Glioblastoma is the most aggressive primary tumor of the central nervous system with a medium overall survival of 7–15 months after diagnosis. Since tumor cells penetrate the surrounding brain tissue, complete surgical resection is impossible and tumor recurrence is almost a certainty. New treatment modalities are therefore needed, and these should be able to trace, identify, and kill dispersed tumor cells with great accuracy. Immunological approaches in principle meet these needs. Unfortunately, due to profound tumor-associated mechanisms of immunosuppression and -evasion, immunotherapeutic strategies like peptide vaccination have so far not been translated into clinical success. If future, peptide-based vaccination approaches shall be successful in glioblastoma therapy, multiple questions need to be solved including identification of suitable antigens, route and mode of vaccination, preparation of the tumor-bearing “host” and antagonizing, as much as this is possible, glioblastoma-associated mechanisms of immune evasion and poor vaccination response. In this review we will address the immunological challenges of glioblastoma and discuss key aspects that have rendered successful immunotherapy difficult in the past.