The role of the conditioning regimen for autologous and ex vivo genetically modified hematopoietic stem cell-based therapies: recommendations from the ISCT stem cell engineering committee.

BACKGROUND: The advent of autologous gene modified cell therapies to treat monogenic disorders has been a major step forward for the field of hematopoietic stem cell transplantation (HCT) and cellular therapies. The need for disease-specific conditioning to enable these products to provide a potential cure has required extrapolation from experience in myeloablative and non-myeloablative HCT for these disorders. METHODS: In this manuscript, we review the current datasets and clinical experience using different conditioning regimens for autologous gene therapies in hemoglobinopathies, metabolic and lysosomal disorders, inborn errors of immunity (IEI) and bone marrow failure (BMF) syndromes. RESULTS: The disease specific and unique conditioning requirements of each disorder are considered in order to achieve maximal benefit while minimizing associated toxicities. CONCLUSIONS: Standardized recommendations based on these data are made for each set of disorders to harmonize treatment. Future directions and the possibility of non-genotoxic conditioning regimens for autologous gene therapies are also discussed. Ethical Statement: The authors followed all relevant ethical considerations in writing this manuscript.

Association of busulfan exposure and outcomes after HCT for patients with an inborn error of immunity.

ABSTRACT: Allogeneic hematopoietic cell transplantation (HCT) is a potentially curative treatment strategy for patients with inborn errors of immunities (IEIs). The objective of this study was to assess the optimal busulfan exposure before allogeneic HCT for patients with an IEI who received an IV busulfan-based conditioning regimen. Patients from 17 international centers were included. The main outcome of interest was event-free survival (EFS). Patients were categorized into 4 IEI subgroups: combined immunodeficiency (CID), severe combined immunodeficiency (SCID), neutrophil disorders, and hemophagocytic lymphohistiocytosis (HLH)-related disorders. Busulfan exposure was calculated by individual centers (area under the curve [AUC]CENTER) and re-estimated using a nonlinear mixed-effects model (NONMEM; exposure defined as AUCNONMEM). Overall, 562 patients were included: 173 (30.8%) with CID, 154 (27.4%) with SCID, 101 (18.0%) with HLH-related disorders, and 134 (23.8%) with neutrophil disorders. The median busulfan AUCNONMEM was 69.0 mg × h/L and correlated poorly with the AUCCENTER (r2 = 0.54). In patients with SCID, HLH-related, and neutrophil disorders with a busulfan AUCNONMEM of 70 to 90 mg × h/L, 2-year EFS was superior to <70 mg × h/L, and >90 mg ×h/L. Full donor chimerism increased with higher busulfan AUCNONMEM, plateauing at 90 mg × h/L. For patients with CID, the optimal AUCNONMEM for donor chimerism was found to be >70 mg × h/L. Improved EFS and higher donor chimerism may be achieved by targeting a cumulative busulfan AUCNONMEM of 80 mg × h/L (range, 70-90). Our study stresses the importance of uniformly using a validated population pharmacokinetic model to estimate AUCNONMEM.

International Society for Cell & Gene Therapy Stem Cell Engineering Committee report on the current state of hematopoietic stem and progenitor cell-based genomic therapies and the challenges faced.

Genetic manipulation of hematopoietic stem cells (HSCs) is being developed as a therapeutic strategy for several inherited disorders. This field is rapidly evolving with several novel tools and techniques being employed to achieve desired genetic changes. While commercial products are now available for sickle cell disease, transfusion-dependent ß-thalassemia, metachromatic leukodystrophy and adrenoleukodystrophy, several challenges remain in patient selection, HSC mobilization and collection, genetic manipulation of stem cells, conditioning, hematologic recovery and post-transplant complications, financial issues, equity of access and institutional and global preparedness. In this report, we explore the current state of development of these therapies and provide a comprehensive assessment of the challenges these therapies face as well as potential solutions.

International Society for Cell and Gene Therapy Clinical Translation Committee recommendations on mesenchymal stromal cells in graft-versus-host disease: easy manufacturing is faced with standardizing and commercialization challenges.

Mesenchymal stromal cells (MSCs) have been used in multiple clinical trials for steroid-refractory moderate-severe (grade II-IV) acute graft-versus-host disease (aGVHD) across the world over the last two decades. Despite very promising results in a variety of trials, it failed to get widespread approval by regulatory agencies such as the U.S. Food and Drug Administration and the European Medicines Agency. What lessons can we learn from this for future studies on MSCs and other cell therapy products? Broad heterogeneity among published trials using MSCs in aGVHD was likely the core problem. We propose a standardized approach in regards to donor-related factors, MSCs-related characteristics, as well as clinical trial design, to limit heterogeneity in trials for aGVHD and to fulfill the requirements of regulatory agencies. This approach may be expanded beyond MSCs to other Cell and Gene therapy products and trials in other diseases.

Developing a membrane-proximal CD33-targeting CAR T cell.

BACKGROUND: CD33 is a tractable target in acute myeloid leukemia (AML) for chimeric antigen receptor (CAR) T cell therapy, but clinical success is lacking. METHODS: We developed 3P14HLh28Z, a novel CD33-directed CD28/CD3Z-based CAR T cell derived from a high-affinity binder obtained through membrane-proximal fragment immunization in humanized mice. RESULTS: We found that immunization exclusively with the membrane-proximal domain of CD33 is necessary for identification of membrane-proximal binders in humanized mice. Compared with clinically validated lintuzumab-based CAR T cells targeting distal CD33 epitopes, 3P14HLh28Z showed enhanced in vitro functionality as well as superior tumor control and increased overall survival in both low antigen density and clinically relevant patient-derived xenograft models. Increased activation and enhanced polyfunctionality led to enhanced efficacy. CONCLUSIONS: Showing for the first time that a membrane-proximal CAR is superior to a membrane-distal one in the setting of CD33 targeting, our results demonstrate the rationale for targeting membrane-proximal epitopes with high-affinity binders. We also demonstrate the importance of optimizing CAR T cells for functionality in settings of both low antigen density and clinically relevant patient-derived models.

Model-Based Antithymocyte Globulin in αßhaplo-Hematopoietic Stem Cell Transplantation Facilitates Engraftment, Expedites T Cell Recovery, and Mitigates the Risk of Acute Graft-versus-Host Disease.

In αß T-cell/CD19 B-cell depleted hematopoietic stem cell transplantation (αßhaplo-HSCT) recipients, antithymocyte globulin (ATG; Thymoglobulin) is used for preventing graft rejection and graft-versus-host disease (GVHD). The optimal dosing remains to be established, however. Here we present the first comparative analysis of 3 different ATG dosing strategies and their impact on immune reconstitution and GVHD. Our study aimed to evaluate the effects of 3 distinct dosing strategies of ATG on engraftment success, αß+ and γδ+ T cell immune reconstitution, and the incidence and severity of acute GVHD in recipients of αßhaplo-HSCT. This comparative analysis included 3 cohorts of pediatric patients with malignant (n = 36) or nonmalignant (n = 8) disease. Cohorts 1 and 2 were given fixed ATG doses, whereas cohort 3 received doses via a new nomogram, based on absolute lymphocyte count (ALC) and body weight (BW). Cohort 3 showed a 0% incidence of day 100 grade II-IV acute GVHD, compared to 48% in cohort 1 and 27% in cohort 2. Furthermore, cohort 3 (the ALC/BW-based cohort) had a significant increase in CD4+ and CD8+ naïve T cells by day 90 (P = .04 and .03, respectively). Additionally, we found that the reconstitution and maturation of γδ+ T cells post-HSCT was not impacted across all 3 cohorts. Cumulative ATG exposure in all cohorts was lower than previously reported in T cell-replete settings, with a lower pre-HSCT exposure (<40 AU*day/mL) correlating with engraftment failure (P = .007). Conversely, a post-HSCT ATG exposure of 10 to 15 AU*day/mL was optimal for improving day 100 CD4+ (P = .058) and CD8+ (P = .03) immune reconstitution without increasing the risk of relapse or nonrelapse mortality. This study represents the first comparative analysis of ATG exposure in αßhaplo-HSCT recipients. Our findings indicate that (1) a 1- to 2-fold ATG to ATLG bioequivalence is more effective than previously established standards, and (2) ATG exposure post-HSCT does not adversely affect γδ+ T cell immune reconstitution. Furthermore, a model-based ATG dosing strategy effectively reduces graft rejection and day 100 acute GVHD while also promoting early CD4+/CD8+ immune reconstitution. These insights suggest that further optimization, including more distal administration of higher ATG doses within an ALC/BW-based strategy, will yield even greater improvements in outcomes.

Progressive demyelinating polyneuropathy after hematopoietic cell transplantation in metachromatic leukodystrophy: a case series.

Metachromatic leukodystrophy (MLD) is a neuro-metabolic disorder due to arylsulfatase A deficiency, causing demyelination of the central and peripheral nervous system. Hematopoietic cell transplantation (HCT) can provide a symptomatic and survival benefit for pre-symptomatic and early symptomatic patients by stabilizing CNS disease. This case series, however, illustrates the occurrence of severely progressive polyneuropathy shortly after HCT in two patients with late-infantile, one with late-juvenile, and one with adult MLD, leading to the inability to walk or sit without support. The patients had demyelinating polyneuropathy before HCT, performed at the ages of 2 years in the first two patients and at 14 and 23 years in the other two patients. The myeloablative conditioning regimen consisted of busulfan, fludarabine and, in one case, rituximab, with anti-thymocyte globulin, cyclosporine, steroids, and/or mycophenolate mofetil for GvHD prophylaxis. Polyneuropathy after HCT progressed parallel with tapering immunosuppression and paralleled bouts of infection and graft-versus-host disease (GvHD). Differential diagnoses included MLD progression, neurological GvHD or another (auto)inflammatory cause. Laboratory, electroneurography and pathology investigations were inconclusive. In two patients, treatment with immunomodulatory drugs led to temporary improvement, but not sustained stabilization of polyneuropathy. One patient showed recovery to pre-HCT functioning, except for a Holmes-like tremor, for which a peripheral origin cannot be excluded. One patient showed marginal response to immunosuppressive treatment and died ten months after HCT due to respiratory failure. The extensive diagnostic and therapeutic attempts highlight the challenge of characterizing and treating progressive polyneuropathy in patients with MLD shortly after HCT. We advise to consider repeat electro-neurography and possibly peripheral nerve biopsy in such patients. Nerve conduction blocks, evidence of the presence of T lymphocytes and macrophages in the neuronal and surrounding nerve tissue, and beneficial effects of immunomodulatory drugs may indicate a partially (auto)immune-mediated pathology. Polyneuropathy may cause major residual disease burden after HCT. MLD patients with progressive polyneuropathy could potentially benefit from a more intensified immunomodulatory drug regime following HCT, especially at times of immune activation.

Persistent or New Cytopenias Predict Relapse Better than Routine Bone Marrow Aspirate Evaluations After Hematopoietic Cell Transplantation for Acute Leukemia or Myelodysplastic Syndrome in Children and Young Adult Patients.

The clinical value of serial routine bone marrow aspirates (rBMAs) in the first year after allogeneic hematopoietic cell transplantation (alloHCT) to detect or predict relapse of acute leukemia (AL) and myelodysplastic syndrome (MDS) in pediatric and young adult patients is unclear. The purpose of this analysis was to determine if assessment of minimal residual disease (MRD) by multiparameter flow cytometry (MFC, MFC-MRD) or donor chimerism (DC) in rBMAs or serial complete blood counts (CBCs) done in the year after alloHCT predicted relapse of AL or MDS in pediatric and young adult patients. We completed a retrospective analysis of patients with AL or MDS who had rBMAs performed after alloHCT between January 2012 and June 2018. Bone marrow (BM) was evaluated at approximately 3, 6, and 12 months for disease recurrence by morphology, MFC-MRD, and percent DC by short tandem repeat molecular testing. CBCs were performed at every clinic visit. The main outcome of interest was an assessment of whether MFC-MRD or DC in rBMAs or serial CBCs done in the year after alloHCT predicted relapse in AL or MDS pediatric and young adult patients. A total of 121 recipients with a median age of 13 years (range 1 to 32) were included: 108 with AL and, 13 with MDS. A total of 423 rBMAs (median 3; 0 to 13) were performed. Relapse at 2 years was 23% (95% CI: 16% to 31%) and at 5 years 25% (95% CI: 18% to 33%). One hundred fifty-four of 157 (98%) rBMAs evaluated for MRD by MFC were negative and did not preclude subsequent relapse. Additionally, low DC (<95%) did not predict relapse and high DC (≥95%) did not preclude relapse. For patients alive without relapse at 1 year, BM DC (P = .74) and peripheral T-cell DC (P = .93) did not predict relapse. Six patients with low-level T-cell and/or BM DC had a total of 8 to 20 BM evaluations, none of these patients relapsed. However, CBC results were informative for relapse; 28 of 31 (90%) relapse patients presented with an abnormal CBC with peripheral blood (PB) blasts (16 patients), cytopenias (9 patients), or extramedullary disease (EMD, 3 patients). Two patients with BM blasts >5% on rBMA had circulating blasts within 5 weeks of rBMA. Neutropenia (ANC <1.5 K/mcl) at 1 year was predictive of relapse (P = .01). Neutropenia and thrombocytopenia (<160 K/mcl) were predictive of disease-free survival (DFS) with inferior DFS for ANC <1.5 K/mcl, P = .001, or platelet count <160 K/mcl (P = .04). These results demonstrate rBMAs after alloHCT assessed for MRD by MFC and/or for level of DC are poor predictors for relapse in pediatric and young adult patients with AL or MDS. Relapse in these patients presents with PB blasts, cytopenias, or EMD. ANC and platelet count at 1-year were highly predictive for DFS.

Consensus guidelines for the monitoring and management of metachromatic leukodystrophy in the United States.

Metachromatic leukodystrophy (MLD) is a fatal, progressive neurodegenerative disorder caused by biallelic pathogenic mutations in the ARSA (Arylsulfatase A) gene. With the advent of presymptomatic diagnosis and the availability of therapies with a narrow window for intervention, it is critical to define a standardized approach to diagnosis, presymptomatic monitoring, and clinical care. To meet the needs of the MLD community, a panel of MLD experts was established to develop disease-specific guidelines based on healthcare resources in the United States. This group developed a consensus opinion for best-practice recommendations, as follows: (i) Diagnosis should include both genetic and biochemical testing; (ii) Early diagnosis and treatment for MLD is associated with improved clinical outcomes; (iii) The panel supported the development of newborn screening to accelerate the time to diagnosis and treatment; (iv) Clinical management of MLD should include specialists familiar with the disease who are able to follow patients longitudinally; (v) In early onset MLD, including late infantile and early juvenile subtypes, ex vivo gene therapy should be considered for presymptomatic patients where available; (vi) In late-onset MLD, including late juvenile and adult subtypes, hematopoietic cell transplant (HCT) should be considered for patients with no or minimal disease involvement. This document summarizes current guidance on the presymptomatic monitoring of children affected by MLD as well as the clinical management of symptomatic patients. Future data-driven evidence and evolution of these recommendations will be important to stratify clinical treatment options and improve clinical care.

Two novel assays demonstrate persistent daratumumab exposure in a pediatric patient with delayed engraftment following allogeneic hematopoietic stem cell transplantation.

BACKGROUND AIMS: Daratumumab, a human IgG monoclonal antibody targeting CD38, is a promising treatment for pediatric patients with relapsed or refractory T-cell acute lymphoblastic leukemia (T-ALL). We describe a case of delayed engraftment following a mismatched, unrelated donor hematopoietic stem cell transplant (HSCT) in a 14-year-old female with relapsed T-ALL, treated with daratumumab and chemotherapy. By Day 28 post-HSCT, the patient had no neutrophil engraftment but full donor myeloid chimerism. METHODS: We developed two novel, semi-quantitative, antibody-based assays to measure the patient's bound and plasma daratumumab levels to determine if prolonged drug exposure may have contributed to her slow engraftment. RESULTS: Daratumumab levels were significantly elevated more than 30 days after the patient's final infusion, and levels inversely correlated with her white blood cell counts. To clear daratumumab, the patient underwent several rounds of plasmapheresis and subsequently engrafted. CONCLUSIONS: This is the first report of both delayed daratumumab clearance and delayed stem cell engraftment following daratumumab treatment in a pediatric patient. Further investigation is needed to elucidate the optimal dosing of daratumumab for treatment of acute leukemias in pediatric populations as well as daratumumab's potential effects on hematopoietic stem cells and stem cell engraftment following allogenic HSCT.

Strategic infection prevention after genetically modified hematopoietic stem cell therapies: recommendations from the International Society for Cell & Gene Therapy Stem Cell Engineering Committee.

There is lack of guidance for immune monitoring and infection prevention after administration of ex vivo genetically modified hematopoietic stem cell therapies (GMHSCT). We reviewed current infection prevention practices as reported by providers experienced with GMHSCTs across North America and Europe, and assessed potential immunologic compromise associated with the therapeutic process of GMHSCTs described to date. Based on these assessments, and with consensus from members of the International Society for Cell & Gene Therapy (ISCT) Stem Cell Engineering Committee, we propose risk-adapted recommendations for immune monitoring, infection surveillance and prophylaxis, and revaccination after receipt of GMHSCTs. Disease-specific and GMHSCT-specific considerations should guide decision making for each therapy.

High-dose individualized antithymocyte globulin with therapeutic drug monitoring in high-risk cord blood transplant.

BACKGROUND: Graft-versus-host disease (GvHD) and rejection are main limitations of cord blood transplantation (CBT), more so in patients with severe inflammation or previous rejections. While rigorous T-cell depletion with antithymocyte globulin (ATG) is needed to prevent GvHD and rejection, overexposure to ATG leads to slow T-cell recovery after transplantation, especially in CBT. OBJECTIVE: To evaluate high-dose, upfront ATG with individualized dosing and therapeutic drug monitoring (TDM) in pediatric CBT for patients at high risk for GvHD and rejection. STUDY DESIGN: Heavily inflamed patients and patients with a recent history of rejection were eligible for individualized high-dose ATG with real-time TDM. The ATG dosing scheme was adjusted to target a post-CBT exposure of <10 AU*day/mL, while achieving a pre-CBT exposure of 60-120 AU*day/mL; exposure levels previously defined for optimal efficacy and safety in terms of reduced GvHD and rejection, respectively. Main outcomes of interest included efficacy (target exposure attainment) and safety (incidence of GvHD and rejection). Other outcomes of interest included T-cell recovery and survival. RESULTS: Twenty-one patients were included ranging from 2 months to 18 years old, receiving an actual median cumulative dose of ATG of 13.3 mg/kg (range 6-30 mg/kg) starting at a median 15 days (range 12-17) prior to CBT. Dosing was adjusted in 14 patients (increased in 3 and decreased in 11 patients). Eighteen (86%) and 19 (91%) patients reached the target pre-CBT and post-CBT exposure, respectively. Cumulative incidence for acute GvHD was 34% (95% CI 23-45) and 5% (95% CI 0-10%) for grade 2-4 and grade 3-4, respectively; cumulative incidence of rejection was 9% (95% CI 2-16%). Overall survival was 75% (95% CI 65-85%). CONCLUSION: Individualized high-dose ATG with TDM is feasible and safe for patients with hyperinflammation in a CBT setting. We observe high target ATG exposure attainment, good immune reconstitution (despite very high doses of ATG) and acceptable rates of GvHD and rejection.

Impact of rabbit anti-thymocyte globulin (ATG) exposure on outcomes after ex vivo T-cell-depleted hematopoietic cell transplantation in pediatric and young adult patients.

BACKGROUND AIMS: Traditional weight-based dosing of rabbit anti-thymocyte globulin (rATG) used in allogeneic hematopoietic cell transplantation (HCT) to prevent graft-versus-host disease (GVHD) and graft rejection leads to variable exposures. High exposures induce delayed CD4+immune reconstitution (CD4+IR) and greater mortality. We sought to determine the impact of rATG exposure in children and young adults receiving various types of EX-VIVO T-cell-depleted (EX-VIVO-TCD) HCT. METHODS: Patients receiving their first EX-VIVO-TCD HCT (CliniMACS CD34+, Isolex or soybean lectin agglutination), with removal of residual T cells by E-rosette depletion (E-) between 2008 and 2018 at Memorial Sloan Kettering Cancer Center were retrospectively analyzed. rATG exposure post-HCT was estimated (AU*d/L) using a validated population pharmacokinetic model. Previously defined rATG-exposures, <30, 30-55, ≥55 AU*d/L, were related with outcomes of interest. Cox proportional hazard and cause-specific models were used for analyses. RESULTS: In total, 180 patients (median age 11 years; range 0.1-44 years) were included, malignant 124 (69%) and nonmalignant 56 (31%). Median post-HCT rATG exposure was 32 (0-104) AU*d/L. Exposure <30 AU*d/L was associated with a 3-fold greater probability of CD4+IR (P < 0.001); 2- to 4-fold lower risk of death (P = 0.002); and 3- to 4-fold lower risk of non-relapse mortality (NRM) (P = 0.02). Cumulative incidence of NRM was 8-fold lower in patients who attained CD4+IR compared with those who did not (P < 0.0001). There was no relation between rATG exposure and aGVHD (P = 0.33) or relapse (P = 0.23). Effect of rATG exposure on outcomes was similar in three EX-VIVO-TCD methods. CONCLUSIONS: Individualizing rATG dosing to target a low rATG exposure post-HCT while maintaining total cumulative exposure may better predict CD4+IR, reduce NRM and increase overall survival, independent of the EX-VIVO-TCD method.

Inventory of current practices regarding hematopoietic stem cell transplantation in metachromatic leukodystrophy in Europe and neighboring countries.

BACKGROUND: For decades, early allogeneic stem cell transplantation (HSCT) has been used to slow neurological decline in metachromatic leukodystrophy (MLD). There is lack of consensus regarding who may benefit, and guidelines are lacking. Clinical practice relies on limited literature and expert opinions. The European Reference Network for Rare Neurological Diseases (ERN-RND) and the MLD initiative facilitate expert panels for treatment advice, but some countries are underrepresented. This study explores organizational and clinical HSCT practices for MLD in Europe and neighboring countries to enhance optimization and harmonization of cross-border MLD care. METHODS: A web-based EUSurvey was distributed through the ERN-RND and the European Society for Blood and Marrow Transplantation Inborn Errors Working Party. Personal invitations were sent to 89 physicians (43 countries) with neurological/metabolic/hematological expertise. The results were analyzed and visualized using Microsoft Excel and IBM SPSS statistics. RESULTS: Of the 30 countries represented by 42 respondents, 23 countries offer HSCT for MLD. The treatment is usually available in 1-3 centers per country (18/23, 78%). Most countries have no or very few MLD patients transplanted during the past 1-5 years. The eligibility criteria regarding MLD subtype, motor function, IQ, and MRI largely differ across countries. CONCLUSION: HSCT for MLD is available in most European countries, but uncertainties exist in Eastern and South-Eastern Europe. Applied eligibility criteria and management vary and may not align with the latest scientific insights, indicating physicians' struggle in providing evidence-based care. Interaction between local physicians and international experts is crucial for adequate treatment decision-making and cross-border care in the rapidly changing MLD field.

T-cell depleted allogeneic hematopoietic stem cell transplant for the treatment of Fanconi anemia and MDS/AML.

The only curative approach for myelodysplastic syndrome (MDS) or acute myeloid leukemia (AML) arising in patients with Fanconi anemia (FA) is allogeneic hematopoietic stem cell transplantation (HCT); however, HCT approaches are inconsistent and limited data on outcomes exist. We retrospectively evaluated outcomes of thirty patients with FA and MDS/AML who underwent first allogeneic HCT with a T-cell depleted (TCD) graft at our institution. Patients were transplanted on successive protocols with stepwise changes in cytoreduction and GVHD prophylaxis. All but two patients (93%) experienced durable hematopoietic engraftment. With median follow-up of 8.7 years, 5-year OS was 66.8% and DFS 53.8%. No significant differences in survival were found in patients with high-risk prognostic features (age ≥20 years, AML diagnosis, alternative donor graft) or when stratified by conditioning regimen. The 5-year cumulative incidences of relapse and NRM were 24.3% and 21.9%, respectively. NRM was higher in patients ≥20 years at HCT but did not otherwise differ. We herein demonstrate promising outcomes following allogeneic HCT for patients with FA and MDS/AML using TCD grafts, particularly in a cohort of high-risk patients with 50% ≥20 years and a majority receiving mismatched grafts. Future prospective studies are needed to compare this approach with other HCT platforms.

Predictors of outcomes in hematopoietic cell transplantation for Fanconi anemia.

Allogeneic hematopoietic cell transplantation (HCT) remains the only cure for the hematologic manifestations of Fanconi anemia (FA). We performed retrospective predictor analyses for HCT outcomes in FA for pediatric and young adult patients transplanted between 2007 and 2020 across three large referral institutions. Eighty-nine patients, 70 with bone marrow failure +/- cytogenetic abnormalities, 19 with MDS/AML, were included. Five-year overall survival (OS) was 83.2% and event-free survival (EFS) was 74%. Age ≥19, HLA mismatch and year of HCT were multivariable predictors (MVPs) for OS, EFS and treatment-related mortality (TRM). In the pediatric group, TCD was a borderline MVP (P = 0.059) with 5-year OS of 73.0% in TCD vs. 100% for T-replete HCT. The cumulative incidence of day 100 grade II-IV aGvHD and 5-year cGvHD were 5.6% and 4.6%, respectively. Relapse in the MDS/AML subgroup occurred in 4 patients (16%). Graft failure was seen in 9 patients (TCD 6/37 [16%]; T-replete 3/52 [5.7%]). Six patients developed malignancy after HCT. Survival chances after HCT for FA are excellent and associated with high engrafted survival and low toxicity. Age ≥19, HLA mismatch, year of transplant and 'TCD in the <19 years group' (although borderline) were found to be negative predictors for survival.

Longitudinal clinical data improve survival prediction after hematopoietic cell transplantation using machine learning.

ABSTRACT: Serial prognostic evaluation after allogeneic hematopoietic cell transplantation (allo-HCT) might help identify patients at high risk of lethal organ dysfunction. Current prediction algorithms based on models that do not incorporate changes to patients' clinical condition after allo-HCT have limited predictive ability. We developed and validated a robust risk-prediction algorithm to predict short- and long-term survival after allo-HCT in pediatric patients that includes baseline biological variables and changes in the patients' clinical status after allo-HCT. The model was developed using clinical data from children and young adults treated at a single academic quaternary-care referral center. The model was created using a randomly split training data set (70% of the cohort), internally validated (remaining 30% of the cohort) and then externally validated on patient data from another tertiary-care referral center. Repeated clinical measurements performed from 30 days before allo-HCT to 30 days afterwards were extracted from the electronic medical record and incorporated into the model to predict survival at 100 days, 1 year, and 2 years after allo-HCT. Naïve-Bayes machine learning models incorporating longitudinal data were significantly better than models constructed from baseline variables alone at predicting whether patients would be alive or deceased at the given time points. This proof-of-concept study demonstrates that unlike traditional prognostic tools that use fixed variables for risk assessment, incorporating dynamic variability using clinical and laboratory data improves the prediction of mortality in patients undergoing allo-HCT.