Mortality In Acute Myeloid Leukemia Research Articles

Preferential Efficacy of Remission Maintenance with HDC/IL-2 in Age Groups of Patients with Chemoresponsive or Normal Karyotype AML

Background Relapse in complete remission (CR) remains a leading cause of morbidity and mortality in acute myeloid leukemia (AML). Previous phase 3 trial data show that combinatorial immunotherapy with histamine dihydrochloride and low-dose interleukin-2 (HDC/IL-2) reduces relapse incidence in the post-consolidation phase, in particular among younger adult AML patients (<60 years old) in CR1 (Brune et al., Blood 108:88). Recent post-hoc analyses identified subgroups of responding patients including (1) those achieving CR1 after the first induction course of chemotherapy (referred to as chemoresponsive AML) and (2) those carrying leukemic cells of normal karyotype (NK-AML), i.e. devoid of structural or numerical chromosomal aberrations. Here we report analyses of treatment efficacy in a phase 3 trial within age groups of patients with chemoresponsive AML or NK-AML. Methods Three hundred and twenty AML patients (18-84 years old, median 57) who were ineligible for allogeneic stem cell transplantation were randomly assigned to receive HDC/IL-2 or no treatment (control) for remission maintenance in a phase 3 trial. Patients were recruited in Europe, Israel, Oceania, and USA. Treatment was initiated in CR after the completion of consolidation chemotherapy and comprised 10 consecutive 3-week cycles of HDC/IL-2 for 18 months (mo) with 3- or 6-week rest periods. In each cycle, patients received HDC at 0.5 mg and human recombinant low-dose IL-2 (aldesleukin; 16,400 IU/kg) s.c. twice daily (n=160) or no treatment (control; n=160). After 18 months of treatment, all patients were monitored for >18 additional mo (median follow-up 48 mo). Results Among study subjects in CR1 203/261 (78%) of evaluable patients attained CR after the first induction cycle (chemoresponsive AML) and 128/220 (58%) had NK-AML. Within both subgroups, analysis of efficacy by age at diagnosis implied preferred benefit of HDC/IL-2 vs. control in patients 40-59 years old (Figure 1) with independent efficacy in patients 40-49 and 50-59 years old (Table 1). The treatment arm was not significantly superior over the control arm for LFS or OS in the youngest patients (18-39 years old) or in patients >60 years old (P>0.2, logrank test, for all comparisons (LFS and OS)) within subgroups of chemosensitive AML or NK-AML. Conclusions We speculate that the observed benefit of HDC/IL-2 in patients responding swiftly to initial chemotherapy may be explained by more efficient immune-mediated elimination of leukemic cells at a lower burden of leukemia. The preferential efficacy of HDC/IL-2 in NK-AML points towards efficacy in subgroups of leukemic mutations that are less common in the youngest group of patients. The reason for the lack of significant efficacy in patients >60 years old, despite achievement of CR1 after the initial chemotherapy and despite having NK-AML, remains unknown but may be explained by fading anti-leukemic immunity in old patients. Overall, these findings may be considered for the selection of patients suitable for immunotherapy with HDC/IL-2 in an emerging landscape of strategies for relapse prevention in AML.

Early Mortality as a Quality Indicator in Frontline and Salvage Acute Myeloid Leukemia

BackgroundEarly mortality is a historical measure of the quality of care incorporated into many quality measure algorithms that mostly account for patient comorbidities but do not incorporate disease characteristics and treatment status which independently increase early mortality. This is particularly significant in leukemia, especially in the refractory and salvage settings. Study AimTo define the independent adverse effect of leukemia salvage vs. frontline therapy on early mortality in acute myeloid leukemia (AML) after accounting for the pretreatment independent adverse effects associated with early mortality. Patients and MethodsA total of 4151 patients with AML were analyzed, 2893 newly diagnosed and 1258 in salvage. Univariate and multivariate analyses (MVA) were conducted to determine the independent adverse effects associated with 8-week mortality. ResultsThe 8-week mortality was 13% in frontline therapy and 18% in salvage therapy. By MVA, older age; therapy-related AML; prior history of myelodysplastic syndrome; poorer performance status; high white blood cell count; lower platelet count; higher percent of peripheral blasts; lower albumin levels; higher bilirubin, creatinine, and lactate dehydrogenase levels; and adverse cytogenetic risk groups were independently associated with a higher 8-week mortality rate. Adding treatment status after accounting for the independent adverse variables still selected salvage status as significantly adverse for 8-week mortality (hazard ratio 1.954; P-value < .001). ConclusionsQuality measure algorithms should incorporate a risk mortality index related to leukemia vs. other tumors and benign conditions and a risk mortality index related to the treatment status of leukemia (salvage vs. frontline therapy).

Venetoclax is safe and tolerable as post-transplant maintenance therapy for AML patients at high risk for relapse.

Relapse is the most common cause of mortality in acute myeloid leukemia (AML) patients after allogeneic stem cell transplant (SCT). Post-SCT maintenance strategies that prevent relapse are desirable but must be well tolerated and convenient to administer. We hypothesized single agent venetoclax (ven) may be an effective maintenance therapy among high relapse risk patients. Between February 2019 and December 2021, we administered post-SCT ven maintenance to 49 AML patients at high-risk for relapse as a prospectively defined off-label practice at our institution. Ven was planned to be administered until 1-year post-SCT. While temporary interruptions were common (67.3% of all patients), of those with >1 year follow up, 22/25 (88%) completed the full year of planned therapy. Cytopenias (40.8%) and gastrointestinal adverse events (34.7%) were the most common toxicities. At 1-year post-SCT, overall survival (OS) and relapse-free survival (RFS) were 70% and 67% respectively. Our experience demonstrates single agent ven is a safe, tolerable, and feasible maintenance therapy that may improve RFS and OS in high relapse risk post-SCT patients.

Early mortality in acute myeloid leukemia with KMT2A rearrangement is associated with high risk of bleeding and disseminated intravascular coagulation.

Acute myeloid leukemia (AML) with rearrangement of lysine methyltransferase 2a gene (KMT2Ar) is characterized by chemotherapy resistance and high rates of relapse. However, additional causes of treatment failure or early mortality have not been well-defined in this entity. In a retrospective analysis, causes and rates of early mortality following induction treatment were compared between a cohort of adults with KMT2Ar AML (N=172) and an age-matched cohort of patients with normal karyotype AML (N=522). The 60-day mortality in patients with KMT2Ar AML was 15% compared with 7% with normal karyotype (p=.04). We found a significantly higher occurrence of major bleeding events (p=.005) and total bleeding events (p=.001) in KMT2Ar AML compared with diploid AML. Among evaluable patients with KMT2Ar AML, 93% exhibited overt disseminated intravascular coagulopathy compared with 54% of patients with a normal karyotype before death (p=.03). In a multivariate analysis, KMT2Ar and a monocytic phenotypic were the only independent predictors of any bleeding event in patients who died within 60days (odds ratio, 3.5; 95% CI, 1.4-10.4; p= .03; odds ratio, 3.2; 95% CI, 1-1-9.4; p= .04, respectively). In conclusion, early recognition and aggressive management of disseminated intravascular coagulopathy and coagulopathy are important considerations that could mitigate the risk of death during induction treatment in KMT2Ar AML. Acute myeloid leukemia (AML) with rearrangement of KMT2A is characterized by chemotherapy resistance and high rates of relapse. However, additional causes of treatment failure or early mortality have not been well-defined in this entity. In this article, that KMT2A-rearranged AML is demonstrably associated with higher early mortality and an increased risk of bleeding and coagulopathy, specifically, disseminated intravascular coagulation, compared with normal karyotype AML. These findings emphasize the importance of monitoring and mitigating coagulopathy in KMT2A-rearranged leukemia similar to what is done in acute promyelocytic leukemia.

Expression of the immune checkpoint modulator OX40 indicates poor survival in acute myeloid leukemia

Despite therapeutic advances, mortality of Acute Myeloid Leukemia (AML) is still high. Currently, the determination of prognosis which guides treatment decisions mainly relies on genetic markers. Besides molecular mechanisms, the ability of malignant cells to evade immune surveillance influences the disease outcome and, among others, the expression of checkpoints modulators contributes to this. In AML, functional expression of the checkpoint molecule OX40 was reported, but the prognostic relevance of OX40 and its ligand OX40L axis has so far not been investigated. Here we described expression and prognostic relevance of the checkpoint modulators OX40 and OX40L, analyzed on primary AML cells obtained from 92 therapy naïve patients. Substantial expression of OX40 and OX40L on AML blasts was detected in 29% and 32% of the investigated subjects, respectively, without correlation between the expression of the receptor and its ligand. Whereas OX40L expression was not associated with different survival, patients with high expression levels of the receptor (OX40high) on AML blasts survived significantly shorter than OX40low patients (p = 0.009, HR 0.46, 95% CI 0.24–0.86), which identifies OX40 as novel prognostic marker and a potential therapeutic target in AML patients.

Abstract 3973: Transcriptomic profiling of in-vitro modelled drug resistant AML cell lines unravel metabolic and stemness gene signatures

Abstract Disease recurrence is the major cause of treatment failure and mortality in acute myeloid leukemia (AML). A small fraction of leukemic cells that withstand chemotherapeutic insult can rewire cellular processes to re-emerge with lethal disease. Approaches to decipher drug resistance mechanisms in these drug persister cells can pave the way for newer therapies and improve survival in patients. To uncover drug resistant mechanisms that are clinically relevant, we generated in-house Cytarabine (AraC) and Arsenic trioxide (ATO) resistant cells from naïve MV4-11 AML cell line by incremental dose exposure method. Single cell clones were generated from AraC (MV411-AraC-R)/ATO (MV411-ATO-R) drug tolerant cells. Both the AraC-R and ATO-R clones were resistant to AraC (600-fold) and ATO (2-fold) respectively compared to the naïve cells. AraC-R cells were marginally cross resistant to daunorubicin (DNR) and ATO. Interestingly, the ATO-R cells were highly cross resistant to both AraC (600-fold) and DNR (4-fold) demonstrating altered drug resistance mechanism. We next subjected the AraC-R, ATO-R and the naïve cells to RNAseq analysis to identify transcriptomic changes post resistance development. 1218 transcripts in AraC-R cells and 7654 transcripts in ATO-R cells were differentially expressed compared to the naïve cells, suggesting ATO exposure alters the transcriptome drastically compared to AraC. We performed gene set enrichment analysis (GSEA) with these differentially expressed gene signatures. AraC-R gene signatures were highly enriched (NES-2.8 &amp; 2.0 FDR-0.00) for oxidative phosphorylation and mitochondrial related signatures, consistent with previous report suggesting AraC exposed residual cells are highly enriched for oxidative phosphorylation and are not enriched for stemness gene signatures (Farge et al). ATO-R gene signatures were positively enriched (NES- 2.0 &amp; 1.4 FDR 0.005 &amp; 0.03) for Leukemic stem cell gene signatures (Eppert_LSC and Gentles_LSC) and relapse (NES- 2.0 &amp; 1.5 FDR 0.0 &amp; 0.01) gene signatures (CALCRL_RIC and Hackles_relapse) and negatively enriched for mitochondrial gene signatures showing reduced reliance on mitochondrial metabolism. Since AraC-R cells showed high dependency on mitochondrial metabolism, we treated these cells with venetoclax (VEN) a known oxphos inhibitor for 48 hours and assessed the viability. Ven markedly sensitized the AraC-R cells compared to naïve cells (IC50 13nM and 323nM), while ATO-R cells were markedly resistant to Ven (IC50 8049nM) corroborating our transcriptomic findings. These finding suggest transcriptome profiling of in-vitro modelled drug resistant cell lines can mimic clinically relevant drug resistant mechanism and could aide in identifying alternate therapies to circumvent disease relapse. Further metabolomic characterization and mechanistic studies are ongoing to substantiate these findings. Citation Format: Raveen Stephen Illangeswaran, Daniel Zechariah Jebanesan, Kezia Kanimozhi Sivakumar, Rakhi Thalayattu Vidhyadharan, Vikram Mathews, Shaji Ramachandran Velayudhan, Poonkuzhali Balasubramanian. Transcriptomic profiling of in-vitro modelled drug resistant AML cell lines unravel metabolic and stemness gene signatures [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 3973.

One-Chip Isolation of Drug-Resistant Acute Myeloid Leukemia Cells with CXCR4-Targeted Magnetic Fluorescent Nanoprobes.

Drug resistance and relapse lead to high mortality in acute myeloid leukemia, and studies have shown that CXCR4 overexpression is highly correlated with poor prognosis and drug resistance in leukemia cells. Isolation and detection of AML cells with CXCR4 overexpression will be crucial to the treatment of AML. In this paper, magnetic nanoparticles were firstly prepared successfully by high-temperature thermal decomposition method, and then characterized by TEM, VSM and DLS. Subsequently CXCR4-targeted magnetic fluorescent nanoprobes conjugated with antibody 12G5 were constructed by stepwise coupling. In cell experiments, the obtained probes demonstrated excellent targeting efficacy to CXCR4 overexpressed AML cells HL-60. In addition, HL-60 cells labelled with the magnetic probes can be magnetic isolated successfully in one microfluidics chip, with efficiency of 82.92 ± 7.03%. Overall, this method utilizes the superiority of superparamagnetic nanomaterials and microfluidic technology to achieve the enrichment and capture of drug-resistant cells in a microfluidic chip, providing a new idea for the isolation and detective of drug-resistant acute myeloid leukemia cells.

T-lymphocytes Expression of Toll-like Receptors 2 and 4 in Acute Myeloid Leukemia Patients with Invasive Fungal Infections.

BackgroundInvasive fungal infections (IFIs) are important cause of mortality in acute myeloid leukemia (AML) patients on treatment with intensive induction chemotherapy. Toll-like receptors, mainly Toll-like receptors 2 and 4 (TLR2 and TLR4), play a considerable role in the host defense against microorganisms. The involvement of TLR signaling in modulation of innate immune response is extensively discussed, but the TLR expressions profiling on adaptive immune cells are not specified. Also, the expressions of TLRs and their association with the occurrence of IFIs in patients with AML are not studied. So, the novel aim of this study was to investigate the associations between the T-lymphocyte expression of TLR2 and TLR4 and the occurrence of IFIs in AML patients treated with intensive induction chemotherapy.Materials and MethodsOne hundred twenty two newly diagnosed AML patients were evaluated. The laboratory diagnostic techniques for IFIs include culture, microscopic examination, histopathology, galactomannan assay and PCR. The expressions of TLR2 and TLR4 were analyzed by flow cytometry. The Control group included 20 age and sex-matched individuals.ResultsThere was a significant increase in the expression of TLR4 in AML patients with IFI compared to healthy controls (p = 0.001). TLR2 and TLR4 expressions increased significantly in AML patients with mixed fungal and bacterial infection compared to healthy controls (p= 0.002 and p=0.001, respectively).ConclusionTLRs expressions could be important biological markers for the occurrence of IFI in non-M3 AML patients after intensive induction chemotherapy.

Immune composition and its association with hematologic recovery after chemotherapeutic injury in acute myeloid leukemia

Chemotherapy-induced bone marrow (BM) injury is a significant cause of morbidity and mortality in acute myeloid leukemia (AML). Time to hematologic recovery after standard ("7+3") myeloablative chemotherapy can vary considerably among patients, but the factors that drive or predict BM recovery remain incompletely understood. Here, we assessed the composition of innate and adaptive immune subsets in the regenerating BM (day 17) after induction chemotherapy and related it to hematologic recovery in AML. T cells, and in particular the CD4 central memory (CD4CM) T-cell subset, were significantly enriched in the BM after chemotherapy, suggesting the relative chemoresistance of cells providing long-term memory for systemic pathogens. In contrast, B cells and other hematopoietic subsets were depleted. Higher frequencies of the CD4CM T-cell subset were associated with delayed hematopoietic recovery, whereas a high frequency of natural killer (NK) cells was related to faster recovery of neutrophil counts. The NK/CD4CM ratio in the BM after chemotherapy was significantly associated with the time to subsequent neutrophil recovery (Spearman's ρ=-0.723, p < 0.001, false discovery rate <0.01). The data provide novel insights into adaptive immune cell recovery after injury and identify the NK/CD4CM index as a putative predictor of hematopoietic recovery in AML.

Prediction of early (4-week) mortality in acute myeloid leukemia with intensive chemotherapy.

The progress with intensive chemotherapy and supportive care measures has improved survival in patients with newly diagnosed acute myeloid leukemia (AML). Given the recent development of effective low intensity therapies, an optimal decision on the therapy intensity may improve survival through the avoidance of early mortality. We reviewed the outcome of 3728 patients with newly diagnosed AML who received intensive chemotherapy between August 1980 and May 2020. Intensive chemotherapy was defined as a cumulative cytarabine dose ≥ 700 mg/m2 during induction therapy. We divided the whole cohort into a training and validation group at a 3:1 ratio. The population was divided into a training (2790 patients) and a validation cohort (938 patients). The median age was 55 years (range, 15-99). Among them, 442 patients (12%) had core-binding factor AML. Binary logistic regression identified older age, worse performance status, hyperbilirubinemia, elevated creatinine, hyperuricemia, cytogenetic abnormalities other than CBF and -Y, and pneumonia as adverse prognostic factors for an early 4-week mortality. This risk classification for early mortality was verified in the validation cohort of patients. In the validation cohort of more recently treated patients from 2000 to 2017, the 4-week mortality rates with intensive chemotherapy were 2%, 14%, and 50% in the low-, high-, and very high-risk group, respectively. The mortality rates with low intensity therapies were 3%, 9%, and 20%, respectively. The risk classification guides treatment intensity by the assessment of age, frailty, organ dysfunction, cytogenetic abnormality, and infection to avoid early mortality.

Induction Mortality in Acute Myeloid Leukemia: A 9 Year Retrospective Study from a Tertiary Care Centre in a Low-Middle Income Country Setting

INTRODUCTION: Literature on acute myeloid Leukemia (AML) from low and middle income countries (LMIC) like India indicate lower uptake of intensive treatment (29%) and higher induction mortality (25%) (Philip C, BJH 2015). This is attributable to delayed presentation, higher infection burden, multi-drug resistant organism (MDRO) infections, poor nutritional status, access to care, and high costs driven by ‘out-of-pocket expenditures’, among other reasons.AIM: To determine the clinical profile, cause and costs of care of patients who die during Induction therapy of newly diagnosed AML.METHODOLOGY: This is a retrospective chart review of newly diagnosed AML patients who died during Induction therapy. Induction mortality was defined as death within 60 days from the date of initiation of therapy for newly diagnosed AML. Secondary and therapy related AML were excluded from the study. Demographic and clinical profiling and analyses of the cause of death was undertaken.RESULTs: Between May 2011-Dec 2019, 825 adult patients were registered with newly diagnosed AML. 455 (55.2%) patients discontinued care primarily due to financial constraints, and 370 (44.8%) continued treatment here. Among these (n=370), 275 (74.3%) chose to receive standard or abbreviated intensive therapy (Group A), 58 (15.7%) chose disease control only with continuous less-intensive regimens like hypomethylating agents (Group B) and 37 (10%) chose palliation and best supportive care (Group C). Median age was 55y (20-84) overall, and 39.5y, 67.5y and 69y respectively in groups A, B and C. Females were 59% of the cohort. The overall induction mortality was 13.78% (n=51). Independently, induction mortality was 9.4% (26/275) in Group A, 24.13% (14/58) in Group B and 29.72% (11/37) in Group C, with a yearly trend suggesting a decrease in Group A (Fig 1) in the last 3 years. The most common cause of death was sepsis, 92.1% (n=47), with or without a bleeding episode. Isolated bleeding episodes were accountable for 4% (n=2) of overall deaths, and disease progression in the rest (n=2). Disease burden at presentation (% blasts in peripheral blood and marrow respectively, median values): 40% & 55% in Group A, 47% & 55% Group B, and 60% & 61% in group C. ELN 2017 risk grouping: 16% standard risk (n=8), 51% intermediate risk (n=26) & 33% high risk (n=17). 33% (n=17) had a lung infection at baseline and this was the commonest source of sepsis. Stool and throat surveillance cultures revealed MDRO (carbapenem resistant enterobacteriaceae) in 47% (n=15/32 tests) and 33% (n=7/21 tests) patients, respectively. 74.5% (n=38/51) developed blood culture positive infections during induction therapy. The median number of days to culture positivity was 11 days overall, 13 days in Group A (n=22), 3 in group B (n=13) and 1 day in Group C (n=3). The most common blood culture-detected organism was MDRO Carbapenem resistant Klebsiella (n=12). Imaging detected fungal infection was observed in 47% (n=24). The common antibiotic regimens included meropenem and colistin, with voriconazole being the most common antifungal used. The median ICU stay was 3.5 days (range of 0-24 days) in Group A and 1.5 days (range 0-13) in Group B. The median direct hospital costs in these patients were $6999 in Group A, $7923 in Group B and $2655 in Group C [per-capita GDP of India in 2019: $2099.6].CONCLUSIONS: The induction mortality in AML varies with age and the intent of treatment. Younger patients undergoing intensive curative therapies have lesser induction mortality. Sepsis is the most common cause of induction deaths, primarily due to a high disease burden and higher infection burden at presentation, and the presence of MDRO bacteremia. Costs of care are very high and contribute to the high drop out rate from curative therapy of newly diagnosed AML. [Display omitted] DisclosuresRadhakrishnan: Novartis India: Membership on an entity's Board of Directors or advisory committees; NATCO Pharmaceuticals: Research Funding; Janssen India: Honoraria; Intas Pharmaceuticals: Research Funding; Bristol-Myers-Squibb India: Membership on an entity's Board of Directors or advisory committees, Research Funding; Roche India: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; AstraZeneca India: Honoraria, Speakers Bureau; Aurigene: Speakers Bureau; Dr Reddy's Laboratories: Honoraria, Membership on an entity's Board of Directors or advisory committees; Cipla Pharmaceuticals India: Research Funding; Emcure Pharmaceuticals: Research Funding. Garg: Dr Reddys Laboratories: Honoraria, Speakers Bureau. Nair: Intas pharmaceuticals: Honoraria, Speakers Bureau; Fresenius Kabi India: Honoraria; Janssen India: Honoraria, Speakers Bureau; Cipla Pharmaceuticals: Honoraria, Speakers Bureau; Novartis India: Honoraria; Dr Reddy's Laboratories: Honoraria, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Mylan pharmaceuticals: Honoraria. Chandy: Intas Pharmaceuticals: Research Funding; Pfizer: Honoraria; Janssen: Honoraria.

Fluid overload is associated with increased 90-day mortality in AML patients undergoing induction chemotherapy

Treatment‐related complications contribute substantially to morbidity and mortality in acute myeloid leukemia (AML) patients undergoing induction chemotherapy. Although AML patients are susceptible to fluid overload (FO) (e.g., in the context of chemotherapy protocols, during sepsis treatment or to prevent tumor lysis syndrome), little attention has been paid to its role in AML patients undergoing induction chemotherapy. AML patients receiving induction chemotherapy between 2014 and 2019 were included in this study. FO was defined as ≥5% weight gain on day 7 of induction chemotherapy compared to baseline weight determined on the day of admission. We found FO in 23 (12%) of 187 AML patients undergoing induction chemotherapy. Application of >100 ml crystalloid fluids/kg body weight until day 7 of induction chemotherapy was identified as an independent risk factor for FO. AML patients with FO suffered from a significantly increased 90-day mortality rate and FO was demonstrated as an independent risk factor for 90-day mortality. Our data suggests an individualized, weight-adjusted calculation of crystalloid fluids in order to prevent FO-related morbidity and mortality in AML patients during induction chemotherapy. Prospective trials are required to determine the adequate fluid management in this patient population.

Abstract 525: Novel DARPin multi-specific T-cell engager with an improved therapeutic window to overcome dose limiting toxicities in AML therapies

Abstract Purpose: The medical need due to high mortality in acute myeloid leukemia (AML) remains high, and the treatment of relapsed or refractory AML continues to be therapeutically challenging. MYLOTARG, the only approved anti-CD33 antibody drug conjugate (ADC), has provided proof-of-concept for targeted immunotherapies in AML. Currently, a plethora of ADCs and T-cell engager (TCE) therapies have entered clinical development in AML, but those therapies are often accompanied by dose limiting toxicities, preventing dose escalation to desired anti-tumor efficacy. The biggest challenges seem to be limited target specificity and hyperstimulation of the immune system leading to e.g. myelotoxicities and cytokine release syndrome, respectively. Therefore, more selective therapies are needed to allow for robust anti-tumor activity with a more acceptable safety profile. Experimental design: To address the selectivity challenge, we have generated multi-specific T-cell engaging DARPin® molecules, targeting two different tumor associated antigens (TAAs) with optimized affinity for their targets. In order to find the right target combination, the optimal affinity to increase tumor specificity via avidity, as well as the best molecular architecture, we took advantage of our unique modular DARPin® platform and screened 1000s of combinations of multi-specific DARPin® molecules, binding simultaneously to multiple TAAs in conjunction with our CD3-binding DARPin® molecule. Results: We constructed multi-specific TCEs targeting two different AML antigens with optimized affinity leading to a substantial avidity gain when both targets are co-expressed on tumor cells. The avidity gain resulted in strongly enhanced in vitro potency as shown by activation of both CD8+ and CD4+ T cells and subsequent killing of AML tumor cells, with bioactivities in the range of established TCE benchmark formats (e.g. BiTE® and DART®). In contrast, in an ex vivo whole blood assay the multi-specific DARPin® constructs induced profoundly less cytokine release as compared to benchmark molecules indicating an improved therapeutic window. Finally, we also demonstrated tumor regression in PMBC humanized mouse models bearing MOLM-13 tumors, using both half-life extended (HLE) and non-HLE lead constructs. In conclusion, we have generated TCEs based on multi-specific DARPin® constructs with high potency, selectivity and ultimately with the potential for an improved therapeutic window for the treatment of AML. Citation Format: Nina Reschke, Thamar Looser, Jennifer Krieg, Matteo Bianchi, Patricia Schildknecht, Nicole Bassler, Yvonne Gruebler, Sebastian Grimm, Laura Jeanbart, Tanja Hospodarsch, Alexandra Neculcea, Daniel Steiner, Bernd Schlereth, Christian Reichen. Novel DARPin multi-specific T-cell engager with an improved therapeutic window to overcome dose limiting toxicities in AML therapies [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 525.

PCN244 Supervised Machine Learning for Predicting Mortality in Acute Myeloid Leukemia Patients Using Electronic Health Record Data

This study implements supervised machine learning (ML) to predict mortality in acute myeloid leukemia (AML) patients and determine the important features in this prediction. Patients were selected from a large US electronic health records database (Cerner Real-World Data) that contains over 87 million patients. We investigated the first visit for patients with an AML ICD-10 diagnosis code, inpatient stay, length of stay of at least 48 hours, and non-missing gender and age. Patient characteristics, hospital characteristics, Charlson Index, quick sequential organ failure assessment (qSOFA), interventions (e.g., mechanical ventilation) and lab values (e.g., minimum white blood cell count) were included in this analysis. Several ML algorithms were compared through 10-fold cross validation; the best performing algorithm was tuned and evaluated with a test dataset. Feature importance was extracted from the final model through permutation importance. There were 8,968 patients included in this study. The ML algorithms that were compared included (mean cross-validation accuracy ± cross-validation standard deviation): logistic regression (72.9% ± 1.6%); random forests (77.5% ± 1.5%); extreme gradient boosting (XGBoost) (78.0% ± 1.2%); k-nearest neighbors (70.8% ± 1.1%); support vector machines (75.8% ± 1.3%). XGBoost was selected for the final model and after hyperparameter tuning had a prediction accuracy of 80.0%. The final model had an F1 score of 0.52, an area under the receiver operator characteristic curve (AUC ROC) of 0.79, a precision of 0.68, and a recall of 0.42. The top five most important features in this prediction were mechanical ventilation, qSOFA, age, intensive care unit admission and minimum white blood cell count. Supervised ML performed well in predicting mortality in AML patients, while identifying the most important features. Similar ML algorithms may identify higher risk AML patients earlier in the hospital to support earlier efforts to modify routine management.

Single-Cell Transcriptomic Profiling of De Novo and Relapsed Acute Myeloid Leukemia Identifies a Leukemic Stemness Program Shared across Diverse Phenotypes

Introduction: Relapse remains the major cause of mortality in acute myeloid leukemia (AML). Prior work indicates that a rare subset of leukemic stem cells (LSCs) self-renew and propagate AML. However, characterizing LSCs is complicated by their scarcity, the lack of universal markers and the heterogeneity across patients. Here, we aim to define a transcriptional program associated with LSCs in patient samples. Methods: We performed single-cell RNA sequencing (scRNA-seq) scRNA-seq of bone marrow from 19 AML samples (14 patients) using the 10X Chromium 3' v2 platform. These samples span multiple morphologies, genetic alterations, and disease stages. Leukemic and normal cells were distinguished based on agreement of three methods: (i) canonical marker expression, (ii) clustering analysis in a multi-sample dataset, and (iii) inferred chromosomal alterations. Leukemic cells were mapped to a panel of signatures from the Human Cell Atlas to infer the most similar normal cell-type, using single-cell gene-set enrichment analysis. Transcription factor activity was inferred at the single-cell level using the SCENIC workflow. Cell state trajectories were constructed using Monocle v2. Common driver mutations were detected at the bulk level using targeted gDNA sequencing and in single cells with targeted amplification of cDNA libraries. A validation cohort of samples was processed with the CITE-seq protocol to capture single-cell gene expression and surface protein levels for CD34, CD38, CD123, CLL1, and TIM3. Results: We captured a total of 55,355 cells meeting quality thresholds, with a median of ~2,800 cells/sample. We observed a large inter-patient heterogeneity with cells segregating largely by sample (Fig. 1A), which was not explained by morphological subtype, treatment received, or driver mutations. As previously described, similarity in gene expression of longitudinal samples did not depend on time before relapse. However, we found transcriptional similarity in a group of samples with relatively silent CNV profiles, suggesting that large chromosomal alterations are a main driver of inter-patient variability. We also observed variation in terms of nearest normal cell assignment: while some samples contained cells resembling diverse mature cell types, others had an abundance of stem-like cells, confirmed by high activity of transcription factors involved in self-renewal (e.g. HOXA9, GATA2). To analyze intrasample variation, we performed Principal Component Analysis and found that, in over half of the samples, LSC and maturation genes were the main source of transcriptional variation. A gradient of activation of known LSC signatures was detected in these samples (Fig 1B). Cell state trajectory reconstruction indicated a continuum of LSC gene expression in leukemic cells. Interestingly, expression of known LSC genes was mostly diffuse is a small subset of samples, a finding that suggests that LSC activity may be widespread in these cases but remains to be validated functionally. Finally, we derived a stemness signature correlated with LSC in our cohort, by extracting concordant genes in a ranked correlation analysis and reconstruction of gene regulatory networks. This yielded a recurrent stemness signature that included previously described LSC-associated genes that were not part of our input, as well as novel factors with expression highly specific to the most LSC-like cells (Fig 1C). To validate this novel stemness signature, we experimentally determined LSC frequencies in a separate cohort (N=5) by xenotransplantation according to expression of CD34 and CD38, and confirmed higher expression of our signature in the LSC fraction. Conclusions: Within a genetically and phenotypically diverse cohort of patients, we could identify, at single-cell resolution, recurrent programs of stemness and myeloid maturation. Altogether, we provide novel candidates for a transcriptional program of putative LSC drivers with therapeutic relevance in AML. Figure Disclosures Johnson: AbbVie: Research Funding; Roche/Genentech, Merck, Bristol-Myers Squibb, AbbVie: Consultancy; Roche/Genentech, Merck: Honoraria. Assouline:Takeda: Research Funding; AbbVie: Consultancy, Honoraria, Speakers Bureau; AstraZeneca: Consultancy, Honoraria, Speakers Bureau; Pfizer: Consultancy, Honoraria; BeiGene: Consultancy, Honoraria, Research Funding; Janssen: Consultancy, Honoraria, Speakers Bureau; F. Hoffmann-La Roche Ltd: Consultancy, Honoraria, Research Funding. Mercier:Sanofi-Genzyme: Consultancy.

The Solute Carrier Family 2 Genes Are Potential Prognostic Biomarkers in Acute Myeloid Leukemia.

Aims:The solute carrier family 2 (SLC2) genes are comprised of 14 members which are essential for the maintenance of glucose uptake and survival of tumour cells. This study was performed to investigate the associations of SLC2 family gene expression with mortality in acute myeloid leukemia (AML).Methods:Clinical features and SLC2 family gene expression data were obtained from The Cancer Genome Atlas and Gene Expression Omnibus database. The associations between SLC2 family gene expression and clinicopathologic features were analyzed using linear regression model. Kaplan-Meier survival, univariate, multivariate survival analyses and validation analysis were performed to analyze the associations between SLC2 family gene expression and patients’ overall survival.Results:Patient mortality was positively associated with age and cytogenetic risk in AML patients. Kaplan-Meier survival analysis suggested that patients with high SLC2A5 and SLC2A10 expression showed poorer survival than those with low SLC2A5 and SLC2A10 expression. In contrast, patients with high SLC2A13 expression exhibited better prognosis than those with low SLC2A13 expression (P < 0.05 for all cases, log rank test). Multivariate survival analysis and validation analysis confirmed that high expression of SLC2A5 and SLC2A10 and low expression of SLC2A13 were associated with increased mortality (P = 0.00, Odd ratio [OR]:4.05, 95% Confidence Interval [CI]: 1.73-10.22; P = 0.00, OR: 3.66, 95% CI: 1.54-9.25; and P = 0.01, OR: 0.26, 95% CI: 0.09-0.68, respectively).Conclusion:SLC family gene expression, such as SLC2A5, SLC2A10 and SLC2A13, was significantly associated with prognosis of AML patients, their expression levels might become useful prognostic biomarkers in AML.

The Value of Adding Surveillance Cultures to Fluoroquinolone Prophylaxis in the Management of Multiresistant Gram Negative Bacterial Infections in Acute Myeloid Leukemia.

Multidrug resistant Gram-Negative Bacterial Infections (MR-GNBI) are an increasing cause of mortality in acute myeloid leukemia (AML), compromising the success of antineoplastic therapy. We prospectively explored a novel strategy, including mandatory fluoroquinolone prophylaxis, weekly surveillance cultures (SC) and targeted antimicrobial therapy for febrile neutropenia, aimed to reduce infectious mortality due to MR-GNBI. Over 146 cycles of chemotherapy, cumulative incidence of colonization was 50%. Half of the colonizations occurred in the consolidation phase of treatment. Application of this strategy led to a significant reduction in the incidence of GNB and carbapenemase-producing Klebisella pneumoniae (cpKp) species, resulting in a reduction of infectious mortality (HR 0.35 [95%, CI 0.13–0.96], p = 0.042). In multivariate analysis, fluroquinolone prophylaxis in addition to SC was associated with improved survival (OR 0.55 [95% CI 0.38–0.79], p = 0.001). Targeted therapy for colonized patients did not overcome the risk of death once cpKp or XDR Pseudomonas aeruginosa infections were developed. Mortality rate after transplant was similar between colonized and not colonized patients. However only 9% of transplanted patients were colonized by cpkp. In conclusion, colonization is a common phenomenon, not limited to the induction phase. This strategy reduces infectious mortality by lowering the global incidence of GN infections and the spread of resistant species.

Induction Related Mortality in Acute Myeloid Leukemia: Multivariate Model of Predictive Score from the Indian Acute Leukemia Research Database (INwARD) of the Hematology Cancer Consortium (HCC)

Background: Remission induction is the most intensive phase of acute myeloid leukemia (AML) treatment, associated with significant morbidity and mortality. Collaborative research and advances in supportive care have steadily improved outcomes in developed countries with induction mortality less than 5%. Challenges for treatment in resource limited settings are varied including delayed presentation, higher disease burden, baseline infections and poor general condition precluding standard intensive therapy, higher rates of resistant infections, and several social and financial constraints. Consequently, a significant proportion of patients do not receive definitive therapy and for those who are treated there is a considerably high risk of induction mortality. In an attempt to identify the subset of patients with highest risk of death during induction, we have developed a multivariate model of induction mortality score using baseline features relevant to our clinical setting by utilizing Indian acute leukemia research database [INwARD] established in 2018 by Hematology Cancer Consortium (HCC). Method: Retrospective data from January 2018 to May 2019 for adult AML was collected from 11 member institutions in a central online data management system. Selection of potential variables that would predict mortality was based on clinical and statistical significance. Thus, 10 variables defining baseline patient and disease characteristics (age, ECOG performance status, duration of symptoms in days, albumin, creatinine, bilirubin, white cell count, platelet, peripheral blood blast percentage, and presence of infection requiring intravenous antibiotic within one week prior of starting induction), were considered for the predictive model using machine learning algorithms: Logistic regression (LR) and Support Vector Machine (SVM). SVM was chosen as the best algorithm based on the AUCs. In order to get robust threshold, sensitivity, specificity and predictive values bootstrapping was done 10,000 times. The final statistics were based on the mean (SD) of bootstrapped sample. R software was used to bootstrap and analyze the data. Result: Of the 611 adult AML cases available during study period, 392 treated with the intensive '3+7' or its abbreviated regimen were considered for analysis. Median age of this cohort was 36 years (range 18 - 67), male to female ratio 1.34. European Leukemia Net (ELN) risk group distribution is shown in Fig 1a. Complete remission was attained in 52.8%. Induction mortality was 16.9 % ranging from 6.1% to 43% across different centers. Most common cause of death was infection (66.7%). Multi-drug resistant blood stream infection was documented in 25.4% cases. For the SVM model for predicting induction mortality using 10 covariates, the AUC based on the bootstrap validation was 91.3% with the best threshold probability being 0.262 (Fig 1b). Thus, a cut off score of 0.262 in the SVM model predicted induction death with sensitivity of 93.6% and specificity of 87.7%. Performance of each variable in the SVM model is shown in Fig 1c and comparison of the LR and SVM model in Fig 1d. Conclusion: Score predicting induction death with high accuracy will be a valuable tool in guiding clinicians against the use of intensive induction therapy, in tailoring of treatment as per individual patients' risk and proper resource allocation. Despite the limitations of retrospective data, wide disparity in resources, patient profile and treatment costs across centers accounting for variability in mortality rates, this study represents one major attempt to find answer to a locally relevant clinical problem of high induction mortality in a cohort of young adult AML, utilizing contemporary pooled data through multi-center collaboration. Optimal cut off point for the score needs to be validated in independent patient cohorts and have to be re-calibrated periodically. Further, an online calculator is being designed on the HCC online system to work as ready reckoner for clinicians. Figure 1 a) ELN risk group distribution b) descriptive statistics of bootstrapped SVM Model c) performance of each covariate in SVM model d) ROC Curve: Comparison of LR and SVM method Figure 1 Disclosures No relevant conflicts of interest to declare.

The Proportional Relationship Between Pretransplant WT1 mRNA Levels and Risk of Mortality After Allogeneic Hematopoietic Cell Transplantation in Acute Myeloid Leukemia Not in Remission.

The relationship between the expression levels of Wilms' tumor-1 gene (WT1) mRNA in peripheral blood before allogeneic hematopoietic cell transplantation (allo-HCT) and risk of mortality in acute myeloid leukemia (AML) patients in noncomplete remission (non-CR) remains quite elusive. We retrospectively assessed the impact of the pretransplant WT1 mRNA level on survival after allo-HCT in non-CR AML patients. A total of 125 AML patients, including 46 non-CR patients (36.8%), were analyzed. On multivariate analysis of non-CR AML patients, WT1 mRNA ≥5000 copies/μg RNA was significantly related to increased risk of mortality (hazard ratio, 2.7; 95% confidence interval, 1.3-5.5; P = 0.008). Furthermore, in the entire cohort, log10-transformed WT1 mRNA before allo-HCT was found to be significantly associated with the increased risk of mortality irrespective of whether the disease status was CR or non-CR, using Akaike's information criterion. As the pretransplant WT1 mRNA level elevated, the hazard ratio of mortality monotonically increased in a nonlinear manner regardless of remission status, suggesting that WT1 mRNA level in peripheral blood might reflect tumor burden. This study demonstrated that the pretransplant WT1 mRNA level was a powerful prognostic factor in allo-HCT even for non-CR AML patients, and there may be a WT1 mRNA threshold in non-CR patients for benefiting from allo-HCT.

PS976 METABOLIC REPROGRAMMING OF ACUTE MYELOID LEUKEMIA STEM‐ PROGENITOR CELLS AT RELAPSE

Background:Primary chemo‐refractory disease and relapse are the most important causes of mortality in acute myeloid leukemia (AML) patients (Döhner et al., N. Engl. J. Med. 2015). Leukemia stem cells (LSCs) are likely responsible for disease relapse (Guzman et al., Blood 2007). Several genomic studies of newly‐diagnosed and relapsed AML have revealed different patterns of clonal evolution, but little is known about metabolic dysregulations among these two groups. Metabolomics can identify endogenous cell metabolite biomarkers to be applied for monitoring disease progression, in order to develop new targeted therapies (D. Wang et al., Journal of Pharm and Biom Analysis 2019).Aims:We aimed to identify relapse‐specific metabolic alterations of CD34+ AML stem‐progenitor cells.Methods:We analyzed metabolites using mass spectrometry (Metabolon) of CD34+ bone‐marrow stem progenitor cells of 22 newly‐diagnosed and 13 relapsed AML patients. To better address the relevance of the results, we also compared our samples with CD34+ cells from umbilical cord blood (21 donors). Statistically‐significant differences in metabolite concentrations between the groups were evaluated by Welch's t‐test. KEGG and SMPDB annotation helped us in identifying the most relevant differences in terms of metabolic pathways.Results:Mass spectrometry analysis identified differences in the concentration of 16 metabolites between newly‐diagnosed and relapsed AML. In particular, we discovered different intermediates of lipid metabolism, as decreased 1‐(1‐enyl‐palmitoyl)‐2‐palmitoyl‐GPC (p = 0,0049), or involved in Kreb's Cycle, like reduced fumarate (p = 0,0393) in relapsed AML. Conversely, we observed a higher concentration of UDP‐N‐acetylglucosamine/galactosamine (p = 0,0044), which is part of carbohydrate metabolism. Moreover, by comparing CD34+ cells from relapsed or newly‐diagnosed AML with cord blood, we defined a set of metabolic features which may suggest metabolic adaptation of the post‐chemotherapy AML clone. It is worth noting the decrease in uridine concentration (p = 0,03), a metabolite of pyrimidine metabolism. Pyrimidine pathway is of great interest thanks to the development of new drugs as dihydroorotate dehydrogenase (DHODH) inhibitor, that reduces leukemic cell burden, decreases levels of leukemia‐initiating cells (Sykes DB et al., Cell 2016). Of note, relapsed cells show reduced orotidine levels compared with cells from newly‐diagnosed patients. Finally, KEGG and SMPDB annotation revealed enrichment for changes in compounds involved in Ketone Body Metabolism, Citric Acid Cycle and Mitochondrial Electron Transport Chain.Summary/Conclusion:Our data are in line with a metabolic reprogramming of AML cells at relapse, including a shift from pyruvate oxidation to fatty acid β‐oxidation, with altered mitochondrial oxidative phosphorylation in leukemia stem progenitor cells. The results suggest potential metabolic vulnerabilities of relapsed patients which could be exploited for novel targeted therapeutic strategies. Supported by EHA research fellowship award.