The paradox of radiation and T cells in tumors

Abstract

In this review we consider what appears to be a paradox in immunotherapies based around radiation therapy. The paradox is based on three main points. 1. That T cells are needed for radiation's efficacy; 2. That tumor-specific T cells are enriched in the field of treatment; and 3. That radiation kills T cells in the treatment field. We discuss evidence of the effect of radiation on T cells in the field given their ongoing movement in and out of tissues and the tumor, and how the movement of T cells impacts the treated primary tumor and untreated distant metastases. Given this evidence, we revisit the paradox to understand how the extraordinary efficacy of radiation and immunity in preclinical models is dependent on this radiation sensitive cell.

Fig. 1

T cells within radiation treatment fields. A) With radiation treatment made up of multiple overlapping fields focused on the tumor and involved lymph nodes, cells moving through the peripheral blood inside the treatment field will also be irradiated. The flow rates of cells through these vessels means that a large proportion are likely to pass beyond the field before treatment is complete. B) Subtypes of CD8 T cells within the tumor, lymph nodes, and blood, and estimates for the time these cells spend in each site depending on whether or not they meet their cognate antigen.^,

Fig. 2

Clearance of irradiated lymphocytes. Radiation treatment of T cells results in cell death, which in turn results in phagocytic clearance of the dying cells. Circulating cells are actively cleared by macrophages in the marginal zone of spleens, though a range of phagocytic cells have the capacity to phagocytose these cells, and local phagocytic mechanisms likely drive clearance of irradiated T cells in tissues and tumors. Failure of macrophage clearance can result in inflammatory modes of death, and uptake from less abundant phagocytic cells such as dendritic cells which may impact the immune response to cell death. Systemic administration of apoptotic lymphocytes can result in systemic immune suppression, though it is unclear whether this occurs following lymphodepleting radiation treatments.

Fig. 3

Accumulation of tumor specific T cells at sites of antigen. Antigen is abundant in the tumor and can be directly presented by cancer cells or cross presented by dendritic cells. Activated dendritic cells traffic through the efferent lymphatics to draining lymph nodes, and can cause an accumulation of specific T cells in these sites by arresting their traffic and driving expansion. Tumor specific T cells may randomly recirculate through distant sites but would be expected to rapidly exit without meeting their cognate antigen.

Fig. 4

Model of tumor antigen-specific T cell accumulation in tumor-draining lymph nodes. Lymphatics draining tumors deliver T cells (effector T cells – Teff and effector memory T cells (Tem) along with dendritic cells (DC) cross-presenting tumor antigens to the draining lymph node. Other sites in the lymphatic drainage basin may also provide T cells and DC cross-presenting irrelevant antigens. In addition, naïve T cells (Tnaive) and central memory T cells (Tcm) can directly enter the lymph node via high endothelial venules. Within the lymph node T cells that fail to find their cognate antigen (blue), randomly screen antigen presenting cells under the competing influence of chemokines such as CCL7 and S1P, which directs lymphocyte exit. Tumor-specific T cells (red) may arrest on meeting their cognate antigen, and adhesion molecule interactions will overcome exit signals to permit the cells to accumulate and potentially proliferate. Over time, these cells may exceed the available antigen presenting capacity and respond to exit signals. Thus, the transit time of tumor-antigen specific cells may be significantly slower than non-specific T cells.

Fig. 5

Trafficking of T cells is not antigen-directed. A) T cells that exit the tumor pass through the lymphatic chain before re-entering the peripheral blood at the thoracic duct. Similar recirculation pathways return all T cell populations except tissue resident memory cells (Trm) to the peripheral circulation. B) Once in the peripheral blood, effector cells and effector memory cells (Teff, Tem) can be recruited back to the tumor, they may enter normal tissues or metastatic sites according to local inflammatory conditions. Naive T cells and central memory T cells (Tnaive, Tcm) can directly enter lymph nodes and recirculate without entering peripheral tissues. C) T cell entry to any particular tumor, tissue or lymph node is more likely if there is local inflammation that results in upregulation of adhesion molecules on the vasculature and chemokines. D) Inflammation in a tissue site can be propagated via the draining lymphatics to increase entry of T cells to the draining lymph node. Together, these features ensure additional surveillance of inflamed tissues and lymph nodes by recirculating T cells, but relative ignorance of tumors and metastases that are not inflamed.