In this issue of Blood, Rowan et al demonstrate that driver mutations for human T-cell leukemia virus (HTLV)–associated adult T-cell leukemia lymphoma (ATLL) can be identified years before the clinical manifestations of disease are apparent.

In fit patients with newly diagnosed acute myeloid leukemia (AML), immediate treatment start is recommended due to the poor prognosis of untreated acute leukemia. We explored the relationship between time from diagnosis to treatment start (TDT) and prognosis in a large real-world data set from the German Study Alliance Leukemia-Acute Myeloid Leukemia (SAL-AML) registry. All registered non-acute promyelocytic leukemia patients with intensive induction treatment and a minimum 12 months of follow-up were selected (n = 2263). We analyzed influence of TDT on remission, early death, and overall survival (OS) in univariable analyses for each day of treatment delay, in groups of 0 to 5, 6 to 10, 11 to 15, and >15 days of TDT, adjusted for influence of established prognostic variables on outcomes. Median TDT was 3 days (interquartile range, 2-7). Unadjusted 2-year OS rates, stratified by TDT of 0 to 5, 6 to 10, 11 to 15, and >15 days, were 51%, 48%, 44%, and 50% (P = .211). In multivariable Cox regression analysis accounting for established prognostic variables, the TDT hazard ratio as a continuous variable was 1.00 (P = .617). In OS analyses, separately stratified for age ≤60 and >60 years and for high vs lower initial white blood cell count, no significant differences between TDT groups were observed. Our study suggests that TDT is not related to survival. As stratification in intensive first-line AML treatment evolves, TDT data suggest that it may be a feasible approach to wait for genetic and other laboratory test results so that clinically stable patients are assigned the best available treatment option. This trial was registered at www.clinicaltrials.gov as #NCT03188874.

Patients with coronavirus disease 2019 (COVID-19) have elevated D-dimer levels. Early reports describe high venous thromboembolism (VTE) and disseminated intravascular coagulation (DIC) rates, but data are limited. This multicenter retrospective study describes the rate and severity of hemostatic and thrombotic complications of 400 hospital-admitted COVID-19 patients (144 critically ill) primarily receiving standard-dose prophylactic anticoagulation. Coagulation and inflammatory parameters were compared between patients with and without coagulation-associated complications. Multivariable logistic models examined the utility of these markers in predicting coagulation-associated complications, critical illness, and death. The radiographically confirmed VTE rate was 4.8% (95% confidence interval [CI], 2.9-7.3), and the overall thrombotic complication rate was 9.5% (95% CI, 6.8-12.8). The overall and major bleeding rates were 4.8% (95% CI, 2.9-7.3) and 2.3% (95% CI, 1.0-4.2), respectively. In the critically ill, radiographically confirmed VTE and major bleeding rates were 7.6% (95% CI, 3.9-13.3) and 5.6% (95% CI, 2.4-10.7), respectively. Elevated D-dimer at initial presentation was predictive of coagulation-associated complications during hospitalization (D-dimer >2500 ng/mL, adjusted odds ratio [OR] for thrombosis, 6.79 [95% CI, 2.39-19.30]; adjusted OR for bleeding, 3.56 [95% CI, 1.01-12.66]), critical illness, and death. Additional markers at initial presentation predictive of thrombosis during hospitalization included platelet count >450 × 109/L (adjusted OR, 3.56 [95% CI, 1.27-9.97]), C-reactive protein (CRP) >100 mg/L (adjusted OR, 2.71 [95% CI, 1.26-5.86]), and erythrocyte sedimentation rate (ESR) >40 mm/h (adjusted OR, 2.64 [95% CI, 1.07-6.51]). ESR, CRP, fibrinogen, ferritin, and procalcitonin were higher in patients with thrombotic complications than in those without. DIC, clinically relevant thrombocytopenia, and reduced fibrinogen were rare and were associated with significant bleeding manifestations. Given the observed bleeding rates, randomized trials are needed to determine any potential benefit of intensified anticoagulant prophylaxis in COVID-19 patients.

Publisher's Note: There is a Blood Commentary on this article in this issue.

Additional sex combs-like 1 (ASXL1), an epigenetic modulator, is frequently mutated in myeloid neoplasms. Recent analyses of mutant ASXL1 conditional knockin (ASXL1-MT-KI) mice suggested that ASXL1-MT alone is insufficient for myeloid transformation. In our previous study, we used retrovirus-mediated insertional mutagenesis, which exhibited the susceptibility of ASXL1-MT-KI hematopoietic cells to transform into myeloid leukemia cells. In this screening, we identified the hematopoietically expressed homeobox (HHEX) gene as one of the common retrovirus integration sites. In this study, we investigated the potential cooperation between ASXL1-MT and HHEX in myeloid leukemogenesis. Expression of HHEX enhanced proliferation of ASXL1-MT-expressing HSPCs by inhibiting apoptosis and blocking differentiation, whereas it showed only modest effect in normal HSPCs. Moreover, ASXL1-MT and HHEX accelerated the development of RUNX1-ETO9a and FLT3-ITD leukemia. Conversely, HHEX depletion profoundly attenuated the colony-forming activity and leukemogenicity of ASXL1-MT-expressing leukemia cells. Mechanistically, we identified MYB and ETV5 as downstream targets for ASXL1-MT and HHEX by using transcriptome and chromatin immunoprecipitation-next-generation sequencing analyses. Moreover, we found that expression of ASXL1-MT enhanced the binding of HHEX to the promoter loci of MYB or ETV5 via reducing H2AK119ub. Depletion of MYB or ETV5 induced apoptosis or differentiation in ASXL1-MT-expressing leukemia cells, respectively. In addition, ectopic expression of MYB or ETV5 reversed the reduced colony-forming activity of HHEX-depleted ASXL1-MT-expressing leukemia cells. These findings indicate that the HHEX-MYB/ETV5 axis promotes myeloid transformation in ASXL1-mutated preleukemia cells.

Yap1 and its paralogue Taz largely control epithelial tissue growth. We have identified that hematopoietic stem cell (HSC) fitness response to stress depends on Yap1 and Taz. Deletion of Yap1 and Taz induces a loss of HSC quiescence, symmetric self-renewal ability, and renders HSC more vulnerable to serial myeloablative 5-fluorouracil treatment. This effect depends on the predominant cytosolic polarization of Yap1 through a PDZ domain-mediated interaction with the scaffold Scribble. Scribble and Yap1 coordinate to control cytoplasmic Cdc42 activity and HSC fate determination in vivo. Deletion of Scribble disrupts Yap1 copolarization with Cdc42 and decreases Cdc42 activity, resulting in increased self-renewing HSC with competitive reconstitution advantages. These data suggest that Scribble/Yap1 copolarization is indispensable for Cdc42-dependent activity on HSC asymmetric division and fate. The combined loss of Scribble, Yap1, and Taz results in transcriptional upregulation of Rac-specific guanine nucleotide exchange factors, Rac activation, and HSC fitness restoration. Scribble links Cdc42 and the cytosolic functions of the Hippo signaling cascade in HSC fate determination.

Hematopoietic cell transplantation (HCT) from HLA-mismatched unrelated donors can cure life-threatening blood disorders, but its success is limited by graft-versus-host disease (GVHD). HLA-B leaders encode methionine (M) or threonine (T) at position 2 and give rise to TT, MT, or MM genotypes. The dimorphic HLA-B leader informs GVHD risk in HLA-B-mismatched HCT. If the leader influences outcome in other HLA-mismatched transplant settings, the success of HCT could be improved for future patients. We determined leader genotypes for 10 415 patients receiving a transplant between 1988 and 2016 from unrelated donors with one HLA-A, HLA-B, HLA-C, HLA-DRB1, or HLA-DQB1 mismatch. Multivariate regression methods were used to evaluate risks associated with patient leader genotype according to the mismatched HLA locus and with HLA-A, HLA-B, HLA-C, HLA-DRB1, or HLA-DQB1 mismatching according to patient leader genotype. The impact of the patient leader genotype on acute GVHD and mortality varied across different mismatched HLA loci. Nonrelapse mortality was higher among HLA-DQB1-mismatched MM patients compared with HLA-DQB1-mismatched TT patients (hazard ratio, 1.35; P = .01). Grades III to IV GVHD risk was higher among HLA-DRB1-mismatched MM or MT patients compared with HLA-DRB1-mismatched TT patients (odds ratio, 2.52 and 1.51, respectively). Patients tolerated a single HLA-DQB1 mismatch better than mismatches at other loci. Outcome after HLA-mismatched transplantation depends on the HLA-B leader dimorphism and the mismatched HLA locus. The patient's leader variant provides new information on the limits of HLA mismatching. The success of HLA-mismatched unrelated transplantation might be enhanced through the judicious selection of mismatched donors for a patient's leader genotype.

T-cell acute lymphoblastic leukemia (T-ALL) is an aggressive malignancy that accounts for ∼20% of ALL cases. Intensive chemotherapy regimens result in cure rates >85% in children and <50% in adults, warranting a search of novel therapeutic strategies. Although immune-based therapies have tremendously improved the treatment of B-ALL and other B-cell malignancies, they are not yet available for T-ALL. We report here that humanized, non-Fcγ receptor (FcγR)-binding monoclonal antibodies (mAbs) to CD3 have antileukemic properties in xenograft (PDX) models of CD3+ T-ALL, resulting in prolonged host survival. We also report that these antibodies cooperate with chemotherapy to enhance antileukemic effects and host survival. Because these antibodies show only minor, manageable adverse effects in humans, they offer a new therapeutic option for the treatment of T-ALL. Our results also show that the antileukemic properties of anti-CD3 mAbs are largely independent of FcγR-mediated pathways in T-ALL PDXs.

Anti-CD19 chimeric antigen receptor (CAR) T cells showed significant antileukemic activity in B-precursor acute lymphoblastic leukemia (ALL). Allogeneic, HLA-mismatched off-the-shelf third-party donors may offer ideal fitness of the effector cells, but carry the risk of graft-versus-host disease. Knockout (KO) of the endogenous T-cell receptor (TCR) in CD19-CAR-T cells may be a promising solution. Here, we induced a CRISPR/Cas9-mediated KO of the TCRβ chain in combination with a second-generation retroviral CAR transduction including a 4-1BB costimulatory domain in primary T cells. This tandem engineering led to a highly functional population of TCR-KO-CAR-T cells with strong activation (CD25, interferon γ), proliferation, and specific killing upon CD19 target recognition. TCR-KO-CAR-T cells had a balanced phenotype of central memory and effector memory T cells. KO of the endogenous TCR in T cells strongly ablated alloreactivity in comparison with TCR-expressing T cells. In a patient-derived xenograft model of childhood ALL, TCR-KO-CAR-T cells clearly controlled CD19+ leukemia burden and improved survival in vivo. However, coexpression of endogenous TCR plus CAR led to superior persistence of T cells and significantly prolonged leukemia control in vivo, confirmed by a second in vivo model using the leukemia cell line NALM6. These results point toward an essential role of the endogenous TCR for longevity of the response at the price of alloreactivity. In conclusion, anti-CD19 CAR T cells with a CRISPR/Cas9-mediated TCR-KO are promising candidates for nonmatched third-party adoptive T-cell transfer with high antileukemic functionality in the absence of alloreactivity, but long-term persistence in vivo is better in the presence of the endogenous TCR.

Tumor-associated antigens (TAAs) are monomorphic self-antigens that are proposed as targets for immunotherapeutic approaches to treat malignancies. We investigated whether T cells with sufficient avidity to recognize naturally overexpressed self-antigens in the context of self-HLA can be found in the T-cell repertoire of healthy donors. Minor histocompatibility antigen (MiHA)-specific T cells were used as a model, as the influence of thymic selection on the T-cell repertoire directed against MiHA can be studied in both self (MiHApos donors) and non-self (MiHAneg donors) backgrounds. T-cell clones directed against the HLA*02:01-restricted MiHA HA-1H were isolated from HA-1Hneg/HLA-A*02:01pos and HA-1Hpos/HLA-A*02:01pos donors. Of the 16 unique HA-1H-specific T-cell clones, five T-cell clones derived from HA-1Hneg/HLA-A*02:01pos donors and one T-cell clone derived from an HA-1Hpos/HLA-A*02:01pos donor showed reactivity against HA-1Hpos target cells. In addition, in total, 663 T-cell clones (containing at least 91 unique clones expressing different T-cell receptors) directed against HLA*02:01-restricted peptides of TAA WT1-RMF, RHAMM-ILS, proteinase-3-VLQ, PRAME-VLD, and NY-eso-1-SLL were isolated from HLA-A*02:01pos donors. Only 3 PRAME-VLD-specific and one NY-eso-1-SLL-specific T-cell clone provoked interferon-γ production and/or cytolysis upon stimulation with HLA-A*02:01pos malignant cell lines (but not primary malignant samples) naturally overexpressing the TAA. These results show that self-HLA-restricted T cells specific for self-antigens such as MiHA in MiHApos donors and TAAs are present in peripheral blood of healthy individuals. However, clinical efficacy would require highly effective in vivo priming by peptide vaccination in the presence of proper adjuvants or in vitro expansion of the low numbers of self-antigen-specific T cells of sufficient avidity to recognize endogenously processed antigen.

Programmed cell death-1 (PD-1)/programmed death ligand-1 blockade may potentially augment graft-vs-tumor effects following allogeneic hematopoietic cell transplantation (alloHCT), but retrospective studies of anti-PD-1 therapy reported substantial toxicity from graft-versus-host-disease (GVHD). Here, we report the results of a prospective clinical trial of PD-1 blockade for relapsed hematologic malignancies (HMs) after alloHCT (NCT01822509). The primary objective in this phase 1 multicenter, investigator-initiated study was to determine maximum tolerated dose and safety. Secondary objectives were to assess efficacy and immunologic activity. Patients with relapsed HMs following alloHCT were eligible. Nivolumab was administered every 2 weeks until progression or unacceptable toxicity, starting with a 1-mg/kg cohort, with planned deescalation based on toxicity to a 0.5-mg/kg cohort. Twenty-eight patients were treated (n = 19 myeloid, n = 9 lymphoid). Median age was 57 years (range 27-76), and median time from alloHCT to enrollment was 21 months (range 5.6-108.5). Two of 6 patients treated at 1 mg/kg experienced dose-limiting toxicity (DLT) from immune-related adverse events (irAEs). Twenty-two patients were treated at 0.5 mg/kg, and 4 DLTs occurred, including 2 irAEs and 2 with fatal GVHD. The overall response rate in efficacy-evaluable patients was 32% (8/25). With a median follow-up of 11 months, the 1-year progression-free survival and overall survival were 23% and 56%, respectively. In this first prospective clinical trial of an anti-PD-1 antibody for post-alloHCT relapse, GVHD and irAEs occurred, requiring dose deescalation, with only modest antitumor activity. Further studies of anti-PD-1 therapy post-alloHCT may require specific toxicity mitigation strategies. This trial was registered at www.clinicaltrials.gov as #NCT01822509.

In this issue of Blood, DeBaun proposes a forward-thinking evidence-based strategy to limit the impact of the SARS-CoV-2 coronavirus disease (COVID-19) pandemic on children with sickle cell anemia (SCA) at high risk for stroke. DeBaun argues that rapid initiation of low- and fixed-dose (10 mg/kg per day) hydroxyurea treatment while the children are receiving prophylactic chronic transfusions, and before blood supply is disrupted, ensures that when blood shortages do occur, the clinical benefit of hydroxyurea will be more quickly established, and the children will be better protected.