spreading in this subject, and CTLA4 blockade reactivated T cells with prior antigen exposure. The DC phenotype and cytokine profile do not correlate with the ability to induce antigen-specific T cells, while determinant spreading after DC immunization may be a marker of an efficient antitumor response. Sequential CTLA4 blockade may enhance the immune activity of DC-based immunotherapy.