Characterization of T-cells in ER + breast tumors and their matching adjacent tissue reveals that classical exhaustion markers are not indicative of T-cell function

The manipulation of patients' tumor-infiltrating T-lymphocytes (TIL) as a form of immunotherapy has shown promising results in the treatment of select cancers; however, limited benefit has been observed in treating invasive estrogen receptor-expressing (ER + ) breast tumors. The development of a dysfunctional, exhausted-like T-cell phenotype in the tumor microenvironment has been linked to immunotherapy failure, but the intrinsic and extrinsic factors that result in its development remain poorly understood. We previously showed that breast tumor-adjacent tissue (TAT), located over three centimeters from the primary tumor, contains an inflammatory signature and, therefore, hypothesized that it may be a reservoir of tumor-reactive T-cells. Further, the comparative study of TILs and TAT-T-cells may reveal important information about the mechanisms of TIL exhaustion and how to overcome them. ER + breast tumor and matching TAT samples were collected from patients undergoing mastectomy procedures, from which the breast cancer cells (BCC), TIL-CD3 + and TAT-CD3 + T-cells were obtained. The cytotoxic CD8 + subset from each CD3 + T-cell sample was isolated and examined for the expression of exhaustion and activation markers and then placed in a three-dimensional co-culture assay with the autologous BCCs. Here, BCC viability and T-cell exhaustion marker expression were assessed and conditioned media was collected for cytokine analysis. Interestingly, the TIL-CD8 + T-cells expanded more rapidly than the TAT-CD8 + T-cells; however, the TAT-CD8 + T-cells were more effective in eliminating autologous BCCs. The enhanced reactivity of the TAT-T-cells was further corroborated with enhanced secretion of activating cytokines such as interferon gamma. T-cells with an increased expression of exhaustion marker TIM-3 (T cell immunoglobulin and mucin-domain containing-3) prior to co-culture were less effective in eliminating BCCs. Further, in co-cultures, up to a 9-fold increase in the expression of exhaustion marker LAG-3 (lymphocyte activation gene 3) was observed in select TIL- and TAT-CD8 + T-cells. Interestingly, the expression of PD-1 (programmed cell death protein 1), the most frequently studied marker in the context of T-cell dysfunction, remained low and unchanged. To our knowledge, this is the first detailed study of the CD8 + T-cells that reside in ER + breast tumors and their adjacent tissues. We have observed that TAT contains CD8 + T-cells with enhanced reactivity against autologous BCCs despite having a similar exhaustion marker profile as the TIL CD8 + T-cells. Interestingly, our findings also suggest that TIM-3 and LAG-3 may contribute to T-cell exhaustion in ER + breast tumors and that blocking these receptors might be more effective than blocking PD-1 as immunotherapy.