A pilot study of implication of machine learning for relapse prediction after allogeneic stem cell transplantation in adults with Ph-positive acute lymphoblastic leukemia.

The posttransplant relapse in Ph-positive ALL increases the risk of death. There is an unmet need for instruments to predict the risk of relapse and plan prophylaxis. In this study, we analyzed posttransplant data by machine learning algorithms. Seventy-four Ph-positive ALL patients with a median age of 30 (range 18-55) years who previously underwent allo-HSCT, were retrospectively enrolled. Ninety-three percent of patients received prophylactic/preemptive TKIs after allo-HSCT. The values of the BCR::ABL1 level at serial assessments and over variables were collected in specified intervals after allo-HSCT. They were used to model relapse risk with several machine-learning approaches. GBM proved superior to the other algorithms and provided a maximal AUC score of 0.91. BCR::ABL1 level before and after allo-HSCT, prediction moment, and chronic GvHD had the highest value in the model. It was shown that after Day + 100, both error rates do not exceed 22%, while before D + 100, the model fails to make accurate predictions. As a result, we determined BCR::ABL1 levels at which the relapse risk remains low. Thus, the current BCR::ABL1 level less than 0.06% in patients with chronic GvHD predicts low risk of relapse. At the same time, patients without chronic GVHD after allo-HSCT should be classified as high risk with any level of BCR::ABL1. GBM model with posttransplant laboratory values of BCR::ABL1 provides a high prediction of relapse after allo-HSCT in the era of TKIs prophylaxis. Validation of this approach is warranted.

High prevalence of CD3, NK, and NKT cells in the graft predicts adverse outcome after matched-related and unrelated transplantations with post transplantation cyclophosphamide.

The predictive value of graft composition and plasma biomarkers on the outcome of allogeneic HSCT is well known for conventional GVHD prophylaxis based on calcineurin inhibitors with or without antithymocyte globulin. Currently, there is limited data whether these results could be translated to post transplantation cyclophosphamide (PTCy). The prospective extension cohort of NCT02294552 trial enrolled 79 adult patients with acute leukemia in CR. Twenty-six received matched-related bone marrow (BM) grafts with single-agent PTCy and 53 received unrelated peripheral blood stem cell graft (PBSC) with PTCy, tacrolimus, and MMF. The grafts were studied by the flow cytometry, and plasma samples were analyzed by ELISA. In the cluster and major component analysis, we determined that transplantation from donors with high content of CD3, NKT, and CD16-CD56 + subpopulations in the PBSC grafts was associated with poor immunological recovery and compromised event-free survival (50% vs. 80%, HR 2.93, p = 0.015) both due to increased relapse incidence and non-relapse mortality. The significant independent predictor of moderate and severe chronic GVHD was the high prevalence of and iNKT, Vß11, and double-positive cells in the PBSC grafts from young donors (HR 2.75, p = 0.0483). No patterns could be identified for BM grafts and for plasma biomarkers.

Risk-adapted GVHD prophylaxis with post-transplantation cyclophosphamide in adults after related, unrelated, and haploidentical transplantations.

INTRODUCTION: Although a number of studies were published on the efficacy of post-transplantation cyclophosphamide (PTCy) for graft-versus-host disease (GVHD) prophylaxis, no large studies prospectively evaluated this strategy in related, unrelated, and haploidentical grafts. METHODS: In this study, GVHD prophylaxis for 57 matched bone marrow (MBM) grafts consisted of single-agent PTCy, for 88 matched PBSC grafts (MPBSC) consisted of PTCy, tacrolimus, and mycophenolate mofetil (MMF) 30 mg/kg, and for 55 mismatched grafts (MMGs) consisted of PTCy, tacrolimus and MMF 45 mg/kg. RESULTS: The study met the primary endpoint to demonstrate equivalent rates of acute GVHD grade II-IV (11%, 17%,19%, P = .46), III-IV (7%, 2%, 6%, P = .41), and moderate and severe chronic GVHD (22%, 11%, 15%, P = .23). There was also no differences in non-relapse mortality (11% vs 15% vs 17%, P = .75), overall survival (63% vs 71% vs 56%, P = .72), event-free-survival (51% vs 66% vs 48%, P = .32) for MBM, MPBSC, and MMG groups, respectively. Toxicity was comparable between groups except higher incidence of nephrotoxicity in combination arms (P = .0005) and higher incidence of graft failures in MMG group (P = .004). CONCLUSION: The suggested risk-adapted PTCy-based prophylaxis is feasible and is associated with low GVHD incidence and mortality in all types of grafts. The study was registered on clinicaltrials.gov (NCT02294552).

Graft-versus-Host Disease Prophylaxis in Unrelated Peripheral Blood Stem Cell Transplantation with Post-Transplantation Cyclophosphamide, Tacrolimus, and Mycophenolate Mofetil.

Clinical efficacy of post-transplantation cyclophosphamide (PTCy) as graft-versus-host disease (GVHD) prophylaxis has been demonstrated in haploidentical and HLA-matched bone marrow but not in unrelated peripheral blood stem cell (PBSC) transplantations. Also, no direct comparisons have been published with current standard of care, combination of antithymocyte globulin (ATG), calcineurin inhibitors, and either methotrexate or mycophenolate mofetil (MMF). Eighty-six adult patients (median age 34 years; range, 18 to 59) with acute myeloblastic and lymphoblastic leukemia underwent unrelated PBSC transplantation with PTCy, tacrolimus, and MMF as GVHD prophylaxis in the single-center trial (clinicaltrial.govNCT02294552). The control group comprised 125 consecutive historical control patients who received ATG, tacrolimus, and methotrexate or MMF. Cumulative incidences of grades II to IV acute (19% versus 45%, P = .0003), grades III to IV acute (4% versus 27%, P < .0001), and chronic GVHD (16% versus 65%, P < .0001) were significantly lower in the PTCy compared with the ATG group. PTCy-based prophylaxis was associated with reduced incidence of nonrelapse mortality (16% versus 36%, P = .005; HR, .55; 95% CI, .34 to .89) and improved overall survival (69% versus 40%, P = .0007; HR, .43; 95% CI, .26 to .70), event-free survival (65% versus 38%, P = .0006; HR, .49; 95% CI, .31 to .78), and GVHD relapse-free survival (52% versus 12%, P < .0001). PTCy-based prophylaxis also had a better safety profile compared with ATG with reduced incidence of veno-occlusive disease, cytomegalovirus reactivation, invasive mycosis, and reduced severity of mucositis. In this study we demonstrated that PTCy in combination with tacrolimus and MMF is a safe and effective GVHD prophylaxis for unrelated PBSC transplantation. Although there are several limitations of the historical control approach, this study suggests the superiority of a PTCy-based approach over an ATG-based prophylaxis.