Endogenous regulation of inflammatory response as a determinant of durable remission without stem cell transplant following brexucabtagene autoleucel (Brexu-cel) therapy in ALL Free

Introduction: Brexu-cel therapy achieved remarkable overall complete remission rates in the pivotal ZUMA-3 (NCT02614066) for adults with relapsed/refractory B-cell acute lymphoblastic leukemia (R/R B-ALL). However, understanding which patients (pts) can maintain long-term remission following brexu-cel without allogeneic stem cell transplant (alloSCT) remains limited. Our previous analyses from the ZUMA-3 pivotal study suggested that early recovery of endogenous immune cells (monocytes and lymphocytes) and a higher peak CAR T-cell expansion were associated with achieving longer duration of responses. Here, we extend our analyses to examine the dynamics of early inflammation to identify high-risk pts (non-responders) in comparison to pts who achieve durable/ long-term (>12 months) responses without alloSCT.

Methods: Pts from ZUMA-3 Phase 1 and 2 study treated at the pivotal dose (1×10⁶ CAR T cells/kg) were stratified by response duration: non-responders (n=18), responders with ≤12 months duration (n=14), and responders with >12 months duration (n=18). Pts who underwent alloSCT within 12 months post brexu-cel were excluded. Kinetics of CAR-T expansion and recovery of platelets, neutrophils, monocytes, and lymphocytes were assessed over a 28-day period post-infusion. Baseline hematologic (PLT, Hgb, ANC) and inflammatory markers (CRP, ferritin) were evaluated and used to calculate CAR-hematotoxicity (HT) scores. Cytokine profiles at baseline and serial time points post-T cell infusion were assessed.

Results: Evaluation of serial serum cytokines demonstrated a distinct pattern between responders and non-responders. Multiple inflammatory chemokines and cytokines - including MCP-1, IL-6, IL-8, and IL-10 – were highly elevated between days 7 and 14 post CAR-T infusion in non-responders, although there was no significant difference in levels of these analytes at baseline across the groups. T cell specific cytokines IFN-g and IL-2 were also surprisingly higher in non-responders at day 7 post infusion despite lack of CAR-T expansion. Elevated levels of these analytes were not associated with the severity of either CRS or neurotoxicity, indicating a distinctly shifted inflammatory response in non-responders. Overall, long-term responders (>12 months) tended to exhibit moderate levels of these inflammatory mediators – higher than short-term but lower than non-responders - along with the highest expansion peak of CAR T cells, highlighting a more calibrated endogenous inflammatory response.

While systemic inflammation defined by CAR-HT scores ≥2 was prevalent at pretreatment in most pts (n=47), the median CAR-HT scores were significantly lower (p<0.05) in pts that achieved longer duration of response (HT=4) than non-responders (HT=6). Univariate analyses of HT score parameters revealed baseline platelet counts to be significantly higher (p<0.05) in long-term responders vs. non-responders. Consistently, post infusion recovery of platelets as well as myeloid cells (neutrophils and monocytes) within first month were also significantly greater in long-term responders vs. non-responders (p<0.05). Baseline bone marrow (BM) blast levels were not significantly different across the three groups. Notably, 10 of 18 patients in the long-responders without alloSCT included pts with high BM blasts (>50%), and they were characterized by higher levels of platelets at both baseline and at day 28 post CAR-T infusion.

Conclusions: Our data demonstrate that the host inflammatory response post CAR-T drive transplant-free outcomes in brexu-cel treated ALL pts. We also found that pts with durable responses following brexu-cel without consolidative alloSCT had higher PLT counts at baseline and at recovery at day 28. Non-responders exhibited a hyperinflammatory response despite low CAR T cell expansion, suggesting dysregulated immune activation, whereas long-term responders without alloSCT showed a more calibrated inflammatory response and early recovery of platelets despite having the highest levels of CAR T cell expansion. These results highlight the importance of dynamic interplay between host immune response and CAR T cells, and future work will address whether the addition of immune-modulatory or anti-inflammatory agents post-CAR could further enhance the curative potential of CAR T cell therapy in ALL. More detailed data of immune recovery and cytokine changes following brexu-cel will be presented at the meeting.