Acute Lymphoblastic Leukemia Subtypes Research Articles

An Intelligent Telediagnosis of Acute Lymphoblastic Leukemia using Histopathological Deep Learning

Leukemia, a global health challenge characterized by malignant blood cell proliferation, demands innovative diagnostic techniques due to its increasing incidence. Among leukemia types, Acute Lymphoblastic Leukemia (ALL) emerges as a particularly aggressive form affecting diverse age groups. This study proposes an advanced mechanized system utilizing Deep Neural Networks for detecting ALL blast cells in microscopic blood smear images. Achieving a remarkable accuracy of 97% using MobileNetV2, our system demonstrates high sensitivity and specificity in identifying multiple ALL sub-types. Furthermore, we introduce cutting-edge telediagnosis software facilitating real-time support for clinicians in promptly and accurately diagnosing various ALL subtypes from microscopic blood smear images. This research aims to enhance leukemia diagnosis efficiency, which is crucial for the timely intervention and managing this life-threatening condition.

Identifying Subtypes of Acute Lymphoblastic Leukemia Using Blood Smear Images: A Hybrid Learning Approach

Identifying Subtypes of Acute Lymphoblastic Leukemia Using Blood Smear Images: A Hybrid Learning Approach

Acute lymphocytic leukemia detection and subtype classification via extended wavelet pooling based-CNNs and statistical-texture features

Acute lymphoblastic leukemia (ALL) is considered the most fatal form of leukemia (also known as blood cancer). It propagates quickly among adults and children and could lead to their death. Early detection of ALL and ALL subtypes is the key factor in selecting effective treatment types and improving survival rates. However, routine diagnostic approaches have several drawbacks. Computer-assisted diagnosis (CAD) is the perfect solution to avoid these challenges and achieve a fast and accurate diagnosis. Current CAD models require enhancement/segmentation processing. Besides, they are either dependent on deep learning (DL) models or handcrafted features along with machine learning. Those CADs that employed DL approaches relied solely on spatial information during the training procedure. However, learning them with spectral temporal and temporal representations could improve performance. Furthermore, integrating deep features from DL models along with handcrafted features can increase the discrimination ability of attributes in medical image classification. This study aims to propose a novel CAD for ALL detection and subtype classification without pre-segmentation or enhancement steps. The proposed CAD extends the conventional DL models of convolutional neural networks by introducing an additional wavelet pooling, accompanied by a dense layer or a long-short-term memory (LSTM) layer, and then a SoftMax layer, acquiring spectral-temporal information along with temporal information. To further improve the framework's ability to discriminate, the introduced CAD then combines the wavelet-based deep features of every CNN with numerous handcrafted attributes. Afterward, a feature selection methodology is utilized to create a model with limited features and improved accuracy. The performance results show that the novel CAD is capable of achieving 100% ALL detection accuracy, as well as 100% ALL-subtype classification accuracy with just 88 and 146 features. Thus, this CAD can be employed to assist pathologists in the rapid and precise ALL identification and subcategories recognition.

Evaluation of next-generation sequencing for measurable residual disease monitoring in three major fusion transcript subtypes of B-precursor acute lymphoblastic leukaemia

The use of next-generation sequencing (NGS) for monitoring measurable residual disease (MRD) in acute lymphoblastic leukaemia (ALL) has been gaining traction. This study aimed to investigate the utility of NGS in MRD monitoring for the three major fusion transcript (FT) subtypes of B-precursor ALL (B-ALL). The MRD results for 104 bone marrow samples from 56 patients were analysed through NGS and real time quantitative reverse transcription PCR (RT-qPCR) for the three major FTs: BCR::ABL1, TCF3::PBX1, and ETV6::RUNX1. To validate the NGS approach, NGS-MRD was initially compared with allele-specific oligonucleotide-qPCR-MRD, and the coefficient of determination was good (R2=0.8158). A subsequent comparison of NGS-MRD with FT-MRD yielded a good coefficient of determination (R2=0.7690), but the coefficient varied by subtype. Specifically, the R2 was excellent for TCF3::PBX1 ALL (R2=0.9157), good for ETV6::RUNX1 ALL (R2=0.8606), and subpar for BCR::ABL1 ALL (R2=0.5763). The overall concordance between the two methods was 83.7%, and an excellent concordance rate of 95.8% was achieved for TCF3::PBX1 ALL. Major discordance, which was defined as a >1 log difference between discordant NGS-MRD and FT-MRD, occurred in 6.7% of the samples, with all but one sample being BCR::ABL1 ALL. Among the four non-transplanted patients with BCR::ABL1-MRD (+)/NGS-MRD (-), three did not relapse after long-term follow-up. Our finding indicates that NGS-MRD has a better prognostic impact than RT-qPCR-MRD in ETV6::RUNX1 and BCR::ABL1 ALL, whereas in TCF3::PBX1 ALL, both methods exhibit comparable efficacy.

Terminal deoxynucleotidyl transferase expression in different subtypes of childhood B-cell acute lymphoblastic leukemia

The lack of expression of terminal deoxynucleotidyl transferase (TdT) is frequently associated with KMT2A-rearranged subtype of pediatric acute lymphoblastic leukemia (ALL). However, this association has not been investigated extensively in the Asian population. A retrospective analysis of TdT expression in pediatric B-cell ALL (B-ALL) was performed in patients treated using the Taiwan Pediatric Oncology Group (TPOG) ALL 2002 and 2013 protocols. Among the 331 patients with B-ALL, 12 patients showed TdT negativity at initial diagnosis. Among these, eight patients showed KMT2A rearrangement (66.7%). Other patients showing negative TdT expression had ETV6::RUNX1, MEF2D-rearranged, and other B-ALL subtypes. However, in the context of KMT2A-rearranged B-ALL (n = 20), only eight patients showed TdT negativity. The 5-year event-free survival and overall survival of patients with and without TdT expression were 83.8% versus 46.8% (P <0.001) and 86.3% versus 55.4% (P = 0.004), respectively. Moreover, several aberrant markers, such as CD2, CD56, CD7, and CD117, were rarely expressed in the B-ALL samples, and if expressed, they were enriched in specific genetic subtypes. The results of this study indicate that immunophenotypic features are correlated with specific genetic subtypes of childhood B-ALL.

Abstract 4752: Enhancer of zeste homolog 2 inhibitors overcome glucocorticoid resistance in acute lymphoblastic leukemia by augmenting pro-apoptotic signaling

Abstract Introduction Relapsed acute lymphoblastic leukemia (ALL) portends a poor prognosis. Glucocorticoids (GCs) are a central component of a multi-agent regimen to treat ALL. GCs mediate ligand-dependent global transcriptional changes activating feedback loop that increases GC receptor (GR) expression and pro-apoptotic signaling. GC resistance is a common mechanism in relapsed/refractory ALL. Altered function of Enhancer of Zeste Homolog 2 (EZH2), a histone methyl transferase (HMT), including increased expression and activating mutations have been reported in ALL. Previously, EZH2 inhibitors have been shown to increase GC efficacy in non-ALL B-cell lymphoid malignancies including non-Hodgkin lymphoma. Furthermore, EZH2 inhibitors reversed GC resistance and increased efficacy of GCs in an ALL subtype with an activating point mutation in an HMT called nuclear set domain 2. Overcoming GC resistance by novel therapeutic strategies may help improve outcomes in relapsed/refractory ALL. Hypothesis EZH2 inhibitors may enhance GC efficacy by augmenting downstream effects of GCs. Methods We performed in-vitro studies of EZH2 inhibitors, EPZ-6438 and CPI-1205, at graded doses in GC resistance human-derived B-cell (SUPB15) and T-cell (SUPT1, KOPTK1, PEER) ALL cell lines followed by graded doses of dexamethasone (dex). Cell viability was performed using CellTiter-Glo assay. Induction of apoptosis was measured with Annexin V flow cytometry. Pro-apoptotic BIM mRNA and protein levels were measured using real-time PCR and immunoblot. Global H3K27me3 protein levels were assessed using immunoblot. Results GC resistant B-cell ALL and T-cell ALL treated with EZH2 inhibitors for 3-7 days followed by the addition of dex for 48-72 hours led to decreased cell viability and increased apoptosis of the ALL cells compared to cells treated with vehicle (DMSO) and dex alone. EZH2 inhibitors on their own did not cause apoptosis. SynergyFinder revealed Bliss synergy scores between 23- 31 in B-cell ALL and T-cell ALL cell lines with greater than 10 considered a synergistic drug interaction. Immunoblotting of cells treated with EZH2 inhibitors alone showed an increase in pro-apoptotic protein, BIM. BIM level was further enhanced when dex was added in combination with EZH2 inhibitors compared to vehicle and dex alone suggesting that EZH2 inhibitors were augmenting BIM expression. Further assessment of the mechanism of how EZH2 inhibitors lead to improved GC efficacy will be performed by integrating transcriptomic and epigenomic profiling in the presence and absence of EZH2 inhibitors. Furthermore, in-vivo testing of EZH2 inhibitors in patient derived xenografts will be performed. Conclusion EZH2 inhibitors can overcome GC resistance in ALL posing as a potential novel therapeutic agent to treat relapsed/refractory ALL. Citation Format: Mansi Dalal, Joanna Yi, Jianping Li, Amin Shobh, Daphne DupereRicher, Richard Bennett, Jonathan Licht. Enhancer of zeste homolog 2 inhibitors overcome glucocorticoid resistance in acute lymphoblastic leukemia by augmenting pro-apoptotic signaling [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 4752.

Abstract 4907: RanBALL: Identifying B-cell acute lymphoblastic leukemia subtypes based on an ensemble random projection model

Abstract As the most common pediatric malignancy, B-cell acute lymphoblastic leukemia (B-ALL) has multiple distinct subtypes characterized by recurrent and sporadic somatic and germline genetic alterations like chromosomal alteration, transcription factor rearrangement or kinase inhibition. The treatment of B-ALL patients is personalized based on specific subtypes, as the treatment responses for different B-ALL subtypes may vary considerably. Identification of B-ALL subtypes can facilitate risk stratification and enable tailored therapeutic approaches. Existing methods for B-ALL subtyping primarily depend on immunophenotypic, cytogenetic and genomic analyses, which would be costly, complicated, and laborious in clinical practice applications. To overcome these challenges, we present RanBALL (an Ensemble Random Projection-Based Model for Identifying B-Cell Acute Lymphoblastic Leukemia Subtypes), an accurate and cost-effective model for B-ALL subtype identification based on transcriptomic profiling only. RanBALL leverages random projection (RP) to construct an ensemble of dimension-reduced multi-class classifiers for B-ALL subtyping. Specifically, the transcriptomic profiling features were projected onto low-dimensional spaces by random projection matrices whose elements conform to a distribution characterized by zero mean and unit variance. To ensure reliable and robust performance, we selected 20 subspace dimensions ranging from 600 to 2500, with intervals of 100. The transformed low dimensional data matrix was used for training an ensemble of multi-class support vector machine (SVM) classifiers, each corresponding to one of the RP matrices of various dimensions. The predicted probabilistic scores of each B-ALL subtype were integrated for determining the final decision. Results based on 10 times 10-fold cross validation tests for &amp;gt;1700 B-ALL patients demonstrated that the proposed model achieved an accuracy of 93.7%, indicating promising prediction capabilities of RanBALL for B-ALL subtyping. Furthermore, the 30% held-out tests suggested that the model was robust and consistent to maintain high confidence levels for accurate predictions. The high accuracies of RanBALL suggested that our model could effectively capture underlying patterns of transcriptomic profiling for accurate B-ALL subtype identification. To extend the impact of RanBALL, we have established a free and publicly available python package for RanBALL available at https://github.com/wan-mlab/RanBALL. We believe RanBALL will facilitate the discovery of B-ALL subtype-specific marker genes and therapeutic targets, and eventually have consequential positive impacts on downstream risk stratification and tailored treatment design. Citation Format: Lusheng Li, Hanyu Xiao, Joseph D. Khoury, Jieqiong Wang, Shibiao Wan. RanBALL: Identifying B-cell acute lymphoblastic leukemia subtypes based on an ensemble random projection model [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 4907.

Automatic classification of acute lymphoblastic leukemia cells and lymphocyte subtypes based on a novel convolutional neural network.

Acute lymphoblastic leukemia (ALL) is a life-threatening disease that commonly affects children and is classified into three subtypes: L1, L2, and L3. Traditionally, ALL is diagnosed through morphological analysis, involving the examination of blood and bone marrow smears by pathologists. However, this manual process is time-consuming, laborious, and prone to errors. Moreover, the significant morphological similarity between ALL and various lymphocyte subtypes, such as normal, atypic, and reactive lymphocytes, further complicates the feature extraction and detection process. The aim of this study is to develop an accurate and efficient automatic system to distinguish ALL cells from these similar lymphocyte subtypes without the need for direct feature extraction. First, the contrast of microscopic images is enhanced using histogram equalization, which improves the visibility of important features. Next, a fuzzy C-means clustering algorithm is employed to segment cell nuclei, as they play a crucial role in ALL diagnosis. Finally, a novel convolutional neural network (CNN) with three convolutional layers is utilized to classify the segmented nuclei into six distinct classes. The CNN is trained on a labeled dataset, allowing it to learn the distinguishing features of each class. To evaluate the performance of the proposed model, quantitative metrics are employed, and a comparison is made with three well-known deep networks: VGG-16, DenseNet, and Xception. The results demonstrate that the proposed model outperforms these networks, achieving an approximate accuracy of 97%. Moreover, the model's performance surpasses that of other studies focused on 6-class classification in the context of ALL diagnosis. RESEARCH HIGHLIGHTS: Deep neural networks eliminate the requirement for feature extraction in ALL classification The proposed convolutional neural network achieves an impressive accuracy of approximately 97% in classifying six ALL and lymphocyte subtypes.

An efficient decision support system for leukemia identification utilizing nature-inspired deep feature optimization.

In the field of medicine, decision support systems play a crucial role by harnessing cutting-edge technology and data analysis to assist doctors in disease diagnosis and treatment. Leukemia is a malignancy that emerges from the uncontrolled growth of immature white blood cells within the human body. An accurate and prompt diagnosis of leukemia is desired due to its swift progression to distant parts of the body. Acute lymphoblastic leukemia (ALL) is an aggressive type of leukemia that affects both children and adults. Computer vision-based identification of leukemia is challenging due to structural irregularities and morphological similarities of blood entities. Deep neural networks have shown promise in extracting valuable information from image datasets, but they have high computational costs due to their extensive feature sets. This work presents an efficient pipeline for binary and subtype classification of acute lymphoblastic leukemia. The proposed method first unveils a novel neighborhood pixel transformation method using differential evolution to improve the clarity and discriminability of blood cell images for better analysis. Next, a hybrid feature extraction approach is presented leveraging transfer learning from selected deep neural network models, InceptionV3 and DenseNet201, to extract comprehensive feature sets. To optimize feature selection, a customized binary Grey Wolf Algorithm is utilized, achieving an impressive 80% reduction in feature size while preserving key discriminative information. These optimized features subsequently empower multiple classifiers, potentially capturing diverse perspectives and amplifying classification accuracy. The proposed pipeline is validated on publicly available standard datasets of ALL images. For binary classification, the best average accuracy of 98.1% is achieved with 98.1% sensitivity and 98% precision. For ALL subtype classifications, the best accuracy of 98.14% was attained with 78.5% sensitivity and 98% precision. The proposed feature selection method shows a better convergence behavior as compared to classical population-based meta-heuristics. The suggested solution also demonstrates comparable or better performance in comparison to several existing techniques.

Abstract P30: Therapeutic activation of pre-B-cell receptor signaling in high-risk pediatric ALL: exploiting TCF3 and MYC-linked vulnerabilities

Abstract P30: Therapeutic activation of pre-B-cell receptor signaling in high-risk pediatric ALL: exploiting TCF3 and MYC-linked vulnerabilities

ETV6::ACSL6 translocation-driven super-enhancer activation leads to eosinophilia in acute lymphoblastic leukemia through IL-3 overexpression.

ETV6::ACSL6 represents a rare genetic aberration in hematopoietic neoplasms and is often associated with severe eosinophilia, which confers an unfavorable prognosis requiring additional anti-inflammatory treatment. However, since the translocation is unlikely to produce a fusion protein, the mechanism of ETV6::ACSL6 action remains unclear. Here, we performed multi-omics analyses of primary leukemia cells and patient-derived xenografts from an acute lymphoblastic leukemia (ALL) patient with ETV6::ACSL6 translocation. We identified a super-enhancer located within the ETV6 gene locus and revealed translocation and activation of the super-enhancer associated with the ETV6::ACSL6 fusion. The translocated super-enhancer exhibited intense interactions with genomic regions adjacent to and distal from the breakpoint at chromosomes 5 and 12, including genes coding inflammatory factors such as IL-3. This led to modulations in DNA methylation, histone modifications, and chromatin structures, triggering transcription of inflammatory factors leading to eosinophilia. Furthermore, the bromodomain and extraterminal domain (BET) inhibitor synergized with standard-of-care drugs for ALL, effectively reducing IL-3 expression and inhibiting ETV6::ACSL6 ALL growth in vitro and in vivo. Overall, our study revealed for the first time a cis-regulatory mechanism of super-enhancer translocation in ETV6::ACSL6 ALL, leading to ALL-accompanying clinical syndrome. These findings may stimulate novel treatment approaches for this challenging ALL subtype.

Raman classification of selected subtypes of acute lymphoblastic leukemia (ALL).

B-cell precursor acute lymphoblastic leukemia (BCP-ALL) with chromosome translocations like KMT2A gene rearrangement (KMT2A-r) and BCR-ABL1 fusion gene have been recognized as crucial drivers in both BCP-ALL leukemogenesis and treatment management. Standard diagnostic protocols for proliferative diseases of the hematopoietic system, like KMT2A-r-ALL, are genetically based and strongly molecularly oriented. Therefore, an efficient diagnostic procedure requires not only experienced and multidisciplinary laboratory staff but also considerable instrumentation and material costs. In recent years, a Raman spectroscopy method has been increasingly used to detect subtle chemical changes in individual cells resulting from stress or disease. Therefore, the objective of this study was to identify Raman signatures for the molecular subtypes and to develop a classification method based on the unique spectroscopic profile of in vitro models that represent specific aberrations aimed at KMT2A-r (RS4;11, and SEM) and the BCR-ABL1 fusion gene (SUP-B15, BV-173, and SD-1). Data analysis was based on chemometric methods, i.e. principal component analysis (PCA), partial least squares discriminant analysis (PLS-DA), and support vector machine (SVM). The PCA-based multivariate model was used for pattern recognition of each investigated group of cells while PLS-DA and SVM were used to build models for the discrimination of spectra from the studied BCP-ALL molecular subtypes. The results showed that the studied molecular subtypes of ALL have characteristic spectroscopic profiles reflecting their peculiar biochemical state. The content of lipids (1600 cm-1), nucleic acids (789 cm-1), and haemoproteins (754, 1130, and 1315 cm-1), which are crucial in cell metabolism, was indicated as the main source of differentiation between subtypes. Identification of spectroscopic markers of cells with BCR-ABL1 or KMT2A-r may be useful in pharmacological studies to monitor the effectiveness of chemotherapy and further to understand differences in molecular responses between leukemia primary cells and cell lines.

Diagnostic and therapeutic advances in adults with acute lymphoblastic leukemia in the era of gene analysis and targeted immunotherapy.

Acute lymphoblastic leukemia (ALL) is one of the most rapidly changing hematological malignancies with advanced understanding of the genetic landscape, detection methods of minimal residual disease (MRD), and the development of immunotherapeutic agents with good clinical outcomes. The annual incidence of adult ALL in Korea is 300-350 patients per year. The WHO classification of ALL was revised in 2022 to reflect the molecular cytogenetic features and suggest new adverse- risk subgroups, such as Ph-like ALL and ETP-ALL. We continue to use traditional adverse-risk features and cytogenetics, with MRD-directed post-remission therapy including allogeneic hematopoietic cell transplantation. However, with the introduction of novel agents, such as ponatinib, blinatumomab, and inotuzumab ozogamicin incorporated into frontline therapy, good MRD responses have been achieved, and overall survival outcomes are improving. Accordingly, some clinical trials have suggested a possible era of chemotherapy-free or transplantation-free approaches in the near future. Nevertheless, relapse of refractory ALL still occurs, and some poor ALL subtypes, such as Ph-like ALL and ETP-ALL, are unsolved problems for which novel agents and treatment strategies are needed. In this review, we summarize the currently applied diagnostic and therapeutic practices in the era of advanced genetic analysis and targeted immunotherapies in United States and Europe and introduce real-world Korean data.

MAJOR REASONS FOR HOSPITALIZATION TO ICU OF CHILDREN WITH ACUTE LYMPHOBLASTIC LEUKEMIA: A LITERATURE REVIEW

Relevance: Acute lymphoblastic leukemia (ALL) is a common malignancy in children. Approximately 85% of ALLs have B-cell origin, and 15% are T-cell ALLs. Many patients diagnosed with hematologic cancer will require hospitalization in the intensive care unit (ICU) at some point in their treatment.&#x0D; The aim was to study the available literature on clinical deterioration in patients with АLL admitted to the ICU, the clinical significance and prognostic value of causes of clinical deterioration, and adverse outcomes in patients with АLL staying in the ICU.&#x0D; Methods: A descriptive cross-sectional study approach was used. We reviewed published sources from 2016 to 2023 to collect data on major reasons for ALL patients’ hospitalization to ICU.&#x0D; Results: First, the patient’s age at the time of initial diagnosis of ALL is crucial. Cure rates for B-cell ALL are higher between 1 and 9 years of age than in other age groups. Second, the initial white blood cell count during diagnosis is a prognostic indicator. Third, the specific subtype of ALL also affects prognosis. The risk factors emphasize the importance of comorbidities and infectious diseases, as well as monitoring and managing pulmonary and cardiovascular function in patients to avoid hospitalization in the ICU. The main causes of hospitalization in the ICU are complications related to chemotherapy, infection, and unplanned hospitalizations. Compared to normal-risk patients, high-risk patients had a higher rate of OIT hospitalization in the ICU. It is important to control chemotherapy and infections to reduce the number of admissions to the ICU in this group.&#x0D; Conclusion: Chemotherapy, concomitant and infectious diseases, hypoxia, and hemodynamic instability are reasons for hospitalization of these patients to ICU. The condition of various organs and systems shall be monitored.

MAJOR REASONS FOR HOSPITALIZATION TO ICU OF CHILDREN&#x0D; WITH ACUTE LYMPHOBLASTIC LEUKEMIA:&#x0D; A LITERATURE REVIEW

Relevance: Acute lymphoblastic leukemia (ALL) is a common malignancy in children. Approximately 85% of ALLs have B-cell origin, and 15% are&#x0D; T-cell ALLs. Many patients diagnosed with hematologic cancer will require hospitalization in the intensive care unit (ICU) at some point in their treatment.&#x0D; The aim was to study the available literature on clinical deterioration in patients with АLL admitted to the ICU, the clinical significance&#x0D; and prognostic value of causes of clinical deterioration, and adverse outcomes in patients with АLL staying in the ICU.&#x0D; Methods: A descriptive cross-sectional study approach was used. We reviewed published sources from 2016 to 2023 to collect data on major&#x0D; reasons for ALL patients’ hospitalization to ICU.&#x0D; Results: First, the patient’s age at the time of initial diagnosis of ALL is crucial. Cure rates for B-cell ALL are higher between 1 and 9 years of&#x0D; age than in other age groups. Second, the initial white blood cell count during diagnosis is a prognostic indicator. Third, the specific subtype of ALL&#x0D; also affects prognosis. The risk factors emphasize the importance of comorbidities and infectious diseases, as well as monitoring and managing&#x0D; pulmonary and cardiovascular function in patients to avoid hospitalization in the ICU. The main causes of hospitalization in the ICU are complications related to chemotherapy, infection, and unplanned hospitalizations. Compared to normal-risk patients, high-risk patients had a higher rate of&#x0D; OIT hospitalization in the ICU. It is important to control chemotherapy and infections to reduce the number of admissions to the ICU in this group.&#x0D; Conclusion: Chemotherapy, concomitant and infectious diseases, hypoxia, and hemodynamic instability are reasons for hospitalization of&#x0D; these patients to ICU. The condition of various organs and systems shall be monitored.

Multimodal classification of molecular subtypes in pediatric acute lymphoblastic leukemia

Genomic analyses have redefined the molecular subgrouping of pediatric acute lymphoblastic leukemia (ALL). Molecular subgroups guide risk-stratification and targeted therapies, but outcomes of recently identified subtypes are often unclear, owing to limited cases with comprehensive profiling and cross-protocol studies. We developed a machine learning tool (ALLIUM) for the molecular subclassification of ALL in retrospective cohorts as well as for up-front diagnostics. ALLIUM uses DNA methylation and gene expression data from 1131 Nordic ALL patients to predict 17 ALL subtypes with high accuracy. ALLIUM was used to revise and verify the molecular subtype of 281 B-cell precursor ALL (BCP-ALL) cases with previously undefined molecular phenotype, resulting in a single revised subtype for 81.5% of these cases. Our study shows the power of combining DNA methylation and gene expression data for resolving ALL subtypes and provides a comprehensive population-based retrospective cohort study of molecular subtype frequencies in the Nordic countries.

Abstract B110: Race/ethnicity, socioeconomic status, and early death among children enrolled in clinical trials for acute lymphoblastic leukemia

Abstract Introduction: Prior work indicates that Hispanic and Black children diagnosed with leukemia may experience a higher risk of death in the period shortly after diagnosis, but no work has examined these disparities taking into account leukemia subtypes and the role socioeconomic status (SES) may play. We investigated reported race/ethnicity (RRE), health insurance status, neighborhood SES, and early death among pediatric acute lymphoblastic leukemia (ALL) patients enrolled on clinical trials. Methods: Using data from the Childhood Cancer Research Network as well as Children’s Oncology Group therapeutic trials, we estimated the associations between RRE (Non-Hispanic [NH] White, NH Black, NH Asian/Pacific Islander [PI], or Hispanic/Native American), health insurance status, neighborhood SES using the Yost Index, and death within 60 days of diagnosis using multivariable logistic regression. All analyses were adjusted for age at diagnosis, sex, and ALL subtype (Standard Risk B-Cell, High Risk B-Cell, Infant ALL, T-Cell). We also estimated associations among non-infant patients diagnosed with B-Cell ALL, additionally adjusting for standard or high-risk disease. All associations were estimated as odds ratios (OR) with 95% confidence intervals (CI). Results: Overall, we analyzed data from 8,634 pediatric ALL cases (7,559 B-Cell; 122 Infant ALL; 953 T-Cell) enrolled across 7 clinical trials (AALL0232, AALL0331, AALL0434, AALL0631, AALL0932, AALL1131, AALL1231). In total, there were 29 deaths within 60 days of diagnosis, including 22 among B-Cell, 4 among infant ALL, and 3 among T-Cell patients. In the full analysis, infant ALL subtype was very strongly associated with early death compared to Standard Risk B-Cell patients (OR: 37.8 [95% CI: 9.22, 137]). We also observed a suggestive association between NH Asian/PI RRE and higher odds of early death compared to NH Whites (OR: 3.10 [95% CI: 0.71, 9.63]), which remained after limiting the analysis to B-Cell patients (OR: 5.63 [95% CI: 1.24, 18.7]). No association was apparent between NH Black or Hispanic/Native American RRE and odds of early death. Neighborhood SES was not significantly associated with early death, though point estimates suggested a potential protective effect of increasing SES. Lack of health insurance was also suggestively associated with greater odds of early death (OR: 2.78 [95% CI: 0.60, 9.47]). Conclusion: Unlike prior work, we observed increased odds of death soon after diagnosis for NH Asian/PI children. Future work incorporating more detailed treatment and cytogenetic subtype information is needed to identify the factors contributing to this disparity. Citation Format: Sarah Lucht, Jeannette M. Sample, Logan Spector, Jenny N. Poynter, Lucie Turcotte, Erin L. Marcotte. Race/ethnicity, socioeconomic status, and early death among children enrolled in clinical trials for acute lymphoblastic leukemia [abstract]. In: Proceedings of the 16th AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2023 Sep 29-Oct 2;Orlando, FL. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2023;32(12 Suppl):Abstract nr B110.

Allogeneic Hematopoietic Stem Cell Transplantation for Elderly Acute Lymphoblastic Leukemia Patients: A Registry Study from the Société Francophone De Greffe De Moelle Et Thérapie Cellulaire (SFGM-TC)

Background There are very scarce data regarding the outcome of elderly patients with acute lymphoblastic leukemia (ALL) who received allogeneic hematopoietic stem cell transplantation (alloHSCT) as consolidation therapy in 1rst or higher complete remission (CR). Most studies evaluating the benefit of alloHSCT in ALL include both young and elderly populations. Thus, the optimal conditioning regimen still need to be determined in a frail population represented by patients over 60 years old. In addition, total body irradiation (TBI) dose is of first importance, as prior reports demonstrated its potential higher anti-leukemic effect. We here present the outcome of ALL patients older than 59 years from the Société Francophone de Greffe de Moelle et Thérapie Cellulaire (SFGM-TC) registry. Method This is a retrospective study. The primary outcome was overall survival (OS). Secondary outcomes were progression free survival (PFS), non-relapse mortality (NRM), relapse incidence (RI), acute Graft-versus-host disease (aGvHD) grade II-IV, chronic GvHD, neutrophil engraftment and GvHD-free relapse-free survival (GRFS). Competing risks analyses were performed to analyze NRM with competing event relapse, and aGvHD grade II-IV, chronic GvHD and neutrophil engraftment with competing events relapse and death. Univariable analyses were performed using the log-rang test for OS and PFS, while Gray's test was used for cumulative incidence (CI). Multivariable analyses were performed using the Cox proportional hazards regression model including age, ALL subtype, time from diagnosis to alloHSCT, disease status at alloHSCT, donor to patient CMV status, donor to patient sex, ATG use, myeloablative conditioning (MAC), TBI use. Results A total of 316 patients ≥ 60 years old transplanted for ALL from 2012 to 2022 in 36 participating centers were included in this study. Patient's characteristics are described in Table 1. With a median follow up of 34.5 months (IQR 29.5-38.8), 3-year OS was 46% (95% CI 40-53%) (Figure 1A) with only the disease status at transplant impacting negatively OS, 53% (95% CI 46-60%) in CR1, 32% (95% CI 21-49%) in CR2, 29% (95% CI 13-63%) in advanced disease, p=0.002 and the ALL subtype, 59% (95% CI 51-68%) in Ph+ ALL, 40% (95% CI 29-54%) in Ph- ALL, 34% (95% CI 20-55%) in T-ALL, 20% (95% CI 9-42%) in other/NA, p&amp;lt;0.001. 3-year PFS was 41% (95% CI 35-48%) (Figure 1B) with the disease status at transplant impacting negatively, 49% (95% CI 42-57%) in CR1, 26% (95% CI 16-42%) in CR2, 22% (95% CI 9-58%) in advanced disease, p&amp;lt;0.001, the ALL subtype, 51% (95% CI 42-61%) in Ph+ ALL, 41% (95% CI 31-54%) in Ph- ALL, 21% (95% CI 10-41%) in T-ALL, 21% (95% CI11-43%) in other/NA, p&amp;lt;0.001, year of HSCT worse &amp;lt; 2018, p=0.007, CMV -/- worse, p=0.033, absence of TBI worse, p=0.018. 3-year NRM was 23% (95% CI 18-28%) (Figure 1C) and none of the factors impacted it. 3-year RI was 36% (95% CI 31-42%) (Figure 1D) with the disease status at transplant impacting negatively, 29% (95% CI 23-36%) in CR1, 50% (95% CI 35-63%) in CR2, 56% (95% CI 26-77%) in advanced disease, p=0.0042, the ALL subtype, 26% (95% CI 19-34%) in Ph+ ALL, 43% (95% CI 31-54%) in Ph- ALL, 57% (95% CI 38-73%) in T-ALL, 42% (95% CI 25-59%) in other/NA, p=0.0064, year of HSCT worse &amp;lt; 2018, p=0.0216, CMV -/- worse, p&amp;lt;0.001, absence of TBI worse, p=0.0069, MRD worse, p=0.0111. 3-year GRFS was 30% (95% CI 25-37%) with the disease status at transplant impacting negatively, 35% (95% CI 28-43%) in CR1, 22% (95% CI 13-37%) in CR2, 23% (95% CI 9-59%) in advanced disease, p=0.029, the ALL subtype, 37% (95% CI 30-47%) in Ph+ ALL, 33% (95% CI 23-46%) in Ph- ALL, 15% (95% CI 7-32%) in T-ALL, 17% (95% CI 8-38%) in other/NA, p=0.0029, year of HSCT worse &amp;lt; 2018, p&amp;lt;0.001. CI of aGVHD grade II-IV was 33% (95% CI 28-38%), grade III-IV 11% (95% CI 8-15%), cGVHD 35% (95% CI30-41%), extensive cGVHD 21% (95% CI 16-26%).Multivariable analyses confirmed a worse OS and PFS for advanced disease, with a HR of 1.79 (95% CI 1.22-2.64), p=0.00322 and ALL subtype with a HR for other than Ph+ ALL of 1.99 (95%CI 1.42-2.79). Conclusion: This study suggests that alloHSCT is a reasonable option for elderly ALL patients without any impact of age but advanced disease and ALL subtype other than Ph+ ALL negatively influenced the outcome.

The Impact of Age and Genomics on Drug Sensitivity in 1,076 Children and Adults with B-Cell Acute Lymphoblastic Leukemia

While the cure rate of pediatric acute lymphoblastic leukemia (ALL) now exceeds ~90% with contemporary combination chemotherapy, the prognosis for adults with ALL remains significantly inferior with long-term overall survival ranging from 50% to 70%. Recent studies have uncovered marked differences in ALL genomics between children and adults, with some high-risk subtypes becoming more prevalent with age. However, the underlying biology of age-related disparities in ALL is not fully understood, especially with regard to differences in leukemia sensitivity to chemotherapy. To address this knowledge gap, we performed exvivo drug sensitivity profiling (i.e., pharmacotyping) of 21 anti-leukemia agents on primary B-ALL diagnostic samples from 767 pediatric (age, 0-18 years) and 309 adult (19-84 years) patients. Drug sensitivity was measured as LC 50: the concentration of drug required to kill 50% of the leukemia cells (PMID: 36604538). A total of 7,975 unique LC 50 values were experimentally determined. RNA-seq was used for subtype classification and gene expression analysis. Among 21 drugs, seven showed significant differences in overall LC 50 between children and adults ( P&amp;lt;0.05 after Bonferroni correction): children displayed higher sensitivity to asparaginase, prednisolone, mercaptopurine, daunorubicin, and inotuzumab, while adults showed higher sensitivity to dasatinib and nelarabine. In multivariate models adjusting for 23 ALL molecular subtypes, only mercaptopurine remained significantly associated with age ( P=1.5×10 -5), suggesting that age-related differences in drug sensitivity can be primarily attributed to the variation in ALL subtypes between children and adults. For mercaptopurine, within KMT2A, CRLF2, and DUX4 subtypes, pediatric samples consistently showed a lower LC 50 than adults carrying the same genomic abnormality ( P=0.032, 0.0045, and 0.02, respectively). To explore intra-subtype heterogeneity, we performed unsupervised clustering using gene expression data for each of these three subtypes. Remarkably, within each of these subtypes, we identified two clusters with distinct transcriptomic profiles that were also largely segregated by age group, i.e., an adult-dominated cluster (C-a) and a pediatric-dominated cluster (C-p). In the KMT2A subtype, cases in C-a exhibited an over-representation of the KMT2A:: AFF1 fusion, and resistance to mercaptopurine ( P=0.029), prednisolone ( P=0.0039), vincristine ( P=0.046) and cytarabine ( P=0.0037). Within CRLF2 ALL, cases in C-a were associated with the presence of BCR:: ABL1-like signature and IGH:: CRLF2 rearrangements, and were more resistant to mercaptopurine ( P=0.0073) and prednisolone ( P=0.00036) compared to those in C-p. For DUX4 ALL, C-a was characterized by an under-representation of ERG deletions and resistance to mercaptopurine ( P=0.0056) and prednisolone ( P=0.0031), compared to C-p within DUX4. To explore the clinical relevance of this heterogeneity, we analyzed the invivo treatment response of KMT2A (N=35), CRLF2 (N=59) and DUX4 (N=118) B-ALL enrolled in six frontline ALL trials. Compared to cases in C-p (usually drug-sensitive), those in C-a (usually drug-resistant) consistently had significantly poorer initial treatment responses as measured by persistent end-of-induction minimal residual disease (≥0.01%) in KMT2A (58% vs 9%; P=0.0063), CRLF2 (74% vs 41%; P=0.030), and DUX4 (66% vs 40%; P=0.0058) ALL. In conclusion, these studies have revealed important new insights into the pharmacogenomic basis of age-related differences in B-ALL treatment response. These results indicate that both inter- and intra-subtype heterogeneity contribute to inferior prognosis in adults with ALL, but also point to therapeutic opportunities to improve their outcomes.

Deep Learning Models for the Diagnosis of Acute Lymphoblastic Leukemia from Bone Marrow Images : A Comprehensive Literature Review

Introduction: Acute lymphoblastic leukemia (ALL) is an aggressive blood cancer that begins in the bone marrow. It is the most common childhood cancer and requires early and accurate diagnosis for optimal treatment outcomes. Through automated image analysis of peripheral blood smears and bone marrow biopsies, artificial intelligence (AI), particularly Deep Learning (DL), have opened new avenues for improving ALL diagnosis. While bone marrow testing is the gold standard for ALL confirmation, AI applications in diagnosing ALL from bone marrow images have received less attention in the literature than peripheral blood smears. This review aims to assess the current state of AI in ALL diagnosis using bone marrow aspirates and biopsies, with a focus on Deep Learning (DL) models such as Convolutional Neural Networks (CNNs). Methods: A comprehensive literature search, conducted on June 11th, 2023, covered major medical databases, including PubMed/MEDLINE, Scopus, Embase, and Web of Science. Relevant keywords like “acute lymphoblastic leukemia,” “deep learning,” and “neural network” were employed, without time frame restrictions. Articles were included if they assessed the metrics of AI applications for ALL diagnosis in bone marrow aspirates. Articles were excluded if they: (1) had different outcomes, (2) were reviews, (3) were abstracts only, or (4) only reported peripheral blood smear metrics. Results: The search yielded 496 articles. After eliminating duplicates (282), title and abstract screening on the Rayyan platform excluded 204 records, leaving 214 articles eligible for full-text screening. Ten relevant articles were ultimately included in the review. Diverse approaches were presented to enhance diagnostic accuracy and efficiency. One study achieved 100% accuracy in classifying leukemic cells using a unique Convolutional Leaky RELU with CatBoost and XGBoost (CLR-CXG) design. Another study employed transfer learning with CNNs, obtaining 95.3% accuracy for AML, ALL, and CML classification with DenseNet121. Furthermore, the novel “i-Net” model demonstrated 99.18% validation accuracy for white blood cancer segmentation and classification. Adaptive Multi-objective CAT algorithms achieved 99.45% accuracy in detecting bone marrow cancer cells. An AI-based system using deep learning exhibited 97.2% accuracy in diagnosing ALL, while a robust fuzzy logic algorithm combined with a radial basis function neural network achieved 82.93% accuracy in diagnosing ALL in developing countries. Additionally, AI technologies based on computer microscopy demonstrated 95% and 97.5% accuracy in diagnosing ALL and minimal residual disease, respectively. A computer-aided system achieved 97.78% accuracy in classifying ALL subtypes. Conclusion: The potential of AI-driven approaches, particularly deep learning models like CNNs, in improving ALL diagnosis from bone marrow images is highlighted in this review. Because of their superior sensitivity and specificity over peripheral blood smears, bone marrow samples are the gold standard for ALL confirmation. The morphological and cellular properties of bone marrow samples provide important information for disease classification and monitoring. While DL and CNNs have shown promise in diagnosing ALL from peripheral blood smears, their use in bone marrow samples has yet to be investigated deeply. More research and external validation are required to fully realize AI's potential in ALL diagnosis from bone marrow specimens.