Abstract Multiple myeloma (MM) is the second most common hematological malignancy and despite the developments of novel therapies, it remains largely incurable. A minority of patients do achieve durable disease control after autologous stem cell transplantation (SCT), and we previously demonstrated the induction of profound T cell-mediated myeloma-specific immunity after SCT in preclinical models. Nonetheless, in both mice and patients, myeloma progression is associated with inhibitory receptor expression on CD8 T cells in the bone marrow (BM). Characterization of CD8 T cell exhaustion and the presence of precursor exhausted T cells (TPEX) has been limited to solid tumor models. Whether CD8 T cells exhibit the same trajectory of exhaustion from TPEX in the MM BM microenvironment after SCT was hitherto unknown. This is an important clinical question as immunotherapies are largely utilized in relapsed/refractory MM and currently do not target this patient population. We performed single cell RNA sequencing in mice with relapsed MM after SCT to provide in-depth characterization of CD8 T cell differentiation and exhaustion. CD8 T cells were sorted via flow cytometry based on CD38 and CD101 expression to capture diverse stages of differentiation. Notably, CD38+CD101+ CD8 T cells have been previously described as irreversibly exhausted and represent over 50% of polyclonal CD8 T cells in our model. We performed unsupervised clustering and identified ten CD8 T cell clusters that spanned T cell differentiation, including a TPEX population with a phenotype akin to solid tumor settings. We also observed two distinct exhausted T cell clusters that both expressed Tox, Pdcd1, and lacked Tcf7 while only one cluster expressed high levels of Prdm1 (Blimp-1), Havcr2 (TIM-3), Maf and Il10. Utilizing flow cytometry, we confirmed expression of TOX and PD-1, with loss of TCF-1, in mice with relapsed MM and expression of c-Maf was largely restricted to the TIM-3+ subset. We next sought to determine whether the exhausted clusters represented degrees of exhaustion or were distinct lineages. Thus, we performed RNA velocity analysis and noted a clear trajectory from the TPEX cluster towards the two exhausted clusters, confirming a common trajectory of CD8 T cell exhaustion states across solid and hematological malignancies. Surprisingly, there was a divergence point between the exhausted clusters suggesting that these could be distinct lineages. To address this, we performed bone marrow aspirates in mice from 3 weeks post-SCT to track exhaustion phenotypes over time in myeloma-bearing mice. We observed concurrent emergence of both exhausted T cell phenotypes from early post-SCT, supporting our hypothesis that these may represent distinct T cell exhaustion lineages. These data highlight a trajectory of CD8 T cell exhaustion from precursor subsets in MM that supports the utilization of immunotherapies in the early stages of disease. Citation Format: Simone A. Minnie, Nicole S. Nemychenkov, Olivia G. Waltner, Kathleen S. Ensbey, Christine R. Schmidt, Shruti S. Bhise, Scott N. Furlan, Geoffrey R. Hill. CD8 T cells display distinct trajectories of T cell exhaustion in the bone marrow of mice with multiple myeloma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 1360.