Finally, able to present ABL

Chronic myelogenous leukemia (CML) has long been a hunting ground for specific therapies of cancer because the disease is initiated by a single, unique chromosomal translocation. The translocation fusion gene product is the target of the first rationally designed tumor inhibitor STI571 (imatinib mesylate, or Gleevec). The unique amino acid sequences at the fusion point of this protein have also interested immunologists, who have seen it as a truly specific cancer antigen. What makes this target so appealing is its specificity for the tumor cell, its limited variability (there are only 2 major breakpoints), and its role as the oncogene product responsible for leukemogenesis. But the exclusively cytoplasmic location for the fusion proteins led to considerable early pessimism that this target would find a use in immunotherapy. Despite this, a number of groups have shown that the amino acid sequences that cross the breakpoint are immunogenic to human T cells in vitro. We also have recently shown that vaccination of patients with CML with these sequences will generate a specific T-cell–mediated immune response to the peptides. What has been most clearly lacking in this field is the direct demonstration that the appropriate amino acid sequences derived from the fusion point could be processed within the cell and incorporated into HLA molecules for presentation on the surface of target leukemia cells where they can be recognized and attacked by cytotoxic T cells.

Clark and colleagues (page 2887) finally provide this direct evidence. The authors do this by using mass spectrometry of sequences of peptides stripped from the HLA molecules on the surface of CML cells. They are able to show that sequences thus derived are identical to sequences predicted (and shown) to be immunogenic. They follow these data with a demonstration that human T cells stimulated to respond to this peptide can be quantitated using HLA tetramers containing the sequence and can kill fresh CML cells in an HLA-restricted, peptide-specific manner. The authors conclude that these data provide a sound basis for immunization of patients against the BCR-ABL protein. As we finalize enrollment in our phase II vaccination trial using these sequences, we heartily and retroactively applaud Clark and colleagues's work and its conclusions.