Revealing the Generation of Human Memory Stem T Cells in Haploidentical T-Replete Hematopoietic Stem Cell Transplantation

Introduction: The T cell memory compartment is multi-faceted and encompasses multiple subsets with divergent properties. In addition to central memory (T_CM) and effector memory (T_EM) cells, the spectrum of immunological memory has been recently extended with the identification of memory stem T cells (T_SCM). Gene expression profiling, corroborated by in vitro and in vivo experimental results, posits T_SCM upstream T_CM and T_EM in T-cell ontogeny. However, T_SCM role in immune reconstitution (IR) following allogeneic HSCT remains largely unknown. Here, we tracked T_SCM dynamics in patients undergoing haploidentical HSCT. By this process, we unconventionally exploited HSCT as model system to provide novel insights into the contribution of T_SCM to a mounting immune response.

Patients and Methods: We characterized T cell dynamics during the first month post HSCT in 20 patients receiving myeloablative conditioning, T-replete haploidentical peripheral blood stem cell graft, and GvHD prophylaxis consisting of cyclophosphamide (PT-Cy day 3,4), MMF and sirolimus from day 5.

Results: Upon infusion, naïve T cells (T_N) cells gradually disappeared from circulation, and by day 8 post HSCT the peripheral compartment was composed only by memory lymphocytes. At day 8, T_SCM cells were the most represented circulating subset, and their frequency was significantly higher than that of the infused graft (LK). To investigate whether such high T_SCM frequency was due to expansion of T_SCM cells infused or to their direct in vivo generation, T cell subset proliferation was analyzed. At day 3 upon infusion (in the absence of immunosuppression), all memory subsets robustly proliferated, with T_SCM cells displaying significantly higher frequencies of Ki-67⁺ cells compared to all other subsets. In sharp contrast, T_N cells were scantly Ki-67⁺, in line with the longer timeframe required for their priming. PT-Cy abrogated proliferation in all subsets. T cell counts however did not drop between day 5 and 8 post HSCT, and T_SCM cells increased in percentages and absolute numbers. We thus explored whether T_SCM cells were resistant to PT-Cy, but failed to record ALDH activity, the major mechanism of Cy inactivation, in CD3⁺ cells. neither within LK nor upon infusion. Accordingly, we detected high percentages of apoptotic cells within all memory subsets at day 5 post HSCT. Nonetheless, at day 8 significantly lower percentages of T_SCM cells were Annexin V⁺ compared to T_CM/T_EM, while by day 15 post-HSCT all memory subsets displayed very low levels of apoptosis. Thus, we reasoned that direct conversion of T_N into T_SCM could be the preferential mechanism underlying T_SCM expansion. To prove this, we exploited the TCR sequences harbored by each individual T cell as surrogate clonal markers. Purified T cell subsets from LK and harvested 30 days post HSCT in 3 consecutive patients were subjected to TCR sequencing. Sequences harbored by LK-T_N were retrieved in all memory subsets at day 30 after HSCT, showing that T_N were able to generate the complete spectrum of immunological memory, including T_SCM. Quantitative analysis revealed that only clonotypes originally harbored by LK-T_N were significantly increased in counts at day 30 post-HSCT. In contrast, sequences unique to LK-T_SCM, LK-T_CM or LK-T_EM were similar or reduced in counts at day 30 post-HSCT compared to LK, indicating that either PT-Cy efficiently dampened their expansion and/or that such memory lymphocytes persisted unchallenged upon in vivo infusion. Overall, in vivo fate mapping through TCR sequencing allowed defining the in vivo differentiation landscapes of human naïve T cells upon HSCT, highlighting T_SCM as privileged players in the diversification of immunological memory. We next validated these results at the antigen-specific level. In suitable 5 patients, we recorded the presence of WT1 or PRAME-specific T_N cells within donor LK. Upon HSCT, tumor-specific T cells increased in frequency and displayed a memory phenotype, comprising T_SCM. Finally, by quantifying patient serum cytokines, we found that the degree of IL-7 availability at day 1 post HSCT correlated with the extent of T_SCM expansion at day 8.

Conclusions: These data provide novel insights into T_SCM biology and ongoing analyses will define correlations with clinical events. Longer immunological follow-up will advance our understanding of the contribution of early T_SCM generation to the overall success of post HSCT IR.