Fratricide-resistant CD7-CAR T cells in T-ALL.

T cell acute lymphoblastic leukemia (T-ALL) is difficult to treat when it relapses after therapy or is chemoresistant; the prognosis of patients with relapsed or refractory T-ALL is generally poor. We report a case series of 17 such patients who received autologous chimeric antigen receptor (CAR) T cells expressing an anti-CD7 CAR and an anti-CD7 protein expression blocker (PEBL), which prevented CAR T cell fratricide. Despite high leukemic burden and low CAR T cell dosing, 16 of the 17 patients attained minimal residual disease-negative complete remission within 1 month. The remaining patient had CD7- T-ALL cells before infusion, which persisted after infusion. Toxicities were mild: cytokine release syndrome grade 1 in ten patients and grade 2 in three patients; immune effector cell-associated neurotoxicity syndrome grade 1 in two patients. Eleven patients remained relapse-free (median follow-up, 15 months), including all nine patients who received an allotransplant. The first patient is in remission 55 months after infusion without further chemotherapy or transplantation; circulating CAR T cells were detectable for 2 years. T cells regenerating after lymphodepletion lacked CD7 expression, were polyclonal and responded to SARS-CoV-2 vaccination; CD7+ immune cells reemerged concomitantly with CAR T cell disappearance. In conclusion, autologous anti-CD7 PEBL-CAR T cells have powerful antileukemic activity and are potentially an effective option for the treatment of T-ALL.

Impact of Age on Pharmacogenomics and Treatment Outcomes of B-Cell Acute Lymphoblastic Leukemia.

PURPOSE: Acute lymphoblastic leukemia (ALL) can occur across all age groups, with a strikingly higher cure rate in children compared with adults. However, the pharmacological basis of age-related differences in ALL treatment response remains unclear. METHODS: Studying 767 children and 309 adults with newly diagnosed B-cell ALL enrolled on frontline trials at St Jude Children's Research Hospital, MD Anderson Cancer Center, the Alliance for Clinical Trials in Oncology, and the ECOG-ACRIN Cancer Research Group, we determined the ex vivo sensitivity of leukemia cells to 21 drugs. Twenty-three ALL molecular subtypes were identified using RNA sequencing. We systematically characterized the associations between drug response and ALL genomics in children, adolescents and young adults, and elderly adults. We evaluated the effect of age-related gene expression signature on ALL treatment outcomes. RESULTS: Seven ALL drugs (asparaginase, prednisolone, mercaptopurine, dasatinib, nelarabine, daunorubicin, and inotuzumab ozogamicin) showed differential activity between children and adults, of which six were explained by age-related differences in leukemia molecular subtypes. Adolescents and young adults showed similar patterns of drug resistance as older adults, relative to young children. Mercaptopurine exhibited subtype-independent greater sensitivity in children. Transcriptomic profiling uncovered subclusters within CRLF2-, DUX4-, and KMT2A-rearranged ALL that were linked to age and cytotoxic drug resistance. In particular, a subset of children had adult-like ALL on the basis of leukemia gene expression patterns across subtypes, despite their chronological age. Resistant to cytotoxic drugs, children with adult-like ALL exhibited poor prognosis in pediatric ALL trials, even after adjusting for age and minimal residual diseases. CONCLUSION: Our results provide pharmacogenomic insights into age-related disparities in ALL cure rates and identify leukemia prognostic features for treatment individualization across age groups.

Oncogenic dependency on SWI/SNF chromatin remodeling factors in T-cell acute lymphoblastic leukemia.

T-cell acute lymphoblastic leukemia (T-ALL) is a hematological malignancy arising from immature thymocytes. Unlike well-known oncogenic transcription factors, such as NOTCH1 and MYC, the involvement of chromatin remodeling factors in T-ALL pathogenesis is poorly understood. Here, we provide compelling evidence on how SWI/SNF chromatin remodeling complex contributes to human T-ALL pathogenesis. Integrative analysis of transcriptomic and ATAC-Seq datasets revealed high expression of SMARCA4, one of the subunits of the SWI/SNF complex, in T-ALL patient samples and cell lines compared to normal T cells. Loss of SMARCA protein function resulted in apoptosis induction and growth inhibition in multiple T-ALL cell lines. ATAC-Seq analysis revealed a massive reduction in chromatin accessibility across the genome after the loss of SMARCA protein function. RUNX1 interacts with SMARCA4 protein and co-occupies the same genomic regions. Importantly, the NOTCH1-MYC pathway was primarily affected when SMARCA protein function was impaired, implicating SWI/SNF as a novel therapeutic target.

Activation of NOTCH signaling impedes cell proliferation and survival in acute megakaryoblastic leukemia.

The genetic lesions that drive acute megakaryoblastic leukemia (AMKL) have not been fully elucidated. To search for genetic alterations in AMKL, we performed targeted deep sequencing in 34 AMKL patient samples and 8 AMKL cell lines and detected frequent genetic mutations in the NOTCH pathway in addition to previously reported alterations in GATA-1 and the JAK-STAT pathway. Pharmacological and genetic NOTCH activation, but not inhibition, significantly suppressed AMKL cell proliferation in both in vitro and in vivo assays employing a patient-derived xenograft model. These results suggest that NOTCH inactivation underlies AMKL leukemogenesis. and NOTCH activation holds the potential for therapeutic application in AMKL.

Additive effects of TPMT and NUDT15 on thiopurine toxicity in children with acute lymphoblastic leukemia across multiethnic populations.

BACKGROUND: Thiopurines such as mercaptopurine (MP) are widely used to treat acute lymphoblastic leukemia (ALL). Thiopurine-S-methyltransferase (TPMT) and Nudix hydrolase 15 (NUDT15) inactivate thiopurines, and no-function variants are associated with drug-induced myelosuppression. Dose adjustment of MP is strongly recommended in patients with intermediate or complete loss of activity of TPMT and NUDT15. However, the extent of dosage reduction recommended for patients with intermediate activity in both enzymes is currently not clear. METHODS: MP dosages during maintenance were collected from 1768 patients with ALL in Singapore, Guatemala, India, and North America. Patients were genotyped for TPMT and NUDT15, and actionable variants defined by the Clinical Pharmacogenetics Implementation Consortium were used to classify patients as TPMT and NUDT15 normal metabolizers (TPMT/NUDT15 NM), TPMT or NUDT15 intermediate metabolizers (TPMT IM or NUDT15 IM), or TPMT and NUDT15 compound intermediate metabolizers (TPMT/NUDT15 IM/IM). In parallel, we evaluated MP toxicity, metabolism, and dose adjustment using a Tpmt/Nudt15 combined heterozygous mouse model (Tpmt+/-/Nudt15+/-). RESULTS: Twenty-two patients (1.2%) were TPMT/NUDT15 IM/IM in the cohort, with the majority self-reported as Hispanics (68.2%, 15/22). TPMT/NUDT15 IM/IM patients tolerated a median daily MP dose of 25.7 mg/m2 (interquartile range = 19.0-31.1 mg/m2), significantly lower than TPMT IM and NUDT15 IM dosage (P < .001). Similarly, Tpmt+/-/Nudt15+/- mice displayed excessive hematopoietic toxicity and accumulated more metabolite (DNA-TG) than wild-type or single heterozygous mice, which was effectively mitigated by a genotype-guided dose titration of MP. CONCLUSION: We recommend more substantial dose reductions to individualize MP therapy and mitigate toxicity in TPMT/NUDT15 IM/IM patients.

Biologic and clinical features of childhood gamma delta T-ALL: identification of STAG2/LMO2 γδ T-ALL as an extremely high risk leukemia in the very young.

PURPOSE: Gamma delta T-cell receptor-positive acute lymphoblastic leukemia (γδ T-ALL) is a high-risk but poorly characterized disease. METHODS: We studied clinical features of 200 pediatric γδ T-ALL, and compared the prognosis of 93 cases to 1,067 protocol-matched non-γδ T-ALL. Genomic features were defined by transcriptome and genome sequencing. Experimental modeling was used to examine the mechanistic impacts of genomic alterations. Therapeutic vulnerabilities were identified by high throughput drug screening of cell lines and xenografts. RESULTS: γδ T-ALL in children under three was extremely high-risk with 5-year event-free survival (33% v. 70% [age 3-<10] and 73% [age ≥10], P =9.5 x 10 -5 ) and 5-year overall survival (49% v. 78% [age 3-<10] and 81% [age ≥10], P =0.002), differences not observed in non-γδ T-ALL. γδ T-ALL in this age group was enriched for genomic alterations activating LMO2 activation and inactivating STAG2 inactivation ( STAG2/LMO2 ). Mechanistically, we show that inactivation of STAG2 profoundly perturbs chromatin organization by altering enhancer-promoter looping resulting in deregulation of gene expression associated with T-cell differentiation. Drug screening showed resistance to prednisolone, consistent with clinical slow treatment response, but identified a vulnerability in DNA repair pathways arising from STAG2 inactivation, which was efficaciously targeted by Poly(ADP-ribose) polymerase (PARP) inhibition, with synergism with HDAC inhibitors. Ex-vivo drug screening on PDX cells validated the efficacy of PARP inhibitors as well as other potential targets including nelarabine. CONCLUSION: γδ T-ALL in children under the age of three is extremely high-risk and enriched for STAG2/LMO2 ALL. STAG2 loss perturbs chromatin conformation and differentiation, and STAG2/LMO2 ALL is sensitive to PARP inhibition. These data provide a diagnostic and therapeutic framework for pediatric γδ T-ALL. SUPPORT: The authors are supported by the American and Lebanese Syrian Associated Charities of St Jude Children's Research Hospital, NCI grants R35 CA197695, P50 CA021765 (C.G.M.), the Henry Schueler 41&9 Foundation (C.G.M.), and a St. Baldrick's Foundation Robert J. Arceci Innovation Award (C.G.M.), Gabriella Miller Kids First X01HD100702 (D.T.T and C.G.M.) and R03CA256550 (D.T.T. and C.G.M.), F32 5F32CA254140 (L.M.), and a Garwood Postdoctoral Fellowship of the Hematological Malignancies Program of the St Jude Children's Research Hospital Comprehensive Cancer Center (S.K.). This project was supported by the National Cancer Institute of the National Institutes of Health under the following award numbers: U10CA180820, UG1CA189859, U24CA114766, U10CA180899, U10CA180866 and U24CA196173. DISCLAIMER: The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The funding agencies were not directly involved in the design of the study, gathering, analysis and interpretation of the data, writing of the manuscript, or decision to submit the manuscript for publication.

TAL1 hijacks MYCN enhancer that induces MYCN expression and dependence on mevalonate pathway in T-cell acute lymphoblastic leukemia.

A hallmark of T-cell acute lymphoblastic leukemia (T-ALL) is the dysregulated expression of oncogenic transcription factors (TFs), including TAL1, NOTCH1 and MYC. Rewiring of the transcriptional program disrupts the tightly controlled spatiotemporal expression of downstream target genes, thereby contributing to leukemogenesis. In this study, we first identify an evolutionarily conserved enhancer element controlling the MYCN oncogene (named enhMYCN) that is aberrantly activated by the TAL1 complex in T-ALL cells. TAL1-positive T-ALL cells are highly dependent on MYCN expression for their maintenance in vitro and in xenograft models. Interestingly, MYCN drives the expression of multiple genes involved in the mevalonate pathway, and T-ALL cells are sensitive to inhibition of HMG-CoA reductase (HMGCR), a rate-limiting enzyme of this pathway. Importantly, MYC and MYCN regulate the same targets and compensate for each other. Thus, MYCN-positive T-ALL cells display a dual dependence on the TAL1-MYCN and NOTCH1-MYC pathways. Together, our results demonstrate that enhMYCN-mediated MYCN expression is required for human T-ALL cells and implicate the TAL1-MYCN-HMGCR axis as a potential therapeutic target in T-ALL.

Outcome for Children and Young Adults With T-Cell ALL and Induction Failure in Contemporary Trials.

PURPOSE: Historically, patients with T-cell acute lymphoblastic leukemia (T-ALL) who fail to achieve remission at the end of induction (EOI) have had poor long-term survival. The goal of this study was to examine the efficacy of contemporary therapy, including allogeneic hematopoietic stem cell transplantation (HSCT) in first remission (CR1). METHODS: Induction failure (IF) was defined as the persistence of at least 5% bone marrow (BM) lymphoblasts and/or extramedullary disease after 4-6 weeks of induction chemotherapy. Disease features and clinical outcomes were reported in 325 of 6,167 (5%) patients age 21 years and younger treated in 14 cooperative study groups between 2000 and 2018. RESULTS: With a median follow-up period of 6.4 years (range, 0.3-17.9 years), the 10-year overall survival (OS) was 54.7% (SE = 2.9), which is significantly higher than the 27.6% (SE = 2.9) observed in the historical cohort from 1985 to 2000. There was no significant impact of sex, age, white blood cell count, central nervous system disease status, T-cell maturity, or BM disease burden at EOI on OS. Postinduction complete remission (CR) was achieved in 93% of patients with 10-year OS of 59.6% (SE = 3.1%) and disease-free survival (DFS) of 56.3% (SE = 3.1%). Among the patients who achieved CR, 72% underwent HSCT and their 10-year DFS (with a 190-day landmark) was significantly better than nontransplanted patients (63.8% [SE = 3.6] v 45.5% [SE = 7.1]; P = .005), with OS of 66.2% (SE = 3.6) versus 50.8% (SE = 6.8); P = .10, respectively. CONCLUSION: Outcomes for patients age 21 years and younger with T-ALL and IF have improved in the contemporary treatment era with a DFS benefit among those undergoing HSCT in CR1. However, outcomes still lag considerably behind those who achieve remission at EOI, warranting investigation of new treatment approaches.

Anthracycline-Free Protocol for Favorable-Risk Childhood ALL: A Noninferiority Comparison Between Malaysia-Singapore ALL 2003 and ALL 2010 Studies.

PURPOSE: To investigate whether, for children with favorable-risk B-cell precursor ALL (BCP-ALL), an anthracycline-free protocol is noninferior to a modified Berlin-Frankfurt-Muenster ALL-IC2002 protocol, which includes 120 mg/m2 of anthracyclines. PATIENTS AND METHODS: Three hundred sixty-nine children with favorable-risk BCP-ALL (age 1-9 years, no extramedullary disease, and no high-risk genetics) who cleared minimal residual disease (≤0.01%) at the end of remission induction were enrolled into Ma-Spore (MS) ALL trials. One hundred sixty-seven standard-risk (SR) patients (34% of Malaysia-Singapore ALL 2003 study [MS2003]) were treated with the MS2003-SR protocol and received 120 mg/m2 of anthracyclines during delayed intensification while 202 patients (42% of MS2010) received an anthracycline-free successor protocol. The primary outcome was a noninferiority margin of 1.15 in 6-year event-free survival (EFS) between the MS2003-SR and MS2010-SR cohorts. RESULTS: The 6-year EFS of MS2003-SR and MS2010-SR (anthracycline-free) cohorts was 95.2% ± 1.7% and 96.5% ± 1.5%, respectively (P = .46). The corresponding 6-year overall survival was 97.6% and 99.0% ± 0.7% (P = .81), respectively. The cumulative incidence of relapse was 3.6% and 2.6%, respectively (P = .42). After adjustment for race, sex, age, presenting WBC, day 8 prednisolone response, and favorable genetic subgroups, the hazard ratio for MS2010-SR EFS was 0.98 (95% CI, 0.84 to 1.14; P = .79), confirming noninferiority. Compared with MS2003-SR, MS2010-SR had significantly lower episodes of bacteremia (30% v 45.6%; P = .04) and intensive care unit admissions (1.5% v 9.5%; P = .004). CONCLUSION: In comparison with MS2003-SR, the anthracycline-free MS2010-SR protocol is not inferior and was less toxic as treatment for favorable-risk childhood BCP-ALL.

Associations of T-Cell Receptor Repertoire Diversity with L-Asparaginase Allergy in Childhood Acute Lymphoblastic Leukemia.

Asparaginase is a critical component of therapy for childhood acute lymphoblastic leukemia (ALL), but it is commonly associated with allergy, which results in morbidity and poorer outcomes. The underlying basis of this allergy is undoubtedly immune-mediated, but the exact components of T-cell immunity have yet to be characterized. We performed longitudinal TCR sequencing of 180 bone marrow samples from 67 children with B-ALL treated as part of the Ma-Spore-ALL-2010 trial, and we evaluated the associations of TCR profile with asparaginase hypersensitivity, with functional validation of asparaginase activity in a separate cohort of 113 children. We found that a more diverse and dynamically changing TCR repertoire was associated with increased risk of clinical hypersensitivity and decreased L-asp activity. Allergic patients had a higher proportion of infrequent clonotypes, as well as a significantly lower degree of shared clonotypes amongst the cohort. Allergic patients also had significantly higher longitudinal variability of clonotypes across timepoints, where a higher dissimilarity between diagnosis and week 5 represented an 8.1-fold increased risk of an allergic event. After an allergy had occurred, there was shaping and convergence of the TCR repertoire towards a common antigen. Understanding the immunological basis of T-cell responses in allergy lays the groundwork for developing predictive biomarkers or strategies to mediate this common toxicity in childhood ALL.

Targeting dual oncogenic machineries driven by TAL1 and PI3K-AKT pathways in T-cell acute lymphoblastic leukemia.

T-cell acute lymphoblastic leukemia (T-ALL) is a malignancy of thymic T-cell precursors. Overexpression of oncogenic transcription factor TAL1 is observed in 40-60% of human T-ALL cases, frequently together with activation of the NOTCH1 and PI3K-AKT pathways. In this study, we performed chemical screening to identify small molecules that can inhibit the enhancer activity driven by TAL1 using the GIMAP enhancer reporter system. Among approximately 3,000 compounds, PIK- 75, a known inhibitor of PI3K and CDK, was found to strongly inhibit the enhancer activity. Mechanistic analysis demonstrated that PIK-75 blocks transcriptional activity, which primarily affects TAL1 target genes as well as AKT activity. TAL1-positive, AKT-activated T-ALL cells were very sensitive to PIK-75, as evidenced by growth inhibition and apoptosis induction, while T-ALL cells that exhibited activation of the JAK-STAT pathway were insensitive to this drug. Together, our study demonstrates a strategy targeting two types of core machineries mediated by oncogenic transcription factors and signaling pathways in T-ALL.

Epigenetic activation of the FLT3 gene by ZNF384 fusion confers a therapeutic susceptibility in acute lymphoblastic leukemia.

FLT3 is an attractive therapeutic target in acute lymphoblastic leukemia (ALL) but the mechanism for its activation in this cancer is incompletely understood. Profiling global gene expression in large ALL cohorts, we identify over-expression of FLT3 in ZNF384-rearranged ALL, consistently across cases harboring different fusion partners with ZNF384. Mechanistically, we discover an intergenic enhancer element at the FLT3 locus that is exclusively activated in ZNF384-rearranged ALL, with the enhancer-promoter looping directly mediated by the fusion protein. There is also a global enrichment of active enhancers within ZNF384 binding sites across the genome in ZNF384-rearranged ALL cells. Downregulation of ZNF384 blunts FLT3 activation and decreases ALL cell sensitivity to FLT3 inhibitor gilteritinib in vitro. In patient-derived xenograft models of ZNF384-rearranged ALL, gilteritinib exhibits significant anti-leukemia efficacy as a monotherapy in vivo. Collectively, our results provide insights into FLT3 regulation in ALL and point to potential genomics-guided targeted therapy for this patient population.

Targeted Gene Sanger Sequencing Should Remain the First-Tier Genetic Test for Children Suspected to Have the Five Common X-Linked Inborn Errors of Immunity.

To address inborn errors of immunity (IEI) which were underdiagnosed in resource-limited regions, our centre developed and offered free genetic testing for the most common IEI by Sanger sequencing (SS) since 2001. With the establishment of The Asian Primary Immunodeficiency (APID) Network in 2009, the awareness and definitive diagnosis of IEI were further improved with collaboration among centres caring for IEI patients from East and Southeast Asia. We also started to use whole exome sequencing (WES) for undiagnosed cases and further extended our collaboration with centres from South Asia and Africa. With the increased use of Next Generation Sequencing (NGS), we have shifted our diagnostic practice from SS to WES. However, SS was still one of the key diagnostic tools for IEI for the past two decades. Our centre has performed 2,024 IEI SS genetic tests, with in-house protocol designed specifically for 84 genes, in 1,376 patients with 744 identified to have disease-causing mutations (54.1%). The high diagnostic rate after just one round of targeted gene SS for each of the 5 common IEI (X-linked agammaglobulinemia (XLA) 77.4%, Wiskott-Aldrich syndrome (WAS) 69.2%, X-linked chronic granulomatous disease (XCGD) 59.5%, X-linked severe combined immunodeficiency (XSCID) 51.1%, and X-linked hyper-IgM syndrome (HIGM1) 58.1%) demonstrated targeted gene SS should remain the first-tier genetic test for the 5 common X-linked IEI.

Transcriptome-wide subtyping of pediatric and adult T cell acute lymphoblastic leukemia in an international study of 707 cases.

T cell acute lymphoblastic leukemia (T-ALL) is an aggressive hematological malignancy of T cell progenitors, known to be a heterogeneous disease in pediatric and adult patients. Here we attempted to better understand the disease at the molecular level based on the transcriptomic landscape of 707 T-ALL patients (510 pediatric, 190 adult patients, and 7 with unknown age; 599 from published cohorts and 108 newly investigated). Leveraging the information of gene expression enabled us to identify 10 subtypes (G1­G10), including the previously undescribed one characterized by GATA3 mutations, with GATA3R276Q capable of affecting lymphocyte development in zebrafish. Through associating with T cell differentiation stages, we found that high expression of LYL1/LMO2/SPI1/HOXA (G1­G6) might represent the early T cell progenitor, pro/precortical/cortical stage with a relatively high age of disease onset, and lymphoblasts with TLX3/TLX1 high expression (G7­G8) could be blocked at the cortical/postcortical stage, while those with high expression of NKX2-1/TAL1/LMO1 (G9­G10) might correspond to cortical/postcortical/mature stages of T cell development. Notably, adult patients harbored more cooperative mutations among epigenetic regulators, and genes involved in JAK-STAT and RAS signaling pathways, with 44% of patients aged 40 y or above in G1 bearing DNMT3A/IDH2 mutations usually seen in acute myeloid leukemia, suggesting the nature of mixed phenotype acute leukemia.

Digital PCR for Minimal Residual Disease Quantitation Using Immunoglobulin/T-Cell Receptor Gene Rearrangements in Acute Lymphoblastic Leukemia: A Proposed Analytic Algorithm.

In minimal residual disease (MRD), where there are exceedingly low target copy numbers, digital PCR (dPCR) can improve MRD quantitation. However, standards for dPCR MRD interpretation in acute lymphoblastic leukemia are lacking. Here, for immunoglobulin/T-cell receptor-based MRD, we propose an objective, statistics-based analytic algorithm. In 161 postinduction samples from 79 children with acute lymphoblastic leukemia, MRD was performed by dPCR and real-time quantitative PCR (qPCR) using the same markers and primer-probe sets. The dPCR raw data were analyzed by using an automated algorithm. dPCR and qPCR results were highly concordant (P < 0.0001): 98% (50 of 51) of qPCR positive were positive by dPCR, whereas 95% (61 of 64) of qPCR negative results were also negative by dPCR. For MRD quantitation, both qPCR and dPCR were tightly correlated (R2 = 0.94). Using more DNA (1 µg × 7 versus 630 ng × 3), dPCR improved sensitivity of MRD quantitation by one log10 (median MRD positive cutoff 1.6 × 10-5). With dPCR, 83% (29 of 35) of positive-not-quantifiable results by qPCR could be assigned positive/negative MRD status. Seven replicates of tested samples and negative controls were optimal. Compared with qPCR, dPCR could improve MRD sensitivity by one log10. We proposed an automatable, statistics-based algorithm that minimized interoperator variance for dPCR MRD.

Life-threatening infections during treatment for acute lymphoblastic leukemia on the Malaysia-Singapore 2003 and 2010 clinical trials: A risk prediction model.

AIM: Life-threatening infections significantly impact the care of children undergoing therapy for acute lymphoblastic leukemia (ALL) who are at risk of severe sepsis due to both host and treatment factors. Our aim was to develop a life-threatening infection risk prediction model that would allow remote rapid triage of patients to reduce time to first dose of antibiotics and sepsis-related mortality. METHODS: A retrospective analysis of 2068 fever episodes during ALL therapy was used for model building and subsequent internal validation. RESULTS: Three hundred and seventy-seven patients were treated for ALL in two institutions with comparable critical and supportive care resources. A total of 55 patients accounted for 71 admissions to the critical care unit for sepsis that led to eight septic deaths during a 16-year study period. A retrospective analysis of risk factors for sepsis enabled us to build a model focused on 13 variables that discriminated admissions requiring critical care well: area under the receiver operating characteristic curve of .82; 95% CI .76-.87, p<.001, and Brier score of .033. Significant univariate predictors included neutropenia, presence of symptoms of abdominal pain, diarrhea, fever during induction or steroid-based phases, and the lack of any localizing source of infection at time of presentation. CONCLUSION: We have developed a risk prediction model that can reliably identify ALL patients undergoing treatment who are at a higher risk of life-threatening sepsis. Clinical applicability can potentially be extended to low-middle income settings, and its utility should be further studied in real-world settings.