The brain (William Say, 1829)
The brain (William Say, 1829)

The other day I talked about Jim Allison’s paper1 that proposed (among other things) that irradiation of tumors might help the immune system detect them. A quite unrelated paper2 seems to have almost accidentally helped test that suggestion.

Allison’s group was looking at the puzzling fact that circulating anti-tumor T cells don’t seem to correlate well with effective anti-tumor immune responses. They suggest that part of the problem is that the blood vessels that supply tumors are often abnormal, and these altered blood vessels don’t allow T cells to enter the tissues properly. However, irradiating those blood vessels made them closer to normal, allowed the T cells to enter the tumor more effectively, and led to more complete immune clearance of the tumor.

The new paper I mentioned2 wasn’t specifically looking at tumor vasculature at all. They were working with a model of brain cancer (a particularly difficult nut to crack, immunologically). They used transgenic mice, that express a tumor antigen (the SV40 virus large T antigen) in some brain cells; these mice develop brain cancer and die by 3 or 4 months of age. They’re immunologically tolerant to the tumor antigen, as you’d expect; but transferring tumor antigen-specific T cells from non-transgenic mice helped them survive longer. In this particular paper, they try to narrow down the specific conditions that allow immune rejection of the tumor.

T cells attacking tumor cellThey found a number of important factors, but the interesting one in light of the Allison paper was that “successful adoptive immunotherapy of T Ag-induced tumors still required prior conditioning of the host with gamma irradiation”.  Just transferring the T cells helped a bit, delaying death by a couple of months; but irradiating the tumors and transfering the T cells completely eliminated the tumors. They didn’t specifically look at the tumor blood vessels (the Allison paper wasn’t out when this work was done, of course), but they did note that irradiation correlated with increased T cell accumulation in the brain.

Another interesting point is that the irradiated tumors were cleared even when only a single T cell epitope was targeted. That’s an important finding, because one might expect that the tumor might be able to recur because of immune escape. (That is, tumor cells with fortuitous mutations in that epitope would not be killed by the T cells, and would be able to regrow and recur.) In fact, that might have happened with the non-irradiated mice, in which the tumor shrank briefly then rapidly recurred. (They didn’t check the epitope, so it’s equally possible that something like tolerance or deletion of the T cells was involved in this situation.) Still, it seems that, at least in this model, you can get rid of a tumor altogether, even if you only have a single tumor epitope to work with. Given the limited number of tumor epitopes we have to work with, that’s pretty encouraging.


  1. S. A. Quezada, K. S. Peggs, T. R. Simpson, Y. Shen, D. R. Littman, J. P. Allison (2008). Limited tumor infiltration by activated T effector cells restricts the therapeutic activity of regulatory T cell depletion against established melanoma Journal of Experimental Medicine, 205 (9), 2125-2138 DOI: 10.1084/jem.20080099[]
  2. Angela M. Tatum, Lawrence M. Mylin, Susan J. Bender, Matthew A. Fischer, Beth A. Vigliotti, M. Judith Tevethia, Satvir S. Tevethia and Todd D. Schell.
    CD8+ T Cells Targeting a Single Immunodominant Epitope are Sufficient for Elimination of Established SV40 T Antigen-Induced Brain Tumors.
    The Journal of Immunology, 2008, 181: 4406-4417.
    [][]