Citations

Key Points Systemic immune microenvironment signatures in CMML indicate an altered T- and natural killer cell balance. CMML bone marrow dendritic cell aggregates associate with disease progression and systemic regulatory T-cell phenotypic switch.

Introduction: Mutations in the gene isocitrate dehydrogenase 2 (IDH2) occur in ~12-15% of patients with acute myeloid leukemia (AML). Enasidenib, an oral small molecule selective targeted inhibitor of IDH2 enzymes, was approved forIDH2mutated relapsed/refractory AML (RR-AML) in 2017 (Stein et. al., 2017). There is limited data regarding its efficacy and toxicity outside the context of clinical trials. Using Mayo Clinic's cohort of AML patients, we provide a real-world look at outcomes forIDH2mutated RR-AML patients treated with enasidenib. Methods: A total of 13 patients treated with enasidenib for RR-AML were identified, all of which experienced at least one relapse following induction chemotherapy. Response assessment was based on European Leukemia Net (ELN) 2017 response criteria.Complete blood counts including blast percentage and transfusion dependency were obtained at AML diagnosis, initiation of enasidenib, 2 months and 6 months following initiation of therapy. Transfusion dependency was defined by the need for either a red cell or platelet transfusion within 4 weeks of the designated time frame. We compared the clinical characteristics of patients who achieved complete response (CR) (n=4) to those not achieving CR (n=9), using the Mann-Whitney U test for numerical variables and the Fischer's Exact test for quantitative data. Results: The median age at diagnosis of our 13 patients was 68 years (range 60-82), with a male predilection (n=11) (Table 1). Cytogenetic and molecular genetic findings are provided in Table 2. Complete response (CR) was achieved in 4 (30.7%) patients by cycle 2 and 2 of those patients (50%) proceeded to an allogeneic hematopoietic stem cell transplant (AHSCT). The median time from diagnosis to initiation of treatment for patients in CR was 3.5 months (range 1-7), and for patients not achieving CR was 13 months (range 6-53; p=0.048), with median time of follow up after treatment initiation of 23.5 months (range, 5-36) and 7 months (range, 1-14), respectively. Among 8 patients who were transfusion dependent at baseline, all the CR patients (2/2) achieved red blood cell and platelet transfusion independence 6 months after treatment, instead only (3/6) 50% of those not in CR achieved transfusion independence (p=0.028). An initial trajectory of improvement in counts is noted from the time of diagnosis to the 2 month mark post treatment with 75% of the responder cohort showing improvement in counts which was durable through the 6 month mark, whereas 55% of those that did not respond had worsening counts at month 2 and died prior to 6 month follow up. At the time of data cut-off, the median duration of CR was 14 months (range, 2-33), with 3 of 4 (75%) of responders maintaining CR. Figure 1 details overall responses and outcomes in all 13 patients. None of the patients in our study experienced differentiation syndrome. Five patients were hospitalized in the course of treatment for infectious complications. Two patients were noted to have hyperbilirubinemia however there was no clear indication that this was secondary to enasidenib Median overall survival for all 13 patients was 21 months (range, 6-53 months) with leukemia free survival of 7 months (range, 5-19 months). Furthermore, median overall survival for patients in CR was 27 months (range, 6-42), and those not in CR was 16 months (range, 6-53) (p=0.20). In addition, we observed that 75% of patients who attained CR are alive compared to 11.1% of those not in CR (p = 0.052). Conclusion: We demonstrate a superior CR rate of 30.7% with enasidenib amongstIDH2mutated RR-AML patients compared with clinical trial data showing CR rate of 19.6% (Stein et. al., 2019) despite our patients being heavily pre-treated with median number of two prior therapies (range; 1-5) including prior AHSCT in 2 patients. Moreover, a subset of patients (2 of 4 in CR) were able to undergo ASCHT. Finally, the achievement of CR at 2 months after initiation of therapy serves as an appropriate time point for response assessment. Disclosures No relevant conflicts of interest to declare.

Introduction Clonal cytopenias of undetermined significance (CCUS) is defined by the presence of somatic driver mutations/copy number variations in hematopoietic cells in patients with low blood counts, in the absence of morphological evidence for myeloid neoplasms. Patients with CCUS are often symptomatic and can be transfusion dependent (TD), with high rates of progression to myeloid neoplasms, especially myelodysplastic syndromes (MDS). In addition, similar to MDS, there are reports of CCUS patients responding to hypomethylating agents. Due to the lack of a formal diagnosis/morphological dysplasia, CCUS patients are often denied therapies or clinical trial enrollment. We carried out this study to validate our hypothesis that CCUS shares similar clinical and survival characteristics with lower risk MDS (LR-MDS). Methods CCUS patients were prospectively identified from the clonal hematopoiesis clinic, whereas LR-MDS patients were retrospectively identified from our institutional database. LR-MDS was defined as very low, low and intermediate risk MDS based on the R-IPSS categorization. Baseline demographics, blood counts, bone marrow (BM) morphology, cytogenetics, and NGS results were abstracted. Transfusion dependency was defined as requiring at least one unit of red cell or platelets every 4 weeks. The Mann-Whitney-U and Fischer's exact test were used to compare quantitative and qualitative data in subgroups. Kaplan-Meier overall survival (OS) estimates were used for survival analysis and compared using the log-rank test. Results 186 patients were included in the study; 74 (40%) with CCUS and 112 (60%) with LR-MDS, median age 66 years, with 63% being male (Table 1). In the CCUS group, 93% had one or more mutations (67% >1 mutation) detected by NGS, while 7% had clonal cytogenetic abnormalities. Common mutations in CCUS included TET2 (30%), SRSF2 (20%), DNMT3A (13%) and ASXL1 (11%); with 28% of patients being red cell TD and 15% being platelet TD. On application of the R-IPSS stratification, 40%, 45%, 12% and 3% were in the very low, low, intermediate and high risk categories, respectively. LR-MDS subtypes included MDS-RS (50%), MDS-MLD (21%), MDS-EB (10%), MDS del5q (7%), and MDS-U (12%). SF3B1 mutations were seen in 67% (95% of MDS-RS), while TET2 and DNMT3A mutations were seen in 29% and 21%, respectively. 60% of LR-MDS patients had >1 mutation. 44% were red cell TD, while 10% were platelet TD. In comparison to patients with CCUS, LR-MDS patients were more likely to have higher white blood cell counts (p=0.002), higher neutrophil counts (p=0.009), higher platelet counts (p<0.001), be red cell TD (p=0.031), have BM ring sideroblasts (p<0.001), have higher BM blast% (p=0.004) and carry SF3B1 mutations (p<0.001); whereas CCUS patients were more likely to have higher monocyte counts (p=0.009) and more likely to carry IDH1 (p=0.005), ZRSR2 (p=0.005) and ATM (p=0.012) mutations. Importantly there were no differences with regards to R-IPSS cytogenetic risk groups, R-IPSS prognostic categories, platelet TD, and mutations involving TET2, DNMT3A, SRSF2 and ASXL1 and TP53. Notably signal pathway mutations and mutations involving tumor suppressor genes were infrequent (<5%) in both groups. The LR-MDS patients had a longer median follow up (53 vs 15 months) and at last follow up 72 (64%) and 11 (15%) deaths had been documented in the LR-MDS and CCUS groups respectively. 13 (18%) CCUS patients progressed to MDS (10) and AML (3) over a median of 15 months, while 9 (8%) LR-MDS patients progressed to higher grade MDS (1) and AML (8), respectively (median follow up 53 months). There was no difference in the median survival between CCUS (median OS not reached) versus LR-MDS (median OS 8.3 years) (p=0.372, Figure 1). Conclusion In spite of subtle phenotypic/molecular differences between LR-MDS and CCUS, both entities had similar prognostication, distribution of high risk mutations and survival outcomes. Within a short follow-up, 18% of CCUS patients progressed to MDS/AML, indicating that regardless of the absence of dysplasia these entities are a continuum of myeloid neoplasms. In order to improve access to treatments and clinical trial opportunities, we recommend that close consideration be given to consider CCUS as a MDS subtype. Disclosures Shah: Dren Bio: Consultancy.

Magnesium is an essential element that is involved in the regulation of cell growth, division and differentiation. Epidemiologic studies have demonstrated that a diet poor in magnesium is associated with an increased risk of developing cancer. There are few studies investigating whether magnesium deficiency is associated with a poor prognosis in lymphomas. The recent discovery of X-linked immunodeficiency with magnesium defect, Epstein-Barr virus, and neoplasia (XMEN) disease, suggests a role of magnesium in the development of hematologic malignancies and lymphocyte function. Furthermore, a recently published study found that hypomagnesemia prior to chemotherapy is associated with inferior overall survival (OS) in patients with Burkitt Lymphoma. We hypothesized that low magnesium levels prior to autologous hematopoietic stem cell transplant (AHSCT) may lead to inferior outcomes via impaired lymphocyte recovery. We evaluated whether low serum magnesium levels were associated with inferior survival in patients with relapsed or refractory diffuse large B cell lymphoma (DLBCL) who underwent autologous hematopoietic stem cell transplant (AHSCT) between January 1998 and May 2020 using a retrospective study design. We cross-referenced our search with patients who had a magnesium level as part of their pre-transplant workup between 7 and 28 days prior to their first transplant. Two-sided Wilcoxon rank sum test and Chi square/Fischer's exact test were used to compare the continuous and categorical variables, respectively. Kaplan-Meier and log-rank tests were used to perform all time to event analysis. Hazard ratios (HR) with confidence intervals (CI) were calculated using Cox-proportional hazards. Our study included 581 patients of which 82 (14.1%) had hypomagnesemia (magnesium < 1.7 mg/dL) prior to AHSCT. Our study found that hypomagnesemia was associated with inferior OS and event-free survival (EFS) defined as re-transplant, relapse or death, compared to those with a normal magnesium level; median OS: 7.3 years (95% CI: 2.91 - upper limit not estimable) versus 9.7 years (95% CI: 6.92 - 12.3 years), p=0.0041; and median EFS: 3.9 years (95% CI: 1.63 - 8.98 years) versus 6.29 years (95% CI: 4.73 - 8.95 years), p = 0.016 respectively(B). Higher serum magnesium (HR 0.199 per doubling, 95% CI 0.078 - 0.503; p<0.001) and higher albumin (HR 0.090 per doubling, 95% CI 0.021 - 0.394; p=0.001) were associated with lower mortality rates, and higher creatinine (HR 1.669 per doubling, 95% CI 1.03 - 2.52; p=0.04) was associated with greater mortality up to 1 year post-transplantation. In a multivariate Cox regression excluding albumin, hypomagnesemia (p < 0.001) and elevated creatinine (p = 0.038) were independent predictors of poor overall survival(A). There was no significant difference between day 15 absolute lymphocyte count between patients with normal magnesium levels and low serum magnesium levels prior to their conditioning regimen. Although we hypothesized that we would find a significant difference between 15 day absolute lymphocyte count recovery and hypomagnesemia, this was not apparent. However, magnesium is an important nutrient for immune function and immune recovery is important after the bone marrow is restored post ablative therapy. Thus, low serum magnesium levels may lead to impaired OS and EFS via impaired lymphocyte function. In vitro studies will be necessary to understand this mechanism further. Magnesium may simply be a marker of illness and replacement of magnesium may not improve outcomes. However, further investigations are needed to determine whether outcomes would improve with replacement of magnesium in patients with hypomagnesemia, which our study did not address. Overall, our preliminary data suggests that hypomagnesemia is an independent prognostic factor and predictor of inferior OS and EFS in this cohort of patients. Figure Disclosures Ansell: ADC Therapeutics:Research Funding;AI Therapeutics:Research Funding;Trillium:Research Funding;Affimed:Research Funding;Regeneron:Research Funding;Seattle Genetics:Research Funding;Bristol Myers Squibb:Research Funding;Takeda:Research Funding.Witzig:Spectrum:Consultancy;Celgene:Consultancy, Research Funding;Immune Design:Research Funding;Karyopharm Therapeutics:Research Funding;Acerta:Research Funding;Incyte:Consultancy;AbbVie:Consultancy;MorphSys:Consultancy.

Introduction Germline predisposition syndromes (GPS) are inherited disorders associated with germinal aberrations that increase the risk of malignancies. While aberrations in certain genes increase the risk for all types of malignancies (Tp53, ATM, CDKN2A, CHEK2), there is a growing list of genes associated specifically with hematological malignancies (GATA2, RUNX1, DDX41, ETV6, ANKRD26). At our institution, we have established a hematology GPS clinic to diagnose and manage GPS and with this report, detail our experience with 130 patients. Methods GPS were investigated in pediatric and adult patients with one or more first degree relatives with hematological/visceral malignancies or in those with antecedent thrombocytopenia (ANKRD26, RUNX1, ETV6), or with specific syndromic features (short telomere syndromes/STS, GATA2 haploinsufficiency, Fanconi anemia/FA, Shwachman-Diamond syndrome/SDS). Depending on the phenotype, specific functional assays such as flow-FISH for telomere length assessment and chromosomal breakage assays were ordered. After informed consent and genetic counselling, germline testing was carried out on peripheral blood mononuclear cell, skin fibroblast, or hair follicle-derived DNA. A custom-designed marrow failure NGS panel (200 genes) was used in most cases and interrogation of variants, in silico studies, and functional assays were carried out as previously described (Mangaonkar et al MC Proc 2019). Copy number variations were identified by aCGH. At the time of progression/worsening cytopenias, bone marrow/lymph node biopsies and NGS (next generation sequencing) were carried out where indicated. Results 130 patients with germline predisposition have been identified to date. The spectrum of disorders seen include STS 29 (22%), FA 17 (13%), GATA2 16 (12%), Diamond Blackfan anemia/DBA 13 (10%), RUNX1-FPD 12 (9%), ATM deletions/mutations 11 (8%), ANKRD26 6 (5%), SDS 5 (4%), DDX41 4 (3%), MPL 3 (2%), CHEK2, MECOM, Tp53 mutations 2 (2%) each, and CBL, CEPBA, ELANE, NF1, CDKN2A, CSF3R, ETV6, and GATA1 mutations, 1 (1%) each. Evidence for clonal evolution (CCUS) and hematological malignancies were seen in 51 (39%) patients, involving all the aforementioned genes/syndromes with the exception of DBA, CBL, ETV6, MPL, CSF3R, and GATA1. Seven (64%) of 11 patients with germline ATM deletions/mutations developed lymphoid malignancies; homozygous ATM (Follicular NHL-1, Burkitt lymphoma-1, T-ALL-1, T-LPD-1) and heterozygous ATM (T-PLL-1, DLBCL-1, CLL-1). Clonal evolution occurred in 11 (69%) of 16 GATA2 haploinsufficient patients (CCUS-2, MDS-3, CMML-1, AML-5) and in 7 (58%) of 12 RUNX1-FPD patients (CCUS-1, MDS-1, MDS/MPN-3, AML-2). Five of 29 (17%) STS patients had clonal progression (CCUS-2, MDS-2, AML-1), and 5 (29%) of 17 FA patients progressed to MDS-2 or AML-3. JMML was seen in one patient with a germline NF1 mutation, while 1 (20%) of 5 SDS patients progressed to AML. NGS data at progression was available in 24 (55%) of 44 myeloid/CCUS progressions, with somatic truncating ASXL1 mutations being most frequent (29%), followed by RAS pathway mutations (15%). AML/MDS progressions in STS, FA, and SDS were universally associated with complex/monosomal karyotypes, translating to refractory disease. Seventeen (39%) of 44 patients with myeloid predisposition underwent allogenic HCT (GATA2-7, FA-3, RUNX1-FPD-3, STS-2, NF1-1, Tp53-1), with 10 (59%) being alive at last follow up (Table 1). Conclusion We demonstrate the spectrum of germline aberrations associated with predisposition to hematological malignancies and outline the phenotypic heterogeneity of clonal transformation. The advent of NGS allows identification of clonal progression earlier than morphological changes, with mutations in ASXL1 and RAS pathway genes being commonly implicated. This study supports the universal development of dedicated germline predisposition clinics. Disclosures Pruthi: CSL Behring: Honoraria; Genentech Inc.: Honoraria; Bayer Healthcare: Honoraria; HEMA Biologics: Honoraria; Instrumentation Laboratory: Honoraria; Merck: Honoraria.

Background: Chronic myelomonocytic leukemia (CMML) is a rare chronic myeloid malignancy with a poor prognosis. There are few epidemiologic studies on CMML. We studied population-based outcomes in CMML using the Surveillance, Epidemiology, and End Results (SEER) and the National Cancer Database (NCDB). Methods: The SEER 18 registries and NCDB Participant User File were used to identify patients with ICD-O-3 diagnosis code 9945/3 from 2004-2015. Incidence was identified using SEER and age-adjusted to the U.S. 2000 standard population. Causes of death were obtained and CMML-related death was defined as death from any myeloid disorder (CMML, acute monocytic leukemia, acute myeloid leukemia [AML], chronic myeloid leukemia, other myeloid/monocytic leukemia, and aleukemic, subleukemic, and NOS) to avoid misattribution. Relative survival (RS) was defined as the ratio of the proportion of observed survivors in a cohort of CMML patients to the proportion of expected survivors in a comparable set of individuals that did not have CMML, adjusting for the general survival of the US population for race, sex, age, and time when the diagnosis was established. Time to AML was calculated using the left-truncated life tables session and a 3 month latency period was used to prevent misattribution. Per the NCDB's system on classifying treatment, hydroxyurea, decitabine, and azacitidine are considered chemotherapy. The Kaplan-Meier method was used to calculate overall survival (OS), and Cox regression model was used to identify predictors of survival. Variables significant in univariate analysis (age, Charlson Deyo score [CDS], insurance type, and treatment facility type) were included in a multivariate analysis. Statistical analyses were performed using SAS version 9.0. Results: In the SEER database (n=4437), the median age at diagnosis was 76 years (interquartile range [IQR] 68-83) and 2,783 patients (63%) were male. Incidence rates (1 case/1,000,000 individuals) were as follows: overall 4.4, male 6.6, female 2.9, Non-Hispanic White 4.9, and Non-Hispanic Black 3.3. The incidence did not significantly change from 2004 to 2015. With a median follow up of 5.8 years (95% CI: 5.5-6.3), the median OS was 1.3 years (95% CI: 1.3-1.4). 3635 patients (82%) died. Among those who died, 2016 patients (55%) deaths were CMML-related. When comparing CMML patients to the US general population, 3,156 patients were matched to the expected survival tables. In the general U.S. population, the expected survival for 12, 24, 36, 48 and 60 months was 95.6%, 91.3%, 87.1%, 83.1%, and 79.2%, respectively. In contrast, the RS for patients with CMML at the same time points was 63.6%, 46.2%, 35.2%, 28.8%, and 23.1% (Figure 1). 229 patients (5.2%) progressed to AML. The median time to AML was 1.2 years (IQR 0.6-2.3). In NCDB (n=6403), the median age at diagnosis was 74 years (IQR 66-82). With the median follow up of 6 years (95% CI 3.5-9), the median OS was 1.3 years (95% CI 0.4-3.3). The distribution of CDS score 0, 1, and >1 was 4518 (71%), 1193 (18%), and 692 (11%) respectively. 4687 (77%) had government insurance, while 1399 (23%) had private insurance. 3750 (60%) were treated at a non-academic center. The year of diagnosis was associated with improved OS (HR 0.99, 95% CI: 0.98-0.99). Patients who were diagnosed in 2012-2015 had improved OS compared to patients diagnosed in 2004-2007 (HR 0.89, 95% CI 0.83-0.95) with OS at 1- and 5- years being 60% and 16% (2012-2015) vs. 56% and 17% (2008-2011) vs. 53% and 15% (2004-2007) respectively. 3029 (48%) patients received chemotherapy, while 270 (4%) patients received stem cell transplant as the first-line therapy. The median time to chemotherapy from diagnosis was 19 days (IQR 6-43). OS at 1-, 5-, and 10-years was 57%, 18%, and 4% (no chemotherapy), 56%, 14%, and 7% (received chemotherapy), 79%, 44%, and 38% (received stem cell transplant), and 56%, 15%, 5% (no stem cell transplant). OS curves are shown in Figure 2. Factors independently predicting inferior OS were age ≥ 65 years at the time of diagnosis (p<0.001), CDS ≥ 1 (p<0.001), government insurance (p<0.001), and treatment at a non-academic center (p<0.001) (Table). Conclusion: Despite an improvement in survival over the years, CMML continues to be a cause of significant morbidity and mortality. Older age, CDS ≥ 1, government insurance, and treatment at non-academic facilities were predictive of inferior survival. Disclosures Shah: Dren Bio: Consultancy.

Background. Experience with sequential hematopoietic stem cell transplant (HSCT) and kidney transplant (KT) is limited. Methods. We conducted a retrospective observational study of adult patients who underwent both HSCT and KT at our center, with a median follow-up of 11 y. Results. In our 54 patients cohort (94% autologous HSCT), 36 (67%) patients received HSCT first followed by KT, while 18 (33%) received KT before HSCT. In both groups, AL amyloidosis represented 50% of hematologic diagnosis. Only 4 patients expired due to hematologic disease relapse (2 patients in each group) and only 3 allografts were lost due to hematologic disease recurrence (HSCT first n = 1 and KT first n = 2). Overall 1, 5, and 10 y death-censored graft survival rates were 94%, 94%, and 94%, respectively, for the HSCT first group and 89%, 89%, and 75%, respectively, for the KT first group. Overall 1, 5, and 10 y patients survival rates were 100%, 97% and 90%, respectively, for the HSCT first group and 100%, 76%, and 63%, respectively, for the KT first group. Conclusions. Our study supports safety of sequential KT and HSCT, with improved overall patient survival compared to recipients of HSCT remaining on dialysis and good long-term kidney allograft outcome.

AbstractVenetoclax and hypomethylating agent (HMA) combination therapy is FDA‐approved for elderly or unfit acute myeloid leukemia (AML) patients unable to withstand intensive chemotherapy. The primary objective of the current study was to impart our institutional experience with the above regimen, outlining response, survival outcomes, and its determinants amongst 86 treatment‐ naïve and relapsed/refractory AML patients. A total of 44 treatment‐naïve AML patients, median age 73.5 years, enriched with secondary, therapy related and ELN adverse risk disease (n = 27) were studied. The CR/CRi rates of 50% (22 of 44 patients) were superior to 23% in a matched AML cohort treated with HMA alone (P = .005). Response rates were similar with TP53, FLT3, NPM1 and IDH mutations (P = .31). Moreover, CEPBA mutations (P = .03) and neutropenia (P = .05) emerged as predictors of complete response. Survivalwas prolonged in patients achieving CR/CRi (17 vs 3 months without CR/CRi, P < .001; conversely adverse ELN risk portended inferior survival. Amongst 42 relapsed/refractory AML patients, half received ≥2 prior therapies excluding transplant, and 15 (35.7%) had received HMA. A group of 14 patients (33.3%) attained CR/CRi; age > 65 years, AML with myelodysplasia, JAK2, DNMT3A, and BCOR mutations predicted complete response. Survival distinctions were based on CR/CRi (median survival 15 vs 3 months with/without CR/CRi; P < .001), and TP53 mutation status (P = .04). In summary, we corroborate existing reports demonstrating superior response and prolonged survival with venetoclax and HMA in treatment ‐naïve and relapsed/refractory AML patients regardless of genotype. Additionally, we identify unique predictors of response to therapy which require validation.

BackgroundExtracorporeal photopheresis (ECP) is an immunomodulatory cellular therapy which has been shown to induce a tolerogenic state in patients with acute and chronic graft‐vs‐host disease. ECOG‐ACRIN explored the activity of ECP as a part of a reduced intensity conditioning regimen in two multicenter trials in patients with MDS (E1902) and lymphomas (E1402). While both studies closed before completing accrual, we report results in 23 patients (17 MDS and 6 lymphoma).Study Design and MethodsPatients received 2 days of ECP followed by pentostatin 4 mg/m2 /day for two consecutive days, followed by 600 cGy of total body irradiation prior to stem cell infusion. Immunosuppression for aGVHD was infusional cyclosporine A or tacrolimus and methotrexate on day +1, +3, with mycophenolate mofetil starting on day 100 for chronic GVHD prophylaxis.ResultsAll patients engrafted, with median time to neutrophil and platelet engraftment of 15‐18 days and 10‐18 days respectively. Grade 3 or 4 aGVHD occurred in 13% and chronic extensive GVHD in 30%.ConclusionsThese studies demonstrate that ECP/pentostatin/TBI is well tolerated and associated with adequate engraftment of neutrophils and platelets in patients with lymphomas and MDS.

AbstractTransplant‐associated thrombotic microangiopathy (TA‐TMA) has a wide range of presentations after hematopoietic stem‐cell transplantation (HSCT). We retrospectively studied the risk factors and outcomes of patients with early (≤day 100) and late (>day 100) TA‐TMA. Among the 1451 HSCT recipients, early TA‐TMA occurred in 45 (3.1%) patients at a median of 27 (3‐91) days, and late TA‐TMA in 39 (2.7%) patients at a median of 303 (122‐2595) days. Patients with early TA‐TMA were more likely to have high blood calcineurin‐inhibitor levels (P < .001) and acute graph‐vs‐host disease (GVHD, P < .001), while late TMA patients were more likely to have chronic GVHD (P < .001). The estimated median overall survival after onset of TMA for the entire cohort was 6 months. The estimated median overall survival was not reached in patients with an improvement of TMA vs 2 months in patients with no improvement (P < .001). In the early TMA group, older age (for every 10 years, HR 1.40; 95% CI 1.00‐1.94; P = .049) and bacterial infection (HR 2.42; 95% CI 0.98‐6.00; P = .056) were positively associated with mortality. Switching to MMF treatment (HR 0.40; 95% CI 0.16‐0.99; P = .047) and improvement of TMA (HR 0.08; 95% CI 0.03‐0.25; P < .001) were negatively associated with mortality in the multivariate analysis. In the late TMA group, the improvement of TMA was the only independent predictor associated with a lower risk of death (HR 0.05; 95% CI 0.02‐0.19; P < .001). Mortality rates in both early and late TMA remain unacceptably high. Future studies are needed for early diagnosis, trigger identifications, and use of targeted treatments.