Tumor growth is blunted in mice immunized with doxorubicin-treated cells.

The way a tumor cell dies determines its immune-stimulating capacity, according to a study by Casares and colleagues on page 1691. Tumor cells that die from exposure to the chemotherapy drug doxorubicin, they show, become highly immunogenic and induce the regression of established tumors when injected into mice.

Inactivated tumor cells have long been considered to be attractive vaccine immunogens, in part because whole-cell vaccines would allow each patient to be immunized against his or her own tumor. But this approach has met with limited success, probably because traditional ways of inducing tumor cell death—by irradiation or freeze-thawing—somehow strip the cells of their immunogenicity.

Many vaccine strategists have instead turned to T cell-inducing dendritic cell (DC) vaccines, in which DCs are grown from a patient's blood, pulsed with tumor antigens, and reinjected. But the new data by Casares and colleagues might help revive interest in the simpler whole-cell approach.

Casares et al. show that tumor cells treated with doxorubicin, but not those treated with the chemotherapy drug mitomycin, elicit a tumor-specific T cell response in mice. Although both treatments induced apoptosis of the injected tumor cells, only the doxorubicin-treated cells were taken up by DCs, which then activated specific T cells. The immune-stimulating effect of doxorubicin required the activity of the death-inducing caspase enzymes in the tumor cells, although the relevant caspase targets were not identified.

The authors are now trying to pin down the doxorubicin-induced changes on the dying tumor cells that render them more palatable to DCs. In the meantime, they suggest that this approach—should it work in humans—might provide a more practical alternative to anticancer vaccine strategies that require the labor-intensive cultivation of autologous DCs.