Contribution of humoral immune responses to the antitumor effects mediated by anthracyclines.

Immunogenic cell death induced by cytotoxic compounds contributes to the success of selected chemotherapies by eliciting a protective anticancer immune response, which is mediated by CD4(+) and CD8(+) T cells producing interferon-γ. In many instances, cancer progression is associated with high titers of tumor-specific antibodies, which become detectable in the serum, but whose functional relevance is elusive. Here, we explored the role of humoral immune responses in the anticancer efficacy of anthracyclines. Chemotherapy reduced the number of tumor-infiltrating B cells, and failed to promote humoral responses against immunodominant tumor antigens. Although anthracycline-based anticancer chemotherapies failed in T cell-deficient mice, they successfully reduced the growth of cancers developing in mice lacking B lymphocytes (due to the injection of a B-cell-depleting anti-CD20 antibody), immunoglobulins (Igs) or Ig receptors (Fc receptor) due to genetic manipulations. These results suggest that the humoral arm of antitumor immunity is dispensable for the immune-dependent therapeutic effect of anthracyclines against mouse sarcoma. In addition, we show here that the titers of IgA and IgG antibodies directed against an autoantigen appearing at the cell surface of tumor cells post chemotherapy (calreticulin, CRT) did not significantly increase in patients treated with anthracyclines, and that anti-CRT antibodies had no prognostic or predictive significance. Collectively, our data indicate that humoral anticancer immune responses differ from cellular responses in, thus far, that they do not contribute to the success of anthracycline-mediated anticancer therapies in human breast cancers and mouse sarcomas.

Immunogenic death of colon cancer cells treated with oxaliplatin.

Both the pre-apoptotic exposure of calreticulin (CRT) and the post-apoptotic release of high-mobility group box 1 protein (HMGB1) are required for immunogenic cell death elicited by anthracyclins. Here, we show that both oxaliplatin (OXP) and cisplatin (CDDP) were equally efficient in triggering HMGB1 release. However, OXP, but not CDDP, stimulates pre-apoptotic CRT exposure in a series of murine and human colon cancer cell lines. Subcutaneous injection of OXP-treated colorectal cancer (CRC), CT26, cells induced an anticancer immune response that was reduced by short interfering RNA-mediated depletion of CRT or HMGB1. In contrast, CDDP-treated CT26 cells failed to induce anticancer immunity, unless recombinant CRT protein was absorbed into the cells. CT26 tumors implanted in immunocompetent mice responded to OXP treatment in vivo, and this therapeutic response was lost when CRT exposure by CT26 cells was inhibited or when CT26 cells were implanted in immunodeficient mice. The knockout of toll-like receptor 4 (TLR4), the receptor for HMGB1, also resulted in a deficient immune response against OXP-treated CT26 cells. In patients with advanced (stage IV, Duke D) CRC, who received an OXP-based chemotherapeutic regimen, the loss-of-function allele of TLR4 (Asp299Gly in linkage disequilibrium with Thr399Ile, reducing its affinity for HMGB1) was as prevalent as in the general population. However, patients carrying the TLR4 loss-of-function allele exhibited reduced progression-free and overall survival, as compared with patients carrying the normal TLR4 allele. In conclusion, OXP induces immunogenic death of CRC cells, and this effect determines its therapeutic efficacy in CRC patients.