Participants successfully desensitized to peanut through the 2-year-long peanut OIT trial POISED were classified as sustained unresponsive (SU) vs. desensitized alone (DS) based upon oral peanut challenge outcome following a 3-month peanut avoidance period post-OIT. Mass cytometry-based phenotyping of PBMCs from these participants showed that a lower frequency of naïve CD8+ T cells positively associates with sustained unresponsiveness. We further probed CD8+ T cells from SU and DS participants to understand molecules and pathways influencing peanut tolerance. PBMCs from a subset of SU (n=8) and DS (n=7) participants were stimulated with peanut solution or with non-specific stimuli, namely anti-CD3/CD28 beads or PMA/Ionomycin. The cells were harvested, stained with a CD8+ and CD4+ subset-focused antibody panel, and acquired on a spectral flow cytometer. Data were analyzed by manual gating. Concomitant with the lower naïve CD8+ T cell frequency, SU participants showed a higher frequency of effector memory (EM) CD8+ T cells compared to DS post-OIT. In addition, post-OIT EM and terminally differentiated CD45RA+ EM (TEMRA) CD8+ T cells from SU participants expressed higher granzyme B and perforin than DS participants in response to both the non-specific stimuli tested. Notably, CD222 (mannose-6-phosphate receptor), a putative receptor for extracellular granzyme B, was exclusively expressed by ∼10% of peanut-reactive (pr) CD4+ T cells compared to other Th subsets and Treg cells. Our preliminary findings suggest a hypothesis that cytotoxic molecules such as granzyme B sourced from memory CD8+ T cells may target pathogenic pr CD4+ T cells, and thus support SU status.