Several current tumor immunotherapies (e.g., PD-1/PD-L1 blockade and tumor vaccine) are designed to boost endogenous tumor-specific T cell responses. To improve the efficacy of current immunotherapies and benefit more cancer patients, it is urgent to advance our knowledge about the cellular and molecular mechanisms underlying the response of endogenous tumor-specific T cells. Importantly, a subset of antigen-specific CD8+ T cells have been identified as stem cell-like or progenitor-like, which are the ones responding to both PD-1/PD- L1 blockade and tumor vaccine. However, we know little about how these stem-like T cells respond to tumor vaccine. Along a different line of research in mouse acute infection models, tissue-resident memory T cells (TRM) have been identified as a unique population of memory T cells. In contrast to other migratory T cell subsets, TRMs do not re-circulate and reside inside a particular tissue (mostly non-lymphoid tissues) for an extended period. We and others have established a critical role for TGF-b in the establishment of TRM after acute infection. Our preliminary findings have demonstrated that tumor draining lymph nodes (TDLNs) function as a unique reservoir to host stem-like tumor-specific CD8+ T cells. Surprisingly, a substantial portion of these TDLN stem-like T cells adopt a TRM phenotype in a TGF-b-dependent manner. Further, we have discovered that wild type TDLN stem- like T cells rapidly, but transiently lose TRM phenotype after tumor vaccine. In contrast, TGF-b receptor deficient TDLN stem-like T cells carry significantly reduced TRM phenotype at baseline and exhibit greatly enhanced and prolonged response to tumor vaccine. The enhanced response in TDLN is translated into increased migration from TDLN to tumor and better tumor control for TGF-b receptor deficient CD8+ T cells. Importantly, inhibition of T cell migration completely abolishes the response to tumor vaccine for TGF-b receptor deficient CD8+ T cells. Together, our results support a working model that a significant portion of stem-like T cells differentiate into TRM inside TDLN and will not migrate to tumor site. Loss of tissue-residency is required for stem-like T cells to differentiate into migratory effectors and elicit robust response to tumor vaccine. Suppression of tissue-residency in TDLN (e.g., deletion of TGF-b receptor or TGF-b downstream molecular targets) will greatly boost the differentiation and migration of stem-like T cells, which will lead to better tumor control in response to certain tumor immunotherapies. In current proposal, we will directly test whether targeting TGF-b or TRM-signature will boost the migration of stem-like T cells and therefore enhance the efficacy of tumor vaccine as well as local irradiation released endogenous tumor antigen. Together, our proposal is primarily focused on the TRM biology of stem-like CD8+ T cells during tumor immunotherapies, with a special emphasis on TDLN. Our results will have great tra...