Combining gene mutation with gene expression analysis improves outcome prediction in acute promyelocytic leukemia

Risk factors and remaining challenges in the treatment of acute promyelocytic leukemia

The treatment of acute promyelocytic leukemia (APL) has evolved with the introduction of all-trans retinoic acid (ATRA) and subsequent arsenic trioxide (ATO), particularly in standard-risk APL with an initial white blood cell count (WBC) < 10,000/μL, where a high cure rate can now be achieved. However, for some patients with risk factors, early death or relapse remains a concern. Insights from the analysis of patients treated with ATRA and chemotherapy have identified risk factors such as WBC, surface antigens, complex karyotypes, FLT3 and other genetic mutations, p73 isoforms, variant rearrangements, and drug resistance mutations. However, in the ATRA + ATO era, the significance of these risk factors is changing. This article provides a comprehensive review of APL risk factors, taking into account the treatment approach, and explores the challenges associated with APL treatments.

Clinical networking results in continuous improvement of the outcome of patients with acute promyelocytic leukemia

ABSTRACT

The introduction of all-trans retinoic acid combined with anthracyclines has significantly improved the outcomes for patients diagnosed with acute promyelocytic leukemia (APL), and this strategy remains the standard of care in countries in which arsenic trioxide is not affordable. However, data from national registries and real-world databases indicate that low- and middle-income countries (LMIC) still face disappointing results, mainly because of high induction mortality and suboptimal management of complications. The American Society of Hematology established the International Consortium on Acute Leukemias (ICAL) to address this challenge through international clinical networking. Here, we present the findings from the International Consortium on Acute Promyelocytic Leukemia study involving 806 patients with APL recruited from 2005 to 2020 in Brazil, Chile, Paraguay, Peru, and Uruguay. The induction mortality rate has notably decreased to 14.6% compared with the pre-ICAL rate of 32%. Multivariable logistic regression analysis revealed as factors associated with induction death: age of ≥40 years, Eastern Cooperative Oncology Group performance status score of 3, high-risk status based on the Programa Español de Tratamiento en Hematologia/Gruppo Italiano Malattie EMatologiche dell'Adulto classification, albumin level of ≤3.5 g/dL, bcr3 PML/RARA isoform, the interval between presenting symptoms to diagnosis exceeding 48 hours, and the occurrence of central nervous system and pulmonary bleeding. With a median follow-up of 53 months, the estimated 4-year overall survival rate is 81%, the 4-year disease-free survival rate is 80%, and the 4-year cumulative incidence of relapse rate is 15%. These results parallel those observed in studies conducted in high-income countries, highlighting the long-term effectiveness of developing clinical networks to improve clinical care and infrastructure in LMIC.

Deep-GenMut: Automated genetic mutation classification in oncology: A deep learning comparative study

Early cancer detection and treatment depend on the discovery of specific genes that cause cancer. The classification of genetic mutations was initially done manually. However, this process relies on pathologists and can be a time-consuming task. Therefore, to improve the precision of clinical interpretation, researchers have developed computational algorithms that leverage next-generation sequencing technologies for automated mutation analysis. This paper utilized four deep learning classification models with training collections of biomedical texts. These models comprise bidirectional encoder representations from transformers for Biomedical text mining (BioBERT), a specialized language model implemented for biological contexts. Impressive results in multiple tasks, including text classification, language inference, and question answering, can be obtained by simply adding an extra layer to the BioBERT model. Moreover, bidirectional encoder representations from transformers (BERT), long short-term memory (LSTM), and bidirectional LSTM (BiLSTM) have been leveraged to produce very good results in categorizing genetic mutations based on textual evidence. The dataset used in the work was created by Memorial Sloan Kettering Cancer Center (MSKCC), which contains several mutations. Furthermore, this dataset poses a major classification challenge in the Kaggle research prediction competitions. In carrying out the work, three challenges were identified: enormous text length, biased representation of the data, and repeated data instances. Based on the commonly used evaluation metrics, the experimental results show that the BioBERT model outperforms other models with an F1 score of 0.87 and 0.850 MCC, which can be considered as improved performance compared to similar results in the literature that have an F1 score of 0.70 achieved with the BERT model.

Acute myeloid leukemia: challenges for diagnosis and treatment in Latin America

OBJECTIVE

to review the current diagnostic and therapeutic landscape of AML in Latin America as a reflection of other low- and middle-income countries and regions of the world. Encompassing both acute promyelocytic and non-promyelocytic disease types.

METHODS

We reviewed the literature and study registries concerning epidemiological features of patients with AML/APL treated in Latin America, as well as evaluated diagnostic and genetic stratification and patient fitness assessment challenges, the importance of early mortality and supportive care capacity, intensive and non-intensive chemotherapy alternatives, consolidation, and maintenance strategies including novel agents and hematopoietic stem cell transplantation.

RESULTS

Although most of the current technologies and treatment options are available in the region, a significant fraction of patients have only limited access to them. In addition, mortality in the first weeks from diagnosis is higher in the region compared to developed countries.

CONCLUSIONS

Disparities in access to technologies, supportive care capacity, and availability of novel agents and HSCT hinder results in our region, reflecting barriers common to other LMICs. Recent developments in the diagnosis and treatment of this disease must be implemented through education, collaborative clinical research, and advocacy to improve outcomes.

A novel network pharmacology approach for leukaemia differentiation therapy using Mogrify®

Acute myeloid leukaemia (AML) is a rapidly fatal blood cancer that is characterised by the accumulation of immature myeloid cells in the blood and bone marrow as a result of blocked differentiation. Methods which identify master transcriptional regulators of AML subtype-specific leukaemia cell states and their combinations could be critical for discovering novel differentiation-inducing therapies. In this proof-of-concept study, we demonstrate a novel utility of the Mogrify® algorithm in identifying combinations of transcription factors (TFs) and drugs, which recapitulate granulocytic differentiation of the NB4 acute promyelocytic leukaemia (APL) cell line, using two different approaches. In the first approach, Connectivity Map (CMAP) analysis of these TFs and their target networks outperformed standard approaches, retrieving ATRA as the top hit. We identify dimaprit and mebendazole as a drug combination which induces myeloid differentiation. In the second approach, we show that genetic manipulation of specific Mogrify®-identified TFs (MYC and IRF1) leads to co-operative induction of APL differentiation, as does pharmacological targeting of these TFs using currently available compounds. We also show that loss of IRF1 blunts ATRA-mediated differentiation, and that MYC represses IRF1 expression through recruitment of PML-RARα, the driver fusion oncoprotein in APL, to the IRF1 promoter. Finally, we demonstrate that these drug combinations can also induce differentiation of primary patient-derived APL cells, and highlight the potential of targeting MYC and IRF1 in high-risk APL. Thus, these results suggest that Mogrify® could be used for drug discovery or repositioning in leukaemia differentiation therapy for other subtypes of leukaemia or cancers.

Tanshinone IIa Induces Autophagy and Apoptosis via PI3K/Akt/mTOR Axis in Acute Promyelocytic Leukemia NB4 Cells

Tanshinone IIa (TanIIa), an ingredient of Radix Salviae Miltiorrhizae, has an anticancer effect on various solid tumors with high efficiency and low toxicity. Nonetheless, the underlying role of TanIIa in acute promyelocytic leukemia (APL) remains unclear. Here, we revealed that TanIIa drastically inhibited NB4 cell viability with an IC50 value of 31.25 μmol/L. Using flow cytometry apoptosis assay, we identified that TanIIa dose-dependently exacerbated NB4 cell apoptosis. Mechanistically, TanIIa upregulated apoptotic factor levels, namely, cleaved-caspase 9, cleaved-caspase 3, and cleaved-PARP-1. Moreover, we noticed that TanIIa dose-dependently suppressed the PI3K/Akt/mTOR axis. This axis not only functions as an essential antiapoptotic modulator but also serves as a suppressant regulator of autophagy. Correspondingly, we detected the levels of autophagic marker, namely, LC3B, which were increased after the TanIIa treatment. Furthermore, the autophagy inhibitor Baf-A1 could effectively reverse the TanIIa-induced apoptosis, manifesting that TanIIa eliminated NB4 cells in an autophagy-dependent manner. In conclusion, tanshinone IIa exerts anti-APL effects through triggering autophagy and apoptosis in NB4 cells.

Identification of variant APL translocations PRKAR1A-RARα and ZBTB16-RARα (PLZF-RARα) through the MI-ONCOSEQ platform

The cornerstone of management in patients with acute promyelocytic leukemia (APL) is early diagnosis and prompt initiation of treatment with an all-trans retinoic acid (ATRA)-based regimen. Identification of the t(15;17)(PML-RARA) chromosomal translocation through conventional cytogenetics fluorescence in-situ hybridization (FISH) or detection of the promyelocytic leukemia-retinoic acid receptor alpha (PML-RARα) fusion through RT-PCR represent the current standard of care for diagnosing APL. However, about 1-2% of patients with APL have a variant translocation involving other fusion partners with RARα besides PML. These patients present a unique diagnostic and clinical challenge in that conventional cytogenetics in addition to FISH and/or RT-PCR for PML-RARα may fail to identify these clinically relevant genetic lesions leading to an inappropriate diagnosis and treatment. We present two cases of patients who had APL with variant translocations whose bone marrow specimens were sent to the University of Michigan for enrollment in the MI-ONCOSEQ study (HUM00067928) after standard testing failed to identify PML-RARα or t(15;17) despite a phenotypic concern for this diagnosis. In these two patients, whole exome and transcriptome profiling via the MI-ONCOSEQ platform identified a PRKAR1A-RARα fusion in one patient and ZBTB16-RARα fusion in another patient. These cases illustrate the utility of whole exome and transcriptome profiling in diagnosing variant translocations in patients in whom there is a high clinical suspicion for APL based on hematopathology review.

Gene Mutation Classification through Text Evidence Facilitating Cancer Tumour Detection

A cancer tumour consists of thousands of genetic mutations. Even after advancement in technology, the task of distinguishing genetic mutations, which act as driver for the growth of tumour with passengers (Neutral Genetic Mutations), is still being done manually. This is a time-consuming process where pathologists interpret every genetic mutation from the clinical evidence manually. These clinical shreds of evidence belong to a total of nine classes, but the criterion of classification is still unknown. The main aim of this research is to propose a multiclass classifier to classify the genetic mutations based on clinical evidence (i.e., the text description of these genetic mutations) using Natural Language Processing (NLP) techniques. The dataset for this research is taken from Kaggle and is provided by the Memorial Sloan Kettering Cancer Center (MSKCC). The world-class researchers and oncologists contribute the dataset. Three text transformation models, namely, CountVectorizer, TfidfVectorizer, and Word2Vec, are utilized for the conversion of text to a matrix of token counts. Three machine learning classification models, namely, Logistic Regression (LR), Random Forest (RF), and XGBoost (XGB), along with the Recurrent Neural Network (RNN) model of deep learning, are applied to the sparse matrix (keywords count representation) of text descriptions. The accuracy score of all the proposed classifiers is evaluated by using the confusion matrix. Finally, the empirical results show that the RNN model of deep learning has performed better than other proposed classifiers with the highest accuracy of 70%.

Comprehensive analysis of cytoskeleton regulatory genes identifies ezrin as a prognostic marker and molecular target in acute myeloid leukemia

PURPOSE

Despite great advances that have been made in the understanding of the molecular complexity of acute myeloid leukemia (AML), very little has been translated into new therapies. Here, we set out to investigate the impact of cytoskeleton regulatory genes on clinical outcomes and their potential as therapeutic targets in AML.

METHODS

Gene expression and clinical data were retrieved from The Cancer Genome Atlas (TCGA) AML study and used for survival and functional genomics analyses. For pharmacological tests, AML cells were exposed to ezrin (EZR) inhibitors and submitted to several cellular and molecular assays.

RESULTS

High EZR expression was identified as an independent marker of worse outcomes in AML patients from the TCGA cohort (p < 0.05). Functional genomics analyses suggested that EZR contributes to responses to stimuli and signal transduction pathways in leukemia cells. EZR pharmacological inhibition with NSC305787 and NSC668394 reduced viability, proliferation, autonomous clonal growth, and cell cycle progression in AML cells (p < 0.05). NSC305787 had a greater potency and efficiency than NSC668394 in leukemia models. At the molecular level, EZR inhibitors reduced EZR, S6 ribosomal protein and 4EBP1 phosphorylation, and induced PARP1 cleavage in AML cells. NSC305787, but not NSC668394, favored a gene network involving cell cycle arrest and apoptosis in Kasumi 1 AML cells.

CONCLUSIONS

From our data we conclude that EZR expression may serve as a prognostic factor in AML. Our preclinical findings indicate that ezrin inhibitors may be employed as a putative novel class of AML targeting drugs.

The effects of solution-focused nursing on leukemia chemotherapy patients' moods, cancer-related fatigue, coping styles, self-efficacy, and quality of life

OBJECTIVE

To explore the effects of solution-focused nursing on leukemia chemotherapy patients' moods, cancer-related fatigue, coping styles, self-efficacy, and quality of life.

METHODS

A total of 103 patients who underwent leukemia chemotherapy in our hospital were analyzed retrospectively and were divided into two groups based on the intervention method. Group A underwent routine nursing intervention, and group B underwent solution-focused nursing. The Hamilton Anxiety Rating Scale (HAMA) scores, the Montgomery-Asberg Depression Rating Scale (MADRS) scores, the Trait Coping Style Questionnaire (TCSQ) scores, the cancer-related fatigue self-rating scores, the General Self-Efficacy Scale (GSES) scores, and the Spitzer Quality of Life Index scores were compared between the two groups.

RESULTS

Compared with group A, group B had lower HAMA scores, lower MADRS scores, lower cognitive, behavioral, perception, and emotional scores, and higher self-efficacy scores (P<0.05). Group B had higher activity scores, and better psychological statuses, support from family and friends, health perception, and outlook on life than group A after the intervention (P<0.05).

CONCLUSION

Solution-focused nursing can alleviate leukemia chemotherapy patients' negative emotions and cancer-related fatigue, improve their coping styles, and increase their self-efficacy and quality of life.

Integration of Genomic and Transcriptomic Markers Improves the Prognosis Prediction of Acute Promyelocytic Leukemia

PURPOSE

The current stratification system for acute promyelocytic leukemia (APL) is based on the white blood cell (WBC) and the platelet counts (i.e., Sanz score) over the past two decades. However, the borderlines among different risk groups are sometimes ambiguous, and for some patients, early death and relapse remained challenges. Besides, with the evolving of the treatment strategy from all-trans-retinoic acid (ATRA) and chemotherapy to ATRA-arsenic trioxide-based synergistic targeted therapy, the precise risk stratification with molecular markers is needed.

EXPERIMENTAL DESIGN

This study performed a systematic analysis of APL genomics and transcriptomics to identify genetic abnormalities in 348 patients mainly from the APL2012 trial (NCT01987297) to illustrate the potential molecular background of Sanz score and further optimize it. The least absolute shrinkage and selection operator algorithm was used to analyze the gene expression in 323 cases to establish a scoring system (i.e., APL9 score).

RESULTS

Through combining NRAS mutations, APL9 score, and WBC, 321 cases can be stratified into two groups with significantly different outcomes. The estimated 5-year overall (P = 0.00031), event-free (P < 0.0001), and disease-free (P = 0.001) survival rates in the revised standard-risk group (95.6%, 93.8%, and 98.1%, respectively) were significantly better than those in the revised high-risk group (82.9%, 77.4%, and 88.4%, respectively), which could be validated using The Cancer Genome Atlas dataset.

CONCLUSIONS

We have proposed a two-category system for improving prognosis in patients with APL. Molecular markers identified in this study may also provide genomic insights into the disease mechanism for improved therapy.

MLL5 improves ATRA driven differentiation and promotes xenotransplant engraftment in acute promyelocytic leukemia model

Although the mixed lineage leukemia 5 (MLL5) gene has prognostic implications in acute promyelocyte leukemia (APL), the underlying mechanism remains to be elucidated. Here, we demonstrate the critical role exerted by MLL5 in APL regarding cell proliferation and resistance to drug-induced apoptosis, through mtROS regulation. Additionally, MLL5 overexpression increased the responsiveness of APL leukemic cells to all-trans retinoic acid (ATRA)-induced differentiation, via regulation of the epigenetic modifiers SETD7 and LSD1. In silico analysis indicated that APL blasts with MLL5^high transcript levels were associated with retinoic acid binding and downstream signaling, while MLL5^low blasts displayed decreased expression of epigenetic modifiers (such as KMT2C, PHF8 and ARID4A). Finally, APL xenograft transplants demonstrated improved engraftment of MLL5-expressing cells and increased myeloid differentiation over time. Concordantly, evaluation of engrafted blasts revealed increased responsiveness of MLL5-expressing cells to ATRA-induced granulocytic differentiation. Together, we describe the epigenetic changes triggered by the interaction of MLL5 and ATRA resulting in enhanced granulocytic differentiation.

Targeting LSD1 for acute myeloid leukemia (AML) treatment

Targeted therapy for acute myeloid leukemia (AML) is an effective strategy, but currently there are very limited therapeutic targets for AML treatment. Histone lysine specific demethylase 1 (LSD1) is highly expressed in many cancers, impedes the differentiation of cancer cells, promotes the proliferation, metastasis and invasion of cancer cells, and is associated with poor prognosis. Targeting LSD1 has been recognized as a promising strategy for AML treatment in recent years. Based on these features, in the review, we discussed the main epigenetic drugs targeting LSD1 for AML therapy. Thus, this review focuses on the progress of LSD1 inhibitors in AML treatment, particularly those such as tranylcypromine (TCP), ORY-1001, GSK2879552, and IMG-7289 in clinical trials. These inhibitors provide novel scaffolds for designing new LSD1 inhibitors. Besides, combined therapies of LSD1 inhibitors with other drugs for AML treatment are also highlighted.

All-trans retinoic acid in non-promyelocytic acute myeloid leukemia: driver lesion dependent effects on leukemic stem cells

Acute myeloid leukemia (AML) is an aggressive, often fatal hematopoietic malignancy. All-trans retinoic acid (atRA), one of the first molecularly targeted drugs in oncology, has greatly improved the outcome of a subtype of AML, acute promyelocytic leukemia (APL). In contrast, atRA has so far provided little therapeutic benefit in the much larger group of patients with non-APL AML. Attempts to identify genetically or molecularly defined subgroups of patients that may respond to atRA have not yielded consistent results. Since AML is a stem cell-driven disease, understanding the effectiveness of atRA may require an appreciation of its impact on AML stem cells. Recent studies reported that atRA decreased stemness of AML with an FLT3-ITD mutation, yet increased it in AML1-ETO driven or EVI1-overexpressing AML. This review summarizes the role of atRA in normal hematopoiesis and in AML, focusing on its impact on AML stem cells.

Genetic biomarkers of drug resistance: A compass of prognosis and targeted therapy in acute myeloid leukemia

Acute myeloid leukemia (AML) is a highly aggressive hematological malignancy with complex heterogenous genetic and biological nature. Thus, prognostic prediction and targeted therapies might contribute to better chemotherapeutic response. However, the emergence of multidrug resistance (MDR) markedly impedes chemotherapeutic efficacy and dictates poor prognosis. Therefore, prior evaluation of chemoresistance is of great importance in therapeutic decision making and prognosis. In recent years, preclinical studies on chemoresistance have unveiled a compendium of underlying molecular basis, which facilitated the development of targetable small molecules. Furthermore, routing genomic sequencing has identified various genomic aberrations driving cellular response during the course of therapeutic treatment through adaptive mechanisms of drug resistance, some of which serve as prognostic biomarkers in risk stratification. However, the underlying mechanisms of MDR have challenged the certainty of the prognostic significance of some mutations. This review aims to provide a comprehensive understanding of the role of MDR in therapeutic decision making and prognostic prediction in AML. We present an updated genetic landscape of the predominant mechanisms of drug resistance with novel targeted therapies and potential prognostic biomarkers from preclinical and clinical chemoresistance studies in AML. We particularly highlight the unfolded protein response (UPR) that has emerged as a critical regulatory pathway in chemoresistance of AML with promising therapeutic horizon. Futhermore, we outline the most prevalent mutations associated with mechanisms of chemoresistance and delineate the future directions to improve the current prognostic tools. The molecular analysis of chemoresistance integrated with genetic profiling will facilitate decision making towards personalized prognostic prediction and enhanced therapeutic efficacy.

Classic and Variants APLs, as Viewed from a Therapy Response

Most acute promyelocytic leukemia (APL) are caused by PML-RARA, a translocation-driven fusion oncoprotein discovered three decades ago. Over the years, several other types of rare X-RARA fusions have been described, while recently, oncogenic fusion proteins involving other retinoic acid receptors (RARB or RARG) have been associated to very rare cases of acute promyelocytic leukemia. PML-RARA driven pathogenesis and the molecular basis for therapy response have been the focus of many studies, which have now converged into an integrated physio-pathological model. The latter is well supported by clinical and molecular studies on patients, making APL one of the rare hematological disorder cured by targeted therapies. Here we review recent data on APL-like diseases not driven by the PML-RARA fusion and discuss these in view of current understanding of "classic" APL pathogenesis and therapy response.

Clinical implications of recurrent gene mutations in acute myeloid leukemia

Acute myeloid leukemia (AML) is a genetically heterogeneous clonal malignancy characterized by recurrent gene mutations. Genomic heterogeneity, patients’ individual variability, and recurrent gene mutations are the major obstacles among many factors that impact treatment efficacy of the AML patients. With the application of cost- and time-effective next-generation sequencing (NGS) technologies, an enormous diversity of genetic mutations has been identified. The recurrent gene mutations and their important roles in acute myeloid leukemia (AML) pathogenesis have been studied extensively. In this review, we summarize the recent development on the gene mutation in patients with AML.

Acute Promyelocytic Leukemia: A Constellation of Molecular Events around a Single PML-RARA Fusion Gene

Although acute promyelocytic leukemia (APL) is one of the most characterized forms of acute myeloid leukemia (AML), the molecular mechanisms involved in the development and progression of this disease are still a matter of study. APL is defined by the PML-RARA rearrangement as a consequence of the translocation t(15;17)(q24;q21). However, this abnormality alone is not able to trigger the whole leukemic phenotype and secondary cooperating events might contribute to APL pathogenesis. Additional somatic mutations are known to occur recurrently in several genes, such as FLT3, WT1, NRAS and KRAS, whereas mutations in other common AML genes are rarely detected, resulting in a different molecular profile compared to other AML subtypes. How this mutational spectrum, including point mutations in the PML-RARA fusion gene, could contribute to the 10%–15% of relapsed or resistant APL patients is still unknown. Moreover, due to the uncertain impact of additional mutations on prognosis, the identification of the APL-specific genetic lesion is still the only method recommended in the routine evaluation/screening at diagnosis and for minimal residual disease (MRD) assessment. However, the gene expression profile of genes, such as ID1, BAALC, ERG, and KMT2E, once combined with the molecular events, might improve future prognostic models, allowing us to predict clinical outcomes and to categorize APL patients in different risk subsets, as recently reported. In this review, we will focus on the molecular characterization of APL patients at diagnosis, relapse and resistance, in both children and adults. We will also describe different standardized molecular approaches to study MRD, including those recently developed. Finally, we will discuss how novel molecular findings can improve the management of this disease.

Acute Promyelocytic Leukemia: Update on the Mechanisms of Leukemogenesis, Resistance and on Innovative Treatment Strategies

This review highlights new findings that have deepened our understanding of the mechanisms of leukemogenesis, therapy and resistance in acute promyelocytic leukemia (APL). Promyelocytic leukemia-retinoic acid receptor α (PML-RARa) sets the cellular landscape of acute promyelocytic leukemia (APL) by repressing the transcription of RARa target genes and disrupting PML-NBs. The RAR receptors control the homeostasis of tissue growth, modeling and regeneration, and PML-NBs are involved in self-renewal of normal and cancer stem cells, DNA damage response, senescence and stress response. The additional somatic mutations in APL mainly involve FLT3, WT1, NRAS, KRAS, ARID1B and ARID1A genes. The treatment outcomes in patients with newly diagnosed APL improved dramatically since the advent of all-trans retinoic acid (ATRA) and arsenic trioxide (ATO). ATRA activates the transcription of blocked genes and degrades PML-RARα, while ATO degrades PML-RARa by promoting apoptosis and has a pro-oxidant effect. The resistance to ATRA and ATO may derive from the mutations in the RARa ligand binding domain (LBD) and in the PML-B2 domain of PML-RARa, but such mutations cannot explain the majority of resistances experienced in the clinic, globally accounting for 5-10% of cases. Several studies are ongoing to unravel clonal evolution and resistance, suggesting the therapeutic potential of new retinoid molecules and combinatorial treatments of ATRA or ATO with different drugs acting through alternative mechanisms of action, which may lead to synergistic effects on growth control or the induction of apoptosis in APL cells.