Identification of reliable surrogate predictors for evaluation of cancer vaccine efficacy is a critical issue in immunotherapy. We analyzed quantitative and qualitative CD8+ T cell parameters in a large pool of BALB/c mice that were DNA-vaccinated against P1A self tumor-specific Ag. After immunization, mice were splenectomized and kept alive for a subsequent tumor challenge to correlate results of immune monitoring assays with tumor regression or progression in each individual animal, and to assess the prognostic value of the assays. The parameters tested were 1) percentage of in vivo vaccine-induced tumor-specific CD8+ T cells; 2) results of ELISPOT tests from fresh splenocytes; 3) percentage of tumor-specific CD8+ T cells in culture after in vitro restimulation; 4) in vitro increase of tumor-specific CD8+ T cell population expressed as fold of expansion; and 5) antitumor lytic activity of restimulated cultures. Except for the ELISPOT assay, each parameter tested was shown by univariate statistical analysis to correlate with tumor regression. However, multivariate analysis revealed that only in vitro percentage of Ag-specific CD8+ T cells was an independent prognostic factor that predicted tumor outcome. These findings should be considered in the design of new immune monitoring systems used in cancer immunotherapy studies.