ALBERT

Description: Background: Umbilical cord blood transplantation (UCBT) is a potentially curative therapy acute leukemia (AL) patients. Transplantation benefit must be balanced against risks, such as transplant related mortality and relapse. The complex nature of hematopoietic stem cell transplantation data (HCT), rich in interactions and possibly nonlinear associations, has motivated us to apply machine learning (ML) for predictive modeling. ML is a field of artificial intelligence and is part of the data mining approach for data analysis. Our group has recently reported on a ML based prediction model for short term HCT outcomes (Shouval R et al; JCO 2015). Using a ML algorithm, the perspective of the current study was prediction of leukemia free survival (LFS) at 2 years after an UCBT, while exploring variables' importance and interactions. Patients & Methods: A cohort of 3,149 UCBT were analyzed. Inclusion criteria encompassed patients at all ages, undergoing an UCBT (single/double unit) in EBMT centers from the year 2004 to 2014, for AL, in all disease status. All conditioning and graft versus host disease prophylaxis regiments were included. A total of 24 variables were considered, including the number of total nucleated cell dose (TNC), donor and recipients HLA typing, as well as recipient, disease and transplant characteristics. The Random Survival Forest (RSF) ML algorithm was applied for model construction and data exploration. RSF is known to be adaptive to data, is able to automatically recover nonlinear effects and complex interactions among variables, and yields nonparametric prediction over test data. The analysis pipeline consisted of prediction model development, assessment of variable importance by their minimal depth from the tree trunk, and exploration of the top ranking variable with dependence plots. The latter promotes understanding of non-trivial associations between variables and outcomes. Results : The 2 years LFS was 49%, with a median follow up of 30 months. A RSF model of 1000 trees was developed, with each tree constructed on a bootstrap sample from the original cohort. A prediction error of 36.0% was calculated. The 10 most predictive variables (in ascending order) were disease status, age, TNC harvested and infused, recipient CMV serostatus, interval from diagnosis to UCBT, transplant year, previous autologous transplant, and use of anti-thymocyte globulin (ATG). Selected findings from exploration of variables-outcome relationship with dependence plots included a varying effect of TNCs in specific subpopulations. Increasing the number of infused TNCs had a positive effect on predicted LFS in patients receiving HLA mismatched (2 or more HLA mismatch) (figure) or single unit CB grafts, and patients in earlier disease status or older age. ATG administration was associated with worse LFS, whether unadjusted or adjusted to all other variables. However, there was an additional negative effect in advanced disease status patients, recipients of HLA mismatched or single CB units grafts, and older patients. Patients in 1st complete remission (CR) had higher predicted LFS as compared to those in 2nd CR. However, in patients receiving a HLA mismatched or a double CB graft, the difference in LFS between CR1 and CR2 was attenuated. Younger age had a favorable impact in early disease status, but lost its positive effect in advanced disease. Conclusions: A prediction model for LFS 2 years post UBCT was developed using the RSF ML algorithm. Variables were ranked according to their predictive contribution. Disease status, age, and TNC count were found to be the most important factors. Dependence plots revealed interactions and nonlinear associations between variables and the outcome, such as the effect of cell dose on HLA disparity. Apart from the study's clinical findings, it carries a methodological significance. A novel ML approach for prediction, variable selection and data exploration, accounting for long term time to event outcomes, has proved useful in the field of HCT. Figure 1. Variable marginal dependence coplot of predicted LFS at 2 years against TNC, conditional on HLA matching. Individual cases are marked with blue circles (alive or censored) and red `x's (event). Linear smooth (a linear extrapolation of the prediction function), with shaded 95% confidence band, indicates trends of variable dependence. Figure 1. Variable marginal dependence coplot of predicted LFS at 2 years against TNC, conditional on HLA matching. Individual cases are marked with blue circles (alive or censored) and red `x's (event). Linear smooth (a linear extrapolation of the prediction function), with shaded 95% confidence band, indicates trends of variable dependence. Disclosures Mohty: Janssen: Honoraria; Celgene: Honoraria. Sanz:JANSSEN CILAG: Honoraria, Research Funding, Speakers Bureau. Bader:Neovii: Other: Institutional grants; Medac: Other: Institutional grants; Riemser: Other: Institutional grants; Amgen: Consultancy; Novartis: Consultancy; Jazz Pharmaceuticals: Consultancy.

Description: Human cancers display substantial intratumoral genetic heterogeneity, which facilitates tumor survival under changing microenvironmental conditions. Tumor substructure and its effect on disease progression and relapse are incompletely understood. In the present study, a high-throughput method that uses neutral somatic mutations accumulated in individual cells to reconstruct cell lineage trees was applied to hundreds of cells of human acute leukemia harvested from multiple patients at diagnosis and at relapse. The reconstructed cell lineage trees of patients with acute myeloid leukemia showed that leukemia cells at relapse were shallow (divide rarely) compared with cells at diagnosis and were closely related to their stem cell subpopulation, implying that in these instances relapse might have originated from rarely dividing stem cells. In contrast, among patients with acute lymphoid leukemia, no differences in cell depth were observed between diagnosis and relapse. In one case of chronic myeloid leukemia, at blast crisis, most of the cells at relapse were mismatch-repair deficient. In almost all leukemia cases, 〉 1 lineage was observed at relapse, indicating that diverse mechanisms can promote relapse in the same patient. In conclusion, diverse relapse mechanisms can be observed by systematic reconstruction of cell lineage trees of patients with leukemia.

Description: Background Allo-HSCT has been shown to increase survival and improve cure in acute leukemia (AL). However, this procedure is accompanied by high rates of morbidity and mortality. Several risk scores based on conventional statistical methods may aid decision regarding whom and how to perform allo-HSCT, but these methods carry inherent limitations, which may lead to sub-optimal candidate selection. Machine learning (ML) is a field in computer science stemming from artificial intelligence and is part of the data mining approach for data analysis. ML algorithms are commonly applied in technological and commercial settings. They allow for coping with complex data scenarios and thus may be suitable for outcome prediction in allo-HSCT. With this background, and using a ML prediction method- the alternating decision tree (ADT) algorithm, we developed an interpretable model for overall mortality (OM) and treatment-related mortality (TRM) at day +100 after allo-HSCT in AL. Patients and Methods 28,995 adult allo-HSCT recipients from the registry of the ALWP of EBMT were analyzed. Twenty two variables were available including year of transplant (range, 2000-2011), diagnosis (Acute Myeloid Leukemia and Acute Lymphoblastic Leukemia), disease status, Karnofsky performance status, conditioning regimen (myeloablative or reduced- intensity conditioning), graft type (peripheral blood, bone marrow or cord blood), donor and recipient HLA compatibility, CMV serostatus, GVHD prophylaxis regimens, etc. Per study definitions, the primary outcomes to be predicted were OM and TRM at day +100 days after allo-HSCT. The complete dataset was split into 3 sets: training set (n=11,600), testing set (n=8,688) and validation set (n=8,707). The ADT prediction model was tested and optimized according to the first 2 subgroups and validated on the last one. Output from the ADT model, included variables selection with assigned scores in a tree-based structure and area under the ROC curve (AUC), a measure for model discrimination Results Each of the ADT models selected 12 out 22 variables for prediction of OM and TRM at day +100. Ten variables were mutual for both prediction models, although different weights were assigned. These included: age, diagnosis, disease status, Karnofsky performance status (all at time of transplant), donor-recipient HLA-matching, number of transplants in each center per year, year of transplant, conditioning regimen and the donor's and patient's CMV serostatus. Variables selected exclusively by the OM prediction model were graft type and donor-patient CMV serostatus match, whereas the TRM model selected time from diagnosis to transplant and donor-recipient sex match. Applying the models on the validation set yielded AUCs of 0.701 (95% confidence interval [CI] 0.691-0.710) for OM prediction and 0.67 (95% [CI] 0.66-0.68) for TRM. The ADT prediction models assigned scores correlating with patient outcome. Patients in each of the validation sets were grouped according to their score range and the prediction success. A Higher score was correlated with higher rate of the measured outcome in both models (figure 1 and figure 2). Conclusions We present two new models, based on the ADT ML algorithm, for prediction of OM and TRM at day +100 after allo-HSCT. The models are robust as they rely on a high number of samples and a large validation set. As shown in the figures, higher scores correlated with a poorer outcome, reaching more than 50% mortality for a score range of 5.76-7 in the OM prediction model. The AUC performance measure was better for OM than TRM, possibly due to a higher event rate in former, making it easier to predict. Improving the predictive ability will probably necessitate evaluation of more variables, as the limitation of the maximal predictive performance is most likely in the information gained from the variables and not from the sample size or algorithm used. This is currently under progress, especially combination with other risk scores linked to comorbidities. In summary, our models can aid candidate selection for allo-HSCT, by providing a measurable score that correlates with transplant success. Disclosures: Schmid: Novartis: Honoraria, Research Funding, travel grant Other; Roche: travel grant, travel grant Other; MSD: Honoraria.

Description: Background Allogeneic hematopoietic stem cell transplantation (HSCT) is a potentially curative therapy for many hematological malignancies, however high mortality associated with the procedure remains a significant obstacle. Predictors of intermediate- and long-term treatment-related mortality (TRM) have been established, including two well-validated scoring systems (HCT-CI, Sorror 2005; EBMT, Gratwohl 2012). These scores reflect overall comorbidity burden (HCT-CI) and disease status together with specifics of the transplant procedure (EBMT). Early mortality post-HSCT may be related closely to the patient's immediate physiological state at the time of HSCT. We hypothesized that clinical laboratory values taken at the time of admission for HSCT may give a useful representation of physiological age and fitness for transplant, and thus be highly predictive of early (100 day) and very-early (30 day) treatment-related mortality (TRM). Methods This was a retrospective study on a cohort of patients who underwent HSCT at a single center for any indication between 2000 and 2014. Hematology and chemistry laboratory data were collected from the time of admission for transplantation, prior to the initiation of conditioning regimens. Variables representing bone marrow function (neutrophils, platelets, hemoglobin), kidney function (creatinine clearance estimated by EPI), liver function (total bilirubin, AST, ALT), nutritional status (albumin) as well as age were included in a Fine & Gray Competing Risk regression at time-points of 30, 100 and 365 days. Cumulative incidence curves were constructed for variables deemed significant at each time-point. Results A total of 1,045 patients were included in this study. The median patient age was 52 (IQR 39-60). Myeloablative conditioning was used for 274 patients (26.2%) while reduced-intensity conditioning was used for 765 patients (73.2%). 568 (54.4%) patients received transplants from sibling donors, and 477 (45.6%) patients received grafts from unrelated donors. Risk associated with individual laboratory biomarkers changed over the course of the first year post-transplantation. Creatinine clearance, albumin and total bilirubin were strong predictors of early mortality (p 〈 0.01) at 30 days, with hazard ratios of 4.5, 3 and 2.6 respectively. Notably, the prognostic weight of these variables decreased over time. In contrast, the prognostic role of age became evident only later in the first year (table 1). Cumulative incidence of TRM in patients with creatinine clearance 〈 60 mL/min (fig. 1a) and with albumin 〈 3.5 g/dL (fig. 1b) showed markedly poorer outcomes over time, with especially strong effect in the immediate aftermath of transplantation. Conclusions Our study demonstrates that common laboratory biomarkers of physiological fitness and biological age are strong determinants of early transplantation-related mortality. These markers are complimentary to existing prognostic models, which utilize broader comorbidity categories (HCT-CI) or specifics of the transplantation procedure (EBMT score) to predict medium and long-term survival trends. Laboratory variables are readily-available objective measures, and may be of particular importance when assessing fitness for the initial transplantation procedure. Table 1 Hazard ratios of the key laboratory markers at 30, 100 and 365 days post-transplantation Table 1. Hazard ratios of the key laboratory markers at 30, 100 and 365 days post-transplantation Figure 1 Cumulative incidence curves stratified by (a) estimated creatinine clearance and (b) albumin pre-transplantation Figure 1. Cumulative incidence curves stratified by (a) estimated creatinine clearance and (b) albumin pre-transplantation Disclosures No relevant conflicts of interest to declare.

Description: Background: Toxicities following allogeneic hematopoietic stem cell transplantation are specific to the conditioning regimen used and its intensity. Therefore, we hypothesized that the hazard of individual comorbidities is dependent on the conditioning agents and their dose. The aim of this analysis was to study, in a regimen specific manner, the impact of individual comorbidities (cardiac disease, severe pulmonary disease, and diabetes mellitus) on mortality. Methods: We included a multi-center cohort of 3,338 patients from the registry of the Acute Leukemia Working Party of the European Society of Blood and Marrow Transplantation (EBMT), who underwent first allogeneic HSCT for treatment of AML in all disease stages, transplanted between 2005 and 2016. Patients included received grafts from a matched sibling or 10/10 HLA-allele matched unrelated donor, and who were conditioned using one of 7 common regimens (Busulfan [BU[/Cyclophposphamide [Cy], Cy/Total Body Irradiation [TBI], Flu [Fludarabine]/Bu at reduced-intensity [RIC] dosage, Flu/Bu at myeloablative [MAC] dosage, Flu/Melphalan [Mel], Flu/Treosulfan [Treo], Flu/TBI). Regimens were excluded from a given analysis if fewer than ten overall mortality events occurred among patients with or without the studied condition. We constructed a multivariable Cox model for the outcome of overall survival, adjusted for key transplantation variables, including patient age, disease status, and donor type among others. For each comorbidity, separate models were constructed within each of the seven regimens, and the comorbidity's hazard ratio was extracted. Using the same method, the hazard ratio across all regimens, as well as in the myeloablative and the reduced-intensity setting, was obtained. Results: The median age was 55 years; 70% of patients were in first complete remission (CR), and 54% of patients received grafts from matched sibling donors. The most common regimens studied were Flu/Bu at a reduced-intensity dose (22%), Flu/Mel (20%) and Bu/Cy (16%, Table 1). We find that the studied comorbidities were associated with different degrees of added risk for overall mortality in each regimen studied. For cardiac disease, hazard ratios (HR) ranged from 1.00 (95% CI: 0.63-1.60) in Flu/Treo to 1.65 (1.15-2.36) in Flu/Bu with myeloablative dose (Figure 1). Among patients with severe pulmonary disease, significant increases in hazard were seen only in patients treated with myeloablative Flu/Bu and Flu/Mel. In diabetes mellitus, no regimens were clearly associated with increased risk of overall mortality. However, Flu+TBI trended strongly toward increased risk with an HR of 1.55 (1.00-2.41, p = 0.051). Greater comorbidity-associated risk was not consistently associated with increasing conditioning, and the hazards associated with each comorbidity in the MAC and RIC settings were broadly overlapping. Similar trends were seen for the outcome of non-relapse mortality, although statistical significance (p 〈 0.05) was not observed in the setting of low event numbers. Conclusions: These results confirm our hypothesis that comorbidities exert an effect on transplantation outcome in a conditioning regimen-specific manner. Additional study, both retrospective, and prospective, may eventually allow for the precision selection of a conditioning regimen based on the individual patient's physiological status. Disclosures Finke: Medac: Consultancy, Honoraria, Other: travel grants, Research Funding; Neovii: Consultancy, Honoraria, Other: travel grants, Research Funding; Riemser: Consultancy, Honoraria, Research Funding; Novartis: Consultancy, Honoraria, Other: travel grants, Research Funding. Malladi:Roche: Membership on an entity's Board of Directors or advisory committees. Mohty:MaaT Pharma: Consultancy, Honoraria.

Description: Background: While intensive consolidation therapy with autologous stem cell transplantation (ASCT) can secure a remission in selected Acute Myeloid Leukemia (AML) patients with intermediate-risk cytogenetics, a substantial proportion will ultimately relapse. Knowledge of the mutational status of FMS like tyrosine kinase 3 internal tandem duplication (FLT3-ITD) and nucleophosmin (NPM) 1 and their possible combinations could further refine a subset of intermediate cytogenetic risk patients who may benefit from ASCT. However, data are limited. We, therefore, set out to evaluate the impact of FLT3-ITD and NPM1 in a large cohort of patients undergoing ASCT. Methods: This was a retrospective analysis of the Acute Leukemia Working Parity of the European Society for Blood and Marrow Transplantation (EBMT) registry. We included 405 de-novo AML patients, from 51 European centers, with intermediate-risk cytogenetics (Grimwade et al., Blood 2010) and complete data on FLT3-ITD and NPM1 status, receiving an ASCT at first complete remission (CR1), between 2000-2014. Leukemia free survival (LFS) was the primary outcome. Secondary outcomes were overall survival (OS), transplantation-related mortality (TRM), and relapse incidence (RI). The latter two were considered as competing events. Univariate and multivariable Cox regression models, adjusted for recipient sex, age, Karnofsky performance status, FLT3/NPM1 combinations, days from diagnosis to transplantation, stem cell source -bone marrow or peripheral blood (PB), and use of total-body irradiation-based conditioning. Results: Patients included had a median age of 52 years and received an autograft at median of 5 months from diagnosis. PB-based autograft and non-TBI conditioning were used in the majority of patients (93% and 90%, respectively). FLT3-/NPM1- was the leading molecular combination (50%), followed by FLT3-/NPM1+ (30%), FLT3+/NPM1+ (11%), and FLT3+/NPM1- (9%). Age, time from diagnosis to transplantation, graft source, and use of TBI were similar between the molecular subgroups. The median year of transplantation was earlier in NPM1- patients (FLT3+/NPM1- 2008, FLT3-/NPM1- 2009, FLT3-/NPM1+ 2010, FLT3+/NPM1+ 2011, p

Description: Background: Haploidentical (Haplo) stem cell transplantation (SCT) provide a curative option for nearly all Acute Myeloid Leukemia (AML) patients lacking an HLA matched donor. However, outcomes following Haplo-SCT vary and are dependent on a number of individual features. Integrative prognostic models for decision support towards a Haplo-SCT are lacking. We sought to develop a prediction model of Leukemia-Free Survival (LFS) for AML patients undergoing a Haplo-SCT. Methods: A total of 1,804 de-novo (80%) and secondary (20%) AML patients who received a non-T-cell depleted Haplo-SCT between the years 2005-2017 were included. All patients were reported to the registry of the Acute Leukemia Working Party (ALWP) of the European Society for Blood and Marrow Transplantation (EBMT). To account for non-linear associations, violation of the proportional hazard assumption, and to reduce bias associated with feature selection, a non-parsimonious non-parametric machine learning algorithm, Random Survival Forest (RSF), was used. RSF provides a continuous probabilistic estimation of LFS by fitting an ensemble of decision trees. Variables included in the model were reflective of patient, disease, and transplantation characteristics. Since RSF models are not readily interpretable (i.e., "black box" models) variable importance (VIMP) of covariates included in the model (Xv), were assessed by calculating the difference in prediction error before and after permuting Xv. The model's generalizability and accuracy were tested through repetitive bootstrapping (5000 iterations) and calculation of the C-index. Results: The median age of the patients was 53 years. The majority had an early disease status (complete remission [CR] 1[44%]) with intermediate cytogenetic risk (43%) and were undergoing allogeneic transplantation for the first time (93%). Reduced-intensity conditioning (RIC) was used in 57% of cases, and grafts were from peripheral blood in 54% of transplants. For graft-versus-host disease (GvHD) prophylaxis, 82% of the patients received post-transplant cyclophosphamide (PTCy) and 18% anti-thymocyte globulin (ATG). The median follow-up duration was 2.0 years. In the RSF prediction model, the top-ranking variables (Figure A) were disease status, GvHD prophylaxis, time from diagnosis to transplantation, and age. Bootstrapped C-index of the prediction model was 0.66. Prognostic discrimination was assessed by dividing the predicted LFS probabilities into quartiles that were then used to plot Kaplan-Meier curves, demonstrating LFS ranging from 24.8%-60.1% at 2-years (Figure B). Differing features of the four prognostic groups are listed in the Table. Conclusions: Our group has developed the first prediction model for LFS in AML patients treated with a Haplo-SCT. The model is based on a machine learning technique and provides an individualized estimation of LFS probability. It is conceivable that once this model is verified, it could serve as an important clinical tool when considering a patient to Haplo-SCT. Figure. Figure. Disclosures Angelucci: Vertex Pharmaceuticals Incorporated (MA) and CRISPR CAS9 Therapeutics AG (CH): Other: Chair DMC; Jazz Pharmaceuticals Italy: Other: Local ( national) advisory board; Celgene: Honoraria, Other: Chair DMC; Novartis: Honoraria, Other: Chair Steering Comiittee TELESTO Protocol; Roche Italy: Other: Local (national) advisory board. Tischer:Jazz Pharmaceuticals: Other: Jazz Advisory Board. Mohty:MaaT Pharma: Consultancy, Honoraria.

Description: Introduction Chimeric Antigen Receptor (CAR) T cells are associated with unique toxicities, including cytokine release syndrome (CRS) and immune effector cells-associated neurotoxicity syndrome (ICANS). Patients (pts) with severe CRS and ICANS exhibit hemodynamic instability and coagulopathy with evidence of endothelial activation and increased blood brain barrier permeability. Increases in inflammatory cytokines and biomarkers of endothelial activation in serum and CSF have been associated with severe CRS and ICANS. The EASIX (Endothelial Activation and Stress Index) score [lactate dehydrogenase (LDH) (U/L) × creatinine (mg/dl) / platelets (PLT) (109 cells/L)] correlates with severe fluid overload and survival in allogeneic transplant pts. Elevated LDH and low PLT levels have been associated with severe ICANS development, and high IL-6 levels are seen in severe CRS and ICANS. We hypothesized that the EASIX and a newly proposed version of it, the modified-EASIX (mEASIX), in which creatinine is replaced by CRP (mg/dL) as an easily available surrogate for IL6, would be associated with CRS and ICANS in CAR T cells pts. Methods We analyzed 2 different populations of adult CAR T cells pts treated at our institution: 1) B-cell acute lymphoblastic leukemia (B-ALL) pts treated with CD1928z CAR T cells from 2010 to 2016 (NCT01044069), and 2) aggressive diffuse large B-cell lymphoma (DLBCL) pts treated with axicabtagene ciloleucel (axi-cel) or tisagenlecleucel (tisa-cel) after FDA approval starting from 2018. EASIX and mEASIX scores were calculated for each patient daily from start of lymphodepletion conditioning to day +14. A log transformation using base 2 (log2) was applied to all EASIX/mEASIX variables to reduce skew. CRS and ICANS were graded according to the ASTCT grading system. Results 87 pts, B-ALL (n=53) and DLBCL (n=34), were analyzed. According to ASTCT grading, 83% (72/87) experienced CRS and 54% (47/87) developed ICANS, grade ≥3 in 23% (20/87) and 40% (35/87) of pts, respectively. When analyzed by disease, CRS and ICANS rates were 87% (46/53) and 55% (29/53) for B-ALL and 76% (26/34) and 53% (18/34) for DLBCL, respectively. CRS and ICANS were grade ≥3 in 28% (15/53) and 45% (24/53) of B-ALL pts and in 15% (5/34) and 32% (11/34) of DLBCL pts, respectively. Median time of onset of CRS after CAR T cell infusion was day +2 and median onset of ICANS was day +6 for the overall population and the subgroups. High EASIX and mEASIX scores at start of conditioning were both associated with development of any grade CRS [OR=1.81 (95% CI 1.09-3.36) p=0.038 and OR=1.94 (95% CI 1.32-3.38) p=0.005] and grade ≥3 CRS [OR=1.47 (95% CI 1.05-2.29) p=0.049 and OR=1.34 (95% CI 1.07-1.80) p=0.024], respectively (Table). Following CAR T cell infusion, high scores of both EASIX [OR=1.60 (95% CI 1.12-2.43) p=0.017] and mEASIX [OR=1.32 (95% CI 1.07-1.69) p=0.014] on day +1 were associated with development of grade ≥3 CRS. Moreover, both high EASIX [OR=1.43 (95% CI 1.08-1.96) p=0.018] and mEASIX [OR=1.29 (95% CI 1.07-1.60) p=0.010] scores on day +3 were associated with grade ≥3 ICANS. When analyzed by disease, results were confirmed for severe CRS and ICANS in B-ALL patients, while in the DLBCL group only mEASIX at start of conditioning and at day +1 was associated with development of any grade CRS. EASIX and mEASIX scores were not associated with response rates to CAR T cells therapy. Conclusions EASIX and mEASIX scores calculated at baseline (before lymphodepletion) are associated with development of CRS and severe CRS. Moreover, both high EASIX and mEASIX scores on day +1 and day +3 are associated with occurrence of grade ≥3 CRS and grade ≥3 ICANS, respectively. We conclude that EASIX and mEASIX, as markers of endothelial damage and inflammation, could be useful as early predictors in guiding treatment decisions before the onset of severe symptoms. Table Disclosures Batlevi: Juno Therapeutics: Consultancy, Membership on an entity's Board of Directors or advisory committees. Brentjens:JUNO Therapeutics: Consultancy, Patents & Royalties, Research Funding; Celgene: Consultancy. Giralt:Miltenyi: Research Funding; Spectrum Pharmaceuticals: Consultancy; Novartis: Consultancy; Jazz Pharmaceuticals: Consultancy; Celgene: Consultancy, Research Funding; Johnson & Johnson: Consultancy, Research Funding; Takeda: Consultancy, Research Funding; Amgen: Consultancy, Research Funding; Actinium: Consultancy, Research Funding; Kite: Consultancy. Palomba:Pharmacyclics: Membership on an entity's Board of Directors or advisory committees; Noble Insights: Consultancy; Seres Therapeutics: Equity Ownership, Membership on an entity's Board of Directors or advisory committees; STRAXIMM: Membership on an entity's Board of Directors or advisory committees; Kite Pharmaceuticals: Membership on an entity's Board of Directors or advisory committees; Evelo: Equity Ownership; MSK (IP for Juno and Seres): Patents & Royalties; Hemedicus: Speakers Bureau; Merck & Co Inc.: Consultancy. Santomasso:Kite/Gilead: Consultancy; Juno/Celgene: Consultancy; Novartis: Consultancy. Sauter:GSK: Consultancy; Kite/Gilead: Consultancy; Celgene: Consultancy; Juno Therapeutics: Consultancy, Research Funding; Sanofi-Genzyme: Consultancy, Research Funding; Spectrum Pharmaceuticals: Consultancy; Novartis: Consultancy; Genmab: Consultancy; Precision Biosciences: Consultancy. Scordo:Angiocrine Bioscience, Inc.: Consultancy; McKinsey & Company: Consultancy. Shah:Amgen: Research Funding; Janssen Pharmaceutica: Research Funding. Park:Allogene: Consultancy; Amgen: Consultancy; AstraZeneca: Consultancy; Autolus: Consultancy; GSK: Consultancy; Incyte: Consultancy; Kite Pharma: Consultancy; Novartis: Consultancy; Takeda: Consultancy. Perales:Takeda: Honoraria, Membership on an entity's Board of Directors or advisory committees; Abbvie: Honoraria, Membership on an entity's Board of Directors or advisory committees; Bellicum: Honoraria, Membership on an entity's Board of Directors or advisory committees; Bristol-Meyers Squibb: Honoraria, Membership on an entity's Board of Directors or advisory committees; Incyte: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; Servier: Membership on an entity's Board of Directors or advisory committees; Kyte/Gilead: Research Funding; Miltenyi: Research Funding; Novartis: Honoraria, Membership on an entity's Board of Directors or advisory committees; Omeros: Honoraria, Membership on an entity's Board of Directors or advisory committees; Nektar Therapeutics: Honoraria, Membership on an entity's Board of Directors or advisory committees; Merck: Consultancy, Honoraria; Medigene: Membership on an entity's Board of Directors or advisory committees; NexImmune: Membership on an entity's Board of Directors or advisory committees; MolMed: Membership on an entity's Board of Directors or advisory committees.

Description: Background: Risk stratification is critical for balancing the potential benefit and harm of allogeneic hematopoietic stem cell transplantation (HSCT). However, our ability to do so is suboptimal and better markers are warranted. Lactate dehydrogenase (LDH) is a readily available biomarker with a well-established prognostic role in the front-line therapy setting of hematological malignancies. Few reports in selected diseases have studied the utility of LDH as a surrogate for pre-transplant risk-stratification. We, therefore, sought to systematically study the prognostic role of LDH in patients with various hematological malignancies undergoing allogeneic - HSCT. Methods : This was a retrospective study on a cohort of patients who underwent allogeneic-HSCT at a single center between 2000 and 2016. LDH levels in the day before conditioning initiation were extracted from the electronic medical record, as well as features related to patients, disease, donor, and transplant. The primary outcome was overall survival (OS), and secondary outcomes included non-relapse mortality (NRM) and relapse (considered as competing events). The median follow-up time was 59 months (IOR 27 - 112). Using Kaplan-Meier estimates or cumulative incidence curves for competing events, we analyzed the association between median LDH levels, per disease category, with the detailed outcomes. Associations meeting a p-value below 0.05 for the primary outcome were further explored in a multivariable Cox regression (cause-specific Cox for competing events). The model was adjusted for age, disease risk index (DRI), year of HSCT, donor type, and conditioning regimens. Results A total of 1,250 adult (median age 52.8 years) patients were included in this study. Acute myeloid leukemia (AML) was the leading transplantation induction (n=594; 48%), followed by non-Hodgkin lymphoma (NHL, 211, 16.9%), acute lymphoblastic leukemia (ALL, n=175, 14%), myelodysplastic syndrome (MDS, n=165, 13%), and plasma cell dyscrasia (PCD, n=105, 8%). The majority of patients had intermediate-risk disease (44%), sibling or match unrelated donor (51% and 31% respectively) and received myeloablative or reduced-toxicity conditioning regimens (27% and 36%, respectively). In the univariate analysis, AML patients with higher than median LDH levels (〉193 IU/L) had lower 3-year OS (51.9% vs. 39.2%, p

Description: Background: Steroid-resistant (SR) intestinal acute graft versus host disease (aGVHD) is a devastating complication of allogeneic hematopoietic stem cell transplantation. Preliminary reports suggest that fecal microbiota transplantation (FMT) administered through a nasogastric tube or colonoscopy may be an effective treatment. We report the results of a single-arm pilot study (NCT 03214289) using FMT in capsules to treat SR or steroid dependent (SD) intestinal aGVHD. Methods: The primary outcome was the occurrence of severe adverse events (SAEs) at 28 days post last FMT course. Secondary outcomes included GVHD response. Complete response (CR) was defined as resolution of gastrointestinal symptoms or reduction of steroid dose to 5 mg of prednisone. Partial response was defined as a decrease in severity of GVHD by at least one stage or a ≥40% reduction in steroid dose. Patients were eligible if they had SR or SD gut aGVHD without active infection or neutropenia. Per-protocol, participants received a course of 30 frozen capsules of fecal matter over two consecutive days. FMT courses could be repeated from the same or a different donor, at the treating physician's discretion. Capsules are produced from healthy unrelated donors who underwent vigorous screening. They are taken orally and are flavorless and odor-free. To characterize the impact of the FMT on the gut microbiota, stool samples of recipients were serially collected and underwent 16s rRNA sequencing. Results: To date, we have enrolled 7 patients with intestinal aGVHD (6 SR, 1 SD) (Table). The median dose of methylprednisolone (MP) was 1 mg/kg (interquartile range [IQR] 0.8-1.3 mg/kg). FMT was administered at a median of 39 days (IQR 21-58 days) from aGVHD diagnosis. A total of 15 courses of FMT were given. Patients received a range of 1-3 FMT courses (median 2). The capsules were well tolerated. Patient #1 developed Enterococcus Faecium bacteremia 2 days following the second FMT. To track the source of bacteremia, we performed targeted metagenomic sequencing. The enterococcus strain from the blood culture was identified in the recipient's pre-FMT stool sample but not in the FMT inoculum (i.e., capsule), confirming that the bacteremia was not an FMT complication. Similarly, patient #6 developed Pseudomonas aeruginosa bacteremia 3 days after the 2nd FMT. 16s rRNA sequencing of the donor capsule failed to demonstrate Pseudomonas taxa. No other SAEs suspected to be related to the FMT were observed. Two patients achieved a CR with complete resolution of GVHD symptoms. Patient #6 had a partial improvement following the 1st FMT, with a reduction of MP from 2 mg/kg to 1.3 mg/kg. Three days after the 2nd FMT, she developed fatal pseudomonas bacteremia, not related to the FMT as detailed above. At last follow-up (median 61 days, IQR 40-99), 3/7 patients were alive. Three patients died from consequences of active GVHD, while one patients who responded to FMT and was free of GVHD, succumbed to an invasive Aspergillus infection of the brain. 16s rRNA sequencing of stool samples revealed bacterial domination (i.e., occupation of at least 40% of the microbiota by a single predominating taxon) of Escherichia(E) coli in four patients before FMT, with a major reduction following therapy. FMT was associated with the introduction of new bacteria and an increase in bacterial diversity in the recipient's stool (Figure). Conclusions: We demonstrate for the first time the utility of fecal microbiota transplantation in orally administered capsules for the treatment of severe intestinal acute GVHD. The capsules were well tolerated and safe. Metagenomic sequencing proved that a bacterial infection following FMT was not related to the procedure. Sequencing of the stool sample revealed bacterial domination with E.coli in 4/7 patients prior to the first FMT. Following FMT, bacterial diversity increased. Finally, 2/7 patients attained a complete response following therapy, suggesting a potential role of FMT in patient management. Figure. (A) Heatmap of operational taxonomics units (OTU). Each column marks a sequenced stool sample at a specific time point and rows individual taxas. The color code indicates relative abundance. Dotted lines represent an FMT course. Before FMT all patients, aside from patient #6, had markedly reduced diversity, with enrichment of OTUs following treatment. (B). Change of bacterial diversity, measured by the Shanon diversity index before and after FMTs. Figure. Figure. Disclosures No relevant conflicts of interest to declare.