Reconstitution Dynamics and Impact on Clinical Outcomes of Mucosal-Associated Invariant T-Cells in Pediatric Patients Receiving Allogeneic Hematopoietic Stem Cell Transplantation

Abstract

Background Understanding the reconstitution of all immune system components after allogeneic hematopoietic stem cell transplantation (HSCT) is crucial to optimize the prevention of infections, leukemia relapse and graft-versus-host disease (GvHD), thus improving patient's outcome. Mucosal-associated invariant T (MAIT) cells have received major interest in the field for their ability to respond to fungal and bacterial infections and control mucosal homeostasis in the GvHD context. However, the impact of MAIT cell recovery on T-cell immunity and on patient outcomes after HSCT remains not fully characterized. Patients and study design We retrospectively analyzed 156 consecutive patients (median age 8.5 years, range 0.6-24.7) affected by malignant hemopathies undergoing HSCT between April 2019 to May 2022, fully engrafted at day+30 after receiving the graft from an unrelated matched (MUD) or mismatched (mMUD) donor (n=52 and n=11, respectively) or from an HLA-haploidentical (Haplo) donor (n= 93) after alpha-beta T-/CD19-cell depletion. Conditioning regimen was myeloablative, TBI- (n=120) or chemo-based (n=36) in all patients. GvHD prophylaxis included short course methotrexate, anti-thymocyte globulin, cyclosporin with or without abatacept 10mg/kg at day -1,+5,+14,+28,+42,+60 for patients transplanted from mMUD or MUD, respectively. Haplo-HSCT patients did not receive any post-HSCT pharmacological prophylaxis. Patient characteristics are listed in Table 1. We investigated T and, in particular, MAIT cell recovery up to 2 years after HSCT and their correlation with clinical outcomes. Results With a median follow-up of 27 months (range 2-49), the overall survival (OS), disease-free survival (DFS) and non-relapse mortality (NRM) of the entire cohort were 78,5% (95%CI 70-84), 70% (95%CI 62-77) and 6% (95%CI 3-11), respectively. The cumulative incidence (CI) of relapse was 23% (95%CI 17-31). CI of grade II-IV acute GvHD (aGvHD) was 19.7% (95%CI 14-27), while CI of chronic GvHD (cGvHD) was 15% (95%CI 10-22) with a CI of moderate-severe cGvHD of 10% (95% CI 6-16), resulting in a cGvHD-free, relapse-free survival (GRFS) of 63% (95%CI 54-69). CI of post-HSCT CMV reactivation was 35% (95%CI 28-43); 13 and 22 patients experienced pre-engraftment or late blood stream infections (BSI) (CI 8%, 95%CI 5-13 and 14%, 95%CI 9-20, respectively). While total abT cells reached normal levels between 1 and 2 years post-transplantation, MAIT cells showed prolonged impaired reconstitution up to 2 years post-HSCT. Compared to UD-HSCT, MAIT cell recovery after Haplo-HSCT was significantly delayed. Interestingly, across all time points analyzed, we observed that a large fraction of MAIT cells were negative for CD161, a marker canonically expressed on MAIT cells. We also found that MAIT cells showed prolonged functional impairment as demonstrated by the increased expression of the activation (CD25, CD38), exhaustion (PD1, TIM3) and senescence (CD57) markers, as well as by the suboptimal response to ex vivo T-cell activation 1 year post HSCT. OS, DFS and NRM were not influenced by the absolute number and proportion of MAIT cells when assessed 30 days post-HSCT. Interestingly, higher absolute count of MAIT at +30 was associated with higher CI of aGvHD (27% vs 11%, p=0.04) in univariate analysis. Furthermore, an higher proportion of MAIT cells at the same time point, as well as the use of mMUD as donor, were associated as independent factors with a higher CI of all grades of cGvHD (25% vs 8%, p=0.01 and 74% vs 12%, p<0.001, respectively) and moderate-severe cGvHD (17% vs 4% p=0.024 and 59% vs 8% p=0.017, respectively), resulting in lower GRFS in patients with an higher proportion of MAIT cells at +30 (72% vs 52%, p=0.019) (Fig. 1). Univariate analysis also showed that patients with higher absolute values of MAIT had a lower CI of late BSI (17% vs 35%, p=0.05) but experienced more frequently CMV reactivation (44% vs 26%, p= 0.029) which occurred more in Haplo- and mMUD-recipients compared to MUD-patients (41.2% and 45.4% vs 21.2%, p=0.029). Conclusions Our study shows that MAIT cell recovery is significantly impaired in terms of absolute number and function in patients receiving HSCT. Evaluation of MAIT cells at early time points after HSCT may represent a useful parameter to identify patients at risk of late BSI, cGvHD and CMV reactivation. Additional studies are needed to further assess the biological function of MAIT cells.