In the past decade, the discovery of tumor antigens recognized by T cells has revolutionized the tumor vaccine field. The appreciation that peptides are bound to and restricted by major histocompatibility class I and II molecules for immune recognition has encouraged a number of early-phase clinical trials of peptide vaccines. I summarize herein the rationale for and the results of a number of clinical trials of peptide vaccines for melanoma, suggesting that immune and clinical responses can be seen in those with metastatic and resected disease using a variety of surrogate assays. The challenge for the future is to correlate the results of immunologic assays with clinical benefit in patients with advanced cancer.