Abstract Some extraordinary clinical successes of immunotherapy have had game-changing effects on the way we view cancer biology and the interaction of the immune system with our tissues. The conventional context for considering cancer immunology is one in which dendritic cells and/or monocytes present cancer antigens, particularly neo-antigens generated by somatic mutation, to systemic T cells within cancer-draining lymph nodes, as a result of which an expansive adaptive, cancer-specific T lymphocyte-driven immune response develops. Furthermore, cancer-specific T cells may become tissue-associated, maintaining a so-called Tissue-Resident Memory (TRM) pool. However, although this perspective readily explains many basic biological and clinical data, it leaves several issues unanswered. For example, does the immune system recognize tumor cells that do not carry high mutational loads, many of which cancers have shown disappointing responses to immunotherapy? Related to this, what is the nature of cancer immunogenicity that is required to first activate the immune system, thereby promoting the presentation of cancer antigens by DC and/ or monocyte? We have addressed these issues by considering a separate T cell compartment that is constitutively tissue-resident, rather than being primed in the lymphoid circulation. Prominent within this T cell compartment are intraepithelial gamma delta T cells that have been particularly well characterized in mice, wherein they have been shown to limit sensitivity to carcinogens, and to promote tissue integrity1,2. Moreover, a recent bio-informatic analysis showed a tumor-associated gamma delta T cell signature to be the strongest gene expression correlate with overall survival in over 18,000 patients across 39 tumor types3. Furthermore, another very strong correlate was the expression of NKG2D, a cell surface receptor expressed on tissue-resident gamma delta T cells and whose ligands are upregulated on tumor cells by dysregulation of the EGF-receptor pathway4. Given these observations, we have asked several questions: [1] what is the nature of tissue-resident gamma delta T cell compartments in human skin, gut, and breast tissue? [2] do such cells become activated solely by the upregulation of NKG2D ligands, thereby offering an explanation for tumor immunogenicity? [3] what are the effector functions and potentials of such cells? [4] what are the molecular checkpoints by which the cells are regulated? [5] can the cells be explanted, grown, and cultured in a manner that preserves their function(s) and that can provide a source of such cells to be used for adoptive immunotherapy? [6] might such cells be applied “off-the-shelf” rather than requiring autologous transfusions? Emerging answers to these questions will be provided. Thus, by examining over 100 donors, we have found that human skin, gut, and breast reproducibly harbor large and complex T cell subsets that include substantive gamma delta T cell compartments clearly distinct from most blood-borne gd T cells. These tissue-resident gamma delta T cells show strong, innate-like responses to NKG2D-ligands and to cytokines that were not shown by CD8+ TRM cells harvested in parallel from the same sites. Additionally, however, tissue-resident gamma delta T cells are uniquely and powerfully regulated at steady-state by novel, CD80/B7/PDL1-like molecules known as butyrophilin-like (BTNL) proteins that are expressed by epithelial cells in a tissue-specific fashion, and that de facto constitute local checkpoints. These emerging insights create a new context in which to consider the interaction of immune cells with developing cancers, and possibly to explain the patient benefit of a strong gamma delta T cell signature. At the same time, we have developed novel methods for maintaining and expanding tissue-derived gamma delta T cells that offer a practical off-the-shelf approach to tissue-targeted cellular immunotherapy, and that may be particularly suited to adenocarcinomas of low mutation load or with high rates of tumor evolution and antigenic drift. Citation Format: Oliver Nussbaumer, Yin Wu, Fernanda Kyle, Rick Woolf, Robin Dart, Deborah Enting, Pierre Vantourout, Adrian Hayday. Targeting molecules mediating the dialogue between epithelial cells and local T cells. [abstract]. In: Proceedings of the AACR Precision Medicine Series: Targeting the Vulnerabilities of Cancer; May 16-19, 2016; Miami, FL. Philadelphia (PA): AACR; Clin Cancer Res 2017;23(1_Suppl):Abstract nr IA20.