Cancer vaccine strategies: translation from mice to human clinical trials.

We translated two cancer vaccine strategies from mice into human clinical trials. (1) In preclinical studies on TARP, an antigen expressed in most prostate cancers, we mapped epitopes presented by HLA-A*0201, modified them to increase affinity and immunogenicity in HLA transgenic mice, and induced human T cells that killed human cancer cells ("epitope enhancement"). In a clinical trial, HLA-A2+ prostate cancer patients with PSA biochemical recurrence (Stage D0) were vaccinated with two peptides either in Montanide-ISA51 or on autologous dendritic cells (DCs). In stage D0, the Prostate-Specific Antigen (PSA) slope is prognostic of time to radiographic evidence of metastases and death. With no difference between arms, 74% of combined subjects had a decreased PSA slope at 1 year compared to their own baseline slopes (p = 0.0004). For patients vaccinated with DCs, response inversely correlated with a tolerogenic DC signature. A randomized placebo-controlled phase II trial is underway. (2) HER2 is a driver surface oncogene product expressed in multiple tumors. We made an adenoviral vector vaccine expressing the extracellular and transmembrane domains of HER2 and cured mice with large established HER2+ tumors, dependent on antibodies to HER2, not T cells. The mechanism differed from that of trastuzumab. We tested a human version in advanced metastatic cancer patients naïve to HER2-directed therapies. At the second and third dose levels, 45% of evaluable patients showed clinical benefit. Circulating tumor cells also declined in some vaccinated patients. Thus, cancer vaccines developed in mice were successfully translated to humans with promising early results.